diff options
Diffstat (limited to 'lib')
181 files changed, 45835 insertions, 0 deletions
diff --git a/lib/.cvsignore b/lib/.cvsignore new file mode 100644 index 0000000..5e41225 --- /dev/null +++ b/lib/.cvsignore @@ -0,0 +1,17 @@ +*.bb +*.bbg +*.da +*.plg +.deps +.pure +Makefile +WinDebug +WinRel +alloca.h +fnmatch.h +glob.h +getopt.h +stdbool.h +stdint.h +getdate +getdate.log diff --git a/lib/ChangeLog b/lib/ChangeLog new file mode 100644 index 0000000..077df81 --- /dev/null +++ b/lib/ChangeLog @@ -0,0 +1,2726 @@ +2005-09-29 Derek Price <derek@ximbiot.com> + + * system.h: #define fd_select select when not already defined. + +2005-09-25 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Add files (mkstemp.c tempname.c) to project. + + * libcvs.dep, libcvs.mak: Regenerated for "libcvs.dsp" changes. + +2005-09-21 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dep: Regenerated for "../windows-NT/fix-msvc-mak.pl" change + to remove "basetsd" references. + +2005-09-20 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Add files (canon-host.c gai_strerror.c getaddrinfo.c + getdelim.c mbchar.c strnlen1.c canon-host.h getaddrinfo.h getdelim.h + glob-libc.h mbchar.h mbuiter.h regex_internal.h strcase.h strnlen1.h + ..\windows-NT\netdb.h ..\windows-NT\sys\socket.h) to project. + + * libcvs.dep libcvs.mak: Regenerated for "libcvs.dsp" changes. + +2005-09-20 Derek Price <derek@ximbiot.com> + + * getaddrinfo.c: Update from GNULIB. + +2005-09-20 Derek Price <derek@ximbiot.com> + Paul Eggert <eggert@cs.ucla.edu> + + * regcomp.c, regex_internal.c, regex_internal.h, regexec.c: New, + simpler replacement fix for the previous. + +2005-09-19 Derek Price <derek@ximbiot.com> + + * regcomp.c, regex_internal.c, regex_internal.h, regexec.c: Fix some + __attribute definition problems (submitted to GNULIB). + + * __fpending.c, __fpending.h, atexit.c, basename.c, chdir-long.c, + closeout.c, cycle-check.c, dirname.c, dup-safer.c, dup2.c, exitfail.c, + fd-safer.c, filenamecat.c, fnmatch.c, ftruncate.c, getaddrinfo.c, + getcwd.c, getdate.h, getdelim.c, getline.c, getlogin_r.c, getndelim2.c, + getnline.c, getopt1.c, getpass.c, gettimeofday.c, lstat.c, malloc.c, + memmove.c, mkdir.c, mkstemp.c, nanosleep.c, openat.c, pipe-safer.c, + quotearg.c, readlink.c, realloc.c, regex.c, rename.c, rpmatch.c, + save-cwd.c, strerror.c, stripslash.c, strncasecmp.c, strnlen1.c, + strstr.c, strtol.c, tempname.c, time_r.c, timespec.h, xalloc-die.c, + xgetcwd.c, xmalloc.c, xreadlink.c, yesno.c: Update from GNULIB. + + * canon-host.h, md5.c, md5.h, mktime.c, timespec.h: Update from GNULIB. + +2005-09-15 Derek Price <derek@ximbiot.com> + + * regcomp.c, regex_internal.c, regex_internal.h, regexec.c: Update from + GNULIB, again. + + * regcomp.c, regex_internal.c, regex_internal.h, regexec.c: Update from + GNULIB. + + * strstr.h: Update from GNULIB. + +2005-09-13 Derek Price <derek@ximbiot.com> + + * canon-host.c: Update from GNULIB. + +2005-09-12 Derek Price <derek@ximbiot.com> + + * gai_strerror.c: Include config.h when available. Include + getaddrinfo.h before other headers to test interface. + +2005-09-10 Larry Jones <lawrence.jones@ugs.com> + + * canon-host.c (canon_host_r): Designated initializers are C99 only. + * getaddrinfo.c (getaddrinfo): Can't do arithmetic on void *. Fix + incorrect hints member name. Fix const mismatch. + +2005-09-09 Derek Price <derek@ximbiot.com> + + Install glibc porting changes. + * glob.c, glob_.h: Update with alpha-GNULIB. + * glob-libc.h: New file from same. + +2005-09-08 Derek Price <derek@ximbiot.com> + + * getaddrinfo.h: Don't assume HAVE_GETADDRINFO will be defined when the + system does not haev getaddrinfo. + +2005-09-06 Derek Price <derek@ximbiot.com> + + * regex_internal.h: Update from GNULIB. + + * getpass.c, regcomp.c, regex_internal.c, regex_internal.h, regexec.c: + Update from GNULIB. + + Update from GNULIB with alpha-quality patch. + * Makefile.am: Add getaddrinfo cruft. + * canon-host.c: Updated to use getaddrinfo module. + * canon-host.h, gai_strerror.c, getaddrinfo.c, getaddrinfo.h: New + files. + +2005-09-05 Derek Price <derek@ximbiot.com> + + * Makefile.am (mbchar): Remove obsolete module cruft. + +2005-09-04 Derek Price <derek@ximbiot.com> + + Update from GNULIB. + * glob.c, glob_.h: Updated. + * Makefile.am: Move glob targets into alpha order. Add size_max + module. + + Add canon-host module from GNULIB. + * canon-host.c: New file. + + Update from GNULIB. + * Makefile.am: Import modifications from... + * Makefile.gnulib: ...here. + * dup-safer.c, fd-safer.c, getpass.c, regcomp.c, regex.c, regex.h, + regex_internal.c, regex_internal.h, regexec.c, strcase.h, + strcasecmp.c, strftime.c, strncasecmp.c, strstr.c, strstr.h, + unistd-safer.h: Updated. + * mbchar.c, mbchar.h, mbuiter.h, memchr.c, pipe-safer.c, strnlen1.c, + strnlen1.h, unistd--.h: New files. + +2005-08-31 Derek Price <derek@ximbiot.com> + + * getdelim.c (getdelim): Return EOF on EOF, revisited. + +2005-08-31 Derek Price <derek@ximbiot.com> + + * getdelim.c (getdelim): Return EOF on EOF. + +2005-08-12 Derek Price <derek@ximbiot.com> + + Update from GNULIB. + * Makefile.am: Update using Makefile.gnulib changes. + * Makefile.gnulib, dup-safer.c, dup2.c, fnmatch.c, getcwd.c, getcwd.h, + getline.c, getline.h, getlogin_r.h, mktime.c, openat.h, + pagealign_alloc.c, quotearg.c, regex.c, regex.h, save-cwd.c, + strftime.c, tempname.c, time_r.h, xmalloc.c: Updated. + * getdelim.c, getdelim.h, regcomp.c, regex_internal.c, + regex_internal.h, regexec.c, size_max.h: New files. + +2005-06-12 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Add files "filenamecat.h" and "lstat.c" to project. + * libcvs.dep, libcvs.mak: Regenerate for "libcvs.dsp" changes. + +2005-06-10 Derek Price <derek@ximbiot.com> + + Sync with GNULIB. + * basename.c, canonicalize.c, canonicalize.h, dirname.c, dirname.h, + fnmatch_.h, fnmatch_loop.c, getcwd.c, getlogin_r.c, mkstemp.c, + openat.c, openat.h, pathmax.h, rename.c, stripslash.c, tempname.c, + xreadlink.c: Update from GNULIB. + * path-concat.c, path-concat.h: Rename to... + * filenamecat.c, filenamecat.h: ...this. + + Test new stat module. + * lstat.h: New file. + * stat.c: Remove this file and... + * lstat.c: ...move most functionality here. + +2005-05-30 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Add file "windows-NT/woe32.h" to project. + * libcvs.dep: Regenerate for "libcvs.dsp" change. + +2005-05-28 Derek Price <derek@ximbiot.com> + + * glob_.h: s/MISSING_SYS_CDEFS_H/_SYS_CDEFS_H/. Add comment. + +2005-05-27 Derek Price <derek@ximbiot.com> + + * getlogin_r.h: Import new version from GNULIB. + +2005-05-26 Derek Price <derek@ximbiot.com> + + * glob_.h: Protect __THROW separately from __BEGIN_DECLS. + +2005-05-26 Derek Price <derek@ximbiot.com> + + * system.h: Move non-system substitute GNULIB headers into cvs.h. + Remove stat macros in favor of GNULIB stat-macros module. Assume + <sys/stat.h> and <stdlib.h> per HACKING. + +2004-05-26 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Add files "getlogin_r.c" and "getlogin_r.h" to project. + * libcvs.dep, libcvs.mak: Regenerated for "libcvs.dsp" change. + +2004-05-26 Conrad T. Pino <Conrad@Pino.com> + + * glob_.h: Windows build fakes "ndir.h" which defines "struct direct" + which differs from "dirent.h" which defines "struct dirent". + +2005-05-25 Derek Price <derek@ximbiot.com> + + * glob.c (LOGIN_NAME_MAX): Wrap sysconf for Windows. + +2005-05-25 Derek Price <derek@ximbiot.com> + + * glob_.h: Simplify <sys/cdefs.h> include and GLIBC macro use. Replace + use of defined _LIBC with !defined GLOB_PREFIX. Restore some GLIBC + stuff we don't care about anyhow. Separate some #defines into their + own section for clarity. + (__glob_pattern_p): Move this definition... + * glob.c (__glob_pattern_p): ...here. + (HAVE_D_TYPE): Remove this macro for simplicity's sake. + (HAVE__POSIX_GETPWNAM_R): Ditto. + * getlogin_r.c, getlogin_r.h: Update from GNULIB. + +2005-05-25 Derek Price <derek@ximbiot.com> + + * glob.c: Remove HAVE_GETLOGIN_R complexity in favor of the GNULIB + getlogin_r module. Redefine "struct_stat64" rather than "stat64" to + avoid macro collisions on Solaris. + * getlogin_r.c, getlogin_r.h: New files from GNULIB. + +2004-05-24 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Add files "canonicalize.c", "cycle-check.c", + "canonicalize.h", "cycle-check.h", "dev-ino.h", "glob.h", "mempcpy.h", + "path-concat.h", "stat-macros.h" and "strdup.h" to project. + * libcvs.dep, libcvs.mak: Regenerated for "libcvs.dsp" change. + +2005-05-24 Conrad T. Pino <Conrad@Pino.com> + + * glob_.h: Remove "struct stat;" forward declaration and update comment. + +2005-05-24 Conrad T. Pino <Conrad@Pino.com> + + * glob_.h: "#include <sys/stat.h>" for Solaris "#define stat stat64" + case. + +2005-05-24 Derek Price <derek@ximbiot.com> + + * stat.c: Use system stat & lstat functions in this file. + * system.h (CVS_STAT, CVS_LSTAT): Remove these macros. + +2005-05-24 Derek Price <derek@ximbiot.com> + + * Makefile.am (BUILT_SOURCES, EXTRA_DIST, MAINTAINERCLEANFILES): Add + getdate.c due to automake bug. + +2005-05-24 Derek Price <derek@ximbiot.com> + + * glob_.h, glob.c: Don't attempt to handle 64 bit file information + explicitly when not compiling as part of glibc. + +2005-05-23 Derek Price <derek@ximbiot.com> + + * canonicalize.c, canonicalize.h, cycle-check.c, cycle-check.h, + dev-ino.h, path-concat.c, path-concat.h: New files from GNULIB. + +2005-05-23 Derek Price <derek@ximbiot.com> + + * Makefile.am: Remove unnecessary comment. + +2005-05-23 Derek Price <derek@ximbiot.com> + + * __fpending.c, alloca_.h, allocsa.c, allocsa.h, asnprintf.c, + asprintf.c, basename.c, chdir-long.c, chdir-long.h, closeout.c, + closeout.h, dirname.c, dirname.h, dup-safer.c, dup2.c, error.c, + error.h, exit.h, exitfail.c, exitfail.h, fd-safer.c, fnmatch.c, + fnmatch_.h, fnmatch_loop.c, getcwd.c, getcwd.h, getdate.c, getdate.h, + getdate.y, gethostname.c, getline.c, getline.h, getndelim2.c, + getndelim2.h, getnline.c, getnline.h, getopt.c, getopt1.c, getopt_.h, + getopt_int.h, getpagesize.h, getpass.c, getpass.h, gettext.h, + gettime.c, gettimeofday.c, lstat.c, malloc.c, md5.c, md5.h, mempcpy.c, + mempcpy.h, memrchr.c, memrchr.h, minmax.h, mkdir.c, mkstemp.c, + mktime.c, nanosleep.c, openat.c, openat.h, pagealign_alloc.c, + pagealign_alloc.h, pathmax.h, printf-args.c, printf-args.h, + printf-parse.c, printf-parse.h, quotearg.c, quotearg.h, readlink.c, + realloc.c, regex.c, regex.h, rename.c, rpmatch.c, save-cwd.c, + save-cwd.h, setenv.c, setenv.h, stat-macros.h, stat.c, stdbool_.h, + stdint_.h, strcase.h, strcasecmp.c, strdup.c, strdup.h, strerror.c, + strftime.c, strftime.h, stripslash.c, strstr.c, strstr.h, strtol.c, + strtoul.c, tempname.c, time_r.c, time_r.h, timespec.h, unistd-safer.h, + unlocked-io.h, unsetenv.c, vasnprintf.c, vasnprintf.h, vasprintf.c, + vasprintf.h, xalloc-die.c, xalloc.h, xgetcwd.c, xgetcwd.h, + xgethostname.c, xmalloc.c, xreadlink.c, xreadlink.h, xsize.h, yesno.c, + yesno.h: Update from GNULIB. + +2005-05-20 Derek Price <derek@ximbiot.com> + + * glob.c (__stat64): Add args to macro definition for clarity. + +2005-05-19 Derek Price <derek@ximbiot.com> + + * glob.c: Use POSIX getpwnam_r on Solaris. + +2005-05-19 Derek Price <derek@ximbiot.com> + + * glob.c: Don't delcare protos for getlogin or getlogin_r if we got one + from <unistd.h>. + +2005-05-18 Derek Price <derek@ximbiot.com> + + * glob.c (GETPW_R_SIZE_MAX): Remove `;' from macro definition. + +2005-05-18 Derek Price <derek@ximbiot.com> + + More misc cleanup for compatibility with GNULIB & glibc, mostly + suggested by Paul Eggert <eggert@cs.ucla.edu>. + * glob.c: Improve comments. + (GETPW_R_SIZE_MAX): Add parens to definition. + (glob): Use GNU coding conventions (formatting change). + * glob_.h: Include <sys/cdefs.h> when possible. + +2005-05-17 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Add "glob.c", "glob_.h" and "mempcpy.c" to project. + * libcvs.dep, libcvs.mak: Regenerated for "libcvs.dsp" change. + +2005-05-17 Derek Price <derek@ximbiot.com> + + More misc cleanup for compatibility with GNULIB & glibc, mostly + suggested by Paul Eggert <eggert@cs.ucla.edu>. + * glob.c: Improve comments. + (GETPW_R_SIZE_MAX): New macro to avoid undefined sysconf macros. + (glob): Use new macro. + * glob_.h: Define __const, __restrict, and __USE_GNU when necessary. + Protect use of __BEGIN_DECLS & __END_DECLS. Remove include of + <sys/types.h>. Reorganize definitions for GNULIB. Remove #undefs + which work around what is now probably a non-existant problem. + +2005-05-15 Derek Price <derek@ximbiot.com> + + * glob_.h, glob.c: More misc cleanup for compatibility with GNULIB & + glibc, mostly suggested by Paul Eggert <eggert@cs.ucla.edu>. Remove + casts to and from void * - they are not needed in C89. + +2005-05-13 Derek Price <derek@ximbiot.com> + + * glob_.h: Enable GNU features by default. Remove copy/pasted getopt + #define. + * glob.c: s/__strdup/strdup/ (Fixes BSD compilation issue). + +2005-05-13 Derek Price <derek@ximbiot.com> + + * glob.c, glob_.h: Misc cleanup for compatibility with GNULIB & glibc, + mostly suggested by Paul Eggert <eggert@cs.ucla.edu>. + +2005-05-12 Derek Price <derek@ximbiot.com> + + * glob.c, glob_.h: Misc improvements suggested by Paul Eggert, bringing + these files closer to their glibc sources. Simplify & reorganize for + clarity. Use protos. + (glob_in_dir): Fix GLOB_ONLYDIR bug/typo. + +2005-05-11 Derek Price <derek@ximbiot.com> + + * .cvsignore: Ignore glob.h. + * Makefile.am (BUILT_SOURCES, EXTRA_DIST, MOSTLYCLEANFILES): Add glob + files. + (glob.h): New target. + * glob.c, glob_.h: New files. + * strdup.c, strdup.h: New files from GNULIB. + +2005-05-09 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dep, libcvs.mak: Regenerated after Windows full rebuild. + +2005-05-08 Mark D. Baushke <mdb@cvshome.org> + + * libcvs.dsp: unistd-safer.h is in lib, not windows-NT. + * libcvs.dep: Regenerate. + +2005-05-07 Mark D. Baushke <mdb@cvshome.org> + + * libcvs.dep: Use a relative path for unistd-safer.h. + + * libcvs.dsp: Update from GNULIB. + Add files dup-safer.c, fd-safer.c, and unistd-safer.h. + * libcvs.dep, libcvs.mak: Regenerate. + +2005-05-03 Derek Price <derek@ximbiot.com> + + Update from GNULIB. + * chdir-long.c, nanosleep.c, save-cwd.c, tempname.c: Updated. + * dup-safer.c, fd-safer.c, unistd-safer.h: New files. + +2005-05-01 Mark D. Baushke <mdb@cvshome.org> + + * regex.c: Update from GNULIB. + +2005-04-25 Derek Price <derek@ximbiot.com> + + * getdate.y: Update from GNULIB. + +2005-04-15 Derek Price <derek@ximbiot.com> + + * Makefile.am (DISTCLEANFILES): Add getdate.log. + * .cvsignore: Ditto. + * test-getdate.sh: Add new tests. + (verify, skip, valid_timezone): New functions. + +2005-04-12 Derek Price <derek@ximbiot.com> + + * getdate.y: Update from GNULIB. + +2005-04-06 Derek Price <derek@ximbiot.com> + + * getdate.y: Update from GNULIB. + +2005-04-03 Mark D. Baushke <mdb@cvshome.org> + + * strtol.c: Update from GNULIB. + +2005-03-29 Mark D. Baushke <mdb@cvshome.org> + + * mktime.c, strftime.c: Update from GNULIB. + +2005-03-23 Derek Price <derek@ximbiot.com> + + * Makefile.am (getdate_SOURCES): Remove sources that are now included + via $(noinst_LIBRARIES). + +2005-03-22 Mark D. Baushke <mdb@cvshome.org> + + * mktime.c, regex.c, stdint_.h, strftime.c, strtol.c: Update from + GNULIB. + +2005-03-10 Mark D. Baushke <mdb@cvshome.org> + + * mktime.c, save-cwd.c, strftime.c, strtol.c: Update from GNULIB. + +2005-03-08 Conrad T. Pino <Conrad@Pino.com> + + * .cvsignore: Add stdint.h entry. + +2005-03-07 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Synchronize with GNULIB updates. + Remove valloc.c file. Add files __fpending.c, __fpending.h, + chdir-long.h, closeout.c, closeout.h, getcwd.h, + pagealign_alloc.c, pagealign_alloc.h, quotearg.c, quotearg.h. + * libcvs.dep, libcvs.mak: Regenerate for libcvs.dsp change. + +2005-03-07 Mark D. Baushke <mdb@cvshome.org> + + * pagealign_alloc.c: Update from GNULIB. + +2005-03-04 Jim Hyslop <jhyslop@ieee.org> + + * xtime.h: added include guards to fix compile errors on IRIX 5.3 + (Patch from Georg Schwarz <georg.scwarz@freenet.de>.) + +2005-03-03 Derek Price <derek@ximbiot.com> + + * pagealign_alloc.c, pagealign_alloc.h: New files from GNULIB. + * valloc.c: Remove this obsolete file. + +2005-03-02 Jim Meyering <jim@meyering.net> + + * Makefile.am (libcvs_a_SOURCES): Add the following to support + the GNULIB closeout module: closeout.c, closeout.h, quotearg.c, + quotearg.h, __fpending.h. + * closeout.c, closeout.h: New files. + * quotearg.c, quotearg.h: New files. + * __fpending.c, __fpending.h: New files. + * Makefile.gnulib: Regenerate. + +2005-03-02 Derek Price <derek@ximbiot.com> + + * gettext.h, setenv.c, vasnprintf.c, vasprintf.c: Update from GNULIB. + +2005-03-01 Derek Price <derek@ximbiot.com> + + Update installed GNULIB modules. + * Makefile.am: Tweak for new GNULIB updates. + * alloca_.h, dup2.c, error.c, ftruncate.c, getdate.c, getdate.y, + getopt_.h, gettext.h, gettime.c, malloc.c, mkdir.c, mktime.c, + realloc.c, rename.c, save-cwd.c, strstr.c, strtoul.c, timespec.h, + xgetcwd.c: Update these files. + * chdir-long.c, chdir-long.h, getcwd.c, getcwd.h, mempcpy.c, mempcpy.h, + memrchr.c, memrchr.h, openat.c, openat.h, strstr.h, strtol.c: New files + from GNULIB. + +2005-03-01 Derek Price <derek@ximbiot.com> + + * xtime.h: Remove obsolete timeb stuff. + +2005-02-20 Mark D. Baushke <mdb@cvshome.org> + + * xgethostname.c (xgethostname): Check for ENOMEM, which is + returned by OSX/Darwin if the specified buffer is not large + enough for the hostname. + (Problem reported by Neil Conway <neilc@samurai.com>.) + +2005-02-08 Derek Price <derek@ximbiot.com> + + * fncase.c (OSX_filename_classes): Mac OSX doesn't need \ mapped to /. + * system.h (FOLD_FN_CASE): Clarify comment. + +2005-01-31 Derek Price <derek@ximbiot.com> + + * Makefile.am: Update copyright notices. + +2004-12-14 Derek Price <derek@ximbiot.com> + + * Makefile.am (EXTRA_DIST): Add Makefile.gnulib. + +2004-11-30 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dep: Regenerated for "../cvsnt.dsp" change. + * libcvs.mak: Regenerated for "../cvsnt.dsp" change. + +2004-11-21 Mark D. Baushke <mdb@cvshome.org> + + * libcvs.dep, libcvs.dsp, libcvs.mak: Add "xgethostname.c" to + project. + +2004-11-19 Derek Price <derek@ximbiot.com> + + * xgethostname.h, xgethostname.c: New files from GNULIB. + * Makefile.am (libcvs_a_SOURCES): Add new files. + +2004-11-17 Derek Price <derek@ximbiot.com> + + * getdate.y, getopt.c, getopt1.c, getopt_.h, getopt_int.h, xgetcwd.c: + Update from GNULIB. + +2004-11-15 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_LIBADD): Add @ALLOCA@. + +2004-11-11 Derek Price <derek@ximbiot.com> + + * Makefile.am (EXTRA_DIST): Add test-getdate.sh. + +2004-11-11 Mark D. Baushke <mdb@cvshome.org> + + * alloca_.h, allocsa.h, mktime.c: Update from GNULIB. + +2004-11-11 Mark D. Baushke <mdb@cvshome.org> + + * strftime.c, getdate.y: Update from GNULIB. + * getdate.c: Regenerated. + +2004-11-10 Mark D. Baushke <mdb@cvshome.org> + + * test-getdate.sh: Remove non-portable 'Feb 29, 2096 8 years' + test case. Add comments about portable dates. + +2004-11-10 Mark D. Baushke <mdb@cvshome.org> + + * test-getdate.sh (TZ): Set to UTC0 instead of UTC to fix MacOS X + problem. + +2004-11-10 Mark D. Baushke <mdb@cvshome.org> + + * mktime.c (SHR): New macro, which is a portable substitute for >> + that should work even on Crays. + Problem reported by Mark D. Baushke. + (TIME_T_MIDPOINT, ydhms_diff, __mktime_internal): Use it. + (Patch from Paul Eggert of GNULIB.) + + * Makefile.am (getdate_LDADD): Add @ALLOCA@. + * Makefile.in: Regenerated. + +2004-11-09 Mark D. Baushke <mdb@cvshome.org> + + * strftime.c: Update from GNULIB. + +2004-11-05 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Add "readlink.c" to project. + * libcvs.dep: Regenerate for "libcvs.dsp" change. + * libcvs.mak: Regenerate for "libcvs.dsp" change. + +2004-11-05 Mark D. Baushke <mdb@cvshome.org> + + * readlink.c: Update from GNULIB. + +2004-11-04 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Add "xreadlink.c" and "xreadlink.h" to project. + * libcvs.dep: Regenerate for "libcvs.dsp" change. + * libcvs.mak: Regenerate for "libcvs.dsp" change. + +2004-11-04 Mark D. Baushke <mdb@cvshome.org> + + * allocsa.h, allocsa.valgrind: Update from GNULIB. + * Makefile.in: Regenerated. + +2004-11-04 Mark D. Baushke <mdb@cvshome.org> + + * sunos57-select.c: Move '#undef select' before system #include + statements to avoid conflicting declarations on BSDI BSD/OS 4.2. + +2004-11-03 Mark D. Baushke <mdb@cvshome.org> + + * Makefile.am: Update from GNULIB. + * readlink.c, xreadlink.c, xreadlink.h: Add from GNULIB. + * Makefile.gnulib, Makefile.in: Regenerated. + +2004-11-03 Derek Price <derek@ximbiot.com> + + * getdate.y: Update from GNULIB. + +2004-11-02 Mark D. Baushke <mdb@cvshome.org> + + * getpass.c, setenv.h: Update from GNULIB. + +2004-11-02 Mark D. Baushke <mdb@cvshome.org> + + * test-getdate.sh (getdate): Do not assume 'diff -u' is + universally available. + * Makefile.am (getdate_SOURCES): Add allocsa.c sources needed by + getdate on Solaris. + * Makefile.in: Regenerated. + +2004-11-01 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Add "setenv.c", "unsetenv.c", "allocsa.h" and "setenv.h" to project. + * libcvs.dep: Regenerate for "libcvs.dsp" change. + * libcvs.mak: Regenerate for "libcvs.dsp" change. + +2004-11-01 Derek Price <derek@ximbiot.com> + + * test-getdate.sh: Test a date relative to a specified date and one of + the failing dates from the rcs2-7 test in src/sanity.sh. + +2004-11-01 Derek Price <derek@ximbiot.com> + + * Makefile.am (TESTS): Add test-getdate.sh. + (check_PROGRAMS): Add getdate. + (getdate_*): Add info to build getdate. + (MOSTLYCLEANFILES): Add files that might be generated by + test-getdate.sh. + * allocsa.c, allocsa.h, setenv.c setenv.h, unsetenv.c: New files from + GNULIB. + * test-getdate.sh: Add some new dates and expect correct output. + +2004-10-27 Mark D. Baushke <mdb@cvshome.org> + + * mktime.c (not_equal_tm): New patch from GNULIB. + +2004-10-26 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Add "rpmatch.c" and "yesno.h" to project. + * libcvs.dep: Regenerate for "libcvs.dsp" change. + * libcvs.mak: Regenerate for "libcvs.dsp" change. + +2004-10-26 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Move error.h to its own section. + +2004-10-21 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Add yesno.h. + * rpmatch.c, yesno.h: New files from GNULIB. + * yesno.c: Update from GNULIB. + +2004-10-23 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Add "../windows-NT/stdint.h" to project. + * libcvs.dep: Regenerate for "libcvs.dsp" change. + +2004-10-22 Mark D. Baushke <mdb@cvshome.org> + + * Makefile.am (libcvs_a_OBJECTS): Fix typo for $(ALLOCA_H) and + ($STDBOOL_H). Also update rules for stdint GNULIB module. + * Makefile.in: Regenerated. + +2004-10-22 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Remove "argmatch.c" from project. + +2004-10-22 Mark D. Baushke <mdb@cvshome.org> + + * Makefile.am (libcvs_a_SOURCES): Add stdint GNULIB module. + Fix typos in alloca and stdbool sections. + * stdint_.h: New file from GNULIB. + * getpagesize.h: Update from GNULIB. + * mktime.c: Update from GNULIB. + +2004-10-22 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Move md5 to its own section. + * md5.c, md5.h: Update from GNULIB. + +2004-10-21 Derek Price <derek@ximbiot.com> + + * Makefile.am, rpmatch.c, yesno.h, yesno.c: Back out recent changes. + +2004-10-21 Derek Price <derek@ximbiot.com> + + * error.c, error.h: Update from GNULIB. + +2004-10-21 Mark Baushke <mdb@cvshome.org> + + * libcvs.dep, libcvs.mak: Remove argmatch references. + +2004-10-21 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Move yesno to its own section. + * rpmatch.c, yesno.h: New files from GNULIB. + * yesno.c: Update from GNULIB. + +2004-10-21 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Move getpagesize to its own section. + * getpagesize.h: Update from GNULIB. + +2004-10-21 Mark Baushke <mdb@cvshome.org> + + * Makefile.in: Regenerated. + +2004-10-21 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Remove argmatch.c. + * argmatch.c: Remove this unused file. + +2004-10-21 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Move strftime to its own section. + * strftime.h: New file. + +2004-10-21 Derek Price <derek@ximbiot.com> + + * malloc.c: Update from GNULIB. + +2004-10-21 Mark Baushke <mdb@cvshome.org> + + * Makefile.am (all-local): Fix typo. + * Makefile.in: Regenerated. + +2004-10-21 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.mak: Regenerated for "../zlib/lib.dsp" change. + +2004-10-20 Mark D. Baushke <mdb@cvshome.org> + + * sunos57-select.c: New file. Work around Solaris 7 select() + hang. + * Makefile.in: Regenerate for new configure.in. + +2004-10-18 Derek Price <derek@ximbiot.com> + + * system.h: Define DEVNULL here when necessary. Don't include + unistd.h twice. + +2004-10-16 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Add "vasprintf.c" and "vasprintf.h" to project. + * libcvs.dep: Regenerate for "libcvs.dsp" change. + * libcvs.mak: Regenerate for "libcvs.dsp" change. + +2004-10-15 Derek Price <derek@ximbiot.com> + + * asprintf.c, vasprintf.c, vasprintf.h: New files from GNULIB. + +2004-10-15 Derek Price <derek@ximbiot.com> + + * getpass.c: Update from GNULIB. + +2004-10-09 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Add stat-macros.h. + * stat.c: Update from GNULIB. + * stat-macros.c: New file from GNULIB. + +2004-10-09 Derek Price <derek@ximbiot.com> + + * nanosleep.c: Update from GNULIB. + +2004-10-09 Derek Price <derek@ximbiot.com> + + * memmove.c: Update from GNULIB. + +2004-10-09 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Move minmax.h to its own section. + +2004-10-07 Derek Price <derek@ximbiot.com> + + * system.h: Reorganize slightly. + +2004-10-07 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Remove "xstrdup.c" from project. + Add "strcasecmp.c", "getopt_.h" and "getopt_int.h" to project. + * libcvs.dep: Regenerate for "libcvs.dsp" changes. + * libcvs.mak: Regenerate for "libcvs.dsp" changes. + +2004-10-06 Derek Price <derek@ximbiot.com> + + * system.h: Include getopt.h. + +2004-10-06 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Add getpass.h. + * getpass.h: New file from GNULIB. + * getpass.c: Update from GNULIB. + +2004-10-06 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Add strcase.h. + * strcase.h, strcasecmp.c, strncasecmp.c: New files from GNULIB. + +2004-10-06 Derek Price <derek@ximbiot.com> + + * gethostname.c: Update from gnulib. + +2004-10-06 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Move getline, getnline, & getndelim2 + sources to their own sections. + * getline.c, getndelim2.c, getndelim2.h, getnline.c, getnline.h: Update + from GNULIB. + +2004-10-06 Derek Price <derek@ximbiot.com> + + * .cvsignore: Add getopt.h. + +2004-10-06 Derek Price <derek@ximbiot.com> + + * Makefile.am: Add new configuration for getopt. + * getopt.h: Remove file. + * getopt_.h, getopt_int.h: New files from GNULIB. + * getopt.c, getopt1.c: Update from GNULIB. + +2004-10-06 Derek Price <derek@ximbiot.com> + + * mkstemp.c, tempname.c: Update from GNULIB. + +2004-10-06 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Remove xstrdup.c. + +2004-10-06 Derek Price <derek@ximbiot.com> + + * xstrdup.c: Remove this obsolete file. + +2004-10-05 Derek Price <derek@ximbiot.com> + + * Makefile.am (fnmatch, fnmatch-posix): Improve comment. + * fnmatch.c, fnmatch_loop.c: Update from GNULIB. + +2004-10-05 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Move exit.h to its own section. + +2004-10-05 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Remove "savecwd.c", "savecwd.h", "xgetwd.c" from project. + Add "save-cwd.c", "save-cwd.h", "xgetcwd.c", "xgetcwd.h" to project. + * libcvs.dep: Regenerated for "libcvs.dsp" change. + * libcvs.mak: Regenerated for "libcvs.dsp" change. + +2004-10-05 Derek Price <derek@ximbiot.com> + + * basename.c, dirname.c, dirname.h, stripslash.c: Updated from GNULIB. + * system.h: s/FILESYSTEM_PREFIX_LEN/FILE_SYSTEM_PREFIX_LEN/. + +2004-10-05 Derek Price <derek@ximbiot.com> + + * Makefile.gnulib: New file. + +2004-10-05 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Add xalloc-die.c. + * xalloc-die.c: New file from GNULIB. + * xalloc.h, xmalloc.c: Update from GNULIB. + +2004-10-05 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Move gettext.h to its own section. + * gettext.h, vasnprintf.c, vasnprintf.h: Update from GNULIB. + +2004-10-05 Derek Price <derek@ximbiot.com> + + * savecwd.c: Rename to... + * save-cwd.c: ...this and replace with the newest version from GNULIB. + * savecwd.h: Ditto, but to... + * save-cwd.h: ...this name. + * Makefile.am (libcvs_a_SOURCES): Remove savecwd.[ch]. Add save-cwd.h + & save-cwd.c in their own section. + +2004-10-05 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Remove xgetwd.c, add xgetcwd.h & + xgetcwd.c in their own section. + * xgetwd.c: Removed. + * xgetcwd.h, xgetcwd.c: New files imported from GNULIB. + +2004-10-05 Derek Price <derek@ximbiot.com> + + * alloca_.h, alloca.c: Import latest versions from GNULIB. + +2004-10-05 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Move pathmax.h to its own section. + +2004-10-05 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Move unlocked-io.h to its own + section. + * unlocked-io.h: Import most recent version from GNULIB. + +2004-10-05 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Move regex.h to its own section. + * regex.c: Import most recent version from GNULIB. + +2004-10-05 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.dsp: Add "dirname.c" to project to resolve link error. + * libcvs.dep: Regenerated for "libcvs.dsp" change. + * libcvs.mak: Regenerated for "libcvs.dsp" change. + +2004-10-05 Mark D. Baushke <mdb@cvshome.org> + + * regex.c (re_comp): Cast gettext return value to char * to + avoid warning in !ENABLE_NLS case. Patch imported from GNULIB. + (Problem report from Martin Neitzel <neitzel@sco.gaertner.de>.) + +2004-09-09 Conrad T. Pino <Conrad@Pino.com> + + * libcvs.mak: Regenerated for "../cvsnt.dsp" changes made 2004-09-08. + +2004-08-27 Derek Price <derek@ximbiot.com> + + Import minmax module from GNULIB. + + * Makefile.am (libcvs_A_SOURCES): Add minmax.h. + * minmax.h: New file. + +2004-07-13 Derek Price <derek@ximbiot.com> + + * .cvsignore: Ignore GCC profiling data. + +2004-06-24 Derek Price <derek@ximbiot.com> + + * libcvs.dsp: Add "./xsize.h" to Header file list. Add "./strftime.c" + & "./time_r.c" to Source file list. + * libcvs.dep, libcvs.mak: Regenerated for "./libcvs.dsp" change. + (Patch submitted by Conrad T. Pino <Conrad@Pino.com>.) + +2004-06-09 Derek Price <derek@ximbiot.com> + + * Makefile.am (xsize): Move GNULIB xsize module into its own section. + +2004-05-20 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Add strftime.c. + * strftime.c: New file. + +2004-05-17 Derek Price <derek@ximbiot.com> + + * libcvs.dsp: Add "lib/xtime.h" to project header files. + * libcvs.dep: Regnerated for "libcvs.dsp" file change. + (Patch from Conrad T. Pino <Conrad@Pino.com>.) + +2004-05-15 Derek Price <derek@ximbiot.com> + + Back out GNULIB getdate update until date parsing problem is fixed. + * Makefile.am: Temporarily comment out getdate.y test targets. + * getdate.y: Restored from version 1.19. + * xtime.h: Readded from version 1.1. + * Makefile.in, getdate.c: Regenerated. + +2004-05-15 Derek Price <derek@ximbiot.com> + + * libcvs.dsp: Move all "../lib/*.c" files to this project. Header file + list updated for GNULIB updates. + * libcvs.dep: Regenerated for "libcvs.dsp" changes. + * libcvs.mak: Regenerated for "libcvs.dsp" changes. + (Patch from Conrad T. Pino <Conrad@Pino.com>.) + +2004-05-14 Derek Price <derek@ximbiot.com> + + * libcvs.dsp: Add basename.c & gettime.c for Windows build. + (Patch from Conrad T. Pino <Conrad@Pino.com>.) + +2004-05-14 Derek Price <derek@ximbiot.com> + + * gettime.c, nanosleep.c: Updated from GNULIB. + +2004-05-03 Mark D. Baushke <mdb@cvshome.org> + + * Makefile.in (check-TESTS): Correct the script to deal with some + make (BSD) that use 'sh -e' which will terminate whenever a + command (such as test -n "") fails. + +2004-05-03 Derek Price <derek@ximbiot.com> + + * Makefile.am (getdate_LDADD): Add $(LIBINTL), needed on some systems. + +2004-05-03 Derek Price <derek@ximbiot.com> + + * Makefile.am (BUILT_SOURCES, MAINTAINERCLEANFILES): Remove getdate.c, + Automake determines this automatically. + * Makefile.in: Regenerated. + +2004-05-02 Derek Price <derek@ximbiot.com> + + * progname.c (progname): Assume = "getdate" for now. + +2004-05-02 Derek Price <derek@ximbiot.com> + + * Makefile.am (getdate_SOURCES): Add xmalloc, xstrdup, error, exitfail, + and progname for getdate test program on some systems. + * error.c, progname.c: New file. + (Report from Mark D. Baushke <mdb@cvshome.org>.) + +2004-04-29 Derek Price <derek@ximbiot.com> + + * getdate.y: Restore unforked version from GNULIB. + * unlocked-io.h: Update to most recent version from GNULIB. + * getdate.y: Regenerated. + +2004-04-28 Derek Price <derek@ximbiot.com> + + * .cvsignore: Add getdate executable. + * Makefile.am: Remove out-of-date comment about regex. Use Automake + comments where the comments won't look good in the Makefile. + (AM_CPPFLAGS): Remove obsolescent include. + (TESTS, MOSTLYCLEANFILES, check_PROGRAMS, getdate_*): Add support for + testing getdate. + * test-getdate.sh: New file. + * getdate.y: Move include to compile in test mode. + * Makefile.in, getdate.c: Regenerated. + +2004-04-28 Derek Price <derek@ximbiot.com> + + * system.h: Include timespec.h instead of xtime.h. + * xtime.h: Removed. + +2004-04-27 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Add getdate.h, getdate.y in their own + section. + (BUILT_SOURCES, MAINTAINERCLEANFILES): Add getdate.c. + * getdate.h, getdate.y: Import updated versions from GNULIB. + * Makefile.in: Regenerated. + +2004-04-27 Derek Price <derek@ximbiot.com> + + * mktime.c: New file. + * Makefile.in: Regenerated. + +2004-04-27 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Add time_r.h. + * time_r.h: New file. + * Makefile.in: Regenerated. + +2004-04-27 Derek Price <derek@ximbiot.com> + + * nanosleep.c: New file. + * Makefile.in: Regenerated. + +2004-04-27 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Add gettime.c. + * gettime.c: New file. + * Makefile.in: Regenerated. + +2004-04-27 Derek Price <derek@ximbiot.com> + + * gettimeofday.c: New file. + * Makefile.in: Regenerated. + +2004-04-27 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Add timespec.h. + * timespec.h: New file. + * Makefile.in: Regenerated. + +2004-04-27 Derek Price <derek@ximbiot.com> + + * system.h: Move definition of FILENAMES_CASE_INSENSITIVE to + windows-NT/config.h. + +2004-04-27 Derek Price <derek@ximbiot.com> + + Add dirname module from GNULIB. + + * Makefile.am (libcvs_SOURCES): Add new files. + * basename.c, dirname.c, dirname.h: New files. + * stripslash.c: Update from GNULIB. + * system.h (ISDIRSEP): Remove. + (ISABSOLUTE): s/ISDIRSEP/ISSLASH/. Use FILESYSTEM_PREFIX_LEN to search + for DOS drive spec rather than reimplementing. + * Makefile.in: Regenerated. + +2004-04-26 Derek Price <derek@ximbiot.com> + + * libcvs.dsp: Add stdbool.h from windows-NT as source dependency. + * libcvs.mak: Regenerated. + +2004-04-23 Derek Price <derek@ximbiot.com> + + * libcvs.dsp, libcvs.dep, libcvs.mak: Back out recent stdbool.h change. + +2004-04-23 Derek Price <derek@ximbiot.com> + + * .cvsignore: Add stdbool.h. + +2004-04-23 Derek Price <derek@ximbiot.com> + + * libcvs.dsp: Create stdbool.h. + * libcvs.dep, libcvs.mak: Regenerated. + +2004-04-22 Derek Price <derek@ximbiot.com> + + * system.h: #include <fnmatch.h>. + +2004-04-22 Derek Price <derek@ximbiot.com> + + * system.h: #include <stdbool.h>. + +2004-04-21 Derek Price <derek@ximbiot.com> + + Add stdbool module from GNULIB. + + * stdbool_.h: New file. + * Makefile.am (stdbool.h, BUILT_SOURCES, EXTRA_DIST, MOSTLYCLEANFILES): + Add stdbool cruft. + * Makefile.in: Regenerated. + +2004-04-20 Derek Price <derek@ximbiot.com> + + * system.h: Correct comments. + +2004-04-19 Derek Price <derek@ximbiot.com> + + * system.h: Gratuitous reformatting. + +2004-04-19 Derek Price <derek@ximbiot.com> + + * Makefile.am (EXTRA_DIST): s/lib\.\w+/libcvs.*/i. + * Makefile.in: Regenerated. + +2004-04-19 Derek Price <derek@ximbiot.com> + + * libcvs.mak: Regenerated for "zlib/libz.dsp" change. + (Patch from Conrad T. Pino <Conrad@Pino.com>.) + +2004-04-17 Derek Price <derek@ximbiot.com> + + * libcvs.dsp, libcvs.dep, libcvs.mak: Move... + * lib.dsp, lib.dep, lib.mak: ...here. + (Patch from Conrad T. Pino <Conrad@Pino.com>.) + +2004-04-17 Derek Price <derek@ximbiot.com> + + * LIB.dsp: Rename to... + * lib.dsp: ...this. + +2004-04-15 Derek Price <derek@ximbiot.com> + + * lib.dsp: Set PROP BASE directories to projet standard + has "Reset" function use project defaults. + Change "..\lib\" paths to ".\" for consistency. + * lib.mak: Regenerated for lib.dsp change. + (Patch from Conrad T. Pino <Conrad@Pino.com>.) + +2004-04-15 Derek Price <derek@ximbiot.com> + + * lib.dsp: Dropped "alloca.c" from project. + * lib.dep: Regenerated for "lib.dsp" change. + * lib.mak: Regenerated for "lib.dsp" change. + (Patch from Conrad T. Pino <Conrad@Pino.com>.) + +2004-04-15 Derek Price <derek@ximbiot.com> + + * Makefile.am (EXTRA_DIST): Add LIB.dep. + * Makefile.in: Regenerated. + +2004-04-15 Derek Price <derek@ximbiot.com> + + * lib.dsp: Update to compile fnmatch. + * LIB.dep: New generated file. + * lib.mak: Regenerated. + (Patch from Conrad T. Pino <conrad@pino.com>.) + +2004-04-14 Derek Price <derek@ximbiot.com> + + Update to current fnmatch module from GNULIB. + + * Makefile.am (BUILT_SOURCES, EXTRA_DIST): Add fnmatch sources. + (MOSTLYCLEANFILES): Initialize & add fnmatch & alloca headers. + (libcvs_a_OBJECTS): Make dependent on $(FNMATCH_H). + (fnmatch.h): New target. + * fnmatch.h.in: Rename to... + * fnmatch_.h: ...this. + * fnmatch.c: New version from GNULIB. + * fnmatch_loop.c: New file. + * Makefile.in: Regenerated. + +2004-04-14 Derek Price <derek@ximbiot.com> + + * Makefile.am (EXTRA_DIST): Remove lib.dep. + * Makefile.in: Regenerated. + +2004-04-07 Derek Price <derek@ximbiot.com> + + * regex.c: Revise "FREE_VAR" macro to eliminate C4090/C4022 warnings + in Windows build with Visual C++ 6.0 compiler. + (Original patch from Conrad T. Pino <Conrad@Pino.com>.) + +2004-04-07 Derek Price <derek@ximbiot.com> + + Update regex module from GNULIB. + + * Makefile.am (libcvs_a_SOURCES): Remove regex.c. + * regex.c, regex.h: Import new versions from GNULIB. + * Makefile.in: Regenerated. + +2004-04-04 Derek Price <derek@ximbiot.com> + + * system.h: Correct comment. + +2004-04-04 Derek Price <derek@ximbiot.com> + + * system.h: Restore complete path folding for Cygwin under Windows. + Add ISABSOLUTE macro for determining whether a path is absolute to + handle X:\ style paths under Windows (& Cygwin). + +2004-03-29 Derek Price <derek@ximbiot.com> + + * lib.mak: Regenerated with VC++ 5.0. + (Sent by Dennis Jones <djones@oregon.com>.) + * LIB.dep: Removed. + +2004-03-28 Derek Price <derek@ximbiot.com> + + * lib.mak: Regenerated. + +2004-03-27 Derek Price <derek@ximbiot.com> + + * LIB.dep, lib.mak: Regenerated. + +2004-03-26 Derek Price <derek@ximbiot.com> + + * Makefile.in: Regenerated. + +2004-03-26 Derek Price <derek@ximbiot.com> + + * LIB.dep, lib.dsp, lib.mak: New files. + * Makefile.am (EXTRA_DIST): Add new files. + * .cvsignore: Ignore MSVC build cruft. + +2004-03-25 Derek Price <derek@ximbiot.com> + + * system.h: No longer fold back slashes in paths into slashes. + +2004-03-25 Derek Price <derek@ximbiot.com> + + * realloc.c: Update to more recent version from GNULIB to fix a WOE32 + compilation problem. + +2004-03-20 Derek Price <derek@ximbiot.com> + + * mkdir.c (mkdir): Declare string args const. + +2004-03-13 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Add error.h. + * Makefile.in: Regenerated. + +2004-03-11 Derek Price <derek@ximbiot.com> + + * xsize.h: Move the default definition of SIZE_MAX into + windows-NT/config.h.in. + +2004-03-07 Derek Price <derek@ximbiot.com> + + * xsize.h: Provide a default definition of SIZE_MAX for Windows. + +2004-02-20 Derek Price <derek@ximbiot.com> + + Import xalloc module from GNULIB, as well as its remaining unimported + dependency, the exitfail module. + + * Makefile.am (libcvs_a_SOURCES): Add new sources. + * exitfail.c, exitfail.h, xalloc.h, xmalloc.c, xstrdup.c: New files. + * Makefile.in: Regenerated. + +2004-02-15 Derek Price <derek@ximbiot.com> + + Import vasnprintf module from GNULIB. + * asnprintf.c, printf-args.c, printf-args.h, printf-parse.c, + printf-parse.h, vasnprintf.c, vasnprintf.h: New files. + * Makefile.am (libcvs_a_SOURCES): Add header files. + +2004-02-15 Derek Price <derek@ximbiot.com> + + Import xsize module from GNULIB for vasnprintf(). + * xsize.h: New file. + * Makefile.am (libcvs_a_SOURCES): Add xsize.h. + +2004-02-15 Derek Price <derek@ximbiot.com> + + Import alloca module from GNULIB for vasnprintf(). + * alloca_.h, alloca.c: New files. + * .cvsignore: Ignore alloca.h. + * Makefile.am: Compile alloca() when necessary. + * Makefile.in: Regenerated. + +2004-02-15 Derek Price <derek@ximbiot.com> + + Make error() accessible to the GNULIB functions. + + * error.h: Move in from ../src. + +2004-02-15 Derek Price <derek@ximbiot.com> + + * strerror.c: Update to latest version from gnulib. + +2003-12-14 Mark D. Baushke <mdb@cvshome.org> + + * fseeko.c: Remove unnecessary #ifdef code. + * ftello.c: Ditto. + * system.h: Ditto. + +2003-12-10 Mark D. Baushke <mdb@cvshome.org> + + * ftello.c: Rework for AC_FUNC_FSEEKO. + * system.h: Rework fseeko and ftello prototypes. + +2003-12-10 Mark D. Baushke <mdb@cvshome.org> + + * xgetwd.c (pathmax.h): Add new include. + (Patch from Rob Clevenger <rob@robsite.org>.) + +2003-12-09 Derek Price <derek@ximbiot.com> + + * system.h: Correct spelling in comment. + +2003-12-09 Mark D. Baushke <mdb@cvshome.org> + + * fseeko.c (fseeko): New file and function to help with largefile + support. + * ftello.c (ftello): Ditto. + * system.h (ftello): Add prototype if !defined(HAVE_FTELLO). + * Makefile.in: Regenerated. + +2003-12-09 Larry Jones <lawrence.jones@eds.com> + + * getdate.c: Regenerated. + +2003-12-08 Mark D. Baushke <mdb@cvshome.org> + + * getdate.y: Remove #include <stdio.h> to avoid problems compiling + AIX with '#define _LARGE_FILES 1' with Redeclaration of fgetpos64 + and ftello64. This may be an autoconf problem with + AC_SYS_LARGEFILE. + * getdate.c: Regenerated. + +2003-12-03 Derek Price <derek@ximbiot.com> + + * fncase.c (OSX_filename_classes): New array. + (fncmp): Use FOLD_FN_CASE rather relying on the fact that it will be + #defined to use WNT_filename_classes. + * system.h: Define FOLD_FN_CASE, fncmp, and fnfold for all case + insensitive filesystems. Share some code between the new generic case + insensitive section and the old WOE32 section. + +2003-11-25 Mark D. Baushke <mdb@cvshome.org> + + * Makefile.in: Regenerate for new configure.in. + +2003-10-27 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Remove obsolete PROTO.h. + * Makefile.in: Regenerated. + +2003-10-22 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_LIBADD): Undo the previous change and move it + into the program linking step in `../src'. + * Makefile.in: Regenerated. + +2003-10-21 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_LIBADD): Add $(LIBINTL) for gettext. + * Makefile.in: Regenerated. + +2003-10-04 Derek Price <derek@ximbiot.com> + + * sighandle.c (SIG_handle): With the last commit, avoid allowing a + second interrupt while this interrupt handler runs. + (SIG_beginCrSect, SIG_endCrSect): With the last change, move these two + functions to the top of this file to avoid prototyping them for + SIG_handle(). + (*): C89ify some prototypes. + +2003-10-04 Derek Price <derek@ximbiot.com> + + * system.h (EXIT_FAILURE): Don't define here since it's defined in + exit.h. + +2003-10-01 Derek Price <derek@ximbiot.com> + + * getpass.c: Update to new version from GNULIB with Larry's fix + incorporated. + +2003-10-01 Derek Price <derek@ximbiot.com> + + * getopt.h, getopt.c, getopt1.c: Update from GNULIB. + +2003-10-01 Derek Price <derek@ximbiot.com> + + * getline.c, getndelim2.c: Merge changes from GNULIB. + +2003-10-01 Derek Price <derek@ximbiot.com> + + * system.h: Assume <string.h> and <sys/types.h> per the notes in + HACKING. + +2003-09-30 Larry Jones <lawrence.jones@eds.com> + + * getpass.c: Fix bug that caused password to be echoed on many + systems (input may not be followed by output on the same stream + without an intervening call to a file positioning function). + (Reported by David Everly <david.everly@mci.com>.) + +2003-09-30 Derek Price <derek@ximbiot.com> + + Provide an atexit() function on systems which provide on_exit() but not + atexit(). + + * atexit.c: New file. + * Makefile.in: Regenerated. + +2003-08-19 Derek Price <derek@ximbiot.com> + + * system.h: Assume more headers, re C89 & GNULIB. + +2003-08-12 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Remove getndelim2 - it is now in + LIBOBJ. + +2003-07-31 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Remove getpass.c per new getpass-gnu + API from GNULIB. + * Makefile: Regenerated. + +2003-07-29 Derek Price <derek@ximbiot.com> + + * getpass.c: New file, almost identical to GNULIB's currect version. + * Makefile.am (libcvs_a_SOURCES): Add getpass.c. + * Makefile.in: Regenerated. + +2003-07-25 Derek Price <derek@ximbiot.com> + + * lstat.c: Sync with GNULIB. + * stat.c: Almost sync'd with GNULIB. + +2003-07-25 Derek Price <derek@ximbiot.com> + + * getline.h: Sync this file with GNULIB. + +2003-07-24 Derek Price <derek@ximbiot.com> + + * getline.c, getline.h, getndelim2.c, getndelim2.h, getnline.c, + getnline.h: Merge some more changes from GNULIB. + +2003-07-24 Derek Price <derek@ximbiot.com> + + * gethostname.c: Update from GNULIB. + +2003-07-23 Derek Price <derek@ximbiot.com> + + * system.h: Move some includes from src/cvs.h. + +2003-07-22 Derek Price <derek@ximbiot.com> + + * system.h: Assume <stddef.h> per C89. + +2003-07-22 Derek Price <derek@ximbiot.com> + + * strerror.c: Import current version from GNULIB. + * Makefile.in: Regenerated. + +2003-07-22 Derek Price <derek@ximbiot.com> + + * Makefile.am: Make a comment more specific about LIBOBJ, while + removing a list of LIBOBJ'd file maintained elsewhere. + * memmove.c: Import current version from GNULIB. + +2003-07-22 Derek Price <derek@ximbiot.com> + + * exit.h: Really add this file this time. + * Makefile.in: Regenerated. + +2003-07-22 Derek Price <derek@ximbiot.com> + + * exit.h: New file from GNULIB. + * Makefile.am (lib_SOURCES): Add exit.h. + +2003-07-19 Derek Price <derek@ximbiot.com> + + * stat.c: Sync us with GNULIB, except for the xalloc.h requirement. + xalloc.h is waiting on $GNULIB/modules/error for import. + +2003-07-19 Derek Price <derek@ximbiot.com> + + * Makefile.am: Add getndelim.c & getnline.c as part of the new API. + * getline.c: Merge more of Bruno Habile's GNULIB changes to this file + so our fork is less divergent. + * getline.h: Ditto. + * getndelim2.c: Ditto. + * getndelim2.h: Ditto. + * getnline.c: Ditto. + * getnline.h: Ditto. + + * Makefile.in: Regenerated. + +2003-07-19 Derek Price <derek@ximbiot.com> + + * savecwd.c: Move the MD5 stuff that shifted here a few commits back... + * md5.c: ...back to here. + +2003-07-19 Derek Price <derek@ximbiot.com> + + * getndelim2.c: Make limit an ssize_t rather than int, as per the + return type. + * getndelim2.h: Ditto. + * getnline.c: Ditto. + * getnline.h: Ditto. + +2003-07-19 Derek Price <derek@ximbiot.com> + + * getline.c (getdelim): Return ssize_t, as per getline.h. + +2003-07-17 Derek Price <derek@ximbiot.com> + + * getline.c: Merge some of Bruno Habile's changes from GNULIB including + some he didn't quite finish, like renaming of n => linesize. + * getline.h: Ditto. + * getndelim2.c: Ditto. + * getndelim2.h: Ditto. + * getnline.c: Ditto. + * getnline.h: Ditto. + +2003-07-17 Derek Price <derek@ximbiot.com> + + * getndelim2.c: New file. + * getndelim2.h: Ditto. + + * Makefile.am (libcvs_a_SOURCES): Add getndelim2.h. + * getline.h: Return ssize_t. + * getnline.h: Ditto. + * getline.c: Ditto and depend on getndelim2. + * getnline.c: Ditto. + + * Makefile.in: Regenerated. + +2003-07-16 Derek Price <derek@ximbiot.com> + + * getline.c: Don't include the unneeded unlocked-io.h. + * getnline.c: Remove extra blank line. Use GETNDELIM_NO_LIMIT rather + than GETNLINE_NO_LIMIT. + * getnline.h: Define GETNDELIM_NO_LIMIT rather than GETNLINE_NO_LIMIT. + +2003-07-16 Derek Price <derek@ximbiot.com> + + * PROTO.h: Remove this file since some of our GNULIB sources no longer + support compilers that can't handle prototypes. + + * md5.c: s/PROTO//. + * md5.h: Ditto. + * savecwd.c: Ditto. + * savecwd.h: Ditto. + * sighandle.c: Ditto. + +2003-07-16 Derek Price <derek@ximbiot.com> + + * getline.h: Include config.h so to use the macros it defines. + (Thanks to Steve McIntyre <stevem@cvshome.org> for the report.) + +2003-07-16 Derek Price <derek@ximbiot.com> + + * pathmax.h: New file. + + * Makefile.am (libcvs_a_SOURCES): Add pathmax.h. + * system.h: Remove PATH_MAX stuff. + + * Makefile.in: Regenerated. + +2003-07-16 Derek Price <derek@ximbiot.com> + + * getnline.c: New file based on GNULIB and resubmitted to GNULIB. + * getnline.h: Ditto. + + * Makefile.am (libcvs_a_SOURCES): Add new getnline.h. + * getline.c: Import and hack version from GNULIB to prevent double + prototype problem on 64-bit machines. Resubmitted to GNULIB. + * getline.h: Ditto. + (Thanks to Steve McIntyre <stevem@cvshome.org> for the report.) + + * Makefile.in: Regenerated. + +2003-07-15 Derek Price <derek@ximbiot.com> + + * unlocked-io.h: s/LGPL/GPL/g, imported from GNULIB. + +2003-06-13 Derek Price <derek@ximbiot.com> + + * mkstemp.c: New file. + * tempname.c: Ditto. + + * Makefile.in: Regenerated. + +2003-06-11 Derek Price <derek@ximbiot.com> + + * Makefile.am (libcvs_a_SOURCES): Add gettext.h. + * system.h: Include gettext.h. + + * gettext.h: New file from GNULIB. + + * Makefile.in: Regenerated. + +2003-06-09 Derek Price <derek@ximbiot.com> + + * system.h: Reference the WIN32 macro only in order to define WOE32, + in accordance with the GNU convention to avoid implying that we + consider the Microsoft Windows Operating Environment any sort of "win". + +2003-05-29 Derek Price <derek@ximbiot.com> + + * Makefile.am: xgssapi.h is no more. + * xgssapi.h: Move its contents to src/gssapi-client.h + (Patch from Alexey Mahotkin <alexm@hsys.msk.ru>.) + + * Makefile.in: Regenerated. + +2003-05-28 Derek Price <derek@ximbiot.com> + + * getline.h, md5.c, sighandle.c, savecwd.h: Use PROTO.h. + * savecwd.c: Use standard PROTO macro. + * Makefile.am: Add PROTO.h + * PROTO.h: PROTO macro definition is finally here, as specified in + Autoconf documentation. + (Original patch from Alexey Mahotkin <alexm@hsys.msk.ru>.) + + * Makefile.in: Regenerated. + +2003-05-28 Derek Price <derek@ximbiot.com> + + * Makefile.am: get rid of $(includeopt); -Isrc/ is ok, so remove + lengthy explanation. + (Patch from Alexey Mahotkin <alexm@hsys.msk.ru>.) + + * Makefile.in: Regenerated. + +2003-05-27 Derek Price <derek@ximbiot.com> + + * README: New file explaining the GNULIB origin of many of the lib + files. + +2003-05-21 Derek Price <derek@ximbiot.com> + + * Makefile.in: Regenerate with Automake version 1.7.5. + +2003-05-21 Derek Price <derek@ximbiot.com> + + * unlocked-io.h: New file from GNULIB. + * system.h: Include unlocked-io.h. + * Makefile.am (libcvs_a_SOURCES): Add unlocked-io.h. + + * Makefile.in: Regenerated. + +2003-05-21 Derek Price <derek@ximbiot.com> + + * realloc.c: New file from GNULIB. + * system.h (HAVE_REALLOC, CVS_REALLOC): Add support for realloc.c. + +2003-05-20 Derek Price <derek@ximbiot.com> + + * stat.c: Don't compile LSTAT code unless configure detected that we + need to be trailing-slash-safe. + +2003-05-20 Derek Price <derek@ximbiot.com> + + * Makefile.in: Regenerated. + +2003-05-20 Derek Price <derek@ximbiot.com> + + * system.h: Move in the standard includes from cvs.h. Define + CVS_MALLOC & CVS_REALLOC. + +2003-05-19 Derek Price <derek@ximbiot.com> + + * malloc.c: New file from <https://savannah.gnu.org/projects/gnulib>. + +2003-05-19 Derek Price <derek@ximbiot.com> + + * lstat.c: New file from <https://savannah.gnu.org/projects/gnulib>. + * stat.c: Ditto. + * system.h: Define CVS_STAT and CVS_LSTAT properly when the system + versions of stat and lstat are broken. + +2003-05-19 Derek Price <derek@ximbiot.com> + + * .cvsignore: Add fnmatch.h for Windows. + +2003-05-09 Derek Price <derek@ximbiot.com> + + * xgssapi.h: #include config.h. + (Reported by Boyd Lynn Gerber <gerberb@zenez.com>.) + +2003-05-09 Derek Price <derek@ximbiot.com> + + * system.h: Define S_ISSOCK on SCO OpenServer. + +2003-05-09 Derek Price <derek@ximbiot.com> + + * Makefile.in: Regenerated. + +2003-04-30 Derek Price <derek@ximbiot.com> + + * Makefile.in: Regenerated. + +2003-04-10 Larry Jones <lawrence.jones@eds.com> + + * Makefile.in: Regenerated. + +2003-04-03 Derek Price <derek@ximbiot.com> + + * Makefile.am (distclean-local): New target to remove fnmatch.h when + necessary. This should be handled by Automake, but until then... + (Resolves issue #100 + <http://ccvs.cvshome.org/issues/show_bug.cgi?id=100> from + Serguei E. Leontiev <Serge3lse@cvshome.org>.) + + * Makefile.in: Regenerated. + +2003-03-24 Derek Price <derek@ximbiot.com> + + * Makefile.am: Update copyright notice. + + * Makefile.in: Regenerated. + +2003-03-19 Mark D. Baushke <mdb@cvshome.org> + + * Makefile.in: Regenerated. + +2003-03-19 Derek Price <derek@ximbiot.com> + + * Makefile.in: Regenerated. + +2003-02-25 Derek Price <derek@ximbiot.com> + + * Makefile.in: Regenerated. + +2003-01-23 Larry Jones <lawrence.jones@eds.com> + + * getdate.y: Add RCS/CVS timestamp format (Y.mm.dd.hh.mm.ss). + * getdate.c: Regenerated. + + * wait.h (WCOREDUMP): New macro. + +2002-12-27 Derek Price <derek@ximbiot.com> + + * getdate.c: Regenerated with Bison 1.35. + +2002-11-04 Derek Price <derek@ximbiot.com> + + * getdate.y (Convert): Add comment as to the effectiveness of + descriptive error messages. + +2002-09-24 Derek Price <derek@ximbiot.com> + + * Makefile.in: Regenerated using Automake 1.6.3. + +2002-09-24 Derek Price <derek@ximbiot.com> + + * Makefile.in: Regenerated. + +2002-09-15 Larry Jones <lawrence.jones@eds.com> + + * system.h: Add FOPEN_BINARY_READWRITE. + (Patch submitted by Josh Lehan <cvs@krellan.com>.) + +2002-08-12 Derek Price <oberon@umich.edu> + + * Makefile.am: Remove obsolete reference to `ftruncate.c'. + (Symptoms reported by + Andrey Aristarkhov <Aristarkhov@bitechnology.ru>.) + * Makefile.in: Regenerated. + +2002-08-08 Derek Price <oberon@umich.edu> + + * regex.c: Removed unused `compile_range' declaration. + (Patch from John Tytgat <John.Tytgat@aaug.net>.) + +2002-05-09 Larry Jones <lawrence.jones@eds.com> + + * getline.c (getstr): Make terminator int instead of char to avoid + promotion problems. + * getline.h (getstr): Change to match. + +2002-05-08 Derek Price <oberon@umich.edu> + + * Makefile.in: Regenerated. + * fnmatch.h: Move this file... + * fnmatch.h.in: here. + +2002-05-08 Derek Price <oberon@umich.edu> + + * strerror.c: Use HAVE_CONFIG_H and put config.h in brackets rather + than quotes. + +2002-05-02 Derek Price <oberon@umich.edu> + + * fnmatch.h: More #defines to avoid Mac OS X namespace conflicts. + +2002-04-30 Derek Price <oberon@umich.edu> + + * hostname.c: Rename to... + * gethostname.c: this. + * Makefile.am: Change comment to reflect above. + + * Makefile.in: Regenerated with automake 1.6. + +2002-04-28 Derek Price <oberon@umich.edu> + + * getopt.h: #define new names for functions and variables when they + might conflict with system definitions (namely on Mac OS X 10.1 with + the most recent dev packages - This should be removable after the Mac + dev packages are fixed.). + * regex.h: Ditto. + * Makefile.am (libcvs_a_SOURCES): Remove fnmatch.h. + +2002-04-20 Larry Jones <larry.jones@sdrc.com> + + * Makefile.am (libcvs_a_SOURCES): Add getpagesize.h. + * Makefile.in: Regenerated. + +2001-09-18 Derek Price <dprice@collab.net> + + * fnmatch.c: The header file for a system function we're replacing with + our own should be #included using double quotes. + (Patch from Corey Minyard <minyard@acm.org> via + Alexey Mahotkin <alexm@hsys.msk.ru>.) + +2001-09-04 Derek Price <dprice@collab.net> + + * Makefile.in: Regenerated with automake 1.5. + +2001-08-09 Derek Price <dprice@collab.net> + + * getpagesize.h: Only include sys/param.h when HAVE_SYS_PARAM_H has + been defined by configure. + +2001-08-07 Derek Price <dprice@collab.net> + + * build_lib.com: Verify. + * getdate.y: Move the include of xtime.h out from underneath the ifdef + so that it is always included. + (Patch from Mike Marciniszyn <Mike.Marciniszyn@sanchez.com>.) + + * getdate.c: Regenerated. + +2001-08-06 Derek Price <dprice@collab.net> + + * Makefile.in: Regenerated. + +2001-07-05 Larry Jones <larry.jones@sdrc.com> + + * getpagesize.h: New file to define getpagesize() for systems that + don't already have it. + * valloc.c (valloc): Use it. + +2001-07-04 Derek Price <dprice@collab.net> + + * Makefile.in: Regenerated with new Automake release candidate 1.4h. + +2001-06-28 Derek Price <dprice@collab.net> + + * Makefile.in: Regenerated with new version of Automake. + +2001-06-15 Derek Price <dprice@collab.net> + + * xselect.h: Don't include xtime.h. + (Thanks to Martin Neitzel <neitzel@sco.gaertner.de>.) + +2001-04-25 Derek Price <dprice@collab.net> + + * Makefile.in: Regenerated using AM 1.4e as of today at 18:10 -0400. + +2001-04-02 Derek Price <dprice@collab.net> + for Alon Ziv <alonz@zapper.com> + + * getdate.y: Add a declaration for yyparse(). + + * getdate.c: Regenerated. + +2001-03-14 Derek Price <derek.price@openavenue.com> + + * Makefile.in: Regenerated + +2001-02-20 Derek Price <derek.price@openavenue.com> + + * xgssapi.h: New file to perform GSSAPI include magic. + * Makefile.am (EXTRA_DIST): Add xgssapi.h. + + * Makefile.in: Regenerated. + +2001-02-14 Derek Price <derek.price@openavenue.com> + + * Makefile.am (libcvs_a_SOURCES): Add xtime.h & xselect.h. + * Makefile.in: Regenerated. + * getdate.y: Include xtime.h. + * getdate.c: Regenerated. + * system.h: Include xtime.h. + * xtime.h: New file to do include magic for time functions. + * xselect.h: New file to do select include magic. + +2001-02-06 Derek Price <derek.price@openavenue.com> + Rex Jolliff <Rex_Jolliff@notes.ymp.gov> + Shawn Smith <Shawn_Smith@notes.ymp.gov> + + * system.h: definitions of CVS_OPENDIR, CVS_READDIR, & CVS_CLOSEDIR + provided here in support of changes to handle VMS DEC C 5.7 + {open,read,close}dir problems. Check today's entry in the vms subdir + for more. + +2001-01-10 Derek Price <derek.price@openavenue.com> + Rex Jolliff <Rex_Jolliff@notes.ymp.gov> + + * rename.c: replace calls to unlink() with CVS_UNLINK() for VMS + +2000-12-22 Derek Price <derek.price@openavenue.com> + + * Makefile.am (INCLUDES): Fixed typo + * Makefile.in: Regenerated + +2000-12-22 Derek Price <derek.price@openavenue.com> + + * Makefile.am (INCLUDES): Added $(top_srcdir)/src + * Makefile.in: Regenerated + +2000-12-21 Derek Price <derek.price@openavenue.com> + + * .cvsignore: Added .deps directory and alphabetized + * Makefile.am: New file needed by Automake + * Makefile.in: Regenerated + +2000-11-15 Derek Price <derek.price@openavenue.com> + + * system.h: Added CVS_FDOPEN to conform to CVS_FOPEN precedent + +2000-07-10 Larry Jones <larry.jones@sdrc.com> + + * savecwd.c: #include <sys/types.h> before <fcntl.h>. + +2000-07-04 Karl Fogel <kfogel@red-bean.com> + + * getline.h, getline.c (getstr): take new limit arg. + (GETLINE_NO_LIMIT): new #define. + (getline_safe): new function, takes limit arg and passes it on. + (getline): pass GETLINE_NO_LIMIT to getstr(). + + See related change of same date in ../src/ChangeLog. + +2000-06-19 Larry Jones <larry.jones@sdrc.com> + + * regex.c, regex.h: Version from emacs 20.7 to plug memory leaks + and avoid potential portability problems. + +2000-03-22 Larry Jones <larry.jones@sdrc.com> + + * getdate.y: Add logic to allow yyyy/mm/dd in addition to mm/dd/yy + (since that is the format CVS frequently uses). + * getdate.c: Regenerated. + +2000-02-16 Jim Meyering <meyering@lucent.com> + + * sighandle.c (SIG_inCrSect): New function. + +2000-01-03 Larry Jones <larry.jones@sdrc.com> + + * getdate.y (Convert): Add window to determine whether 2-digit dates + are 19xx (69-99) or 20xx (00-68). + (get_date): Fix y2k bug: initialize yyYear to tm_year + 1900, + not just tm_year. + * getdate.c: Regenerated. + +1999-12-29 Jim Kingdon <http://developer.redhat.com/> + + * Makefile.in: There was a comment here which referred to a long + comment in configure.in about regex.o (the configure.in comment + isn't there any more). Replace our comment with a conciser + version of the former configure.in comment. + +1999-03-26 Jim Kingdon <http://www.cyclic.com> + + * getopt.h: Don't declare the arguments to getopt. + +1999-02-09 Jim Kingdon <http://www.cyclic.com> + + * vasprintf.c: Removed; there is apparently no clean, portable + solution to the VA_LIST_IS_ARRAY problem (C9X drafts have va_copy, + but we aren't even assuming C90 yet!). + * Makefile.in (SOURCES): Remove vasprintf.c. + * build_lib.com: Remove vasprintf.c and vasprintf.obj. + +1999-01-26 Jim Kingdon <http://www.cyclic.com> + and Joerg Bullmann <http://www.glink.net.hk/~jb/MacCVSClient/> + + * fnmatch.c: Use FOLD_FN_CHAR in two cases where it had been + omitted. + +1999-01-22 Jim Kingdon <http://www.cyclic.com> + + * fnmatch.c: Include system.h; FOLD_FN_CHAR has moved there from + config.h (from Alexey Milov). Don't define our own FOLD_FN_CHAR; + that just masks cases in which we got the includes tangled up. + +1999-01-12 Jim Kingdon <http://www.cyclic.com> + + * memmove.c: Remove paragraph which contained the FSF's old + snail mail address; it has changed. + +1999-01-05 Jim Kingdon <http://www.cyclic.com> + + * md5.c, md5.h: Rename all the external interfaces to start with + cvs_* to avoid namespace pollution problems. Include string.h + unconditionally, to avoid gcc -Wall warnings on memset. + +1998-12-29 Jim Kingdon <http://www.cyclic.com> + + * getdate.y (RelativeMonth): Add 1900 to tm_year, so that in 2000, + we pass 2000, not 100, to Convert. + (Convert): Add comment about Year argument. + * getdate.c: Regenerated using byacc. + +Tue Mar 24 16:08:00 1998 Ian Lance Taylor <ian@cygnus.com> + + * Makefile.in (CFLAGS): Set to @CFLAGS@, not -g. + +1998-02-20 Jim Kingdon <kingdon@harvey.cyclic.com> + + * regex.c: Partial merge with version from emacs 20.2. Brings + over some trivial changes (whitespace and so on) (most such + changes I didn't bother with, for this time). Don't cast to int + before comparing old_regend[r] to regstart[r] (this is the point + of bothering; the old code was broken for 64 bit machines. + Reported by Paul Vixie). + +Tue Feb 17 18:33:26 1998 Ian Lance Taylor <ian@cygnus.com> + + * memmove.c: New file, resurrecting the old one. + * Makefile.in (SOURCES): Add memmove.c. + +1998-02-03 Tim Pierce <twp@skepsis.com> + + * system.h (CVS_LSTAT): New macro. + +Sat Feb 7 17:33:39 1998 Ian Lance Taylor <ian@cygnus.com> + + * getline.h (getstr): Declare. + +13 Jan 1998 Jim Kingdon + + * fncase.c: Include config.h before system.h. + + * system.h: Just include string.h unconditionally. We already + include it unconditionally elsewhere. + +Tue Jan 13 16:51:59 1998 Ian Lance Taylor <ian@cygnus.com> + + * fncase.c: New file, taken from windows-NT/filesubr.c. + * system.h: If __CYGWIN32__ or WIN32 are defined, define + FOLD_FN_CHAR, FILENAMES_CASE_INSENSITIVE, and ISDIRSEP, and + declare fncmp and fnfold. Taken from windows-NT/config.h. + * Makefile.in (SOURCES): Add fncase.c. + +Sat Jan 10 10:51:26 1998 Jim Kingdon <kingdon@harvey.cyclic.com> + + * getline.c (getstr): Make sure to set errno when appropriate. I + didn't test the error case for the new code but inspection shows + the old code was rather broken. + +Sat Nov 29 22:03:39 1997 Jim Kingdon <kingdon@harvey.cyclic.com> + + getwd and getcwd were a big big mess. Although Jim's fix might + indeed be fixing a typo, the code is so tangled that I would guess + it probably breaks some system. So clean this up: + * xgetwd.c: Always assume we have getcwd (we had been anyway, + before Jim's change). + * getwd.c: Removed. + * Makefile.in: Remove getwd.c + * system.h: Remove declarations of getwd and getcwd. Move getcwd + declaration to the !HAVE_UNISTD_H section. + +1997-11-29 Jim Meyering <meyering@na-net.ornl.gov> + + * xgetwd.c: Fix typo s/ifndef/ifdef/ in test of HAVE_GETWD. + +Wed Nov 26 10:12:33 1997 Jim Kingdon <kingdon@harvey.cyclic.com> + + * system.h: Always use "rb" and "wb". Check for O_BINARY with an + #ifdef, not the error-prone LINES_CRLF_TERMINATED. + +Thu Sep 25 10:57:39 1997 Jim Kingdon <kingdon@harvey.cyclic.com> + + * getdate.y (get_date): If gmtime returns NULL, try to cope. + * getdate.c: Regenerated using byacc. + + * getdate.y: Remove comment about sending email concerning this file + to Rich Salz. + * getdate.c: Regenerated using byacc. + +Wed Sep 24 10:35:38 1997 Jim Kingdon <kingdon@harvey.cyclic.com> + + * Makefile.in (OBJECTS): Add regex.o. + +Wed Sep 17 16:37:17 1997 Jim Kingdon <kingdon@harvey.cyclic.com> + + * getdate.y (ToSeconds): For am or pm, a hour of "12" really means 0. + * getdate.c: Regenerated using byacc (not bison per comment). + +Tue Sep 9 20:51:45 1997 Jim Kingdon <kingdon@harvey.cyclic.com> + + * build_lib.com: Add vasprintf.c and vasprintf.obj. + + * build_lib.com: Remove strippath.obj from library/create command. + +Sun Sep 7 17:35:27 1997 Jim Kingdon <kingdon@harvey.cyclic.com> + + * system.h: Replace comment referring to ChangeLog with a + comment based on the ChangeLog entries. + + * strdup.c: Removed, per change to ../configure.in + * Makefile.in (SOURCES): Remove strdup.c. + +Mon Jun 16 18:59:50 1997 Jim Kingdon <kingdon@harvey.cyclic.com> + + * system.h: Add CVS_FNMATCH. + +Sun Jun 8 23:41:11 1997 Jim Kingdon <kingdon@harvey.cyclic.com> + + * system.h (mkfifo): Remove; not used anywhere. + +Thu Mar 6 17:14:49 1997 Jim Kingdon <kingdon@harvey.cyclic.com> + + * regex.c: Partial merge with version from emacs 19.34. I brought + over most trivial changes (whitespace and so on). Most of the + changes to portability cruft I did not bring over, on the theory + of sticking to the devil that we know. I did bring over the + change to undef MAX and MIN (this is a better solution to a + problem we had been handling a different way). There were a + variety of changes I probably could/should have brought over, but + elected not to try to understand them and whether they would cause + trouble (printchar -> putchar, changes to output format in + print_partial_compiled_pattern, internationalization, + FREE_STACK_RETURN and friends which would appear to be fixing + memory leaks in error cases, RE_TRANSLATE_TYPE, and others). I + did merge the changes (union fail_stack_elt, PUSH_FAILURE_POINTER, + etc.) to use a union for the failure stack rather than playing + games with pointers and integers (that was my reason for + bothering; the code had been broken on the Alpha). + +Mon Feb 10 18:52:18 1997 Ullrich von Bassewitz <uz@musoftware.com> + + * md5.c: Make the parameter to getu32 const since the function will + only read the values and this will avoid compiler warnings in other + places. + +Mon Feb 10 18:29:04 1997 Ullrich von Bassewitz <uz@musoftware.com> + + * vasprintf.c: Added a #define for systems where a va_list is + defined as an array, not as a pointer. + +Mon Feb 10 09:31:38 1997 Ken Raeburn <raeburn@cygnus.com> + + * md5.c (MD5STEP): Truncate to 32 bits before shifting right. + +Thu Jan 30 11:35:26 1997 Jim Kingdon <kingdon@harvey.cyclic.com> + + * regex.h: Don't prototype re_comp and re_exec. + +Tue Jan 28 17:45:46 1997 Jim Kingdon <kingdon@harvey.cyclic.com> + + * md5.c, md5.h: Changes so these work without having an integer + type which is exactly 32 bits. Modeled after changes by Tatu Ylonen + <ylo@cs.hut.fi> as part of SSH but rewritten. + +Wed Jan 8 14:50:47 1997 Jim Kingdon <kingdon@harvey.cyclic.com> + + * Makefile.in, getopt.h, sighandle.c, system.h: Remove CVSid; we + decided to get rid of these some time ago. + +Thu Jan 2 13:30:56 1997 Jim Kingdon <kingdon@harvey.cyclic.com> + + * Makefile.in, argmatch.c, fnmatch.c, fnmatch.h, getline.c, + getopt.c, getopt.h, getopt1.c, getwd.c, hostname.c, mkdir.c, + regex.c, regex.h, rename.c, sighandle.c, strdup.c, strerror.c, + stripslash.c, system.h, vasprintf.c, wait.h, xgetwd.c, yesno.c: + Remove "675" paragraph; see ../ChangeLog for rationale. + +Sun Nov 24 13:34:25 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * getdate.y (Convert): Change last acceptable year from 1999 to + 2038. + * getdate.c: Regenerated using byacc 1.9. + +Tue Nov 19 17:11:17 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * Makefile.in (OBJECTS): Remove strippath.o; we don't use + strip_path anymore. + (SOURCES): Remove strippath.c. + * strippath.c: Removed. + * build_lib.com: Remove strippath.c. + +Wed Oct 2 10:43:35 1996 Norbert Kiesel <nk@col.sw-ley.de> + + * getdate.y: removed CVSid variable + + * getdate.c: regenerated (using byacc 1.9) + +Wed Sep 25 10:25:00 1996 Larry Jones <larry.jones@sdrc.com> + + * vasprintf.c: Fix type clashes in calls to strtoul. + +Wed Sep 11 15:55:31 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * build_lib.com: Add valloc.c. + +Tue Sep 10 23:04:34 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * Makefile.in (DISTFILES): Add build_lib.com. + +Fri Aug 16 16:01:57 1996 Norbert Kiesel <nk@col.sw-ley.de> + + * Makefile.in (installdirs): new (empty) target + +Mon Aug 12 11:03:43 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * system.h: Don't use #elif. It is said to cause problems with + one of the HP compilers on HPUX 9.01. + +Sun Jul 7 23:25:46 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * memmove.c: Removed. The memove function was used by a very old + version of the CVS server for nefarious purposes and it has been + long gone. + * Makefile.in (SOURCES): Remove memmove.c. + +Thu Jun 6 15:12:59 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * vasprintf.c: If STDC_HEADERS, include stdlib.h rather than + declaring its functions ourself. + +Wed Jun 05 10:14:29 1996 Mike Ladwig <mike@twinpeaks.prc.com> + and Jim Kingdon <kingdon@cyclic.com> + + * system.h: If ERRNO_H_MISSING is defined, don't include errno.h. + +Wed Jun 05 10:14:29 1996 Mike Ladwig <mike@twinpeaks.prc.com> + + * regex.c: Don't define MAX and MIN if already defined. + +Sun May 12 09:40:08 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * getdate.y: Replace alloca.h include with a comment explaining + why we avoid alloca and the consequences of that. + * getdate.c: Regenerated. + +Wed May 8 09:31:03 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * getdate.c: Regenerate with the version of byacc in Red Hat 3.0.3 + (which I believe is byacc 1.9). byacc, unlike bison, does not + require alloca in the generated parser. + +Thu Apr 25 18:26:34 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * getdate.y (get_date): Set Start from nowtime, not now->time, + which may not be set. + * getdate.c: Regenerated. + +Wed Apr 10 17:55:02 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * getdate.y (get_date): Use a time_t variable rather than a field + in a struct timeb. Works around Solaris compiler bug. Sure, it + is a compiler bug, but the workaround is completely painless. + * getdate.c: Regenerated. + +Fri Mar 22 11:17:05 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * system.h: If EXIT_FAILURE is not defined by stdlib.h, define it + ourself. + +Thu Mar 14 16:27:53 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * system.h: Remove alloca cruft. + +Wed Feb 28 03:16:48 1996 Benjamin J. Lee <benjamin@cyclic.com> + + * build_lib.com: Changed definition of symbol CC to search + for include files in [-.VMS] so VMS config.h can be picked + up without copying. + +Tue Feb 27 21:26:34 1996 Benjamin J. Lee <benjamin@cyclic.com> + + * build_lib.com: Added. DCL File to build contents of [.lib] + +Tue Feb 27 21:18:38 1996 Benjamin J. Lee <benjamin@cyclic.com> + + * system.h: added an existence_error macro check for EVMSERR + necessary for happiness under VMS + +Thu Feb 22 22:30:04 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * Makefile.in (OBJECTS): Remove @ALLOCA@ + (SOURCES): Remove alloca.c + * alloca.c: Removed. + * regex.c (REGEX_MALLOC): Define. + +Thu Feb 15 14:00:00 Jim Kingdon <kingdon@cyclic.com> + + * vasprintf.c: Declare abs(). + +Wed Feb 14 14:48:31 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * vasprintf.c (int_vasprintf): Don't cast arguments to memcpy. + * vasprintf.c, strtoul.c: Don't include ansidecl.h. Do include + config.h if HAVE_CONFIG_H (for const). + * strtoul.c: Change CONST to const. + +Tue Feb 13 20:04:39 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * strtoul.c: Added (needed by vasprintf.c, and missing on SunOS4). + * Makefile.in (SOURCES): Add strtoul.c. + +Mon Feb 12 10:04:46 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * vasprintf.c: Added (same contents as before). + * Makefile.in (SOURCES): Add vasprintf.c. + +Thu Feb 1 14:33:17 1996 Karl Fogel <kfogel@floss.red-bean.com> + + * Makefile.in (xlint): new rule; does nothing, as I'm not sure + running lint is actually advisable in here, but the top-level + Makefile thinks it can `make xlint' here. + +Thu Feb 1 15:07:42 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * getopt.c: Remove rcsid. + +Tue Jan 30 18:20:27 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * getline.c: Don't define NDEBUG. + (getstr): Rewrite assertions in a way which should stay clear of + signed/unsigned problems and compiler warnings thereof. + +Thu Jan 25 00:14:06 1996 Jim Kingdon <kingdon@beezley.cyclic.com> + + * yesno.c (yesno): fflush stdout as well as stderr. + +Wed Jan 3 18:16:50 1996 Jim Kingdon <kingdon@harvey.cyclic.com> + + * sighandle.c (SIG_register): Use memset not bzero. + * system.h: Remove defines for index, rindex, bcmp, and bzero. + All the calls to those functions are gone from CVS. + +Tue Jan 2 13:00:00 1996 Jim Kingdon <kingdon@peary.cyclic.com> + + Visual C++ lint: + * sighandle.c: Prototype SIG_handle and SIG_defaults. + Use SIG_ERR where appropriate. + +Mon Dec 18 10:15:05 1995 Jim Kingdon <kingdon@harvey.cyclic.com> + + * rename.c: Check ENOENT rather than existence_error. The latter + is undefined in this file, and including system.h is said to cause + (unspecified) problems. + +Sun Dec 17 23:58:06 1995 Jim Kingdon <kingdon@harvey.cyclic.com> + + * vasprintf.c: Removed (it is no longer used). + * Makefile.in (SOURCES): Remove vasprintf.c. + +Sat Dec 16 17:18:33 1995 Jim Kingdon <kingdon@harvey.cyclic.com> + + * vasprintf.c: Added. + * Makefile.in (SOURCES): Add vasprintf.c + +Mon Dec 4 10:54:04 1995 Jim Kingdon <kingdon@harvey.cyclic.com> + + * getdate.c: Remove #line directives. I know, this is a kludge, + but Visual C++ 2.1 seems to require it (why, I have no idea. It + has no trouble with the #line directives in getdate in CVS 1.6). + +Sat Nov 18 16:20:37 1995 Karl Fogel <kfogel@floss.cyclic.com> + + * rename.c: same. + + * mkdir.c: Use new macro `existence_error', instead of comparing + errno to ENOENT directly. + + * system.h (existence_error): new macro, tries to portably ask if + errno represents a file-not-exist error. + +Fri Nov 17 20:08:58 1995 Karl Fogel <kfogel@floss.cyclic.com> + + * system.h (NEED_DECOY_PERMISSIONS): moved this section to where + it belongs, duh. + + * getdate.c: if STDC_HEADERS, then just include <stdlib.h> instead + of declaring malloc() and realloc() to be char *. + + * system.h: ifdef NEED_DECOY_PERMISSIONS, then define the S_I* + permission masks for USR, GRP, and OTH in terms of the simpler + OS/2 masks. + +Wed Nov 15 15:36:03 1995 Karl Fogel <kfogel@floss.cyclic.com> + + * system.h: ifdef USE_OWN_TCPIP_H, then include "tcpip.h". Only + OS/2 does this right now. + +Tue Nov 14 18:44:57 1995 Greg A. Woods <woods@most.weird.com> + + * getdate.c: OK, this one is from SunOS-4.1 yacc and may be more + portable -- at least it compiles silently here! ;-) + +Mon Nov 13 03:53:45 1995 Karl Fogel <kfogel@floss.cyclic.com> + + * fnmatch.c: conform to 80 column standard (yes, I'm a pedant). + +Wed Nov 8 11:10:59 1995 Karl Fogel <kfogel@floss.cyclic.com> + + * system.h (STAT_MACROS): ifdef S_IFMT, then use it as before; but + if it's not defined, then just do a single mask and assume + acceptance any of non-zero result. Norbert, I trust you'll let me + know if this is unsatisfactory. :-) + Ifdef HAVE_SYS_UTIME_H, then include <sys/utime.h>. Only OS/2 + defines this right now. + +Wed Nov 8 13:18:51 1995 Norbert Kiesel <nk@col.sw-ley.de> + + * valloc.c: omit malloc declaration (it's already in system.h + which is included and conflicts with <stdlib.h> on some + systems). + +Tue Nov 7 19:38:48 1995 Norbert Kiesel <nk@col.sw-ley.de> + + * system.h (STAT_MACROS_BROKEN): undo previous change, because + else all regular files will be identified as links (the mask for + links is S_IFREG|S_IFCHR). + +Mon Nov 6 19:20:56 1995 Karl Fogel <kfogel@floss.cyclic.com> + + * system.h (STAT_MACROS_BROKEN): in defining the S_IF* macros, + don't fold to 1 or 0 by first masking with S_IFMT; not all + systems have that macro, and anyway it's only necessary that we + return non-zero. + +Fri Oct 27 13:43:35 1995 Karl Fogel <kfogel@floss.cyclic.com> + + * save-cwd.c: use __PROTO instead of __P (see below). + + * getline.h (__PROTO): same as below. + + * save-cwd.h (__PROTO): replaces __P. New name, so don't ask if + already defined. The conflict was that OS/2 w/ IBM C/C++ uses + `__P' for something else, in <ctype.h> of all places. + + * system.h: do nothing about alloca ifdef ALLOCA_IN_STDLIB (see + ../src/ChangeLog). + +Tue Oct 24 13:01:25 1995 Norbert Kiesel <nk@col.sw-ley.de> + + * wait.h: include sys/resource.h if available. This is needed at + least under AIX-3.2 where <sys/wait.h> doesn't include it. + +Mon Oct 23 17:39:11 1995 Norbert Kiesel <nk@col.sw-ley.de> + + * valloc.c (valloc): change parameter definition + +Sun Oct 22 14:15:44 1995 Jim Meyering (meyering@comco.com) + + * getline.c, getline.h: New files. + * Makefile.in (SOURCES, OBJECTS, HEADERS): Add getline.c, getline.o, + and getline.h, respectively. + +Tue Oct 10 18:01:50 1995 Karl Fogel <kfogel@totoro.cyclic.com> + + * Makefile.in (cvs_srcdir): define cvs_srcdir to be ../src, then + include it with -I so save_cwd.c can find error.h (for example). + +Sun Oct 8 12:27:57 1995 Peter Wemm <peter@haywire.DIALix.COM> + + * system.h: define POSIX_SIGNALS or BSD_SIGNALS if configure has + located all the necessary functions for each "type". + * sighandle.c: detect/use POSIX/BSD reliable signals (especially + for blocking signals in critical sections). Helps prevent stray + locks on interruption. + +Mon Oct 2 18:11:23 1995 Jim Blandy <jimb@totoro.cyclic.com> + + * system.h: Doc fix. + +Mon Oct 2 18:10:35 1995 Larry Jones <larry.jones@sdrc.com> + + * regex.c: compile 4.2 BSD compatible functions even when + _POSIX_SOURCE is defined since we need them and we wouldn't be + compiling this file unless they don't exist. + +Mon Oct 2 10:32:20 1995 Michael Finken <finken@conware.de> + + * strstr.c (strstr): new file and func. + + * Makefile.in (SOURCES): added strstr.c. + +Sun Oct 1 21:03:40 1995 Karl Fogel <kfogel@totoro.cyclic.com> + + * regex.c: reverted below change. + +Thu Sep 28 13:37:04 1995 Larry Jones <larry.jones@sdrc.com> + + * regexp.c: check for ISC. + +Thu Sep 7 19:18:00 1995 Jim Blandy <jimb@cyclic.com> + + * save-cwd.c: #include <direct.h> and <io.h>, on systems that + have them. + + * getopt.c (_getopt_internal): Cast the return value of strlen, + which is unsigned, before comparing it with the difference between + two pointers, which is unsigned. + +Thu Aug 31 11:31:42 1995 Jim Blandy <jimb@totoro.cyclic.com> + + * getdate.y [STDC_HEADERS]: #include <stdlib.h>, for abort. + [HAVE_ALLOCA_H]: #include <alloca.h>, for alloca on Windows NT. + +Wed Aug 30 18:48:44 1995 Jim Blandy <jimb@totoro.cyclic.com> + + * system.h [HAVE_IO_H]: #include <io.h>, for Windows NT. + [HAVE_DIRECT_H]: #include <direct.h>, for Windows NT. + (CVS_MKDIR, FOLD_FN_CHAR, fnfold, fncmp, ISDIRSEP, OPEN_BINARY, + FOPEN_BINARY_READ, FOPEN_BINARY_WRITE): New macros/functions, for + use in system-sensitive code. + + * regex.c (re_set_registers): start and end are pointers, not + integers. Cast the initializing value appropriately. + + * getopt.c [HAVE_STRING_H]: #include <string.h>, to avoid + warnings. + + * fnmatch.c (FOLD_FN_CHAR): Give this a dummy #definition if + config.h didn't #define it. + (fnmatch): Pass filename characters through FOLD_FN_CHAR before + comparing them. + + * argmatch.c: #include <sys/types.h>. + (argmatch): Declare arglen to be a size_t, rather than an int, + to avoid signed/unsigned comparison "problems". + + * .cvsignore: Remove getdate.c from this file. We want to + distribute it, for systems that don't have a Yacc-equivalent + installed (like Windows NT). + +Sat Aug 19 22:00:51 1995 Jim Blandy <jimb@totoro.cyclic.com> + + * error.c: Don't #define CVS_SUPPORT here. config.h takes care of + that for us. + [CVS_SUPPORT] (error_use_protocol): New variable, with apology. + (error): If error_use_protocol is set, report errors using the + client/server protocol. + * error.h [CVS_SUPPORT]: Extern decl for error_use_protocol. + +Fri Aug 4 00:01:24 1995 Jim Meyering (meyering@comco.com) + + * xgetwd.c: Don't declare free. A K&R style declaration gets + a conflict on some Sun systems when compiling with acc. + + * save-cwd.c: New file. + * save-cwd.h: New file. + * Makefile.in (SOURCES): Add save-cwd.c + (OBJECTS): Add save-cwd.o. + (HEADERS): Add save-cwd.h. + +Thu Aug 3 00:55:54 1995 Jim Meyering (meyering@comco.com) + + * error.h: New file. + * Makefile.in (HEADERS): Add error.h. + +Sat Jul 29 15:53:55 1995 James Kingdon <kingdon@harvey.cyclic.com> + + * Makefile.in (SOURCES): Add getdate.c. + +Thu Jul 27 09:11:41 1995 Robert Lipe <robertl@rjlhome.arnet.com> + + * system.h: Check for PATHSIZE before falling back to _POSIX_PATH_MAX. + +Thu Jul 20 12:38:03 1995 James Kingdon <kingdon@harvey.cyclic.com> + + * error.c: Instead of calling cvs functions to clean up, allow cvs + to register a callback via error_set_cleanup. Avoids hassles with + include files and SERVER_SUPPORT and so on. + +Tue Jul 18 21:18:00 1995 Jim Blandy <jimb@cyclic.com> + + * system.h: Include <sys/param.h> only if HAVE_SYS_PARAM_H + is #defined. We've added a test to configure.in to #define this + on most systems. + +Thu Jul 13 11:22:21 1995 Jim Meyering (meyering@comco.com) + + * xgetwd.c: New file. + * Makefile.in (SOURCES): Add xgetwd.c + (OBJECTS): Add xgetwd.o. + +Wed Jul 12 09:18:49 1995 Jim Meyering (meyering@comco.com) + + * Makefile.in (OBJECTS): Remove fnmatch.o. Now configure adds it + to LIBOBJS when necessary. + +Fri Jun 30 16:27:18 1995 James Kingdon <kingdon@harvey.cyclic.com> + + * rename.c (rename): If MVDIR is not defined, just give an error + on attempt to rename a directory. + +Thu Jun 29 00:46:31 1995 James Kingdon <kingdon@harvey.cyclic.com> + + * system.h: Check HAVE_SYS_TIMEB_H not non-existent HAVE_TIMEB_H. + + * system.h: Don't define alloca if it is already defined. + +Wed Jun 28 15:24:51 1995 James Kingdon <kingdon@harvey.cyclic.com> + + * system.h: If NeXT, define utimbuf ourself. + +Mon May 29 22:32:40 1995 J.T. Conklin <jtc@rtl.cygnus.com> + + * system.h: Handle time and directory headers as recommended in + the autoconf manual. + Undefine the S_FOO() macros if STAT_MACROS_BROKEN is set. + Don't define mode_t, as it is handled by config.h. + +Sat May 27 08:46:00 1995 Jim Meyering (meyering@comco.com) + + * Makefile.in (Makefile): Regenerate only Makefile in current + directory when Makefile.in is out of date. Depend on ../config.status. + +Fri Apr 28 22:49:25 1995 Jim Blandy <jimb@totoro.bio.indiana.edu> + + * Makefile.in (SOURCES, OBJECTS): Updated. + (HEADERS): New variable. + (DISTFILES): Updated. + (dist-dir): Renamed from dist; changed to work with DISTDIR + variable passed from parent. + +Wed Feb 8 06:37:53 1995 Roland McGrath <roland@churchy.gnu.ai.mit.edu> + + * system.h (S_IRUSR et al): Define if not already defined. + + * waitpid.c [HAVE_CONFIG_H]: Include "config.h". + (ualloc): Return OLDPTR rather than running off the end. + +Mon Aug 22 22:48:19 1994 Ken Raeburn (raeburn@kr-pc.cygnus.com) + + * error.c (strerror): Replaced conditional static definition + (always used, since the condition variable was never set) with an + extern declaration, since it's provided by libc or strerror.c. + +Wed Aug 10 14:54:25 1994 Ken Raeburn (raeburn@cujo.cygnus.com) + + * Makefile.in (SOURCES): Add waitpid.c. + * waitpid.c: New file. + +Tue Aug 9 16:00:12 1994 Ken Raeburn (raeburn@cujo.cygnus.com) + + * md5.h (uint32): If SIZEOF_LONG isn't 4, don't define this to be + "unsigned long"; try SIZEOF_INT and "unsigned int", otherwise + complain. + + * md5.c: Include config.h. + (const): Don't bother defining here, config.h should take care of + it. + + * valloc.c (malloc): Declare. + +Fri Jul 15 12:57:20 1994 Ian Lance Taylor (ian@sanguine.cygnus.com) + + * getopt.c: Do not include <stdlib.h> unless __GNU_LIBRARY__ is + defined. On Irix 5.2, <stdlib.h> includes <getopt.h>, which + causes a multiple definition of struct option. + +Fri Jul 8 10:04:59 1994 Jim Kingdon (kingdon@lioth.cygnus.com) + + * md5.h, md5.c: Remove ANSI-isms. + +Thu Jul 7 20:24:18 1994 Ian Lance Taylor (ian@sanguine.cygnus.com) + + * md5.h, md5.c: New files. + * Makefile.in (SOURCES): Add md5.c. + (OBJECTS): Add md5.o. + (DISTFILES): Add md5.h. + (md5.o): New target; depend upon md5.h. + +Fri May 27 18:15:34 1994 Ian Lance Taylor (ian@tweedledumb.cygnus.com) + + * valloc.c: New file. + +Tue May 17 08:18:26 1994 Jim Kingdon (kingdon@lioth.cygnus.com) + + * error.c (error, fperror): If server_active, call server_cleanup + as well as Lock_Cleanup. + +Thu Jan 6 13:45:04 1994 Ken Raeburn (raeburn@cujo.cygnus.com) + + * system.h: Fix Dec 27 change to work correctly. Makes Sep 9 + change unnecessary, so backed that one out. Never define PATH_MAX + in terms of pathconf, because that doesn't produce a constant, and + PATH_MAX is used to set array sizes. + +Mon Dec 27 14:22:07 1993 Mark Eichin (eichin@cygnus.com) + + * system.h: don't touch PATH_MAX or MAXPATHLEN if *both* of them + are already defined, as one may be defined in terms of the other. diff --git a/lib/ChangeLog.fsf b/lib/ChangeLog.fsf new file mode 100644 index 0000000..176d791 --- /dev/null +++ b/lib/ChangeLog.fsf @@ -0,0 +1,90 @@ +Thu Sep 15 00:18:26 1994 david d `zoo' zuhn <zoo@monad.armadillo.com> + + * system.h: remove a bunch of "extern int " declarations of system + functions (could conflict with vendor header files, and didn't + do anything *too* useful to begin with). + + * Makefile.in: update getdate.y message (now has 10 s/r conflicts) + +Wed Sep 14 22:12:21 1994 david d `zoo' zuhn <zoo@monad.armadillo.com> + + * strerror.c: more complete, from the Cygnus libiberty package + + * error.c (strerror): removed, functionality is in strerror.c + + * cvs.h: remove duplicate prototype for Reader_Lock + * history.c: printf argument mismatch + (Both fixes thanks to J.T. Conklin (jtc@cygnus.com) + +Sat Jul 30 13:50:11 1994 david d `zoo' zuhn (zoo@monad.armadillo.com) + + * getopt1.c, getopt.c, getopt.h, getdate.y: latest versions from FSF + +Wed Jul 13 22:11:17 1994 david d `zoo' zuhn (zoo@monad.armadillo.com) + + * system.h: don't set PATH_MAX to pathconf(), since PATH_MAX is + used to size arrays. (thanks to kingdon@cygnus.com) + + * getopt1.c: remove #ifdef __STDC__ around const usages (which + isn't correct and weren't complete) + +Wed Apr 20 14:57:16 1994 Ian Lance Taylor (ian@tweedledumb.cygnus.com) + + * getopt.h: Prevent multiple inclusion. + +Tue Jan 25 17:34:42 1994 david d zuhn (zoo@monad.armadillo.com) + + * Makefile.in: make sure that no blank lines are in the $(OBJECTS) + list (from Brad Figg) + +Mon Jan 24 12:27:13 1994 david d zuhn (zoo@monad.armadillo.com) + + * system.h: remove alloca checks (added to src/cvs.h); revamped + the MAXPATHLEN and PATH_MAX tests (from Brad Figg + <bradf@wv.MENTORG.COM>); handle index,rindex,bcmp,bzero better + (don't redefine if already defined); added S_IWRITE, S_IWGRP, + S_IWOTH definitions (header file reorganization) + + * strippath.c: use strchr, not index + + * getopt1.c: match prototypes when __STDC__ compiler (lint fixes) + + * getdate.c: alloca checks for when using bison + + * Makefile.in: added CC and YACC definitions; use YACC not BISON; + better getdate.c tests (also from Brad Figg) + +Sat Dec 18 00:55:43 1993 david d zuhn (zoo@monad.armadillo.com) + + * Makefile.in (VPATH): don't use $(srcdir), but @srcdir@ instead + + * memmove.c: new file, implements memmove in terms of bcopy + + * wait.h: include <sys/wait.h> if HAVE_SYS_WAIT_H, not if POSIX + +Thu Sep 9 18:02:11 1993 david d `zoo' zuhn (zoo@rtl.cygnus.com) + + * system.h: only #undef PATH_MAX if not on an Alpha. The #undef + causes problems with the Alpha C compiler. + +Thu Apr 8 12:39:56 1993 Ian Lance Taylor (ian@cygnus.com) + + * system.h: Removed several incorrect declarations which fail + on Solaris. + +Wed Jan 20 17:57:24 1993 K. Richard Pixley (rich@rtl.cygnus.com) + + * system.h: add externs for sun4 so that gcc -Wall becomes useful + again. + +Wed Feb 26 18:04:40 1992 K. Richard Pixley (rich@cygnus.com) + + * Makefile.in, configure.in: removed traces of namesubdir, + -subdirs, $(subdir), $(unsubdir), some rcs triggers. Forced + copyrights to '92, changed some from Cygnus to FSF. + +Sat Dec 28 02:42:06 1991 K. Richard Pixley (rich at cygnus.com) + + * mkdir.c, rename.c: change fork() to vfork(). + + diff --git a/lib/Makefile.am b/lib/Makefile.am new file mode 100644 index 0000000..1b8ffa5 --- /dev/null +++ b/lib/Makefile.am @@ -0,0 +1,321 @@ +## Process this file with automake to produce Makefile.in +# Makefile for library files used by GNU CVS. +# +# Copyright (C) 1986-2005 The Free Software Foundation, Inc. +# +# Portions Copyright (C) 1998-2005 Derek Price, Ximbiot <http://ximbiot.com>, +# and others. + +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2, or (at your option) +# any later version. + +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. + +noinst_LIBRARIES = libcvs.a + +# Should look into unifying regular expression matching in CVS +# with the diff library (perhaps to have the caller, CVS, do the +# matching?) +libcvs_a_SOURCES = \ + sighandle.c \ + system.h \ + wait.h \ + xselect.h \ + xtime.h +## Because @LIBOBJS@ is included below and AC_LIBOBJ is sometimes called on +## these files from configure.in, automake automatically knows about many of +## the *.c files in this directory. +libcvs_a_LIBADD = \ + @LIBOBJS@ + + + +## Zero some variables so Automake will allow us to use += on them. +BUILT_SOURCES = +MOSTLYCLEANFILES = +DISTCLEANFILES = +MAINTAINERCLEANFILES = +TESTS = +check_PROGRAMS = + +EXTRA_DIST = \ + .cvsignore \ + ChangeLog.fsf \ + Makefile.gnulib \ + build_lib.com \ + libcvs.dep \ + libcvs.dsp \ + libcvs.mak \ + test-getdate.sh + +## begin gnulib module alloca + + +libcvs_a_LIBADD += @ALLOCA@ +## end gnulib module alloca + +## begin gnulib module alloca-opt + +BUILT_SOURCES += $(ALLOCA_H) +EXTRA_DIST += alloca_.h + +# We need the following in order to create <alloca.h> when the system +# doesn't have one that works with the given compiler. +alloca.h: alloca_.h + cp $(srcdir)/alloca_.h $@-t + mv $@-t $@ +MOSTLYCLEANFILES += alloca.h alloca.h-t + +## end gnulib module alloca-opt + +## begin gnulib module allocsa + +libcvs_a_SOURCES += allocsa.h allocsa.c +EXTRA_DIST += allocsa.valgrind + +## end gnulib module allocsa + +## begin gnulib module cycle-check + +libcvs_a_SOURCES += cycle-check.c cycle-check.h dev-ino.h + +## end gnulib module cycle-check + +## begin gnulib module dirname + +libcvs_a_SOURCES += basename.c stripslash.c + +## end gnulib module dirname + +## For GNULIB's error module. +## +## This module isn't fully imported since GNULIB's error.c would conflict with +## src/error.c (which knows how to send error messages over the network). We +## don't compile error.c into libcvs on purpose. It can get compiled later as +## part of the getdate test program, +libcvs_a_SOURCES += error.h + +## begin gnulib module exit + +libcvs_a_SOURCES += exit.h + +## end gnulib module exit + +## begin gnulib module fnmatch + +BUILT_SOURCES += $(FNMATCH_H) +EXTRA_DIST += fnmatch_.h fnmatch_loop.c + +# We need the following in order to create <fnmatch.h> when the system +# doesn't have one that supports the required API. +fnmatch.h: fnmatch_.h + cp $(srcdir)/fnmatch_.h $@-t + mv $@-t $@ +MOSTLYCLEANFILES += fnmatch.h fnmatch.h-t + +## end gnulib module fnmatch + +## begin gnulib module getaddrinfo + +libcvs_a_SOURCES += getaddrinfo.h + +## end gnulib module getaddrinfo + +## begin gnulib module getdate + +## CVS test scripts for getdate. +TESTS += test-getdate.sh +MOSTLYCLEANFILES += getdate-expected getdate-got getdate.diff +DISTCLEANFILES += getdate.log +# Program required by test-getdate.sh for testing getdate.y. +check_PROGRAMS += getdate +getdate_SOURCES = \ + error.c \ + getdate.y +## This source file was added only for the getdate test program when compiled +## with GNULIB's error.c. +getdate_SOURCES += \ + progname.c +getdate_CPPFLAGS = -DTEST +getdate_LDADD = \ + $(noinst_LIBRARIES) \ + $(LIB_CLOCK_GETTIME) \ + $(LIBINTL) + +BUILT_SOURCES += getdate.c +MAINTAINERCLEANFILES += getdate.c +EXTRA_DIST += getdate.c + +## end gnulib module getdate + +## begin gnulib module getndelim2 + +EXTRA_DIST += getndelim2.h getndelim2.c + +## end gnulib module getndelim2 + +## begin gnulib module getnline + +libcvs_a_SOURCES += getnline.h getnline.c + +## end gnulib module getnline + +## begin gnulib module getopt + +BUILT_SOURCES += $(GETOPT_H) +EXTRA_DIST += getopt_.h getopt_int.h + +# We need the following in order to create <getopt.h> when the system +# doesn't have one that works with the given compiler. +getopt.h: getopt_.h + cp $(srcdir)/getopt_.h $@-t + mv $@-t $@ +MOSTLYCLEANFILES += getopt.h getopt.h-t + +## end gnulib module getopt + +## begin gnulib module gettext-h + +libcvs_a_SOURCES += gettext.h + +## end gnulib module gettext-h + +## begin gnulib module glob + +BUILT_SOURCES += $(GLOB_H) + +# We need the following in order to create an <getopt.h> when the system +# doesn't have one that works with the given compiler. +all-local $(libcvs_a_OBJECTS): $(GLOB_H) +glob.h: glob_.h + cp $(srcdir)/glob_.h $@-t + mv $@-t $@ +MOSTLYCLEANFILES += glob.h glob.h-t + +## end gnulib module glob.c + +## begin gnulib module mbuiter + +libcvs_a_SOURCES += mbuiter.h + +## end gnulib module mbuiter + +## begin gnulib module minmax + +libcvs_a_SOURCES += minmax.h + +## end gnulib module minmax + +## begin gnulib module setenv + +libcvs_a_SOURCES += setenv.h + +## end gnulib module setenv + +## begin gnulib module size_max + +libcvs_a_SOURCES += size_max.h + +## end gnulib module size_max + +## begin gnulib module stdbool + +BUILT_SOURCES += $(STDBOOL_H) +EXTRA_DIST += stdbool_.h + +# We need the following in order to create <stdbool.h> when the system +# doesn't have one that works. +stdbool.h: stdbool_.h + sed -e 's/@''HAVE__BOOL''@/$(HAVE__BOOL)/g' \ + < $(srcdir)/stdbool_.h > $@-t + mv $@-t $@ +MOSTLYCLEANFILES += stdbool.h stdbool.h-t + +## end gnulib module stdbool + +## begin gnulib module stdint + +BUILT_SOURCES += $(STDINT_H) +EXTRA_DIST += stdint_.h + +# We need the following in order to create <stdint.h> when the system +# doesn't have one that works with the given compiler. +stdint.h: stdint_.h + sed -e 's/@''HAVE_LONG_64BIT''@/$(HAVE_LONG_64BIT)/g;s/@''HAVE_LONG_LONG_64BIT@/$(HAVE_LONG_LONG_64BIT)/g' < $(srcdir)/stdint_.h > $@-t + mv $@-t $@ +MOSTLYCLEANFILES += stdint.h stdint.h-t + +## end gnulib module stdint + +## begin gnulib module strcase + +libcvs_a_SOURCES += strcase.h + +## end gnulib module strcase + +## begin gnulib module strnlen1 + +libcvs_a_SOURCES += strnlen1.h strnlen1.c + +## end gnulib module strnlen1 + +## begin gnulib module strstr + +libcvs_a_SOURCES += strstr.h + +## end gnulib module strstr + +## begin gnulib module time_r + +libcvs_a_SOURCES += time_r.h + +## end gnulib module time_r + +## begin gnulib module vasnprintf + +libcvs_a_SOURCES += printf-args.h printf-parse.h vasnprintf.h + +## end gnulib module vasnprintf + +## begin gnulib module vasprintf + +libcvs_a_SOURCES += vasprintf.h + +## end gnulib module vasprintf + +## begin gnulib module xalloc-die + +libcvs_a_SOURCES += xalloc-die.c + +## end gnulib module xalloc-die + +## begin gnulib module xgethostname + +libcvs_a_SOURCES += xgethostname.h xgethostname.c + +## end gnulib module xgethostname + +## begin gnulib module xreadlink + +libcvs_a_SOURCES += xreadlink.h xreadlink.c + +## end gnulib module xreadlink + +## begin gnulib module xsize + +libcvs_a_SOURCES += xsize.h + +## end gnulib module xsize + +# Until Automake gets its act together +distclean-local: + rm -f fnmatch.h + +# for backwards compatibility with the old makefiles +realclean: maintainer-clean +.PHONY: realclean diff --git a/lib/Makefile.gnulib b/lib/Makefile.gnulib new file mode 100644 index 0000000..0ec0d06 --- /dev/null +++ b/lib/Makefile.gnulib @@ -0,0 +1,258 @@ +## Process this file with automake to produce Makefile.in. +# Copyright (C) 2004 Free Software Foundation, Inc. +# +# This file is free software, distributed under the terms of the GNU +# General Public License. As a special exception to the GNU General +# Public License, this file may be distributed as part of a program +# that contains a configuration script generated by Automake, under +# the same distribution terms as the rest of that program. +# +# Generated by gnulib-tool. +# Reproduce by: gnulib-tool --import --dir=. --lib=libgnu --source-base=lib --m4-base=m4 --aux-dir=build-aux --macro-prefix=gl allocsa atexit canon-host canonicalize closeout dirname dup2 error exit exitfail extensions fnmatch fnmatch-posix ftruncate getdate gethostname getline getlogin_r getndelim2 getnline getopt getpagesize getpass-gnu gettext gettime gettimeofday glob lstat malloc md5 memmove minmax mkdir mkstemp mktime nanosleep pagealign_alloc pathmax quotearg readlink realloc regex rename restrict save-cwd setenv stat-macros stdbool stdint strcase strdup strerror strftime strstr strtoul time_r timespec tzset unlocked-io vasnprintf vasprintf xalloc-die xgethostname xreadlink xsize yesno + +AUTOMAKE_OPTIONS = 1.5 gnits no-dependencies + +noinst_LIBRARIES = libgnu.a + +libgnu_a_SOURCES = +libgnu_a_LIBADD = @LIBOBJS@ +EXTRA_DIST = +BUILT_SOURCES = +SUFFIXES = +MOSTLYCLEANFILES = +CLEANFILES = +DISTCLEANFILES = +MAINTAINERCLEANFILES = + +## begin gnulib module alloca + + +libgnu_a_LIBADD += @ALLOCA@ +## end gnulib module alloca + +## begin gnulib module alloca-opt + +BUILT_SOURCES += $(ALLOCA_H) +EXTRA_DIST += alloca_.h + +# We need the following in order to create <alloca.h> when the system +# doesn't have one that works with the given compiler. +alloca.h: alloca_.h + cp $(srcdir)/alloca_.h $@-t + mv $@-t $@ +MOSTLYCLEANFILES += alloca.h alloca.h-t + +## end gnulib module alloca-opt + +## begin gnulib module allocsa + +libgnu_a_SOURCES += allocsa.h allocsa.c +EXTRA_DIST += allocsa.valgrind + +## end gnulib module allocsa + +## begin gnulib module cycle-check + +libgnu_a_SOURCES += cycle-check.c cycle-check.h dev-ino.h + +## end gnulib module cycle-check + +## begin gnulib module dirname + +libgnu_a_SOURCES += basename.c stripslash.c + +## end gnulib module dirname + +## begin gnulib module exit + +libgnu_a_SOURCES += exit.h + +## end gnulib module exit + +## begin gnulib module fnmatch + +BUILT_SOURCES += $(FNMATCH_H) +EXTRA_DIST += fnmatch_.h fnmatch_loop.c + +# We need the following in order to create <fnmatch.h> when the system +# doesn't have one that supports the required API. +fnmatch.h: fnmatch_.h + cp $(srcdir)/fnmatch_.h $@-t + mv $@-t $@ +MOSTLYCLEANFILES += fnmatch.h fnmatch.h-t + +## end gnulib module fnmatch + +## begin gnulib module getaddrinfo + +libgnu_a_SOURCES += getaddrinfo.h + +## end gnulib module getaddrinfo + +## begin gnulib module getdate + +BUILT_SOURCES += getdate.c +MAINTAINERCLEANFILES += getdate.c +EXTRA_DIST += getdate.c + +## end gnulib module getdate + +## begin gnulib module getndelim2 + +EXTRA_DIST += getndelim2.h getndelim2.c + +## end gnulib module getndelim2 + +## begin gnulib module getnline + +libgnu_a_SOURCES += getnline.h getnline.c + +## end gnulib module getnline + +## begin gnulib module getopt + +BUILT_SOURCES += $(GETOPT_H) +EXTRA_DIST += getopt_.h getopt_int.h + +# We need the following in order to create <getopt.h> when the system +# doesn't have one that works with the given compiler. +getopt.h: getopt_.h + cp $(srcdir)/getopt_.h $@-t + mv $@-t $@ +MOSTLYCLEANFILES += getopt.h getopt.h-t + +## end gnulib module getopt + +## begin gnulib module gettext-h + +libgnu_a_SOURCES += gettext.h + +## end gnulib module gettext-h + +## begin gnulib module glob + +BUILT_SOURCES += $(GLOB_H) + +# We need the following in order to create <glob.h> when the system +# doesn't have one that works with the given compiler. +glob.h: glob_.h + cp $(srcdir)/glob_.h $@-t + mv $@-t $@ +MOSTLYCLEANFILES += glob.h glob.h-t + +## end gnulib module glob + +## begin gnulib module mbuiter + +libgnu_a_SOURCES += mbuiter.h + +## end gnulib module mbuiter + +## begin gnulib module minmax + +libgnu_a_SOURCES += minmax.h + +## end gnulib module minmax + +## begin gnulib module setenv + +libgnu_a_SOURCES += setenv.h + +## end gnulib module setenv + +## begin gnulib module size_max + +libgnu_a_SOURCES += size_max.h + +## end gnulib module size_max + +## begin gnulib module stdbool + +BUILT_SOURCES += $(STDBOOL_H) +EXTRA_DIST += stdbool_.h + +# We need the following in order to create <stdbool.h> when the system +# doesn't have one that works. +stdbool.h: stdbool_.h + sed -e 's/@''HAVE__BOOL''@/$(HAVE__BOOL)/g' < $(srcdir)/stdbool_.h > $@-t + mv $@-t $@ +MOSTLYCLEANFILES += stdbool.h stdbool.h-t + +## end gnulib module stdbool + +## begin gnulib module stdint + +BUILT_SOURCES += $(STDINT_H) +EXTRA_DIST += stdint_.h + +# We need the following in order to create <stdint.h> when the system +# doesn't have one that works with the given compiler. +stdint.h: stdint_.h + sed -e 's/@''HAVE_LONG_64BIT''@/$(HAVE_LONG_64BIT)/g;s/@''HAVE_LONG_LONG_64BIT@/$(HAVE_LONG_LONG_64BIT)/g' < $(srcdir)/stdint_.h > $@-t + mv $@-t $@ +MOSTLYCLEANFILES += stdint.h stdint.h-t + +## end gnulib module stdint + +## begin gnulib module strcase + +libgnu_a_SOURCES += strcase.h + +## end gnulib module strcase + +## begin gnulib module strnlen1 + +libgnu_a_SOURCES += strnlen1.h strnlen1.c + +## end gnulib module strnlen1 + +## begin gnulib module strstr + +libgnu_a_SOURCES += strstr.h + +## end gnulib module strstr + +## begin gnulib module time_r + +libgnu_a_SOURCES += time_r.h + +## end gnulib module time_r + +## begin gnulib module vasnprintf + +libgnu_a_SOURCES += printf-args.h printf-parse.h vasnprintf.h + +## end gnulib module vasnprintf + +## begin gnulib module vasprintf + +libgnu_a_SOURCES += vasprintf.h + +## end gnulib module vasprintf + +## begin gnulib module xalloc-die + +libgnu_a_SOURCES += xalloc-die.c + +## end gnulib module xalloc-die + +## begin gnulib module xgethostname + +libgnu_a_SOURCES += xgethostname.h xgethostname.c + +## end gnulib module xgethostname + +## begin gnulib module xreadlink + +libgnu_a_SOURCES += xreadlink.h xreadlink.c + +## end gnulib module xreadlink + +## begin gnulib module xsize + +libgnu_a_SOURCES += xsize.h + +## end gnulib module xsize + + +# Makefile.am ends here diff --git a/lib/Makefile.in b/lib/Makefile.in new file mode 100644 index 0000000..1dc29df --- /dev/null +++ b/lib/Makefile.in @@ -0,0 +1,855 @@ +# Makefile.in generated by automake 1.9.6 from Makefile.am. +# @configure_input@ + +# Copyright (C) 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, +# 2003, 2004, 2005 Free Software Foundation, Inc. +# This Makefile.in is free software; the Free Software Foundation +# gives unlimited permission to copy and/or distribute it, +# with or without modifications, as long as this notice is preserved. + +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY, to the extent permitted by law; without +# even the implied warranty of MERCHANTABILITY or FITNESS FOR A +# PARTICULAR PURPOSE. + +@SET_MAKE@ + +# Makefile for library files used by GNU CVS. +# +# Copyright (C) 1986-2005 The Free Software Foundation, Inc. +# +# Portions Copyright (C) 1998-2005 Derek Price, Ximbiot <http://ximbiot.com>, +# and others. + +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2, or (at your option) +# any later version. + +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. + +srcdir = @srcdir@ +top_srcdir = @top_srcdir@ +VPATH = @srcdir@ +pkgdatadir = $(datadir)/@PACKAGE@ +pkglibdir = $(libdir)/@PACKAGE@ +pkgincludedir = $(includedir)/@PACKAGE@ +top_builddir = .. +am__cd = CDPATH="$${ZSH_VERSION+.}$(PATH_SEPARATOR)" && cd +INSTALL = @INSTALL@ +install_sh_DATA = $(install_sh) -c -m 644 +install_sh_PROGRAM = $(install_sh) -c +install_sh_SCRIPT = $(install_sh) -c +INSTALL_HEADER = $(INSTALL_DATA) +transform = $(program_transform_name) +NORMAL_INSTALL = : +PRE_INSTALL = : +POST_INSTALL = : +NORMAL_UNINSTALL = : +PRE_UNINSTALL = : +POST_UNINSTALL = : +build_triplet = @build@ +host_triplet = @host@ +check_PROGRAMS = getdate$(EXEEXT) +subdir = lib +DIST_COMMON = README $(srcdir)/Makefile.am $(srcdir)/Makefile.in \ + ChangeLog __fpending.c __fpending.h alloca.c asnprintf.c \ + asprintf.c atexit.c canon-host.c canon-host.h canonicalize.c \ + canonicalize.h chdir-long.c chdir-long.h closeout.c closeout.h \ + dirname.c dirname.h dup-safer.c dup2.c exitfail.c exitfail.h \ + fd-safer.c filenamecat.c filenamecat.h fncase.c fnmatch.c \ + fseeko.c ftello.c ftruncate.c gai_strerror.c getaddrinfo.c \ + getcwd.c getcwd.h getdate.c getdate.h getdate.y getdelim.c \ + getdelim.h gethostname.c getline.c getline.h getlogin_r.c \ + getlogin_r.h getndelim2.c getopt.c getopt1.c getpagesize.h \ + getpass.c getpass.h gettime.c gettimeofday.c glob-libc.h \ + glob.c glob_.h lstat.c lstat.h malloc.c mbchar.c mbchar.h \ + md5.c md5.h memchr.c memmove.c mempcpy.c mempcpy.h memrchr.c \ + memrchr.h mkdir.c mkstemp.c mktime.c nanosleep.c openat.c \ + openat.h pagealign_alloc.c pagealign_alloc.h pathmax.h \ + pipe-safer.c printf-args.c printf-parse.c quotearg.c \ + quotearg.h readlink.c realloc.c regcomp.c regex.c regex.h \ + regex_internal.c regex_internal.h regexec.c rename.c rpmatch.c \ + save-cwd.c save-cwd.h setenv.c stat-macros.h strcasecmp.c \ + strdup.c strdup.h strerror.c strftime.c strftime.h \ + strncasecmp.c strstr.c strtol.c strtoul.c sunos57-select.c \ + tempname.c time_r.c timespec.h unistd--.h unistd-safer.h \ + unlocked-io.h unsetenv.c vasnprintf.c vasprintf.c waitpid.c \ + xalloc.h xgetcwd.c xgetcwd.h xmalloc.c yesno.c yesno.h +ACLOCAL_M4 = $(top_srcdir)/aclocal.m4 +am__aclocal_m4_deps = $(top_srcdir)/m4/acx_extract_cpp_defn.m4 \ + $(top_srcdir)/m4/acx_with_external_zlib.m4 \ + $(top_srcdir)/m4/acx_with_gssapi.m4 $(top_srcdir)/m4/alloca.m4 \ + $(top_srcdir)/m4/allocsa.m4 \ + $(top_srcdir)/m4/asx_version_compare.m4 \ + $(top_srcdir)/m4/atexit.m4 $(top_srcdir)/m4/bison.m4 \ + $(top_srcdir)/m4/canon-host.m4 \ + $(top_srcdir)/m4/canonicalize.m4 \ + $(top_srcdir)/m4/chdir-long.m4 $(top_srcdir)/m4/clock_time.m4 \ + $(top_srcdir)/m4/closeout.m4 $(top_srcdir)/m4/codeset.m4 \ + $(top_srcdir)/m4/cvs_func_printf_ptr.m4 \ + $(top_srcdir)/m4/d-ino.m4 $(top_srcdir)/m4/d-type.m4 \ + $(top_srcdir)/m4/dirname.m4 $(top_srcdir)/m4/dos.m4 \ + $(top_srcdir)/m4/dup2.m4 $(top_srcdir)/m4/eealloc.m4 \ + $(top_srcdir)/m4/eoverflow.m4 $(top_srcdir)/m4/error.m4 \ + $(top_srcdir)/m4/exitfail.m4 $(top_srcdir)/m4/extensions.m4 \ + $(top_srcdir)/m4/filenamecat.m4 $(top_srcdir)/m4/fnmatch.m4 \ + $(top_srcdir)/m4/fpending.m4 $(top_srcdir)/m4/ftruncate.m4 \ + $(top_srcdir)/m4/getaddrinfo.m4 \ + $(top_srcdir)/m4/getcwd-path-max.m4 $(top_srcdir)/m4/getcwd.m4 \ + $(top_srcdir)/m4/getdate.m4 $(top_srcdir)/m4/getdelim.m4 \ + $(top_srcdir)/m4/gethostname.m4 $(top_srcdir)/m4/getline.m4 \ + $(top_srcdir)/m4/getlogin_r.m4 $(top_srcdir)/m4/getndelim2.m4 \ + $(top_srcdir)/m4/getnline.m4 $(top_srcdir)/m4/getopt.m4 \ + $(top_srcdir)/m4/getpagesize.m4 $(top_srcdir)/m4/getpass.m4 \ + $(top_srcdir)/m4/gettext.m4 $(top_srcdir)/m4/gettime.m4 \ + $(top_srcdir)/m4/gettimeofday.m4 $(top_srcdir)/m4/glob.m4 \ + $(top_srcdir)/m4/gnulib-comp.m4 $(top_srcdir)/m4/iconv.m4 \ + $(top_srcdir)/m4/intmax_t.m4 $(top_srcdir)/m4/inttypes.m4 \ + $(top_srcdir)/m4/inttypes_h.m4 $(top_srcdir)/m4/lib-ld.m4 \ + $(top_srcdir)/m4/lib-link.m4 $(top_srcdir)/m4/lib-prefix.m4 \ + $(top_srcdir)/m4/longdouble.m4 $(top_srcdir)/m4/longlong.m4 \ + $(top_srcdir)/m4/lstat.m4 $(top_srcdir)/m4/mbchar.m4 \ + $(top_srcdir)/m4/mbiter.m4 $(top_srcdir)/m4/mbrtowc.m4 \ + $(top_srcdir)/m4/mbstate_t.m4 $(top_srcdir)/m4/md5.m4 \ + $(top_srcdir)/m4/memchr.m4 $(top_srcdir)/m4/memmove.m4 \ + $(top_srcdir)/m4/mempcpy.m4 $(top_srcdir)/m4/memrchr.m4 \ + $(top_srcdir)/m4/minmax.m4 $(top_srcdir)/m4/mkdir-slash.m4 \ + $(top_srcdir)/m4/mkstemp.m4 $(top_srcdir)/m4/mktime.m4 \ + $(top_srcdir)/m4/mmap-anon.m4 $(top_srcdir)/m4/nanosleep.m4 \ + $(top_srcdir)/m4/nls.m4 $(top_srcdir)/m4/onceonly_2_57.m4 \ + $(top_srcdir)/m4/openat.m4 $(top_srcdir)/m4/pagealign_alloc.m4 \ + $(top_srcdir)/m4/pathmax.m4 $(top_srcdir)/m4/po.m4 \ + $(top_srcdir)/m4/progtest.m4 $(top_srcdir)/m4/quotearg.m4 \ + $(top_srcdir)/m4/readlink.m4 $(top_srcdir)/m4/regex.m4 \ + $(top_srcdir)/m4/rename.m4 $(top_srcdir)/m4/restrict.m4 \ + $(top_srcdir)/m4/rpmatch.m4 $(top_srcdir)/m4/save-cwd.m4 \ + $(top_srcdir)/m4/setenv.m4 $(top_srcdir)/m4/signed.m4 \ + $(top_srcdir)/m4/size_max.m4 $(top_srcdir)/m4/sockpfaf.m4 \ + $(top_srcdir)/m4/ssize_t.m4 $(top_srcdir)/m4/stat-macros.m4 \ + $(top_srcdir)/m4/stdbool.m4 $(top_srcdir)/m4/stdint.m4 \ + $(top_srcdir)/m4/stdint_h.m4 $(top_srcdir)/m4/strcase.m4 \ + $(top_srcdir)/m4/strdup.m4 $(top_srcdir)/m4/strerror.m4 \ + $(top_srcdir)/m4/strftime.m4 $(top_srcdir)/m4/strstr.m4 \ + $(top_srcdir)/m4/strtol.m4 $(top_srcdir)/m4/strtoul.m4 \ + $(top_srcdir)/m4/sunos57-select.m4 $(top_srcdir)/m4/time_r.m4 \ + $(top_srcdir)/m4/timespec.m4 $(top_srcdir)/m4/tm_gmtoff.m4 \ + $(top_srcdir)/m4/tzset.m4 $(top_srcdir)/m4/uint32_t.m4 \ + $(top_srcdir)/m4/uintmax_t.m4 $(top_srcdir)/m4/ulonglong.m4 \ + $(top_srcdir)/m4/unistd-safer.m4 \ + $(top_srcdir)/m4/unlocked-io.m4 $(top_srcdir)/m4/vasnprintf.m4 \ + $(top_srcdir)/m4/vasprintf.m4 $(top_srcdir)/m4/wchar_t.m4 \ + $(top_srcdir)/m4/wint_t.m4 $(top_srcdir)/m4/xalloc.m4 \ + $(top_srcdir)/m4/xgetcwd.m4 $(top_srcdir)/m4/xreadlink.m4 \ + $(top_srcdir)/m4/xsize.m4 $(top_srcdir)/m4/yesno.m4 \ + $(top_srcdir)/configure.in +am__configure_deps = $(am__aclocal_m4_deps) $(CONFIGURE_DEPENDENCIES) \ + $(ACLOCAL_M4) +mkinstalldirs = $(install_sh) -d +CONFIG_HEADER = $(top_builddir)/config.h +CONFIG_CLEAN_FILES = +LIBRARIES = $(noinst_LIBRARIES) +AR = ar +ARFLAGS = cru +libcvs_a_AR = $(AR) $(ARFLAGS) +libcvs_a_DEPENDENCIES = @LIBOBJS@ @ALLOCA@ +am_libcvs_a_OBJECTS = sighandle.$(OBJEXT) allocsa.$(OBJEXT) \ + cycle-check.$(OBJEXT) basename.$(OBJEXT) stripslash.$(OBJEXT) \ + getnline.$(OBJEXT) strnlen1.$(OBJEXT) xalloc-die.$(OBJEXT) \ + xgethostname.$(OBJEXT) xreadlink.$(OBJEXT) +libcvs_a_OBJECTS = $(am_libcvs_a_OBJECTS) +am_getdate_OBJECTS = getdate-error.$(OBJEXT) getdate-getdate.$(OBJEXT) \ + getdate-progname.$(OBJEXT) +getdate_OBJECTS = $(am_getdate_OBJECTS) +am__DEPENDENCIES_1 = libcvs.a +am__DEPENDENCIES_2 = +getdate_DEPENDENCIES = $(am__DEPENDENCIES_1) $(am__DEPENDENCIES_2) \ + $(am__DEPENDENCIES_2) +DEFAULT_INCLUDES = -I. -I$(srcdir) -I$(top_builddir) +depcomp = $(SHELL) $(top_srcdir)/build-aux/depcomp +am__depfiles_maybe = depfiles +COMPILE = $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(AM_CPPFLAGS) \ + $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) +CCLD = $(CC) +LINK = $(CCLD) $(AM_CFLAGS) $(CFLAGS) $(AM_LDFLAGS) $(LDFLAGS) -o $@ +YACCCOMPILE = $(YACC) $(YFLAGS) $(AM_YFLAGS) +YLWRAP = $(top_srcdir)/build-aux/ylwrap +SOURCES = $(libcvs_a_SOURCES) $(getdate_SOURCES) +DIST_SOURCES = $(libcvs_a_SOURCES) $(getdate_SOURCES) +ETAGS = etags +CTAGS = ctags +DISTFILES = $(DIST_COMMON) $(DIST_SOURCES) $(TEXINFOS) $(EXTRA_DIST) +ACLOCAL = @ACLOCAL@ +ALLOCA = @ALLOCA@ +ALLOCA_H = @ALLOCA_H@ +AMDEP_FALSE = @AMDEP_FALSE@ +AMDEP_TRUE = @AMDEP_TRUE@ +AMTAR = @AMTAR@ +AUTOCONF = @AUTOCONF@ +AUTOHEADER = @AUTOHEADER@ +AUTOMAKE = @AUTOMAKE@ +AWK = @AWK@ +CC = @CC@ +CCDEPMODE = @CCDEPMODE@ +CFLAGS = @CFLAGS@ +CPP = @CPP@ +CPPFLAGS = @CPPFLAGS@ +CSH = @CSH@ +CYGPATH_W = @CYGPATH_W@ +DEFS = @DEFS@ +DEPDIR = @DEPDIR@ +ECHO_C = @ECHO_C@ +ECHO_N = @ECHO_N@ +ECHO_T = @ECHO_T@ +EDITOR = @EDITOR@ +EGREP = @EGREP@ +EOVERFLOW = @EOVERFLOW@ +EXEEXT = @EXEEXT@ +FNMATCH_H = @FNMATCH_H@ +GETOPT_H = @GETOPT_H@ +GLOB_H = @GLOB_H@ +GMSGFMT = @GMSGFMT@ +HAVE_LONG_64BIT = @HAVE_LONG_64BIT@ +HAVE_LONG_LONG_64BIT = @HAVE_LONG_LONG_64BIT@ +HAVE__BOOL = @HAVE__BOOL@ +INSTALL_DATA = @INSTALL_DATA@ +INSTALL_PROGRAM = @INSTALL_PROGRAM@ +INSTALL_SCRIPT = @INSTALL_SCRIPT@ +INSTALL_STRIP_PROGRAM = @INSTALL_STRIP_PROGRAM@ +INTLLIBS = @INTLLIBS@ +INTL_MACOSX_LIBS = @INTL_MACOSX_LIBS@ +KRB4 = @KRB4@ +LDFLAGS = @LDFLAGS@ +LIBICONV = @LIBICONV@ +LIBINTL = @LIBINTL@ +LIBOBJS = @LIBOBJS@ +LIBS = @LIBS@ +LIB_CLOCK_GETTIME = @LIB_CLOCK_GETTIME@ +LIB_NANOSLEEP = @LIB_NANOSLEEP@ +LN_S = @LN_S@ +LTLIBICONV = @LTLIBICONV@ +LTLIBINTL = @LTLIBINTL@ +LTLIBOBJS = @LTLIBOBJS@ +MAINT = @MAINT@ +MAINTAINER_MODE_FALSE = @MAINTAINER_MODE_FALSE@ +MAINTAINER_MODE_TRUE = @MAINTAINER_MODE_TRUE@ +MAKEINFO = @MAKEINFO@ +MAKE_TARGETS_IN_VPATH_FALSE = @MAKE_TARGETS_IN_VPATH_FALSE@ +MAKE_TARGETS_IN_VPATH_TRUE = @MAKE_TARGETS_IN_VPATH_TRUE@ +MKINSTALLDIRS = @MKINSTALLDIRS@ +MKTEMP = @MKTEMP@ +MSGFMT = @MSGFMT@ +MSGMERGE = @MSGMERGE@ +OBJEXT = @OBJEXT@ +PACKAGE = @PACKAGE@ +PACKAGE_BUGREPORT = @PACKAGE_BUGREPORT@ +PACKAGE_NAME = @PACKAGE_NAME@ +PACKAGE_STRING = @PACKAGE_STRING@ +PACKAGE_TARNAME = @PACKAGE_TARNAME@ +PACKAGE_VERSION = @PACKAGE_VERSION@ +PATH_SEPARATOR = @PATH_SEPARATOR@ +PERL = @PERL@ +POSUB = @POSUB@ +PR = @PR@ +PS2PDF = @PS2PDF@ +RANLIB = @RANLIB@ +ROFF = @ROFF@ +RSH_DFLT = @RSH_DFLT@ +SENDMAIL = @SENDMAIL@ +SET_MAKE = @SET_MAKE@ +SHELL = @SHELL@ +STDBOOL_H = @STDBOOL_H@ +STDINT_H = @STDINT_H@ +STRIP = @STRIP@ +TEXI2DVI = @TEXI2DVI@ +USE_NLS = @USE_NLS@ +VERSION = @VERSION@ +XGETTEXT = @XGETTEXT@ +YACC = @YACC@ +YFLAGS = @YFLAGS@ +ZLIB_CPPFLAGS = @ZLIB_CPPFLAGS@ +ZLIB_LIBS = @ZLIB_LIBS@ +ZLIB_SUBDIRS = @ZLIB_SUBDIRS@ +ac_ct_CC = @ac_ct_CC@ +ac_ct_RANLIB = @ac_ct_RANLIB@ +ac_ct_STRIP = @ac_ct_STRIP@ +ac_prefix_program = @ac_prefix_program@ +am__fastdepCC_FALSE = @am__fastdepCC_FALSE@ +am__fastdepCC_TRUE = @am__fastdepCC_TRUE@ +am__include = @am__include@ +am__leading_dot = @am__leading_dot@ +am__quote = @am__quote@ +am__tar = @am__tar@ +am__untar = @am__untar@ +bindir = @bindir@ +build = @build@ +build_alias = @build_alias@ +build_cpu = @build_cpu@ +build_os = @build_os@ +build_vendor = @build_vendor@ +cvs_client_objects = @cvs_client_objects@ +datadir = @datadir@ +exec_prefix = @exec_prefix@ +host = @host@ +host_alias = @host_alias@ +host_cpu = @host_cpu@ +host_os = @host_os@ +host_vendor = @host_vendor@ +includedir = @includedir@ +infodir = @infodir@ +install_sh = @install_sh@ +libdir = @libdir@ +libexecdir = @libexecdir@ +localstatedir = @localstatedir@ +mandir = @mandir@ +mkdir_p = @mkdir_p@ +oldincludedir = @oldincludedir@ +prefix = @prefix@ +program_transform_name = @program_transform_name@ +sbindir = @sbindir@ +sharedstatedir = @sharedstatedir@ +sysconfdir = @sysconfdir@ +target_alias = @target_alias@ +with_default_rsh = @with_default_rsh@ +noinst_LIBRARIES = libcvs.a + +# Should look into unifying regular expression matching in CVS +# with the diff library (perhaps to have the caller, CVS, do the +# matching?) +libcvs_a_SOURCES = sighandle.c system.h wait.h xselect.h xtime.h \ + allocsa.h allocsa.c cycle-check.c cycle-check.h dev-ino.h \ + basename.c stripslash.c error.h exit.h getaddrinfo.h \ + getnline.h getnline.c gettext.h mbuiter.h minmax.h setenv.h \ + size_max.h strcase.h strnlen1.h strnlen1.c strstr.h time_r.h \ + printf-args.h printf-parse.h vasnprintf.h vasprintf.h \ + xalloc-die.c xgethostname.h xgethostname.c xreadlink.h \ + xreadlink.c xsize.h +libcvs_a_LIBADD = @LIBOBJS@ @ALLOCA@ +BUILT_SOURCES = $(ALLOCA_H) $(FNMATCH_H) getdate.c $(GETOPT_H) \ + $(GLOB_H) $(STDBOOL_H) $(STDINT_H) +MOSTLYCLEANFILES = alloca.h alloca.h-t fnmatch.h fnmatch.h-t \ + getdate-expected getdate-got getdate.diff getopt.h getopt.h-t \ + glob.h glob.h-t stdbool.h stdbool.h-t stdint.h stdint.h-t +DISTCLEANFILES = getdate.log +MAINTAINERCLEANFILES = getdate.c +TESTS = test-getdate.sh +EXTRA_DIST = .cvsignore ChangeLog.fsf Makefile.gnulib build_lib.com \ + libcvs.dep libcvs.dsp libcvs.mak test-getdate.sh alloca_.h \ + allocsa.valgrind fnmatch_.h fnmatch_loop.c getdate.c \ + getndelim2.h getndelim2.c getopt_.h getopt_int.h stdbool_.h \ + stdint_.h +getdate_SOURCES = error.c getdate.y progname.c +getdate_CPPFLAGS = -DTEST +getdate_LDADD = \ + $(noinst_LIBRARIES) \ + $(LIB_CLOCK_GETTIME) \ + $(LIBINTL) + +all: $(BUILT_SOURCES) + $(MAKE) $(AM_MAKEFLAGS) all-am + +.SUFFIXES: +.SUFFIXES: .c .o .obj .y +$(srcdir)/Makefile.in: @MAINTAINER_MODE_TRUE@ $(srcdir)/Makefile.am $(am__configure_deps) + @for dep in $?; do \ + case '$(am__configure_deps)' in \ + *$$dep*) \ + cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh \ + && exit 0; \ + exit 1;; \ + esac; \ + done; \ + echo ' cd $(top_srcdir) && $(AUTOMAKE) --gnu lib/Makefile'; \ + cd $(top_srcdir) && \ + $(AUTOMAKE) --gnu lib/Makefile +.PRECIOUS: Makefile +Makefile: $(srcdir)/Makefile.in $(top_builddir)/config.status + @case '$?' in \ + *config.status*) \ + cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh;; \ + *) \ + echo ' cd $(top_builddir) && $(SHELL) ./config.status $(subdir)/$@ $(am__depfiles_maybe)'; \ + cd $(top_builddir) && $(SHELL) ./config.status $(subdir)/$@ $(am__depfiles_maybe);; \ + esac; + +$(top_builddir)/config.status: $(top_srcdir)/configure $(CONFIG_STATUS_DEPENDENCIES) + cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh + +$(top_srcdir)/configure: @MAINTAINER_MODE_TRUE@ $(am__configure_deps) + cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh +$(ACLOCAL_M4): @MAINTAINER_MODE_TRUE@ $(am__aclocal_m4_deps) + cd $(top_builddir) && $(MAKE) $(AM_MAKEFLAGS) am--refresh + +clean-noinstLIBRARIES: + -test -z "$(noinst_LIBRARIES)" || rm -f $(noinst_LIBRARIES) +libcvs.a: $(libcvs_a_OBJECTS) $(libcvs_a_DEPENDENCIES) + -rm -f libcvs.a + $(libcvs_a_AR) libcvs.a $(libcvs_a_OBJECTS) $(libcvs_a_LIBADD) + $(RANLIB) libcvs.a + +clean-checkPROGRAMS: + -test -z "$(check_PROGRAMS)" || rm -f $(check_PROGRAMS) +getdate$(EXEEXT): $(getdate_OBJECTS) $(getdate_DEPENDENCIES) + @rm -f getdate$(EXEEXT) + $(LINK) $(getdate_LDFLAGS) $(getdate_OBJECTS) $(getdate_LDADD) $(LIBS) + +mostlyclean-compile: + -rm -f *.$(OBJEXT) + +distclean-compile: + -rm -f *.tab.c + +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/__fpending.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/alloca.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/asnprintf.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/asprintf.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/atexit.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/canon-host.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/canonicalize.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/chdir-long.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/closeout.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/dirname.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/dup-safer.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/dup2.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/exitfail.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/fd-safer.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/filenamecat.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/fncase.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/fnmatch.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/fseeko.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/ftello.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/ftruncate.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/gai_strerror.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/getaddrinfo.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/getcwd.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/getdate.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/getdate.y@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/getdelim.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/gethostname.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/getline.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/getlogin_r.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/getndelim2.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/getopt.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/getopt1.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/getpass.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/gettime.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/gettimeofday.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/glob.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/lstat.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/malloc.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/mbchar.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/md5.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/memchr.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/memmove.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/mempcpy.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/memrchr.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/mkdir.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/mkstemp.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/mktime.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/nanosleep.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/openat.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/pagealign_alloc.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/pipe-safer.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/printf-args.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/printf-parse.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/quotearg.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/readlink.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/realloc.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/regcomp.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/regex.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/regex_internal.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/regexec.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/rename.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/rpmatch.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/save-cwd.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/setenv.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/strcasecmp.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/strdup.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/strerror.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/strftime.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/strncasecmp.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/strstr.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/strtol.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/strtoul.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/sunos57-select.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/tempname.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/time_r.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/unsetenv.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/vasnprintf.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/vasprintf.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/waitpid.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/xgetcwd.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/xmalloc.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@$(DEPDIR)/yesno.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/allocsa.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/basename.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/cycle-check.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/getdate-error.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/getdate-getdate.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/getdate-progname.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/getnline.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/sighandle.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/stripslash.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/strnlen1.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xalloc-die.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xgethostname.Po@am__quote@ +@AMDEP_TRUE@@am__include@ @am__quote@./$(DEPDIR)/xreadlink.Po@am__quote@ + +.c.o: +@am__fastdepCC_TRUE@ if $(COMPILE) -MT $@ -MD -MP -MF "$(DEPDIR)/$*.Tpo" -c -o $@ $<; \ +@am__fastdepCC_TRUE@ then mv -f "$(DEPDIR)/$*.Tpo" "$(DEPDIR)/$*.Po"; else rm -f "$(DEPDIR)/$*.Tpo"; exit 1; fi +@AMDEP_TRUE@@am__fastdepCC_FALSE@ source='$<' object='$@' libtool=no @AMDEPBACKSLASH@ +@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@ +@am__fastdepCC_FALSE@ $(COMPILE) -c $< + +.c.obj: +@am__fastdepCC_TRUE@ if $(COMPILE) -MT $@ -MD -MP -MF "$(DEPDIR)/$*.Tpo" -c -o $@ `$(CYGPATH_W) '$<'`; \ +@am__fastdepCC_TRUE@ then mv -f "$(DEPDIR)/$*.Tpo" "$(DEPDIR)/$*.Po"; else rm -f "$(DEPDIR)/$*.Tpo"; exit 1; fi +@AMDEP_TRUE@@am__fastdepCC_FALSE@ source='$<' object='$@' libtool=no @AMDEPBACKSLASH@ +@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@ +@am__fastdepCC_FALSE@ $(COMPILE) -c `$(CYGPATH_W) '$<'` + +getdate-error.o: error.c +@am__fastdepCC_TRUE@ if $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(getdate_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT getdate-error.o -MD -MP -MF "$(DEPDIR)/getdate-error.Tpo" -c -o getdate-error.o `test -f 'error.c' || echo '$(srcdir)/'`error.c; \ +@am__fastdepCC_TRUE@ then mv -f "$(DEPDIR)/getdate-error.Tpo" "$(DEPDIR)/getdate-error.Po"; else rm -f "$(DEPDIR)/getdate-error.Tpo"; exit 1; fi +@AMDEP_TRUE@@am__fastdepCC_FALSE@ source='error.c' object='getdate-error.o' libtool=no @AMDEPBACKSLASH@ +@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@ +@am__fastdepCC_FALSE@ $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(getdate_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o getdate-error.o `test -f 'error.c' || echo '$(srcdir)/'`error.c + +getdate-error.obj: error.c +@am__fastdepCC_TRUE@ if $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(getdate_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT getdate-error.obj -MD -MP -MF "$(DEPDIR)/getdate-error.Tpo" -c -o getdate-error.obj `if test -f 'error.c'; then $(CYGPATH_W) 'error.c'; else $(CYGPATH_W) '$(srcdir)/error.c'; fi`; \ +@am__fastdepCC_TRUE@ then mv -f "$(DEPDIR)/getdate-error.Tpo" "$(DEPDIR)/getdate-error.Po"; else rm -f "$(DEPDIR)/getdate-error.Tpo"; exit 1; fi +@AMDEP_TRUE@@am__fastdepCC_FALSE@ source='error.c' object='getdate-error.obj' libtool=no @AMDEPBACKSLASH@ +@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@ +@am__fastdepCC_FALSE@ $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(getdate_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o getdate-error.obj `if test -f 'error.c'; then $(CYGPATH_W) 'error.c'; else $(CYGPATH_W) '$(srcdir)/error.c'; fi` + +getdate-getdate.o: getdate.c +@am__fastdepCC_TRUE@ if $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(getdate_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT getdate-getdate.o -MD -MP -MF "$(DEPDIR)/getdate-getdate.Tpo" -c -o getdate-getdate.o `test -f 'getdate.c' || echo '$(srcdir)/'`getdate.c; \ +@am__fastdepCC_TRUE@ then mv -f "$(DEPDIR)/getdate-getdate.Tpo" "$(DEPDIR)/getdate-getdate.Po"; else rm -f "$(DEPDIR)/getdate-getdate.Tpo"; exit 1; fi +@AMDEP_TRUE@@am__fastdepCC_FALSE@ source='getdate.c' object='getdate-getdate.o' libtool=no @AMDEPBACKSLASH@ +@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@ +@am__fastdepCC_FALSE@ $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(getdate_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o getdate-getdate.o `test -f 'getdate.c' || echo '$(srcdir)/'`getdate.c + +getdate-getdate.obj: getdate.c +@am__fastdepCC_TRUE@ if $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(getdate_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT getdate-getdate.obj -MD -MP -MF "$(DEPDIR)/getdate-getdate.Tpo" -c -o getdate-getdate.obj `if test -f 'getdate.c'; then $(CYGPATH_W) 'getdate.c'; else $(CYGPATH_W) '$(srcdir)/getdate.c'; fi`; \ +@am__fastdepCC_TRUE@ then mv -f "$(DEPDIR)/getdate-getdate.Tpo" "$(DEPDIR)/getdate-getdate.Po"; else rm -f "$(DEPDIR)/getdate-getdate.Tpo"; exit 1; fi +@AMDEP_TRUE@@am__fastdepCC_FALSE@ source='getdate.c' object='getdate-getdate.obj' libtool=no @AMDEPBACKSLASH@ +@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@ +@am__fastdepCC_FALSE@ $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(getdate_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o getdate-getdate.obj `if test -f 'getdate.c'; then $(CYGPATH_W) 'getdate.c'; else $(CYGPATH_W) '$(srcdir)/getdate.c'; fi` + +getdate-progname.o: progname.c +@am__fastdepCC_TRUE@ if $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(getdate_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT getdate-progname.o -MD -MP -MF "$(DEPDIR)/getdate-progname.Tpo" -c -o getdate-progname.o `test -f 'progname.c' || echo '$(srcdir)/'`progname.c; \ +@am__fastdepCC_TRUE@ then mv -f "$(DEPDIR)/getdate-progname.Tpo" "$(DEPDIR)/getdate-progname.Po"; else rm -f "$(DEPDIR)/getdate-progname.Tpo"; exit 1; fi +@AMDEP_TRUE@@am__fastdepCC_FALSE@ source='progname.c' object='getdate-progname.o' libtool=no @AMDEPBACKSLASH@ +@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@ +@am__fastdepCC_FALSE@ $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(getdate_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o getdate-progname.o `test -f 'progname.c' || echo '$(srcdir)/'`progname.c + +getdate-progname.obj: progname.c +@am__fastdepCC_TRUE@ if $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(getdate_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -MT getdate-progname.obj -MD -MP -MF "$(DEPDIR)/getdate-progname.Tpo" -c -o getdate-progname.obj `if test -f 'progname.c'; then $(CYGPATH_W) 'progname.c'; else $(CYGPATH_W) '$(srcdir)/progname.c'; fi`; \ +@am__fastdepCC_TRUE@ then mv -f "$(DEPDIR)/getdate-progname.Tpo" "$(DEPDIR)/getdate-progname.Po"; else rm -f "$(DEPDIR)/getdate-progname.Tpo"; exit 1; fi +@AMDEP_TRUE@@am__fastdepCC_FALSE@ source='progname.c' object='getdate-progname.obj' libtool=no @AMDEPBACKSLASH@ +@AMDEP_TRUE@@am__fastdepCC_FALSE@ DEPDIR=$(DEPDIR) $(CCDEPMODE) $(depcomp) @AMDEPBACKSLASH@ +@am__fastdepCC_FALSE@ $(CC) $(DEFS) $(DEFAULT_INCLUDES) $(INCLUDES) $(getdate_CPPFLAGS) $(CPPFLAGS) $(AM_CFLAGS) $(CFLAGS) -c -o getdate-progname.obj `if test -f 'progname.c'; then $(CYGPATH_W) 'progname.c'; else $(CYGPATH_W) '$(srcdir)/progname.c'; fi` + +.y.c: + $(SHELL) $(YLWRAP) $< y.tab.c $@ y.tab.h $*.h y.output $*.output -- $(YACCCOMPILE) +uninstall-info-am: + +ID: $(HEADERS) $(SOURCES) $(LISP) $(TAGS_FILES) + list='$(SOURCES) $(HEADERS) $(LISP) $(TAGS_FILES)'; \ + unique=`for i in $$list; do \ + if test -f "$$i"; then echo $$i; else echo $(srcdir)/$$i; fi; \ + done | \ + $(AWK) ' { files[$$0] = 1; } \ + END { for (i in files) print i; }'`; \ + mkid -fID $$unique +tags: TAGS + +TAGS: $(HEADERS) $(SOURCES) $(TAGS_DEPENDENCIES) \ + $(TAGS_FILES) $(LISP) + tags=; \ + here=`pwd`; \ + list='$(SOURCES) $(HEADERS) $(LISP) $(TAGS_FILES)'; \ + unique=`for i in $$list; do \ + if test -f "$$i"; then echo $$i; else echo $(srcdir)/$$i; fi; \ + done | \ + $(AWK) ' { files[$$0] = 1; } \ + END { for (i in files) print i; }'`; \ + if test -z "$(ETAGS_ARGS)$$tags$$unique"; then :; else \ + test -n "$$unique" || unique=$$empty_fix; \ + $(ETAGS) $(ETAGSFLAGS) $(AM_ETAGSFLAGS) $(ETAGS_ARGS) \ + $$tags $$unique; \ + fi +ctags: CTAGS +CTAGS: $(HEADERS) $(SOURCES) $(TAGS_DEPENDENCIES) \ + $(TAGS_FILES) $(LISP) + tags=; \ + here=`pwd`; \ + list='$(SOURCES) $(HEADERS) $(LISP) $(TAGS_FILES)'; \ + unique=`for i in $$list; do \ + if test -f "$$i"; then echo $$i; else echo $(srcdir)/$$i; fi; \ + done | \ + $(AWK) ' { files[$$0] = 1; } \ + END { for (i in files) print i; }'`; \ + test -z "$(CTAGS_ARGS)$$tags$$unique" \ + || $(CTAGS) $(CTAGSFLAGS) $(AM_CTAGSFLAGS) $(CTAGS_ARGS) \ + $$tags $$unique + +GTAGS: + here=`$(am__cd) $(top_builddir) && pwd` \ + && cd $(top_srcdir) \ + && gtags -i $(GTAGS_ARGS) $$here + +distclean-tags: + -rm -f TAGS ID GTAGS GRTAGS GSYMS GPATH tags + +check-TESTS: $(TESTS) + @failed=0; all=0; xfail=0; xpass=0; skip=0; \ + srcdir=$(srcdir); export srcdir; \ + list='$(TESTS)'; \ + if test -n "$$list"; then \ + for tst in $$list; do \ + if test -f ./$$tst; then dir=./; \ + elif test -f $$tst; then dir=; \ + else dir="$(srcdir)/"; fi; \ + if $(TESTS_ENVIRONMENT) $${dir}$$tst; then \ + all=`expr $$all + 1`; \ + case " $(XFAIL_TESTS) " in \ + *" $$tst "*) \ + xpass=`expr $$xpass + 1`; \ + failed=`expr $$failed + 1`; \ + echo "XPASS: $$tst"; \ + ;; \ + *) \ + echo "PASS: $$tst"; \ + ;; \ + esac; \ + elif test $$? -ne 77; then \ + all=`expr $$all + 1`; \ + case " $(XFAIL_TESTS) " in \ + *" $$tst "*) \ + xfail=`expr $$xfail + 1`; \ + echo "XFAIL: $$tst"; \ + ;; \ + *) \ + failed=`expr $$failed + 1`; \ + echo "FAIL: $$tst"; \ + ;; \ + esac; \ + else \ + skip=`expr $$skip + 1`; \ + echo "SKIP: $$tst"; \ + fi; \ + done; \ + if test "$$failed" -eq 0; then \ + if test "$$xfail" -eq 0; then \ + banner="All $$all tests passed"; \ + else \ + banner="All $$all tests behaved as expected ($$xfail expected failures)"; \ + fi; \ + else \ + if test "$$xpass" -eq 0; then \ + banner="$$failed of $$all tests failed"; \ + else \ + banner="$$failed of $$all tests did not behave as expected ($$xpass unexpected passes)"; \ + fi; \ + fi; \ + dashes="$$banner"; \ + skipped=""; \ + if test "$$skip" -ne 0; then \ + skipped="($$skip tests were not run)"; \ + test `echo "$$skipped" | wc -c` -le `echo "$$banner" | wc -c` || \ + dashes="$$skipped"; \ + fi; \ + report=""; \ + if test "$$failed" -ne 0 && test -n "$(PACKAGE_BUGREPORT)"; then \ + report="Please report to $(PACKAGE_BUGREPORT)"; \ + test `echo "$$report" | wc -c` -le `echo "$$banner" | wc -c` || \ + dashes="$$report"; \ + fi; \ + dashes=`echo "$$dashes" | sed s/./=/g`; \ + echo "$$dashes"; \ + echo "$$banner"; \ + test -z "$$skipped" || echo "$$skipped"; \ + test -z "$$report" || echo "$$report"; \ + echo "$$dashes"; \ + test "$$failed" -eq 0; \ + else :; fi + +distdir: $(DISTFILES) + @srcdirstrip=`echo "$(srcdir)" | sed 's|.|.|g'`; \ + topsrcdirstrip=`echo "$(top_srcdir)" | sed 's|.|.|g'`; \ + list='$(DISTFILES)'; for file in $$list; do \ + case $$file in \ + $(srcdir)/*) file=`echo "$$file" | sed "s|^$$srcdirstrip/||"`;; \ + $(top_srcdir)/*) file=`echo "$$file" | sed "s|^$$topsrcdirstrip/|$(top_builddir)/|"`;; \ + esac; \ + if test -f $$file || test -d $$file; then d=.; else d=$(srcdir); fi; \ + dir=`echo "$$file" | sed -e 's,/[^/]*$$,,'`; \ + if test "$$dir" != "$$file" && test "$$dir" != "."; then \ + dir="/$$dir"; \ + $(mkdir_p) "$(distdir)$$dir"; \ + else \ + dir=''; \ + fi; \ + if test -d $$d/$$file; then \ + if test -d $(srcdir)/$$file && test $$d != $(srcdir); then \ + cp -pR $(srcdir)/$$file $(distdir)$$dir || exit 1; \ + fi; \ + cp -pR $$d/$$file $(distdir)$$dir || exit 1; \ + else \ + test -f $(distdir)/$$file \ + || cp -p $$d/$$file $(distdir)/$$file \ + || exit 1; \ + fi; \ + done +check-am: all-am + $(MAKE) $(AM_MAKEFLAGS) $(check_PROGRAMS) + $(MAKE) $(AM_MAKEFLAGS) check-TESTS +check: $(BUILT_SOURCES) + $(MAKE) $(AM_MAKEFLAGS) check-am +all-am: Makefile $(LIBRARIES) +installdirs: +install: $(BUILT_SOURCES) + $(MAKE) $(AM_MAKEFLAGS) install-am +install-exec: install-exec-am +install-data: install-data-am +uninstall: uninstall-am + +install-am: all-am + @$(MAKE) $(AM_MAKEFLAGS) install-exec-am install-data-am + +installcheck: installcheck-am +install-strip: + $(MAKE) $(AM_MAKEFLAGS) INSTALL_PROGRAM="$(INSTALL_STRIP_PROGRAM)" \ + install_sh_PROGRAM="$(INSTALL_STRIP_PROGRAM)" INSTALL_STRIP_FLAG=-s \ + `test -z '$(STRIP)' || \ + echo "INSTALL_PROGRAM_ENV=STRIPPROG='$(STRIP)'"` install +mostlyclean-generic: + -test -z "$(MOSTLYCLEANFILES)" || rm -f $(MOSTLYCLEANFILES) + +clean-generic: + +distclean-generic: + -test -z "$(CONFIG_CLEAN_FILES)" || rm -f $(CONFIG_CLEAN_FILES) + -test -z "$(DISTCLEANFILES)" || rm -f $(DISTCLEANFILES) + +maintainer-clean-generic: + @echo "This command is intended for maintainers to use" + @echo "it deletes files that may require special tools to rebuild." + -rm -f getdate.c + -test -z "$(BUILT_SOURCES)" || rm -f $(BUILT_SOURCES) + -test -z "$(MAINTAINERCLEANFILES)" || rm -f $(MAINTAINERCLEANFILES) +clean: clean-am + +clean-am: clean-checkPROGRAMS clean-generic clean-noinstLIBRARIES \ + mostlyclean-am + +distclean: distclean-am + -rm -rf $(DEPDIR) ./$(DEPDIR) + -rm -f Makefile +distclean-am: clean-am distclean-compile distclean-generic \ + distclean-local distclean-tags + +dvi: dvi-am + +dvi-am: + +html: html-am + +info: info-am + +info-am: + +install-data-am: + +install-exec-am: + +install-info: install-info-am + +install-man: + +installcheck-am: + +maintainer-clean: maintainer-clean-am + -rm -rf $(DEPDIR) ./$(DEPDIR) + -rm -f Makefile +maintainer-clean-am: distclean-am maintainer-clean-generic + +mostlyclean: mostlyclean-am + +mostlyclean-am: mostlyclean-compile mostlyclean-generic + +pdf: pdf-am + +pdf-am: + +ps: ps-am + +ps-am: + +uninstall-am: uninstall-info-am + +.PHONY: CTAGS GTAGS all all-am check check-TESTS check-am clean \ + clean-checkPROGRAMS clean-generic clean-noinstLIBRARIES ctags \ + distclean distclean-compile distclean-generic distclean-local \ + distclean-tags distdir dvi dvi-am html html-am info info-am \ + install install-am install-data install-data-am install-exec \ + install-exec-am install-info install-info-am install-man \ + install-strip installcheck installcheck-am installdirs \ + maintainer-clean maintainer-clean-generic mostlyclean \ + mostlyclean-compile mostlyclean-generic pdf pdf-am ps ps-am \ + tags uninstall uninstall-am uninstall-info-am + + +# We need the following in order to create <alloca.h> when the system +# doesn't have one that works with the given compiler. +alloca.h: alloca_.h + cp $(srcdir)/alloca_.h $@-t + mv $@-t $@ + +# We need the following in order to create <fnmatch.h> when the system +# doesn't have one that supports the required API. +fnmatch.h: fnmatch_.h + cp $(srcdir)/fnmatch_.h $@-t + mv $@-t $@ + +# We need the following in order to create <getopt.h> when the system +# doesn't have one that works with the given compiler. +getopt.h: getopt_.h + cp $(srcdir)/getopt_.h $@-t + mv $@-t $@ + +# We need the following in order to create an <getopt.h> when the system +# doesn't have one that works with the given compiler. +all-local $(libcvs_a_OBJECTS): $(GLOB_H) +glob.h: glob_.h + cp $(srcdir)/glob_.h $@-t + mv $@-t $@ + +# We need the following in order to create <stdbool.h> when the system +# doesn't have one that works. +stdbool.h: stdbool_.h + sed -e 's/@''HAVE__BOOL''@/$(HAVE__BOOL)/g' \ + < $(srcdir)/stdbool_.h > $@-t + mv $@-t $@ + +# We need the following in order to create <stdint.h> when the system +# doesn't have one that works with the given compiler. +stdint.h: stdint_.h + sed -e 's/@''HAVE_LONG_64BIT''@/$(HAVE_LONG_64BIT)/g;s/@''HAVE_LONG_LONG_64BIT@/$(HAVE_LONG_LONG_64BIT)/g' < $(srcdir)/stdint_.h > $@-t + mv $@-t $@ + +# Until Automake gets its act together +distclean-local: + rm -f fnmatch.h + +# for backwards compatibility with the old makefiles +realclean: maintainer-clean +.PHONY: realclean +# Tell versions [3.59,3.63) of GNU make to not export all variables. +# Otherwise a system limit (for SysV at least) may be exceeded. +.NOEXPORT: diff --git a/lib/README b/lib/README new file mode 100644 index 0000000..ce1cc7f --- /dev/null +++ b/lib/README @@ -0,0 +1,9 @@ +Many of the source files in this directory come from the GNULIB project +<http://savannah.gnu.org/projects/gnulib/>/<mailto:bug-gnulib@gnu.org>, and, +if they don't, they should. + +What this means is that there is often at least one corresponding Autoconf +macro in the CVS/m4 directory and that bug fixes and enhancements to this code +should be sent to the GNULIB project and then reimported here after the GNULIB +developers approve and adopt the change. Changes should not be made locally +without good reason! diff --git a/lib/__fpending.c b/lib/__fpending.c new file mode 100644 index 0000000..63fb7d7 --- /dev/null +++ b/lib/__fpending.c @@ -0,0 +1,32 @@ +/* __fpending.c -- return the number of pending output bytes on a stream + Copyright (C) 2000, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Jim Meyering. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "__fpending.h" + +/* Return the number of pending (aka buffered, unflushed) + bytes on the stream, FP, that is open for writing. */ +size_t +__fpending (FILE *fp) +{ + return PENDING_OUTPUT_N_BYTES; +} diff --git a/lib/__fpending.h b/lib/__fpending.h new file mode 100644 index 0000000..36a842e --- /dev/null +++ b/lib/__fpending.h @@ -0,0 +1,13 @@ +#include <stddef.h> +#include <stdio.h> + +#if HAVE_STDIO_EXT_H +# include <stdio_ext.h> +#endif + +#ifndef HAVE_DECL___FPENDING +"this configure-time declaration test was not run" +#endif +#if !HAVE_DECL___FPENDING +size_t __fpending (FILE *); +#endif diff --git a/lib/alloca.c b/lib/alloca.c new file mode 100644 index 0000000..d1d5447 --- /dev/null +++ b/lib/alloca.c @@ -0,0 +1,491 @@ +/* alloca.c -- allocate automatically reclaimed memory + (Mostly) portable public-domain implementation -- D A Gwyn + + This implementation of the PWB library alloca function, + which is used to allocate space off the run-time stack so + that it is automatically reclaimed upon procedure exit, + was inspired by discussions with J. Q. Johnson of Cornell. + J.Otto Tennant <jot@cray.com> contributed the Cray support. + + There are some preprocessor constants that can + be defined when compiling for your specific system, for + improved efficiency; however, the defaults should be okay. + + The general concept of this implementation is to keep + track of all alloca-allocated blocks, and reclaim any + that are found to be deeper in the stack than the current + invocation. This heuristic does not reclaim storage as + soon as it becomes invalid, but it will do so eventually. + + As a special case, alloca(0) reclaims storage without + allocating any. It is a good idea to use alloca(0) in + your main control loop, etc. to force garbage collection. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include <alloca.h> + +#include <string.h> +#include <stdlib.h> + +#ifdef emacs +# include "lisp.h" +# include "blockinput.h" +# ifdef EMACS_FREE +# undef free +# define free EMACS_FREE +# endif +#else +# define memory_full() abort () +#endif + +/* If compiling with GCC 2, this file's not needed. */ +#if !defined (__GNUC__) || __GNUC__ < 2 + +/* If someone has defined alloca as a macro, + there must be some other way alloca is supposed to work. */ +# ifndef alloca + +# ifdef emacs +# ifdef static +/* actually, only want this if static is defined as "" + -- this is for usg, in which emacs must undefine static + in order to make unexec workable + */ +# ifndef STACK_DIRECTION +you +lose +-- must know STACK_DIRECTION at compile-time +/* Using #error here is not wise since this file should work for + old and obscure compilers. */ +# endif /* STACK_DIRECTION undefined */ +# endif /* static */ +# endif /* emacs */ + +/* If your stack is a linked list of frames, you have to + provide an "address metric" ADDRESS_FUNCTION macro. */ + +# if defined (CRAY) && defined (CRAY_STACKSEG_END) +long i00afunc (); +# define ADDRESS_FUNCTION(arg) (char *) i00afunc (&(arg)) +# else +# define ADDRESS_FUNCTION(arg) &(arg) +# endif + +/* Define STACK_DIRECTION if you know the direction of stack + growth for your system; otherwise it will be automatically + deduced at run-time. + + STACK_DIRECTION > 0 => grows toward higher addresses + STACK_DIRECTION < 0 => grows toward lower addresses + STACK_DIRECTION = 0 => direction of growth unknown */ + +# ifndef STACK_DIRECTION +# define STACK_DIRECTION 0 /* Direction unknown. */ +# endif + +# if STACK_DIRECTION != 0 + +# define STACK_DIR STACK_DIRECTION /* Known at compile-time. */ + +# else /* STACK_DIRECTION == 0; need run-time code. */ + +static int stack_dir; /* 1 or -1 once known. */ +# define STACK_DIR stack_dir + +static void +find_stack_direction (void) +{ + static char *addr = NULL; /* Address of first `dummy', once known. */ + auto char dummy; /* To get stack address. */ + + if (addr == NULL) + { /* Initial entry. */ + addr = ADDRESS_FUNCTION (dummy); + + find_stack_direction (); /* Recurse once. */ + } + else + { + /* Second entry. */ + if (ADDRESS_FUNCTION (dummy) > addr) + stack_dir = 1; /* Stack grew upward. */ + else + stack_dir = -1; /* Stack grew downward. */ + } +} + +# endif /* STACK_DIRECTION == 0 */ + +/* An "alloca header" is used to: + (a) chain together all alloca'ed blocks; + (b) keep track of stack depth. + + It is very important that sizeof(header) agree with malloc + alignment chunk size. The following default should work okay. */ + +# ifndef ALIGN_SIZE +# define ALIGN_SIZE sizeof(double) +# endif + +typedef union hdr +{ + char align[ALIGN_SIZE]; /* To force sizeof(header). */ + struct + { + union hdr *next; /* For chaining headers. */ + char *deep; /* For stack depth measure. */ + } h; +} header; + +static header *last_alloca_header = NULL; /* -> last alloca header. */ + +/* Return a pointer to at least SIZE bytes of storage, + which will be automatically reclaimed upon exit from + the procedure that called alloca. Originally, this space + was supposed to be taken from the current stack frame of the + caller, but that method cannot be made to work for some + implementations of C, for example under Gould's UTX/32. */ + +void * +alloca (size_t size) +{ + auto char probe; /* Probes stack depth: */ + register char *depth = ADDRESS_FUNCTION (probe); + +# if STACK_DIRECTION == 0 + if (STACK_DIR == 0) /* Unknown growth direction. */ + find_stack_direction (); +# endif + + /* Reclaim garbage, defined as all alloca'd storage that + was allocated from deeper in the stack than currently. */ + + { + register header *hp; /* Traverses linked list. */ + +# ifdef emacs + BLOCK_INPUT; +# endif + + for (hp = last_alloca_header; hp != NULL;) + if ((STACK_DIR > 0 && hp->h.deep > depth) + || (STACK_DIR < 0 && hp->h.deep < depth)) + { + register header *np = hp->h.next; + + free (hp); /* Collect garbage. */ + + hp = np; /* -> next header. */ + } + else + break; /* Rest are not deeper. */ + + last_alloca_header = hp; /* -> last valid storage. */ + +# ifdef emacs + UNBLOCK_INPUT; +# endif + } + + if (size == 0) + return NULL; /* No allocation required. */ + + /* Allocate combined header + user data storage. */ + + { + /* Address of header. */ + register header *new; + + size_t combined_size = sizeof (header) + size; + if (combined_size < sizeof (header)) + memory_full (); + + new = malloc (combined_size); + + if (! new) + memory_full (); + + new->h.next = last_alloca_header; + new->h.deep = depth; + + last_alloca_header = new; + + /* User storage begins just after header. */ + + return (void *) (new + 1); + } +} + +# if defined (CRAY) && defined (CRAY_STACKSEG_END) + +# ifdef DEBUG_I00AFUNC +# include <stdio.h> +# endif + +# ifndef CRAY_STACK +# define CRAY_STACK +# ifndef CRAY2 +/* Stack structures for CRAY-1, CRAY X-MP, and CRAY Y-MP */ +struct stack_control_header + { + long shgrow:32; /* Number of times stack has grown. */ + long shaseg:32; /* Size of increments to stack. */ + long shhwm:32; /* High water mark of stack. */ + long shsize:32; /* Current size of stack (all segments). */ + }; + +/* The stack segment linkage control information occurs at + the high-address end of a stack segment. (The stack + grows from low addresses to high addresses.) The initial + part of the stack segment linkage control information is + 0200 (octal) words. This provides for register storage + for the routine which overflows the stack. */ + +struct stack_segment_linkage + { + long ss[0200]; /* 0200 overflow words. */ + long sssize:32; /* Number of words in this segment. */ + long ssbase:32; /* Offset to stack base. */ + long:32; + long sspseg:32; /* Offset to linkage control of previous + segment of stack. */ + long:32; + long sstcpt:32; /* Pointer to task common address block. */ + long sscsnm; /* Private control structure number for + microtasking. */ + long ssusr1; /* Reserved for user. */ + long ssusr2; /* Reserved for user. */ + long sstpid; /* Process ID for pid based multi-tasking. */ + long ssgvup; /* Pointer to multitasking thread giveup. */ + long sscray[7]; /* Reserved for Cray Research. */ + long ssa0; + long ssa1; + long ssa2; + long ssa3; + long ssa4; + long ssa5; + long ssa6; + long ssa7; + long sss0; + long sss1; + long sss2; + long sss3; + long sss4; + long sss5; + long sss6; + long sss7; + }; + +# else /* CRAY2 */ +/* The following structure defines the vector of words + returned by the STKSTAT library routine. */ +struct stk_stat + { + long now; /* Current total stack size. */ + long maxc; /* Amount of contiguous space which would + be required to satisfy the maximum + stack demand to date. */ + long high_water; /* Stack high-water mark. */ + long overflows; /* Number of stack overflow ($STKOFEN) calls. */ + long hits; /* Number of internal buffer hits. */ + long extends; /* Number of block extensions. */ + long stko_mallocs; /* Block allocations by $STKOFEN. */ + long underflows; /* Number of stack underflow calls ($STKRETN). */ + long stko_free; /* Number of deallocations by $STKRETN. */ + long stkm_free; /* Number of deallocations by $STKMRET. */ + long segments; /* Current number of stack segments. */ + long maxs; /* Maximum number of stack segments so far. */ + long pad_size; /* Stack pad size. */ + long current_address; /* Current stack segment address. */ + long current_size; /* Current stack segment size. This + number is actually corrupted by STKSTAT to + include the fifteen word trailer area. */ + long initial_address; /* Address of initial segment. */ + long initial_size; /* Size of initial segment. */ + }; + +/* The following structure describes the data structure which trails + any stack segment. I think that the description in 'asdef' is + out of date. I only describe the parts that I am sure about. */ + +struct stk_trailer + { + long this_address; /* Address of this block. */ + long this_size; /* Size of this block (does not include + this trailer). */ + long unknown2; + long unknown3; + long link; /* Address of trailer block of previous + segment. */ + long unknown5; + long unknown6; + long unknown7; + long unknown8; + long unknown9; + long unknown10; + long unknown11; + long unknown12; + long unknown13; + long unknown14; + }; + +# endif /* CRAY2 */ +# endif /* not CRAY_STACK */ + +# ifdef CRAY2 +/* Determine a "stack measure" for an arbitrary ADDRESS. + I doubt that "lint" will like this much. */ + +static long +i00afunc (long *address) +{ + struct stk_stat status; + struct stk_trailer *trailer; + long *block, size; + long result = 0; + + /* We want to iterate through all of the segments. The first + step is to get the stack status structure. We could do this + more quickly and more directly, perhaps, by referencing the + $LM00 common block, but I know that this works. */ + + STKSTAT (&status); + + /* Set up the iteration. */ + + trailer = (struct stk_trailer *) (status.current_address + + status.current_size + - 15); + + /* There must be at least one stack segment. Therefore it is + a fatal error if "trailer" is null. */ + + if (trailer == 0) + abort (); + + /* Discard segments that do not contain our argument address. */ + + while (trailer != 0) + { + block = (long *) trailer->this_address; + size = trailer->this_size; + if (block == 0 || size == 0) + abort (); + trailer = (struct stk_trailer *) trailer->link; + if ((block <= address) && (address < (block + size))) + break; + } + + /* Set the result to the offset in this segment and add the sizes + of all predecessor segments. */ + + result = address - block; + + if (trailer == 0) + { + return result; + } + + do + { + if (trailer->this_size <= 0) + abort (); + result += trailer->this_size; + trailer = (struct stk_trailer *) trailer->link; + } + while (trailer != 0); + + /* We are done. Note that if you present a bogus address (one + not in any segment), you will get a different number back, formed + from subtracting the address of the first block. This is probably + not what you want. */ + + return (result); +} + +# else /* not CRAY2 */ +/* Stack address function for a CRAY-1, CRAY X-MP, or CRAY Y-MP. + Determine the number of the cell within the stack, + given the address of the cell. The purpose of this + routine is to linearize, in some sense, stack addresses + for alloca. */ + +static long +i00afunc (long address) +{ + long stkl = 0; + + long size, pseg, this_segment, stack; + long result = 0; + + struct stack_segment_linkage *ssptr; + + /* Register B67 contains the address of the end of the + current stack segment. If you (as a subprogram) store + your registers on the stack and find that you are past + the contents of B67, you have overflowed the segment. + + B67 also points to the stack segment linkage control + area, which is what we are really interested in. */ + + stkl = CRAY_STACKSEG_END (); + ssptr = (struct stack_segment_linkage *) stkl; + + /* If one subtracts 'size' from the end of the segment, + one has the address of the first word of the segment. + + If this is not the first segment, 'pseg' will be + nonzero. */ + + pseg = ssptr->sspseg; + size = ssptr->sssize; + + this_segment = stkl - size; + + /* It is possible that calling this routine itself caused + a stack overflow. Discard stack segments which do not + contain the target address. */ + + while (!(this_segment <= address && address <= stkl)) + { +# ifdef DEBUG_I00AFUNC + fprintf (stderr, "%011o %011o %011o\n", this_segment, address, stkl); +# endif + if (pseg == 0) + break; + stkl = stkl - pseg; + ssptr = (struct stack_segment_linkage *) stkl; + size = ssptr->sssize; + pseg = ssptr->sspseg; + this_segment = stkl - size; + } + + result = address - this_segment; + + /* If you subtract pseg from the current end of the stack, + you get the address of the previous stack segment's end. + This seems a little convoluted to me, but I'll bet you save + a cycle somewhere. */ + + while (pseg != 0) + { +# ifdef DEBUG_I00AFUNC + fprintf (stderr, "%011o %011o\n", pseg, size); +# endif + stkl = stkl - pseg; + ssptr = (struct stack_segment_linkage *) stkl; + size = ssptr->sssize; + pseg = ssptr->sspseg; + result += size; + } + return (result); +} + +# endif /* not CRAY2 */ +# endif /* CRAY */ + +# endif /* no alloca */ +#endif /* not GCC version 2 */ diff --git a/lib/alloca_.h b/lib/alloca_.h new file mode 100644 index 0000000..3e3fdf4 --- /dev/null +++ b/lib/alloca_.h @@ -0,0 +1,52 @@ +/* Memory allocation on the stack. + + Copyright (C) 1995, 1999, 2001, 2002, 2003, 2004 Free Software + Foundation, Inc. + + This program is free software; you can redistribute it and/or modify it + under the terms of the GNU General Public License as published + by the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + General Public License for more details. + + You should have received a copy of the GNU General Public + License along with this program; if not, write to the Free Software + Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, + USA. */ + +/* Avoid using the symbol _ALLOCA_H here, as Bison assumes _ALLOCA_H + means there is a real alloca function. */ +#ifndef _GNULIB_ALLOCA_H +# define _GNULIB_ALLOCA_H + +/* alloca (N) returns a pointer to N bytes of memory + allocated on the stack, which will last until the function returns. + Use of alloca should be avoided: + - inside arguments of function calls - undefined behaviour, + - in inline functions - the allocation may actually last until the + calling function returns, + - for huge N (say, N >= 65536) - you never know how large (or small) + the stack is, and when the stack cannot fulfill the memory allocation + request, the program just crashes. + */ + +#ifdef __GNUC__ +# define alloca __builtin_alloca +#elif defined _AIX +# define alloca __alloca +#elif defined _MSC_VER +# include <malloc.h> +# define alloca _alloca +#else +# include <stddef.h> +# ifdef __cplusplus +extern "C" +# endif +void *alloca (size_t); +#endif + +#endif /* _GNULIB_ALLOCA_H */ diff --git a/lib/allocsa.c b/lib/allocsa.c new file mode 100644 index 0000000..a54e2d0 --- /dev/null +++ b/lib/allocsa.c @@ -0,0 +1,139 @@ +/* Safe automatic memory allocation. + Copyright (C) 2003 Free Software Foundation, Inc. + Written by Bruno Haible <bruno@clisp.org>, 2003. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Specification. */ +#include "allocsa.h" + +/* The speed critical point in this file is freesa() applied to an alloca() + result: it must be fast, to match the speed of alloca(). The speed of + mallocsa() and freesa() in the other case are not critical, because they + are only invoked for big memory sizes. */ + +#if HAVE_ALLOCA + +/* Store the mallocsa() results in a hash table. This is needed to reliably + distinguish a mallocsa() result and an alloca() result. + + Although it is possible that the same pointer is returned by alloca() and + by mallocsa() at different times in the same application, it does not lead + to a bug in freesa(), because: + - Before a pointer returned by alloca() can point into malloc()ed memory, + the function must return, and once this has happened the programmer must + not call freesa() on it anyway. + - Before a pointer returned by mallocsa() can point into the stack, it + must be freed. The only function that can free it is freesa(), and + when freesa() frees it, it also removes it from the hash table. */ + +#define MAGIC_NUMBER 0x1415fb4a +#define MAGIC_SIZE sizeof (int) +/* This is how the header info would look like without any alignment + considerations. */ +struct preliminary_header { void *next; char room[MAGIC_SIZE]; }; +/* But the header's size must be a multiple of sa_alignment_max. */ +#define HEADER_SIZE \ + (((sizeof (struct preliminary_header) + sa_alignment_max - 1) / sa_alignment_max) * sa_alignment_max) +struct header { void *next; char room[HEADER_SIZE - sizeof (struct preliminary_header) + MAGIC_SIZE]; }; +/* Verify that HEADER_SIZE == sizeof (struct header). */ +typedef int verify1[2 * (HEADER_SIZE == sizeof (struct header)) - 1]; +/* We make the hash table quite big, so that during lookups the probability + of empty hash buckets is quite high. There is no need to make the hash + table resizable, because when the hash table gets filled so much that the + lookup becomes slow, it means that the application has memory leaks. */ +#define HASH_TABLE_SIZE 257 +static void * mallocsa_results[HASH_TABLE_SIZE]; + +#endif + +void * +mallocsa (size_t n) +{ +#if HAVE_ALLOCA + /* Allocate one more word, that serves as an indicator for malloc()ed + memory, so that freesa() of an alloca() result is fast. */ + size_t nplus = n + HEADER_SIZE; + + if (nplus >= n) + { + char *p = (char *) malloc (nplus); + + if (p != NULL) + { + size_t slot; + + p += HEADER_SIZE; + + /* Put a magic number into the indicator word. */ + ((int *) p)[-1] = MAGIC_NUMBER; + + /* Enter p into the hash table. */ + slot = (unsigned long) p % HASH_TABLE_SIZE; + ((struct header *) (p - HEADER_SIZE))->next = mallocsa_results[slot]; + mallocsa_results[slot] = p; + + return p; + } + } + /* Out of memory. */ + return NULL; +#else +# if !MALLOC_0_IS_NONNULL + if (n == 0) + n = 1; +# endif + return malloc (n); +#endif +} + +#if HAVE_ALLOCA +void +freesa (void *p) +{ + /* mallocsa() may have returned NULL. */ + if (p != NULL) + { + /* Attempt to quickly distinguish the mallocsa() result - which has + a magic indicator word - and the alloca() result - which has an + uninitialized indicator word. It is for this test that sa_increment + additional bytes are allocated in the alloca() case. */ + if (((int *) p)[-1] == MAGIC_NUMBER) + { + /* Looks like a mallocsa() result. To see whether it really is one, + perform a lookup in the hash table. */ + size_t slot = (unsigned long) p % HASH_TABLE_SIZE; + void **chain = &mallocsa_results[slot]; + for (; *chain != NULL;) + { + if (*chain == p) + { + /* Found it. Remove it from the hash table and free it. */ + char *p_begin = (char *) p - HEADER_SIZE; + *chain = ((struct header *) p_begin)->next; + free (p_begin); + return; + } + chain = &((struct header *) ((char *) *chain - HEADER_SIZE))->next; + } + } + /* At this point, we know it was not a mallocsa() result. */ + } +} +#endif diff --git a/lib/allocsa.h b/lib/allocsa.h new file mode 100644 index 0000000..cb6893a --- /dev/null +++ b/lib/allocsa.h @@ -0,0 +1,113 @@ +/* Safe automatic memory allocation. + Copyright (C) 2003-2004 Free Software Foundation, Inc. + Written by Bruno Haible <bruno@clisp.org>, 2003. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _ALLOCSA_H +#define _ALLOCSA_H + +#include <alloca.h> +#include <stddef.h> +#include <stdlib.h> + +/* safe_alloca(N) is equivalent to alloca(N) when it is safe to call + alloca(N); otherwise it returns NULL. It either returns N bytes of + memory allocated on the stack, that lasts until the function returns, + or NULL. + Use of safe_alloca should be avoided: + - inside arguments of function calls - undefined behaviour, + - in inline functions - the allocation may actually last until the + calling function returns. +*/ +#if HAVE_ALLOCA +/* The OS usually guarantees only one guard page at the bottom of the stack, + and a page size can be as small as 4096 bytes. So we cannot safely + allocate anything larger than 4096 bytes. Also care for the possibility + of a few compiler-allocated temporary stack slots. + This must be a macro, not an inline function. */ +# define safe_alloca(N) ((N) < 4032 ? alloca (N) : NULL) +#else +# define safe_alloca(N) ((N), NULL) +#endif + +/* allocsa(N) is a safe variant of alloca(N). It allocates N bytes of + memory allocated on the stack, that must be freed using freesa() before + the function returns. Upon failure, it returns NULL. */ +#if HAVE_ALLOCA +# define allocsa(N) \ + ((N) < 4032 - sa_increment \ + ? (void *) ((char *) alloca ((N) + sa_increment) + sa_increment) \ + : mallocsa (N)) +#else +# define allocsa(N) \ + mallocsa (N) +#endif +extern void * mallocsa (size_t n); + +/* Free a block of memory allocated through allocsa(). */ +#if HAVE_ALLOCA +extern void freesa (void *p); +#else +# define freesa free +#endif + +/* Maybe we should also define a variant + nallocsa (size_t n, size_t s) - behaves like allocsa (n * s) + If this would be useful in your application. please speak up. */ + + +/* ------------------- Auxiliary, non-public definitions ------------------- */ + +/* Determine the alignment of a type at compile time. */ +#if defined __GNUC__ +# define sa_alignof __alignof__ +#elif defined __cplusplus + template <class type> struct sa_alignof_helper { char __slot1; type __slot2; }; +# define sa_alignof(type) offsetof (sa_alignof_helper<type>, __slot2) +#elif defined __hpux + /* Work around a HP-UX 10.20 cc bug with enums constants defined as offsetof + values. */ +# define sa_alignof(type) (sizeof (type) <= 4 ? 4 : 8) +#else +# define sa_alignof(type) offsetof (struct { char __slot1; type __slot2; }, __slot2) +#endif + +enum +{ +/* The desired alignment of memory allocations is the maximum alignment + among all elementary types. */ + sa_alignment_long = sa_alignof (long), + sa_alignment_double = sa_alignof (double), +#ifdef HAVE_LONG_LONG + sa_alignment_longlong = sa_alignof (long long), +#endif +#ifdef HAVE_LONG_DOUBLE + sa_alignment_longdouble = sa_alignof (long double), +#endif + sa_alignment_max = ((sa_alignment_long - 1) | (sa_alignment_double - 1) +#ifdef HAVE_LONG_LONG + | (sa_alignment_longlong - 1) +#endif +#ifdef HAVE_LONG_DOUBLE + | (sa_alignment_longdouble - 1) +#endif + ) + 1, +/* The increment that guarantees room for a magic word must be >= sizeof (int) + and a multiple of sa_alignment_max. */ + sa_increment = ((sizeof (int) + sa_alignment_max - 1) / sa_alignment_max) * sa_alignment_max +}; + +#endif /* _ALLOCSA_H */ diff --git a/lib/allocsa.valgrind b/lib/allocsa.valgrind new file mode 100644 index 0000000..f4c77d6 --- /dev/null +++ b/lib/allocsa.valgrind @@ -0,0 +1,7 @@ +# Suppress a valgrind message about use of uninitialized memory in freesa(). +# This use is OK because it provides only a speedup. +{ + freesa + Memcheck:Cond + fun:freesa +} diff --git a/lib/asnprintf.c b/lib/asnprintf.c new file mode 100644 index 0000000..1b7f4ba --- /dev/null +++ b/lib/asnprintf.c @@ -0,0 +1,37 @@ +/* Formatted output to strings. + Copyright (C) 1999, 2002 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Specification. */ +#include "vasnprintf.h" + +#include <stdarg.h> + +char * +asnprintf (char *resultbuf, size_t *lengthp, const char *format, ...) +{ + va_list args; + char *result; + + va_start (args, format); + result = vasnprintf (resultbuf, lengthp, format, args); + va_end (args); + return result; +} diff --git a/lib/asprintf.c b/lib/asprintf.c new file mode 100644 index 0000000..7c4e64a --- /dev/null +++ b/lib/asprintf.c @@ -0,0 +1,37 @@ +/* Formatted output to strings. + Copyright (C) 1999, 2002 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Specification. */ +#include "vasprintf.h" + +#include <stdarg.h> + +int +asprintf (char **resultp, const char *format, ...) +{ + va_list args; + int result; + + va_start (args, format); + result = vasprintf (resultp, format, args); + va_end (args); + return result; +} diff --git a/lib/atexit.c b/lib/atexit.c new file mode 100644 index 0000000..f4873c9 --- /dev/null +++ b/lib/atexit.c @@ -0,0 +1,15 @@ +/* Wrapper to implement ANSI C's atexit using SunOS's on_exit. */ +/* This function is in the public domain. --Mike Stump. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +int +atexit (void (*f) (void)) +{ + /* If the system doesn't provide a definition for atexit, use on_exit + if the system provides that. */ + on_exit (f, 0); + return 0; +} diff --git a/lib/basename.c b/lib/basename.c new file mode 100644 index 0000000..5cc97cd --- /dev/null +++ b/lib/basename.c @@ -0,0 +1,79 @@ +/* basename.c -- return the last element in a file name + + Copyright (C) 1990, 1998, 1999, 2000, 2001, 2003, 2004, 2005 Free + Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "dirname.h" +#include <string.h> + +/* In general, we can't use the builtin `basename' function if available, + since it has different meanings in different environments. + In some environments the builtin `basename' modifies its argument. + + Return the address of the last file name component of NAME. If + NAME has no file name components because it is all slashes, return + NAME if it is empty, the address of its last slash otherwise. */ + +char * +base_name (char const *name) +{ + char const *base = name + FILE_SYSTEM_PREFIX_LEN (name); + char const *p; + + for (p = base; *p; p++) + { + if (ISSLASH (*p)) + { + /* Treat multiple adjacent slashes like a single slash. */ + do p++; + while (ISSLASH (*p)); + + /* If the file name ends in slash, use the trailing slash as + the basename if no non-slashes have been found. */ + if (! *p) + { + if (ISSLASH (*base)) + base = p - 1; + break; + } + + /* *P is a non-slash preceded by a slash. */ + base = p; + } + } + + return (char *) base; +} + +/* Return the length of of the basename NAME. Typically NAME is the + value returned by base_name. Act like strlen (NAME), except omit + redundant trailing slashes. */ + +size_t +base_len (char const *name) +{ + size_t len; + + for (len = strlen (name); 1 < len && ISSLASH (name[len - 1]); len--) + continue; + + return len; +} diff --git a/lib/build_lib.com b/lib/build_lib.com new file mode 100755 index 0000000..2b19926 --- /dev/null +++ b/lib/build_lib.com @@ -0,0 +1,20 @@ +$ set verify +$ CC :== CC/DEBUG/NOOPTIMIZE/STANDARD=VAXC/DEFINE=HAVE_CONFIG_H- +/INCLUDE_DIRECTORY=([-],[-.LIB],[-.SRC],[-.VMS])/PREFIX_LIBRARY_ENTRIES=ALL_ENTRIES +$ CC fnmatch.c +$ CC getdate.c +$ CC getline.c +$ CC getopt.c +$ CC getopt1.c +$ CC md5.c +$ CC regex.c +$ CC savecwd.c +$ CC sighandle.c +$ CC stripslash.c +$ CC valloc.c +$ CC xgetwd.c +$ CC yesno.c +$ library/create gnulib.olb fnmatch.obj,getdate.obj,getline.obj,- +getopt.obj,getopt1.obj,md5.obj,regex.obj,savecwd.obj,sighandle.obj,- +stripslash.obj,valloc.obj,xgetwd.obj,yesno.obj +$ set noverify diff --git a/lib/canon-host.c b/lib/canon-host.c new file mode 100644 index 0000000..e2bd1a7 --- /dev/null +++ b/lib/canon-host.c @@ -0,0 +1,90 @@ +/* Host name canonicalization + + Copyright (C) 2005 Free Software Foundation, Inc. + + Written by Derek Price <derek@ximbiot.com>. + + This program is free software; you can redistribute it and/or + modify it under the terms of the GNU General Public License as + published by the Free Software Foundation; either version 2, or (at + your option) any later version. + + This program is distributed in the hope that it will be useful, but + WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "canon-host.h" + +#include "getaddrinfo.h" +#include "strdup.h" + +/* Store the last error for the single-threaded version of this function. */ +static int last_cherror; + +/* Single-threaded of wrapper for canon_host_r. After a NULL return, error + messages may be retrieved via ch_strerror(). */ +char * +canon_host (const char *host) +{ + return canon_host_r (host, &last_cherror); +} + +/* Return a malloc'd string containing the canonical hostname associated with + HOST, or NULL if a canonical name cannot be determined. On NULL return, + if CHERROR is not NULL, set *CHERROR to an error code as returned by + getaddrinfo(). Use ch_strerror_r() or gai_strerror() to convert a *CHERROR + value to a string suitable for error messages. + + WARNINGS + HOST must be a string representation of a resolvable name for this host. + Strings containing an IP address in dotted decimal notation will be + returned as-is, without further resolution. + + The use of the word "canonical" in this context is unfortunate but + entrenched. The value returned by this function will be the end result + of the resolution of any CNAME chains in the DNS. There may only be one + such value for any given hostname, though the actual IP address + referenced by this value and the device using that IP address may each + actually have any number of such "canonical" hostnames. See the POSIX + getaddrinfo spec <http://www.opengroup.org/susv3xsh/getaddrinfo.html">, + RFC 1034 <http://www.faqs.org/rfcs/rfc1034.html>, & RFC 2181 + <http://www.faqs.org/rfcs/rfc2181.html> for more on what this confusing + term really refers to. */ +char * +canon_host_r (char const *host, int *cherror) +{ + char *retval = NULL; + static struct addrinfo hints; + struct addrinfo *res = NULL; + int status; + + hints.ai_flags = AI_CANONNAME; + status = getaddrinfo (host, NULL, &hints, &res); + if (!status) + { + retval = strdup (res->ai_canonname); + if (!retval && cherror) + *cherror = EAI_MEMORY; + freeaddrinfo (res); + } + else if (cherror) + *cherror = status; + + return retval; +} + +/* Return a string describing the last error encountered by canon_host. */ +const char * +ch_strerror (void) +{ + return gai_strerror (last_cherror); +} diff --git a/lib/canon-host.h b/lib/canon-host.h new file mode 100644 index 0000000..c1f6cb1 --- /dev/null +++ b/lib/canon-host.h @@ -0,0 +1,30 @@ +/* Host name canonicalization + + Copyright (C) 2005 Free Software Foundation, Inc. + + Written by Derek Price <derek@ximbiot.com> + + This program is free software; you can redistribute it and/or + modify it under the terms of the GNU General Public License as + published by the Free Software Foundation; either version 2, or (at + your option) any later version. + + This program is distributed in the hope that it will be useful, but + WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef CANON_HOST_H +# define CANON_HOST_H 1 + +char *canon_host (char const *host); +char *canon_host_r (char const *host, int *cherror); + +const char *ch_strerror (void); +# define ch_strerror_r(cherror) gai_strerror (cherror); + +#endif /* !CANON_HOST_H */ diff --git a/lib/canonicalize.c b/lib/canonicalize.c new file mode 100644 index 0000000..571fa19 --- /dev/null +++ b/lib/canonicalize.c @@ -0,0 +1,318 @@ +/* Return the canonical absolute name of a given file. + Copyright (C) 1996-2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; see the file COPYING. + If not, write to the Free Software Foundation, + 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "canonicalize.h" + +#ifdef STDC_HEADERS +# include <stdlib.h> +#else +void free (); +#endif + +#if defined STDC_HEADERS || defined HAVE_STRING_H +# include <string.h> +#else +# include <strings.h> +#endif + +#if HAVE_SYS_PARAM_H +# include <sys/param.h> +#endif + +#include <sys/stat.h> + +#if HAVE_UNISTD_H +# include <unistd.h> +#endif + +#include <errno.h> +#include <stddef.h> + +#include "cycle-check.h" +#include "filenamecat.h" +#include "stat-macros.h" +#include "xalloc.h" +#include "xgetcwd.h" + +#ifndef __set_errno +# define __set_errno(Val) errno = (Val) +#endif + +#include "pathmax.h" +#include "xreadlink.h" + +#if !HAVE_CANONICALIZE_FILE_NAME +/* Return the canonical absolute name of file NAME. A canonical name + does not contain any `.', `..' components nor any repeated file name + separators ('/') or symlinks. All components must exist. + The result is malloc'd. */ + +char * +canonicalize_file_name (const char *name) +{ +# if HAVE_RESOLVEPATH + + char *resolved, *extra_buf = NULL; + size_t resolved_size; + ssize_t resolved_len; + + if (name == NULL) + { + __set_errno (EINVAL); + return NULL; + } + + if (name[0] == '\0') + { + __set_errno (ENOENT); + return NULL; + } + + /* All known hosts with resolvepath (e.g. Solaris 7) don't turn + relative names into absolute ones, so prepend the working + directory if the file name is not absolute. */ + if (name[0] != '/') + { + char *wd; + + if (!(wd = xgetcwd ())) + return NULL; + + extra_buf = file_name_concat (wd, name, NULL); + name = extra_buf; + free (wd); + } + + resolved_size = strlen (name); + while (1) + { + resolved_size = 2 * resolved_size + 1; + resolved = xmalloc (resolved_size); + resolved_len = resolvepath (name, resolved, resolved_size); + if (resolved_len < 0) + { + free (resolved); + free (extra_buf); + return NULL; + } + if (resolved_len < resolved_size) + break; + free (resolved); + } + + free (extra_buf); + + /* NUL-terminate the resulting name. */ + resolved[resolved_len] = '\0'; + + return resolved; + +# else + + return canonicalize_filename_mode (name, CAN_EXISTING); + +# endif /* !HAVE_RESOLVEPATH */ +} +#endif /* !HAVE_CANONICALIZE_FILE_NAME */ + +/* Return the canonical absolute name of file NAME. A canonical name + does not contain any `.', `..' components nor any repeated file name + separators ('/') or symlinks. Whether components must exist + or not depends on canonicalize mode. The result is malloc'd. */ + +char * +canonicalize_filename_mode (const char *name, canonicalize_mode_t can_mode) +{ + char *rname, *dest, *extra_buf = NULL; + char const *start; + char const *end; + char const *rname_limit; + size_t extra_len = 0; + struct cycle_check_state cycle_state; + + if (name == NULL) + { + __set_errno (EINVAL); + return NULL; + } + + if (name[0] == '\0') + { + __set_errno (ENOENT); + return NULL; + } + + if (name[0] != '/') + { + rname = xgetcwd (); + if (!rname) + return NULL; + dest = strchr (rname, '\0'); + if (dest - rname < PATH_MAX) + { + char *p = xrealloc (rname, PATH_MAX); + dest = p + (dest - rname); + rname = p; + rname_limit = rname + PATH_MAX; + } + else + { + rname_limit = dest; + } + } + else + { + rname = xmalloc (PATH_MAX); + rname_limit = rname + PATH_MAX; + rname[0] = '/'; + dest = rname + 1; + } + + cycle_check_init (&cycle_state); + for (start = end = name; *start; start = end) + { + /* Skip sequence of multiple file name separators. */ + while (*start == '/') + ++start; + + /* Find end of component. */ + for (end = start; *end && *end != '/'; ++end) + /* Nothing. */; + + if (end - start == 0) + break; + else if (end - start == 1 && start[0] == '.') + /* nothing */; + else if (end - start == 2 && start[0] == '.' && start[1] == '.') + { + /* Back up to previous component, ignore if at root already. */ + if (dest > rname + 1) + while ((--dest)[-1] != '/'); + } + else + { + struct stat st; + + if (dest[-1] != '/') + *dest++ = '/'; + + if (dest + (end - start) >= rname_limit) + { + ptrdiff_t dest_offset = dest - rname; + size_t new_size = rname_limit - rname; + + if (end - start + 1 > PATH_MAX) + new_size += end - start + 1; + else + new_size += PATH_MAX; + rname = xrealloc (rname, new_size); + rname_limit = rname + new_size; + + dest = rname + dest_offset; + } + + dest = memcpy (dest, start, end - start); + dest += end - start; + *dest = '\0'; + + if (lstat (rname, &st) != 0) + { + if (can_mode == CAN_EXISTING) + goto error; + if (can_mode == CAN_ALL_BUT_LAST && *end) + goto error; + st.st_mode = 0; + } + + if (S_ISLNK (st.st_mode)) + { + char *buf; + size_t n, len; + + if (cycle_check (&cycle_state, &st)) + { + __set_errno (ELOOP); + if (can_mode == CAN_MISSING) + continue; + else + goto error; + } + + buf = xreadlink (rname, st.st_size); + if (!buf) + { + if (can_mode == CAN_MISSING) + continue; + else + goto error; + } + + n = strlen (buf); + len = strlen (end); + + if (!extra_len) + { + extra_len = + ((n + len + 1) > PATH_MAX) ? (n + len + 1) : PATH_MAX; + extra_buf = xmalloc (extra_len); + } + else if ((n + len + 1) > extra_len) + { + extra_len = n + len + 1; + extra_buf = xrealloc (extra_buf, extra_len); + } + + /* Careful here, end may be a pointer into extra_buf... */ + memmove (&extra_buf[n], end, len + 1); + name = end = memcpy (extra_buf, buf, n); + + if (buf[0] == '/') + dest = rname + 1; /* It's an absolute symlink */ + else + /* Back up to previous component, ignore if at root already: */ + if (dest > rname + 1) + while ((--dest)[-1] != '/'); + + free (buf); + } + else + { + if (!S_ISDIR (st.st_mode) && *end && (can_mode != CAN_MISSING)) + { + errno = ENOTDIR; + goto error; + } + } + } + } + if (dest > rname + 1 && dest[-1] == '/') + --dest; + *dest = '\0'; + + free (extra_buf); + return rname; + +error: + free (extra_buf); + free (rname); + return NULL; +} diff --git a/lib/canonicalize.h b/lib/canonicalize.h new file mode 100644 index 0000000..4cae3c5 --- /dev/null +++ b/lib/canonicalize.h @@ -0,0 +1,41 @@ +/* Return the canonical absolute name of a given file. + Copyright (C) 1996-2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; see the file COPYING. + If not, write to the Free Software Foundation, + 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef CANONICALIZE_H_ +# define CANONICALIZE_H_ + +enum canonicalize_mode_t + { + /* All components must exist. */ + CAN_EXISTING = 0, + + /* All components excluding last one must exist. */ + CAN_ALL_BUT_LAST = 1, + + /* No requirements on components existence. */ + CAN_MISSING = 2 + }; +typedef enum canonicalize_mode_t canonicalize_mode_t; + +char *canonicalize_filename_mode (const char *, canonicalize_mode_t); + +# if !HAVE_CANONICALIZE_FILE_NAME +char *canonicalize_file_name (const char *); +# endif + +#endif /* !CANONICALIZE_H_ */ diff --git a/lib/chdir-long.c b/lib/chdir-long.c new file mode 100644 index 0000000..a727817 --- /dev/null +++ b/lib/chdir-long.c @@ -0,0 +1,276 @@ +/* provide a chdir function that tries not to fail due to ENAMETOOLONG + Copyright (C) 2004, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* written by Jim Meyering */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "chdir-long.h" + +#include <stdlib.h> +#include <stdbool.h> +#include <string.h> +#include <unistd.h> +#include <errno.h> +#include <stdio.h> +#include <assert.h> +#include <limits.h> + +#include "memrchr.h" +#include "openat.h" + +#ifndef O_DIRECTORY +# define O_DIRECTORY 0 +#endif + +#ifndef PATH_MAX +# error "compile this file only if your system defines PATH_MAX" +#endif + +struct cd_buf +{ + int fd; +}; + +static inline void +cdb_init (struct cd_buf *cdb) +{ + cdb->fd = AT_FDCWD; +} + +static inline int +cdb_fchdir (struct cd_buf const *cdb) +{ + return fchdir (cdb->fd); +} + +static inline void +cdb_free (struct cd_buf const *cdb) +{ + if (0 <= cdb->fd) + { + bool close_fail = close (cdb->fd); + assert (! close_fail); + } +} + +/* Given a file descriptor of an open directory (or AT_FDCWD), CDB->fd, + try to open the CDB->fd-relative directory, DIR. If the open succeeds, + update CDB->fd with the resulting descriptor, close the incoming file + descriptor, and return zero. Upon failure, return -1 and set errno. */ +static int +cdb_advance_fd (struct cd_buf *cdb, char const *dir) +{ + int new_fd = openat (cdb->fd, dir, O_RDONLY | O_DIRECTORY); + if (new_fd < 0) + { + new_fd = openat (cdb->fd, dir, O_WRONLY | O_DIRECTORY); + if (new_fd < 0) + return -1; + } + + cdb_free (cdb); + cdb->fd = new_fd; + + return 0; +} + +/* Return a pointer to the first non-slash in S. */ +static inline char * +find_non_slash (char const *s) +{ + size_t n_slash = strspn (s, "/"); + return (char *) s + n_slash; +} + +/* This is a function much like chdir, but without the PATH_MAX limitation + on the length of the directory name. A significant difference is that + it must be able to modify (albeit only temporarily) the directory + name. It handles an arbitrarily long directory name by operating + on manageable portions of the name. On systems without the openat + syscall, this means changing the working directory to more and more + `distant' points along the long directory name and then restoring + the working directory. If any of those attempts to save or restore + the working directory fails, this function exits nonzero. + + Note that this function may still fail with errno == ENAMETOOLONG, but + only if the specified directory name contains a component that is long + enough to provoke such a failure all by itself (e.g. if the component + has length PATH_MAX or greater on systems that define PATH_MAX). */ + +int +chdir_long (char *dir) +{ + int e = chdir (dir); + if (e == 0 || errno != ENAMETOOLONG) + return e; + + { + size_t len = strlen (dir); + char *dir_end = dir + len; + struct cd_buf cdb; + size_t n_leading_slash; + + cdb_init (&cdb); + + /* If DIR is the empty string, then the chdir above + must have failed and set errno to ENOENT. */ + assert (0 < len); + assert (PATH_MAX <= len); + + /* Count leading slashes. */ + n_leading_slash = strspn (dir, "/"); + + /* Handle any leading slashes as well as any name that matches + the regular expression, m!^//hostname[/]*! . Handling this + prefix separately usually results in a single additional + cdb_advance_fd call, but it's worthwhile, since it makes the + code in the following loop cleaner. */ + if (n_leading_slash == 2) + { + int err; + /* Find next slash. + We already know that dir[2] is neither a slash nor '\0'. */ + char *slash = memchr (dir + 3, '/', dir_end - (dir + 3)); + if (slash == NULL) + { + errno = ENAMETOOLONG; + return -1; + } + *slash = '\0'; + err = cdb_advance_fd (&cdb, dir); + *slash = '/'; + if (err != 0) + goto Fail; + dir = find_non_slash (slash + 1); + } + else if (n_leading_slash) + { + if (cdb_advance_fd (&cdb, "/") != 0) + goto Fail; + dir += n_leading_slash; + } + + assert (*dir != '/'); + assert (dir <= dir_end); + + while (PATH_MAX <= dir_end - dir) + { + int err; + /* Find a slash that is PATH_MAX or fewer bytes away from dir. + I.e. see if there is a slash that will give us a name of + length PATH_MAX-1 or less. */ + char *slash = memrchr (dir, '/', PATH_MAX); + if (slash == NULL) + { + errno = ENAMETOOLONG; + return -1; + } + + *slash = '\0'; + assert (slash - dir < PATH_MAX); + err = cdb_advance_fd (&cdb, dir); + *slash = '/'; + if (err != 0) + goto Fail; + + dir = find_non_slash (slash + 1); + } + + if (dir < dir_end) + { + if (cdb_advance_fd (&cdb, dir) != 0) + goto Fail; + } + + if (cdb_fchdir (&cdb) != 0) + goto Fail; + + cdb_free (&cdb); + return 0; + + Fail: + { + int saved_errno = errno; + cdb_free (&cdb); + errno = saved_errno; + return -1; + } + } +} + +#if TEST_CHDIR + +# include <stdio.h> +# include "closeout.h" +# include "error.h" + +char *program_name; + +int +main (int argc, char *argv[]) +{ + char *line = NULL; + size_t n = 0; + int len; + + program_name = argv[0]; + atexit (close_stdout); + + len = getline (&line, &n, stdin); + if (len < 0) + { + int saved_errno = errno; + if (feof (stdin)) + exit (0); + + error (EXIT_FAILURE, saved_errno, + "reading standard input"); + } + else if (len == 0) + exit (0); + + if (line[len-1] == '\n') + line[len-1] = '\0'; + + if (chdir_long (line) != 0) + error (EXIT_FAILURE, errno, + "chdir_long failed: %s", line); + + if (argc <= 1) + { + /* Using `pwd' here makes sense only if it is a robust implementation, + like the one in coreutils after the 2004-04-19 changes. */ + char const *cmd = "pwd"; + execlp (cmd, (char *) NULL); + error (EXIT_FAILURE, errno, "%s", cmd); + } + + fclose (stdin); + fclose (stderr); + + exit (EXIT_SUCCESS); +} +#endif + +/* +Local Variables: +compile-command: "gcc -DTEST_CHDIR=1 -DHAVE_CONFIG_H -I.. -g -O -W -Wall chdir-long.c libcoreutils.a" +End: +*/ diff --git a/lib/chdir-long.h b/lib/chdir-long.h new file mode 100644 index 0000000..4852b40 --- /dev/null +++ b/lib/chdir-long.h @@ -0,0 +1,35 @@ +/* provide a chdir function that tries not to fail due to ENAMETOOLONG + Copyright (C) 2004, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Jim Meyering. */ + +#include <unistd.h> +#include <limits.h> + +#ifndef PATH_MAX +# ifdef MAXPATHLEN +# define PATH_MAX MAXPATHLEN +# endif +#endif + +/* On systems without PATH_MAX, presume that chdir accepts + arbitrarily long directory names. */ +#ifndef PATH_MAX +# define chdir_long(Dir) chdir (Dir) +#else +int chdir_long (char *dir); +#endif diff --git a/lib/closeout.c b/lib/closeout.c new file mode 100644 index 0000000..5d0509d --- /dev/null +++ b/lib/closeout.c @@ -0,0 +1,101 @@ +/* closeout.c - close standard output + + Copyright (C) 1998, 1999, 2000, 2001, 2002, 2004 Free Software + Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "closeout.h" + +#include <stdio.h> +#include <stdbool.h> +#include <errno.h> + +#include "gettext.h" +#define _(msgid) gettext (msgid) + +#include "error.h" +#include "exitfail.h" +#include "quotearg.h" +#include "__fpending.h" + +#if USE_UNLOCKED_IO +# include "unlocked-io.h" +#endif + +static const char *file_name; + +/* Set the file name to be reported in the event an error is detected + by close_stdout. */ +void +close_stdout_set_file_name (const char *file) +{ + file_name = file; +} + +/* Close standard output, exiting with status 'exit_failure' on failure. + If a program writes *anything* to stdout, that program should close + stdout and make sure that it succeeds before exiting. Otherwise, + suppose that you go to the extreme of checking the return status + of every function that does an explicit write to stdout. The last + printf can succeed in writing to the internal stream buffer, and yet + the fclose(stdout) could still fail (due e.g., to a disk full error) + when it tries to write out that buffered data. Thus, you would be + left with an incomplete output file and the offending program would + exit successfully. Even calling fflush is not always sufficient, + since some file systems (NFS and CODA) buffer written/flushed data + until an actual close call. + + Besides, it's wasteful to check the return value from every call + that writes to stdout -- just let the internal stream state record + the failure. That's what the ferror test is checking below. + + It's important to detect such failures and exit nonzero because many + tools (most notably `make' and other build-management systems) depend + on being able to detect failure in other tools via their exit status. */ + +void +close_stdout (void) +{ + bool prev_fail = ferror (stdout); + bool none_pending = (0 == __fpending (stdout)); + bool fclose_fail = fclose (stdout); + + if (prev_fail || fclose_fail) + { + int e = fclose_fail ? errno : 0; + char const *write_error; + + /* If ferror returned zero, no data remains to be flushed, and we'd + otherwise fail with EBADF due to a failed fclose, then assume that + it's ok to ignore the fclose failure. That can happen when a + program like cp is invoked like this `cp a b >&-' (i.e., with + stdout closed) and doesn't generate any output (hence no previous + error and nothing to be flushed). */ + if (e == EBADF && !prev_fail && none_pending) + return; + + write_error = _("write error"); + if (file_name) + error (exit_failure, e, "%s: %s", quotearg_colon (file_name), + write_error); + else + error (exit_failure, e, "%s", write_error); + } +} diff --git a/lib/closeout.h b/lib/closeout.h new file mode 100644 index 0000000..589a14f --- /dev/null +++ b/lib/closeout.h @@ -0,0 +1,33 @@ +/* Close standard output. + + Copyright (C) 1998, 2000, 2003, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef CLOSEOUT_H +# define CLOSEOUT_H 1 + +# ifdef __cplusplus +extern "C" { +# endif + +void close_stdout_set_file_name (const char *file); +void close_stdout (void); + +# ifdef __cplusplus +} +# endif + +#endif diff --git a/lib/cycle-check.c b/lib/cycle-check.c new file mode 100644 index 0000000..401dd86 --- /dev/null +++ b/lib/cycle-check.c @@ -0,0 +1,93 @@ +/* help detect directory cycles efficiently + + Copyright (C) 2003, 2004, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; see the file COPYING. + If not, write to the Free Software Foundation, + 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Jim Meyering */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include <sys/types.h> +#include <sys/stat.h> +#include <stdio.h> +#include <assert.h> +#include <stdlib.h> + +#include <stdbool.h> + +#include "cycle-check.h" + +#define SAME_INODE(Stat_buf_1, Stat_buf_2) \ + ((Stat_buf_1).st_ino == (Stat_buf_2).st_ino \ + && (Stat_buf_1).st_dev == (Stat_buf_2).st_dev) + +#define CC_MAGIC 9827862 + +/* Return true if I is a power of 2, or is zero. */ + +static inline bool +is_zero_or_power_of_two (uintmax_t i) +{ + return (i & (i - 1)) == 0; +} + +void +cycle_check_init (struct cycle_check_state *state) +{ + state->chdir_counter = 0; + state->magic = CC_MAGIC; +} + +/* In traversing a directory hierarchy, call this function once for each + descending chdir call, with SB corresponding to the chdir operand. + If SB corresponds to a directory that has already been seen, + return true to indicate that there is a directory cycle. + Note that this is done `lazily', which means that some of + the directories in the cycle may be processed twice before + the cycle is detected. */ + +bool +cycle_check (struct cycle_check_state *state, struct stat const *sb) +{ + assert (state->magic == CC_MAGIC); + + /* If the current directory ever happens to be the same + as the one we last recorded for the cycle detection, + then it's obviously part of a cycle. */ + if (state->chdir_counter && SAME_INODE (*sb, state->dev_ino)) + return true; + + /* If the number of `descending' chdir calls is a power of two, + record the dev/ino of the current directory. */ + if (is_zero_or_power_of_two (++(state->chdir_counter))) + { + /* On all architectures that we know about, if the counter + overflows then there is a directory cycle here somewhere, + even if we haven't detected it yet. Typically this happens + only after the counter is incremented 2**64 times, so it's a + fairly theoretical point. */ + if (state->chdir_counter == 0) + return true; + + state->dev_ino.st_dev = sb->st_dev; + state->dev_ino.st_ino = sb->st_ino; + } + + return false; +} diff --git a/lib/cycle-check.h b/lib/cycle-check.h new file mode 100644 index 0000000..0551bf2 --- /dev/null +++ b/lib/cycle-check.h @@ -0,0 +1,44 @@ +/* help detect directory cycles efficiently + + Copyright (C) 2003, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; see the file COPYING. + If not, write to the Free Software Foundation, + 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Jim Meyering */ + +#ifndef CYCLE_CHECK_H +# define CYCLE_CHECK_H 1 + +# if HAVE_INTTYPES_H +# include <inttypes.h> +# endif +# if HAVE_STDINT_H +# include <stdint.h> +# endif +# include <stdbool.h> +# include "dev-ino.h" + +struct cycle_check_state +{ + struct dev_ino dev_ino; + uintmax_t chdir_counter; + int magic; +}; + +void cycle_check_init (struct cycle_check_state *state); +bool cycle_check (struct cycle_check_state *state, struct stat const *sb); + +#endif diff --git a/lib/dev-ino.h b/lib/dev-ino.h new file mode 100644 index 0000000..695d38c --- /dev/null +++ b/lib/dev-ino.h @@ -0,0 +1,13 @@ +#ifndef DEV_INO_H +# define DEV_INO_H 1 + +# include <sys/types.h> +# include <sys/stat.h> + +struct dev_ino +{ + ino_t st_ino; + dev_t st_dev; +}; + +#endif diff --git a/lib/dirname.c b/lib/dirname.c new file mode 100644 index 0000000..e2b9d64 --- /dev/null +++ b/lib/dirname.c @@ -0,0 +1,121 @@ +/* dirname.c -- return all but the last element in a file name + + Copyright (C) 1990, 1998, 2000, 2001, 2003, 2004, 2005 Free Software + Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "dirname.h" + +#include <string.h> +#include "xalloc.h" + +/* Return the length of `dirname (FILE)', or zero if FILE is + in the working directory. Works properly even if + there are trailing slashes (by effectively ignoring them). */ +size_t +dir_len (char const *file) +{ + size_t prefix_length = FILE_SYSTEM_PREFIX_LEN (file); + size_t length; + + /* Strip the basename and any redundant slashes before it. */ + for (length = base_name (file) - file; prefix_length < length; length--) + if (! ISSLASH (file[length - 1])) + return length; + + /* But don't strip the only slash from "/". */ + return prefix_length + ISSLASH (file[prefix_length]); +} + +/* Return the leading directories part of FILE, + allocated with xmalloc. + Works properly even if there are trailing slashes + (by effectively ignoring them). */ + +char * +dir_name (char const *file) +{ + size_t length = dir_len (file); + bool append_dot = (length == FILE_SYSTEM_PREFIX_LEN (file)); + char *dir = xmalloc (length + append_dot + 1); + memcpy (dir, file, length); + if (append_dot) + dir[length++] = '.'; + dir[length] = 0; + return dir; +} + +#ifdef TEST_DIRNAME +/* + +Run the test like this (expect no output): + gcc -DHAVE_CONFIG_H -DTEST_DIRNAME -I.. -O -Wall \ + basename.c dirname.c xmalloc.c error.c + sed -n '/^BEGIN-DATA$/,/^END-DATA$/p' dirname.c|grep -v DATA|./a.out + +If it's been built on a DOS or Windows platforms, run another test like +this (again, expect no output): + sed -n '/^BEGIN-DOS-DATA$/,/^END-DOS-DATA$/p' dirname.c|grep -v DATA|./a.out + +BEGIN-DATA +foo//// . +bar/foo//// bar +foo/ . +/ / +. . +a . +END-DATA + +BEGIN-DOS-DATA +c:///// c:/ +c:/ c:/ +c:/. c:/ +c:foo c:. +c:foo/bar c:foo +END-DOS-DATA + +*/ + +# define MAX_BUFF_LEN 1024 +# include <stdio.h> + +char *program_name; + +int +main (int argc, char *argv[]) +{ + char buff[MAX_BUFF_LEN + 1]; + + program_name = argv[0]; + + buff[MAX_BUFF_LEN] = 0; + while (fgets (buff, MAX_BUFF_LEN, stdin) && buff[0]) + { + char file[MAX_BUFF_LEN]; + char expected_result[MAX_BUFF_LEN]; + char const *result; + sscanf (buff, "%s %s", file, expected_result); + result = dir_name (file); + if (strcmp (result, expected_result)) + printf ("%s: got %s, expected %s\n", file, result, expected_result); + } + return 0; +} +#endif diff --git a/lib/dirname.h b/lib/dirname.h new file mode 100644 index 0000000..1688ae8 --- /dev/null +++ b/lib/dirname.h @@ -0,0 +1,47 @@ +/* Take file names apart into directory and base names. + + Copyright (C) 1998, 2001, 2003, 2004, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef DIRNAME_H_ +# define DIRNAME_H_ 1 + +# include <stdbool.h> +# include <stddef.h> + +# ifndef DIRECTORY_SEPARATOR +# define DIRECTORY_SEPARATOR '/' +# endif + +# ifndef ISSLASH +# define ISSLASH(C) ((C) == DIRECTORY_SEPARATOR) +# endif + +# ifndef FILE_SYSTEM_PREFIX_LEN +# define FILE_SYSTEM_PREFIX_LEN(File_name) 0 +# endif + +# define IS_ABSOLUTE_FILE_NAME(F) ISSLASH ((F)[FILE_SYSTEM_PREFIX_LEN (F)]) +# define IS_RELATIVE_FILE_NAME(F) (! IS_ABSOLUTE_FILE_NAME (F)) + +char *base_name (char const *file); +char *dir_name (char const *file); +size_t base_len (char const *file); +size_t dir_len (char const *file); + +bool strip_trailing_slashes (char *file); + +#endif /* not DIRNAME_H_ */ diff --git a/lib/dup-safer.c b/lib/dup-safer.c new file mode 100644 index 0000000..8cbee70 --- /dev/null +++ b/lib/dup-safer.c @@ -0,0 +1,46 @@ +/* Invoke dup, but avoid some glitches. + Copyright (C) 2001, 2004, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Paul Eggert. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "unistd-safer.h" + +#include <fcntl.h> + +#include <unistd.h> +#ifndef STDERR_FILENO +# define STDERR_FILENO 2 +#endif + +/* Like dup, but do not return STDIN_FILENO, STDOUT_FILENO, or + STDERR_FILENO. */ + +int +dup_safer (int fd) +{ +#ifdef F_DUPFD + return fcntl (fd, F_DUPFD, STDERR_FILENO + 1); +#else + /* fd_safer calls us back, but eventually the recursion unwinds and + does the right thing. */ + return fd_safer (dup (fd)); +#endif +} diff --git a/lib/dup2.c b/lib/dup2.c new file mode 100644 index 0000000..d934da1 --- /dev/null +++ b/lib/dup2.c @@ -0,0 +1,61 @@ +/* Duplicate an open file descriptor to a specified file descriptor. + Copyright (C) 1999, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* written by Paul Eggert */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include <errno.h> + +#include <fcntl.h> + +#if HAVE_UNISTD_H +# include <unistd.h> +#endif + +#ifndef F_DUPFD +static int +dupfd (int fd, int desired_fd) +{ + int duplicated_fd = dup (fd); + if (duplicated_fd < 0 || duplicated_fd == desired_fd) + return duplicated_fd; + else + { + int r = dupfd (fd, desired_fd); + int e = errno; + close (duplicated_fd); + errno = e; + return r; + } +} +#endif + +int +dup2 (int fd, int desired_fd) +{ + if (fd == desired_fd) + return fd; + close (desired_fd); +#ifdef F_DUPFD + return fcntl (fd, F_DUPFD, desired_fd); +#else + return dupfd (fd, desired_fd); +#endif +} diff --git a/lib/error.c b/lib/error.c new file mode 100644 index 0000000..45698be --- /dev/null +++ b/lib/error.c @@ -0,0 +1,304 @@ +/* Error handler for noninteractive utilities + Copyright (C) 1990-1998, 2000-2003, 2004 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by David MacKenzie <djm@gnu.ai.mit.edu>. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "error.h" + +#include <stdarg.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +#if !_LIBC && ENABLE_NLS +# include "gettext.h" +#endif + +#ifdef _LIBC +# include <wchar.h> +# define mbsrtowcs __mbsrtowcs +#endif + +#if USE_UNLOCKED_IO +# include "unlocked-io.h" +#endif + +#ifndef _ +# define _(String) String +#endif + +/* If NULL, error will flush stdout, then print on stderr the program + name, a colon and a space. Otherwise, error will call this + function without parameters instead. */ +void (*error_print_progname) (void); + +/* This variable is incremented each time `error' is called. */ +unsigned int error_message_count; + +#ifdef _LIBC +/* In the GNU C library, there is a predefined variable for this. */ + +# define program_name program_invocation_name +# include <errno.h> +# include <libio/libioP.h> + +/* In GNU libc we want do not want to use the common name `error' directly. + Instead make it a weak alias. */ +extern void __error (int status, int errnum, const char *message, ...) + __attribute__ ((__format__ (__printf__, 3, 4))); +extern void __error_at_line (int status, int errnum, const char *file_name, + unsigned int line_number, const char *message, + ...) + __attribute__ ((__format__ (__printf__, 5, 6)));; +# define error __error +# define error_at_line __error_at_line + +# include <libio/iolibio.h> +# define fflush(s) INTUSE(_IO_fflush) (s) +# undef putc +# define putc(c, fp) INTUSE(_IO_putc) (c, fp) + +# include <bits/libc-lock.h> + +#else /* not _LIBC */ + +# if !HAVE_DECL_STRERROR_R && STRERROR_R_CHAR_P +# ifndef HAVE_DECL_STRERROR_R +"this configure-time declaration test was not run" +# endif +char *strerror_r (); +# endif + +# ifndef SIZE_MAX +# define SIZE_MAX ((size_t) -1) +# endif + +/* The calling program should define program_name and set it to the + name of the executing program. */ +extern char *program_name; + +# if HAVE_STRERROR_R || defined strerror_r +# define __strerror_r strerror_r +# endif +#endif /* not _LIBC */ + +static void +print_errno_message (int errnum) +{ + char const *s = NULL; + +#if defined HAVE_STRERROR_R || _LIBC + char errbuf[1024]; +# if STRERROR_R_CHAR_P || _LIBC + s = __strerror_r (errnum, errbuf, sizeof errbuf); +# else + if (__strerror_r (errnum, errbuf, sizeof errbuf) == 0) + s = errbuf; +# endif +#endif + +#if !_LIBC + if (! s && ! (s = strerror (errnum))) + s = _("Unknown system error"); +#endif + +#if _LIBC + if (_IO_fwide (stderr, 0) > 0) + { + __fwprintf (stderr, L": %s", s); + return; + } +#endif + + fprintf (stderr, ": %s", s); +} + +static void +error_tail (int status, int errnum, const char *message, va_list args) +{ +#if _LIBC + if (_IO_fwide (stderr, 0) > 0) + { +# define ALLOCA_LIMIT 2000 + size_t len = strlen (message) + 1; + const wchar_t *wmessage = L"out of memory"; + wchar_t *wbuf = (len < ALLOCA_LIMIT + ? alloca (len * sizeof *wbuf) + : len <= SIZE_MAX / sizeof *wbuf + ? malloc (len * sizeof *wbuf) + : NULL); + + if (wbuf) + { + size_t res; + mbstate_t st; + const char *tmp = message; + memset (&st, '\0', sizeof (st)); + res = mbsrtowcs (wbuf, &tmp, len, &st); + wmessage = res == (size_t) -1 ? L"???" : wbuf; + } + + __vfwprintf (stderr, wmessage, args); + if (! (len < ALLOCA_LIMIT)) + free (wbuf); + } + else +#endif + vfprintf (stderr, message, args); + va_end (args); + + ++error_message_count; + if (errnum) + print_errno_message (errnum); +#if _LIBC + if (_IO_fwide (stderr, 0) > 0) + putwc (L'\n', stderr); + else +#endif + putc ('\n', stderr); + fflush (stderr); + if (status) + exit (status); +} + + +/* Print the program name and error message MESSAGE, which is a printf-style + format string with optional args. + If ERRNUM is nonzero, print its corresponding system error message. + Exit with status STATUS if it is nonzero. */ +void +error (int status, int errnum, const char *message, ...) +{ + va_list args; + +#if defined _LIBC && defined __libc_ptf_call + /* We do not want this call to be cut short by a thread + cancellation. Therefore disable cancellation for now. */ + int state = PTHREAD_CANCEL_ENABLE; + __libc_ptf_call (pthread_setcancelstate, (PTHREAD_CANCEL_DISABLE, &state), + 0); +#endif + + fflush (stdout); +#ifdef _LIBC + _IO_flockfile (stderr); +#endif + if (error_print_progname) + (*error_print_progname) (); + else + { +#if _LIBC + if (_IO_fwide (stderr, 0) > 0) + __fwprintf (stderr, L"%s: ", program_name); + else +#endif + fprintf (stderr, "%s: ", program_name); + } + + va_start (args, message); + error_tail (status, errnum, message, args); + +#ifdef _LIBC + _IO_funlockfile (stderr); +# ifdef __libc_ptf_call + __libc_ptf_call (pthread_setcancelstate, (state, NULL), 0); +# endif +#endif +} + +/* Sometimes we want to have at most one error per line. This + variable controls whether this mode is selected or not. */ +int error_one_per_line; + +void +error_at_line (int status, int errnum, const char *file_name, + unsigned int line_number, const char *message, ...) +{ + va_list args; + + if (error_one_per_line) + { + static const char *old_file_name; + static unsigned int old_line_number; + + if (old_line_number == line_number + && (file_name == old_file_name + || strcmp (old_file_name, file_name) == 0)) + /* Simply return and print nothing. */ + return; + + old_file_name = file_name; + old_line_number = line_number; + } + +#if defined _LIBC && defined __libc_ptf_call + /* We do not want this call to be cut short by a thread + cancellation. Therefore disable cancellation for now. */ + int state = PTHREAD_CANCEL_ENABLE; + __libc_ptf_call (pthread_setcancelstate, (PTHREAD_CANCEL_DISABLE, &state), + 0); +#endif + + fflush (stdout); +#ifdef _LIBC + _IO_flockfile (stderr); +#endif + if (error_print_progname) + (*error_print_progname) (); + else + { +#if _LIBC + if (_IO_fwide (stderr, 0) > 0) + __fwprintf (stderr, L"%s: ", program_name); + else +#endif + fprintf (stderr, "%s:", program_name); + } + + if (file_name != NULL) + { +#if _LIBC + if (_IO_fwide (stderr, 0) > 0) + __fwprintf (stderr, L"%s:%d: ", file_name, line_number); + else +#endif + fprintf (stderr, "%s:%d: ", file_name, line_number); + } + + va_start (args, message); + error_tail (status, errnum, message, args); + +#ifdef _LIBC + _IO_funlockfile (stderr); +# ifdef __libc_ptf_call + __libc_ptf_call (pthread_setcancelstate, (state, NULL), 0); +# endif +#endif +} + +#ifdef _LIBC +/* Make the weak alias. */ +# undef error +# undef error_at_line +weak_alias (__error, error) +weak_alias (__error_at_line, error_at_line) +#endif diff --git a/lib/error.h b/lib/error.h new file mode 100644 index 0000000..e5220a2 --- /dev/null +++ b/lib/error.h @@ -0,0 +1,66 @@ +/* Declaration for error-reporting function + Copyright (C) 1995, 1996, 1997, 2003 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _ERROR_H +#define _ERROR_H 1 + +#ifndef __attribute__ +/* This feature is available in gcc versions 2.5 and later. */ +# if __GNUC__ < 2 || (__GNUC__ == 2 && __GNUC_MINOR__ < 5) +# define __attribute__(Spec) /* empty */ +# endif +/* The __-protected variants of `format' and `printf' attributes + are accepted by gcc versions 2.6.4 (effectively 2.7) and later. */ +# if __GNUC__ < 2 || (__GNUC__ == 2 && __GNUC_MINOR__ < 7) +# define __format__ format +# define __printf__ printf +# endif +#endif + +#ifdef __cplusplus +extern "C" { +#endif + +/* Print a message with `fprintf (stderr, FORMAT, ...)'; + if ERRNUM is nonzero, follow it with ": " and strerror (ERRNUM). + If STATUS is nonzero, terminate the program with `exit (STATUS)'. */ + +extern void error (int __status, int __errnum, const char *__format, ...) + __attribute__ ((__format__ (__printf__, 3, 4))); + +extern void error_at_line (int __status, int __errnum, const char *__fname, + unsigned int __lineno, const char *__format, ...) + __attribute__ ((__format__ (__printf__, 5, 6))); + +/* If NULL, error will flush stdout, then print on stderr the program + name, a colon and a space. Otherwise, error will call this + function without parameters instead. */ +extern void (*error_print_progname) (void); + +/* This variable is incremented each time `error' is called. */ +extern unsigned int error_message_count; + +/* Sometimes we want to have at most one error per line. This + variable controls whether this mode is selected or not. */ +extern int error_one_per_line; + +#ifdef __cplusplus +} +#endif + +#endif /* error.h */ diff --git a/lib/exit.h b/lib/exit.h new file mode 100644 index 0000000..9dbfb7c --- /dev/null +++ b/lib/exit.h @@ -0,0 +1,32 @@ +/* exit() function. + Copyright (C) 1995, 2001 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _EXIT_H +#define _EXIT_H + +/* Get exit() declaration. */ +#include <stdlib.h> + +/* Some systems do not define EXIT_*, even with STDC_HEADERS. */ +#ifndef EXIT_SUCCESS +# define EXIT_SUCCESS 0 +#endif +#ifndef EXIT_FAILURE +# define EXIT_FAILURE 1 +#endif + +#endif /* _EXIT_H */ diff --git a/lib/exitfail.c b/lib/exitfail.c new file mode 100644 index 0000000..27d38c3 --- /dev/null +++ b/lib/exitfail.c @@ -0,0 +1,27 @@ +/* Failure exit status + + Copyright (C) 2002, 2003 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; see the file COPYING. + If not, write to the Free Software Foundation, + 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "exitfail.h" +#include "exit.h" + +int volatile exit_failure = EXIT_FAILURE; diff --git a/lib/exitfail.h b/lib/exitfail.h new file mode 100644 index 0000000..e46cf9c --- /dev/null +++ b/lib/exitfail.h @@ -0,0 +1,20 @@ +/* Failure exit status + + Copyright (C) 2002 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; see the file COPYING. + If not, write to the Free Software Foundation, + 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +extern int volatile exit_failure; diff --git a/lib/fd-safer.c b/lib/fd-safer.c new file mode 100644 index 0000000..5933bcb --- /dev/null +++ b/lib/fd-safer.c @@ -0,0 +1,59 @@ +/* Return a safer copy of a file descriptor. + + Copyright (C) 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Paul Eggert. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "unistd-safer.h" + +#include <errno.h> + +#include <unistd.h> +#ifndef STDIN_FILENO +# define STDIN_FILENO 0 +#endif +#ifndef STDERR_FILENO +# define STDERR_FILENO 2 +#endif + +/* Return FD, unless FD would be a copy of standard input, output, or + error; in that case, return a duplicate of FD, closing FD. On + failure to duplicate, close FD, set errno, and return -1. Preserve + errno if FD is negative, so that the caller can always inspect + errno when the returned value is negative. + + This function is usefully wrapped around functions that return file + descriptors, e.g., fd_safer (open ("file", O_RDONLY)). */ + +int +fd_safer (int fd) +{ + if (STDIN_FILENO <= fd && fd <= STDERR_FILENO) + { + int f = dup_safer (fd); + int e = errno; + close (fd); + errno = e; + fd = f; + } + + return fd; +} diff --git a/lib/filenamecat.c b/lib/filenamecat.c new file mode 100644 index 0000000..f947346 --- /dev/null +++ b/lib/filenamecat.c @@ -0,0 +1,126 @@ +/* Concatenate two arbitrary file names. + + Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005 Free + Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Jim Meyering. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Specification. */ +#include "filenamecat.h" + +#include <string.h> + +#include "dirname.h" +#include "xalloc.h" + +#if ! HAVE_MEMPCPY && ! defined mempcpy +# define mempcpy(D, S, N) ((void *) ((char *) memcpy (D, S, N) + (N))) +#endif + +/* Return the longest suffix of F that is a relative file name. + If it has no such suffix, return the empty string. */ + +static char const * +longest_relative_suffix (char const *f) +{ + for (f += FILE_SYSTEM_PREFIX_LEN (f); ISSLASH (*f); f++) + continue; + return f; +} + +/* Concatenate two file name components, DIR and ABASE, in + newly-allocated storage and return the result. + The resulting file name F is such that the commands "ls F" and "(cd + DIR; ls BASE)" refer to the same file, where BASE is ABASE with any + file system prefixes and leading separators removed. + Arrange for a directory separator if necessary between DIR and BASE + in the result, removing any redundant separators. + In any case, if BASE_IN_RESULT is non-NULL, set + *BASE_IN_RESULT to point to the copy of ABASE in the returned + concatenation. However, if ABASE begins with more than one slash, + set *BASE_IN_RESULT to point to the sole corresponding slash that + is copied into the result buffer. + + Report an error if memory is exhausted. */ + +char * +file_name_concat (char const *dir, char const *abase, char **base_in_result) +{ + char const *dirbase = base_name (dir); + size_t dirbaselen = base_len (dirbase); + size_t dirlen = dirbase - dir + dirbaselen; + size_t needs_separator = (dirbaselen && ! ISSLASH (dirbase[dirbaselen - 1])); + + char const *base = longest_relative_suffix (abase); + size_t baselen = strlen (base); + + char *p_concat = xmalloc (dirlen + needs_separator + baselen + 1); + char *p; + + p = mempcpy (p_concat, dir, dirlen); + *p = DIRECTORY_SEPARATOR; + p += needs_separator; + + if (base_in_result) + *base_in_result = p - IS_ABSOLUTE_FILE_NAME (abase); + + p = mempcpy (p, base, baselen); + *p = '\0'; + + return p_concat; +} + +#ifdef TEST_FILE_NAME_CONCAT +# include <stdlib.h> +# include <stdio.h> +int +main () +{ + static char const *const tests[][3] = + { + {"a", "b", "a/b"}, + {"a/", "b", "a/b"}, + {"a/", "/b", "a/b"}, + {"a", "/b", "a/b"}, + + {"/", "b", "/b"}, + {"/", "/b", "/b"}, + {"/", "/", "/"}, + {"a", "/", "a/"}, /* this might deserve a diagnostic */ + {"/a", "/", "/a/"}, /* this might deserve a diagnostic */ + {"a", "//b", "a/b"}, + }; + size_t i; + bool fail = false; + for (i = 0; i < sizeof tests / sizeof tests[0]; i++) + { + char *base_in_result; + char const *const *t = tests[i]; + char *res = file_name_concat (t[0], t[1], &base_in_result); + if (strcmp (res, t[2]) != 0) + { + printf ("got %s, expected %s\n", res, t[2]); + fail = true; + } + } + exit (fail ? EXIT_FAILURE : EXIT_SUCCESS); +} +#endif diff --git a/lib/filenamecat.h b/lib/filenamecat.h new file mode 100644 index 0000000..c943b67 --- /dev/null +++ b/lib/filenamecat.h @@ -0,0 +1,22 @@ +/* Concatenate two arbitrary file names. + + Copyright (C) 1996, 1997, 2003, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Jim Meyering. */ + +char *file_name_concat (char const *dir, char const *base, + char **base_in_result); diff --git a/lib/fncase.c b/lib/fncase.c new file mode 100644 index 0000000..2842428 --- /dev/null +++ b/lib/fncase.c @@ -0,0 +1,155 @@ +/* fncase.c -- CVS support for case insensitive file systems. + Jim Blandy <jimb@cyclic.com> + + This file is part of GNU CVS. + + GNU CVS is free software; you can redistribute it and/or modify it + under the terms of the GNU General Public License as published by the + Free Software Foundation; either version 2, or (at your option) any + later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. */ + +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif + +#include "system.h" + +/* The equivalence class mapping for filenames. + Windows NT filenames are case-insensitive, but case-preserving. + Both / and \ are path element separators. + Thus, this table maps both upper and lower case to lower case, and + both / and \ to /. */ + +#if 0 +main () +{ + int c; + + for (c = 0; c < 256; c++) + { + int t; + + if (c == '\\') + t = '/'; + else + t = tolower (c); + + if ((c & 0x7) == 0x0) + printf (" "); + printf ("0x%02x,", t); + if ((c & 0x7) == 0x7) + putchar ('\n'); + else if ((c & 0x7) == 0x3) + putchar (' '); + } +} +#endif + +/* Under Windows NT, filenames are case-insensitive but case-preserving, + and both \ and / are path element separators. */ +unsigned char +WNT_filename_classes[] = +{ + 0x00,0x01,0x02,0x03, 0x04,0x05,0x06,0x07, + 0x08,0x09,0x0a,0x0b, 0x0c,0x0d,0x0e,0x0f, + 0x10,0x11,0x12,0x13, 0x14,0x15,0x16,0x17, + 0x18,0x19,0x1a,0x1b, 0x1c,0x1d,0x1e,0x1f, + 0x20,0x21,0x22,0x23, 0x24,0x25,0x26,0x27, + 0x28,0x29,0x2a,0x2b, 0x2c,0x2d,0x2e,0x2f, + 0x30,0x31,0x32,0x33, 0x34,0x35,0x36,0x37, + 0x38,0x39,0x3a,0x3b, 0x3c,0x3d,0x3e,0x3f, + 0x40,0x61,0x62,0x63, 0x64,0x65,0x66,0x67, + 0x68,0x69,0x6a,0x6b, 0x6c,0x6d,0x6e,0x6f, + 0x70,0x71,0x72,0x73, 0x74,0x75,0x76,0x77, + 0x78,0x79,0x7a,0x5b, 0x2f,0x5d,0x5e,0x5f, + 0x60,0x61,0x62,0x63, 0x64,0x65,0x66,0x67, + 0x68,0x69,0x6a,0x6b, 0x6c,0x6d,0x6e,0x6f, + 0x70,0x71,0x72,0x73, 0x74,0x75,0x76,0x77, + 0x78,0x79,0x7a,0x7b, 0x7c,0x7d,0x7e,0x7f, + 0x80,0x81,0x82,0x83, 0x84,0x85,0x86,0x87, + 0x88,0x89,0x8a,0x8b, 0x8c,0x8d,0x8e,0x8f, + 0x90,0x91,0x92,0x93, 0x94,0x95,0x96,0x97, + 0x98,0x99,0x9a,0x9b, 0x9c,0x9d,0x9e,0x9f, + 0xa0,0xa1,0xa2,0xa3, 0xa4,0xa5,0xa6,0xa7, + 0xa8,0xa9,0xaa,0xab, 0xac,0xad,0xae,0xaf, + 0xb0,0xb1,0xb2,0xb3, 0xb4,0xb5,0xb6,0xb7, + 0xb8,0xb9,0xba,0xbb, 0xbc,0xbd,0xbe,0xbf, + 0xc0,0xc1,0xc2,0xc3, 0xc4,0xc5,0xc6,0xc7, + 0xc8,0xc9,0xca,0xcb, 0xcc,0xcd,0xce,0xcf, + 0xd0,0xd1,0xd2,0xd3, 0xd4,0xd5,0xd6,0xd7, + 0xd8,0xd9,0xda,0xdb, 0xdc,0xdd,0xde,0xdf, + 0xe0,0xe1,0xe2,0xe3, 0xe4,0xe5,0xe6,0xe7, + 0xe8,0xe9,0xea,0xeb, 0xec,0xed,0xee,0xef, + 0xf0,0xf1,0xf2,0xf3, 0xf4,0xf5,0xf6,0xf7, + 0xf8,0xf9,0xfa,0xfb, 0xfc,0xfd,0xfe,0xff, +}; + +/* Same as WNT_filename_classes, but do not fold `\' into `/'. */ +unsigned char +OSX_filename_classes[] = +{ + 0x00,0x01,0x02,0x03, 0x04,0x05,0x06,0x07, + 0x08,0x09,0x0a,0x0b, 0x0c,0x0d,0x0e,0x0f, + 0x10,0x11,0x12,0x13, 0x14,0x15,0x16,0x17, + 0x18,0x19,0x1a,0x1b, 0x1c,0x1d,0x1e,0x1f, + 0x20,0x21,0x22,0x23, 0x24,0x25,0x26,0x27, + 0x28,0x29,0x2a,0x2b, 0x2c,0x2d,0x2e,0x2f, + 0x30,0x31,0x32,0x33, 0x34,0x35,0x36,0x37, + 0x38,0x39,0x3a,0x3b, 0x3c,0x3d,0x3e,0x3f, + 0x40,0x61,0x62,0x63, 0x64,0x65,0x66,0x67, + 0x68,0x69,0x6a,0x6b, 0x6c,0x6d,0x6e,0x6f, + 0x70,0x71,0x72,0x73, 0x74,0x75,0x76,0x77, + 0x78,0x79,0x7a,0x5b, 0x5c,0x5d,0x5e,0x5f, + 0x60,0x61,0x62,0x63, 0x64,0x65,0x66,0x67, + 0x68,0x69,0x6a,0x6b, 0x6c,0x6d,0x6e,0x6f, + 0x70,0x71,0x72,0x73, 0x74,0x75,0x76,0x77, + 0x78,0x79,0x7a,0x7b, 0x7c,0x7d,0x7e,0x7f, + 0x80,0x81,0x82,0x83, 0x84,0x85,0x86,0x87, + 0x88,0x89,0x8a,0x8b, 0x8c,0x8d,0x8e,0x8f, + 0x90,0x91,0x92,0x93, 0x94,0x95,0x96,0x97, + 0x98,0x99,0x9a,0x9b, 0x9c,0x9d,0x9e,0x9f, + 0xa0,0xa1,0xa2,0xa3, 0xa4,0xa5,0xa6,0xa7, + 0xa8,0xa9,0xaa,0xab, 0xac,0xad,0xae,0xaf, + 0xb0,0xb1,0xb2,0xb3, 0xb4,0xb5,0xb6,0xb7, + 0xb8,0xb9,0xba,0xbb, 0xbc,0xbd,0xbe,0xbf, + 0xc0,0xc1,0xc2,0xc3, 0xc4,0xc5,0xc6,0xc7, + 0xc8,0xc9,0xca,0xcb, 0xcc,0xcd,0xce,0xcf, + 0xd0,0xd1,0xd2,0xd3, 0xd4,0xd5,0xd6,0xd7, + 0xd8,0xd9,0xda,0xdb, 0xdc,0xdd,0xde,0xdf, + 0xe0,0xe1,0xe2,0xe3, 0xe4,0xe5,0xe6,0xe7, + 0xe8,0xe9,0xea,0xeb, 0xec,0xed,0xee,0xef, + 0xf0,0xf1,0xf2,0xf3, 0xf4,0xf5,0xf6,0xf7, + 0xf8,0xf9,0xfa,0xfb, 0xfc,0xfd,0xfe,0xff, +}; + +/* Like strcmp, but with the appropriate tweaks for file names. + Under Windows NT, filenames are case-insensitive but case-preserving, + and both \ and / are path element separators. Under Mac OS X, filenames + are case-insensitive but case-preserving. */ +int +fncmp (const char *n1, const char *n2) +{ + while (*n1 && *n2 + && (FOLD_FN_CHAR(*n1) + == FOLD_FN_CHAR(*n2))) + n1++, n2++; + return (FOLD_FN_CHAR(*n1) - FOLD_FN_CHAR(*n2)); +} + +/* Fold characters in FILENAME to their canonical forms. + If FOLD_FN_CHAR is not #defined, the system provides a default + definition for this. */ +void +fnfold (char *filename) +{ + while (*filename) + { + *filename = FOLD_FN_CHAR (*filename); + filename++; + } +} diff --git a/lib/fnmatch.c b/lib/fnmatch.c new file mode 100644 index 0000000..06416f6 --- /dev/null +++ b/lib/fnmatch.c @@ -0,0 +1,383 @@ +/* Copyright (C) 1991,1992,1993,1996,1997,1998,1999,2000,2001,2002,2003,2004,2005 + Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Enable GNU extensions in fnmatch.h. */ +#ifndef _GNU_SOURCE +# define _GNU_SOURCE 1 +#endif + +#if ! defined __builtin_expect && __GNUC__ < 3 +# define __builtin_expect(expr, expected) (expr) +#endif + +#include <fnmatch.h> + +#include <alloca.h> +#include <assert.h> +#include <ctype.h> +#include <errno.h> +#include <stddef.h> +#include <stdbool.h> +#include <stdlib.h> +#include <string.h> + +#define WIDE_CHAR_SUPPORT \ + (HAVE_WCTYPE_H && HAVE_WCHAR_H && HAVE_BTOWC \ + && HAVE_WMEMCHR && (HAVE_WMEMCPY || HAVE_WMEMPCPY)) + +/* For platform which support the ISO C amendement 1 functionality we + support user defined character classes. */ +#if defined _LIBC || WIDE_CHAR_SUPPORT +/* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>. */ +# include <wchar.h> +# include <wctype.h> +#endif + +/* We need some of the locale data (the collation sequence information) + but there is no interface to get this information in general. Therefore + we support a correct implementation only in glibc. */ +#ifdef _LIBC +# include "../locale/localeinfo.h" +# include "../locale/elem-hash.h" +# include "../locale/coll-lookup.h" +# include <shlib-compat.h> + +# define CONCAT(a,b) __CONCAT(a,b) +# define mbsrtowcs __mbsrtowcs +# define fnmatch __fnmatch +extern int fnmatch (const char *pattern, const char *string, int flags); +#endif + +#ifndef SIZE_MAX +# define SIZE_MAX ((size_t) -1) +#endif + +/* We often have to test for FNM_FILE_NAME and FNM_PERIOD being both set. */ +#define NO_LEADING_PERIOD(flags) \ + ((flags & (FNM_FILE_NAME | FNM_PERIOD)) == (FNM_FILE_NAME | FNM_PERIOD)) + +/* Comment out all this code if we are using the GNU C Library, and are not + actually compiling the library itself, and have not detected a bug + in the library. This code is part of the GNU C + Library, but also included in many other GNU distributions. Compiling + and linking in this code is a waste when using the GNU C library + (especially if it is a shared library). Rather than having every GNU + program understand `configure --with-gnu-libc' and omit the object files, + it is simpler to just do this in the source for each such file. */ + +#if defined _LIBC || !defined __GNU_LIBRARY__ || !HAVE_FNMATCH_GNU + + +# if defined STDC_HEADERS || !defined isascii +# define ISASCII(c) 1 +# else +# define ISASCII(c) isascii(c) +# endif + +# ifdef isblank +# define ISBLANK(c) (ISASCII (c) && isblank (c)) +# else +# define ISBLANK(c) ((c) == ' ' || (c) == '\t') +# endif +# ifdef isgraph +# define ISGRAPH(c) (ISASCII (c) && isgraph (c)) +# else +# define ISGRAPH(c) (ISASCII (c) && isprint (c) && !isspace (c)) +# endif + +# define ISPRINT(c) (ISASCII (c) && isprint (c)) +# define ISDIGIT(c) (ISASCII (c) && isdigit (c)) +# define ISALNUM(c) (ISASCII (c) && isalnum (c)) +# define ISALPHA(c) (ISASCII (c) && isalpha (c)) +# define ISCNTRL(c) (ISASCII (c) && iscntrl (c)) +# define ISLOWER(c) (ISASCII (c) && islower (c)) +# define ISPUNCT(c) (ISASCII (c) && ispunct (c)) +# define ISSPACE(c) (ISASCII (c) && isspace (c)) +# define ISUPPER(c) (ISASCII (c) && isupper (c)) +# define ISXDIGIT(c) (ISASCII (c) && isxdigit (c)) + +# define STREQ(s1, s2) ((strcmp (s1, s2) == 0)) + +# if defined _LIBC || WIDE_CHAR_SUPPORT +/* The GNU C library provides support for user-defined character classes + and the functions from ISO C amendement 1. */ +# ifdef CHARCLASS_NAME_MAX +# define CHAR_CLASS_MAX_LENGTH CHARCLASS_NAME_MAX +# else +/* This shouldn't happen but some implementation might still have this + problem. Use a reasonable default value. */ +# define CHAR_CLASS_MAX_LENGTH 256 +# endif + +# ifdef _LIBC +# define IS_CHAR_CLASS(string) __wctype (string) +# else +# define IS_CHAR_CLASS(string) wctype (string) +# endif + +# ifdef _LIBC +# define ISWCTYPE(WC, WT) __iswctype (WC, WT) +# else +# define ISWCTYPE(WC, WT) iswctype (WC, WT) +# endif + +# if (HAVE_MBSTATE_T && HAVE_MBSRTOWCS) || _LIBC +/* In this case we are implementing the multibyte character handling. */ +# define HANDLE_MULTIBYTE 1 +# endif + +# else +# define CHAR_CLASS_MAX_LENGTH 6 /* Namely, `xdigit'. */ + +# define IS_CHAR_CLASS(string) \ + (STREQ (string, "alpha") || STREQ (string, "upper") \ + || STREQ (string, "lower") || STREQ (string, "digit") \ + || STREQ (string, "alnum") || STREQ (string, "xdigit") \ + || STREQ (string, "space") || STREQ (string, "print") \ + || STREQ (string, "punct") || STREQ (string, "graph") \ + || STREQ (string, "cntrl") || STREQ (string, "blank")) +# endif + +/* Avoid depending on library functions or files + whose names are inconsistent. */ + +/* Global variable. */ +static int posixly_correct; + +# ifndef internal_function +/* Inside GNU libc we mark some function in a special way. In other + environments simply ignore the marking. */ +# define internal_function +# endif + +/* Note that this evaluates C many times. */ +# ifdef _LIBC +# define FOLD(c) ((flags & FNM_CASEFOLD) ? tolower (c) : (c)) +# else +# define FOLD(c) ((flags & FNM_CASEFOLD) && ISUPPER (c) ? tolower (c) : (c)) +# endif +# define CHAR char +# define UCHAR unsigned char +# define INT int +# define FCT internal_fnmatch +# define EXT ext_match +# define END end_pattern +# define L(CS) CS +# ifdef _LIBC +# define BTOWC(C) __btowc (C) +# else +# define BTOWC(C) btowc (C) +# endif +# define STRLEN(S) strlen (S) +# define STRCAT(D, S) strcat (D, S) +# ifdef _LIBC +# define MEMPCPY(D, S, N) __mempcpy (D, S, N) +# else +# if HAVE_MEMPCPY +# define MEMPCPY(D, S, N) mempcpy (D, S, N) +# else +# define MEMPCPY(D, S, N) ((void *) ((char *) memcpy (D, S, N) + (N))) +# endif +# endif +# define MEMCHR(S, C, N) memchr (S, C, N) +# define STRCOLL(S1, S2) strcoll (S1, S2) +# include "fnmatch_loop.c" + + +# if HANDLE_MULTIBYTE +# define FOLD(c) ((flags & FNM_CASEFOLD) ? towlower (c) : (c)) +# define CHAR wchar_t +# define UCHAR wint_t +# define INT wint_t +# define FCT internal_fnwmatch +# define EXT ext_wmatch +# define END end_wpattern +# define L(CS) L##CS +# define BTOWC(C) (C) +# ifdef _LIBC +# define STRLEN(S) __wcslen (S) +# define STRCAT(D, S) __wcscat (D, S) +# define MEMPCPY(D, S, N) __wmempcpy (D, S, N) +# else +# define STRLEN(S) wcslen (S) +# define STRCAT(D, S) wcscat (D, S) +# if HAVE_WMEMPCPY +# define MEMPCPY(D, S, N) wmempcpy (D, S, N) +# else +# define MEMPCPY(D, S, N) (wmemcpy (D, S, N) + (N)) +# endif +# endif +# define MEMCHR(S, C, N) wmemchr (S, C, N) +# define STRCOLL(S1, S2) wcscoll (S1, S2) +# define WIDE_CHAR_VERSION 1 + +# undef IS_CHAR_CLASS +/* We have to convert the wide character string in a multibyte string. But + we know that the character class names consist of alphanumeric characters + from the portable character set, and since the wide character encoding + for a member of the portable character set is the same code point as + its single-byte encoding, we can use a simplified method to convert the + string to a multibyte character string. */ +static wctype_t +is_char_class (const wchar_t *wcs) +{ + char s[CHAR_CLASS_MAX_LENGTH + 1]; + char *cp = s; + + do + { + /* Test for a printable character from the portable character set. */ +# ifdef _LIBC + if (*wcs < 0x20 || *wcs > 0x7e + || *wcs == 0x24 || *wcs == 0x40 || *wcs == 0x60) + return (wctype_t) 0; +# else + switch (*wcs) + { + case L' ': case L'!': case L'"': case L'#': case L'%': + case L'&': case L'\'': case L'(': case L')': case L'*': + case L'+': case L',': case L'-': case L'.': case L'/': + case L'0': case L'1': case L'2': case L'3': case L'4': + case L'5': case L'6': case L'7': case L'8': case L'9': + case L':': case L';': case L'<': case L'=': case L'>': + case L'?': + case L'A': case L'B': case L'C': case L'D': case L'E': + case L'F': case L'G': case L'H': case L'I': case L'J': + case L'K': case L'L': case L'M': case L'N': case L'O': + case L'P': case L'Q': case L'R': case L'S': case L'T': + case L'U': case L'V': case L'W': case L'X': case L'Y': + case L'Z': + case L'[': case L'\\': case L']': case L'^': case L'_': + case L'a': case L'b': case L'c': case L'd': case L'e': + case L'f': case L'g': case L'h': case L'i': case L'j': + case L'k': case L'l': case L'm': case L'n': case L'o': + case L'p': case L'q': case L'r': case L's': case L't': + case L'u': case L'v': case L'w': case L'x': case L'y': + case L'z': case L'{': case L'|': case L'}': case L'~': + break; + default: + return (wctype_t) 0; + } +# endif + + /* Avoid overrunning the buffer. */ + if (cp == s + CHAR_CLASS_MAX_LENGTH) + return (wctype_t) 0; + + *cp++ = (char) *wcs++; + } + while (*wcs != L'\0'); + + *cp = '\0'; + +# ifdef _LIBC + return __wctype (s); +# else + return wctype (s); +# endif +} +# define IS_CHAR_CLASS(string) is_char_class (string) + +# include "fnmatch_loop.c" +# endif + + +int +fnmatch (const char *pattern, const char *string, int flags) +{ +# if HANDLE_MULTIBYTE +# define ALLOCA_LIMIT 2000 + if (__builtin_expect (MB_CUR_MAX, 1) != 1) + { + mbstate_t ps; + size_t patsize; + size_t strsize; + size_t totsize; + wchar_t *wpattern; + wchar_t *wstring; + int res; + + /* Calculate the size needed to convert the strings to + wide characters. */ + memset (&ps, '\0', sizeof (ps)); + patsize = mbsrtowcs (NULL, &pattern, 0, &ps) + 1; + if (__builtin_expect (patsize != 0, 1)) + { + assert (mbsinit (&ps)); + strsize = mbsrtowcs (NULL, &string, 0, &ps) + 1; + if (__builtin_expect (strsize != 0, 1)) + { + assert (mbsinit (&ps)); + totsize = patsize + strsize; + if (__builtin_expect (! (patsize <= totsize + && totsize <= SIZE_MAX / sizeof (wchar_t)), + 0)) + { + errno = ENOMEM; + return -1; + } + + /* Allocate room for the wide characters. */ + if (__builtin_expect (totsize < ALLOCA_LIMIT, 1)) + wpattern = (wchar_t *) alloca (totsize * sizeof (wchar_t)); + else + { + wpattern = malloc (totsize * sizeof (wchar_t)); + if (__builtin_expect (! wpattern, 0)) + { + errno = ENOMEM; + return -1; + } + } + wstring = wpattern + patsize; + + /* Convert the strings into wide characters. */ + mbsrtowcs (wpattern, &pattern, patsize, &ps); + assert (mbsinit (&ps)); + mbsrtowcs (wstring, &string, strsize, &ps); + + res = internal_fnwmatch (wpattern, wstring, wstring + strsize - 1, + flags & FNM_PERIOD, flags); + + if (__builtin_expect (! (totsize < ALLOCA_LIMIT), 0)) + free (wpattern); + return res; + } + } + } + +# endif /* HANDLE_MULTIBYTE */ + + return internal_fnmatch (pattern, string, string + strlen (string), + flags & FNM_PERIOD, flags); +} + +# ifdef _LIBC +# undef fnmatch +versioned_symbol (libc, __fnmatch, fnmatch, GLIBC_2_2_3); +# if SHLIB_COMPAT(libc, GLIBC_2_0, GLIBC_2_2_3) +strong_alias (__fnmatch, __fnmatch_old) +compat_symbol (libc, __fnmatch_old, fnmatch, GLIBC_2_0); +# endif +libc_hidden_ver (__fnmatch, fnmatch) +# endif + +#endif /* _LIBC or not __GNU_LIBRARY__. */ diff --git a/lib/fnmatch_.h b/lib/fnmatch_.h new file mode 100644 index 0000000..fecada5 --- /dev/null +++ b/lib/fnmatch_.h @@ -0,0 +1,63 @@ +/* Copyright (C) 1991, 1992, 1993, 1996, 1997, 1998, 1999, 2001, 2002, 2003, + 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _FNMATCH_H +# define _FNMATCH_H 1 + +# ifdef __cplusplus +extern "C" { +# endif + +/* We #undef these before defining them because some losing systems + (HP-UX A.08.07 for example) define these in <unistd.h>. */ +# undef FNM_PATHNAME +# undef FNM_NOESCAPE +# undef FNM_PERIOD + +/* Bits set in the FLAGS argument to `fnmatch'. */ +# define FNM_PATHNAME (1 << 0) /* No wildcard can ever match `/'. */ +# define FNM_NOESCAPE (1 << 1) /* Backslashes don't quote special chars. */ +# define FNM_PERIOD (1 << 2) /* Leading `.' is matched only explicitly. */ + +# if !defined _POSIX_C_SOURCE || _POSIX_C_SOURCE < 2 || defined _GNU_SOURCE +# define FNM_FILE_NAME FNM_PATHNAME /* Preferred GNU name. */ +# define FNM_LEADING_DIR (1 << 3) /* Ignore `/...' after a match. */ +# define FNM_CASEFOLD (1 << 4) /* Compare without regard to case. */ +# define FNM_EXTMATCH (1 << 5) /* Use ksh-like extended matching. */ +# endif + +/* Value returned by `fnmatch' if STRING does not match PATTERN. */ +# define FNM_NOMATCH 1 + +/* This value is returned if the implementation does not support + `fnmatch'. Since this is not the case here it will never be + returned but the conformance test suites still require the symbol + to be defined. */ +# ifdef _XOPEN_SOURCE +# define FNM_NOSYS (-1) +# endif + +/* Match NAME against the file name pattern PATTERN, + returning zero if it matches, FNM_NOMATCH if not. */ +extern int fnmatch (const char *__pattern, const char *__name, + int __flags); + +# ifdef __cplusplus +} +# endif + +#endif /* fnmatch.h */ diff --git a/lib/fnmatch_loop.c b/lib/fnmatch_loop.c new file mode 100644 index 0000000..35d5980 --- /dev/null +++ b/lib/fnmatch_loop.c @@ -0,0 +1,1192 @@ +/* Copyright (C) 1991,1992,1993,1996,1997,1998,1999,2000,2001,2002,2003,2004,2005 + Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Match STRING against the file name pattern PATTERN, returning zero if + it matches, nonzero if not. */ +static int EXT (INT opt, const CHAR *pattern, const CHAR *string, + const CHAR *string_end, bool no_leading_period, int flags) + internal_function; +static const CHAR *END (const CHAR *patternp) internal_function; + +static int +internal_function +FCT (const CHAR *pattern, const CHAR *string, const CHAR *string_end, + bool no_leading_period, int flags) +{ + register const CHAR *p = pattern, *n = string; + register UCHAR c; +#ifdef _LIBC +# if WIDE_CHAR_VERSION + const char *collseq = (const char *) + _NL_CURRENT(LC_COLLATE, _NL_COLLATE_COLLSEQWC); +# else + const UCHAR *collseq = (const UCHAR *) + _NL_CURRENT(LC_COLLATE, _NL_COLLATE_COLLSEQMB); +# endif +#endif + + while ((c = *p++) != L('\0')) + { + bool new_no_leading_period = false; + c = FOLD (c); + + switch (c) + { + case L('?'): + if (__builtin_expect (flags & FNM_EXTMATCH, 0) && *p == '(') + { + int res; + + res = EXT (c, p, n, string_end, no_leading_period, + flags); + if (res != -1) + return res; + } + + if (n == string_end) + return FNM_NOMATCH; + else if (*n == L('/') && (flags & FNM_FILE_NAME)) + return FNM_NOMATCH; + else if (*n == L('.') && no_leading_period) + return FNM_NOMATCH; + break; + + case L('\\'): + if (!(flags & FNM_NOESCAPE)) + { + c = *p++; + if (c == L('\0')) + /* Trailing \ loses. */ + return FNM_NOMATCH; + c = FOLD (c); + } + if (n == string_end || FOLD ((UCHAR) *n) != c) + return FNM_NOMATCH; + break; + + case L('*'): + if (__builtin_expect (flags & FNM_EXTMATCH, 0) && *p == '(') + { + int res; + + res = EXT (c, p, n, string_end, no_leading_period, + flags); + if (res != -1) + return res; + } + + if (n != string_end && *n == L('.') && no_leading_period) + return FNM_NOMATCH; + + for (c = *p++; c == L('?') || c == L('*'); c = *p++) + { + if (*p == L('(') && (flags & FNM_EXTMATCH) != 0) + { + const CHAR *endp = END (p); + if (endp != p) + { + /* This is a pattern. Skip over it. */ + p = endp; + continue; + } + } + + if (c == L('?')) + { + /* A ? needs to match one character. */ + if (n == string_end) + /* There isn't another character; no match. */ + return FNM_NOMATCH; + else if (*n == L('/') + && __builtin_expect (flags & FNM_FILE_NAME, 0)) + /* A slash does not match a wildcard under + FNM_FILE_NAME. */ + return FNM_NOMATCH; + else + /* One character of the string is consumed in matching + this ? wildcard, so *??? won't match if there are + less than three characters. */ + ++n; + } + } + + if (c == L('\0')) + /* The wildcard(s) is/are the last element of the pattern. + If the name is a file name and contains another slash + this means it cannot match, unless the FNM_LEADING_DIR + flag is set. */ + { + int result = (flags & FNM_FILE_NAME) == 0 ? 0 : FNM_NOMATCH; + + if (flags & FNM_FILE_NAME) + { + if (flags & FNM_LEADING_DIR) + result = 0; + else + { + if (MEMCHR (n, L('/'), string_end - n) == NULL) + result = 0; + } + } + + return result; + } + else + { + const CHAR *endp; + + endp = MEMCHR (n, (flags & FNM_FILE_NAME) ? L('/') : L('\0'), + string_end - n); + if (endp == NULL) + endp = string_end; + + if (c == L('[') + || (__builtin_expect (flags & FNM_EXTMATCH, 0) != 0 + && (c == L('@') || c == L('+') || c == L('!')) + && *p == L('('))) + { + int flags2 = ((flags & FNM_FILE_NAME) + ? flags : (flags & ~FNM_PERIOD)); + bool no_leading_period2 = no_leading_period; + + for (--p; n < endp; ++n, no_leading_period2 = false) + if (FCT (p, n, string_end, no_leading_period2, flags2) + == 0) + return 0; + } + else if (c == L('/') && (flags & FNM_FILE_NAME)) + { + while (n < string_end && *n != L('/')) + ++n; + if (n < string_end && *n == L('/') + && (FCT (p, n + 1, string_end, flags & FNM_PERIOD, flags) + == 0)) + return 0; + } + else + { + int flags2 = ((flags & FNM_FILE_NAME) + ? flags : (flags & ~FNM_PERIOD)); + int no_leading_period2 = no_leading_period; + + if (c == L('\\') && !(flags & FNM_NOESCAPE)) + c = *p; + c = FOLD (c); + for (--p; n < endp; ++n, no_leading_period2 = false) + if (FOLD ((UCHAR) *n) == c + && (FCT (p, n, string_end, no_leading_period2, flags2) + == 0)) + return 0; + } + } + + /* If we come here no match is possible with the wildcard. */ + return FNM_NOMATCH; + + case L('['): + { + /* Nonzero if the sense of the character class is inverted. */ + register bool not; + CHAR cold; + UCHAR fn; + + if (posixly_correct == 0) + posixly_correct = getenv ("POSIXLY_CORRECT") != NULL ? 1 : -1; + + if (n == string_end) + return FNM_NOMATCH; + + if (*n == L('.') && no_leading_period) + return FNM_NOMATCH; + + if (*n == L('/') && (flags & FNM_FILE_NAME)) + /* `/' cannot be matched. */ + return FNM_NOMATCH; + + not = (*p == L('!') || (posixly_correct < 0 && *p == L('^'))); + if (not) + ++p; + + fn = FOLD ((UCHAR) *n); + + c = *p++; + for (;;) + { + if (!(flags & FNM_NOESCAPE) && c == L('\\')) + { + if (*p == L('\0')) + return FNM_NOMATCH; + c = FOLD ((UCHAR) *p); + ++p; + + if (c == fn) + goto matched; + } + else if (c == L('[') && *p == L(':')) + { + /* Leave room for the null. */ + CHAR str[CHAR_CLASS_MAX_LENGTH + 1]; + size_t c1 = 0; +#if defined _LIBC || WIDE_CHAR_SUPPORT + wctype_t wt; +#endif + const CHAR *startp = p; + + for (;;) + { + if (c1 == CHAR_CLASS_MAX_LENGTH) + /* The name is too long and therefore the pattern + is ill-formed. */ + return FNM_NOMATCH; + + c = *++p; + if (c == L(':') && p[1] == L(']')) + { + p += 2; + break; + } + if (c < L('a') || c >= L('z')) + { + /* This cannot possibly be a character class name. + Match it as a normal range. */ + p = startp; + c = L('['); + goto normal_bracket; + } + str[c1++] = c; + } + str[c1] = L('\0'); + +#if defined _LIBC || WIDE_CHAR_SUPPORT + wt = IS_CHAR_CLASS (str); + if (wt == 0) + /* Invalid character class name. */ + return FNM_NOMATCH; + +# if defined _LIBC && ! WIDE_CHAR_VERSION + /* The following code is glibc specific but does + there a good job in speeding up the code since + we can avoid the btowc() call. */ + if (_ISCTYPE ((UCHAR) *n, wt)) + goto matched; +# else + if (ISWCTYPE (BTOWC ((UCHAR) *n), wt)) + goto matched; +# endif +#else + if ((STREQ (str, L("alnum")) && ISALNUM ((UCHAR) *n)) + || (STREQ (str, L("alpha")) && ISALPHA ((UCHAR) *n)) + || (STREQ (str, L("blank")) && ISBLANK ((UCHAR) *n)) + || (STREQ (str, L("cntrl")) && ISCNTRL ((UCHAR) *n)) + || (STREQ (str, L("digit")) && ISDIGIT ((UCHAR) *n)) + || (STREQ (str, L("graph")) && ISGRAPH ((UCHAR) *n)) + || (STREQ (str, L("lower")) && ISLOWER ((UCHAR) *n)) + || (STREQ (str, L("print")) && ISPRINT ((UCHAR) *n)) + || (STREQ (str, L("punct")) && ISPUNCT ((UCHAR) *n)) + || (STREQ (str, L("space")) && ISSPACE ((UCHAR) *n)) + || (STREQ (str, L("upper")) && ISUPPER ((UCHAR) *n)) + || (STREQ (str, L("xdigit")) && ISXDIGIT ((UCHAR) *n))) + goto matched; +#endif + c = *p++; + } +#ifdef _LIBC + else if (c == L('[') && *p == L('=')) + { + UCHAR str[1]; + uint32_t nrules = + _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES); + const CHAR *startp = p; + + c = *++p; + if (c == L('\0')) + { + p = startp; + c = L('['); + goto normal_bracket; + } + str[0] = c; + + c = *++p; + if (c != L('=') || p[1] != L(']')) + { + p = startp; + c = L('['); + goto normal_bracket; + } + p += 2; + + if (nrules == 0) + { + if ((UCHAR) *n == str[0]) + goto matched; + } + else + { + const int32_t *table; +# if WIDE_CHAR_VERSION + const int32_t *weights; + const int32_t *extra; +# else + const unsigned char *weights; + const unsigned char *extra; +# endif + const int32_t *indirect; + int32_t idx; + const UCHAR *cp = (const UCHAR *) str; + + /* This #include defines a local function! */ +# if WIDE_CHAR_VERSION +# include <locale/weightwc.h> +# else +# include <locale/weight.h> +# endif + +# if WIDE_CHAR_VERSION + table = (const int32_t *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEWC); + weights = (const int32_t *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_WEIGHTWC); + extra = (const int32_t *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_EXTRAWC); + indirect = (const int32_t *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_INDIRECTWC); +# else + table = (const int32_t *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB); + weights = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_WEIGHTMB); + extra = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_EXTRAMB); + indirect = (const int32_t *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_INDIRECTMB); +# endif + + idx = findidx (&cp); + if (idx != 0) + { + /* We found a table entry. Now see whether the + character we are currently at has the same + equivalance class value. */ + int len = weights[idx]; + int32_t idx2; + const UCHAR *np = (const UCHAR *) n; + + idx2 = findidx (&np); + if (idx2 != 0 && len == weights[idx2]) + { + int cnt = 0; + + while (cnt < len + && (weights[idx + 1 + cnt] + == weights[idx2 + 1 + cnt])) + ++cnt; + + if (cnt == len) + goto matched; + } + } + } + + c = *p++; + } +#endif + else if (c == L('\0')) + /* [ (unterminated) loses. */ + return FNM_NOMATCH; + else + { + bool is_range = false; + +#ifdef _LIBC + bool is_seqval = false; + + if (c == L('[') && *p == L('.')) + { + uint32_t nrules = + _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES); + const CHAR *startp = p; + size_t c1 = 0; + + while (1) + { + c = *++p; + if (c == L('.') && p[1] == L(']')) + { + p += 2; + break; + } + if (c == '\0') + return FNM_NOMATCH; + ++c1; + } + + /* We have to handling the symbols differently in + ranges since then the collation sequence is + important. */ + is_range = *p == L('-') && p[1] != L('\0'); + + if (nrules == 0) + { + /* There are no names defined in the collation + data. Therefore we only accept the trivial + names consisting of the character itself. */ + if (c1 != 1) + return FNM_NOMATCH; + + if (!is_range && *n == startp[1]) + goto matched; + + cold = startp[1]; + c = *p++; + } + else + { + int32_t table_size; + const int32_t *symb_table; +# ifdef WIDE_CHAR_VERSION + char str[c1]; + size_t strcnt; +# else +# define str (startp + 1) +# endif + const unsigned char *extra; + int32_t idx; + int32_t elem; + int32_t second; + int32_t hash; + +# ifdef WIDE_CHAR_VERSION + /* We have to convert the name to a single-byte + string. This is possible since the names + consist of ASCII characters and the internal + representation is UCS4. */ + for (strcnt = 0; strcnt < c1; ++strcnt) + str[strcnt] = startp[1 + strcnt]; +# endif + + table_size = + _NL_CURRENT_WORD (LC_COLLATE, + _NL_COLLATE_SYMB_HASH_SIZEMB); + symb_table = (const int32_t *) + _NL_CURRENT (LC_COLLATE, + _NL_COLLATE_SYMB_TABLEMB); + extra = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, + _NL_COLLATE_SYMB_EXTRAMB); + + /* Locate the character in the hashing table. */ + hash = elem_hash (str, c1); + + idx = 0; + elem = hash % table_size; + second = hash % (table_size - 2); + while (symb_table[2 * elem] != 0) + { + /* First compare the hashing value. */ + if (symb_table[2 * elem] == hash + && c1 == extra[symb_table[2 * elem + 1]] + && memcmp (str, + &extra[symb_table[2 * elem + 1] + + 1], c1) == 0) + { + /* Yep, this is the entry. */ + idx = symb_table[2 * elem + 1]; + idx += 1 + extra[idx]; + break; + } + + /* Next entry. */ + elem += second; + } + + if (symb_table[2 * elem] != 0) + { + /* Compare the byte sequence but only if + this is not part of a range. */ +# ifdef WIDE_CHAR_VERSION + int32_t *wextra; + + idx += 1 + extra[idx]; + /* Adjust for the alignment. */ + idx = (idx + 3) & ~3; + + wextra = (int32_t *) &extra[idx + 4]; +# endif + + if (! is_range) + { +# ifdef WIDE_CHAR_VERSION + for (c1 = 0; + (int32_t) c1 < wextra[idx]; + ++c1) + if (n[c1] != wextra[1 + c1]) + break; + + if ((int32_t) c1 == wextra[idx]) + goto matched; +# else + for (c1 = 0; c1 < extra[idx]; ++c1) + if (n[c1] != extra[1 + c1]) + break; + + if (c1 == extra[idx]) + goto matched; +# endif + } + + /* Get the collation sequence value. */ + is_seqval = true; +# ifdef WIDE_CHAR_VERSION + cold = wextra[1 + wextra[idx]]; +# else + /* Adjust for the alignment. */ + idx += 1 + extra[idx]; + idx = (idx + 3) & ~4; + cold = *((int32_t *) &extra[idx]); +# endif + + c = *p++; + } + else if (c1 == 1) + { + /* No valid character. Match it as a + single byte. */ + if (!is_range && *n == str[0]) + goto matched; + + cold = str[0]; + c = *p++; + } + else + return FNM_NOMATCH; + } + } + else +# undef str +#endif + { + c = FOLD (c); + normal_bracket: + + /* We have to handling the symbols differently in + ranges since then the collation sequence is + important. */ + is_range = (*p == L('-') && p[1] != L('\0') + && p[1] != L(']')); + + if (!is_range && c == fn) + goto matched; + + cold = c; + c = *p++; + } + + if (c == L('-') && *p != L(']')) + { +#if _LIBC + /* We have to find the collation sequence + value for C. Collation sequence is nothing + we can regularly access. The sequence + value is defined by the order in which the + definitions of the collation values for the + various characters appear in the source + file. A strange concept, nowhere + documented. */ + uint32_t fcollseq; + uint32_t lcollseq; + UCHAR cend = *p++; + +# ifdef WIDE_CHAR_VERSION + /* Search in the `names' array for the characters. */ + fcollseq = __collseq_table_lookup (collseq, fn); + if (fcollseq == ~((uint32_t) 0)) + /* XXX We don't know anything about the character + we are supposed to match. This means we are + failing. */ + goto range_not_matched; + + if (is_seqval) + lcollseq = cold; + else + lcollseq = __collseq_table_lookup (collseq, cold); +# else + fcollseq = collseq[fn]; + lcollseq = is_seqval ? cold : collseq[(UCHAR) cold]; +# endif + + is_seqval = false; + if (cend == L('[') && *p == L('.')) + { + uint32_t nrules = + _NL_CURRENT_WORD (LC_COLLATE, + _NL_COLLATE_NRULES); + const CHAR *startp = p; + size_t c1 = 0; + + while (1) + { + c = *++p; + if (c == L('.') && p[1] == L(']')) + { + p += 2; + break; + } + if (c == '\0') + return FNM_NOMATCH; + ++c1; + } + + if (nrules == 0) + { + /* There are no names defined in the + collation data. Therefore we only + accept the trivial names consisting + of the character itself. */ + if (c1 != 1) + return FNM_NOMATCH; + + cend = startp[1]; + } + else + { + int32_t table_size; + const int32_t *symb_table; +# ifdef WIDE_CHAR_VERSION + char str[c1]; + size_t strcnt; +# else +# define str (startp + 1) +# endif + const unsigned char *extra; + int32_t idx; + int32_t elem; + int32_t second; + int32_t hash; + +# ifdef WIDE_CHAR_VERSION + /* We have to convert the name to a single-byte + string. This is possible since the names + consist of ASCII characters and the internal + representation is UCS4. */ + for (strcnt = 0; strcnt < c1; ++strcnt) + str[strcnt] = startp[1 + strcnt]; +# endif + + table_size = + _NL_CURRENT_WORD (LC_COLLATE, + _NL_COLLATE_SYMB_HASH_SIZEMB); + symb_table = (const int32_t *) + _NL_CURRENT (LC_COLLATE, + _NL_COLLATE_SYMB_TABLEMB); + extra = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, + _NL_COLLATE_SYMB_EXTRAMB); + + /* Locate the character in the hashing + table. */ + hash = elem_hash (str, c1); + + idx = 0; + elem = hash % table_size; + second = hash % (table_size - 2); + while (symb_table[2 * elem] != 0) + { + /* First compare the hashing value. */ + if (symb_table[2 * elem] == hash + && (c1 + == extra[symb_table[2 * elem + 1]]) + && memcmp (str, + &extra[symb_table[2 * elem + 1] + + 1], c1) == 0) + { + /* Yep, this is the entry. */ + idx = symb_table[2 * elem + 1]; + idx += 1 + extra[idx]; + break; + } + + /* Next entry. */ + elem += second; + } + + if (symb_table[2 * elem] != 0) + { + /* Compare the byte sequence but only if + this is not part of a range. */ +# ifdef WIDE_CHAR_VERSION + int32_t *wextra; + + idx += 1 + extra[idx]; + /* Adjust for the alignment. */ + idx = (idx + 3) & ~4; + + wextra = (int32_t *) &extra[idx + 4]; +# endif + /* Get the collation sequence value. */ + is_seqval = true; +# ifdef WIDE_CHAR_VERSION + cend = wextra[1 + wextra[idx]]; +# else + /* Adjust for the alignment. */ + idx += 1 + extra[idx]; + idx = (idx + 3) & ~4; + cend = *((int32_t *) &extra[idx]); +# endif + } + else if (symb_table[2 * elem] != 0 && c1 == 1) + { + cend = str[0]; + c = *p++; + } + else + return FNM_NOMATCH; + } +# undef str + } + else + { + if (!(flags & FNM_NOESCAPE) && cend == L('\\')) + cend = *p++; + if (cend == L('\0')) + return FNM_NOMATCH; + cend = FOLD (cend); + } + + /* XXX It is not entirely clear to me how to handle + characters which are not mentioned in the + collation specification. */ + if ( +# ifdef WIDE_CHAR_VERSION + lcollseq == 0xffffffff || +# endif + lcollseq <= fcollseq) + { + /* We have to look at the upper bound. */ + uint32_t hcollseq; + + if (is_seqval) + hcollseq = cend; + else + { +# ifdef WIDE_CHAR_VERSION + hcollseq = + __collseq_table_lookup (collseq, cend); + if (hcollseq == ~((uint32_t) 0)) + { + /* Hum, no information about the upper + bound. The matching succeeds if the + lower bound is matched exactly. */ + if (lcollseq != fcollseq) + goto range_not_matched; + + goto matched; + } +# else + hcollseq = collseq[cend]; +# endif + } + + if (lcollseq <= hcollseq && fcollseq <= hcollseq) + goto matched; + } +# ifdef WIDE_CHAR_VERSION + range_not_matched: +# endif +#else + /* We use a boring value comparison of the character + values. This is better than comparing using + `strcoll' since the latter would have surprising + and sometimes fatal consequences. */ + UCHAR cend = *p++; + + if (!(flags & FNM_NOESCAPE) && cend == L('\\')) + cend = *p++; + if (cend == L('\0')) + return FNM_NOMATCH; + + /* It is a range. */ + if (cold <= fn && fn <= cend) + goto matched; +#endif + + c = *p++; + } + } + + if (c == L(']')) + break; + } + + if (!not) + return FNM_NOMATCH; + break; + + matched: + /* Skip the rest of the [...] that already matched. */ + do + { + ignore_next: + c = *p++; + + if (c == L('\0')) + /* [... (unterminated) loses. */ + return FNM_NOMATCH; + + if (!(flags & FNM_NOESCAPE) && c == L('\\')) + { + if (*p == L('\0')) + return FNM_NOMATCH; + /* XXX 1003.2d11 is unclear if this is right. */ + ++p; + } + else if (c == L('[') && *p == L(':')) + { + int c1 = 0; + const CHAR *startp = p; + + while (1) + { + c = *++p; + if (++c1 == CHAR_CLASS_MAX_LENGTH) + return FNM_NOMATCH; + + if (*p == L(':') && p[1] == L(']')) + break; + + if (c < L('a') || c >= L('z')) + { + p = startp; + goto ignore_next; + } + } + p += 2; + c = *p++; + } + else if (c == L('[') && *p == L('=')) + { + c = *++p; + if (c == L('\0')) + return FNM_NOMATCH; + c = *++p; + if (c != L('=') || p[1] != L(']')) + return FNM_NOMATCH; + p += 2; + c = *p++; + } + else if (c == L('[') && *p == L('.')) + { + ++p; + while (1) + { + c = *++p; + if (c == '\0') + return FNM_NOMATCH; + + if (*p == L('.') && p[1] == L(']')) + break; + } + p += 2; + c = *p++; + } + } + while (c != L(']')); + if (not) + return FNM_NOMATCH; + } + break; + + case L('+'): + case L('@'): + case L('!'): + if (__builtin_expect (flags & FNM_EXTMATCH, 0) && *p == '(') + { + int res; + + res = EXT (c, p, n, string_end, no_leading_period, flags); + if (res != -1) + return res; + } + goto normal_match; + + case L('/'): + if (NO_LEADING_PERIOD (flags)) + { + if (n == string_end || c != (UCHAR) *n) + return FNM_NOMATCH; + + new_no_leading_period = true; + break; + } + /* FALLTHROUGH */ + default: + normal_match: + if (n == string_end || c != FOLD ((UCHAR) *n)) + return FNM_NOMATCH; + } + + no_leading_period = new_no_leading_period; + ++n; + } + + if (n == string_end) + return 0; + + if ((flags & FNM_LEADING_DIR) && n != string_end && *n == L('/')) + /* The FNM_LEADING_DIR flag says that "foo*" matches "foobar/frobozz". */ + return 0; + + return FNM_NOMATCH; +} + + +static const CHAR * +internal_function +END (const CHAR *pattern) +{ + const CHAR *p = pattern; + + while (1) + if (*++p == L('\0')) + /* This is an invalid pattern. */ + return pattern; + else if (*p == L('[')) + { + /* Handle brackets special. */ + if (posixly_correct == 0) + posixly_correct = getenv ("POSIXLY_CORRECT") != NULL ? 1 : -1; + + /* Skip the not sign. We have to recognize it because of a possibly + following ']'. */ + if (*++p == L('!') || (posixly_correct < 0 && *p == L('^'))) + ++p; + /* A leading ']' is recognized as such. */ + if (*p == L(']')) + ++p; + /* Skip over all characters of the list. */ + while (*p != L(']')) + if (*p++ == L('\0')) + /* This is no valid pattern. */ + return pattern; + } + else if ((*p == L('?') || *p == L('*') || *p == L('+') || *p == L('@') + || *p == L('!')) && p[1] == L('(')) + p = END (p + 1); + else if (*p == L(')')) + break; + + return p + 1; +} + + +static int +internal_function +EXT (INT opt, const CHAR *pattern, const CHAR *string, const CHAR *string_end, + bool no_leading_period, int flags) +{ + const CHAR *startp; + size_t level; + struct patternlist + { + struct patternlist *next; + CHAR str[1]; + } *list = NULL; + struct patternlist **lastp = &list; + size_t pattern_len = STRLEN (pattern); + const CHAR *p; + const CHAR *rs; + enum { ALLOCA_LIMIT = 8000 }; + + /* Parse the pattern. Store the individual parts in the list. */ + level = 0; + for (startp = p = pattern + 1; ; ++p) + if (*p == L('\0')) + /* This is an invalid pattern. */ + return -1; + else if (*p == L('[')) + { + /* Handle brackets special. */ + if (posixly_correct == 0) + posixly_correct = getenv ("POSIXLY_CORRECT") != NULL ? 1 : -1; + + /* Skip the not sign. We have to recognize it because of a possibly + following ']'. */ + if (*++p == L('!') || (posixly_correct < 0 && *p == L('^'))) + ++p; + /* A leading ']' is recognized as such. */ + if (*p == L(']')) + ++p; + /* Skip over all characters of the list. */ + while (*p != L(']')) + if (*p++ == L('\0')) + /* This is no valid pattern. */ + return -1; + } + else if ((*p == L('?') || *p == L('*') || *p == L('+') || *p == L('@') + || *p == L('!')) && p[1] == L('(')) + /* Remember the nesting level. */ + ++level; + else if (*p == L(')')) + { + if (level-- == 0) + { + /* This means we found the end of the pattern. */ +#define NEW_PATTERN \ + struct patternlist *newp; \ + size_t plen; \ + size_t plensize; \ + size_t newpsize; \ + \ + plen = (opt == L('?') || opt == L('@') \ + ? pattern_len \ + : p - startp + 1); \ + plensize = plen * sizeof (CHAR); \ + newpsize = offsetof (struct patternlist, str) + plensize; \ + if ((size_t) -1 / sizeof (CHAR) < plen \ + || newpsize < offsetof (struct patternlist, str) \ + || ALLOCA_LIMIT <= newpsize) \ + return -1; \ + newp = (struct patternlist *) alloca (newpsize); \ + *((CHAR *) MEMPCPY (newp->str, startp, p - startp)) = L('\0'); \ + newp->next = NULL; \ + *lastp = newp; \ + lastp = &newp->next + NEW_PATTERN; + break; + } + } + else if (*p == L('|')) + { + if (level == 0) + { + NEW_PATTERN; + startp = p + 1; + } + } + assert (list != NULL); + assert (p[-1] == L(')')); +#undef NEW_PATTERN + + switch (opt) + { + case L('*'): + if (FCT (p, string, string_end, no_leading_period, flags) == 0) + return 0; + /* FALLTHROUGH */ + + case L('+'): + do + { + for (rs = string; rs <= string_end; ++rs) + /* First match the prefix with the current pattern with the + current pattern. */ + if (FCT (list->str, string, rs, no_leading_period, + flags & FNM_FILE_NAME ? flags : flags & ~FNM_PERIOD) == 0 + /* This was successful. Now match the rest with the rest + of the pattern. */ + && (FCT (p, rs, string_end, + rs == string + ? no_leading_period + : rs[-1] == '/' && NO_LEADING_PERIOD (flags), + flags & FNM_FILE_NAME + ? flags : flags & ~FNM_PERIOD) == 0 + /* This didn't work. Try the whole pattern. */ + || (rs != string + && FCT (pattern - 1, rs, string_end, + rs == string + ? no_leading_period + : rs[-1] == '/' && NO_LEADING_PERIOD (flags), + flags & FNM_FILE_NAME + ? flags : flags & ~FNM_PERIOD) == 0))) + /* It worked. Signal success. */ + return 0; + } + while ((list = list->next) != NULL); + + /* None of the patterns lead to a match. */ + return FNM_NOMATCH; + + case L('?'): + if (FCT (p, string, string_end, no_leading_period, flags) == 0) + return 0; + /* FALLTHROUGH */ + + case L('@'): + do + /* I cannot believe it but `strcat' is actually acceptable + here. Match the entire string with the prefix from the + pattern list and the rest of the pattern following the + pattern list. */ + if (FCT (STRCAT (list->str, p), string, string_end, + no_leading_period, + flags & FNM_FILE_NAME ? flags : flags & ~FNM_PERIOD) == 0) + /* It worked. Signal success. */ + return 0; + while ((list = list->next) != NULL); + + /* None of the patterns lead to a match. */ + return FNM_NOMATCH; + + case L('!'): + for (rs = string; rs <= string_end; ++rs) + { + struct patternlist *runp; + + for (runp = list; runp != NULL; runp = runp->next) + if (FCT (runp->str, string, rs, no_leading_period, + flags & FNM_FILE_NAME ? flags : flags & ~FNM_PERIOD) == 0) + break; + + /* If none of the patterns matched see whether the rest does. */ + if (runp == NULL + && (FCT (p, rs, string_end, + rs == string + ? no_leading_period + : rs[-1] == '/' && NO_LEADING_PERIOD (flags), + flags & FNM_FILE_NAME ? flags : flags & ~FNM_PERIOD) + == 0)) + /* This is successful. */ + return 0; + } + + /* None of the patterns together with the rest of the pattern + lead to a match. */ + return FNM_NOMATCH; + + default: + assert (! "Invalid extended matching operator"); + break; + } + + return -1; +} + + +#undef FOLD +#undef CHAR +#undef UCHAR +#undef INT +#undef FCT +#undef EXT +#undef END +#undef MEMPCPY +#undef MEMCHR +#undef STRCOLL +#undef STRLEN +#undef STRCAT +#undef L +#undef BTOWC diff --git a/lib/fseeko.c b/lib/fseeko.c new file mode 100644 index 0000000..d4c0ee0 --- /dev/null +++ b/lib/fseeko.c @@ -0,0 +1,48 @@ +/* fseeko.c -- an implementation of fseek() with an off_t argument. + Copyright (C) 2003, Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. */ + +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif + +#include <stdio.h> +#include <sys/types.h> + +#ifdef HAVE_LIMITS_H +#include <limits.h> +#endif +#ifndef LONG_MAX +#define LONG_MAX ((long) ((unsigned long) ~0 >> 1)) +#endif +#ifndef LONG_MIN +#define LONG_MIN (-1 - LONG_MAX) +#endif + +/* + * A replacement/substitute for fseeko, for hosts that don't have it. + */ + +int +fseeko (FILE *stream, off_t offset, int whence) +{ + while (offset != (long) offset) + { + long pos = (offset < 0) ? LONG_MIN : LONG_MAX; + + if (fseek (stream, pos, whence) != 0) + return -1; + offset -= pos; + whence = SEEK_CUR; + } + return fseek (stream, (long) offset, whence); +} diff --git a/lib/ftello.c b/lib/ftello.c new file mode 100644 index 0000000..82fe70e --- /dev/null +++ b/lib/ftello.c @@ -0,0 +1,28 @@ +/* ftello.c -- an implementation of ftell() that returns an off_t + Copyright (C) 2003, Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. */ + +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif + +#include <stdio.h> +#include <sys/types.h> + +off_t +ftello (FILE *stream) +{ + long pos; + pos = ftell (stream); + + return (off_t) pos; +} diff --git a/lib/ftruncate.c b/lib/ftruncate.c new file mode 100644 index 0000000..d5e1add --- /dev/null +++ b/lib/ftruncate.c @@ -0,0 +1,92 @@ +/* ftruncate emulations that work on some System V's. + This file is in the public domain. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include <sys/types.h> +#include <fcntl.h> + +#ifdef F_CHSIZE + +int +ftruncate (int fd, off_t length) +{ + return fcntl (fd, F_CHSIZE, length); +} + +#else /* not F_CHSIZE */ +# ifdef F_FREESP + +/* By William Kucharski <kucharsk@netcom.com>. */ + +# include <sys/stat.h> +# include <errno.h> +# if HAVE_UNISTD_H +# include <unistd.h> +# endif + +int +ftruncate (int fd, off_t length) +{ + struct flock fl; + struct stat filebuf; + + if (fstat (fd, &filebuf) < 0) + return -1; + + if (filebuf.st_size < length) + { + /* Extend file length. */ + if (lseek (fd, (length - 1), SEEK_SET) < 0) + return -1; + + /* Write a "0" byte. */ + if (write (fd, "", 1) != 1) + return -1; + } + else + { + + /* Truncate length. */ + + fl.l_whence = 0; + fl.l_len = 0; + fl.l_start = length; + fl.l_type = F_WRLCK; /* write lock on file space */ + + /* This relies on the *undocumented* F_FREESP argument to fcntl, + which truncates the file so that it ends at the position + indicated by fl.l_start. Will minor miracles never cease? */ + + if (fcntl (fd, F_FREESP, &fl) < 0) + return -1; + } + + return 0; +} + +# else /* not F_CHSIZE nor F_FREESP */ +# if HAVE_CHSIZE + +int +ftruncate (int fd, off_t length) +{ + return chsize (fd, length); +} + +# else /* not F_CHSIZE nor F_FREESP nor HAVE_CHSIZE */ + +# include <errno.h> + +int +ftruncate (int fd, off_t length) +{ + errno = EIO; + return -1; +} + +# endif /* not HAVE_CHSIZE */ +# endif /* not F_FREESP */ +#endif /* not F_CHSIZE */ diff --git a/lib/gai_strerror.c b/lib/gai_strerror.c new file mode 100644 index 0000000..f289870 --- /dev/null +++ b/lib/gai_strerror.c @@ -0,0 +1,78 @@ +/* Copyright (C) 1997, 2001, 2002, 2004, 2005 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by Philip Blundell <pjb27@cam.ac.uk>, 1997. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#ifndef _LIBC +# include "getaddrinfo.h" +#endif + +#include <stdio.h> +#include <netdb.h> + +#ifdef _LIBC +# include <libintl.h> +#else +# include "gettext.h" +# define _(String) gettext (String) +# define N_(String) String +#endif + +static struct + { + int code; + const char *msg; + } +values[] = + { + { EAI_ADDRFAMILY, N_("Address family for hostname not supported") }, + { EAI_AGAIN, N_("Temporary failure in name resolution") }, + { EAI_BADFLAGS, N_("Bad value for ai_flags") }, + { EAI_FAIL, N_("Non-recoverable failure in name resolution") }, + { EAI_FAMILY, N_("ai_family not supported") }, + { EAI_MEMORY, N_("Memory allocation failure") }, + { EAI_NODATA, N_("No address associated with hostname") }, + { EAI_NONAME, N_("Name or service not known") }, + { EAI_SERVICE, N_("Servname not supported for ai_socktype") }, + { EAI_SOCKTYPE, N_("ai_socktype not supported") }, + { EAI_SYSTEM, N_("System error") }, +#ifdef __USE_GNU + { EAI_INPROGRESS, N_("Processing request in progress") }, + { EAI_CANCELED, N_("Request canceled") }, + { EAI_NOTCANCELED, N_("Request not canceled") }, + { EAI_ALLDONE, N_("All requests done") }, + { EAI_INTR, N_("Interrupted by a signal") }, + { EAI_IDN_ENCODE, N_("Parameter string not correctly encoded") } +#endif + }; + +const char * +gai_strerror (int code) +{ + size_t i; + for (i = 0; i < sizeof (values) / sizeof (values[0]); ++i) + if (values[i].code == code) + return _(values[i].msg); + + return _("Unknown error"); +} +#ifdef _LIBC +libc_hidden_def (gai_strerror) +#endif diff --git a/lib/getaddrinfo.c b/lib/getaddrinfo.c new file mode 100644 index 0000000..594b764 --- /dev/null +++ b/lib/getaddrinfo.c @@ -0,0 +1,210 @@ +/* Get address information (partial implementation). + Copyright (C) 1997, 2001, 2002, 2004, 2005 Free Software Foundation, Inc. + Contributed by Simon Josefsson <simon@josefsson.org>. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "getaddrinfo.h" + +/* Get calloc. */ +#include <stdlib.h> + +/* Get memcpy. */ +#include <string.h> + +#include <stdbool.h> + +#include "gettext.h" +#define _(String) gettext (String) +#define N_(String) String + +#include "strdup.h" + +static inline bool +validate_family (int family) +{ + /* FIXME: Support more families. */ +#if HAVE_IPV4 + if (family == PF_INET) + return true; +#endif +#if HAVE_IPV6 + if (family == PF_INET6) + return true; +#endif + if (family == PF_UNSPEC) + return true; + return false; +} + +/* Translate name of a service location and/or a service name to set of + socket addresses. */ +int +getaddrinfo (const char *restrict nodename, + const char *restrict servname, + const struct addrinfo *restrict hints, + struct addrinfo **restrict res) +{ + struct addrinfo *tmp; + struct servent *se; + struct hostent *he; + size_t sinlen; + + if (hints && (hints->ai_flags & ~AI_CANONNAME)) + /* FIXME: Support more flags. */ + return EAI_BADFLAGS; + + if (hints && !validate_family (hints->ai_family)) + return EAI_FAMILY; + + if (hints && + hints->ai_socktype != SOCK_STREAM && hints->ai_socktype != SOCK_DGRAM) + /* FIXME: Support other socktype. */ + return EAI_SOCKTYPE; /* FIXME: Better return code? */ + + if (!nodename) + /* FIXME: Support server bind mode. */ + return EAI_NONAME; + + if (servname) + { + const char *proto = + (hints && hints->ai_socktype == SOCK_DGRAM) ? "udp" : "tcp"; + + /* FIXME: Use getservbyname_r if available. */ + se = getservbyname (servname, proto); + + if (!se) + return EAI_SERVICE; + } + + /* FIXME: Use gethostbyname_r if available. */ + he = gethostbyname (nodename); + if (!he || he->h_addr_list[0] == NULL) + return EAI_NONAME; + + switch (he->h_addrtype) + { +#if HAVE_IPV6 + case PF_INET6: + sinlen = sizeof (struct sockaddr_in6); + break; +#endif + +#if HAVE_IPV4 + case PF_INET: + sinlen = sizeof (struct sockaddr_in); + break; +#endif + + default: + return EAI_NODATA; + } + + tmp = calloc (1, sizeof (*tmp) + sinlen); + if (!tmp) + return EAI_MEMORY; + + switch (he->h_addrtype) + { +#if HAVE_IPV6 + case PF_INET6: + { + struct sockaddr_in6 *sinp = (char *) tmp + sizeof (*tmp); + + if (se) + sinp->sin6_port = se->s_port; + + if (he->h_length != sizeof (sinp->sin6_addr)) + return EAI_SYSTEM; /* FIXME: Better return code? Set errno? */ + + memcpy (&sinp->sin6_addr, he->h_addr_list[0], he->h_length); + + tmp->ai_addr = (struct sockaddr *) sinp; + tmp->ai_addrlen = sinlen; + } + break; +#endif + +#if HAVE_IPV4 + case PF_INET: + { + struct sockaddr_in *sinp = (char *) tmp + sizeof (*tmp); + + if (se) + sinp->sin_port = se->s_port; + + if (he->h_length != sizeof (sinp->sin_addr)) + return EAI_SYSTEM; /* FIXME: Better return code? Set errno? */ + + memcpy (&sinp->sin_addr, he->h_addr_list[0], he->h_length); + + tmp->ai_addr = (struct sockaddr *) sinp; + tmp->ai_addrlen = sinlen; + } + break; +#endif + + default: + free (tmp); + return EAI_NODATA; + } + + if (hints && hints->ai_flags & AI_CANONNAME) + { + const char *cn; + if (he->h_name) + cn = he->h_name; + else + cn = nodename; + + tmp->ai_canonname = strdup (cn); + if (!tmp->ai_canonname) + { + free (tmp); + return EAI_MEMORY; + } + } + + tmp->ai_protocol = (hints) ? hints->ai_protocol : 0; + tmp->ai_socktype = (hints) ? hints->ai_socktype : 0; + tmp->ai_addr->sa_family = he->h_addrtype; + + /* FIXME: If more than one address, create linked list of addrinfo's. */ + + *res = tmp; + + return 0; +} + +/* Free `addrinfo' structure AI including associated storage. */ +void +freeaddrinfo (struct addrinfo *ai) +{ + while (ai) + { + struct addrinfo *cur; + + cur = ai; + ai = ai->ai_next; + + if (cur->ai_canonname) free (cur->ai_canonname); + free (cur); + } +} diff --git a/lib/getaddrinfo.h b/lib/getaddrinfo.h new file mode 100644 index 0000000..5d14562 --- /dev/null +++ b/lib/getaddrinfo.h @@ -0,0 +1,95 @@ +/* Get address information. + Copyright (C) 1996-2002, 2003, 2004, 2005 Free Software Foundation, Inc. + Contributed by Simon Josefsson <simon@josefsson.org>. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef GETADDRINFO_H +# define GETADDRINFO_H + +/* Get getaddrinfo declarations, if available. Also get 'socklen_t', + and 'struct sockaddr' via sys/types.h which are used below. */ +# include <sys/types.h> +# include <sys/socket.h> +# include <netdb.h> + +# if !HAVE_GETADDRINFO + +/* Structure to contain information about address of a service provider. */ +struct addrinfo +{ + int ai_flags; /* Input flags. */ + int ai_family; /* Protocol family for socket. */ + int ai_socktype; /* Socket type. */ + int ai_protocol; /* Protocol for socket. */ + socklen_t ai_addrlen; /* Length of socket address. */ + struct sockaddr *ai_addr; /* Socket address for socket. */ + char *ai_canonname; /* Canonical name for service location. */ + struct addrinfo *ai_next; /* Pointer to next in list. */ +}; + +/* Possible values for `ai_flags' field in `addrinfo' structure. */ +# define AI_PASSIVE 0x0001 /* Socket address is intended for `bind'. */ +# define AI_CANONNAME 0x0002 /* Request for canonical name. */ +# define AI_NUMERICHOST 0x0004 /* Don't use name resolution. */ +# define AI_V4MAPPED 0x0008 /* IPv4 mapped addresses are acceptable. */ +# define AI_ALL 0x0010 /* Return IPv4 mapped and IPv6 addresses. */ +# define AI_ADDRCONFIG 0x0020 /* Use configuration of this host to choose + returned address type.. */ + +/* Error values for `getaddrinfo' function. */ +# define EAI_BADFLAGS -1 /* Invalid value for `ai_flags' field. */ +# define EAI_NONAME -2 /* NAME or SERVICE is unknown. */ +# define EAI_AGAIN -3 /* Temporary failure in name resolution. */ +# define EAI_FAIL -4 /* Non-recoverable failure in name res. */ +# define EAI_NODATA -5 /* No address associated with NAME. */ +# define EAI_FAMILY -6 /* `ai_family' not supported. */ +# define EAI_SOCKTYPE -7 /* `ai_socktype' not supported. */ +# define EAI_SERVICE -8 /* SERVICE not supported for `ai_socktype'. */ +# define EAI_ADDRFAMILY -9 /* Address family for NAME not supported. */ +# define EAI_MEMORY -10 /* Memory allocation failure. */ +# define EAI_SYSTEM -11 /* System error returned in `errno'. */ +# define EAI_OVERFLOW -12 /* Argument buffer overflow. */ +# ifdef __USE_GNU +# define EAI_INPROGRESS -100 /* Processing request in progress. */ +# define EAI_CANCELED -101 /* Request canceled. */ +# define EAI_NOTCANCELED -102 /* Request not canceled. */ +# define EAI_ALLDONE -103 /* All requests done. */ +# define EAI_INTR -104 /* Interrupted by a signal. */ +# define EAI_IDN_ENCODE -105 /* IDN encoding failed. */ +# endif + +/* Translate name of a service location and/or a service name to set of + socket addresses. + For more details, see the POSIX:2001 specification + <http://www.opengroup.org/susv3xsh/getaddrinfo.html>. */ +extern int getaddrinfo (const char *restrict nodename, + const char *restrict servname, + const struct addrinfo *restrict hints, + struct addrinfo **restrict res); + +/* Free `addrinfo' structure AI including associated storage. + For more details, see the POSIX:2001 specification + <http://www.opengroup.org/susv3xsh/getaddrinfo.html>. */ +extern void freeaddrinfo (struct addrinfo *ai); + +/* Convert error return from getaddrinfo() to a string. + For more details, see the POSIX:2001 specification + <http://www.opengroup.org/susv3xsh/gai_strerror.html>. */ +extern const char *gai_strerror (int ecode); + +# endif /* !HAVE_GETADDRINFO */ + +#endif /* GETADDRINFO_H */ diff --git a/lib/getcwd.c b/lib/getcwd.c new file mode 100644 index 0000000..9f8c4cc --- /dev/null +++ b/lib/getcwd.c @@ -0,0 +1,407 @@ +/* Copyright (C) 1991,92,93,94,95,96,97,98,99,2004,2005 Free Software + Foundation, Inc. + This file is part of the GNU C Library. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#if !_LIBC +# include "getcwd.h" +#endif + +#include <errno.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <stdbool.h> +#include <stddef.h> + +#include <fcntl.h> /* For AT_FDCWD on Solaris 9. */ + +#ifndef __set_errno +# define __set_errno(val) (errno = (val)) +#endif + +#if HAVE_DIRENT_H || _LIBC +# include <dirent.h> +# ifndef _D_EXACT_NAMLEN +# define _D_EXACT_NAMLEN(d) strlen ((d)->d_name) +# endif +#else +# define dirent direct +# if HAVE_SYS_NDIR_H +# include <sys/ndir.h> +# endif +# if HAVE_SYS_DIR_H +# include <sys/dir.h> +# endif +# if HAVE_NDIR_H +# include <ndir.h> +# endif +#endif +#ifndef _D_EXACT_NAMLEN +# define _D_EXACT_NAMLEN(d) ((d)->d_namlen) +#endif +#ifndef _D_ALLOC_NAMLEN +# define _D_ALLOC_NAMLEN(d) (_D_EXACT_NAMLEN (d) + 1) +#endif + +#if HAVE_UNISTD_H || _LIBC +# include <unistd.h> +#endif + +#include <stdlib.h> +#include <string.h> + +#if _LIBC +# ifndef mempcpy +# define mempcpy __mempcpy +# endif +#else +# include "mempcpy.h" +#endif + +#include <limits.h> + +#ifdef ENAMETOOLONG +# define is_ENAMETOOLONG(x) ((x) == ENAMETOOLONG) +#else +# define is_ENAMETOOLONG(x) 0 +#endif + +#ifndef MAX +# define MAX(a, b) ((a) < (b) ? (b) : (a)) +#endif +#ifndef MIN +# define MIN(a, b) ((a) < (b) ? (a) : (b)) +#endif + +#ifndef PATH_MAX +# ifdef MAXPATHLEN +# define PATH_MAX MAXPATHLEN +# else +# define PATH_MAX 1024 +# endif +#endif + +#if D_INO_IN_DIRENT +# define MATCHING_INO(dp, ino) ((dp)->d_ino == (ino)) +#else +# define MATCHING_INO(dp, ino) true +#endif + +#if !_LIBC +# define __getcwd getcwd +# define __lstat lstat +# define __closedir closedir +# define __opendir opendir +# define __readdir readdir +#endif + +/* Get the name of the current working directory, and put it in SIZE + bytes of BUF. Returns NULL if the directory couldn't be determined or + SIZE was too small. If successful, returns BUF. In GNU, if BUF is + NULL, an array is allocated with `malloc'; the array is SIZE bytes long, + unless SIZE == 0, in which case it is as big as necessary. */ + +char * +__getcwd (char *buf, size_t size) +{ + /* Lengths of big file name components and entire file names, and a + deep level of file name nesting. These numbers are not upper + bounds; they are merely large values suitable for initial + allocations, designed to be large enough for most real-world + uses. */ + enum + { + BIG_FILE_NAME_COMPONENT_LENGTH = 255, + BIG_FILE_NAME_LENGTH = MIN (4095, PATH_MAX - 1), + DEEP_NESTING = 100 + }; + +#ifdef AT_FDCWD + int fd = AT_FDCWD; + bool fd_needs_closing = false; +#else + char dots[DEEP_NESTING * sizeof ".." + BIG_FILE_NAME_COMPONENT_LENGTH + 1]; + char *dotlist = dots; + size_t dotsize = sizeof dots; + size_t dotlen = 0; +#endif + DIR *dirstream = NULL; + dev_t rootdev, thisdev; + ino_t rootino, thisino; + char *dir; + register char *dirp; + struct stat st; + size_t allocated = size; + size_t used; + +#if HAVE_PARTLY_WORKING_GETCWD && !defined AT_FDCWD + /* The system getcwd works, except it sometimes fails when it + shouldn't, setting errno to ERANGE, ENAMETOOLONG, or ENOENT. If + AT_FDCWD is not defined, the algorithm below is O(N**2) and this + is much slower than the system getcwd (at least on GNU/Linux). + So trust the system getcwd's results unless they look + suspicious. */ +# undef getcwd + dir = getcwd (buf, size); + if (dir || (errno != ERANGE && !is_ENAMETOOLONG (errno) && errno != ENOENT)) + return dir; +#endif + + if (size == 0) + { + if (buf != NULL) + { + __set_errno (EINVAL); + return NULL; + } + + allocated = BIG_FILE_NAME_LENGTH + 1; + } + + if (buf == NULL) + { + dir = malloc (allocated); + if (dir == NULL) + return NULL; + } + else + dir = buf; + + dirp = dir + allocated; + *--dirp = '\0'; + + if (__lstat (".", &st) < 0) + goto lose; + thisdev = st.st_dev; + thisino = st.st_ino; + + if (__lstat ("/", &st) < 0) + goto lose; + rootdev = st.st_dev; + rootino = st.st_ino; + + while (!(thisdev == rootdev && thisino == rootino)) + { + struct dirent *d; + dev_t dotdev; + ino_t dotino; + bool mount_point; + int parent_status; + + /* Look at the parent directory. */ +#ifdef AT_FDCWD + fd = openat (fd, "..", O_RDONLY); + if (fd < 0) + goto lose; + fd_needs_closing = true; + parent_status = fstat (fd, &st); +#else + dotlist[dotlen++] = '.'; + dotlist[dotlen++] = '.'; + dotlist[dotlen] = '\0'; + parent_status = __lstat (dotlist, &st); +#endif + if (parent_status != 0) + goto lose; + + if (dirstream && __closedir (dirstream) != 0) + { + dirstream = NULL; + goto lose; + } + + /* Figure out if this directory is a mount point. */ + dotdev = st.st_dev; + dotino = st.st_ino; + mount_point = dotdev != thisdev; + + /* Search for the last directory. */ +#ifdef AT_FDCWD + dirstream = fdopendir (fd); + if (dirstream == NULL) + goto lose; + fd_needs_closing = false; +#else + dirstream = __opendir (dotlist); + if (dirstream == NULL) + goto lose; + dotlist[dotlen++] = '/'; +#endif + /* Clear errno to distinguish EOF from error if readdir returns + NULL. */ + __set_errno (0); + while ((d = __readdir (dirstream)) != NULL) + { + if (d->d_name[0] == '.' && + (d->d_name[1] == '\0' || + (d->d_name[1] == '.' && d->d_name[2] == '\0'))) + continue; + if (MATCHING_INO (d, thisino) || mount_point) + { + int entry_status; +#ifdef AT_FDCWD + entry_status = fstatat (fd, d->d_name, &st, AT_SYMLINK_NOFOLLOW); +#else + /* Compute size needed for this file name, or for the file + name ".." in the same directory, whichever is larger. + Room for ".." might be needed the next time through + the outer loop. */ + size_t name_alloc = _D_ALLOC_NAMLEN (d); + size_t filesize = dotlen + MAX (sizeof "..", name_alloc); + + if (filesize < dotlen) + goto memory_exhausted; + + if (dotsize < filesize) + { + /* My, what a deep directory tree you have, Grandma. */ + size_t newsize = MAX (filesize, dotsize * 2); + size_t i; + if (newsize < dotsize) + goto memory_exhausted; + if (dotlist != dots) + free (dotlist); + dotlist = malloc (newsize); + if (dotlist == NULL) + goto lose; + dotsize = newsize; + + i = 0; + do + { + dotlist[i++] = '.'; + dotlist[i++] = '.'; + dotlist[i++] = '/'; + } + while (i < dotlen); + } + + strcpy (dotlist + dotlen, d->d_name); + entry_status = __lstat (dotlist, &st); +#endif + /* We don't fail here if we cannot stat() a directory entry. + This can happen when (network) file systems fail. If this + entry is in fact the one we are looking for we will find + out soon as we reach the end of the directory without + having found anything. */ + if (entry_status == 0 && S_ISDIR (st.st_mode) + && st.st_dev == thisdev && st.st_ino == thisino) + break; + } + } + if (d == NULL) + { + if (errno == 0) + /* EOF on dirstream, which means that the current directory + has been removed. */ + __set_errno (ENOENT); + goto lose; + } + else + { + size_t dirroom = dirp - dir; + size_t namlen = _D_EXACT_NAMLEN (d); + + if (dirroom <= namlen) + { + if (size != 0) + { + __set_errno (ERANGE); + goto lose; + } + else + { + char *tmp; + size_t oldsize = allocated; + + allocated += MAX (allocated, namlen); + if (allocated < oldsize + || ! (tmp = realloc (dir, allocated))) + goto memory_exhausted; + + /* Move current contents up to the end of the buffer. + This is guaranteed to be non-overlapping. */ + dirp = memcpy (tmp + allocated - (oldsize - dirroom), + tmp + dirroom, + oldsize - dirroom); + dir = tmp; + } + } + dirp -= namlen; + memcpy (dirp, d->d_name, namlen); + *--dirp = '/'; + } + + thisdev = dotdev; + thisino = dotino; + } + + if (dirstream && __closedir (dirstream) != 0) + { + dirstream = NULL; + goto lose; + } + + if (dirp == &dir[allocated - 1]) + *--dirp = '/'; + +#ifndef AT_FDCWD + if (dotlist != dots) + free (dotlist); +#endif + + used = dir + allocated - dirp; + memmove (dir, dirp, used); + + if (buf == NULL && size == 0) + /* Ensure that the buffer is only as large as necessary. */ + buf = realloc (dir, used); + + if (buf == NULL) + /* Either buf was NULL all along, or `realloc' failed but + we still have the original string. */ + buf = dir; + + return buf; + + memory_exhausted: + __set_errno (ENOMEM); + lose: + { + int save = errno; + if (dirstream) + __closedir (dirstream); +#ifdef AT_FDCWD + if (fd_needs_closing) + close (fd); +#else + if (dotlist != dots) + free (dotlist); +#endif + if (buf == NULL) + free (dir); + __set_errno (save); + } + return NULL; +} + +#ifdef weak_alias +weak_alias (__getcwd, getcwd) +#endif diff --git a/lib/getcwd.h b/lib/getcwd.h new file mode 100644 index 0000000..200a3d3 --- /dev/null +++ b/lib/getcwd.h @@ -0,0 +1,42 @@ +/* Get the working directory, compatibly with the GNU C Library. + + Copyright (C) 2004-2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Paul Eggert. */ + +/* Include the headers that might declare getcwd so that they will not + cause confusion if included after this file. */ + +#include <stdlib.h> +#if HAVE_UNISTD_H +# include <unistd.h> +#endif + +/* If necessary, systematically rename identifiers so that they do not + collide with the system function. Renaming avoids problems with + some compilers and linkers. */ + +#ifdef __GETCWD_PREFIX +# undef getcwd +# define __GETCWD_CONCAT(x, y) x ## y +# define __GETCWD_XCONCAT(x, y) __GETCWD_CONCAT (x, y) +# define __GETCWD_ID(y) __GETCWD_XCONCAT (__GETCWD_PREFIX, y) +# define getcwd __GETCWD_ID (getcwd) +/* See the POSIX:2001 specification + <http://www.opengroup.org/susv3xsh/getcwd.html>. */ +char *getcwd (char *, size_t); +#endif diff --git a/lib/getdate.c b/lib/getdate.c new file mode 100644 index 0000000..7e0f763 --- /dev/null +++ b/lib/getdate.c @@ -0,0 +1,2936 @@ +/* A Bison parser, made by GNU Bison 1.875c. */ + +/* Skeleton parser for Yacc-like parsing with Bison, + Copyright (C) 1984, 1989, 1990, 2000, 2001, 2002, 2003 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 59 Temple Place - Suite 330, + Boston, MA 02111-1307, USA. */ + +/* As a special exception, when this file is copied by Bison into a + Bison output file, you may use that output file without restriction. + This special exception was added by the Free Software Foundation + in version 1.24 of Bison. */ + +/* Written by Richard Stallman by simplifying the original so called + ``semantic'' parser. */ + +/* All symbols defined below should begin with yy or YY, to avoid + infringing on user name space. This should be done even for local + variables, as they might otherwise be expanded by user macros. + There are some unavoidable exceptions within include files to + define necessary library symbols; they are noted "INFRINGES ON + USER NAME SPACE" below. */ + +/* Identify Bison output. */ +#define YYBISON 1 + +/* Skeleton name. */ +#define YYSKELETON_NAME "yacc.c" + +/* Pure parsers. */ +#define YYPURE 1 + +/* Using locations. */ +#define YYLSP_NEEDED 0 + + + +/* Tokens. */ +#ifndef YYTOKENTYPE +# define YYTOKENTYPE + /* Put the tokens into the symbol table, so that GDB and other debuggers + know about them. */ + enum yytokentype { + tAGO = 258, + tDST = 259, + tDAY = 260, + tDAY_UNIT = 261, + tDAYZONE = 262, + tHOUR_UNIT = 263, + tLOCAL_ZONE = 264, + tMERIDIAN = 265, + tMINUTE_UNIT = 266, + tMONTH = 267, + tMONTH_UNIT = 268, + tORDINAL = 269, + tSEC_UNIT = 270, + tYEAR_UNIT = 271, + tZONE = 272, + tSNUMBER = 273, + tUNUMBER = 274, + tSDECIMAL_NUMBER = 275, + tUDECIMAL_NUMBER = 276 + }; +#endif +#define tAGO 258 +#define tDST 259 +#define tDAY 260 +#define tDAY_UNIT 261 +#define tDAYZONE 262 +#define tHOUR_UNIT 263 +#define tLOCAL_ZONE 264 +#define tMERIDIAN 265 +#define tMINUTE_UNIT 266 +#define tMONTH 267 +#define tMONTH_UNIT 268 +#define tORDINAL 269 +#define tSEC_UNIT 270 +#define tYEAR_UNIT 271 +#define tZONE 272 +#define tSNUMBER 273 +#define tUNUMBER 274 +#define tSDECIMAL_NUMBER 275 +#define tUDECIMAL_NUMBER 276 + + + + +/* Copy the first part of user declarations. */ +#line 1 "getdate.y" + +/* Parse a string into an internal time stamp. + + Copyright (C) 1999, 2000, 2002, 2003, 2004, 2005 Free Software + Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Originally written by Steven M. Bellovin <smb@research.att.com> while + at the University of North Carolina at Chapel Hill. Later tweaked by + a couple of people on Usenet. Completely overhauled by Rich $alz + <rsalz@bbn.com> and Jim Berets <jberets@bbn.com> in August, 1990. + + Modified by Paul Eggert <eggert@twinsun.com> in August 1999 to do + the right thing about local DST. Also modified by Paul Eggert + <eggert@cs.ucla.edu> in February 2004 to support + nanosecond-resolution time stamps, and in October 2004 to support + TZ strings in dates. */ + +/* FIXME: Check for arithmetic overflow in all cases, not just + some of them. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "getdate.h" + +/* There's no need to extend the stack, so there's no need to involve + alloca. */ +#define YYSTACK_USE_ALLOCA 0 + +/* Tell Bison how much stack space is needed. 20 should be plenty for + this grammar, which is not right recursive. Beware setting it too + high, since that might cause problems on machines whose + implementations have lame stack-overflow checking. */ +#define YYMAXDEPTH 20 +#define YYINITDEPTH YYMAXDEPTH + +/* Since the code of getdate.y is not included in the Emacs executable + itself, there is no need to #define static in this file. Even if + the code were included in the Emacs executable, it probably + wouldn't do any harm to #undef it here; this will only cause + problems if we try to write to a static variable, which I don't + think this code needs to do. */ +#ifdef emacs +# undef static +#endif + +#include <ctype.h> +#include <limits.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +#include "setenv.h" +#include "xalloc.h" + +#if STDC_HEADERS || (! defined isascii && ! HAVE_ISASCII) +# define IN_CTYPE_DOMAIN(c) 1 +#else +# define IN_CTYPE_DOMAIN(c) isascii (c) +#endif + +#define ISSPACE(c) (IN_CTYPE_DOMAIN (c) && isspace (c)) +#define ISALPHA(c) (IN_CTYPE_DOMAIN (c) && isalpha (c)) +#define ISLOWER(c) (IN_CTYPE_DOMAIN (c) && islower (c)) + +/* ISDIGIT differs from isdigit, as follows: + - Its arg may be any int or unsigned int; it need not be an unsigned char. + - It's guaranteed to evaluate its argument exactly once. + - It's typically faster. + POSIX says that only '0' through '9' are digits. Prefer ISDIGIT to + isdigit unless it's important to use the locale's definition + of `digit' even when the host does not conform to POSIX. */ +#define ISDIGIT(c) ((unsigned int) (c) - '0' <= 9) + +#if __GNUC__ < 2 || (__GNUC__ == 2 && __GNUC_MINOR__ < 8) || __STRICT_ANSI__ +# define __attribute__(x) +#endif + +#ifndef ATTRIBUTE_UNUSED +# define ATTRIBUTE_UNUSED __attribute__ ((__unused__)) +#endif + +/* Shift A right by B bits portably, by dividing A by 2**B and + truncating towards minus infinity. A and B should be free of side + effects, and B should be in the range 0 <= B <= INT_BITS - 2, where + INT_BITS is the number of useful bits in an int. GNU code can + assume that INT_BITS is at least 32. + + ISO C99 says that A >> B is implementation-defined if A < 0. Some + implementations (e.g., UNICOS 9.0 on a Cray Y-MP EL) don't shift + right in the usual way when A < 0, so SHR falls back on division if + ordinary A >> B doesn't seem to be the usual signed shift. */ +#define SHR(a, b) \ + (-1 >> 1 == -1 \ + ? (a) >> (b) \ + : (a) / (1 << (b)) - ((a) % (1 << (b)) < 0)) + +#define EPOCH_YEAR 1970 +#define TM_YEAR_BASE 1900 + +#define HOUR(x) ((x) * 60) + +/* An integer value, and the number of digits in its textual + representation. */ +typedef struct +{ + bool negative; + long int value; + size_t digits; +} textint; + +/* An entry in the lexical lookup table. */ +typedef struct +{ + char const *name; + int type; + int value; +} table; + +/* Meridian: am, pm, or 24-hour style. */ +enum { MERam, MERpm, MER24 }; + +enum { BILLION = 1000000000, LOG10_BILLION = 9 }; + +/* Information passed to and from the parser. */ +typedef struct +{ + /* The input string remaining to be parsed. */ + const char *input; + + /* N, if this is the Nth Tuesday. */ + long int day_ordinal; + + /* Day of week; Sunday is 0. */ + int day_number; + + /* tm_isdst flag for the local zone. */ + int local_isdst; + + /* Time zone, in minutes east of UTC. */ + long int time_zone; + + /* Style used for time. */ + int meridian; + + /* Gregorian year, month, day, hour, minutes, seconds, and nanoseconds. */ + textint year; + long int month; + long int day; + long int hour; + long int minutes; + struct timespec seconds; /* includes nanoseconds */ + + /* Relative year, month, day, hour, minutes, seconds, and nanoseconds. */ + long int rel_year; + long int rel_month; + long int rel_day; + long int rel_hour; + long int rel_minutes; + long int rel_seconds; + long int rel_ns; + + /* Presence or counts of nonterminals of various flavors parsed so far. */ + bool timespec_seen; + bool rels_seen; + size_t dates_seen; + size_t days_seen; + size_t local_zones_seen; + size_t dsts_seen; + size_t times_seen; + size_t zones_seen; + + /* Table of local time zone abbrevations, terminated by a null entry. */ + table local_time_zone_table[3]; +} parser_control; + +union YYSTYPE; +static int yylex (union YYSTYPE *, parser_control *); +static int yyerror (parser_control *, char *); +static long int time_zone_hhmm (textint, long int); + + + +/* Enabling traces. */ +#ifndef YYDEBUG +# define YYDEBUG 0 +#endif + +/* Enabling verbose error messages. */ +#ifdef YYERROR_VERBOSE +# undef YYERROR_VERBOSE +# define YYERROR_VERBOSE 1 +#else +# define YYERROR_VERBOSE 0 +#endif + +#if ! defined (YYSTYPE) && ! defined (YYSTYPE_IS_DECLARED) +#line 209 "getdate.y" +typedef union YYSTYPE { + long int intval; + textint textintval; + struct timespec timespec; +} YYSTYPE; +/* Line 191 of yacc.c. */ +#line 322 "getdate.c" +# define yystype YYSTYPE /* obsolescent; will be withdrawn */ +# define YYSTYPE_IS_DECLARED 1 +# define YYSTYPE_IS_TRIVIAL 1 +#endif + + + +/* Copy the second part of user declarations. */ + + +/* Line 214 of yacc.c. */ +#line 334 "getdate.c" + +#if ! defined (yyoverflow) || YYERROR_VERBOSE + +# ifndef YYFREE +# define YYFREE free +# endif +# ifndef YYMALLOC +# define YYMALLOC malloc +# endif + +/* The parser invokes alloca or malloc; define the necessary symbols. */ + +# ifdef YYSTACK_USE_ALLOCA +# if YYSTACK_USE_ALLOCA +# define YYSTACK_ALLOC alloca +# endif +# else +# if defined (alloca) || defined (_ALLOCA_H) +# define YYSTACK_ALLOC alloca +# else +# ifdef __GNUC__ +# define YYSTACK_ALLOC __builtin_alloca +# endif +# endif +# endif + +# ifdef YYSTACK_ALLOC + /* Pacify GCC's `empty if-body' warning. */ +# define YYSTACK_FREE(Ptr) do { /* empty */; } while (0) +# else +# if defined (__STDC__) || defined (__cplusplus) +# include <stdlib.h> /* INFRINGES ON USER NAME SPACE */ +# define YYSIZE_T size_t +# endif +# define YYSTACK_ALLOC YYMALLOC +# define YYSTACK_FREE YYFREE +# endif +#endif /* ! defined (yyoverflow) || YYERROR_VERBOSE */ + + +#if (! defined (yyoverflow) \ + && (! defined (__cplusplus) \ + || (defined (YYSTYPE_IS_TRIVIAL) && YYSTYPE_IS_TRIVIAL))) + +/* A type that is properly aligned for any stack member. */ +union yyalloc +{ + short yyss; + YYSTYPE yyvs; + }; + +/* The size of the maximum gap between one aligned stack and the next. */ +# define YYSTACK_GAP_MAXIMUM (sizeof (union yyalloc) - 1) + +/* The size of an array large to enough to hold all stacks, each with + N elements. */ +# define YYSTACK_BYTES(N) \ + ((N) * (sizeof (short) + sizeof (YYSTYPE)) \ + + YYSTACK_GAP_MAXIMUM) + +/* Copy COUNT objects from FROM to TO. The source and destination do + not overlap. */ +# ifndef YYCOPY +# if defined (__GNUC__) && 1 < __GNUC__ +# define YYCOPY(To, From, Count) \ + __builtin_memcpy (To, From, (Count) * sizeof (*(From))) +# else +# define YYCOPY(To, From, Count) \ + do \ + { \ + register YYSIZE_T yyi; \ + for (yyi = 0; yyi < (Count); yyi++) \ + (To)[yyi] = (From)[yyi]; \ + } \ + while (0) +# endif +# endif + +/* Relocate STACK from its old location to the new one. The + local variables YYSIZE and YYSTACKSIZE give the old and new number of + elements in the stack, and YYPTR gives the new location of the + stack. Advance YYPTR to a properly aligned location for the next + stack. */ +# define YYSTACK_RELOCATE(Stack) \ + do \ + { \ + YYSIZE_T yynewbytes; \ + YYCOPY (&yyptr->Stack, Stack, yysize); \ + Stack = &yyptr->Stack; \ + yynewbytes = yystacksize * sizeof (*Stack) + YYSTACK_GAP_MAXIMUM; \ + yyptr += yynewbytes / sizeof (*yyptr); \ + } \ + while (0) + +#endif + +#if defined (__STDC__) || defined (__cplusplus) + typedef signed char yysigned_char; +#else + typedef short yysigned_char; +#endif + +/* YYFINAL -- State number of the termination state. */ +#define YYFINAL 12 +/* YYLAST -- Last index in YYTABLE. */ +#define YYLAST 88 + +/* YYNTOKENS -- Number of terminals. */ +#define YYNTOKENS 26 +/* YYNNTS -- Number of nonterminals. */ +#define YYNNTS 19 +/* YYNRULES -- Number of rules. */ +#define YYNRULES 78 +/* YYNRULES -- Number of states. */ +#define YYNSTATES 96 + +/* YYTRANSLATE(YYLEX) -- Bison symbol number corresponding to YYLEX. */ +#define YYUNDEFTOK 2 +#define YYMAXUTOK 276 + +#define YYTRANSLATE(YYX) \ + ((unsigned int) (YYX) <= YYMAXUTOK ? yytranslate[YYX] : YYUNDEFTOK) + +/* YYTRANSLATE[YYLEX] -- Bison symbol number corresponding to YYLEX. */ +static const unsigned char yytranslate[] = +{ + 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 24, 2, 2, 25, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 23, 2, + 2, 2, 2, 2, 22, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 1, 2, 3, 4, + 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, + 15, 16, 17, 18, 19, 20, 21 +}; + +#if YYDEBUG +/* YYPRHS[YYN] -- Index of the first RHS symbol of rule number YYN in + YYRHS. */ +static const unsigned char yyprhs[] = +{ + 0, 0, 3, 5, 7, 10, 11, 14, 16, 18, + 20, 22, 24, 26, 28, 31, 36, 42, 49, 57, + 59, 62, 64, 67, 71, 73, 76, 78, 81, 84, + 87, 91, 97, 101, 105, 109, 112, 117, 120, 124, + 127, 129, 132, 135, 137, 140, 143, 145, 148, 151, + 153, 156, 159, 161, 164, 167, 169, 172, 175, 178, + 181, 183, 185, 188, 191, 194, 197, 200, 203, 205, + 207, 209, 211, 213, 215, 217, 218, 221, 222 +}; + +/* YYRHS -- A `-1'-separated list of the rules' RHS. */ +static const yysigned_char yyrhs[] = +{ + 27, 0, -1, 28, -1, 29, -1, 22, 39, -1, + -1, 29, 30, -1, 31, -1, 32, -1, 33, -1, + 35, -1, 34, -1, 36, -1, 42, -1, 19, 10, + -1, 19, 23, 19, 44, -1, 19, 23, 19, 18, + 43, -1, 19, 23, 19, 23, 41, 44, -1, 19, + 23, 19, 23, 41, 18, 43, -1, 9, -1, 9, + 4, -1, 17, -1, 17, 38, -1, 17, 18, 43, + -1, 7, -1, 17, 4, -1, 5, -1, 5, 24, + -1, 14, 5, -1, 19, 5, -1, 19, 25, 19, + -1, 19, 25, 19, 25, 19, -1, 19, 18, 18, + -1, 19, 12, 18, -1, 12, 18, 18, -1, 12, + 19, -1, 12, 19, 24, 19, -1, 19, 12, -1, + 19, 12, 19, -1, 37, 3, -1, 37, -1, 14, + 16, -1, 19, 16, -1, 16, -1, 14, 13, -1, + 19, 13, -1, 13, -1, 14, 6, -1, 19, 6, + -1, 6, -1, 14, 8, -1, 19, 8, -1, 8, + -1, 14, 11, -1, 19, 11, -1, 11, -1, 14, + 15, -1, 19, 15, -1, 20, 15, -1, 21, 15, + -1, 15, -1, 38, -1, 18, 16, -1, 18, 13, + -1, 18, 6, -1, 18, 8, -1, 18, 11, -1, + 18, 15, -1, 40, -1, 41, -1, 20, -1, 18, + -1, 21, -1, 19, -1, 19, -1, -1, 23, 19, + -1, -1, 10, -1 +}; + +/* YYRLINE[YYN] -- source line where rule number YYN was defined. */ +static const unsigned short yyrline[] = +{ + 0, 230, 230, 231, 235, 242, 244, 248, 250, 252, + 254, 256, 258, 260, 264, 272, 280, 290, 297, 309, + 314, 322, 324, 326, 328, 330, 335, 340, 345, 350, + 358, 363, 383, 390, 398, 406, 411, 417, 422, 431, + 441, 445, 447, 449, 451, 453, 455, 457, 459, 461, + 463, 465, 467, 469, 471, 473, 475, 477, 479, 481, + 483, 485, 489, 491, 493, 495, 497, 499, 503, 503, + 506, 507, 512, 513, 518, 556, 557, 563, 564 +}; +#endif + +#if YYDEBUG || YYERROR_VERBOSE +/* YYTNME[SYMBOL-NUM] -- String name of the symbol SYMBOL-NUM. + First, the terminals, then, starting at YYNTOKENS, nonterminals. */ +static const char *const yytname[] = +{ + "$end", "error", "$undefined", "tAGO", "tDST", "tDAY", "tDAY_UNIT", + "tDAYZONE", "tHOUR_UNIT", "tLOCAL_ZONE", "tMERIDIAN", "tMINUTE_UNIT", + "tMONTH", "tMONTH_UNIT", "tORDINAL", "tSEC_UNIT", "tYEAR_UNIT", "tZONE", + "tSNUMBER", "tUNUMBER", "tSDECIMAL_NUMBER", "tUDECIMAL_NUMBER", "'@'", + "':'", "','", "'/'", "$accept", "spec", "timespec", "items", "item", + "time", "local_zone", "zone", "day", "date", "rel", "relunit", + "relunit_snumber", "seconds", "signed_seconds", "unsigned_seconds", + "number", "o_colon_minutes", "o_merid", 0 +}; +#endif + +# ifdef YYPRINT +/* YYTOKNUM[YYLEX-NUM] -- Internal token number corresponding to + token YYLEX-NUM. */ +static const unsigned short yytoknum[] = +{ + 0, 256, 257, 258, 259, 260, 261, 262, 263, 264, + 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, + 275, 276, 64, 58, 44, 47 +}; +# endif + +/* YYR1[YYN] -- Symbol number of symbol that rule YYN derives. */ +static const unsigned char yyr1[] = +{ + 0, 26, 27, 27, 28, 29, 29, 30, 30, 30, + 30, 30, 30, 30, 31, 31, 31, 31, 31, 32, + 32, 33, 33, 33, 33, 33, 34, 34, 34, 34, + 35, 35, 35, 35, 35, 35, 35, 35, 35, 36, + 36, 37, 37, 37, 37, 37, 37, 37, 37, 37, + 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, + 37, 37, 38, 38, 38, 38, 38, 38, 39, 39, + 40, 40, 41, 41, 42, 43, 43, 44, 44 +}; + +/* YYR2[YYN] -- Number of symbols composing right hand side of rule YYN. */ +static const unsigned char yyr2[] = +{ + 0, 2, 1, 1, 2, 0, 2, 1, 1, 1, + 1, 1, 1, 1, 2, 4, 5, 6, 7, 1, + 2, 1, 2, 3, 1, 2, 1, 2, 2, 2, + 3, 5, 3, 3, 3, 2, 4, 2, 3, 2, + 1, 2, 2, 1, 2, 2, 1, 2, 2, 1, + 2, 2, 1, 2, 2, 1, 2, 2, 2, 2, + 1, 1, 2, 2, 2, 2, 2, 2, 1, 1, + 1, 1, 1, 1, 1, 0, 2, 0, 1 +}; + +/* YYDEFACT[STATE-NAME] -- Default rule to reduce with in state + STATE-NUM when YYTABLE doesn't specify something else to do. Zero + means the default is an error. */ +static const unsigned char yydefact[] = +{ + 5, 0, 0, 2, 3, 71, 73, 70, 72, 4, + 68, 69, 1, 26, 49, 24, 52, 19, 55, 0, + 46, 0, 60, 43, 21, 0, 74, 0, 0, 6, + 7, 8, 9, 11, 10, 12, 40, 61, 13, 27, + 20, 0, 35, 28, 47, 50, 53, 44, 56, 41, + 25, 75, 22, 64, 65, 66, 63, 67, 62, 29, + 48, 51, 14, 54, 37, 45, 57, 42, 0, 0, + 0, 58, 59, 39, 34, 0, 0, 23, 33, 38, + 32, 77, 30, 36, 76, 78, 75, 0, 15, 0, + 16, 77, 31, 75, 17, 18 +}; + +/* YYDEFGOTO[NTERM-NUM]. */ +static const yysigned_char yydefgoto[] = +{ + -1, 2, 3, 4, 29, 30, 31, 32, 33, 34, + 35, 36, 37, 9, 10, 11, 38, 77, 88 +}; + +/* YYPACT[STATE-NUM] -- Index in YYTABLE of the portion describing + STATE-NUM. */ +#define YYPACT_NINF -43 +static const yysigned_char yypact[] = +{ + -18, 48, 9, -43, 19, -43, -43, -43, -43, -43, + -43, -43, -43, 32, -43, -43, -43, 54, -43, 28, + -43, 37, -43, -43, -2, 49, -5, 57, 58, -43, + -43, -43, -43, -43, -43, -43, 60, -43, -43, -43, + -43, 56, 51, -43, -43, -43, -43, -43, -43, -43, + -43, 6, -43, -43, -43, -43, -43, -43, -43, -43, + -43, -43, -43, -43, 52, -43, -43, -43, 59, 61, + 62, -43, -43, -43, -43, 63, 64, -43, -43, -43, + -43, 31, 53, -43, -43, -43, 65, 40, -43, 66, + -43, 5, -43, 65, -43, -43 +}; + +/* YYPGOTO[NTERM-NUM]. */ +static const yysigned_char yypgoto[] = +{ + -43, -43, -43, -43, -43, -43, -43, -43, -43, -43, + -43, -43, 55, -43, -43, -11, -43, -42, -7 +}; + +/* YYTABLE[YYPACT[STATE-NUM]]. What to do in state STATE-NUM. If + positive, shift that token. If negative, reduce the rule which + number is the opposite. If zero, do what YYDEFACT says. + If YYTABLE_NINF, syntax error. */ +#define YYTABLE_NINF -1 +static const unsigned char yytable[] = +{ + 59, 60, 50, 61, 1, 62, 63, 64, 65, 12, + 66, 67, 53, 68, 54, 85, 51, 55, 69, 56, + 70, 57, 58, 93, 13, 14, 15, 16, 17, 76, + 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, + 28, 85, 43, 44, 90, 45, 41, 42, 46, 86, + 47, 95, 48, 49, 87, 53, 39, 54, 40, 6, + 55, 8, 56, 73, 57, 58, 5, 6, 7, 8, + 78, 79, 71, 72, 74, 75, 91, 80, 89, 52, + 81, 82, 83, 84, 94, 92, 0, 0, 76 +}; + +static const yysigned_char yycheck[] = +{ + 5, 6, 4, 8, 22, 10, 11, 12, 13, 0, + 15, 16, 6, 18, 8, 10, 18, 11, 23, 13, + 25, 15, 16, 18, 5, 6, 7, 8, 9, 23, + 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, + 21, 10, 5, 6, 86, 8, 18, 19, 11, 18, + 13, 93, 15, 16, 23, 6, 24, 8, 4, 19, + 11, 21, 13, 3, 15, 16, 18, 19, 20, 21, + 18, 19, 15, 15, 18, 24, 87, 18, 25, 24, + 19, 19, 19, 19, 91, 19, -1, -1, 23 +}; + +/* YYSTOS[STATE-NUM] -- The (internal number of the) accessing + symbol of state STATE-NUM. */ +static const unsigned char yystos[] = +{ + 0, 22, 27, 28, 29, 18, 19, 20, 21, 39, + 40, 41, 0, 5, 6, 7, 8, 9, 11, 12, + 13, 14, 15, 16, 17, 18, 19, 20, 21, 30, + 31, 32, 33, 34, 35, 36, 37, 38, 42, 24, + 4, 18, 19, 5, 6, 8, 11, 13, 15, 16, + 4, 18, 38, 6, 8, 11, 13, 15, 16, 5, + 6, 8, 10, 11, 12, 13, 15, 16, 18, 23, + 25, 15, 15, 3, 18, 24, 23, 43, 18, 19, + 18, 19, 19, 19, 19, 10, 18, 23, 44, 25, + 43, 41, 19, 18, 44, 43 +}; + +#if ! defined (YYSIZE_T) && defined (__SIZE_TYPE__) +# define YYSIZE_T __SIZE_TYPE__ +#endif +#if ! defined (YYSIZE_T) && defined (size_t) +# define YYSIZE_T size_t +#endif +#if ! defined (YYSIZE_T) +# if defined (__STDC__) || defined (__cplusplus) +# include <stddef.h> /* INFRINGES ON USER NAME SPACE */ +# define YYSIZE_T size_t +# endif +#endif +#if ! defined (YYSIZE_T) +# define YYSIZE_T unsigned int +#endif + +#define yyerrok (yyerrstatus = 0) +#define yyclearin (yychar = YYEMPTY) +#define YYEMPTY (-2) +#define YYEOF 0 + +#define YYACCEPT goto yyacceptlab +#define YYABORT goto yyabortlab +#define YYERROR goto yyerrorlab + + +/* Like YYERROR except do call yyerror. This remains here temporarily + to ease the transition to the new meaning of YYERROR, for GCC. + Once GCC version 2 has supplanted version 1, this can go. */ + +#define YYFAIL goto yyerrlab + +#define YYRECOVERING() (!!yyerrstatus) + +#define YYBACKUP(Token, Value) \ +do \ + if (yychar == YYEMPTY && yylen == 1) \ + { \ + yychar = (Token); \ + yylval = (Value); \ + yytoken = YYTRANSLATE (yychar); \ + YYPOPSTACK; \ + goto yybackup; \ + } \ + else \ + { \ + yyerror (pc, "syntax error: cannot back up");\ + YYERROR; \ + } \ +while (0) + +#define YYTERROR 1 +#define YYERRCODE 256 + +/* YYLLOC_DEFAULT -- Compute the default location (before the actions + are run). */ + +#ifndef YYLLOC_DEFAULT +# define YYLLOC_DEFAULT(Current, Rhs, N) \ + ((Current).first_line = (Rhs)[1].first_line, \ + (Current).first_column = (Rhs)[1].first_column, \ + (Current).last_line = (Rhs)[N].last_line, \ + (Current).last_column = (Rhs)[N].last_column) +#endif + +/* YYLEX -- calling `yylex' with the right arguments. */ + +#ifdef YYLEX_PARAM +# define YYLEX yylex (&yylval, YYLEX_PARAM) +#else +# define YYLEX yylex (&yylval, pc) +#endif + +/* Enable debugging if requested. */ +#if YYDEBUG + +# ifndef YYFPRINTF +# include <stdio.h> /* INFRINGES ON USER NAME SPACE */ +# define YYFPRINTF fprintf +# endif + +# define YYDPRINTF(Args) \ +do { \ + if (yydebug) \ + YYFPRINTF Args; \ +} while (0) + +# define YYDSYMPRINT(Args) \ +do { \ + if (yydebug) \ + yysymprint Args; \ +} while (0) + +# define YYDSYMPRINTF(Title, Token, Value, Location) \ +do { \ + if (yydebug) \ + { \ + YYFPRINTF (stderr, "%s ", Title); \ + yysymprint (stderr, \ + Token, Value); \ + YYFPRINTF (stderr, "\n"); \ + } \ +} while (0) + +/*------------------------------------------------------------------. +| yy_stack_print -- Print the state stack from its BOTTOM up to its | +| TOP (included). | +`------------------------------------------------------------------*/ + +#if defined (__STDC__) || defined (__cplusplus) +static void +yy_stack_print (short *bottom, short *top) +#else +static void +yy_stack_print (bottom, top) + short *bottom; + short *top; +#endif +{ + YYFPRINTF (stderr, "Stack now"); + for (/* Nothing. */; bottom <= top; ++bottom) + YYFPRINTF (stderr, " %d", *bottom); + YYFPRINTF (stderr, "\n"); +} + +# define YY_STACK_PRINT(Bottom, Top) \ +do { \ + if (yydebug) \ + yy_stack_print ((Bottom), (Top)); \ +} while (0) + + +/*------------------------------------------------. +| Report that the YYRULE is going to be reduced. | +`------------------------------------------------*/ + +#if defined (__STDC__) || defined (__cplusplus) +static void +yy_reduce_print (int yyrule) +#else +static void +yy_reduce_print (yyrule) + int yyrule; +#endif +{ + int yyi; + unsigned int yylno = yyrline[yyrule]; + YYFPRINTF (stderr, "Reducing stack by rule %d (line %u), ", + yyrule - 1, yylno); + /* Print the symbols being reduced, and their result. */ + for (yyi = yyprhs[yyrule]; 0 <= yyrhs[yyi]; yyi++) + YYFPRINTF (stderr, "%s ", yytname [yyrhs[yyi]]); + YYFPRINTF (stderr, "-> %s\n", yytname [yyr1[yyrule]]); +} + +# define YY_REDUCE_PRINT(Rule) \ +do { \ + if (yydebug) \ + yy_reduce_print (Rule); \ +} while (0) + +/* Nonzero means print parse trace. It is left uninitialized so that + multiple parsers can coexist. */ +int yydebug; +#else /* !YYDEBUG */ +# define YYDPRINTF(Args) +# define YYDSYMPRINT(Args) +# define YYDSYMPRINTF(Title, Token, Value, Location) +# define YY_STACK_PRINT(Bottom, Top) +# define YY_REDUCE_PRINT(Rule) +#endif /* !YYDEBUG */ + + +/* YYINITDEPTH -- initial size of the parser's stacks. */ +#ifndef YYINITDEPTH +# define YYINITDEPTH 200 +#endif + +/* YYMAXDEPTH -- maximum size the stacks can grow to (effective only + if the built-in stack extension method is used). + + Do not make this value too large; the results are undefined if + SIZE_MAX < YYSTACK_BYTES (YYMAXDEPTH) + evaluated with infinite-precision integer arithmetic. */ + +#if defined (YYMAXDEPTH) && YYMAXDEPTH == 0 +# undef YYMAXDEPTH +#endif + +#ifndef YYMAXDEPTH +# define YYMAXDEPTH 10000 +#endif + + + +#if YYERROR_VERBOSE + +# ifndef yystrlen +# if defined (__GLIBC__) && defined (_STRING_H) +# define yystrlen strlen +# else +/* Return the length of YYSTR. */ +static YYSIZE_T +# if defined (__STDC__) || defined (__cplusplus) +yystrlen (const char *yystr) +# else +yystrlen (yystr) + const char *yystr; +# endif +{ + register const char *yys = yystr; + + while (*yys++ != '\0') + continue; + + return yys - yystr - 1; +} +# endif +# endif + +# ifndef yystpcpy +# if defined (__GLIBC__) && defined (_STRING_H) && defined (_GNU_SOURCE) +# define yystpcpy stpcpy +# else +/* Copy YYSRC to YYDEST, returning the address of the terminating '\0' in + YYDEST. */ +static char * +# if defined (__STDC__) || defined (__cplusplus) +yystpcpy (char *yydest, const char *yysrc) +# else +yystpcpy (yydest, yysrc) + char *yydest; + const char *yysrc; +# endif +{ + register char *yyd = yydest; + register const char *yys = yysrc; + + while ((*yyd++ = *yys++) != '\0') + continue; + + return yyd - 1; +} +# endif +# endif + +#endif /* !YYERROR_VERBOSE */ + + + +#if YYDEBUG +/*--------------------------------. +| Print this symbol on YYOUTPUT. | +`--------------------------------*/ + +#if defined (__STDC__) || defined (__cplusplus) +static void +yysymprint (FILE *yyoutput, int yytype, YYSTYPE *yyvaluep) +#else +static void +yysymprint (yyoutput, yytype, yyvaluep) + FILE *yyoutput; + int yytype; + YYSTYPE *yyvaluep; +#endif +{ + /* Pacify ``unused variable'' warnings. */ + (void) yyvaluep; + + if (yytype < YYNTOKENS) + { + YYFPRINTF (yyoutput, "token %s (", yytname[yytype]); +# ifdef YYPRINT + YYPRINT (yyoutput, yytoknum[yytype], *yyvaluep); +# endif + } + else + YYFPRINTF (yyoutput, "nterm %s (", yytname[yytype]); + + switch (yytype) + { + default: + break; + } + YYFPRINTF (yyoutput, ")"); +} + +#endif /* ! YYDEBUG */ +/*-----------------------------------------------. +| Release the memory associated to this symbol. | +`-----------------------------------------------*/ + +#if defined (__STDC__) || defined (__cplusplus) +static void +yydestruct (int yytype, YYSTYPE *yyvaluep) +#else +static void +yydestruct (yytype, yyvaluep) + int yytype; + YYSTYPE *yyvaluep; +#endif +{ + /* Pacify ``unused variable'' warnings. */ + (void) yyvaluep; + + switch (yytype) + { + + default: + break; + } +} + + +/* Prevent warnings from -Wmissing-prototypes. */ + +#ifdef YYPARSE_PARAM +# if defined (__STDC__) || defined (__cplusplus) +int yyparse (void *YYPARSE_PARAM); +# else +int yyparse (); +# endif +#else /* ! YYPARSE_PARAM */ +#if defined (__STDC__) || defined (__cplusplus) +int yyparse ( parser_control *pc ); +#else +int yyparse (); +#endif +#endif /* ! YYPARSE_PARAM */ + + + + + + +/*----------. +| yyparse. | +`----------*/ + +#ifdef YYPARSE_PARAM +# if defined (__STDC__) || defined (__cplusplus) +int yyparse (void *YYPARSE_PARAM) +# else +int yyparse (YYPARSE_PARAM) + void *YYPARSE_PARAM; +# endif +#else /* ! YYPARSE_PARAM */ +#if defined (__STDC__) || defined (__cplusplus) +int +yyparse ( parser_control *pc ) +#else +int +yyparse (pc) + parser_control *pc ; +#endif +#endif +{ + /* The lookahead symbol. */ +int yychar; + +/* The semantic value of the lookahead symbol. */ +YYSTYPE yylval; + +/* Number of syntax errors so far. */ +int yynerrs; + + register int yystate; + register int yyn; + int yyresult; + /* Number of tokens to shift before error messages enabled. */ + int yyerrstatus; + /* Lookahead token as an internal (translated) token number. */ + int yytoken = 0; + + /* Three stacks and their tools: + `yyss': related to states, + `yyvs': related to semantic values, + `yyls': related to locations. + + Refer to the stacks thru separate pointers, to allow yyoverflow + to reallocate them elsewhere. */ + + /* The state stack. */ + short yyssa[YYINITDEPTH]; + short *yyss = yyssa; + register short *yyssp; + + /* The semantic value stack. */ + YYSTYPE yyvsa[YYINITDEPTH]; + YYSTYPE *yyvs = yyvsa; + register YYSTYPE *yyvsp; + + + +#define YYPOPSTACK (yyvsp--, yyssp--) + + YYSIZE_T yystacksize = YYINITDEPTH; + + /* The variables used to return semantic value and location from the + action routines. */ + YYSTYPE yyval; + + + /* When reducing, the number of symbols on the RHS of the reduced + rule. */ + int yylen; + + YYDPRINTF ((stderr, "Starting parse\n")); + + yystate = 0; + yyerrstatus = 0; + yynerrs = 0; + yychar = YYEMPTY; /* Cause a token to be read. */ + + /* Initialize stack pointers. + Waste one element of value and location stack + so that they stay on the same level as the state stack. + The wasted elements are never initialized. */ + + yyssp = yyss; + yyvsp = yyvs; + + goto yysetstate; + +/*------------------------------------------------------------. +| yynewstate -- Push a new state, which is found in yystate. | +`------------------------------------------------------------*/ + yynewstate: + /* In all cases, when you get here, the value and location stacks + have just been pushed. so pushing a state here evens the stacks. + */ + yyssp++; + + yysetstate: + *yyssp = yystate; + + if (yyss + yystacksize - 1 <= yyssp) + { + /* Get the current used size of the three stacks, in elements. */ + YYSIZE_T yysize = yyssp - yyss + 1; + +#ifdef yyoverflow + { + /* Give user a chance to reallocate the stack. Use copies of + these so that the &'s don't force the real ones into + memory. */ + YYSTYPE *yyvs1 = yyvs; + short *yyss1 = yyss; + + + /* Each stack pointer address is followed by the size of the + data in use in that stack, in bytes. This used to be a + conditional around just the two extra args, but that might + be undefined if yyoverflow is a macro. */ + yyoverflow ("parser stack overflow", + &yyss1, yysize * sizeof (*yyssp), + &yyvs1, yysize * sizeof (*yyvsp), + + &yystacksize); + + yyss = yyss1; + yyvs = yyvs1; + } +#else /* no yyoverflow */ +# ifndef YYSTACK_RELOCATE + goto yyoverflowlab; +# else + /* Extend the stack our own way. */ + if (YYMAXDEPTH <= yystacksize) + goto yyoverflowlab; + yystacksize *= 2; + if (YYMAXDEPTH < yystacksize) + yystacksize = YYMAXDEPTH; + + { + short *yyss1 = yyss; + union yyalloc *yyptr = + (union yyalloc *) YYSTACK_ALLOC (YYSTACK_BYTES (yystacksize)); + if (! yyptr) + goto yyoverflowlab; + YYSTACK_RELOCATE (yyss); + YYSTACK_RELOCATE (yyvs); + +# undef YYSTACK_RELOCATE + if (yyss1 != yyssa) + YYSTACK_FREE (yyss1); + } +# endif +#endif /* no yyoverflow */ + + yyssp = yyss + yysize - 1; + yyvsp = yyvs + yysize - 1; + + + YYDPRINTF ((stderr, "Stack size increased to %lu\n", + (unsigned long int) yystacksize)); + + if (yyss + yystacksize - 1 <= yyssp) + YYABORT; + } + + YYDPRINTF ((stderr, "Entering state %d\n", yystate)); + + goto yybackup; + +/*-----------. +| yybackup. | +`-----------*/ +yybackup: + +/* Do appropriate processing given the current state. */ +/* Read a lookahead token if we need one and don't already have one. */ +/* yyresume: */ + + /* First try to decide what to do without reference to lookahead token. */ + + yyn = yypact[yystate]; + if (yyn == YYPACT_NINF) + goto yydefault; + + /* Not known => get a lookahead token if don't already have one. */ + + /* YYCHAR is either YYEMPTY or YYEOF or a valid lookahead symbol. */ + if (yychar == YYEMPTY) + { + YYDPRINTF ((stderr, "Reading a token: ")); + yychar = YYLEX; + } + + if (yychar <= YYEOF) + { + yychar = yytoken = YYEOF; + YYDPRINTF ((stderr, "Now at end of input.\n")); + } + else + { + yytoken = YYTRANSLATE (yychar); + YYDSYMPRINTF ("Next token is", yytoken, &yylval, &yylloc); + } + + /* If the proper action on seeing token YYTOKEN is to reduce or to + detect an error, take that action. */ + yyn += yytoken; + if (yyn < 0 || YYLAST < yyn || yycheck[yyn] != yytoken) + goto yydefault; + yyn = yytable[yyn]; + if (yyn <= 0) + { + if (yyn == 0 || yyn == YYTABLE_NINF) + goto yyerrlab; + yyn = -yyn; + goto yyreduce; + } + + if (yyn == YYFINAL) + YYACCEPT; + + /* Shift the lookahead token. */ + YYDPRINTF ((stderr, "Shifting token %s, ", yytname[yytoken])); + + /* Discard the token being shifted unless it is eof. */ + if (yychar != YYEOF) + yychar = YYEMPTY; + + *++yyvsp = yylval; + + + /* Count tokens shifted since error; after three, turn off error + status. */ + if (yyerrstatus) + yyerrstatus--; + + yystate = yyn; + goto yynewstate; + + +/*-----------------------------------------------------------. +| yydefault -- do the default action for the current state. | +`-----------------------------------------------------------*/ +yydefault: + yyn = yydefact[yystate]; + if (yyn == 0) + goto yyerrlab; + goto yyreduce; + + +/*-----------------------------. +| yyreduce -- Do a reduction. | +`-----------------------------*/ +yyreduce: + /* yyn is the number of a rule to reduce with. */ + yylen = yyr2[yyn]; + + /* If YYLEN is nonzero, implement the default value of the action: + `$$ = $1'. + + Otherwise, the following line sets YYVAL to garbage. + This behavior is undocumented and Bison + users should not rely upon it. Assigning to YYVAL + unconditionally makes the parser a bit smaller, and it avoids a + GCC warning that YYVAL may be used uninitialized. */ + yyval = yyvsp[1-yylen]; + + + YY_REDUCE_PRINT (yyn); + switch (yyn) + { + case 4: +#line 236 "getdate.y" + { + pc->seconds = yyvsp[0].timespec; + pc->timespec_seen = true; + } + break; + + case 7: +#line 249 "getdate.y" + { pc->times_seen++; } + break; + + case 8: +#line 251 "getdate.y" + { pc->local_zones_seen++; } + break; + + case 9: +#line 253 "getdate.y" + { pc->zones_seen++; } + break; + + case 10: +#line 255 "getdate.y" + { pc->dates_seen++; } + break; + + case 11: +#line 257 "getdate.y" + { pc->days_seen++; } + break; + + case 12: +#line 259 "getdate.y" + { pc->rels_seen = true; } + break; + + case 14: +#line 265 "getdate.y" + { + pc->hour = yyvsp[-1].textintval.value; + pc->minutes = 0; + pc->seconds.tv_sec = 0; + pc->seconds.tv_nsec = 0; + pc->meridian = yyvsp[0].intval; + } + break; + + case 15: +#line 273 "getdate.y" + { + pc->hour = yyvsp[-3].textintval.value; + pc->minutes = yyvsp[-1].textintval.value; + pc->seconds.tv_sec = 0; + pc->seconds.tv_nsec = 0; + pc->meridian = yyvsp[0].intval; + } + break; + + case 16: +#line 281 "getdate.y" + { + pc->hour = yyvsp[-4].textintval.value; + pc->minutes = yyvsp[-2].textintval.value; + pc->seconds.tv_sec = 0; + pc->seconds.tv_nsec = 0; + pc->meridian = MER24; + pc->zones_seen++; + pc->time_zone = time_zone_hhmm (yyvsp[-1].textintval, yyvsp[0].intval); + } + break; + + case 17: +#line 291 "getdate.y" + { + pc->hour = yyvsp[-5].textintval.value; + pc->minutes = yyvsp[-3].textintval.value; + pc->seconds = yyvsp[-1].timespec; + pc->meridian = yyvsp[0].intval; + } + break; + + case 18: +#line 298 "getdate.y" + { + pc->hour = yyvsp[-6].textintval.value; + pc->minutes = yyvsp[-4].textintval.value; + pc->seconds = yyvsp[-2].timespec; + pc->meridian = MER24; + pc->zones_seen++; + pc->time_zone = time_zone_hhmm (yyvsp[-1].textintval, yyvsp[0].intval); + } + break; + + case 19: +#line 310 "getdate.y" + { + pc->local_isdst = yyvsp[0].intval; + pc->dsts_seen += (0 < yyvsp[0].intval); + } + break; + + case 20: +#line 315 "getdate.y" + { + pc->local_isdst = 1; + pc->dsts_seen += (0 < yyvsp[-1].intval) + 1; + } + break; + + case 21: +#line 323 "getdate.y" + { pc->time_zone = yyvsp[0].intval; } + break; + + case 22: +#line 325 "getdate.y" + { pc->time_zone = yyvsp[-1].intval; pc->rels_seen = true; } + break; + + case 23: +#line 327 "getdate.y" + { pc->time_zone = yyvsp[-2].intval + time_zone_hhmm (yyvsp[-1].textintval, yyvsp[0].intval); } + break; + + case 24: +#line 329 "getdate.y" + { pc->time_zone = yyvsp[0].intval + 60; } + break; + + case 25: +#line 331 "getdate.y" + { pc->time_zone = yyvsp[-1].intval + 60; } + break; + + case 26: +#line 336 "getdate.y" + { + pc->day_ordinal = 1; + pc->day_number = yyvsp[0].intval; + } + break; + + case 27: +#line 341 "getdate.y" + { + pc->day_ordinal = 1; + pc->day_number = yyvsp[-1].intval; + } + break; + + case 28: +#line 346 "getdate.y" + { + pc->day_ordinal = yyvsp[-1].intval; + pc->day_number = yyvsp[0].intval; + } + break; + + case 29: +#line 351 "getdate.y" + { + pc->day_ordinal = yyvsp[-1].textintval.value; + pc->day_number = yyvsp[0].intval; + } + break; + + case 30: +#line 359 "getdate.y" + { + pc->month = yyvsp[-2].textintval.value; + pc->day = yyvsp[0].textintval.value; + } + break; + + case 31: +#line 364 "getdate.y" + { + /* Interpret as YYYY/MM/DD if the first value has 4 or more digits, + otherwise as MM/DD/YY. + The goal in recognizing YYYY/MM/DD is solely to support legacy + machine-generated dates like those in an RCS log listing. If + you want portability, use the ISO 8601 format. */ + if (4 <= yyvsp[-4].textintval.digits) + { + pc->year = yyvsp[-4].textintval; + pc->month = yyvsp[-2].textintval.value; + pc->day = yyvsp[0].textintval.value; + } + else + { + pc->month = yyvsp[-4].textintval.value; + pc->day = yyvsp[-2].textintval.value; + pc->year = yyvsp[0].textintval; + } + } + break; + + case 32: +#line 384 "getdate.y" + { + /* ISO 8601 format. YYYY-MM-DD. */ + pc->year = yyvsp[-2].textintval; + pc->month = -yyvsp[-1].textintval.value; + pc->day = -yyvsp[0].textintval.value; + } + break; + + case 33: +#line 391 "getdate.y" + { + /* e.g. 17-JUN-1992. */ + pc->day = yyvsp[-2].textintval.value; + pc->month = yyvsp[-1].intval; + pc->year.value = -yyvsp[0].textintval.value; + pc->year.digits = yyvsp[0].textintval.digits; + } + break; + + case 34: +#line 399 "getdate.y" + { + /* e.g. JUN-17-1992. */ + pc->month = yyvsp[-2].intval; + pc->day = -yyvsp[-1].textintval.value; + pc->year.value = -yyvsp[0].textintval.value; + pc->year.digits = yyvsp[0].textintval.digits; + } + break; + + case 35: +#line 407 "getdate.y" + { + pc->month = yyvsp[-1].intval; + pc->day = yyvsp[0].textintval.value; + } + break; + + case 36: +#line 412 "getdate.y" + { + pc->month = yyvsp[-3].intval; + pc->day = yyvsp[-2].textintval.value; + pc->year = yyvsp[0].textintval; + } + break; + + case 37: +#line 418 "getdate.y" + { + pc->day = yyvsp[-1].textintval.value; + pc->month = yyvsp[0].intval; + } + break; + + case 38: +#line 423 "getdate.y" + { + pc->day = yyvsp[-2].textintval.value; + pc->month = yyvsp[-1].intval; + pc->year = yyvsp[0].textintval; + } + break; + + case 39: +#line 432 "getdate.y" + { + pc->rel_ns = -pc->rel_ns; + pc->rel_seconds = -pc->rel_seconds; + pc->rel_minutes = -pc->rel_minutes; + pc->rel_hour = -pc->rel_hour; + pc->rel_day = -pc->rel_day; + pc->rel_month = -pc->rel_month; + pc->rel_year = -pc->rel_year; + } + break; + + case 41: +#line 446 "getdate.y" + { pc->rel_year += yyvsp[-1].intval * yyvsp[0].intval; } + break; + + case 42: +#line 448 "getdate.y" + { pc->rel_year += yyvsp[-1].textintval.value * yyvsp[0].intval; } + break; + + case 43: +#line 450 "getdate.y" + { pc->rel_year += yyvsp[0].intval; } + break; + + case 44: +#line 452 "getdate.y" + { pc->rel_month += yyvsp[-1].intval * yyvsp[0].intval; } + break; + + case 45: +#line 454 "getdate.y" + { pc->rel_month += yyvsp[-1].textintval.value * yyvsp[0].intval; } + break; + + case 46: +#line 456 "getdate.y" + { pc->rel_month += yyvsp[0].intval; } + break; + + case 47: +#line 458 "getdate.y" + { pc->rel_day += yyvsp[-1].intval * yyvsp[0].intval; } + break; + + case 48: +#line 460 "getdate.y" + { pc->rel_day += yyvsp[-1].textintval.value * yyvsp[0].intval; } + break; + + case 49: +#line 462 "getdate.y" + { pc->rel_day += yyvsp[0].intval; } + break; + + case 50: +#line 464 "getdate.y" + { pc->rel_hour += yyvsp[-1].intval * yyvsp[0].intval; } + break; + + case 51: +#line 466 "getdate.y" + { pc->rel_hour += yyvsp[-1].textintval.value * yyvsp[0].intval; } + break; + + case 52: +#line 468 "getdate.y" + { pc->rel_hour += yyvsp[0].intval; } + break; + + case 53: +#line 470 "getdate.y" + { pc->rel_minutes += yyvsp[-1].intval * yyvsp[0].intval; } + break; + + case 54: +#line 472 "getdate.y" + { pc->rel_minutes += yyvsp[-1].textintval.value * yyvsp[0].intval; } + break; + + case 55: +#line 474 "getdate.y" + { pc->rel_minutes += yyvsp[0].intval; } + break; + + case 56: +#line 476 "getdate.y" + { pc->rel_seconds += yyvsp[-1].intval * yyvsp[0].intval; } + break; + + case 57: +#line 478 "getdate.y" + { pc->rel_seconds += yyvsp[-1].textintval.value * yyvsp[0].intval; } + break; + + case 58: +#line 480 "getdate.y" + { pc->rel_seconds += yyvsp[-1].timespec.tv_sec * yyvsp[0].intval; pc->rel_ns += yyvsp[-1].timespec.tv_nsec * yyvsp[0].intval; } + break; + + case 59: +#line 482 "getdate.y" + { pc->rel_seconds += yyvsp[-1].timespec.tv_sec * yyvsp[0].intval; pc->rel_ns += yyvsp[-1].timespec.tv_nsec * yyvsp[0].intval; } + break; + + case 60: +#line 484 "getdate.y" + { pc->rel_seconds += yyvsp[0].intval; } + break; + + case 62: +#line 490 "getdate.y" + { pc->rel_year += yyvsp[-1].textintval.value * yyvsp[0].intval; } + break; + + case 63: +#line 492 "getdate.y" + { pc->rel_month += yyvsp[-1].textintval.value * yyvsp[0].intval; } + break; + + case 64: +#line 494 "getdate.y" + { pc->rel_day += yyvsp[-1].textintval.value * yyvsp[0].intval; } + break; + + case 65: +#line 496 "getdate.y" + { pc->rel_hour += yyvsp[-1].textintval.value * yyvsp[0].intval; } + break; + + case 66: +#line 498 "getdate.y" + { pc->rel_minutes += yyvsp[-1].textintval.value * yyvsp[0].intval; } + break; + + case 67: +#line 500 "getdate.y" + { pc->rel_seconds += yyvsp[-1].textintval.value * yyvsp[0].intval; } + break; + + case 71: +#line 508 "getdate.y" + { yyval.timespec.tv_sec = yyvsp[0].textintval.value; yyval.timespec.tv_nsec = 0; } + break; + + case 73: +#line 514 "getdate.y" + { yyval.timespec.tv_sec = yyvsp[0].textintval.value; yyval.timespec.tv_nsec = 0; } + break; + + case 74: +#line 519 "getdate.y" + { + if (pc->dates_seen && ! pc->year.digits + && ! pc->rels_seen && (pc->times_seen || 2 < yyvsp[0].textintval.digits)) + pc->year = yyvsp[0].textintval; + else + { + if (4 < yyvsp[0].textintval.digits) + { + pc->dates_seen++; + pc->day = yyvsp[0].textintval.value % 100; + pc->month = (yyvsp[0].textintval.value / 100) % 100; + pc->year.value = yyvsp[0].textintval.value / 10000; + pc->year.digits = yyvsp[0].textintval.digits - 4; + } + else + { + pc->times_seen++; + if (yyvsp[0].textintval.digits <= 2) + { + pc->hour = yyvsp[0].textintval.value; + pc->minutes = 0; + } + else + { + pc->hour = yyvsp[0].textintval.value / 100; + pc->minutes = yyvsp[0].textintval.value % 100; + } + pc->seconds.tv_sec = 0; + pc->seconds.tv_nsec = 0; + pc->meridian = MER24; + } + } + } + break; + + case 75: +#line 556 "getdate.y" + { yyval.intval = -1; } + break; + + case 76: +#line 558 "getdate.y" + { yyval.intval = yyvsp[0].textintval.value; } + break; + + case 77: +#line 563 "getdate.y" + { yyval.intval = MER24; } + break; + + case 78: +#line 565 "getdate.y" + { yyval.intval = yyvsp[0].intval; } + break; + + + } + +/* Line 1000 of yacc.c. */ +#line 1782 "getdate.c" + + yyvsp -= yylen; + yyssp -= yylen; + + + YY_STACK_PRINT (yyss, yyssp); + + *++yyvsp = yyval; + + + /* Now `shift' the result of the reduction. Determine what state + that goes to, based on the state we popped back to and the rule + number reduced by. */ + + yyn = yyr1[yyn]; + + yystate = yypgoto[yyn - YYNTOKENS] + *yyssp; + if (0 <= yystate && yystate <= YYLAST && yycheck[yystate] == *yyssp) + yystate = yytable[yystate]; + else + yystate = yydefgoto[yyn - YYNTOKENS]; + + goto yynewstate; + + +/*------------------------------------. +| yyerrlab -- here on detecting error | +`------------------------------------*/ +yyerrlab: + /* If not already recovering from an error, report this error. */ + if (!yyerrstatus) + { + ++yynerrs; +#if YYERROR_VERBOSE + yyn = yypact[yystate]; + + if (YYPACT_NINF < yyn && yyn < YYLAST) + { + YYSIZE_T yysize = 0; + int yytype = YYTRANSLATE (yychar); + const char* yyprefix; + char *yymsg; + int yyx; + + /* Start YYX at -YYN if negative to avoid negative indexes in + YYCHECK. */ + int yyxbegin = yyn < 0 ? -yyn : 0; + + /* Stay within bounds of both yycheck and yytname. */ + int yychecklim = YYLAST - yyn; + int yyxend = yychecklim < YYNTOKENS ? yychecklim : YYNTOKENS; + int yycount = 0; + + yyprefix = ", expecting "; + for (yyx = yyxbegin; yyx < yyxend; ++yyx) + if (yycheck[yyx + yyn] == yyx && yyx != YYTERROR) + { + yysize += yystrlen (yyprefix) + yystrlen (yytname [yyx]); + yycount += 1; + if (yycount == 5) + { + yysize = 0; + break; + } + } + yysize += (sizeof ("syntax error, unexpected ") + + yystrlen (yytname[yytype])); + yymsg = (char *) YYSTACK_ALLOC (yysize); + if (yymsg != 0) + { + char *yyp = yystpcpy (yymsg, "syntax error, unexpected "); + yyp = yystpcpy (yyp, yytname[yytype]); + + if (yycount < 5) + { + yyprefix = ", expecting "; + for (yyx = yyxbegin; yyx < yyxend; ++yyx) + if (yycheck[yyx + yyn] == yyx && yyx != YYTERROR) + { + yyp = yystpcpy (yyp, yyprefix); + yyp = yystpcpy (yyp, yytname[yyx]); + yyprefix = " or "; + } + } + yyerror (pc, yymsg); + YYSTACK_FREE (yymsg); + } + else + yyerror (pc, "syntax error; also virtual memory exhausted"); + } + else +#endif /* YYERROR_VERBOSE */ + yyerror (pc, "syntax error"); + } + + + + if (yyerrstatus == 3) + { + /* If just tried and failed to reuse lookahead token after an + error, discard it. */ + + if (yychar <= YYEOF) + { + /* If at end of input, pop the error token, + then the rest of the stack, then return failure. */ + if (yychar == YYEOF) + for (;;) + { + YYPOPSTACK; + if (yyssp == yyss) + YYABORT; + YYDSYMPRINTF ("Error: popping", yystos[*yyssp], yyvsp, yylsp); + yydestruct (yystos[*yyssp], yyvsp); + } + } + else + { + YYDSYMPRINTF ("Error: discarding", yytoken, &yylval, &yylloc); + yydestruct (yytoken, &yylval); + yychar = YYEMPTY; + + } + } + + /* Else will try to reuse lookahead token after shifting the error + token. */ + goto yyerrlab1; + + +/*---------------------------------------------------. +| yyerrorlab -- error raised explicitly by YYERROR. | +`---------------------------------------------------*/ +yyerrorlab: + +#ifdef __GNUC__ + /* Pacify GCC when the user code never invokes YYERROR and the label + yyerrorlab therefore never appears in user code. */ + if (0) + goto yyerrorlab; +#endif + + yyvsp -= yylen; + yyssp -= yylen; + yystate = *yyssp; + goto yyerrlab1; + + +/*-------------------------------------------------------------. +| yyerrlab1 -- common code for both syntax error and YYERROR. | +`-------------------------------------------------------------*/ +yyerrlab1: + yyerrstatus = 3; /* Each real token shifted decrements this. */ + + for (;;) + { + yyn = yypact[yystate]; + if (yyn != YYPACT_NINF) + { + yyn += YYTERROR; + if (0 <= yyn && yyn <= YYLAST && yycheck[yyn] == YYTERROR) + { + yyn = yytable[yyn]; + if (0 < yyn) + break; + } + } + + /* Pop the current state because it cannot handle the error token. */ + if (yyssp == yyss) + YYABORT; + + YYDSYMPRINTF ("Error: popping", yystos[*yyssp], yyvsp, yylsp); + yydestruct (yystos[yystate], yyvsp); + YYPOPSTACK; + yystate = *yyssp; + YY_STACK_PRINT (yyss, yyssp); + } + + if (yyn == YYFINAL) + YYACCEPT; + + YYDPRINTF ((stderr, "Shifting error token, ")); + + *++yyvsp = yylval; + + + yystate = yyn; + goto yynewstate; + + +/*-------------------------------------. +| yyacceptlab -- YYACCEPT comes here. | +`-------------------------------------*/ +yyacceptlab: + yyresult = 0; + goto yyreturn; + +/*-----------------------------------. +| yyabortlab -- YYABORT comes here. | +`-----------------------------------*/ +yyabortlab: + yyresult = 1; + goto yyreturn; + +#ifndef yyoverflow +/*----------------------------------------------. +| yyoverflowlab -- parser overflow comes here. | +`----------------------------------------------*/ +yyoverflowlab: + yyerror (pc, "parser stack overflow"); + yyresult = 2; + /* Fall through. */ +#endif + +yyreturn: +#ifndef yyoverflow + if (yyss != yyssa) + YYSTACK_FREE (yyss); +#endif + return yyresult; +} + + +#line 568 "getdate.y" + + +static table const meridian_table[] = +{ + { "AM", tMERIDIAN, MERam }, + { "A.M.", tMERIDIAN, MERam }, + { "PM", tMERIDIAN, MERpm }, + { "P.M.", tMERIDIAN, MERpm }, + { NULL, 0, 0 } +}; + +static table const dst_table[] = +{ + { "DST", tDST, 0 } +}; + +static table const month_and_day_table[] = +{ + { "JANUARY", tMONTH, 1 }, + { "FEBRUARY", tMONTH, 2 }, + { "MARCH", tMONTH, 3 }, + { "APRIL", tMONTH, 4 }, + { "MAY", tMONTH, 5 }, + { "JUNE", tMONTH, 6 }, + { "JULY", tMONTH, 7 }, + { "AUGUST", tMONTH, 8 }, + { "SEPTEMBER",tMONTH, 9 }, + { "SEPT", tMONTH, 9 }, + { "OCTOBER", tMONTH, 10 }, + { "NOVEMBER", tMONTH, 11 }, + { "DECEMBER", tMONTH, 12 }, + { "SUNDAY", tDAY, 0 }, + { "MONDAY", tDAY, 1 }, + { "TUESDAY", tDAY, 2 }, + { "TUES", tDAY, 2 }, + { "WEDNESDAY",tDAY, 3 }, + { "WEDNES", tDAY, 3 }, + { "THURSDAY", tDAY, 4 }, + { "THUR", tDAY, 4 }, + { "THURS", tDAY, 4 }, + { "FRIDAY", tDAY, 5 }, + { "SATURDAY", tDAY, 6 }, + { NULL, 0, 0 } +}; + +static table const time_units_table[] = +{ + { "YEAR", tYEAR_UNIT, 1 }, + { "MONTH", tMONTH_UNIT, 1 }, + { "FORTNIGHT",tDAY_UNIT, 14 }, + { "WEEK", tDAY_UNIT, 7 }, + { "DAY", tDAY_UNIT, 1 }, + { "HOUR", tHOUR_UNIT, 1 }, + { "MINUTE", tMINUTE_UNIT, 1 }, + { "MIN", tMINUTE_UNIT, 1 }, + { "SECOND", tSEC_UNIT, 1 }, + { "SEC", tSEC_UNIT, 1 }, + { NULL, 0, 0 } +}; + +/* Assorted relative-time words. */ +static table const relative_time_table[] = +{ + { "TOMORROW", tDAY_UNIT, 1 }, + { "YESTERDAY",tDAY_UNIT, -1 }, + { "TODAY", tDAY_UNIT, 0 }, + { "NOW", tDAY_UNIT, 0 }, + { "LAST", tORDINAL, -1 }, + { "THIS", tORDINAL, 0 }, + { "NEXT", tORDINAL, 1 }, + { "FIRST", tORDINAL, 1 }, +/*{ "SECOND", tORDINAL, 2 }, */ + { "THIRD", tORDINAL, 3 }, + { "FOURTH", tORDINAL, 4 }, + { "FIFTH", tORDINAL, 5 }, + { "SIXTH", tORDINAL, 6 }, + { "SEVENTH", tORDINAL, 7 }, + { "EIGHTH", tORDINAL, 8 }, + { "NINTH", tORDINAL, 9 }, + { "TENTH", tORDINAL, 10 }, + { "ELEVENTH", tORDINAL, 11 }, + { "TWELFTH", tORDINAL, 12 }, + { "AGO", tAGO, 1 }, + { NULL, 0, 0 } +}; + +/* The universal time zone table. These labels can be used even for + time stamps that would not otherwise be valid, e.g., GMT time + stamps in London during summer. */ +static table const universal_time_zone_table[] = +{ + { "GMT", tZONE, HOUR ( 0) }, /* Greenwich Mean */ + { "UT", tZONE, HOUR ( 0) }, /* Universal (Coordinated) */ + { "UTC", tZONE, HOUR ( 0) }, + { NULL, 0, 0 } +}; + +/* The time zone table. This table is necessarily incomplete, as time + zone abbreviations are ambiguous; e.g. Australians interpret "EST" + as Eastern time in Australia, not as US Eastern Standard Time. + You cannot rely on getdate to handle arbitrary time zone + abbreviations; use numeric abbreviations like `-0500' instead. */ +static table const time_zone_table[] = +{ + { "WET", tZONE, HOUR ( 0) }, /* Western European */ + { "WEST", tDAYZONE, HOUR ( 0) }, /* Western European Summer */ + { "BST", tDAYZONE, HOUR ( 0) }, /* British Summer */ + { "ART", tZONE, -HOUR ( 3) }, /* Argentina */ + { "BRT", tZONE, -HOUR ( 3) }, /* Brazil */ + { "BRST", tDAYZONE, -HOUR ( 3) }, /* Brazil Summer */ + { "NST", tZONE, -(HOUR ( 3) + 30) }, /* Newfoundland Standard */ + { "NDT", tDAYZONE,-(HOUR ( 3) + 30) }, /* Newfoundland Daylight */ + { "AST", tZONE, -HOUR ( 4) }, /* Atlantic Standard */ + { "ADT", tDAYZONE, -HOUR ( 4) }, /* Atlantic Daylight */ + { "CLT", tZONE, -HOUR ( 4) }, /* Chile */ + { "CLST", tDAYZONE, -HOUR ( 4) }, /* Chile Summer */ + { "EST", tZONE, -HOUR ( 5) }, /* Eastern Standard */ + { "EDT", tDAYZONE, -HOUR ( 5) }, /* Eastern Daylight */ + { "CST", tZONE, -HOUR ( 6) }, /* Central Standard */ + { "CDT", tDAYZONE, -HOUR ( 6) }, /* Central Daylight */ + { "MST", tZONE, -HOUR ( 7) }, /* Mountain Standard */ + { "MDT", tDAYZONE, -HOUR ( 7) }, /* Mountain Daylight */ + { "PST", tZONE, -HOUR ( 8) }, /* Pacific Standard */ + { "PDT", tDAYZONE, -HOUR ( 8) }, /* Pacific Daylight */ + { "AKST", tZONE, -HOUR ( 9) }, /* Alaska Standard */ + { "AKDT", tDAYZONE, -HOUR ( 9) }, /* Alaska Daylight */ + { "HST", tZONE, -HOUR (10) }, /* Hawaii Standard */ + { "HAST", tZONE, -HOUR (10) }, /* Hawaii-Aleutian Standard */ + { "HADT", tDAYZONE, -HOUR (10) }, /* Hawaii-Aleutian Daylight */ + { "SST", tZONE, -HOUR (12) }, /* Samoa Standard */ + { "WAT", tZONE, HOUR ( 1) }, /* West Africa */ + { "CET", tZONE, HOUR ( 1) }, /* Central European */ + { "CEST", tDAYZONE, HOUR ( 1) }, /* Central European Summer */ + { "MET", tZONE, HOUR ( 1) }, /* Middle European */ + { "MEZ", tZONE, HOUR ( 1) }, /* Middle European */ + { "MEST", tDAYZONE, HOUR ( 1) }, /* Middle European Summer */ + { "MESZ", tDAYZONE, HOUR ( 1) }, /* Middle European Summer */ + { "EET", tZONE, HOUR ( 2) }, /* Eastern European */ + { "EEST", tDAYZONE, HOUR ( 2) }, /* Eastern European Summer */ + { "CAT", tZONE, HOUR ( 2) }, /* Central Africa */ + { "SAST", tZONE, HOUR ( 2) }, /* South Africa Standard */ + { "EAT", tZONE, HOUR ( 3) }, /* East Africa */ + { "MSK", tZONE, HOUR ( 3) }, /* Moscow */ + { "MSD", tDAYZONE, HOUR ( 3) }, /* Moscow Daylight */ + { "IST", tZONE, (HOUR ( 5) + 30) }, /* India Standard */ + { "SGT", tZONE, HOUR ( 8) }, /* Singapore */ + { "KST", tZONE, HOUR ( 9) }, /* Korea Standard */ + { "JST", tZONE, HOUR ( 9) }, /* Japan Standard */ + { "GST", tZONE, HOUR (10) }, /* Guam Standard */ + { "NZST", tZONE, HOUR (12) }, /* New Zealand Standard */ + { "NZDT", tDAYZONE, HOUR (12) }, /* New Zealand Daylight */ + { NULL, 0, 0 } +}; + +/* Military time zone table. */ +static table const military_table[] = +{ + { "A", tZONE, -HOUR ( 1) }, + { "B", tZONE, -HOUR ( 2) }, + { "C", tZONE, -HOUR ( 3) }, + { "D", tZONE, -HOUR ( 4) }, + { "E", tZONE, -HOUR ( 5) }, + { "F", tZONE, -HOUR ( 6) }, + { "G", tZONE, -HOUR ( 7) }, + { "H", tZONE, -HOUR ( 8) }, + { "I", tZONE, -HOUR ( 9) }, + { "K", tZONE, -HOUR (10) }, + { "L", tZONE, -HOUR (11) }, + { "M", tZONE, -HOUR (12) }, + { "N", tZONE, HOUR ( 1) }, + { "O", tZONE, HOUR ( 2) }, + { "P", tZONE, HOUR ( 3) }, + { "Q", tZONE, HOUR ( 4) }, + { "R", tZONE, HOUR ( 5) }, + { "S", tZONE, HOUR ( 6) }, + { "T", tZONE, HOUR ( 7) }, + { "U", tZONE, HOUR ( 8) }, + { "V", tZONE, HOUR ( 9) }, + { "W", tZONE, HOUR (10) }, + { "X", tZONE, HOUR (11) }, + { "Y", tZONE, HOUR (12) }, + { "Z", tZONE, HOUR ( 0) }, + { NULL, 0, 0 } +}; + + + +/* Convert a time zone expressed as HH:MM into an integer count of + minutes. If MM is negative, then S is of the form HHMM and needs + to be picked apart; otherwise, S is of the form HH. */ + +static long int +time_zone_hhmm (textint s, long int mm) +{ + if (mm < 0) + return (s.value / 100) * 60 + s.value % 100; + else + return s.value * 60 + (s.negative ? -mm : mm); +} + +static int +to_hour (long int hours, int meridian) +{ + switch (meridian) + { + default: /* Pacify GCC. */ + case MER24: + return 0 <= hours && hours < 24 ? hours : -1; + case MERam: + return 0 < hours && hours < 12 ? hours : hours == 12 ? 0 : -1; + case MERpm: + return 0 < hours && hours < 12 ? hours + 12 : hours == 12 ? 12 : -1; + } +} + +static long int +to_year (textint textyear) +{ + long int year = textyear.value; + + if (year < 0) + year = -year; + + /* XPG4 suggests that years 00-68 map to 2000-2068, and + years 69-99 map to 1969-1999. */ + else if (textyear.digits == 2) + year += year < 69 ? 2000 : 1900; + + return year; +} + +static table const * +lookup_zone (parser_control const *pc, char const *name) +{ + table const *tp; + + for (tp = universal_time_zone_table; tp->name; tp++) + if (strcmp (name, tp->name) == 0) + return tp; + + /* Try local zone abbreviations before those in time_zone_table, as + the local ones are more likely to be right. */ + for (tp = pc->local_time_zone_table; tp->name; tp++) + if (strcmp (name, tp->name) == 0) + return tp; + + for (tp = time_zone_table; tp->name; tp++) + if (strcmp (name, tp->name) == 0) + return tp; + + return NULL; +} + +#if ! HAVE_TM_GMTOFF +/* Yield the difference between *A and *B, + measured in seconds, ignoring leap seconds. + The body of this function is taken directly from the GNU C Library; + see src/strftime.c. */ +static long int +tm_diff (struct tm const *a, struct tm const *b) +{ + /* Compute intervening leap days correctly even if year is negative. + Take care to avoid int overflow in leap day calculations. */ + int a4 = SHR (a->tm_year, 2) + SHR (TM_YEAR_BASE, 2) - ! (a->tm_year & 3); + int b4 = SHR (b->tm_year, 2) + SHR (TM_YEAR_BASE, 2) - ! (b->tm_year & 3); + int a100 = a4 / 25 - (a4 % 25 < 0); + int b100 = b4 / 25 - (b4 % 25 < 0); + int a400 = SHR (a100, 2); + int b400 = SHR (b100, 2); + int intervening_leap_days = (a4 - b4) - (a100 - b100) + (a400 - b400); + long int ayear = a->tm_year; + long int years = ayear - b->tm_year; + long int days = (365 * years + intervening_leap_days + + (a->tm_yday - b->tm_yday)); + return (60 * (60 * (24 * days + (a->tm_hour - b->tm_hour)) + + (a->tm_min - b->tm_min)) + + (a->tm_sec - b->tm_sec)); +} +#endif /* ! HAVE_TM_GMTOFF */ + +static table const * +lookup_word (parser_control const *pc, char *word) +{ + char *p; + char *q; + size_t wordlen; + table const *tp; + bool period_found; + bool abbrev; + + /* Make it uppercase. */ + for (p = word; *p; p++) + { + unsigned char ch = *p; + if (ISLOWER (ch)) + *p = toupper (ch); + } + + for (tp = meridian_table; tp->name; tp++) + if (strcmp (word, tp->name) == 0) + return tp; + + /* See if we have an abbreviation for a month. */ + wordlen = strlen (word); + abbrev = wordlen == 3 || (wordlen == 4 && word[3] == '.'); + + for (tp = month_and_day_table; tp->name; tp++) + if ((abbrev ? strncmp (word, tp->name, 3) : strcmp (word, tp->name)) == 0) + return tp; + + if ((tp = lookup_zone (pc, word))) + return tp; + + if (strcmp (word, dst_table[0].name) == 0) + return dst_table; + + for (tp = time_units_table; tp->name; tp++) + if (strcmp (word, tp->name) == 0) + return tp; + + /* Strip off any plural and try the units table again. */ + if (word[wordlen - 1] == 'S') + { + word[wordlen - 1] = '\0'; + for (tp = time_units_table; tp->name; tp++) + if (strcmp (word, tp->name) == 0) + return tp; + word[wordlen - 1] = 'S'; /* For "this" in relative_time_table. */ + } + + for (tp = relative_time_table; tp->name; tp++) + if (strcmp (word, tp->name) == 0) + return tp; + + /* Military time zones. */ + if (wordlen == 1) + for (tp = military_table; tp->name; tp++) + if (word[0] == tp->name[0]) + return tp; + + /* Drop out any periods and try the time zone table again. */ + for (period_found = false, p = q = word; (*p = *q); q++) + if (*q == '.') + period_found = true; + else + p++; + if (period_found && (tp = lookup_zone (pc, word))) + return tp; + + return NULL; +} + +static int +yylex (YYSTYPE *lvalp, parser_control *pc) +{ + unsigned char c; + size_t count; + + for (;;) + { + while (c = *pc->input, ISSPACE (c)) + pc->input++; + + if (ISDIGIT (c) || c == '-' || c == '+') + { + char const *p; + int sign; + unsigned long int value; + if (c == '-' || c == '+') + { + sign = c == '-' ? -1 : 1; + while (c = *++pc->input, ISSPACE (c)) + continue; + if (! ISDIGIT (c)) + /* skip the '-' sign */ + continue; + } + else + sign = 0; + p = pc->input; + for (value = 0; ; value *= 10) + { + unsigned long int value1 = value + (c - '0'); + if (value1 < value) + return '?'; + value = value1; + c = *++p; + if (! ISDIGIT (c)) + break; + if (ULONG_MAX / 10 < value) + return '?'; + } + if ((c == '.' || c == ',') && ISDIGIT (p[1])) + { + time_t s; + int ns; + int digits; + unsigned long int value1; + + /* Check for overflow when converting value to time_t. */ + if (sign < 0) + { + s = - value; + if (0 < s) + return '?'; + value1 = -s; + } + else + { + s = value; + if (s < 0) + return '?'; + value1 = s; + } + if (value != value1) + return '?'; + + /* Accumulate fraction, to ns precision. */ + p++; + ns = *p++ - '0'; + for (digits = 2; digits <= LOG10_BILLION; digits++) + { + ns *= 10; + if (ISDIGIT (*p)) + ns += *p++ - '0'; + } + + /* Skip excess digits, truncating toward -Infinity. */ + if (sign < 0) + for (; ISDIGIT (*p); p++) + if (*p != '0') + { + ns++; + break; + } + while (ISDIGIT (*p)) + p++; + + /* Adjust to the timespec convention, which is that + tv_nsec is always a positive offset even if tv_sec is + negative. */ + if (sign < 0 && ns) + { + s--; + if (! (s < 0)) + return '?'; + ns = BILLION - ns; + } + + lvalp->timespec.tv_sec = s; + lvalp->timespec.tv_nsec = ns; + pc->input = p; + return sign ? tSDECIMAL_NUMBER : tUDECIMAL_NUMBER; + } + else + { + lvalp->textintval.negative = sign < 0; + if (sign < 0) + { + lvalp->textintval.value = - value; + if (0 < lvalp->textintval.value) + return '?'; + } + else + { + lvalp->textintval.value = value; + if (lvalp->textintval.value < 0) + return '?'; + } + lvalp->textintval.digits = p - pc->input; + pc->input = p; + return sign ? tSNUMBER : tUNUMBER; + } + } + + if (ISALPHA (c)) + { + char buff[20]; + char *p = buff; + table const *tp; + + do + { + if (p < buff + sizeof buff - 1) + *p++ = c; + c = *++pc->input; + } + while (ISALPHA (c) || c == '.'); + + *p = '\0'; + tp = lookup_word (pc, buff); + if (! tp) + return '?'; + lvalp->intval = tp->value; + return tp->type; + } + + if (c != '(') + return *pc->input++; + count = 0; + do + { + c = *pc->input++; + if (c == '\0') + return c; + if (c == '(') + count++; + else if (c == ')') + count--; + } + while (count != 0); + } +} + +/* Do nothing if the parser reports an error. */ +static int +yyerror (parser_control *pc ATTRIBUTE_UNUSED, char *s ATTRIBUTE_UNUSED) +{ + return 0; +} + +/* If *TM0 is the old and *TM1 is the new value of a struct tm after + passing it to mktime, return true if it's OK that mktime returned T. + It's not OK if *TM0 has out-of-range members. */ + +static bool +mktime_ok (struct tm const *tm0, struct tm const *tm1, time_t t) +{ + if (t == (time_t) -1) + { + /* Guard against falsely reporting an error when parsing a time + stamp that happens to equal (time_t) -1, on a host that + supports such a time stamp. */ + tm1 = localtime (&t); + if (!tm1) + return false; + } + + return ! ((tm0->tm_sec ^ tm1->tm_sec) + | (tm0->tm_min ^ tm1->tm_min) + | (tm0->tm_hour ^ tm1->tm_hour) + | (tm0->tm_mday ^ tm1->tm_mday) + | (tm0->tm_mon ^ tm1->tm_mon) + | (tm0->tm_year ^ tm1->tm_year)); +} + +/* A reasonable upper bound for the size of ordinary TZ strings. + Use heap allocation if TZ's length exceeds this. */ +enum { TZBUFSIZE = 100 }; + +/* Return a copy of TZ, stored in TZBUF if it fits, and heap-allocated + otherwise. */ +static char * +get_tz (char tzbuf[TZBUFSIZE]) +{ + char *tz = getenv ("TZ"); + if (tz) + { + size_t tzsize = strlen (tz) + 1; + tz = (tzsize <= TZBUFSIZE + ? memcpy (tzbuf, tz, tzsize) + : xmemdup (tz, tzsize)); + } + return tz; +} + +/* Parse a date/time string, storing the resulting time value into *RESULT. + The string itself is pointed to by P. Return true if successful. + P can be an incomplete or relative time specification; if so, use + *NOW as the basis for the returned time. */ +bool +get_date (struct timespec *result, char const *p, struct timespec const *now) +{ + time_t Start; + long int Start_ns; + struct tm const *tmp; + struct tm tm; + struct tm tm0; + parser_control pc; + struct timespec gettime_buffer; + unsigned char c; + bool tz_was_altered = false; + char *tz0 = NULL; + char tz0buf[TZBUFSIZE]; + bool ok = true; + + if (! now) + { + gettime (&gettime_buffer); + now = &gettime_buffer; + } + + Start = now->tv_sec; + Start_ns = now->tv_nsec; + + tmp = localtime (&now->tv_sec); + if (! tmp) + return false; + + while (c = *p, ISSPACE (c)) + p++; + + if (strncmp (p, "TZ=\"", 4) == 0) + { + char const *tzbase = p + 4; + size_t tzsize = 1; + char const *s; + + for (s = tzbase; *s; s++, tzsize++) + if (*s == '\\') + { + s++; + if (! (*s == '\\' || *s == '"')) + break; + } + else if (*s == '"') + { + char *z; + char *tz1; + char tz1buf[TZBUFSIZE]; + bool large_tz = TZBUFSIZE < tzsize; + bool setenv_ok; + tz0 = get_tz (tz0buf); + z = tz1 = large_tz ? xmalloc (tzsize) : tz1buf; + for (s = tzbase; *s != '"'; s++) + *z++ = *(s += *s == '\\'); + *z = '\0'; + setenv_ok = setenv ("TZ", tz1, 1) == 0; + if (large_tz) + free (tz1); + if (!setenv_ok) + goto fail; + tz_was_altered = true; + p = s + 1; + } + } + + pc.input = p; + pc.year.value = tmp->tm_year; + pc.year.value += TM_YEAR_BASE; + pc.year.digits = 0; + pc.month = tmp->tm_mon + 1; + pc.day = tmp->tm_mday; + pc.hour = tmp->tm_hour; + pc.minutes = tmp->tm_min; + pc.seconds.tv_sec = tmp->tm_sec; + pc.seconds.tv_nsec = Start_ns; + tm.tm_isdst = tmp->tm_isdst; + + pc.meridian = MER24; + pc.rel_ns = 0; + pc.rel_seconds = 0; + pc.rel_minutes = 0; + pc.rel_hour = 0; + pc.rel_day = 0; + pc.rel_month = 0; + pc.rel_year = 0; + pc.timespec_seen = false; + pc.rels_seen = false; + pc.dates_seen = 0; + pc.days_seen = 0; + pc.times_seen = 0; + pc.local_zones_seen = 0; + pc.dsts_seen = 0; + pc.zones_seen = 0; + +#if HAVE_STRUCT_TM_TM_ZONE + pc.local_time_zone_table[0].name = tmp->tm_zone; + pc.local_time_zone_table[0].type = tLOCAL_ZONE; + pc.local_time_zone_table[0].value = tmp->tm_isdst; + pc.local_time_zone_table[1].name = NULL; + + /* Probe the names used in the next three calendar quarters, looking + for a tm_isdst different from the one we already have. */ + { + int quarter; + for (quarter = 1; quarter <= 3; quarter++) + { + time_t probe = Start + quarter * (90 * 24 * 60 * 60); + struct tm const *probe_tm = localtime (&probe); + if (probe_tm && probe_tm->tm_zone + && probe_tm->tm_isdst != pc.local_time_zone_table[0].value) + { + { + pc.local_time_zone_table[1].name = probe_tm->tm_zone; + pc.local_time_zone_table[1].type = tLOCAL_ZONE; + pc.local_time_zone_table[1].value = probe_tm->tm_isdst; + pc.local_time_zone_table[2].name = NULL; + } + break; + } + } + } +#else +#if HAVE_TZNAME + { +# ifndef tzname + extern char *tzname[]; +# endif + int i; + for (i = 0; i < 2; i++) + { + pc.local_time_zone_table[i].name = tzname[i]; + pc.local_time_zone_table[i].type = tLOCAL_ZONE; + pc.local_time_zone_table[i].value = i; + } + pc.local_time_zone_table[i].name = NULL; + } +#else + pc.local_time_zone_table[0].name = NULL; +#endif +#endif + + if (pc.local_time_zone_table[0].name && pc.local_time_zone_table[1].name + && ! strcmp (pc.local_time_zone_table[0].name, + pc.local_time_zone_table[1].name)) + { + /* This locale uses the same abbrevation for standard and + daylight times. So if we see that abbreviation, we don't + know whether it's daylight time. */ + pc.local_time_zone_table[0].value = -1; + pc.local_time_zone_table[1].name = NULL; + } + + if (yyparse (&pc) != 0) + goto fail; + + if (pc.timespec_seen) + *result = pc.seconds; + else + { + if (1 < (pc.times_seen | pc.dates_seen | pc.days_seen | pc.dsts_seen + | (pc.local_zones_seen + pc.zones_seen))) + goto fail; + + tm.tm_year = to_year (pc.year) - TM_YEAR_BASE; + tm.tm_mon = pc.month - 1; + tm.tm_mday = pc.day; + if (pc.times_seen || (pc.rels_seen && ! pc.dates_seen && ! pc.days_seen)) + { + tm.tm_hour = to_hour (pc.hour, pc.meridian); + if (tm.tm_hour < 0) + goto fail; + tm.tm_min = pc.minutes; + tm.tm_sec = pc.seconds.tv_sec; + } + else + { + tm.tm_hour = tm.tm_min = tm.tm_sec = 0; + pc.seconds.tv_nsec = 0; + } + + /* Let mktime deduce tm_isdst if we have an absolute time stamp. */ + if (!pc.rels_seen) + tm.tm_isdst = -1; + + /* But if the input explicitly specifies local time with or without + DST, give mktime that information. */ + if (pc.local_zones_seen) + tm.tm_isdst = pc.local_isdst; + + tm0 = tm; + + Start = mktime (&tm); + + if (! mktime_ok (&tm0, &tm, Start)) + { + if (! pc.zones_seen) + goto fail; + else + { + /* Guard against falsely reporting errors near the time_t + boundaries when parsing times in other time zones. For + example, suppose the input string "1969-12-31 23:00:00 -0100", + the current time zone is 8 hours ahead of UTC, and the min + time_t value is 1970-01-01 00:00:00 UTC. Then the min + localtime value is 1970-01-01 08:00:00, and mktime will + therefore fail on 1969-12-31 23:00:00. To work around the + problem, set the time zone to 1 hour behind UTC temporarily + by setting TZ="XXX1:00" and try mktime again. */ + + long int time_zone = pc.time_zone; + long int abs_time_zone = time_zone < 0 ? - time_zone : time_zone; + long int abs_time_zone_hour = abs_time_zone / 60; + int abs_time_zone_min = abs_time_zone % 60; + char tz1buf[sizeof "XXX+0:00" + + sizeof pc.time_zone * CHAR_BIT / 3]; + if (!tz_was_altered) + tz0 = get_tz (tz0buf); + sprintf (tz1buf, "XXX%s%ld:%02d", "-" + (time_zone < 0), + abs_time_zone_hour, abs_time_zone_min); + if (setenv ("TZ", tz1buf, 1) != 0) + goto fail; + tz_was_altered = true; + tm = tm0; + Start = mktime (&tm); + if (! mktime_ok (&tm0, &tm, Start)) + goto fail; + } + } + + if (pc.days_seen && ! pc.dates_seen) + { + tm.tm_mday += ((pc.day_number - tm.tm_wday + 7) % 7 + + 7 * (pc.day_ordinal - (0 < pc.day_ordinal))); + tm.tm_isdst = -1; + Start = mktime (&tm); + if (Start == (time_t) -1) + goto fail; + } + + if (pc.zones_seen) + { + long int delta = pc.time_zone * 60; + time_t t1; +#ifdef HAVE_TM_GMTOFF + delta -= tm.tm_gmtoff; +#else + time_t t = Start; + struct tm const *gmt = gmtime (&t); + if (! gmt) + goto fail; + delta -= tm_diff (&tm, gmt); +#endif + t1 = Start - delta; + if ((Start < t1) != (delta < 0)) + goto fail; /* time_t overflow */ + Start = t1; + } + + /* Add relative date. */ + if (pc.rel_year | pc.rel_month | pc.rel_day) + { + int year = tm.tm_year + pc.rel_year; + int month = tm.tm_mon + pc.rel_month; + int day = tm.tm_mday + pc.rel_day; + if (((year < tm.tm_year) ^ (pc.rel_year < 0)) + | ((month < tm.tm_mon) ^ (pc.rel_month < 0)) + | ((day < tm.tm_mday) ^ (pc.rel_day < 0))) + goto fail; + tm.tm_year = year; + tm.tm_mon = month; + tm.tm_mday = day; + Start = mktime (&tm); + if (Start == (time_t) -1) + goto fail; + } + + /* Add relative hours, minutes, and seconds. On hosts that support + leap seconds, ignore the possibility of leap seconds; e.g., + "+ 10 minutes" adds 600 seconds, even if one of them is a + leap second. Typically this is not what the user wants, but it's + too hard to do it the other way, because the time zone indicator + must be applied before relative times, and if mktime is applied + again the time zone will be lost. */ + { + long int sum_ns = pc.seconds.tv_nsec + pc.rel_ns; + long int normalized_ns = (sum_ns % BILLION + BILLION) % BILLION; + time_t t0 = Start; + long int d1 = 60 * 60 * pc.rel_hour; + time_t t1 = t0 + d1; + long int d2 = 60 * pc.rel_minutes; + time_t t2 = t1 + d2; + long int d3 = pc.rel_seconds; + time_t t3 = t2 + d3; + long int d4 = (sum_ns - normalized_ns) / BILLION; + time_t t4 = t3 + d4; + + if ((d1 / (60 * 60) ^ pc.rel_hour) + | (d2 / 60 ^ pc.rel_minutes) + | ((t1 < t0) ^ (d1 < 0)) + | ((t2 < t1) ^ (d2 < 0)) + | ((t3 < t2) ^ (d3 < 0)) + | ((t4 < t3) ^ (d4 < 0))) + goto fail; + + result->tv_sec = t4; + result->tv_nsec = normalized_ns; + } + } + + goto done; + + fail: + ok = false; + done: + if (tz_was_altered) + ok &= (tz0 ? setenv ("TZ", tz0, 1) : unsetenv ("TZ")) == 0; + if (tz0 != tz0buf) + free (tz0); + return ok; +} + +#if TEST + +int +main (int ac, char **av) +{ + char buff[BUFSIZ]; + + printf ("Enter date, or blank line to exit.\n\t> "); + fflush (stdout); + + buff[BUFSIZ - 1] = '\0'; + while (fgets (buff, BUFSIZ - 1, stdin) && buff[0]) + { + struct timespec d; + struct tm const *tm; + if (! get_date (&d, buff, NULL)) + printf ("Bad format - couldn't convert.\n"); + else if (! (tm = localtime (&d.tv_sec))) + { + long int sec = d.tv_sec; + printf ("localtime (%ld) failed\n", sec); + } + else + { + int ns = d.tv_nsec; + printf ("%04ld-%02d-%02d %02d:%02d:%02d.%09d\n", + tm->tm_year + 1900L, tm->tm_mon + 1, tm->tm_mday, + tm->tm_hour, tm->tm_min, tm->tm_sec, ns); + } + printf ("\t> "); + fflush (stdout); + } + return 0; +} +#endif /* TEST */ + + diff --git a/lib/getdate.h b/lib/getdate.h new file mode 100644 index 0000000..4694cdb --- /dev/null +++ b/lib/getdate.h @@ -0,0 +1,22 @@ +/* Parse a string into an internal time stamp. + + Copyright (C) 1995, 1997, 1998, 2003, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#include <stdbool.h> +#include "timespec.h" + +bool get_date (struct timespec *, char const *, struct timespec const *); diff --git a/lib/getdate.y b/lib/getdate.y new file mode 100644 index 0000000..868f983 --- /dev/null +++ b/lib/getdate.y @@ -0,0 +1,1495 @@ +%{ +/* Parse a string into an internal time stamp. + + Copyright (C) 1999, 2000, 2002, 2003, 2004, 2005 Free Software + Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Originally written by Steven M. Bellovin <smb@research.att.com> while + at the University of North Carolina at Chapel Hill. Later tweaked by + a couple of people on Usenet. Completely overhauled by Rich $alz + <rsalz@bbn.com> and Jim Berets <jberets@bbn.com> in August, 1990. + + Modified by Paul Eggert <eggert@twinsun.com> in August 1999 to do + the right thing about local DST. Also modified by Paul Eggert + <eggert@cs.ucla.edu> in February 2004 to support + nanosecond-resolution time stamps, and in October 2004 to support + TZ strings in dates. */ + +/* FIXME: Check for arithmetic overflow in all cases, not just + some of them. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "getdate.h" + +/* There's no need to extend the stack, so there's no need to involve + alloca. */ +#define YYSTACK_USE_ALLOCA 0 + +/* Tell Bison how much stack space is needed. 20 should be plenty for + this grammar, which is not right recursive. Beware setting it too + high, since that might cause problems on machines whose + implementations have lame stack-overflow checking. */ +#define YYMAXDEPTH 20 +#define YYINITDEPTH YYMAXDEPTH + +/* Since the code of getdate.y is not included in the Emacs executable + itself, there is no need to #define static in this file. Even if + the code were included in the Emacs executable, it probably + wouldn't do any harm to #undef it here; this will only cause + problems if we try to write to a static variable, which I don't + think this code needs to do. */ +#ifdef emacs +# undef static +#endif + +#include <ctype.h> +#include <limits.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +#include "setenv.h" +#include "xalloc.h" + +#if STDC_HEADERS || (! defined isascii && ! HAVE_ISASCII) +# define IN_CTYPE_DOMAIN(c) 1 +#else +# define IN_CTYPE_DOMAIN(c) isascii (c) +#endif + +#define ISSPACE(c) (IN_CTYPE_DOMAIN (c) && isspace (c)) +#define ISALPHA(c) (IN_CTYPE_DOMAIN (c) && isalpha (c)) +#define ISLOWER(c) (IN_CTYPE_DOMAIN (c) && islower (c)) + +/* ISDIGIT differs from isdigit, as follows: + - Its arg may be any int or unsigned int; it need not be an unsigned char. + - It's guaranteed to evaluate its argument exactly once. + - It's typically faster. + POSIX says that only '0' through '9' are digits. Prefer ISDIGIT to + isdigit unless it's important to use the locale's definition + of `digit' even when the host does not conform to POSIX. */ +#define ISDIGIT(c) ((unsigned int) (c) - '0' <= 9) + +#if __GNUC__ < 2 || (__GNUC__ == 2 && __GNUC_MINOR__ < 8) || __STRICT_ANSI__ +# define __attribute__(x) +#endif + +#ifndef ATTRIBUTE_UNUSED +# define ATTRIBUTE_UNUSED __attribute__ ((__unused__)) +#endif + +/* Shift A right by B bits portably, by dividing A by 2**B and + truncating towards minus infinity. A and B should be free of side + effects, and B should be in the range 0 <= B <= INT_BITS - 2, where + INT_BITS is the number of useful bits in an int. GNU code can + assume that INT_BITS is at least 32. + + ISO C99 says that A >> B is implementation-defined if A < 0. Some + implementations (e.g., UNICOS 9.0 on a Cray Y-MP EL) don't shift + right in the usual way when A < 0, so SHR falls back on division if + ordinary A >> B doesn't seem to be the usual signed shift. */ +#define SHR(a, b) \ + (-1 >> 1 == -1 \ + ? (a) >> (b) \ + : (a) / (1 << (b)) - ((a) % (1 << (b)) < 0)) + +#define EPOCH_YEAR 1970 +#define TM_YEAR_BASE 1900 + +#define HOUR(x) ((x) * 60) + +/* An integer value, and the number of digits in its textual + representation. */ +typedef struct +{ + bool negative; + long int value; + size_t digits; +} textint; + +/* An entry in the lexical lookup table. */ +typedef struct +{ + char const *name; + int type; + int value; +} table; + +/* Meridian: am, pm, or 24-hour style. */ +enum { MERam, MERpm, MER24 }; + +enum { BILLION = 1000000000, LOG10_BILLION = 9 }; + +/* Information passed to and from the parser. */ +typedef struct +{ + /* The input string remaining to be parsed. */ + const char *input; + + /* N, if this is the Nth Tuesday. */ + long int day_ordinal; + + /* Day of week; Sunday is 0. */ + int day_number; + + /* tm_isdst flag for the local zone. */ + int local_isdst; + + /* Time zone, in minutes east of UTC. */ + long int time_zone; + + /* Style used for time. */ + int meridian; + + /* Gregorian year, month, day, hour, minutes, seconds, and nanoseconds. */ + textint year; + long int month; + long int day; + long int hour; + long int minutes; + struct timespec seconds; /* includes nanoseconds */ + + /* Relative year, month, day, hour, minutes, seconds, and nanoseconds. */ + long int rel_year; + long int rel_month; + long int rel_day; + long int rel_hour; + long int rel_minutes; + long int rel_seconds; + long int rel_ns; + + /* Presence or counts of nonterminals of various flavors parsed so far. */ + bool timespec_seen; + bool rels_seen; + size_t dates_seen; + size_t days_seen; + size_t local_zones_seen; + size_t dsts_seen; + size_t times_seen; + size_t zones_seen; + + /* Table of local time zone abbrevations, terminated by a null entry. */ + table local_time_zone_table[3]; +} parser_control; + +union YYSTYPE; +static int yylex (union YYSTYPE *, parser_control *); +static int yyerror (parser_control *, char *); +static long int time_zone_hhmm (textint, long int); + +%} + +/* We want a reentrant parser, even if the TZ manipulation and the calls to + localtime and gmtime are not reentrant. */ +%pure-parser +%parse-param { parser_control *pc } +%lex-param { parser_control *pc } + +/* This grammar has 20 shift/reduce conflicts. */ +%expect 20 + +%union +{ + long int intval; + textint textintval; + struct timespec timespec; +} + +%token tAGO tDST + +%token <intval> tDAY tDAY_UNIT tDAYZONE tHOUR_UNIT tLOCAL_ZONE tMERIDIAN +%token <intval> tMINUTE_UNIT tMONTH tMONTH_UNIT tORDINAL +%token <intval> tSEC_UNIT tYEAR_UNIT tZONE + +%token <textintval> tSNUMBER tUNUMBER +%token <timespec> tSDECIMAL_NUMBER tUDECIMAL_NUMBER + +%type <intval> o_colon_minutes o_merid +%type <timespec> seconds signed_seconds unsigned_seconds + +%% + +spec: + timespec + | items + ; + +timespec: + '@' seconds + { + pc->seconds = $2; + pc->timespec_seen = true; + } + ; + +items: + /* empty */ + | items item + ; + +item: + time + { pc->times_seen++; } + | local_zone + { pc->local_zones_seen++; } + | zone + { pc->zones_seen++; } + | date + { pc->dates_seen++; } + | day + { pc->days_seen++; } + | rel + { pc->rels_seen = true; } + | number + ; + +time: + tUNUMBER tMERIDIAN + { + pc->hour = $1.value; + pc->minutes = 0; + pc->seconds.tv_sec = 0; + pc->seconds.tv_nsec = 0; + pc->meridian = $2; + } + | tUNUMBER ':' tUNUMBER o_merid + { + pc->hour = $1.value; + pc->minutes = $3.value; + pc->seconds.tv_sec = 0; + pc->seconds.tv_nsec = 0; + pc->meridian = $4; + } + | tUNUMBER ':' tUNUMBER tSNUMBER o_colon_minutes + { + pc->hour = $1.value; + pc->minutes = $3.value; + pc->seconds.tv_sec = 0; + pc->seconds.tv_nsec = 0; + pc->meridian = MER24; + pc->zones_seen++; + pc->time_zone = time_zone_hhmm ($4, $5); + } + | tUNUMBER ':' tUNUMBER ':' unsigned_seconds o_merid + { + pc->hour = $1.value; + pc->minutes = $3.value; + pc->seconds = $5; + pc->meridian = $6; + } + | tUNUMBER ':' tUNUMBER ':' unsigned_seconds tSNUMBER o_colon_minutes + { + pc->hour = $1.value; + pc->minutes = $3.value; + pc->seconds = $5; + pc->meridian = MER24; + pc->zones_seen++; + pc->time_zone = time_zone_hhmm ($6, $7); + } + ; + +local_zone: + tLOCAL_ZONE + { + pc->local_isdst = $1; + pc->dsts_seen += (0 < $1); + } + | tLOCAL_ZONE tDST + { + pc->local_isdst = 1; + pc->dsts_seen += (0 < $1) + 1; + } + ; + +zone: + tZONE + { pc->time_zone = $1; } + | tZONE relunit_snumber + { pc->time_zone = $1; pc->rels_seen = true; } + | tZONE tSNUMBER o_colon_minutes + { pc->time_zone = $1 + time_zone_hhmm ($2, $3); } + | tDAYZONE + { pc->time_zone = $1 + 60; } + | tZONE tDST + { pc->time_zone = $1 + 60; } + ; + +day: + tDAY + { + pc->day_ordinal = 1; + pc->day_number = $1; + } + | tDAY ',' + { + pc->day_ordinal = 1; + pc->day_number = $1; + } + | tORDINAL tDAY + { + pc->day_ordinal = $1; + pc->day_number = $2; + } + | tUNUMBER tDAY + { + pc->day_ordinal = $1.value; + pc->day_number = $2; + } + ; + +date: + tUNUMBER '/' tUNUMBER + { + pc->month = $1.value; + pc->day = $3.value; + } + | tUNUMBER '/' tUNUMBER '/' tUNUMBER + { + /* Interpret as YYYY/MM/DD if the first value has 4 or more digits, + otherwise as MM/DD/YY. + The goal in recognizing YYYY/MM/DD is solely to support legacy + machine-generated dates like those in an RCS log listing. If + you want portability, use the ISO 8601 format. */ + if (4 <= $1.digits) + { + pc->year = $1; + pc->month = $3.value; + pc->day = $5.value; + } + else + { + pc->month = $1.value; + pc->day = $3.value; + pc->year = $5; + } + } + | tUNUMBER tSNUMBER tSNUMBER + { + /* ISO 8601 format. YYYY-MM-DD. */ + pc->year = $1; + pc->month = -$2.value; + pc->day = -$3.value; + } + | tUNUMBER tMONTH tSNUMBER + { + /* e.g. 17-JUN-1992. */ + pc->day = $1.value; + pc->month = $2; + pc->year.value = -$3.value; + pc->year.digits = $3.digits; + } + | tMONTH tSNUMBER tSNUMBER + { + /* e.g. JUN-17-1992. */ + pc->month = $1; + pc->day = -$2.value; + pc->year.value = -$3.value; + pc->year.digits = $3.digits; + } + | tMONTH tUNUMBER + { + pc->month = $1; + pc->day = $2.value; + } + | tMONTH tUNUMBER ',' tUNUMBER + { + pc->month = $1; + pc->day = $2.value; + pc->year = $4; + } + | tUNUMBER tMONTH + { + pc->day = $1.value; + pc->month = $2; + } + | tUNUMBER tMONTH tUNUMBER + { + pc->day = $1.value; + pc->month = $2; + pc->year = $3; + } + ; + +rel: + relunit tAGO + { + pc->rel_ns = -pc->rel_ns; + pc->rel_seconds = -pc->rel_seconds; + pc->rel_minutes = -pc->rel_minutes; + pc->rel_hour = -pc->rel_hour; + pc->rel_day = -pc->rel_day; + pc->rel_month = -pc->rel_month; + pc->rel_year = -pc->rel_year; + } + | relunit + ; + +relunit: + tORDINAL tYEAR_UNIT + { pc->rel_year += $1 * $2; } + | tUNUMBER tYEAR_UNIT + { pc->rel_year += $1.value * $2; } + | tYEAR_UNIT + { pc->rel_year += $1; } + | tORDINAL tMONTH_UNIT + { pc->rel_month += $1 * $2; } + | tUNUMBER tMONTH_UNIT + { pc->rel_month += $1.value * $2; } + | tMONTH_UNIT + { pc->rel_month += $1; } + | tORDINAL tDAY_UNIT + { pc->rel_day += $1 * $2; } + | tUNUMBER tDAY_UNIT + { pc->rel_day += $1.value * $2; } + | tDAY_UNIT + { pc->rel_day += $1; } + | tORDINAL tHOUR_UNIT + { pc->rel_hour += $1 * $2; } + | tUNUMBER tHOUR_UNIT + { pc->rel_hour += $1.value * $2; } + | tHOUR_UNIT + { pc->rel_hour += $1; } + | tORDINAL tMINUTE_UNIT + { pc->rel_minutes += $1 * $2; } + | tUNUMBER tMINUTE_UNIT + { pc->rel_minutes += $1.value * $2; } + | tMINUTE_UNIT + { pc->rel_minutes += $1; } + | tORDINAL tSEC_UNIT + { pc->rel_seconds += $1 * $2; } + | tUNUMBER tSEC_UNIT + { pc->rel_seconds += $1.value * $2; } + | tSDECIMAL_NUMBER tSEC_UNIT + { pc->rel_seconds += $1.tv_sec * $2; pc->rel_ns += $1.tv_nsec * $2; } + | tUDECIMAL_NUMBER tSEC_UNIT + { pc->rel_seconds += $1.tv_sec * $2; pc->rel_ns += $1.tv_nsec * $2; } + | tSEC_UNIT + { pc->rel_seconds += $1; } + | relunit_snumber + ; + +relunit_snumber: + tSNUMBER tYEAR_UNIT + { pc->rel_year += $1.value * $2; } + | tSNUMBER tMONTH_UNIT + { pc->rel_month += $1.value * $2; } + | tSNUMBER tDAY_UNIT + { pc->rel_day += $1.value * $2; } + | tSNUMBER tHOUR_UNIT + { pc->rel_hour += $1.value * $2; } + | tSNUMBER tMINUTE_UNIT + { pc->rel_minutes += $1.value * $2; } + | tSNUMBER tSEC_UNIT + { pc->rel_seconds += $1.value * $2; } + ; + +seconds: signed_seconds | unsigned_seconds; + +signed_seconds: + tSDECIMAL_NUMBER + | tSNUMBER + { $$.tv_sec = $1.value; $$.tv_nsec = 0; } + ; + +unsigned_seconds: + tUDECIMAL_NUMBER + | tUNUMBER + { $$.tv_sec = $1.value; $$.tv_nsec = 0; } + ; + +number: + tUNUMBER + { + if (pc->dates_seen && ! pc->year.digits + && ! pc->rels_seen && (pc->times_seen || 2 < $1.digits)) + pc->year = $1; + else + { + if (4 < $1.digits) + { + pc->dates_seen++; + pc->day = $1.value % 100; + pc->month = ($1.value / 100) % 100; + pc->year.value = $1.value / 10000; + pc->year.digits = $1.digits - 4; + } + else + { + pc->times_seen++; + if ($1.digits <= 2) + { + pc->hour = $1.value; + pc->minutes = 0; + } + else + { + pc->hour = $1.value / 100; + pc->minutes = $1.value % 100; + } + pc->seconds.tv_sec = 0; + pc->seconds.tv_nsec = 0; + pc->meridian = MER24; + } + } + } + ; + +o_colon_minutes: + /* empty */ + { $$ = -1; } + | ':' tUNUMBER + { $$ = $2.value; } + ; + +o_merid: + /* empty */ + { $$ = MER24; } + | tMERIDIAN + { $$ = $1; } + ; + +%% + +static table const meridian_table[] = +{ + { "AM", tMERIDIAN, MERam }, + { "A.M.", tMERIDIAN, MERam }, + { "PM", tMERIDIAN, MERpm }, + { "P.M.", tMERIDIAN, MERpm }, + { NULL, 0, 0 } +}; + +static table const dst_table[] = +{ + { "DST", tDST, 0 } +}; + +static table const month_and_day_table[] = +{ + { "JANUARY", tMONTH, 1 }, + { "FEBRUARY", tMONTH, 2 }, + { "MARCH", tMONTH, 3 }, + { "APRIL", tMONTH, 4 }, + { "MAY", tMONTH, 5 }, + { "JUNE", tMONTH, 6 }, + { "JULY", tMONTH, 7 }, + { "AUGUST", tMONTH, 8 }, + { "SEPTEMBER",tMONTH, 9 }, + { "SEPT", tMONTH, 9 }, + { "OCTOBER", tMONTH, 10 }, + { "NOVEMBER", tMONTH, 11 }, + { "DECEMBER", tMONTH, 12 }, + { "SUNDAY", tDAY, 0 }, + { "MONDAY", tDAY, 1 }, + { "TUESDAY", tDAY, 2 }, + { "TUES", tDAY, 2 }, + { "WEDNESDAY",tDAY, 3 }, + { "WEDNES", tDAY, 3 }, + { "THURSDAY", tDAY, 4 }, + { "THUR", tDAY, 4 }, + { "THURS", tDAY, 4 }, + { "FRIDAY", tDAY, 5 }, + { "SATURDAY", tDAY, 6 }, + { NULL, 0, 0 } +}; + +static table const time_units_table[] = +{ + { "YEAR", tYEAR_UNIT, 1 }, + { "MONTH", tMONTH_UNIT, 1 }, + { "FORTNIGHT",tDAY_UNIT, 14 }, + { "WEEK", tDAY_UNIT, 7 }, + { "DAY", tDAY_UNIT, 1 }, + { "HOUR", tHOUR_UNIT, 1 }, + { "MINUTE", tMINUTE_UNIT, 1 }, + { "MIN", tMINUTE_UNIT, 1 }, + { "SECOND", tSEC_UNIT, 1 }, + { "SEC", tSEC_UNIT, 1 }, + { NULL, 0, 0 } +}; + +/* Assorted relative-time words. */ +static table const relative_time_table[] = +{ + { "TOMORROW", tDAY_UNIT, 1 }, + { "YESTERDAY",tDAY_UNIT, -1 }, + { "TODAY", tDAY_UNIT, 0 }, + { "NOW", tDAY_UNIT, 0 }, + { "LAST", tORDINAL, -1 }, + { "THIS", tORDINAL, 0 }, + { "NEXT", tORDINAL, 1 }, + { "FIRST", tORDINAL, 1 }, +/*{ "SECOND", tORDINAL, 2 }, */ + { "THIRD", tORDINAL, 3 }, + { "FOURTH", tORDINAL, 4 }, + { "FIFTH", tORDINAL, 5 }, + { "SIXTH", tORDINAL, 6 }, + { "SEVENTH", tORDINAL, 7 }, + { "EIGHTH", tORDINAL, 8 }, + { "NINTH", tORDINAL, 9 }, + { "TENTH", tORDINAL, 10 }, + { "ELEVENTH", tORDINAL, 11 }, + { "TWELFTH", tORDINAL, 12 }, + { "AGO", tAGO, 1 }, + { NULL, 0, 0 } +}; + +/* The universal time zone table. These labels can be used even for + time stamps that would not otherwise be valid, e.g., GMT time + stamps in London during summer. */ +static table const universal_time_zone_table[] = +{ + { "GMT", tZONE, HOUR ( 0) }, /* Greenwich Mean */ + { "UT", tZONE, HOUR ( 0) }, /* Universal (Coordinated) */ + { "UTC", tZONE, HOUR ( 0) }, + { NULL, 0, 0 } +}; + +/* The time zone table. This table is necessarily incomplete, as time + zone abbreviations are ambiguous; e.g. Australians interpret "EST" + as Eastern time in Australia, not as US Eastern Standard Time. + You cannot rely on getdate to handle arbitrary time zone + abbreviations; use numeric abbreviations like `-0500' instead. */ +static table const time_zone_table[] = +{ + { "WET", tZONE, HOUR ( 0) }, /* Western European */ + { "WEST", tDAYZONE, HOUR ( 0) }, /* Western European Summer */ + { "BST", tDAYZONE, HOUR ( 0) }, /* British Summer */ + { "ART", tZONE, -HOUR ( 3) }, /* Argentina */ + { "BRT", tZONE, -HOUR ( 3) }, /* Brazil */ + { "BRST", tDAYZONE, -HOUR ( 3) }, /* Brazil Summer */ + { "NST", tZONE, -(HOUR ( 3) + 30) }, /* Newfoundland Standard */ + { "NDT", tDAYZONE,-(HOUR ( 3) + 30) }, /* Newfoundland Daylight */ + { "AST", tZONE, -HOUR ( 4) }, /* Atlantic Standard */ + { "ADT", tDAYZONE, -HOUR ( 4) }, /* Atlantic Daylight */ + { "CLT", tZONE, -HOUR ( 4) }, /* Chile */ + { "CLST", tDAYZONE, -HOUR ( 4) }, /* Chile Summer */ + { "EST", tZONE, -HOUR ( 5) }, /* Eastern Standard */ + { "EDT", tDAYZONE, -HOUR ( 5) }, /* Eastern Daylight */ + { "CST", tZONE, -HOUR ( 6) }, /* Central Standard */ + { "CDT", tDAYZONE, -HOUR ( 6) }, /* Central Daylight */ + { "MST", tZONE, -HOUR ( 7) }, /* Mountain Standard */ + { "MDT", tDAYZONE, -HOUR ( 7) }, /* Mountain Daylight */ + { "PST", tZONE, -HOUR ( 8) }, /* Pacific Standard */ + { "PDT", tDAYZONE, -HOUR ( 8) }, /* Pacific Daylight */ + { "AKST", tZONE, -HOUR ( 9) }, /* Alaska Standard */ + { "AKDT", tDAYZONE, -HOUR ( 9) }, /* Alaska Daylight */ + { "HST", tZONE, -HOUR (10) }, /* Hawaii Standard */ + { "HAST", tZONE, -HOUR (10) }, /* Hawaii-Aleutian Standard */ + { "HADT", tDAYZONE, -HOUR (10) }, /* Hawaii-Aleutian Daylight */ + { "SST", tZONE, -HOUR (12) }, /* Samoa Standard */ + { "WAT", tZONE, HOUR ( 1) }, /* West Africa */ + { "CET", tZONE, HOUR ( 1) }, /* Central European */ + { "CEST", tDAYZONE, HOUR ( 1) }, /* Central European Summer */ + { "MET", tZONE, HOUR ( 1) }, /* Middle European */ + { "MEZ", tZONE, HOUR ( 1) }, /* Middle European */ + { "MEST", tDAYZONE, HOUR ( 1) }, /* Middle European Summer */ + { "MESZ", tDAYZONE, HOUR ( 1) }, /* Middle European Summer */ + { "EET", tZONE, HOUR ( 2) }, /* Eastern European */ + { "EEST", tDAYZONE, HOUR ( 2) }, /* Eastern European Summer */ + { "CAT", tZONE, HOUR ( 2) }, /* Central Africa */ + { "SAST", tZONE, HOUR ( 2) }, /* South Africa Standard */ + { "EAT", tZONE, HOUR ( 3) }, /* East Africa */ + { "MSK", tZONE, HOUR ( 3) }, /* Moscow */ + { "MSD", tDAYZONE, HOUR ( 3) }, /* Moscow Daylight */ + { "IST", tZONE, (HOUR ( 5) + 30) }, /* India Standard */ + { "SGT", tZONE, HOUR ( 8) }, /* Singapore */ + { "KST", tZONE, HOUR ( 9) }, /* Korea Standard */ + { "JST", tZONE, HOUR ( 9) }, /* Japan Standard */ + { "GST", tZONE, HOUR (10) }, /* Guam Standard */ + { "NZST", tZONE, HOUR (12) }, /* New Zealand Standard */ + { "NZDT", tDAYZONE, HOUR (12) }, /* New Zealand Daylight */ + { NULL, 0, 0 } +}; + +/* Military time zone table. */ +static table const military_table[] = +{ + { "A", tZONE, -HOUR ( 1) }, + { "B", tZONE, -HOUR ( 2) }, + { "C", tZONE, -HOUR ( 3) }, + { "D", tZONE, -HOUR ( 4) }, + { "E", tZONE, -HOUR ( 5) }, + { "F", tZONE, -HOUR ( 6) }, + { "G", tZONE, -HOUR ( 7) }, + { "H", tZONE, -HOUR ( 8) }, + { "I", tZONE, -HOUR ( 9) }, + { "K", tZONE, -HOUR (10) }, + { "L", tZONE, -HOUR (11) }, + { "M", tZONE, -HOUR (12) }, + { "N", tZONE, HOUR ( 1) }, + { "O", tZONE, HOUR ( 2) }, + { "P", tZONE, HOUR ( 3) }, + { "Q", tZONE, HOUR ( 4) }, + { "R", tZONE, HOUR ( 5) }, + { "S", tZONE, HOUR ( 6) }, + { "T", tZONE, HOUR ( 7) }, + { "U", tZONE, HOUR ( 8) }, + { "V", tZONE, HOUR ( 9) }, + { "W", tZONE, HOUR (10) }, + { "X", tZONE, HOUR (11) }, + { "Y", tZONE, HOUR (12) }, + { "Z", tZONE, HOUR ( 0) }, + { NULL, 0, 0 } +}; + + + +/* Convert a time zone expressed as HH:MM into an integer count of + minutes. If MM is negative, then S is of the form HHMM and needs + to be picked apart; otherwise, S is of the form HH. */ + +static long int +time_zone_hhmm (textint s, long int mm) +{ + if (mm < 0) + return (s.value / 100) * 60 + s.value % 100; + else + return s.value * 60 + (s.negative ? -mm : mm); +} + +static int +to_hour (long int hours, int meridian) +{ + switch (meridian) + { + default: /* Pacify GCC. */ + case MER24: + return 0 <= hours && hours < 24 ? hours : -1; + case MERam: + return 0 < hours && hours < 12 ? hours : hours == 12 ? 0 : -1; + case MERpm: + return 0 < hours && hours < 12 ? hours + 12 : hours == 12 ? 12 : -1; + } +} + +static long int +to_year (textint textyear) +{ + long int year = textyear.value; + + if (year < 0) + year = -year; + + /* XPG4 suggests that years 00-68 map to 2000-2068, and + years 69-99 map to 1969-1999. */ + else if (textyear.digits == 2) + year += year < 69 ? 2000 : 1900; + + return year; +} + +static table const * +lookup_zone (parser_control const *pc, char const *name) +{ + table const *tp; + + for (tp = universal_time_zone_table; tp->name; tp++) + if (strcmp (name, tp->name) == 0) + return tp; + + /* Try local zone abbreviations before those in time_zone_table, as + the local ones are more likely to be right. */ + for (tp = pc->local_time_zone_table; tp->name; tp++) + if (strcmp (name, tp->name) == 0) + return tp; + + for (tp = time_zone_table; tp->name; tp++) + if (strcmp (name, tp->name) == 0) + return tp; + + return NULL; +} + +#if ! HAVE_TM_GMTOFF +/* Yield the difference between *A and *B, + measured in seconds, ignoring leap seconds. + The body of this function is taken directly from the GNU C Library; + see src/strftime.c. */ +static long int +tm_diff (struct tm const *a, struct tm const *b) +{ + /* Compute intervening leap days correctly even if year is negative. + Take care to avoid int overflow in leap day calculations. */ + int a4 = SHR (a->tm_year, 2) + SHR (TM_YEAR_BASE, 2) - ! (a->tm_year & 3); + int b4 = SHR (b->tm_year, 2) + SHR (TM_YEAR_BASE, 2) - ! (b->tm_year & 3); + int a100 = a4 / 25 - (a4 % 25 < 0); + int b100 = b4 / 25 - (b4 % 25 < 0); + int a400 = SHR (a100, 2); + int b400 = SHR (b100, 2); + int intervening_leap_days = (a4 - b4) - (a100 - b100) + (a400 - b400); + long int ayear = a->tm_year; + long int years = ayear - b->tm_year; + long int days = (365 * years + intervening_leap_days + + (a->tm_yday - b->tm_yday)); + return (60 * (60 * (24 * days + (a->tm_hour - b->tm_hour)) + + (a->tm_min - b->tm_min)) + + (a->tm_sec - b->tm_sec)); +} +#endif /* ! HAVE_TM_GMTOFF */ + +static table const * +lookup_word (parser_control const *pc, char *word) +{ + char *p; + char *q; + size_t wordlen; + table const *tp; + bool period_found; + bool abbrev; + + /* Make it uppercase. */ + for (p = word; *p; p++) + { + unsigned char ch = *p; + if (ISLOWER (ch)) + *p = toupper (ch); + } + + for (tp = meridian_table; tp->name; tp++) + if (strcmp (word, tp->name) == 0) + return tp; + + /* See if we have an abbreviation for a month. */ + wordlen = strlen (word); + abbrev = wordlen == 3 || (wordlen == 4 && word[3] == '.'); + + for (tp = month_and_day_table; tp->name; tp++) + if ((abbrev ? strncmp (word, tp->name, 3) : strcmp (word, tp->name)) == 0) + return tp; + + if ((tp = lookup_zone (pc, word))) + return tp; + + if (strcmp (word, dst_table[0].name) == 0) + return dst_table; + + for (tp = time_units_table; tp->name; tp++) + if (strcmp (word, tp->name) == 0) + return tp; + + /* Strip off any plural and try the units table again. */ + if (word[wordlen - 1] == 'S') + { + word[wordlen - 1] = '\0'; + for (tp = time_units_table; tp->name; tp++) + if (strcmp (word, tp->name) == 0) + return tp; + word[wordlen - 1] = 'S'; /* For "this" in relative_time_table. */ + } + + for (tp = relative_time_table; tp->name; tp++) + if (strcmp (word, tp->name) == 0) + return tp; + + /* Military time zones. */ + if (wordlen == 1) + for (tp = military_table; tp->name; tp++) + if (word[0] == tp->name[0]) + return tp; + + /* Drop out any periods and try the time zone table again. */ + for (period_found = false, p = q = word; (*p = *q); q++) + if (*q == '.') + period_found = true; + else + p++; + if (period_found && (tp = lookup_zone (pc, word))) + return tp; + + return NULL; +} + +static int +yylex (YYSTYPE *lvalp, parser_control *pc) +{ + unsigned char c; + size_t count; + + for (;;) + { + while (c = *pc->input, ISSPACE (c)) + pc->input++; + + if (ISDIGIT (c) || c == '-' || c == '+') + { + char const *p; + int sign; + unsigned long int value; + if (c == '-' || c == '+') + { + sign = c == '-' ? -1 : 1; + while (c = *++pc->input, ISSPACE (c)) + continue; + if (! ISDIGIT (c)) + /* skip the '-' sign */ + continue; + } + else + sign = 0; + p = pc->input; + for (value = 0; ; value *= 10) + { + unsigned long int value1 = value + (c - '0'); + if (value1 < value) + return '?'; + value = value1; + c = *++p; + if (! ISDIGIT (c)) + break; + if (ULONG_MAX / 10 < value) + return '?'; + } + if ((c == '.' || c == ',') && ISDIGIT (p[1])) + { + time_t s; + int ns; + int digits; + unsigned long int value1; + + /* Check for overflow when converting value to time_t. */ + if (sign < 0) + { + s = - value; + if (0 < s) + return '?'; + value1 = -s; + } + else + { + s = value; + if (s < 0) + return '?'; + value1 = s; + } + if (value != value1) + return '?'; + + /* Accumulate fraction, to ns precision. */ + p++; + ns = *p++ - '0'; + for (digits = 2; digits <= LOG10_BILLION; digits++) + { + ns *= 10; + if (ISDIGIT (*p)) + ns += *p++ - '0'; + } + + /* Skip excess digits, truncating toward -Infinity. */ + if (sign < 0) + for (; ISDIGIT (*p); p++) + if (*p != '0') + { + ns++; + break; + } + while (ISDIGIT (*p)) + p++; + + /* Adjust to the timespec convention, which is that + tv_nsec is always a positive offset even if tv_sec is + negative. */ + if (sign < 0 && ns) + { + s--; + if (! (s < 0)) + return '?'; + ns = BILLION - ns; + } + + lvalp->timespec.tv_sec = s; + lvalp->timespec.tv_nsec = ns; + pc->input = p; + return sign ? tSDECIMAL_NUMBER : tUDECIMAL_NUMBER; + } + else + { + lvalp->textintval.negative = sign < 0; + if (sign < 0) + { + lvalp->textintval.value = - value; + if (0 < lvalp->textintval.value) + return '?'; + } + else + { + lvalp->textintval.value = value; + if (lvalp->textintval.value < 0) + return '?'; + } + lvalp->textintval.digits = p - pc->input; + pc->input = p; + return sign ? tSNUMBER : tUNUMBER; + } + } + + if (ISALPHA (c)) + { + char buff[20]; + char *p = buff; + table const *tp; + + do + { + if (p < buff + sizeof buff - 1) + *p++ = c; + c = *++pc->input; + } + while (ISALPHA (c) || c == '.'); + + *p = '\0'; + tp = lookup_word (pc, buff); + if (! tp) + return '?'; + lvalp->intval = tp->value; + return tp->type; + } + + if (c != '(') + return *pc->input++; + count = 0; + do + { + c = *pc->input++; + if (c == '\0') + return c; + if (c == '(') + count++; + else if (c == ')') + count--; + } + while (count != 0); + } +} + +/* Do nothing if the parser reports an error. */ +static int +yyerror (parser_control *pc ATTRIBUTE_UNUSED, char *s ATTRIBUTE_UNUSED) +{ + return 0; +} + +/* If *TM0 is the old and *TM1 is the new value of a struct tm after + passing it to mktime, return true if it's OK that mktime returned T. + It's not OK if *TM0 has out-of-range members. */ + +static bool +mktime_ok (struct tm const *tm0, struct tm const *tm1, time_t t) +{ + if (t == (time_t) -1) + { + /* Guard against falsely reporting an error when parsing a time + stamp that happens to equal (time_t) -1, on a host that + supports such a time stamp. */ + tm1 = localtime (&t); + if (!tm1) + return false; + } + + return ! ((tm0->tm_sec ^ tm1->tm_sec) + | (tm0->tm_min ^ tm1->tm_min) + | (tm0->tm_hour ^ tm1->tm_hour) + | (tm0->tm_mday ^ tm1->tm_mday) + | (tm0->tm_mon ^ tm1->tm_mon) + | (tm0->tm_year ^ tm1->tm_year)); +} + +/* A reasonable upper bound for the size of ordinary TZ strings. + Use heap allocation if TZ's length exceeds this. */ +enum { TZBUFSIZE = 100 }; + +/* Return a copy of TZ, stored in TZBUF if it fits, and heap-allocated + otherwise. */ +static char * +get_tz (char tzbuf[TZBUFSIZE]) +{ + char *tz = getenv ("TZ"); + if (tz) + { + size_t tzsize = strlen (tz) + 1; + tz = (tzsize <= TZBUFSIZE + ? memcpy (tzbuf, tz, tzsize) + : xmemdup (tz, tzsize)); + } + return tz; +} + +/* Parse a date/time string, storing the resulting time value into *RESULT. + The string itself is pointed to by P. Return true if successful. + P can be an incomplete or relative time specification; if so, use + *NOW as the basis for the returned time. */ +bool +get_date (struct timespec *result, char const *p, struct timespec const *now) +{ + time_t Start; + long int Start_ns; + struct tm const *tmp; + struct tm tm; + struct tm tm0; + parser_control pc; + struct timespec gettime_buffer; + unsigned char c; + bool tz_was_altered = false; + char *tz0 = NULL; + char tz0buf[TZBUFSIZE]; + bool ok = true; + + if (! now) + { + gettime (&gettime_buffer); + now = &gettime_buffer; + } + + Start = now->tv_sec; + Start_ns = now->tv_nsec; + + tmp = localtime (&now->tv_sec); + if (! tmp) + return false; + + while (c = *p, ISSPACE (c)) + p++; + + if (strncmp (p, "TZ=\"", 4) == 0) + { + char const *tzbase = p + 4; + size_t tzsize = 1; + char const *s; + + for (s = tzbase; *s; s++, tzsize++) + if (*s == '\\') + { + s++; + if (! (*s == '\\' || *s == '"')) + break; + } + else if (*s == '"') + { + char *z; + char *tz1; + char tz1buf[TZBUFSIZE]; + bool large_tz = TZBUFSIZE < tzsize; + bool setenv_ok; + tz0 = get_tz (tz0buf); + z = tz1 = large_tz ? xmalloc (tzsize) : tz1buf; + for (s = tzbase; *s != '"'; s++) + *z++ = *(s += *s == '\\'); + *z = '\0'; + setenv_ok = setenv ("TZ", tz1, 1) == 0; + if (large_tz) + free (tz1); + if (!setenv_ok) + goto fail; + tz_was_altered = true; + p = s + 1; + } + } + + pc.input = p; + pc.year.value = tmp->tm_year; + pc.year.value += TM_YEAR_BASE; + pc.year.digits = 0; + pc.month = tmp->tm_mon + 1; + pc.day = tmp->tm_mday; + pc.hour = tmp->tm_hour; + pc.minutes = tmp->tm_min; + pc.seconds.tv_sec = tmp->tm_sec; + pc.seconds.tv_nsec = Start_ns; + tm.tm_isdst = tmp->tm_isdst; + + pc.meridian = MER24; + pc.rel_ns = 0; + pc.rel_seconds = 0; + pc.rel_minutes = 0; + pc.rel_hour = 0; + pc.rel_day = 0; + pc.rel_month = 0; + pc.rel_year = 0; + pc.timespec_seen = false; + pc.rels_seen = false; + pc.dates_seen = 0; + pc.days_seen = 0; + pc.times_seen = 0; + pc.local_zones_seen = 0; + pc.dsts_seen = 0; + pc.zones_seen = 0; + +#if HAVE_STRUCT_TM_TM_ZONE + pc.local_time_zone_table[0].name = tmp->tm_zone; + pc.local_time_zone_table[0].type = tLOCAL_ZONE; + pc.local_time_zone_table[0].value = tmp->tm_isdst; + pc.local_time_zone_table[1].name = NULL; + + /* Probe the names used in the next three calendar quarters, looking + for a tm_isdst different from the one we already have. */ + { + int quarter; + for (quarter = 1; quarter <= 3; quarter++) + { + time_t probe = Start + quarter * (90 * 24 * 60 * 60); + struct tm const *probe_tm = localtime (&probe); + if (probe_tm && probe_tm->tm_zone + && probe_tm->tm_isdst != pc.local_time_zone_table[0].value) + { + { + pc.local_time_zone_table[1].name = probe_tm->tm_zone; + pc.local_time_zone_table[1].type = tLOCAL_ZONE; + pc.local_time_zone_table[1].value = probe_tm->tm_isdst; + pc.local_time_zone_table[2].name = NULL; + } + break; + } + } + } +#else +#if HAVE_TZNAME + { +# ifndef tzname + extern char *tzname[]; +# endif + int i; + for (i = 0; i < 2; i++) + { + pc.local_time_zone_table[i].name = tzname[i]; + pc.local_time_zone_table[i].type = tLOCAL_ZONE; + pc.local_time_zone_table[i].value = i; + } + pc.local_time_zone_table[i].name = NULL; + } +#else + pc.local_time_zone_table[0].name = NULL; +#endif +#endif + + if (pc.local_time_zone_table[0].name && pc.local_time_zone_table[1].name + && ! strcmp (pc.local_time_zone_table[0].name, + pc.local_time_zone_table[1].name)) + { + /* This locale uses the same abbrevation for standard and + daylight times. So if we see that abbreviation, we don't + know whether it's daylight time. */ + pc.local_time_zone_table[0].value = -1; + pc.local_time_zone_table[1].name = NULL; + } + + if (yyparse (&pc) != 0) + goto fail; + + if (pc.timespec_seen) + *result = pc.seconds; + else + { + if (1 < (pc.times_seen | pc.dates_seen | pc.days_seen | pc.dsts_seen + | (pc.local_zones_seen + pc.zones_seen))) + goto fail; + + tm.tm_year = to_year (pc.year) - TM_YEAR_BASE; + tm.tm_mon = pc.month - 1; + tm.tm_mday = pc.day; + if (pc.times_seen || (pc.rels_seen && ! pc.dates_seen && ! pc.days_seen)) + { + tm.tm_hour = to_hour (pc.hour, pc.meridian); + if (tm.tm_hour < 0) + goto fail; + tm.tm_min = pc.minutes; + tm.tm_sec = pc.seconds.tv_sec; + } + else + { + tm.tm_hour = tm.tm_min = tm.tm_sec = 0; + pc.seconds.tv_nsec = 0; + } + + /* Let mktime deduce tm_isdst if we have an absolute time stamp. */ + if (!pc.rels_seen) + tm.tm_isdst = -1; + + /* But if the input explicitly specifies local time with or without + DST, give mktime that information. */ + if (pc.local_zones_seen) + tm.tm_isdst = pc.local_isdst; + + tm0 = tm; + + Start = mktime (&tm); + + if (! mktime_ok (&tm0, &tm, Start)) + { + if (! pc.zones_seen) + goto fail; + else + { + /* Guard against falsely reporting errors near the time_t + boundaries when parsing times in other time zones. For + example, suppose the input string "1969-12-31 23:00:00 -0100", + the current time zone is 8 hours ahead of UTC, and the min + time_t value is 1970-01-01 00:00:00 UTC. Then the min + localtime value is 1970-01-01 08:00:00, and mktime will + therefore fail on 1969-12-31 23:00:00. To work around the + problem, set the time zone to 1 hour behind UTC temporarily + by setting TZ="XXX1:00" and try mktime again. */ + + long int time_zone = pc.time_zone; + long int abs_time_zone = time_zone < 0 ? - time_zone : time_zone; + long int abs_time_zone_hour = abs_time_zone / 60; + int abs_time_zone_min = abs_time_zone % 60; + char tz1buf[sizeof "XXX+0:00" + + sizeof pc.time_zone * CHAR_BIT / 3]; + if (!tz_was_altered) + tz0 = get_tz (tz0buf); + sprintf (tz1buf, "XXX%s%ld:%02d", "-" + (time_zone < 0), + abs_time_zone_hour, abs_time_zone_min); + if (setenv ("TZ", tz1buf, 1) != 0) + goto fail; + tz_was_altered = true; + tm = tm0; + Start = mktime (&tm); + if (! mktime_ok (&tm0, &tm, Start)) + goto fail; + } + } + + if (pc.days_seen && ! pc.dates_seen) + { + tm.tm_mday += ((pc.day_number - tm.tm_wday + 7) % 7 + + 7 * (pc.day_ordinal - (0 < pc.day_ordinal))); + tm.tm_isdst = -1; + Start = mktime (&tm); + if (Start == (time_t) -1) + goto fail; + } + + if (pc.zones_seen) + { + long int delta = pc.time_zone * 60; + time_t t1; +#ifdef HAVE_TM_GMTOFF + delta -= tm.tm_gmtoff; +#else + time_t t = Start; + struct tm const *gmt = gmtime (&t); + if (! gmt) + goto fail; + delta -= tm_diff (&tm, gmt); +#endif + t1 = Start - delta; + if ((Start < t1) != (delta < 0)) + goto fail; /* time_t overflow */ + Start = t1; + } + + /* Add relative date. */ + if (pc.rel_year | pc.rel_month | pc.rel_day) + { + int year = tm.tm_year + pc.rel_year; + int month = tm.tm_mon + pc.rel_month; + int day = tm.tm_mday + pc.rel_day; + if (((year < tm.tm_year) ^ (pc.rel_year < 0)) + | ((month < tm.tm_mon) ^ (pc.rel_month < 0)) + | ((day < tm.tm_mday) ^ (pc.rel_day < 0))) + goto fail; + tm.tm_year = year; + tm.tm_mon = month; + tm.tm_mday = day; + Start = mktime (&tm); + if (Start == (time_t) -1) + goto fail; + } + + /* Add relative hours, minutes, and seconds. On hosts that support + leap seconds, ignore the possibility of leap seconds; e.g., + "+ 10 minutes" adds 600 seconds, even if one of them is a + leap second. Typically this is not what the user wants, but it's + too hard to do it the other way, because the time zone indicator + must be applied before relative times, and if mktime is applied + again the time zone will be lost. */ + { + long int sum_ns = pc.seconds.tv_nsec + pc.rel_ns; + long int normalized_ns = (sum_ns % BILLION + BILLION) % BILLION; + time_t t0 = Start; + long int d1 = 60 * 60 * pc.rel_hour; + time_t t1 = t0 + d1; + long int d2 = 60 * pc.rel_minutes; + time_t t2 = t1 + d2; + long int d3 = pc.rel_seconds; + time_t t3 = t2 + d3; + long int d4 = (sum_ns - normalized_ns) / BILLION; + time_t t4 = t3 + d4; + + if ((d1 / (60 * 60) ^ pc.rel_hour) + | (d2 / 60 ^ pc.rel_minutes) + | ((t1 < t0) ^ (d1 < 0)) + | ((t2 < t1) ^ (d2 < 0)) + | ((t3 < t2) ^ (d3 < 0)) + | ((t4 < t3) ^ (d4 < 0))) + goto fail; + + result->tv_sec = t4; + result->tv_nsec = normalized_ns; + } + } + + goto done; + + fail: + ok = false; + done: + if (tz_was_altered) + ok &= (tz0 ? setenv ("TZ", tz0, 1) : unsetenv ("TZ")) == 0; + if (tz0 != tz0buf) + free (tz0); + return ok; +} + +#if TEST + +int +main (int ac, char **av) +{ + char buff[BUFSIZ]; + + printf ("Enter date, or blank line to exit.\n\t> "); + fflush (stdout); + + buff[BUFSIZ - 1] = '\0'; + while (fgets (buff, BUFSIZ - 1, stdin) && buff[0]) + { + struct timespec d; + struct tm const *tm; + if (! get_date (&d, buff, NULL)) + printf ("Bad format - couldn't convert.\n"); + else if (! (tm = localtime (&d.tv_sec))) + { + long int sec = d.tv_sec; + printf ("localtime (%ld) failed\n", sec); + } + else + { + int ns = d.tv_nsec; + printf ("%04ld-%02d-%02d %02d:%02d:%02d.%09d\n", + tm->tm_year + 1900L, tm->tm_mon + 1, tm->tm_mday, + tm->tm_hour, tm->tm_min, tm->tm_sec, ns); + } + printf ("\t> "); + fflush (stdout); + } + return 0; +} +#endif /* TEST */ diff --git a/lib/getdelim.c b/lib/getdelim.c new file mode 100644 index 0000000..8498b75 --- /dev/null +++ b/lib/getdelim.c @@ -0,0 +1,119 @@ +/* getdelim.c --- Implementation of replacement getdelim function. + Copyright (C) 1994, 1996, 1997, 1998, 2001, 2003, 2005 Free + Software Foundation, Inc. + + This program is free software; you can redistribute it and/or + modify it under the terms of the GNU General Public License as + published by the Free Software Foundation; either version 2, or (at + your option) any later version. + + This program is distributed in the hope that it will be useful, but + WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + 02110-1301, USA. */ + +/* Ported from glibc by Simon Josefsson. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include <stdlib.h> +#include <errno.h> + +#include "getdelim.h" + +#if !HAVE_FLOCKFILE +# undef flockfile +# define flockfile(x) ((void) 0) +#endif +#if !HAVE_FUNLOCKFILE +# undef funlockfile +# define funlockfile(x) ((void) 0) +#endif + +/* Read up to (and including) a DELIMITER from FP into *LINEPTR (and + NUL-terminate it). *LINEPTR is a pointer returned from malloc (or + NULL), pointing to *N characters of space. It is realloc'ed as + necessary. Returns the number of characters read (not including + the null terminator), or -1 on error or EOF. */ + +ssize_t +getdelim (char **lineptr, size_t *n, int delimiter, FILE *fp) +{ + int result = 0; + ssize_t cur_len = 0; + ssize_t len; + + if (lineptr == NULL || n == NULL || fp == NULL) + { + errno = EINVAL; + return -1; + } + + flockfile (fp); + + if (*lineptr == NULL || *n == 0) + { + *n = 120; + *lineptr = (char *) malloc (*n); + if (*lineptr == NULL) + { + result = -1; + goto unlock_return; + } + } + + for (;;) + { + char *t; + int i; + + i = getc (fp); + if (i == EOF) + { + result = -1; + break; + } + + /* Make enough space for len+1 (for final NUL) bytes. */ + if (cur_len + 1 >= *n) + { + size_t needed = 2 * (cur_len + 1) + 1; /* Be generous. */ + char *new_lineptr; + + if (needed < cur_len) + { + result = -1; + goto unlock_return; + } + + new_lineptr = (char *) realloc (*lineptr, needed); + if (new_lineptr == NULL) + { + result = -1; + goto unlock_return; + } + + *lineptr = new_lineptr; + *n = needed; + } + + (*lineptr)[cur_len] = i; + cur_len++; + + if (i == delimiter) + break; + } + (*lineptr)[cur_len] = '\0'; + result = cur_len ? cur_len : result; + + unlock_return: + funlockfile (fp); + return result; +} diff --git a/lib/getdelim.h b/lib/getdelim.h new file mode 100644 index 0000000..8bc6130 --- /dev/null +++ b/lib/getdelim.h @@ -0,0 +1,28 @@ +/* getdelim.h --- Prototype for replacement getdelim function. + Copyright (C) 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or + modify it under the terms of the GNU General Public License as + published by the Free Software Foundation; either version 2, or (at + your option) any later version. + + This program is distributed in the hope that it will be useful, but + WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + 02110-1301, USA. */ + +/* Written by Simon Josefsson. */ + +/* Get size_t, FILE, ssize_t. And getdelim, if available. */ +# include <stddef.h> +# include <stdio.h> +# include <sys/types.h> + +#if !HAVE_DECL_GETDELIM +ssize_t getdelim (char **lineptr, size_t *n, int delimiter, FILE *stream); +#endif /* !HAVE_GETDELIM */ diff --git a/lib/gethostname.c b/lib/gethostname.c new file mode 100644 index 0000000..f05ed85 --- /dev/null +++ b/lib/gethostname.c @@ -0,0 +1,53 @@ +/* gethostname emulation for SysV and POSIX.1. + Copyright (C) 1992, 2003 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* David MacKenzie <djm@gnu.ai.mit.edu> */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#ifdef HAVE_UNAME +# include <sys/utsname.h> +#endif + +/* Put up to LEN chars of the host name into NAME. + Null terminate it if the name is shorter than LEN. + Return 0 if ok, -1 if error. */ + +#include <stddef.h> + +int +gethostname (char *name, size_t len) +{ +#ifdef HAVE_UNAME + struct utsname uts; + + if (uname (&uts) == -1) + return -1; + if (len > sizeof (uts.nodename)) + { + /* More space than we need is available. */ + name[sizeof (uts.nodename)] = '\0'; + len = sizeof (uts.nodename); + } + strncpy (name, uts.nodename, len); +#else + strcpy (name, ""); /* Hardcode your system name if you want. */ +#endif + return 0; +} diff --git a/lib/getline.c b/lib/getline.c new file mode 100644 index 0000000..ab054fb --- /dev/null +++ b/lib/getline.c @@ -0,0 +1,32 @@ +/* getline.c --- Implementation of replacement getline function. + Copyright (C) 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or + modify it under the terms of the GNU General Public License as + published by the Free Software Foundation; either version 2, or (at + your option) any later version. + + This program is distributed in the hope that it will be useful, but + WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + 02110-1301, USA. */ + +/* Written by Simon Josefsson. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "getdelim.h" +#include "getline.h" + +ssize_t +getline (char **lineptr, size_t *n, FILE *stream) +{ + return getdelim (lineptr, n, '\n', stream); +} diff --git a/lib/getline.h b/lib/getline.h new file mode 100644 index 0000000..f0f61c4 --- /dev/null +++ b/lib/getline.h @@ -0,0 +1,28 @@ +/* getline.h --- Prototype for replacement getline function. + Copyright (C) 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or + modify it under the terms of the GNU General Public License as + published by the Free Software Foundation; either version 2, or (at + your option) any later version. + + This program is distributed in the hope that it will be useful, but + WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA + 02110-1301, USA. */ + +/* Written by Simon Josefsson. */ + +/* Get size_t, FILE, ssize_t. And getline, if available. */ +# include <stddef.h> +# include <stdio.h> +# include <sys/types.h> + +#if !HAVE_DECL_GETLINE +ssize_t getline (char **lineptr, size_t *n, FILE *stream); +#endif /* !HAVE_GETLINE */ diff --git a/lib/getlogin_r.c b/lib/getlogin_r.c new file mode 100644 index 0000000..b9b76c6 --- /dev/null +++ b/lib/getlogin_r.c @@ -0,0 +1,61 @@ +/* Provide a working getlogin_r for systems which lack it. + + Copyright (C) 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* written by Paul Eggert and Derek Price */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "getlogin_r.h" + +#include <errno.h> +#include <string.h> + +#if HAVE_UNISTD_H +# include <unistd.h> +#endif + +#if !HAVE_DECL_GETLOGIN +char *getlogin (void); +#endif + +/* See getlogin_r.h for documentation. */ +int +getlogin_r (char *name, size_t size) +{ + char *n; + size_t nlen; + + errno = 0; + n = getlogin (); + + /* A system function like getlogin_r is never supposed to set errno + to zero, so make sure errno is nonzero here. ENOENT is a + reasonable errno value if getlogin returns NULL. */ + if (!errno) + errno = ENOENT; + + if (!n) + return errno; + nlen = strlen (n); + if (size <= nlen) + return ERANGE; + memcpy (name, n, nlen + 1); + return 0; +} diff --git a/lib/getlogin_r.h b/lib/getlogin_r.h new file mode 100644 index 0000000..3ff3caf --- /dev/null +++ b/lib/getlogin_r.h @@ -0,0 +1,38 @@ +/* getlogin_r declaration + + Copyright (C) 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Paul Eggert and Derek Price. */ + +#include <stddef.h> + +#if HAVE_UNISTD_H +# include <unistd.h> +#endif + +/* Copies the user's login name to NAME. + The array pointed to by NAME has room for SIZE bytes. + + Returns 0 if successful. Upon error, an error number is returned, or -1 in + the case that the login name cannot be found but no specific error is + provided (this case is hopefully rare but is left open by the POSIX spec). + + See <http://www.opengroup.org/susv3xsh/getlogin.html>. + */ +#if !HAVE_DECL_GETLOGIN_R +int getlogin_r (char *name, size_t size); +#endif diff --git a/lib/getndelim2.c b/lib/getndelim2.c new file mode 100644 index 0000000..3c0fa3f --- /dev/null +++ b/lib/getndelim2.c @@ -0,0 +1,155 @@ +/* getndelim2 - Read a line from a stream, stopping at one of 2 delimiters, + with bounded memory allocation. + + Copyright (C) 1993, 1996, 1997, 1998, 2000, 2003, 2004 Free Software + Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Originally written by Jan Brittenson, bson@gnu.ai.mit.edu. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "getndelim2.h" + +#include <stdlib.h> +#include <stddef.h> + +#if USE_UNLOCKED_IO +# include "unlocked-io.h" +#endif + +#include <limits.h> +#if HAVE_INTTYPES_H +# include <inttypes.h> +#endif +#if HAVE_STDINT_H +# include <stdint.h> +#endif +#ifndef PTRDIFF_MAX +# define PTRDIFF_MAX ((ptrdiff_t) (SIZE_MAX / 2)) +#endif +#ifndef SIZE_MAX +# define SIZE_MAX ((size_t) -1) +#endif +#ifndef SSIZE_MAX +# define SSIZE_MAX ((ssize_t) (SIZE_MAX / 2)) +#endif + +/* The maximum value that getndelim2 can return without suffering from + overflow problems, either internally (because of pointer + subtraction overflow) or due to the API (because of ssize_t). */ +#define GETNDELIM2_MAXIMUM (PTRDIFF_MAX < SSIZE_MAX ? PTRDIFF_MAX : SSIZE_MAX) + +/* Try to add at least this many bytes when extending the buffer. + MIN_CHUNK must be no greater than GETNDELIM2_MAXIMUM. */ +#define MIN_CHUNK 64 + +ssize_t +getndelim2 (char **lineptr, size_t *linesize, size_t offset, size_t nmax, + int delim1, int delim2, FILE *stream) +{ + size_t nbytes_avail; /* Allocated but unused bytes in *LINEPTR. */ + char *read_pos; /* Where we're reading into *LINEPTR. */ + ssize_t bytes_stored = -1; + char *ptr = *lineptr; + size_t size = *linesize; + + if (!ptr) + { + size = nmax < MIN_CHUNK ? nmax : MIN_CHUNK; + ptr = malloc (size); + if (!ptr) + return -1; + } + + if (size < offset) + goto done; + + nbytes_avail = size - offset; + read_pos = ptr + offset; + + if (nbytes_avail == 0 && nmax <= size) + goto done; + + for (;;) + { + /* Here always ptr + size == read_pos + nbytes_avail. */ + + int c; + + /* We always want at least one byte left in the buffer, since we + always (unless we get an error while reading the first byte) + NUL-terminate the line buffer. */ + + if (nbytes_avail < 2 && size < nmax) + { + size_t newsize = size < MIN_CHUNK ? size + MIN_CHUNK : 2 * size; + char *newptr; + + if (! (size < newsize && newsize <= nmax)) + newsize = nmax; + + if (GETNDELIM2_MAXIMUM < newsize - offset) + { + size_t newsizemax = offset + GETNDELIM2_MAXIMUM + 1; + if (size == newsizemax) + goto done; + newsize = newsizemax; + } + + nbytes_avail = newsize - (read_pos - ptr); + newptr = realloc (ptr, newsize); + if (!newptr) + goto done; + ptr = newptr; + size = newsize; + read_pos = size - nbytes_avail + ptr; + } + + c = getc (stream); + if (c == EOF) + { + /* Return partial line, if any. */ + if (read_pos == ptr) + goto done; + else + break; + } + + if (nbytes_avail >= 2) + { + *read_pos++ = c; + nbytes_avail--; + } + + if (c == delim1 || c == delim2) + /* Return the line. */ + break; + } + + /* Done - NUL terminate and return the number of bytes read. + At this point we know that nbytes_avail >= 1. */ + *read_pos = '\0'; + + bytes_stored = read_pos - (ptr + offset); + + done: + *lineptr = ptr; + *linesize = size; + return bytes_stored; +} diff --git a/lib/getndelim2.h b/lib/getndelim2.h new file mode 100644 index 0000000..fb0a8c3 --- /dev/null +++ b/lib/getndelim2.h @@ -0,0 +1,43 @@ +/* getndelim2 - Read a line from a stream, stopping at one of 2 delimiters, + with bounded memory allocation. + + Copyright (C) 2003, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef GETNDELIM2_H +#define GETNDELIM2_H 1 + +#include <stdio.h> +#include <sys/types.h> + +#define GETNLINE_NO_LIMIT ((size_t) -1) + +/* Read into a buffer *LINEPTR returned from malloc (or NULL), + pointing to *LINESIZE bytes of space. Store the input bytes + starting at *LINEPTR + OFFSET, and null-terminate them. Reallocate + the buffer as necessary, but if NMAX is not GETNLINE_NO_LIMIT + then do not allocate more than NMAX bytes; if the line is longer + than that, read and discard the extra bytes. Stop reading after + after the first occurrence of DELIM1 or DELIM2, whichever comes + first; a delimiter equal to EOF stands for no delimiter. Read the + input bytes from STREAM. + Return the number of bytes read and stored at *LINEPTR + OFFSET (not + including the NUL terminator), or -1 on error or EOF. */ +extern ssize_t getndelim2 (char **lineptr, size_t *linesize, size_t offset, + size_t nmax, int delim1, int delim2, + FILE *stream); + +#endif /* GETNDELIM2_H */ diff --git a/lib/getnline.c b/lib/getnline.c new file mode 100644 index 0000000..c8cd38c --- /dev/null +++ b/lib/getnline.c @@ -0,0 +1,39 @@ +/* getnline - Read a line from a stream, with bounded memory allocation. + + Copyright (C) 2003, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Specification. */ +#include "getnline.h" + +#include "getndelim2.h" + +ssize_t +getndelim (char **lineptr, size_t *linesize, size_t nmax, + int delimiter, FILE *stream) +{ + return getndelim2 (lineptr, linesize, 0, nmax, delimiter, EOF, stream); +} + +ssize_t +getnline (char **lineptr, size_t *linesize, size_t nmax, FILE *stream) +{ + return getndelim (lineptr, linesize, nmax, '\n', stream); +} diff --git a/lib/getnline.h b/lib/getnline.h new file mode 100644 index 0000000..ac389bc --- /dev/null +++ b/lib/getnline.h @@ -0,0 +1,48 @@ +/* getnline - Read a line from a stream, with bounded memory allocation. + + Copyright (C) 2003, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef GETNLINE_H +#define GETNLINE_H 1 + +#include <stdio.h> +#include <sys/types.h> + +#define GETNLINE_NO_LIMIT ((size_t) -1) + +/* Read a line, up to the next newline, from STREAM, and store it in *LINEPTR. + *LINEPTR is a pointer returned from malloc (or NULL), pointing to *LINESIZE + bytes of space. It is realloc'd as necessary. Reallocation is limited to + NMAX bytes; if the line is longer than that, the extra bytes are read but + thrown away. + Return the number of bytes read and stored at *LINEPTR (not including the + NUL terminator), or -1 on error or EOF. */ +extern ssize_t getnline (char **lineptr, size_t *linesize, size_t nmax, + FILE *stream); + +/* Read a line, up to the next occurrence of DELIMITER, from STREAM, and store + it in *LINEPTR. + *LINEPTR is a pointer returned from malloc (or NULL), pointing to *LINESIZE + bytes of space. It is realloc'd as necessary. Reallocation is limited to + NMAX bytes; if the line is longer than that, the extra bytes are read but + thrown away. + Return the number of bytes read and stored at *LINEPTR (not including the + NUL terminator), or -1 on error or EOF. */ +extern ssize_t getndelim (char **lineptr, size_t *linesize, size_t nmax, + int delimiter, FILE *stream); + +#endif /* GETNLINE_H */ diff --git a/lib/getopt.c b/lib/getopt.c new file mode 100644 index 0000000..bcb81c8 --- /dev/null +++ b/lib/getopt.c @@ -0,0 +1,1241 @@ +/* Getopt for GNU. + NOTE: getopt is now part of the C library, so if you don't know what + "Keep this file name-space clean" means, talk to drepper@gnu.org + before changing it! + Copyright (C) 1987,88,89,90,91,92,93,94,95,96,98,99,2000,2001,2002,2003,2004 + Free Software Foundation, Inc. + This file is part of the GNU C Library. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* This tells Alpha OSF/1 not to define a getopt prototype in <stdio.h>. + Ditto for AIX 3.2 and <stdlib.h>. */ +#ifndef _NO_PROTO +# define _NO_PROTO +#endif + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include <stdio.h> + +/* This needs to come after some library #include + to get __GNU_LIBRARY__ defined. */ +#ifdef __GNU_LIBRARY__ +/* Don't include stdlib.h for non-GNU C libraries because some of them + contain conflicting prototypes for getopt. */ +# include <stdlib.h> +# include <unistd.h> +#endif /* GNU C library. */ + +#include <string.h> + +#ifdef VMS +# include <unixlib.h> +#endif + +#ifdef _LIBC +# include <libintl.h> +#else +# include "gettext.h" +# define _(msgid) gettext (msgid) +#endif + +#if defined _LIBC && defined USE_IN_LIBIO +# include <wchar.h> +#endif + +#ifndef attribute_hidden +# define attribute_hidden +#endif + +/* Unlike standard Unix `getopt', functions like `getopt_long' + let the user intersperse the options with the other arguments. + + As `getopt_long' works, it permutes the elements of ARGV so that, + when it is done, all the options precede everything else. Thus + all application programs are extended to handle flexible argument order. + + Using `getopt' or setting the environment variable POSIXLY_CORRECT + disables permutation. + Then the application's behavior is completely standard. + + GNU application programs can use a third alternative mode in which + they can distinguish the relative order of options and other arguments. */ + +#include "getopt.h" +#include "getopt_int.h" + +/* For communication from `getopt' to the caller. + When `getopt' finds an option that takes an argument, + the argument value is returned here. + Also, when `ordering' is RETURN_IN_ORDER, + each non-option ARGV-element is returned here. */ + +char *optarg; + +/* Index in ARGV of the next element to be scanned. + This is used for communication to and from the caller + and for communication between successive calls to `getopt'. + + On entry to `getopt', zero means this is the first call; initialize. + + When `getopt' returns -1, this is the index of the first of the + non-option elements that the caller should itself scan. + + Otherwise, `optind' communicates from one call to the next + how much of ARGV has been scanned so far. */ + +/* 1003.2 says this must be 1 before any call. */ +int optind = 1; + +/* Callers store zero here to inhibit the error message + for unrecognized options. */ + +int opterr = 1; + +/* Set to an option character which was unrecognized. + This must be initialized on some systems to avoid linking in the + system's own getopt implementation. */ + +int optopt = '?'; + +/* Keep a global copy of all internal members of getopt_data. */ + +static struct _getopt_data getopt_data; + + +#ifndef __GNU_LIBRARY__ + +/* Avoid depending on library functions or files + whose names are inconsistent. */ + +#ifndef getenv +extern char *getenv (); +#endif + +#endif /* not __GNU_LIBRARY__ */ + +#ifdef _LIBC +/* Stored original parameters. + XXX This is no good solution. We should rather copy the args so + that we can compare them later. But we must not use malloc(3). */ +extern int __libc_argc; +extern char **__libc_argv; + +/* Bash 2.0 gives us an environment variable containing flags + indicating ARGV elements that should not be considered arguments. */ + +# ifdef USE_NONOPTION_FLAGS +/* Defined in getopt_init.c */ +extern char *__getopt_nonoption_flags; +# endif + +# ifdef USE_NONOPTION_FLAGS +# define SWAP_FLAGS(ch1, ch2) \ + if (d->__nonoption_flags_len > 0) \ + { \ + char __tmp = __getopt_nonoption_flags[ch1]; \ + __getopt_nonoption_flags[ch1] = __getopt_nonoption_flags[ch2]; \ + __getopt_nonoption_flags[ch2] = __tmp; \ + } +# else +# define SWAP_FLAGS(ch1, ch2) +# endif +#else /* !_LIBC */ +# define SWAP_FLAGS(ch1, ch2) +#endif /* _LIBC */ + +/* Exchange two adjacent subsequences of ARGV. + One subsequence is elements [first_nonopt,last_nonopt) + which contains all the non-options that have been skipped so far. + The other is elements [last_nonopt,optind), which contains all + the options processed since those non-options were skipped. + + `first_nonopt' and `last_nonopt' are relocated so that they describe + the new indices of the non-options in ARGV after they are moved. */ + +static void +exchange (char **argv, struct _getopt_data *d) +{ + int bottom = d->__first_nonopt; + int middle = d->__last_nonopt; + int top = d->optind; + char *tem; + + /* Exchange the shorter segment with the far end of the longer segment. + That puts the shorter segment into the right place. + It leaves the longer segment in the right place overall, + but it consists of two parts that need to be swapped next. */ + +#if defined _LIBC && defined USE_NONOPTION_FLAGS + /* First make sure the handling of the `__getopt_nonoption_flags' + string can work normally. Our top argument must be in the range + of the string. */ + if (d->__nonoption_flags_len > 0 && top >= d->__nonoption_flags_max_len) + { + /* We must extend the array. The user plays games with us and + presents new arguments. */ + char *new_str = malloc (top + 1); + if (new_str == NULL) + d->__nonoption_flags_len = d->__nonoption_flags_max_len = 0; + else + { + memset (__mempcpy (new_str, __getopt_nonoption_flags, + d->__nonoption_flags_max_len), + '\0', top + 1 - d->__nonoption_flags_max_len); + d->__nonoption_flags_max_len = top + 1; + __getopt_nonoption_flags = new_str; + } + } +#endif + + while (top > middle && middle > bottom) + { + if (top - middle > middle - bottom) + { + /* Bottom segment is the short one. */ + int len = middle - bottom; + register int i; + + /* Swap it with the top part of the top segment. */ + for (i = 0; i < len; i++) + { + tem = argv[bottom + i]; + argv[bottom + i] = argv[top - (middle - bottom) + i]; + argv[top - (middle - bottom) + i] = tem; + SWAP_FLAGS (bottom + i, top - (middle - bottom) + i); + } + /* Exclude the moved bottom segment from further swapping. */ + top -= len; + } + else + { + /* Top segment is the short one. */ + int len = top - middle; + register int i; + + /* Swap it with the bottom part of the bottom segment. */ + for (i = 0; i < len; i++) + { + tem = argv[bottom + i]; + argv[bottom + i] = argv[middle + i]; + argv[middle + i] = tem; + SWAP_FLAGS (bottom + i, middle + i); + } + /* Exclude the moved top segment from further swapping. */ + bottom += len; + } + } + + /* Update records for the slots the non-options now occupy. */ + + d->__first_nonopt += (d->optind - d->__last_nonopt); + d->__last_nonopt = d->optind; +} + +/* Initialize the internal data when the first call is made. */ + +static const char * +_getopt_initialize (int argc, char **argv, const char *optstring, + int posixly_correct, struct _getopt_data *d) +{ + /* Start processing options with ARGV-element 1 (since ARGV-element 0 + is the program name); the sequence of previously skipped + non-option ARGV-elements is empty. */ + + d->__first_nonopt = d->__last_nonopt = d->optind; + + d->__nextchar = NULL; + + d->__posixly_correct = posixly_correct || !!getenv ("POSIXLY_CORRECT"); + + /* Determine how to handle the ordering of options and nonoptions. */ + + if (optstring[0] == '-') + { + d->__ordering = RETURN_IN_ORDER; + ++optstring; + } + else if (optstring[0] == '+') + { + d->__ordering = REQUIRE_ORDER; + ++optstring; + } + else if (d->__posixly_correct) + d->__ordering = REQUIRE_ORDER; + else + d->__ordering = PERMUTE; + +#if defined _LIBC && defined USE_NONOPTION_FLAGS + if (!d->__posixly_correct + && argc == __libc_argc && argv == __libc_argv) + { + if (d->__nonoption_flags_max_len == 0) + { + if (__getopt_nonoption_flags == NULL + || __getopt_nonoption_flags[0] == '\0') + d->__nonoption_flags_max_len = -1; + else + { + const char *orig_str = __getopt_nonoption_flags; + int len = d->__nonoption_flags_max_len = strlen (orig_str); + if (d->__nonoption_flags_max_len < argc) + d->__nonoption_flags_max_len = argc; + __getopt_nonoption_flags = + (char *) malloc (d->__nonoption_flags_max_len); + if (__getopt_nonoption_flags == NULL) + d->__nonoption_flags_max_len = -1; + else + memset (__mempcpy (__getopt_nonoption_flags, orig_str, len), + '\0', d->__nonoption_flags_max_len - len); + } + } + d->__nonoption_flags_len = d->__nonoption_flags_max_len; + } + else + d->__nonoption_flags_len = 0; +#endif + + return optstring; +} + +/* Scan elements of ARGV (whose length is ARGC) for option characters + given in OPTSTRING. + + If an element of ARGV starts with '-', and is not exactly "-" or "--", + then it is an option element. The characters of this element + (aside from the initial '-') are option characters. If `getopt' + is called repeatedly, it returns successively each of the option characters + from each of the option elements. + + If `getopt' finds another option character, it returns that character, + updating `optind' and `nextchar' so that the next call to `getopt' can + resume the scan with the following option character or ARGV-element. + + If there are no more option characters, `getopt' returns -1. + Then `optind' is the index in ARGV of the first ARGV-element + that is not an option. (The ARGV-elements have been permuted + so that those that are not options now come last.) + + OPTSTRING is a string containing the legitimate option characters. + If an option character is seen that is not listed in OPTSTRING, + return '?' after printing an error message. If you set `opterr' to + zero, the error message is suppressed but we still return '?'. + + If a char in OPTSTRING is followed by a colon, that means it wants an arg, + so the following text in the same ARGV-element, or the text of the following + ARGV-element, is returned in `optarg'. Two colons mean an option that + wants an optional arg; if there is text in the current ARGV-element, + it is returned in `optarg', otherwise `optarg' is set to zero. + + If OPTSTRING starts with `-' or `+', it requests different methods of + handling the non-option ARGV-elements. + See the comments about RETURN_IN_ORDER and REQUIRE_ORDER, above. + + Long-named options begin with `--' instead of `-'. + Their names may be abbreviated as long as the abbreviation is unique + or is an exact match for some defined option. If they have an + argument, it follows the option name in the same ARGV-element, separated + from the option name by a `=', or else the in next ARGV-element. + When `getopt' finds a long-named option, it returns 0 if that option's + `flag' field is nonzero, the value of the option's `val' field + if the `flag' field is zero. + + LONGOPTS is a vector of `struct option' terminated by an + element containing a name which is zero. + + LONGIND returns the index in LONGOPT of the long-named option found. + It is only valid when a long-named option has been found by the most + recent call. + + If LONG_ONLY is nonzero, '-' as well as '--' can introduce + long-named options. + + If POSIXLY_CORRECT is nonzero, behave as if the POSIXLY_CORRECT + environment variable were set. */ + +int +_getopt_internal_r (int argc, char **argv, const char *optstring, + const struct option *longopts, int *longind, + int long_only, int posixly_correct, struct _getopt_data *d) +{ + int print_errors = d->opterr; + if (optstring[0] == ':') + print_errors = 0; + + if (argc < 1) + return -1; + + d->optarg = NULL; + + if (d->optind == 0 || !d->__initialized) + { + if (d->optind == 0) + d->optind = 1; /* Don't scan ARGV[0], the program name. */ + optstring = _getopt_initialize (argc, argv, optstring, + posixly_correct, d); + d->__initialized = 1; + } + + /* Test whether ARGV[optind] points to a non-option argument. + Either it does not have option syntax, or there is an environment flag + from the shell indicating it is not an option. The later information + is only used when the used in the GNU libc. */ +#if defined _LIBC && defined USE_NONOPTION_FLAGS +# define NONOPTION_P (argv[d->optind][0] != '-' || argv[d->optind][1] == '\0' \ + || (d->optind < d->__nonoption_flags_len \ + && __getopt_nonoption_flags[d->optind] == '1')) +#else +# define NONOPTION_P (argv[d->optind][0] != '-' || argv[d->optind][1] == '\0') +#endif + + if (d->__nextchar == NULL || *d->__nextchar == '\0') + { + /* Advance to the next ARGV-element. */ + + /* Give FIRST_NONOPT & LAST_NONOPT rational values if OPTIND has been + moved back by the user (who may also have changed the arguments). */ + if (d->__last_nonopt > d->optind) + d->__last_nonopt = d->optind; + if (d->__first_nonopt > d->optind) + d->__first_nonopt = d->optind; + + if (d->__ordering == PERMUTE) + { + /* If we have just processed some options following some non-options, + exchange them so that the options come first. */ + + if (d->__first_nonopt != d->__last_nonopt + && d->__last_nonopt != d->optind) + exchange ((char **) argv, d); + else if (d->__last_nonopt != d->optind) + d->__first_nonopt = d->optind; + + /* Skip any additional non-options + and extend the range of non-options previously skipped. */ + + while (d->optind < argc && NONOPTION_P) + d->optind++; + d->__last_nonopt = d->optind; + } + + /* The special ARGV-element `--' means premature end of options. + Skip it like a null option, + then exchange with previous non-options as if it were an option, + then skip everything else like a non-option. */ + + if (d->optind != argc && !strcmp (argv[d->optind], "--")) + { + d->optind++; + + if (d->__first_nonopt != d->__last_nonopt + && d->__last_nonopt != d->optind) + exchange ((char **) argv, d); + else if (d->__first_nonopt == d->__last_nonopt) + d->__first_nonopt = d->optind; + d->__last_nonopt = argc; + + d->optind = argc; + } + + /* If we have done all the ARGV-elements, stop the scan + and back over any non-options that we skipped and permuted. */ + + if (d->optind == argc) + { + /* Set the next-arg-index to point at the non-options + that we previously skipped, so the caller will digest them. */ + if (d->__first_nonopt != d->__last_nonopt) + d->optind = d->__first_nonopt; + return -1; + } + + /* If we have come to a non-option and did not permute it, + either stop the scan or describe it to the caller and pass it by. */ + + if (NONOPTION_P) + { + if (d->__ordering == REQUIRE_ORDER) + return -1; + d->optarg = argv[d->optind++]; + return 1; + } + + /* We have found another option-ARGV-element. + Skip the initial punctuation. */ + + d->__nextchar = (argv[d->optind] + 1 + + (longopts != NULL && argv[d->optind][1] == '-')); + } + + /* Decode the current option-ARGV-element. */ + + /* Check whether the ARGV-element is a long option. + + If long_only and the ARGV-element has the form "-f", where f is + a valid short option, don't consider it an abbreviated form of + a long option that starts with f. Otherwise there would be no + way to give the -f short option. + + On the other hand, if there's a long option "fubar" and + the ARGV-element is "-fu", do consider that an abbreviation of + the long option, just like "--fu", and not "-f" with arg "u". + + This distinction seems to be the most useful approach. */ + + if (longopts != NULL + && (argv[d->optind][1] == '-' + || (long_only && (argv[d->optind][2] + || !strchr (optstring, argv[d->optind][1]))))) + { + char *nameend; + const struct option *p; + const struct option *pfound = NULL; + int exact = 0; + int ambig = 0; + int indfound = -1; + int option_index; + + for (nameend = d->__nextchar; *nameend && *nameend != '='; nameend++) + /* Do nothing. */ ; + + /* Test all long options for either exact match + or abbreviated matches. */ + for (p = longopts, option_index = 0; p->name; p++, option_index++) + if (!strncmp (p->name, d->__nextchar, nameend - d->__nextchar)) + { + if ((unsigned int) (nameend - d->__nextchar) + == (unsigned int) strlen (p->name)) + { + /* Exact match found. */ + pfound = p; + indfound = option_index; + exact = 1; + break; + } + else if (pfound == NULL) + { + /* First nonexact match found. */ + pfound = p; + indfound = option_index; + } + else if (long_only + || pfound->has_arg != p->has_arg + || pfound->flag != p->flag + || pfound->val != p->val) + /* Second or later nonexact match found. */ + ambig = 1; + } + + if (ambig && !exact) + { + if (print_errors) + { +#if defined _LIBC && defined USE_IN_LIBIO + char *buf; + + if (__asprintf (&buf, _("%s: option `%s' is ambiguous\n"), + argv[0], argv[d->optind]) >= 0) + { + _IO_flockfile (stderr); + + int old_flags2 = ((_IO_FILE *) stderr)->_flags2; + ((_IO_FILE *) stderr)->_flags2 |= _IO_FLAGS2_NOTCANCEL; + + if (_IO_fwide (stderr, 0) > 0) + __fwprintf (stderr, L"%s", buf); + else + fputs (buf, stderr); + + ((_IO_FILE *) stderr)->_flags2 = old_flags2; + _IO_funlockfile (stderr); + + free (buf); + } +#else + fprintf (stderr, _("%s: option `%s' is ambiguous\n"), + argv[0], argv[d->optind]); +#endif + } + d->__nextchar += strlen (d->__nextchar); + d->optind++; + d->optopt = 0; + return '?'; + } + + if (pfound != NULL) + { + option_index = indfound; + d->optind++; + if (*nameend) + { + /* Don't test has_arg with >, because some C compilers don't + allow it to be used on enums. */ + if (pfound->has_arg) + d->optarg = nameend + 1; + else + { + if (print_errors) + { +#if defined _LIBC && defined USE_IN_LIBIO + char *buf; + int n; +#endif + + if (argv[d->optind - 1][1] == '-') + { + /* --option */ +#if defined _LIBC && defined USE_IN_LIBIO + n = __asprintf (&buf, _("\ +%s: option `--%s' doesn't allow an argument\n"), + argv[0], pfound->name); +#else + fprintf (stderr, _("\ +%s: option `--%s' doesn't allow an argument\n"), + argv[0], pfound->name); +#endif + } + else + { + /* +option or -option */ +#if defined _LIBC && defined USE_IN_LIBIO + n = __asprintf (&buf, _("\ +%s: option `%c%s' doesn't allow an argument\n"), + argv[0], argv[d->optind - 1][0], + pfound->name); +#else + fprintf (stderr, _("\ +%s: option `%c%s' doesn't allow an argument\n"), + argv[0], argv[d->optind - 1][0], + pfound->name); +#endif + } + +#if defined _LIBC && defined USE_IN_LIBIO + if (n >= 0) + { + _IO_flockfile (stderr); + + int old_flags2 = ((_IO_FILE *) stderr)->_flags2; + ((_IO_FILE *) stderr)->_flags2 + |= _IO_FLAGS2_NOTCANCEL; + + if (_IO_fwide (stderr, 0) > 0) + __fwprintf (stderr, L"%s", buf); + else + fputs (buf, stderr); + + ((_IO_FILE *) stderr)->_flags2 = old_flags2; + _IO_funlockfile (stderr); + + free (buf); + } +#endif + } + + d->__nextchar += strlen (d->__nextchar); + + d->optopt = pfound->val; + return '?'; + } + } + else if (pfound->has_arg == 1) + { + if (d->optind < argc) + d->optarg = argv[d->optind++]; + else + { + if (print_errors) + { +#if defined _LIBC && defined USE_IN_LIBIO + char *buf; + + if (__asprintf (&buf, _("\ +%s: option `%s' requires an argument\n"), + argv[0], argv[d->optind - 1]) >= 0) + { + _IO_flockfile (stderr); + + int old_flags2 = ((_IO_FILE *) stderr)->_flags2; + ((_IO_FILE *) stderr)->_flags2 + |= _IO_FLAGS2_NOTCANCEL; + + if (_IO_fwide (stderr, 0) > 0) + __fwprintf (stderr, L"%s", buf); + else + fputs (buf, stderr); + + ((_IO_FILE *) stderr)->_flags2 = old_flags2; + _IO_funlockfile (stderr); + + free (buf); + } +#else + fprintf (stderr, + _("%s: option `%s' requires an argument\n"), + argv[0], argv[d->optind - 1]); +#endif + } + d->__nextchar += strlen (d->__nextchar); + d->optopt = pfound->val; + return optstring[0] == ':' ? ':' : '?'; + } + } + d->__nextchar += strlen (d->__nextchar); + if (longind != NULL) + *longind = option_index; + if (pfound->flag) + { + *(pfound->flag) = pfound->val; + return 0; + } + return pfound->val; + } + + /* Can't find it as a long option. If this is not getopt_long_only, + or the option starts with '--' or is not a valid short + option, then it's an error. + Otherwise interpret it as a short option. */ + if (!long_only || argv[d->optind][1] == '-' + || strchr (optstring, *d->__nextchar) == NULL) + { + if (print_errors) + { +#if defined _LIBC && defined USE_IN_LIBIO + char *buf; + int n; +#endif + + if (argv[d->optind][1] == '-') + { + /* --option */ +#if defined _LIBC && defined USE_IN_LIBIO + n = __asprintf (&buf, _("%s: unrecognized option `--%s'\n"), + argv[0], d->__nextchar); +#else + fprintf (stderr, _("%s: unrecognized option `--%s'\n"), + argv[0], d->__nextchar); +#endif + } + else + { + /* +option or -option */ +#if defined _LIBC && defined USE_IN_LIBIO + n = __asprintf (&buf, _("%s: unrecognized option `%c%s'\n"), + argv[0], argv[d->optind][0], d->__nextchar); +#else + fprintf (stderr, _("%s: unrecognized option `%c%s'\n"), + argv[0], argv[d->optind][0], d->__nextchar); +#endif + } + +#if defined _LIBC && defined USE_IN_LIBIO + if (n >= 0) + { + _IO_flockfile (stderr); + + int old_flags2 = ((_IO_FILE *) stderr)->_flags2; + ((_IO_FILE *) stderr)->_flags2 |= _IO_FLAGS2_NOTCANCEL; + + if (_IO_fwide (stderr, 0) > 0) + __fwprintf (stderr, L"%s", buf); + else + fputs (buf, stderr); + + ((_IO_FILE *) stderr)->_flags2 = old_flags2; + _IO_funlockfile (stderr); + + free (buf); + } +#endif + } + d->__nextchar = (char *) ""; + d->optind++; + d->optopt = 0; + return '?'; + } + } + + /* Look at and handle the next short option-character. */ + + { + char c = *d->__nextchar++; + char *temp = strchr (optstring, c); + + /* Increment `optind' when we start to process its last character. */ + if (*d->__nextchar == '\0') + ++d->optind; + + if (temp == NULL || c == ':') + { + if (print_errors) + { +#if defined _LIBC && defined USE_IN_LIBIO + char *buf; + int n; +#endif + + if (d->__posixly_correct) + { + /* 1003.2 specifies the format of this message. */ +#if defined _LIBC && defined USE_IN_LIBIO + n = __asprintf (&buf, _("%s: illegal option -- %c\n"), + argv[0], c); +#else + fprintf (stderr, _("%s: illegal option -- %c\n"), argv[0], c); +#endif + } + else + { +#if defined _LIBC && defined USE_IN_LIBIO + n = __asprintf (&buf, _("%s: invalid option -- %c\n"), + argv[0], c); +#else + fprintf (stderr, _("%s: invalid option -- %c\n"), argv[0], c); +#endif + } + +#if defined _LIBC && defined USE_IN_LIBIO + if (n >= 0) + { + _IO_flockfile (stderr); + + int old_flags2 = ((_IO_FILE *) stderr)->_flags2; + ((_IO_FILE *) stderr)->_flags2 |= _IO_FLAGS2_NOTCANCEL; + + if (_IO_fwide (stderr, 0) > 0) + __fwprintf (stderr, L"%s", buf); + else + fputs (buf, stderr); + + ((_IO_FILE *) stderr)->_flags2 = old_flags2; + _IO_funlockfile (stderr); + + free (buf); + } +#endif + } + d->optopt = c; + return '?'; + } + /* Convenience. Treat POSIX -W foo same as long option --foo */ + if (temp[0] == 'W' && temp[1] == ';') + { + char *nameend; + const struct option *p; + const struct option *pfound = NULL; + int exact = 0; + int ambig = 0; + int indfound = 0; + int option_index; + + /* This is an option that requires an argument. */ + if (*d->__nextchar != '\0') + { + d->optarg = d->__nextchar; + /* If we end this ARGV-element by taking the rest as an arg, + we must advance to the next element now. */ + d->optind++; + } + else if (d->optind == argc) + { + if (print_errors) + { + /* 1003.2 specifies the format of this message. */ +#if defined _LIBC && defined USE_IN_LIBIO + char *buf; + + if (__asprintf (&buf, + _("%s: option requires an argument -- %c\n"), + argv[0], c) >= 0) + { + _IO_flockfile (stderr); + + int old_flags2 = ((_IO_FILE *) stderr)->_flags2; + ((_IO_FILE *) stderr)->_flags2 |= _IO_FLAGS2_NOTCANCEL; + + if (_IO_fwide (stderr, 0) > 0) + __fwprintf (stderr, L"%s", buf); + else + fputs (buf, stderr); + + ((_IO_FILE *) stderr)->_flags2 = old_flags2; + _IO_funlockfile (stderr); + + free (buf); + } +#else + fprintf (stderr, _("%s: option requires an argument -- %c\n"), + argv[0], c); +#endif + } + d->optopt = c; + if (optstring[0] == ':') + c = ':'; + else + c = '?'; + return c; + } + else + /* We already incremented `d->optind' once; + increment it again when taking next ARGV-elt as argument. */ + d->optarg = argv[d->optind++]; + + /* optarg is now the argument, see if it's in the + table of longopts. */ + + for (d->__nextchar = nameend = d->optarg; *nameend && *nameend != '='; + nameend++) + /* Do nothing. */ ; + + /* Test all long options for either exact match + or abbreviated matches. */ + for (p = longopts, option_index = 0; p->name; p++, option_index++) + if (!strncmp (p->name, d->__nextchar, nameend - d->__nextchar)) + { + if ((unsigned int) (nameend - d->__nextchar) == strlen (p->name)) + { + /* Exact match found. */ + pfound = p; + indfound = option_index; + exact = 1; + break; + } + else if (pfound == NULL) + { + /* First nonexact match found. */ + pfound = p; + indfound = option_index; + } + else + /* Second or later nonexact match found. */ + ambig = 1; + } + if (ambig && !exact) + { + if (print_errors) + { +#if defined _LIBC && defined USE_IN_LIBIO + char *buf; + + if (__asprintf (&buf, _("%s: option `-W %s' is ambiguous\n"), + argv[0], argv[d->optind]) >= 0) + { + _IO_flockfile (stderr); + + int old_flags2 = ((_IO_FILE *) stderr)->_flags2; + ((_IO_FILE *) stderr)->_flags2 |= _IO_FLAGS2_NOTCANCEL; + + if (_IO_fwide (stderr, 0) > 0) + __fwprintf (stderr, L"%s", buf); + else + fputs (buf, stderr); + + ((_IO_FILE *) stderr)->_flags2 = old_flags2; + _IO_funlockfile (stderr); + + free (buf); + } +#else + fprintf (stderr, _("%s: option `-W %s' is ambiguous\n"), + argv[0], argv[d->optind]); +#endif + } + d->__nextchar += strlen (d->__nextchar); + d->optind++; + return '?'; + } + if (pfound != NULL) + { + option_index = indfound; + if (*nameend) + { + /* Don't test has_arg with >, because some C compilers don't + allow it to be used on enums. */ + if (pfound->has_arg) + d->optarg = nameend + 1; + else + { + if (print_errors) + { +#if defined _LIBC && defined USE_IN_LIBIO + char *buf; + + if (__asprintf (&buf, _("\ +%s: option `-W %s' doesn't allow an argument\n"), + argv[0], pfound->name) >= 0) + { + _IO_flockfile (stderr); + + int old_flags2 = ((_IO_FILE *) stderr)->_flags2; + ((_IO_FILE *) stderr)->_flags2 + |= _IO_FLAGS2_NOTCANCEL; + + if (_IO_fwide (stderr, 0) > 0) + __fwprintf (stderr, L"%s", buf); + else + fputs (buf, stderr); + + ((_IO_FILE *) stderr)->_flags2 = old_flags2; + _IO_funlockfile (stderr); + + free (buf); + } +#else + fprintf (stderr, _("\ +%s: option `-W %s' doesn't allow an argument\n"), + argv[0], pfound->name); +#endif + } + + d->__nextchar += strlen (d->__nextchar); + return '?'; + } + } + else if (pfound->has_arg == 1) + { + if (d->optind < argc) + d->optarg = argv[d->optind++]; + else + { + if (print_errors) + { +#if defined _LIBC && defined USE_IN_LIBIO + char *buf; + + if (__asprintf (&buf, _("\ +%s: option `%s' requires an argument\n"), + argv[0], argv[d->optind - 1]) >= 0) + { + _IO_flockfile (stderr); + + int old_flags2 = ((_IO_FILE *) stderr)->_flags2; + ((_IO_FILE *) stderr)->_flags2 + |= _IO_FLAGS2_NOTCANCEL; + + if (_IO_fwide (stderr, 0) > 0) + __fwprintf (stderr, L"%s", buf); + else + fputs (buf, stderr); + + ((_IO_FILE *) stderr)->_flags2 = old_flags2; + _IO_funlockfile (stderr); + + free (buf); + } +#else + fprintf (stderr, + _("%s: option `%s' requires an argument\n"), + argv[0], argv[d->optind - 1]); +#endif + } + d->__nextchar += strlen (d->__nextchar); + return optstring[0] == ':' ? ':' : '?'; + } + } + d->__nextchar += strlen (d->__nextchar); + if (longind != NULL) + *longind = option_index; + if (pfound->flag) + { + *(pfound->flag) = pfound->val; + return 0; + } + return pfound->val; + } + d->__nextchar = NULL; + return 'W'; /* Let the application handle it. */ + } + if (temp[1] == ':') + { + if (temp[2] == ':') + { + /* This is an option that accepts an argument optionally. */ + if (*d->__nextchar != '\0') + { + d->optarg = d->__nextchar; + d->optind++; + } + else + d->optarg = NULL; + d->__nextchar = NULL; + } + else + { + /* This is an option that requires an argument. */ + if (*d->__nextchar != '\0') + { + d->optarg = d->__nextchar; + /* If we end this ARGV-element by taking the rest as an arg, + we must advance to the next element now. */ + d->optind++; + } + else if (d->optind == argc) + { + if (print_errors) + { + /* 1003.2 specifies the format of this message. */ +#if defined _LIBC && defined USE_IN_LIBIO + char *buf; + + if (__asprintf (&buf, _("\ +%s: option requires an argument -- %c\n"), + argv[0], c) >= 0) + { + _IO_flockfile (stderr); + + int old_flags2 = ((_IO_FILE *) stderr)->_flags2; + ((_IO_FILE *) stderr)->_flags2 |= _IO_FLAGS2_NOTCANCEL; + + if (_IO_fwide (stderr, 0) > 0) + __fwprintf (stderr, L"%s", buf); + else + fputs (buf, stderr); + + ((_IO_FILE *) stderr)->_flags2 = old_flags2; + _IO_funlockfile (stderr); + + free (buf); + } +#else + fprintf (stderr, + _("%s: option requires an argument -- %c\n"), + argv[0], c); +#endif + } + d->optopt = c; + if (optstring[0] == ':') + c = ':'; + else + c = '?'; + } + else + /* We already incremented `optind' once; + increment it again when taking next ARGV-elt as argument. */ + d->optarg = argv[d->optind++]; + d->__nextchar = NULL; + } + } + return c; + } +} + +int +_getopt_internal (int argc, char **argv, const char *optstring, + const struct option *longopts, int *longind, + int long_only, int posixly_correct) +{ + int result; + + getopt_data.optind = optind; + getopt_data.opterr = opterr; + + result = _getopt_internal_r (argc, argv, optstring, longopts, longind, + long_only, posixly_correct, &getopt_data); + + optind = getopt_data.optind; + optarg = getopt_data.optarg; + optopt = getopt_data.optopt; + + return result; +} + +/* glibc gets a LSB-compliant getopt. + Standalone applications get a POSIX-compliant getopt. */ +#if _LIBC +enum { POSIXLY_CORRECT = 0 }; +#else +enum { POSIXLY_CORRECT = 1 }; +#endif + +int +getopt (int argc, char *const *argv, const char *optstring) +{ + return _getopt_internal (argc, (char **) argv, optstring, NULL, NULL, 0, + POSIXLY_CORRECT); +} + + +#ifdef TEST + +/* Compile with -DTEST to make an executable for use in testing + the above definition of `getopt'. */ + +int +main (int argc, char **argv) +{ + int c; + int digit_optind = 0; + + while (1) + { + int this_option_optind = optind ? optind : 1; + + c = getopt (argc, argv, "abc:d:0123456789"); + if (c == -1) + break; + + switch (c) + { + case '0': + case '1': + case '2': + case '3': + case '4': + case '5': + case '6': + case '7': + case '8': + case '9': + if (digit_optind != 0 && digit_optind != this_option_optind) + printf ("digits occur in two different argv-elements.\n"); + digit_optind = this_option_optind; + printf ("option %c\n", c); + break; + + case 'a': + printf ("option a\n"); + break; + + case 'b': + printf ("option b\n"); + break; + + case 'c': + printf ("option c with value `%s'\n", optarg); + break; + + case '?': + break; + + default: + printf ("?? getopt returned character code 0%o ??\n", c); + } + } + + if (optind < argc) + { + printf ("non-option ARGV-elements: "); + while (optind < argc) + printf ("%s ", argv[optind++]); + printf ("\n"); + } + + exit (0); +} + +#endif /* TEST */ diff --git a/lib/getopt1.c b/lib/getopt1.c new file mode 100644 index 0000000..25d7926 --- /dev/null +++ b/lib/getopt1.c @@ -0,0 +1,174 @@ +/* getopt_long and getopt_long_only entry points for GNU getopt. + Copyright (C) 1987,88,89,90,91,92,93,94,96,97,98,2004 + Free Software Foundation, Inc. + This file is part of the GNU C Library. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#ifdef _LIBC +# include <getopt.h> +#else +# include "getopt.h" +#endif +#include "getopt_int.h" + +#include <stdio.h> + +/* This needs to come after some library #include + to get __GNU_LIBRARY__ defined. */ +#ifdef __GNU_LIBRARY__ +#include <stdlib.h> +#endif + +#ifndef NULL +#define NULL 0 +#endif + +int +getopt_long (int argc, char *__getopt_argv_const *argv, const char *options, + const struct option *long_options, int *opt_index) +{ + return _getopt_internal (argc, (char **) argv, options, long_options, + opt_index, 0, 0); +} + +int +_getopt_long_r (int argc, char **argv, const char *options, + const struct option *long_options, int *opt_index, + struct _getopt_data *d) +{ + return _getopt_internal_r (argc, argv, options, long_options, opt_index, + 0, 0, d); +} + +/* Like getopt_long, but '-' as well as '--' can indicate a long option. + If an option that starts with '-' (not '--') doesn't match a long option, + but does match a short option, it is parsed as a short option + instead. */ + +int +getopt_long_only (int argc, char *__getopt_argv_const *argv, + const char *options, + const struct option *long_options, int *opt_index) +{ + return _getopt_internal (argc, (char **) argv, options, long_options, + opt_index, 1, 0); +} + +int +_getopt_long_only_r (int argc, char **argv, const char *options, + const struct option *long_options, int *opt_index, + struct _getopt_data *d) +{ + return _getopt_internal_r (argc, argv, options, long_options, opt_index, + 1, 0, d); +} + + +#ifdef TEST + +#include <stdio.h> + +int +main (int argc, char **argv) +{ + int c; + int digit_optind = 0; + + while (1) + { + int this_option_optind = optind ? optind : 1; + int option_index = 0; + static struct option long_options[] = + { + {"add", 1, 0, 0}, + {"append", 0, 0, 0}, + {"delete", 1, 0, 0}, + {"verbose", 0, 0, 0}, + {"create", 0, 0, 0}, + {"file", 1, 0, 0}, + {0, 0, 0, 0} + }; + + c = getopt_long (argc, argv, "abc:d:0123456789", + long_options, &option_index); + if (c == -1) + break; + + switch (c) + { + case 0: + printf ("option %s", long_options[option_index].name); + if (optarg) + printf (" with arg %s", optarg); + printf ("\n"); + break; + + case '0': + case '1': + case '2': + case '3': + case '4': + case '5': + case '6': + case '7': + case '8': + case '9': + if (digit_optind != 0 && digit_optind != this_option_optind) + printf ("digits occur in two different argv-elements.\n"); + digit_optind = this_option_optind; + printf ("option %c\n", c); + break; + + case 'a': + printf ("option a\n"); + break; + + case 'b': + printf ("option b\n"); + break; + + case 'c': + printf ("option c with value `%s'\n", optarg); + break; + + case 'd': + printf ("option d with value `%s'\n", optarg); + break; + + case '?': + break; + + default: + printf ("?? getopt returned character code 0%o ??\n", c); + } + } + + if (optind < argc) + { + printf ("non-option ARGV-elements: "); + while (optind < argc) + printf ("%s ", argv[optind++]); + printf ("\n"); + } + + exit (0); +} + +#endif /* TEST */ diff --git a/lib/getopt_.h b/lib/getopt_.h new file mode 100644 index 0000000..a4e78cb --- /dev/null +++ b/lib/getopt_.h @@ -0,0 +1,227 @@ +/* Declarations for getopt. + Copyright (C) 1989-1994,1996-1999,2001,2003,2004 + Free Software Foundation, Inc. + This file is part of the GNU C Library. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _GETOPT_H + +#ifndef __need_getopt +# define _GETOPT_H 1 +#endif + +/* Standalone applications should #define __GETOPT_PREFIX to an + identifier that prefixes the external functions and variables + defined in this header. When this happens, include the + headers that might declare getopt so that they will not cause + confusion if included after this file. Then systematically rename + identifiers so that they do not collide with the system functions + and variables. Renaming avoids problems with some compilers and + linkers. */ +#if defined __GETOPT_PREFIX && !defined __need_getopt +# include <stdlib.h> +# include <stdio.h> +# if HAVE_UNISTD_H +# include <unistd.h> +# endif +# undef __need_getopt +# undef getopt +# undef getopt_long +# undef getopt_long_only +# undef optarg +# undef opterr +# undef optind +# undef optopt +# define __GETOPT_CONCAT(x, y) x ## y +# define __GETOPT_XCONCAT(x, y) __GETOPT_CONCAT (x, y) +# define __GETOPT_ID(y) __GETOPT_XCONCAT (__GETOPT_PREFIX, y) +# define getopt __GETOPT_ID (getopt) +# define getopt_long __GETOPT_ID (getopt_long) +# define getopt_long_only __GETOPT_ID (getopt_long_only) +# define optarg __GETOPT_ID (optarg) +# define opterr __GETOPT_ID (opterr) +# define optind __GETOPT_ID (optind) +# define optopt __GETOPT_ID (optopt) +#endif + +/* Standalone applications get correct prototypes for getopt_long and + getopt_long_only; they declare "char **argv". libc uses prototypes + with "char *const *argv" that are incorrect because getopt_long and + getopt_long_only can permute argv; this is required for backward + compatibility (e.g., for LSB 2.0.1). + + This used to be `#if defined __GETOPT_PREFIX && !defined __need_getopt', + but it caused redefinition warnings if both unistd.h and getopt.h were + included, since unistd.h includes getopt.h having previously defined + __need_getopt. + + The only place where __getopt_argv_const is used is in definitions + of getopt_long and getopt_long_only below, but these are visible + only if __need_getopt is not defined, so it is quite safe to rewrite + the conditional as follows: +*/ +#if !defined __need_getopt +# if defined __GETOPT_PREFIX +# define __getopt_argv_const /* empty */ +# else +# define __getopt_argv_const const +# endif +#endif + +/* If __GNU_LIBRARY__ is not already defined, either we are being used + standalone, or this is the first header included in the source file. + If we are being used with glibc, we need to include <features.h>, but + that does not exist if we are standalone. So: if __GNU_LIBRARY__ is + not defined, include <ctype.h>, which will pull in <features.h> for us + if it's from glibc. (Why ctype.h? It's guaranteed to exist and it + doesn't flood the namespace with stuff the way some other headers do.) */ +#if !defined __GNU_LIBRARY__ +# include <ctype.h> +#endif + +#ifndef __THROW +# ifndef __GNUC_PREREQ +# define __GNUC_PREREQ(maj, min) (0) +# endif +# if defined __cplusplus && __GNUC_PREREQ (2,8) +# define __THROW throw () +# else +# define __THROW +# endif +#endif + +#ifdef __cplusplus +extern "C" { +#endif + +/* For communication from `getopt' to the caller. + When `getopt' finds an option that takes an argument, + the argument value is returned here. + Also, when `ordering' is RETURN_IN_ORDER, + each non-option ARGV-element is returned here. */ + +extern char *optarg; + +/* Index in ARGV of the next element to be scanned. + This is used for communication to and from the caller + and for communication between successive calls to `getopt'. + + On entry to `getopt', zero means this is the first call; initialize. + + When `getopt' returns -1, this is the index of the first of the + non-option elements that the caller should itself scan. + + Otherwise, `optind' communicates from one call to the next + how much of ARGV has been scanned so far. */ + +extern int optind; + +/* Callers store zero here to inhibit the error message `getopt' prints + for unrecognized options. */ + +extern int opterr; + +/* Set to an option character which was unrecognized. */ + +extern int optopt; + +#ifndef __need_getopt +/* Describe the long-named options requested by the application. + The LONG_OPTIONS argument to getopt_long or getopt_long_only is a vector + of `struct option' terminated by an element containing a name which is + zero. + + The field `has_arg' is: + no_argument (or 0) if the option does not take an argument, + required_argument (or 1) if the option requires an argument, + optional_argument (or 2) if the option takes an optional argument. + + If the field `flag' is not NULL, it points to a variable that is set + to the value given in the field `val' when the option is found, but + left unchanged if the option is not found. + + To have a long-named option do something other than set an `int' to + a compiled-in constant, such as set a value from `optarg', set the + option's `flag' field to zero and its `val' field to a nonzero + value (the equivalent single-letter option character, if there is + one). For long options that have a zero `flag' field, `getopt' + returns the contents of the `val' field. */ + +struct option +{ + const char *name; + /* has_arg can't be an enum because some compilers complain about + type mismatches in all the code that assumes it is an int. */ + int has_arg; + int *flag; + int val; +}; + +/* Names for the values of the `has_arg' field of `struct option'. */ + +# define no_argument 0 +# define required_argument 1 +# define optional_argument 2 +#endif /* need getopt */ + + +/* Get definitions and prototypes for functions to process the + arguments in ARGV (ARGC of them, minus the program name) for + options given in OPTS. + + Return the option character from OPTS just read. Return -1 when + there are no more options. For unrecognized options, or options + missing arguments, `optopt' is set to the option letter, and '?' is + returned. + + The OPTS string is a list of characters which are recognized option + letters, optionally followed by colons, specifying that that letter + takes an argument, to be placed in `optarg'. + + If a letter in OPTS is followed by two colons, its argument is + optional. This behavior is specific to the GNU `getopt'. + + The argument `--' causes premature termination of argument + scanning, explicitly telling `getopt' that there are no more + options. + + If OPTS begins with `--', then non-option arguments are treated as + arguments to the option '\0'. This behavior is specific to the GNU + `getopt'. */ + +extern int getopt (int ___argc, char *const *___argv, const char *__shortopts) + __THROW; + +#ifndef __need_getopt +extern int getopt_long (int ___argc, char *__getopt_argv_const *___argv, + const char *__shortopts, + const struct option *__longopts, int *__longind) + __THROW; +extern int getopt_long_only (int ___argc, char *__getopt_argv_const *___argv, + const char *__shortopts, + const struct option *__longopts, int *__longind) + __THROW; + +#endif + +#ifdef __cplusplus +} +#endif + +/* Make sure we later can get all the definitions and declarations. */ +#undef __need_getopt + +#endif /* getopt.h */ diff --git a/lib/getopt_int.h b/lib/getopt_int.h new file mode 100644 index 0000000..401579f --- /dev/null +++ b/lib/getopt_int.h @@ -0,0 +1,131 @@ +/* Internal declarations for getopt. + Copyright (C) 1989-1994,1996-1999,2001,2003,2004 + Free Software Foundation, Inc. + This file is part of the GNU C Library. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _GETOPT_INT_H +#define _GETOPT_INT_H 1 + +extern int _getopt_internal (int ___argc, char **___argv, + const char *__shortopts, + const struct option *__longopts, int *__longind, + int __long_only, int __posixly_correct); + + +/* Reentrant versions which can handle parsing multiple argument + vectors at the same time. */ + +/* Data type for reentrant functions. */ +struct _getopt_data +{ + /* These have exactly the same meaning as the corresponding global + variables, except that they are used for the reentrant + versions of getopt. */ + int optind; + int opterr; + int optopt; + char *optarg; + + /* Internal members. */ + + /* True if the internal members have been initialized. */ + int __initialized; + + /* The next char to be scanned in the option-element + in which the last option character we returned was found. + This allows us to pick up the scan where we left off. + + If this is zero, or a null string, it means resume the scan + by advancing to the next ARGV-element. */ + char *__nextchar; + + /* Describe how to deal with options that follow non-option ARGV-elements. + + If the caller did not specify anything, + the default is REQUIRE_ORDER if the environment variable + POSIXLY_CORRECT is defined, PERMUTE otherwise. + + REQUIRE_ORDER means don't recognize them as options; + stop option processing when the first non-option is seen. + This is what Unix does. + This mode of operation is selected by either setting the environment + variable POSIXLY_CORRECT, or using `+' as the first character + of the list of option characters, or by calling getopt. + + PERMUTE is the default. We permute the contents of ARGV as we + scan, so that eventually all the non-options are at the end. + This allows options to be given in any order, even with programs + that were not written to expect this. + + RETURN_IN_ORDER is an option available to programs that were + written to expect options and other ARGV-elements in any order + and that care about the ordering of the two. We describe each + non-option ARGV-element as if it were the argument of an option + with character code 1. Using `-' as the first character of the + list of option characters selects this mode of operation. + + The special argument `--' forces an end of option-scanning regardless + of the value of `ordering'. In the case of RETURN_IN_ORDER, only + `--' can cause `getopt' to return -1 with `optind' != ARGC. */ + + enum + { + REQUIRE_ORDER, PERMUTE, RETURN_IN_ORDER + } __ordering; + + /* If the POSIXLY_CORRECT environment variable is set + or getopt was called. */ + int __posixly_correct; + + + /* Handle permutation of arguments. */ + + /* Describe the part of ARGV that contains non-options that have + been skipped. `first_nonopt' is the index in ARGV of the first + of them; `last_nonopt' is the index after the last of them. */ + + int __first_nonopt; + int __last_nonopt; + +#if defined _LIBC && defined USE_NONOPTION_FLAGS + int __nonoption_flags_max_len; + int __nonoption_flags_len; +# endif +}; + +/* The initializer is necessary to set OPTIND and OPTERR to their + default values and to clear the initialization flag. */ +#define _GETOPT_DATA_INITIALIZER { 1, 1 } + +extern int _getopt_internal_r (int ___argc, char **___argv, + const char *__shortopts, + const struct option *__longopts, int *__longind, + int __long_only, int __posixly_correct, + struct _getopt_data *__data); + +extern int _getopt_long_r (int ___argc, char **___argv, + const char *__shortopts, + const struct option *__longopts, int *__longind, + struct _getopt_data *__data); + +extern int _getopt_long_only_r (int ___argc, char **___argv, + const char *__shortopts, + const struct option *__longopts, + int *__longind, + struct _getopt_data *__data); + +#endif /* getopt_int.h */ diff --git a/lib/getpagesize.h b/lib/getpagesize.h new file mode 100644 index 0000000..17f070f --- /dev/null +++ b/lib/getpagesize.h @@ -0,0 +1,65 @@ +/* Emulate getpagesize on systems that lack it. + Copyright (C) 1999, 2000, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, + USA. */ + +#ifndef HAVE_GETPAGESIZE + +#ifdef HAVE_UNISTD_H +# include <unistd.h> +#endif + +#if !defined getpagesize && defined _SC_PAGESIZE +# if !(defined VMS && __VMS_VER < 70000000) +# define getpagesize() sysconf (_SC_PAGESIZE) +# endif +#endif + +#if !defined getpagesize && defined VMS +# ifdef __ALPHA +# define getpagesize() 8192 +# else +# define getpagesize() 512 +# endif +#endif + +/* This is for BeOS. */ +#if !defined getpagesize && HAVE_OS_H +# include <OS.h> +# if defined B_PAGE_SIZE +# define getpagesize() B_PAGE_SIZE +# endif +#endif + +#if !defined getpagesize && HAVE_SYS_PARAM_H +# include <sys/param.h> +# ifdef EXEC_PAGESIZE +# define getpagesize() EXEC_PAGESIZE +# else +# ifdef NBPG +# ifndef CLSIZE +# define CLSIZE 1 +# endif +# define getpagesize() (NBPG * CLSIZE) +# else +# ifdef NBPC +# define getpagesize() NBPC +# endif +# endif +# endif +#endif + +#endif /* not HAVE_GETPAGESIZE */ diff --git a/lib/getpass.c b/lib/getpass.c new file mode 100644 index 0000000..f0b3184 --- /dev/null +++ b/lib/getpass.c @@ -0,0 +1,226 @@ +/* Copyright (C) 1992-2001, 2003, 2004, 2005 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "getpass.h" + +#include <stdio.h> + +#if !defined _WIN32 + +#include <stdbool.h> + +#if HAVE_STDIO_EXT_H +# include <stdio_ext.h> +#endif +#if !HAVE___FSETLOCKING +# define __fsetlocking(stream, type) /* empty */ +#endif + +#if HAVE_TERMIOS_H +# include <termios.h> +#endif + +#include "getline.h" + +#if USE_UNLOCKED_IO +# include "unlocked-io.h" +#else +# if !HAVE_DECL_FFLUSH_UNLOCKED +# undef fflush_unlocked +# define fflush_unlocked(x) fflush (x) +# endif +# if !HAVE_DECL_FLOCKFILE +# undef flockfile +# define flockfile(x) ((void) 0) +# endif +# if !HAVE_DECL_FUNLOCKFILE +# undef funlockfile +# define funlockfile(x) ((void) 0) +# endif +# if !HAVE_DECL_FPUTS_UNLOCKED +# undef fputs_unlocked +# define fputs_unlocked(str,stream) fputs (str, stream) +# endif +# if !HAVE_DECL_PUTC_UNLOCKED +# undef putc_unlocked +# define putc_unlocked(c,stream) putc (c, stream) +# endif +#endif + +/* It is desirable to use this bit on systems that have it. + The only bit of terminal state we want to twiddle is echoing, which is + done in software; there is no need to change the state of the terminal + hardware. */ + +#ifndef TCSASOFT +# define TCSASOFT 0 +#endif + +static void +call_fclose (void *arg) +{ + if (arg != NULL) + fclose (arg); +} + +char * +getpass (const char *prompt) +{ + FILE *tty; + FILE *in, *out; + struct termios s, t; + bool tty_changed = false; + static char *buf; + static size_t bufsize; + ssize_t nread; + + /* Try to write to and read from the terminal if we can. + If we can't open the terminal, use stderr and stdin. */ + + tty = fopen ("/dev/tty", "w+"); + if (tty == NULL) + { + in = stdin; + out = stderr; + } + else + { + /* We do the locking ourselves. */ + __fsetlocking (tty, FSETLOCKING_BYCALLER); + + out = in = tty; + } + + flockfile (out); + + /* Turn echoing off if it is on now. */ +#if HAVE_TCGETATTR + if (tcgetattr (fileno (in), &t) == 0) + { + /* Save the old one. */ + s = t; + /* Tricky, tricky. */ + t.c_lflag &= ~(ECHO | ISIG); + tty_changed = (tcsetattr (fileno (in), TCSAFLUSH | TCSASOFT, &t) == 0); + } +#endif + + /* Write the prompt. */ + fputs_unlocked (prompt, out); + fflush_unlocked (out); + + /* Read the password. */ + nread = getline (&buf, &bufsize, in); + + /* According to the C standard, input may not be followed by output + on the same stream without an intervening call to a file + positioning function. Suppose in == out; then without this fseek + call, on Solaris, HP-UX, AIX, OSF/1, the previous input gets + echoed, whereas on IRIX, the following newline is not output as + it should be. POSIX imposes similar restrictions if fileno (in) + == fileno (out). The POSIX restrictions are tricky and change + from POSIX version to POSIX version, so play it safe and invoke + fseek even if in != out. */ + fseek (out, 0, SEEK_CUR); + + if (buf != NULL) + { + if (nread < 0) + buf[0] = '\0'; + else if (buf[nread - 1] == '\n') + { + /* Remove the newline. */ + buf[nread - 1] = '\0'; + if (tty_changed) + { + /* Write the newline that was not echoed. */ + putc_unlocked ('\n', out); + } + } + } + + /* Restore the original setting. */ +#if HAVE_TCSETATTR + if (tty_changed) + tcsetattr (fileno (in), TCSAFLUSH | TCSASOFT, &s); +#endif + + funlockfile (out); + + call_fclose (tty); + + return buf; +} + +#else /* WIN32 */ + +/* Windows implementation by Martin Lambers <marlam@marlam.de>, + improved by Simon Josefsson. */ + +/* For PASS_MAX. */ +#include <limits.h> + +#ifndef PASS_MAX +# define PASS_MAX 512 +#endif + +char * +getpass (const char *prompt) +{ + char getpassbuf[PASS_MAX + 1]; + size_t i = 0; + int c; + + if (prompt) + { + fputs (prompt, stderr); + fflush (stderr); + } + + for (;;) + { + c = _getch (); + if (c == '\r') + { + getpassbuf[i] = '\0'; + break; + } + else if (i < PASS_MAX) + { + getpassbuf[i++] = c; + } + + if (i >= PASS_MAX) + { + getpassbuf[i] = '\0'; + break; + } + } + + if (prompt) + { + fputs ("\r\n", stderr); + fflush (stderr); + } + + return strdup (getpassbuf); +} +#endif diff --git a/lib/getpass.h b/lib/getpass.h new file mode 100644 index 0000000..bdff875 --- /dev/null +++ b/lib/getpass.h @@ -0,0 +1,31 @@ +/* getpass.h -- Read a password of arbitrary length from /dev/tty or stdin. + Copyright (C) 2004 Free Software Foundation, Inc. + Contributed by Simon Josefsson <jas@extundo.com>, 2004. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef GETPASS_H +# define GETPASS_H + +/* Get getpass declaration, if available. */ +# include <unistd.h> + +# if defined HAVE_DECL_GETPASS && !HAVE_DECL_GETPASS +/* Read a password of arbitrary length from /dev/tty or stdin. */ +char *getpass (const char *prompt); + +# endif + +#endif /* GETPASS_H */ diff --git a/lib/gettext.h b/lib/gettext.h new file mode 100644 index 0000000..285cb31 --- /dev/null +++ b/lib/gettext.h @@ -0,0 +1,78 @@ +/* Convenience header for conditional use of GNU <libintl.h>. + Copyright (C) 1995-1998, 2000-2002, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _LIBGETTEXT_H +#define _LIBGETTEXT_H 1 + +/* NLS can be disabled through the configure --disable-nls option. */ +#if ENABLE_NLS + +/* Get declarations of GNU message catalog functions. */ +# include <libintl.h> + +#else + +/* Solaris /usr/include/locale.h includes /usr/include/libintl.h, which + chokes if dcgettext is defined as a macro. So include it now, to make + later inclusions of <locale.h> a NOP. We don't include <libintl.h> + as well because people using "gettext.h" will not include <libintl.h>, + and also including <libintl.h> would fail on SunOS 4, whereas <locale.h> + is OK. */ +#if defined(__sun) +# include <locale.h> +#endif + +/* Many header files from the libstdc++ coming with g++ 3.3 or newer include + <libintl.h>, which chokes if dcgettext is defined as a macro. So include + it now, to make later inclusions of <libintl.h> a NOP. */ +#if defined(__cplusplus) && defined(__GNUG__) && (__GNUC__ >= 3) +# include <cstdlib> +# if (__GLIBC__ >= 2) || _GLIBCXX_HAVE_LIBINTL_H +# include <libintl.h> +# endif +#endif + +/* Disabled NLS. + The casts to 'const char *' serve the purpose of producing warnings + for invalid uses of the value returned from these functions. + On pre-ANSI systems without 'const', the config.h file is supposed to + contain "#define const". */ +# define gettext(Msgid) ((const char *) (Msgid)) +# define dgettext(Domainname, Msgid) ((const char *) (Msgid)) +# define dcgettext(Domainname, Msgid, Category) ((const char *) (Msgid)) +# define ngettext(Msgid1, Msgid2, N) \ + ((N) == 1 ? (const char *) (Msgid1) : (const char *) (Msgid2)) +# define dngettext(Domainname, Msgid1, Msgid2, N) \ + ((N) == 1 ? (const char *) (Msgid1) : (const char *) (Msgid2)) +# define dcngettext(Domainname, Msgid1, Msgid2, N, Category) \ + ((N) == 1 ? (const char *) (Msgid1) : (const char *) (Msgid2)) +# define textdomain(Domainname) ((const char *) (Domainname)) +# define bindtextdomain(Domainname, Dirname) ((const char *) (Dirname)) +# define bind_textdomain_codeset(Domainname, Codeset) ((const char *) (Codeset)) + +#endif + +/* A pseudo function call that serves as a marker for the automated + extraction of messages, but does not call gettext(). The run-time + translation is done at a different place in the code. + The argument, String, should be a literal string. Concatenated strings + and other string expressions won't work. + The macro's expansion is not parenthesized, so that it is suitable as + initializer for static 'char[]' or 'const char[]' variables. */ +#define gettext_noop(String) String + +#endif /* _LIBGETTEXT_H */ diff --git a/lib/gettime.c b/lib/gettime.c new file mode 100644 index 0000000..0f94297 --- /dev/null +++ b/lib/gettime.c @@ -0,0 +1,53 @@ +/* gettime -- get the system clock + Copyright (C) 2002, 2004, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Paul Eggert. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "timespec.h" + +/* Get the system time into *TS. */ + +void +gettime (struct timespec *ts) +{ +#if HAVE_NANOTIME + nanotime (ts); +#else + +# if defined CLOCK_REALTIME && HAVE_CLOCK_GETTIME + if (clock_gettime (CLOCK_REALTIME, ts) == 0) + return; +# endif + +# if HAVE_GETTIMEOFDAY + { + struct timeval tv; + gettimeofday (&tv, NULL); + ts->tv_sec = tv.tv_sec; + ts->tv_nsec = tv.tv_usec * 1000; + } +# else + ts->tv_sec = time (NULL); + ts->tv_nsec = 0; +# endif + +#endif +} diff --git a/lib/gettimeofday.c b/lib/gettimeofday.c new file mode 100644 index 0000000..eb6de94 --- /dev/null +++ b/lib/gettimeofday.c @@ -0,0 +1,123 @@ +/* Work around the bug in some systems whereby gettimeofday clobbers the + static buffer that localtime uses for it's return value. The gettimeofday + function from Mac OS X 10.0.4, i.e. Darwin 1.3.7 has this problem. + The tzset replacement is necessary for at least Solaris 2.5, 2.5.1, and 2.6. + Copyright (C) 2001, 2002, 2003 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* written by Jim Meyering */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Disable the definitions of these functions (from config.h) + so we can use the library versions here. */ +#undef gettimeofday +#undef gmtime +#undef localtime +#undef tzset + +#include <sys/types.h> + +#if TIME_WITH_SYS_TIME +# include <sys/time.h> +# include <time.h> +#else +# if HAVE_SYS_TIME_H +# include <sys/time.h> +# else +# include <time.h> +# endif +#endif + +#include <stdlib.h> + +static struct tm *localtime_buffer_addr; + +/* This is a wrapper for localtime. It is used only on systems for which + gettimeofday clobbers the static buffer used for localtime's result. + + On the first call, record the address of the static buffer that + localtime uses for its result. */ + +struct tm * +rpl_localtime (const time_t *timep) +{ + struct tm *tm = localtime (timep); + + if (! localtime_buffer_addr) + localtime_buffer_addr = tm; + + return tm; +} + +/* Same as above, since gmtime and localtime use the same buffer. */ +struct tm * +rpl_gmtime (const time_t *timep) +{ + struct tm *tm = gmtime (timep); + + if (! localtime_buffer_addr) + localtime_buffer_addr = tm; + + return tm; +} + +/* This is a wrapper for gettimeofday. It is used only on systems for which + gettimeofday clobbers the static buffer used for localtime's result. + + Save and restore the contents of the buffer used for localtime's result + around the call to gettimeofday. */ + +int +rpl_gettimeofday (struct timeval *tv, struct timezone *tz) +{ + struct tm save; + int result; + + if (! localtime_buffer_addr) + { + time_t t = 0; + localtime_buffer_addr = localtime (&t); + } + + save = *localtime_buffer_addr; + result = gettimeofday (tv, tz); + *localtime_buffer_addr = save; + + return result; +} + +/* This is a wrapper for tzset. It is used only on systems for which + tzset may clobber the static buffer used for localtime's result. + Save and restore the contents of the buffer used for localtime's + result around the call to tzset. */ +void +rpl_tzset (void) +{ + struct tm save; + + if (! localtime_buffer_addr) + { + time_t t = 0; + localtime_buffer_addr = localtime (&t); + } + + save = *localtime_buffer_addr; + tzset (); + *localtime_buffer_addr = save; +} diff --git a/lib/glob-libc.h b/lib/glob-libc.h new file mode 100644 index 0000000..95b76c8 --- /dev/null +++ b/lib/glob-libc.h @@ -0,0 +1,196 @@ +/* Copyright (C) 1991,92,95-98,2000,2001,2004 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, write to the Free + Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA + 02111-1307 USA. */ + +#ifndef _GLOB_H +#define _GLOB_H 1 + +#ifndef __GLOB_GNULIB +# include <sys/cdefs.h> +#endif + +__BEGIN_DECLS + +/* We need `size_t' for the following definitions. */ +#ifndef __size_t +# if defined __GNUC__ && __GNUC__ >= 2 +typedef __SIZE_TYPE__ __size_t; +# ifdef __USE_XOPEN +typedef __SIZE_TYPE__ size_t; +# endif +# else +# include <stddef.h> +# ifndef __size_t +# define __size_t size_t +# endif +# endif +#else +/* The GNU CC stddef.h version defines __size_t as empty. We need a real + definition. */ +# undef __size_t +# define __size_t size_t +#endif + +/* Bits set in the FLAGS argument to `glob'. */ +#define GLOB_ERR (1 << 0)/* Return on read errors. */ +#define GLOB_MARK (1 << 1)/* Append a slash to each name. */ +#define GLOB_NOSORT (1 << 2)/* Don't sort the names. */ +#define GLOB_DOOFFS (1 << 3)/* Insert PGLOB->gl_offs NULLs. */ +#define GLOB_NOCHECK (1 << 4)/* If nothing matches, return the pattern. */ +#define GLOB_APPEND (1 << 5)/* Append to results of a previous call. */ +#define GLOB_NOESCAPE (1 << 6)/* Backslashes don't quote metacharacters. */ +#define GLOB_PERIOD (1 << 7)/* Leading `.' can be matched by metachars. */ + +#if !defined __USE_POSIX2 || defined __USE_BSD || defined __USE_GNU +# define GLOB_MAGCHAR (1 << 8)/* Set in gl_flags if any metachars seen. */ +# define GLOB_ALTDIRFUNC (1 << 9)/* Use gl_opendir et al functions. */ +# define GLOB_BRACE (1 << 10)/* Expand "{a,b}" to "a" "b". */ +# define GLOB_NOMAGIC (1 << 11)/* If no magic chars, return the pattern. */ +# define GLOB_TILDE (1 << 12)/* Expand ~user and ~ to home directories. */ +# define GLOB_ONLYDIR (1 << 13)/* Match only directories. */ +# define GLOB_TILDE_CHECK (1 << 14)/* Like GLOB_TILDE but return an error + if the user name is not available. */ +# define __GLOB_FLAGS (GLOB_ERR|GLOB_MARK|GLOB_NOSORT|GLOB_DOOFFS| \ + GLOB_NOESCAPE|GLOB_NOCHECK|GLOB_APPEND| \ + GLOB_PERIOD|GLOB_ALTDIRFUNC|GLOB_BRACE| \ + GLOB_NOMAGIC|GLOB_TILDE|GLOB_ONLYDIR|GLOB_TILDE_CHECK) +#else +# define __GLOB_FLAGS (GLOB_ERR|GLOB_MARK|GLOB_NOSORT|GLOB_DOOFFS| \ + GLOB_NOESCAPE|GLOB_NOCHECK|GLOB_APPEND| \ + GLOB_PERIOD) +#endif + +/* Error returns from `glob'. */ +#define GLOB_NOSPACE 1 /* Ran out of memory. */ +#define GLOB_ABORTED 2 /* Read error. */ +#define GLOB_NOMATCH 3 /* No matches found. */ +#define GLOB_NOSYS 4 /* Not implemented. */ +#ifdef __USE_GNU +/* Previous versions of this file defined GLOB_ABEND instead of + GLOB_ABORTED. Provide a compatibility definition here. */ +# define GLOB_ABEND GLOB_ABORTED +#endif + +/* Structure describing a globbing run. */ +#ifdef __USE_GNU +struct stat; +#endif +typedef struct + { + __size_t gl_pathc; /* Count of paths matched by the pattern. */ + char **gl_pathv; /* List of matched pathnames. */ + __size_t gl_offs; /* Slots to reserve in `gl_pathv'. */ + int gl_flags; /* Set to FLAGS, maybe | GLOB_MAGCHAR. */ + + /* If the GLOB_ALTDIRFUNC flag is set, the following functions + are used instead of the normal file access functions. */ + void (*gl_closedir) (void *); +#ifdef __USE_GNU + struct dirent *(*gl_readdir) (void *); +#else + void *(*gl_readdir) (void *); +#endif + void *(*gl_opendir) (__const char *); +#ifdef __USE_GNU + int (*gl_lstat) (__const char *__restrict, struct stat *__restrict); + int (*gl_stat) (__const char *__restrict, struct stat *__restrict); +#else + int (*gl_lstat) (__const char *__restrict, void *__restrict); + int (*gl_stat) (__const char *__restrict, void *__restrict); +#endif + } glob_t; + +#if defined __USE_LARGEFILE64 && !defined __GLOB_GNULIB +# ifdef __USE_GNU +struct stat64; +# endif +typedef struct + { + __size_t gl_pathc; + char **gl_pathv; + __size_t gl_offs; + int gl_flags; + + /* If the GLOB_ALTDIRFUNC flag is set, the following functions + are used instead of the normal file access functions. */ + void (*gl_closedir) (void *); +# ifdef __USE_GNU + struct dirent64 *(*gl_readdir) (void *); +# else + void *(*gl_readdir) (void *); +# endif + void *(*gl_opendir) (__const char *); +# ifdef __USE_GNU + int (*gl_lstat) (__const char *__restrict, struct stat64 *__restrict); + int (*gl_stat) (__const char *__restrict, struct stat64 *__restrict); +# else + int (*gl_lstat) (__const char *__restrict, void *__restrict); + int (*gl_stat) (__const char *__restrict, void *__restrict); +# endif + } glob64_t; +#endif + +#if __USE_FILE_OFFSET64 && __GNUC__ < 2 && !defined __GLOB_GNULIB +# define glob glob64 +# define globfree globfree64 +#endif + +/* Do glob searching for PATTERN, placing results in PGLOB. + The bits defined above may be set in FLAGS. + If a directory cannot be opened or read and ERRFUNC is not nil, + it is called with the pathname that caused the error, and the + `errno' value from the failing call; if it returns non-zero + `glob' returns GLOB_ABEND; if it returns zero, the error is ignored. + If memory cannot be allocated for PGLOB, GLOB_NOSPACE is returned. + Otherwise, `glob' returns zero. */ +#if !defined __USE_FILE_OFFSET64 || __GNUC__ < 2 || defined __GLOB_GNULIB +extern int glob (__const char *__restrict __pattern, int __flags, + int (*__errfunc) (__const char *, int), + glob_t *__restrict __pglob) __THROW; + +/* Free storage allocated in PGLOB by a previous `glob' call. */ +extern void globfree (glob_t *__pglob) __THROW; +#else +extern int __REDIRECT_NTH (glob, (__const char *__restrict __pattern, + int __flags, + int (*__errfunc) (__const char *, int), + glob_t *__restrict __pglob), glob64); + +extern void __REDIRECT_NTH (globfree, (glob_t *__pglob), globfree64); +#endif + +#if defined __USE_LARGEFILE64 && !defined __GLOB_GNULIB +extern int glob64 (__const char *__restrict __pattern, int __flags, + int (*__errfunc) (__const char *, int), + glob64_t *__restrict __pglob) __THROW; + +extern void globfree64 (glob64_t *__pglob) __THROW; +#endif + + +#ifdef __USE_GNU +/* Return nonzero if PATTERN contains any metacharacters. + Metacharacters can be quoted with backslashes if QUOTE is nonzero. + + This function is not part of the interface specified by POSIX.2 + but several programs want to use it. */ +extern int glob_pattern_p (__const char *__pattern, int __quote) __THROW; +#endif + +__END_DECLS + +#endif /* glob.h */ diff --git a/lib/glob.c b/lib/glob.c new file mode 100644 index 0000000..0567902 --- /dev/null +++ b/lib/glob.c @@ -0,0 +1,1289 @@ +/* Copyright (C) 1991-2002,2003,2004,2005 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, write to the Free + Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA + 02111-1307 USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include <glob.h> + +#include <errno.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <stddef.h> + +/* Outcomment the following line for production quality code. */ +/* #define NDEBUG 1 */ +#include <assert.h> + +#include <stdio.h> /* Needed on stupid SunOS for assert. */ + +#if !defined _LIBC || !defined GLOB_ONLY_P +#if defined HAVE_UNISTD_H || defined _LIBC +# include <unistd.h> +# ifndef POSIX +# ifdef _POSIX_VERSION +# define POSIX +# endif +# endif +#endif + +#include <pwd.h> + +#include <errno.h> +#ifndef __set_errno +# define __set_errno(val) errno = (val) +#endif + +#if defined HAVE_DIRENT_H || defined __GNU_LIBRARY__ +# include <dirent.h> +# define NAMLEN(dirent) strlen((dirent)->d_name) +#else +# define dirent direct +# define NAMLEN(dirent) (dirent)->d_namlen +# ifdef HAVE_SYS_NDIR_H +# include <sys/ndir.h> +# endif +# ifdef HAVE_SYS_DIR_H +# include <sys/dir.h> +# endif +# ifdef HAVE_NDIR_H +# include <ndir.h> +# endif +# ifdef HAVE_VMSDIR_H +# include "vmsdir.h" +# endif /* HAVE_VMSDIR_H */ +#endif + + +/* In GNU systems, <dirent.h> defines this macro for us. */ +#ifdef _D_NAMLEN +# undef NAMLEN +# define NAMLEN(d) _D_NAMLEN(d) +#endif + +/* When used in the GNU libc the symbol _DIRENT_HAVE_D_TYPE is available + if the `d_type' member for `struct dirent' is available. + HAVE_STRUCT_DIRENT_D_TYPE plays the same role in GNULIB. */ +#if defined _DIRENT_HAVE_D_TYPE || defined HAVE_STRUCT_DIRENT_D_TYPE +/* True if the directory entry D must be of type T. */ +# define DIRENT_MUST_BE(d, t) ((d)->d_type == (t)) + +/* True if the directory entry D might be a symbolic link. */ +# define DIRENT_MIGHT_BE_SYMLINK(d) \ + ((d)->d_type == DT_UNKNOWN || (d)->d_type == DT_LNK) + +/* True if the directory entry D might be a directory. */ +# define DIRENT_MIGHT_BE_DIR(d) \ + ((d)->d_type == DT_DIR || DIRENT_MIGHT_BE_SYMLINK (d)) + +#else /* !HAVE_D_TYPE */ +# define DIRENT_MUST_BE(d, t) false +# define DIRENT_MIGHT_BE_SYMLINK(d) true +# define DIRENT_MIGHT_BE_DIR(d) true +#endif /* HAVE_D_TYPE */ + +/* If the system has the `struct dirent64' type we use it internally. */ +#if defined _LIBC && !defined COMPILE_GLOB64 +# if defined HAVE_DIRENT_H || defined __GNU_LIBRARY__ +# define CONVERT_D_NAMLEN(d64, d32) +# else +# define CONVERT_D_NAMLEN(d64, d32) \ + (d64)->d_namlen = (d32)->d_namlen; +# endif + +# if (defined POSIX || defined WINDOWS32) && !defined __GNU_LIBRARY__ +# define CONVERT_D_INO(d64, d32) +# else +# define CONVERT_D_INO(d64, d32) \ + (d64)->d_ino = (d32)->d_ino; +# endif + +# ifdef _DIRENT_HAVE_D_TYPE +# define CONVERT_D_TYPE(d64, d32) \ + (d64)->d_type = (d32)->d_type; +# else +# define CONVERT_D_TYPE(d64, d32) +# endif + +# define CONVERT_DIRENT_DIRENT64(d64, d32) \ + memcpy ((d64)->d_name, (d32)->d_name, NAMLEN (d32) + 1); \ + CONVERT_D_NAMLEN (d64, d32) \ + CONVERT_D_INO (d64, d32) \ + CONVERT_D_TYPE (d64, d32) +#endif + + +#if (defined POSIX || defined WINDOWS32) && !defined __GNU_LIBRARY__ +/* Posix does not require that the d_ino field be present, and some + systems do not provide it. */ +# define REAL_DIR_ENTRY(dp) 1 +#else +# define REAL_DIR_ENTRY(dp) (dp->d_ino != 0) +#endif /* POSIX */ + +#include <stdlib.h> +#include <string.h> + +/* NAME_MAX is usually defined in <dirent.h> or <limits.h>. */ +#include <limits.h> +#ifndef NAME_MAX +# define NAME_MAX (sizeof (((struct dirent *) 0)->d_name)) +#endif + +#include <alloca.h> + +#ifdef _LIBC +# undef strdup +# define strdup(str) __strdup (str) +# define sysconf(id) __sysconf (id) +# define closedir(dir) __closedir (dir) +# define opendir(name) __opendir (name) +# define readdir(str) __readdir64 (str) +# define getpwnam_r(name, bufp, buf, len, res) \ + __getpwnam_r (name, bufp, buf, len, res) +# ifndef __stat64 +# define __stat64(fname, buf) __xstat64 (_STAT_VER, fname, buf) +# endif +# define struct_stat64 struct stat64 +#else /* !_LIBC */ +# include "getlogin_r.h" +# include "mempcpy.h" +# include "stat-macros.h" +# include "strdup.h" +# define __stat64(fname, buf) stat (fname, buf) +# define struct_stat64 struct stat +# define __stat(fname, buf) stat (fname, buf) +# define __alloca alloca +# define __readdir readdir +# define __readdir64 readdir64 +# define __glob_pattern_p glob_pattern_p +#endif /* _LIBC */ + +#include <stdbool.h> +#include <fnmatch.h> + +#ifdef _SC_GETPW_R_SIZE_MAX +# define GETPW_R_SIZE_MAX() sysconf (_SC_GETPW_R_SIZE_MAX) +#else +# define GETPW_R_SIZE_MAX() (-1) +#endif +#ifdef _SC_LOGIN_NAME_MAX +# define GET_LOGIN_NAME_MAX() sysconf (_SC_LOGIN_NAME_MAX) +#else +# define GET_LOGIN_NAME_MAX() (-1) +#endif + +static const char *next_brace_sub (const char *begin, int flags) __THROW; + +#endif /* !defined _LIBC || !defined GLOB_ONLY_P */ + +static int glob_in_dir (const char *pattern, const char *directory, + int flags, int (*errfunc) (const char *, int), + glob_t *pglob); + +#if !defined _LIBC || !defined GLOB_ONLY_P +static int prefix_array (const char *prefix, char **array, size_t n) __THROW; +static int collated_compare (const void *, const void *) __THROW; + + +/* Find the end of the sub-pattern in a brace expression. */ +static const char * +next_brace_sub (const char *cp, int flags) +{ + unsigned int depth = 0; + while (*cp != '\0') + if ((flags & GLOB_NOESCAPE) == 0 && *cp == '\\') + { + if (*++cp == '\0') + break; + ++cp; + } + else + { + if ((*cp == '}' && depth-- == 0) || (*cp == ',' && depth == 0)) + break; + + if (*cp++ == '{') + depth++; + } + + return *cp != '\0' ? cp : NULL; +} + +#endif /* !defined _LIBC || !defined GLOB_ONLY_P */ + +/* Do glob searching for PATTERN, placing results in PGLOB. + The bits defined above may be set in FLAGS. + If a directory cannot be opened or read and ERRFUNC is not nil, + it is called with the pathname that caused the error, and the + `errno' value from the failing call; if it returns non-zero + `glob' returns GLOB_ABORTED; if it returns zero, the error is ignored. + If memory cannot be allocated for PGLOB, GLOB_NOSPACE is returned. + Otherwise, `glob' returns zero. */ +int +#ifdef GLOB_ATTRIBUTE +GLOB_ATTRIBUTE +#endif +glob (pattern, flags, errfunc, pglob) + const char *pattern; + int flags; + int (*errfunc) (const char *, int); + glob_t *pglob; +{ + const char *filename; + const char *dirname; + size_t dirlen; + int status; + size_t oldcount; + + if (pattern == NULL || pglob == NULL || (flags & ~__GLOB_FLAGS) != 0) + { + __set_errno (EINVAL); + return -1; + } + + if (!(flags & GLOB_DOOFFS)) + /* Have to do this so `globfree' knows where to start freeing. It + also makes all the code that uses gl_offs simpler. */ + pglob->gl_offs = 0; + + if (flags & GLOB_BRACE) + { + const char *begin; + + if (flags & GLOB_NOESCAPE) + begin = strchr (pattern, '{'); + else + { + begin = pattern; + while (1) + { + if (*begin == '\0') + { + begin = NULL; + break; + } + + if (*begin == '\\' && begin[1] != '\0') + ++begin; + else if (*begin == '{') + break; + + ++begin; + } + } + + if (begin != NULL) + { + /* Allocate working buffer large enough for our work. Note that + we have at least an opening and closing brace. */ + size_t firstc; + char *alt_start; + const char *p; + const char *next; + const char *rest; + size_t rest_len; +#ifdef __GNUC__ + char onealt[strlen (pattern) - 1]; +#else + char *onealt = malloc (strlen (pattern) - 1); + if (onealt == NULL) + { + if (!(flags & GLOB_APPEND)) + { + pglob->gl_pathc = 0; + pglob->gl_pathv = NULL; + } + return GLOB_NOSPACE; + } +#endif + + /* We know the prefix for all sub-patterns. */ + alt_start = mempcpy (onealt, pattern, begin - pattern); + + /* Find the first sub-pattern and at the same time find the + rest after the closing brace. */ + next = next_brace_sub (begin + 1, flags); + if (next == NULL) + { + /* It is an illegal expression. */ +#ifndef __GNUC__ + free (onealt); +#endif + return glob (pattern, flags & ~GLOB_BRACE, errfunc, pglob); + } + + /* Now find the end of the whole brace expression. */ + rest = next; + while (*rest != '}') + { + rest = next_brace_sub (rest + 1, flags); + if (rest == NULL) + { + /* It is an illegal expression. */ +#ifndef __GNUC__ + free (onealt); +#endif + return glob (pattern, flags & ~GLOB_BRACE, errfunc, pglob); + } + } + /* Please note that we now can be sure the brace expression + is well-formed. */ + rest_len = strlen (++rest) + 1; + + /* We have a brace expression. BEGIN points to the opening {, + NEXT points past the terminator of the first element, and END + points past the final }. We will accumulate result names from + recursive runs for each brace alternative in the buffer using + GLOB_APPEND. */ + + if (!(flags & GLOB_APPEND)) + { + /* This call is to set a new vector, so clear out the + vector so we can append to it. */ + pglob->gl_pathc = 0; + pglob->gl_pathv = NULL; + } + firstc = pglob->gl_pathc; + + p = begin + 1; + while (1) + { + int result; + + /* Construct the new glob expression. */ + mempcpy (mempcpy (alt_start, p, next - p), rest, rest_len); + + result = glob (onealt, + ((flags & ~(GLOB_NOCHECK | GLOB_NOMAGIC)) + | GLOB_APPEND), errfunc, pglob); + + /* If we got an error, return it. */ + if (result && result != GLOB_NOMATCH) + { +#ifndef __GNUC__ + free (onealt); +#endif + if (!(flags & GLOB_APPEND)) + { + globfree (pglob); + pglob->gl_pathc = 0; + } + return result; + } + + if (*next == '}') + /* We saw the last entry. */ + break; + + p = next + 1; + next = next_brace_sub (p, flags); + assert (next != NULL); + } + +#ifndef __GNUC__ + free (onealt); +#endif + + if (pglob->gl_pathc != firstc) + /* We found some entries. */ + return 0; + else if (!(flags & (GLOB_NOCHECK|GLOB_NOMAGIC))) + return GLOB_NOMATCH; + } + } + + /* Find the filename. */ + filename = strrchr (pattern, '/'); +#if defined __MSDOS__ || defined WINDOWS32 + /* The case of "d:pattern". Since `:' is not allowed in + file names, we can safely assume that wherever it + happens in pattern, it signals the filename part. This + is so we could some day support patterns like "[a-z]:foo". */ + if (filename == NULL) + filename = strchr (pattern, ':'); +#endif /* __MSDOS__ || WINDOWS32 */ + if (filename == NULL) + { + /* This can mean two things: a simple name or "~name". The latter + case is nothing but a notation for a directory. */ + if ((flags & (GLOB_TILDE|GLOB_TILDE_CHECK)) && pattern[0] == '~') + { + dirname = pattern; + dirlen = strlen (pattern); + + /* Set FILENAME to NULL as a special flag. This is ugly but + other solutions would require much more code. We test for + this special case below. */ + filename = NULL; + } + else + { + filename = pattern; +#ifdef _AMIGA + dirname = ""; +#else + dirname = "."; +#endif + dirlen = 0; + } + } + else if (filename == pattern) + { + /* "/pattern". */ + dirname = "/"; + dirlen = 1; + ++filename; + } + else + { + char *newp; + dirlen = filename - pattern; +#if defined __MSDOS__ || defined WINDOWS32 + if (*filename == ':' + || (filename > pattern + 1 && filename[-1] == ':')) + { + char *drive_spec; + + ++dirlen; + drive_spec = __alloca (dirlen + 1); + *((char *) mempcpy (drive_spec, pattern, dirlen)) = '\0'; + /* For now, disallow wildcards in the drive spec, to + prevent infinite recursion in glob. */ + if (__glob_pattern_p (drive_spec, !(flags & GLOB_NOESCAPE))) + return GLOB_NOMATCH; + /* If this is "d:pattern", we need to copy `:' to DIRNAME + as well. If it's "d:/pattern", don't remove the slash + from "d:/", since "d:" and "d:/" are not the same.*/ + } +#endif + newp = __alloca (dirlen + 1); + *((char *) mempcpy (newp, pattern, dirlen)) = '\0'; + dirname = newp; + ++filename; + + if (filename[0] == '\0' +#if defined __MSDOS__ || defined WINDOWS32 + && dirname[dirlen - 1] != ':' + && (dirlen < 3 || dirname[dirlen - 2] != ':' + || dirname[dirlen - 1] != '/') +#endif + && dirlen > 1) + /* "pattern/". Expand "pattern", appending slashes. */ + { + int val = glob (dirname, flags | GLOB_MARK, errfunc, pglob); + if (val == 0) + pglob->gl_flags = ((pglob->gl_flags & ~GLOB_MARK) + | (flags & GLOB_MARK)); + return val; + } + } + + if (!(flags & GLOB_APPEND)) + { + pglob->gl_pathc = 0; + if (!(flags & GLOB_DOOFFS)) + pglob->gl_pathv = NULL; + else + { + size_t i; + pglob->gl_pathv = malloc ((pglob->gl_offs + 1) * sizeof (char *)); + if (pglob->gl_pathv == NULL) + return GLOB_NOSPACE; + + for (i = 0; i <= pglob->gl_offs; ++i) + pglob->gl_pathv[i] = NULL; + } + } + + oldcount = pglob->gl_pathc + pglob->gl_offs; + +#ifndef VMS + if ((flags & (GLOB_TILDE|GLOB_TILDE_CHECK)) && dirname[0] == '~') + { + if (dirname[1] == '\0' || dirname[1] == '/') + { + /* Look up home directory. */ + const char *home_dir = getenv ("HOME"); +# ifdef _AMIGA + if (home_dir == NULL || home_dir[0] == '\0') + home_dir = "SYS:"; +# else +# ifdef WINDOWS32 + if (home_dir == NULL || home_dir[0] == '\0') + home_dir = "c:/users/default"; /* poor default */ +# else + if (home_dir == NULL || home_dir[0] == '\0') + { + int success; + char *name; + size_t buflen = GET_LOGIN_NAME_MAX () + 1; + + if (buflen == 0) + /* `sysconf' does not support _SC_LOGIN_NAME_MAX. Try + a moderate value. */ + buflen = 20; + name = __alloca (buflen); + + success = getlogin_r (name, buflen) == 0; + if (success) + { + struct passwd *p; +# if defined HAVE_GETPWNAM_R || defined _LIBC + long int pwbuflen = GETPW_R_SIZE_MAX (); + char *pwtmpbuf; + struct passwd pwbuf; + int save = errno; + +# ifndef _LIBC + if (pwbuflen == -1) + /* `sysconf' does not support _SC_GETPW_R_SIZE_MAX. + Try a moderate value. */ + pwbuflen = 1024; +# endif + pwtmpbuf = __alloca (pwbuflen); + + while (getpwnam_r (name, &pwbuf, pwtmpbuf, pwbuflen, &p) + != 0) + { + if (errno != ERANGE) + { + p = NULL; + break; + } +# ifdef _LIBC + pwtmpbuf = extend_alloca (pwtmpbuf, pwbuflen, + 2 * pwbuflen); +# else + pwbuflen *= 2; + pwtmpbuf = __alloca (pwbuflen); +# endif + __set_errno (save); + } +# else + p = getpwnam (name); +# endif + if (p != NULL) + home_dir = p->pw_dir; + } + } + if (home_dir == NULL || home_dir[0] == '\0') + { + if (flags & GLOB_TILDE_CHECK) + return GLOB_NOMATCH; + else + home_dir = "~"; /* No luck. */ + } +# endif /* WINDOWS32 */ +# endif + /* Now construct the full directory. */ + if (dirname[1] == '\0') + dirname = home_dir; + else + { + char *newp; + size_t home_len = strlen (home_dir); + newp = __alloca (home_len + dirlen); + mempcpy (mempcpy (newp, home_dir, home_len), + &dirname[1], dirlen); + dirname = newp; + } + } +# if !defined _AMIGA && !defined WINDOWS32 + else + { + char *end_name = strchr (dirname, '/'); + const char *user_name; + const char *home_dir; + + if (end_name == NULL) + user_name = dirname + 1; + else + { + char *newp; + newp = __alloca (end_name - dirname); + *((char *) mempcpy (newp, dirname + 1, end_name - dirname)) + = '\0'; + user_name = newp; + } + + /* Look up specific user's home directory. */ + { + struct passwd *p; +# if defined HAVE_GETPWNAM_R || defined _LIBC + long int buflen = GETPW_R_SIZE_MAX (); + char *pwtmpbuf; + struct passwd pwbuf; + int save = errno; + +# ifndef _LIBC + if (buflen == -1) + /* `sysconf' does not support _SC_GETPW_R_SIZE_MAX. Try a + moderate value. */ + buflen = 1024; +# endif + pwtmpbuf = __alloca (buflen); + + while (getpwnam_r (user_name, &pwbuf, pwtmpbuf, buflen, &p) != 0) + { + if (errno != ERANGE) + { + p = NULL; + break; + } +# ifdef _LIBC + pwtmpbuf = extend_alloca (pwtmpbuf, buflen, 2 * buflen); +# else + buflen *= 2; + pwtmpbuf = __alloca (buflen); +# endif + __set_errno (save); + } +# else + p = getpwnam (user_name); +# endif + if (p != NULL) + home_dir = p->pw_dir; + else + home_dir = NULL; + } + /* If we found a home directory use this. */ + if (home_dir != NULL) + { + char *newp; + size_t home_len = strlen (home_dir); + size_t rest_len = end_name == NULL ? 0 : strlen (end_name); + newp = __alloca (home_len + rest_len + 1); + *((char *) mempcpy (mempcpy (newp, home_dir, home_len), + end_name, rest_len)) = '\0'; + dirname = newp; + } + else + if (flags & GLOB_TILDE_CHECK) + /* We have to regard it as an error if we cannot find the + home directory. */ + return GLOB_NOMATCH; + } +# endif /* Not Amiga && not WINDOWS32. */ + } +#endif /* Not VMS. */ + + /* Now test whether we looked for "~" or "~NAME". In this case we + can give the answer now. */ + if (filename == NULL) + { + struct stat st; + struct_stat64 st64; + + /* Return the directory if we don't check for error or if it exists. */ + if ((flags & GLOB_NOCHECK) + || (((flags & GLOB_ALTDIRFUNC) + ? ((*pglob->gl_stat) (dirname, &st) == 0 + && S_ISDIR (st.st_mode)) + : (__stat64 (dirname, &st64) == 0 && S_ISDIR (st64.st_mode))))) + { + int newcount = pglob->gl_pathc + pglob->gl_offs; + char **new_gl_pathv; + + new_gl_pathv + = realloc (pglob->gl_pathv, (newcount + 1 + 1) * sizeof (char *)); + if (new_gl_pathv == NULL) + { + nospace: + free (pglob->gl_pathv); + pglob->gl_pathv = NULL; + pglob->gl_pathc = 0; + return GLOB_NOSPACE; + } + pglob->gl_pathv = new_gl_pathv; + + pglob->gl_pathv[newcount] = strdup (dirname); + if (pglob->gl_pathv[newcount] == NULL) + goto nospace; + pglob->gl_pathv[++newcount] = NULL; + ++pglob->gl_pathc; + pglob->gl_flags = flags; + + return 0; + } + + /* Not found. */ + return GLOB_NOMATCH; + } + + if (__glob_pattern_p (dirname, !(flags & GLOB_NOESCAPE))) + { + /* The directory name contains metacharacters, so we + have to glob for the directory, and then glob for + the pattern in each directory found. */ + glob_t dirs; + size_t i; + + if ((flags & GLOB_ALTDIRFUNC) != 0) + { + /* Use the alternative access functions also in the recursive + call. */ + dirs.gl_opendir = pglob->gl_opendir; + dirs.gl_readdir = pglob->gl_readdir; + dirs.gl_closedir = pglob->gl_closedir; + dirs.gl_stat = pglob->gl_stat; + dirs.gl_lstat = pglob->gl_lstat; + } + + status = glob (dirname, + ((flags & (GLOB_ERR | GLOB_NOCHECK | GLOB_NOESCAPE + | GLOB_ALTDIRFUNC)) + | GLOB_NOSORT | GLOB_ONLYDIR), + errfunc, &dirs); + if (status != 0) + return status; + + /* We have successfully globbed the preceding directory name. + For each name we found, call glob_in_dir on it and FILENAME, + appending the results to PGLOB. */ + for (i = 0; i < dirs.gl_pathc; ++i) + { + int old_pathc; + +#ifdef SHELL + { + /* Make globbing interruptible in the bash shell. */ + extern int interrupt_state; + + if (interrupt_state) + { + globfree (&dirs); + return GLOB_ABORTED; + } + } +#endif /* SHELL. */ + + old_pathc = pglob->gl_pathc; + status = glob_in_dir (filename, dirs.gl_pathv[i], + ((flags | GLOB_APPEND) + & ~(GLOB_NOCHECK | GLOB_NOMAGIC)), + errfunc, pglob); + if (status == GLOB_NOMATCH) + /* No matches in this directory. Try the next. */ + continue; + + if (status != 0) + { + globfree (&dirs); + globfree (pglob); + pglob->gl_pathc = 0; + return status; + } + + /* Stick the directory on the front of each name. */ + if (prefix_array (dirs.gl_pathv[i], + &pglob->gl_pathv[old_pathc + pglob->gl_offs], + pglob->gl_pathc - old_pathc)) + { + globfree (&dirs); + globfree (pglob); + pglob->gl_pathc = 0; + return GLOB_NOSPACE; + } + } + + flags |= GLOB_MAGCHAR; + + /* We have ignored the GLOB_NOCHECK flag in the `glob_in_dir' calls. + But if we have not found any matching entry and the GLOB_NOCHECK + flag was set we must return the input pattern itself. */ + if (pglob->gl_pathc + pglob->gl_offs == oldcount) + { + /* No matches. */ + if (flags & GLOB_NOCHECK) + { + int newcount = pglob->gl_pathc + pglob->gl_offs; + char **new_gl_pathv; + + new_gl_pathv = realloc (pglob->gl_pathv, + (newcount + 2) * sizeof (char *)); + if (new_gl_pathv == NULL) + { + globfree (&dirs); + return GLOB_NOSPACE; + } + pglob->gl_pathv = new_gl_pathv; + + pglob->gl_pathv[newcount] = strdup (pattern); + if (pglob->gl_pathv[newcount] == NULL) + { + globfree (&dirs); + globfree (pglob); + pglob->gl_pathc = 0; + return GLOB_NOSPACE; + } + + ++pglob->gl_pathc; + ++newcount; + + pglob->gl_pathv[newcount] = NULL; + pglob->gl_flags = flags; + } + else + { + globfree (&dirs); + return GLOB_NOMATCH; + } + } + + globfree (&dirs); + } + else + { + int old_pathc = pglob->gl_pathc; + + status = glob_in_dir (filename, dirname, flags, errfunc, pglob); + if (status != 0) + return status; + + if (dirlen > 0) + { + /* Stick the directory on the front of each name. */ + if (prefix_array (dirname, + &pglob->gl_pathv[old_pathc + pglob->gl_offs], + pglob->gl_pathc - old_pathc)) + { + globfree (pglob); + pglob->gl_pathc = 0; + return GLOB_NOSPACE; + } + } + } + + if (!(flags & GLOB_NOSORT)) + { + /* Sort the vector. */ + qsort (&pglob->gl_pathv[oldcount], + pglob->gl_pathc + pglob->gl_offs - oldcount, + sizeof (char *), collated_compare); + } + + return 0; +} +#if defined _LIBC && !defined glob +libc_hidden_def (glob) +#endif + + +#if !defined _LIBC || !defined GLOB_ONLY_P + +/* Free storage allocated in PGLOB by a previous `glob' call. */ +void +globfree (pglob) + register glob_t *pglob; +{ + if (pglob->gl_pathv != NULL) + { + size_t i; + for (i = 0; i < pglob->gl_pathc; ++i) + if (pglob->gl_pathv[pglob->gl_offs + i] != NULL) + free (pglob->gl_pathv[pglob->gl_offs + i]); + free (pglob->gl_pathv); + pglob->gl_pathv = NULL; + } +} +#if defined _LIBC && !defined globfree +libc_hidden_def (globfree) +#endif + + +/* Do a collated comparison of A and B. */ +static int +collated_compare (const void *a, const void *b) +{ + const char *const s1 = *(const char *const * const) a; + const char *const s2 = *(const char *const * const) b; + + if (s1 == s2) + return 0; + if (s1 == NULL) + return 1; + if (s2 == NULL) + return -1; + return strcoll (s1, s2); +} + + +/* Prepend DIRNAME to each of N members of ARRAY, replacing ARRAY's + elements in place. Return nonzero if out of memory, zero if successful. + A slash is inserted between DIRNAME and each elt of ARRAY, + unless DIRNAME is just "/". Each old element of ARRAY is freed. */ +static int +prefix_array (const char *dirname, char **array, size_t n) +{ + register size_t i; + size_t dirlen = strlen (dirname); +#if defined __MSDOS__ || defined WINDOWS32 + int sep_char = '/'; +# define DIRSEP_CHAR sep_char +#else +# define DIRSEP_CHAR '/' +#endif + + if (dirlen == 1 && dirname[0] == '/') + /* DIRNAME is just "/", so normal prepending would get us "//foo". + We want "/foo" instead, so don't prepend any chars from DIRNAME. */ + dirlen = 0; +#if defined __MSDOS__ || defined WINDOWS32 + else if (dirlen > 1) + { + if (dirname[dirlen - 1] == '/' && dirname[dirlen - 2] == ':') + /* DIRNAME is "d:/". Don't prepend the slash from DIRNAME. */ + --dirlen; + else if (dirname[dirlen - 1] == ':') + { + /* DIRNAME is "d:". Use `:' instead of `/'. */ + --dirlen; + sep_char = ':'; + } + } +#endif + + for (i = 0; i < n; ++i) + { + size_t eltlen = strlen (array[i]) + 1; + char *new = malloc (dirlen + 1 + eltlen); + if (new == NULL) + { + while (i > 0) + free (array[--i]); + return 1; + } + + { + char *endp = mempcpy (new, dirname, dirlen); + *endp++ = DIRSEP_CHAR; + mempcpy (endp, array[i], eltlen); + } + free (array[i]); + array[i] = new; + } + + return 0; +} + + +/* We must not compile this function twice. */ +#if !defined _LIBC || !defined NO_GLOB_PATTERN_P +/* Return nonzero if PATTERN contains any metacharacters. + Metacharacters can be quoted with backslashes if QUOTE is nonzero. */ +int +__glob_pattern_p (pattern, quote) + const char *pattern; + int quote; +{ + register const char *p; + int open = 0; + + for (p = pattern; *p != '\0'; ++p) + switch (*p) + { + case '?': + case '*': + return 1; + + case '\\': + if (quote && p[1] != '\0') + ++p; + break; + + case '[': + open = 1; + break; + + case ']': + if (open) + return 1; + break; + } + + return 0; +} +# ifdef _LIBC +weak_alias (__glob_pattern_p, glob_pattern_p) +# endif +#endif + +#endif /* !GLOB_ONLY_P */ + + +/* We put this in a separate function mainly to allow the memory + allocated with alloca to be recycled. */ +#if !defined _LIBC || !defined GLOB_ONLY_P +static bool +is_dir_p (const char *dir, size_t dirlen, const char *fname, + glob_t *pglob, int flags) +{ + size_t fnamelen = strlen (fname); + char *fullname = __alloca (dirlen + 1 + fnamelen + 1); + struct stat st; + struct_stat64 st64; + + mempcpy (mempcpy (mempcpy (fullname, dir, dirlen), "/", 1), + fname, fnamelen + 1); + + return ((flags & GLOB_ALTDIRFUNC) + ? (*pglob->gl_stat) (fullname, &st) == 0 && S_ISDIR (st.st_mode) + : __stat64 (fullname, &st64) == 0 && S_ISDIR (st64.st_mode)); +} +#endif + + +/* Like `glob', but PATTERN is a final pathname component, + and matches are searched for in DIRECTORY. + The GLOB_NOSORT bit in FLAGS is ignored. No sorting is ever done. + The GLOB_APPEND flag is assumed to be set (always appends). */ +static int +glob_in_dir (const char *pattern, const char *directory, int flags, + int (*errfunc) (const char *, int), + glob_t *pglob) +{ + size_t dirlen = strlen (directory); + void *stream = NULL; + struct globlink + { + struct globlink *next; + char *name; + }; + struct globlink *names = NULL; + size_t nfound; + int meta; + int save; + + meta = __glob_pattern_p (pattern, !(flags & GLOB_NOESCAPE)); + if (meta == 0 && (flags & (GLOB_NOCHECK|GLOB_NOMAGIC))) + { + /* We need not do any tests. The PATTERN contains no meta + characters and we must not return an error therefore the + result will always contain exactly one name. */ + flags |= GLOB_NOCHECK; + nfound = 0; + } + else if (meta == 0 && + ((flags & GLOB_NOESCAPE) || strchr (pattern, '\\') == NULL)) + { + /* Since we use the normal file functions we can also use stat() + to verify the file is there. */ + struct stat st; + struct_stat64 st64; + size_t patlen = strlen (pattern); + char *fullname = __alloca (dirlen + 1 + patlen + 1); + + mempcpy (mempcpy (mempcpy (fullname, directory, dirlen), + "/", 1), + pattern, patlen + 1); + if (((flags & GLOB_ALTDIRFUNC) + ? (*pglob->gl_stat) (fullname, &st) + : __stat64 (fullname, &st64)) == 0) + /* We found this file to be existing. Now tell the rest + of the function to copy this name into the result. */ + flags |= GLOB_NOCHECK; + + nfound = 0; + } + else + { + if (pattern[0] == '\0') + { + /* This is a special case for matching directories like in + "*a/". */ + names = __alloca (sizeof (struct globlink)); + names->name = malloc (1); + if (names->name == NULL) + goto memory_error; + names->name[0] = '\0'; + names->next = NULL; + nfound = 1; + meta = 0; + } + else + { + stream = ((flags & GLOB_ALTDIRFUNC) + ? (*pglob->gl_opendir) (directory) + : opendir (directory)); + if (stream == NULL) + { + if (errno != ENOTDIR + && ((errfunc != NULL && (*errfunc) (directory, errno)) + || (flags & GLOB_ERR))) + return GLOB_ABORTED; + nfound = 0; + meta = 0; + } + else + { + int fnm_flags = ((!(flags & GLOB_PERIOD) ? FNM_PERIOD : 0) + | ((flags & GLOB_NOESCAPE) ? FNM_NOESCAPE : 0) +#if defined _AMIGA || defined VMS + | FNM_CASEFOLD +#endif + ); + nfound = 0; + flags |= GLOB_MAGCHAR; + + while (1) + { + const char *name; + size_t len; +#if defined _LIBC && !defined COMPILE_GLOB64 + struct dirent64 *d; + union + { + struct dirent64 d64; + char room [offsetof (struct dirent64, d_name[0]) + + NAME_MAX + 1]; + } + d64buf; + + if (flags & GLOB_ALTDIRFUNC) + { + struct dirent *d32 = (*pglob->gl_readdir) (stream); + if (d32 != NULL) + { + CONVERT_DIRENT_DIRENT64 (&d64buf.d64, d32); + d = &d64buf.d64; + } + else + d = NULL; + } + else + d = __readdir64 (stream); +#else + struct dirent *d = ((flags & GLOB_ALTDIRFUNC) + ? ((*pglob->gl_readdir) (stream)) + : __readdir (stream)); +#endif + if (d == NULL) + break; + if (! REAL_DIR_ENTRY (d)) + continue; + + /* If we shall match only directories use the information + provided by the dirent call if possible. */ + if ((flags & GLOB_ONLYDIR) && !DIRENT_MIGHT_BE_DIR (d)) + continue; + + name = d->d_name; + + if (fnmatch (pattern, name, fnm_flags) == 0) + { + /* ISDIR will often be incorrectly set to false + when not in GLOB_ONLYDIR || GLOB_MARK mode, but we + don't care. It won't be used and we save the + expensive call to stat. */ + int need_dir_test = + (GLOB_MARK | (DIRENT_MIGHT_BE_SYMLINK (d) + ? GLOB_ONLYDIR : 0)); + bool isdir = (DIRENT_MUST_BE (d, DT_DIR) + || ((flags & need_dir_test) + && is_dir_p (directory, dirlen, name, + pglob, flags))); + + /* In GLOB_ONLYDIR mode, skip non-dirs. */ + if ((flags & GLOB_ONLYDIR) && !isdir) + continue; + + { + struct globlink *new = + __alloca (sizeof (struct globlink)); + char *p; + len = NAMLEN (d); + new->name = + malloc (len + 1 + ((flags & GLOB_MARK) && isdir)); + if (new->name == NULL) + goto memory_error; + p = mempcpy (new->name, name, len); + if ((flags & GLOB_MARK) && isdir) + *p++ = '/'; + *p = '\0'; + new->next = names; + names = new; + ++nfound; + } + } + } + } + } + } + + if (nfound == 0 && (flags & GLOB_NOCHECK)) + { + size_t len = strlen (pattern); + nfound = 1; + names = __alloca (sizeof (struct globlink)); + names->next = NULL; + names->name = malloc (len + 1); + if (names->name == NULL) + goto memory_error; + *((char *) mempcpy (names->name, pattern, len)) = '\0'; + } + + if (nfound != 0) + { + char **new_gl_pathv; + + new_gl_pathv + = realloc (pglob->gl_pathv, + (pglob->gl_pathc + pglob->gl_offs + nfound + 1) + * sizeof (char *)); + if (new_gl_pathv == NULL) + goto memory_error; + pglob->gl_pathv = new_gl_pathv; + + for (; names != NULL; names = names->next) + pglob->gl_pathv[pglob->gl_offs + pglob->gl_pathc++] = names->name; + pglob->gl_pathv[pglob->gl_offs + pglob->gl_pathc] = NULL; + + pglob->gl_flags = flags; + } + + save = errno; + if (stream != NULL) + { + if (flags & GLOB_ALTDIRFUNC) + (*pglob->gl_closedir) (stream); + else + closedir (stream); + } + __set_errno (save); + + return nfound == 0 ? GLOB_NOMATCH : 0; + + memory_error: + { + int save = errno; + if (flags & GLOB_ALTDIRFUNC) + (*pglob->gl_closedir) (stream); + else + closedir (stream); + __set_errno (save); + } + while (names != NULL) + { + if (names->name != NULL) + free (names->name); + names = names->next; + } + return GLOB_NOSPACE; +} diff --git a/lib/glob_.h b/lib/glob_.h new file mode 100644 index 0000000..a1bd70c --- /dev/null +++ b/lib/glob_.h @@ -0,0 +1,65 @@ +/* glob_.h -- Find a path matching a pattern. + + Copyright (C) 2005 Free Software Foundation, Inc. + + Written by Derek Price <derek@ximbiot.com> & Paul Eggert <eggert@CS.UCLA.EDU> + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef GLOB_H +#define GLOB_H 1 + +#ifdef HAVE_SYS_CDEFS_H +# include <sys/cdefs.h> +#endif + +#include <stddef.h> + +#ifndef __BEGIN_DECLS +# define __BEGIN_DECLS +# define __END_DECLS +#endif +#ifndef __THROW +# define __THROW +#endif + +#ifndef __size_t +# define __size_t size_t +#endif +#ifndef __const +# define __const const +#endif +#ifndef __restrict +# define __restrict restrict +#endif +#ifndef __USE_GNU +# define __USE_GNU 1 +#endif + + +#ifndef HAVE_DIRENT_H +# define dirent direct +#endif + +#define glob rpl_glob +#define globfree rpl_globfree +#define glob_pattern_p rpl_glob_pattern_p + +#define __GLOB_GNULIB 1 + +/* Now the standard GNU C Library header should work. */ +#include "glob-libc.h" + +#endif /* GLOB_H */ diff --git a/lib/libcvs.dep b/lib/libcvs.dep new file mode 100644 index 0000000..3330a17 --- /dev/null +++ b/lib/libcvs.dep @@ -0,0 +1,548 @@ +# Microsoft Developer Studio Generated Dependency File, included by libcvs.mak + +.\__fpending.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\__fpending.h"\ + ".\timespec.h"\ + + +.\asnprintf.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\timespec.h"\ + ".\vasnprintf.h"\ + + +.\basename.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\woe32.h"\ + ".\dirname.h"\ + ".\timespec.h"\ + + +".\canon-host.c" : \ + "..\windows-NT\config.h"\ + "..\windows-NT\netdb.h"\ + "..\windows-NT\sys\socket.h"\ + "..\windows-NT\woe32.h"\ + ".\canon-host.h"\ + ".\getaddrinfo.h"\ + ".\strdup.h"\ + ".\timespec.h"\ + + +.\canonicalize.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\stdint.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\canonicalize.h"\ + ".\cycle-check.h"\ + ".\dev-ino.h"\ + ".\filenamecat.h"\ + ".\pathmax.h"\ + ".\stat-macros.h"\ + ".\timespec.h"\ + ".\xalloc.h"\ + ".\xgetcwd.h"\ + ".\xreadlink.h"\ + + +.\closeout.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\woe32.h"\ + ".\__fpending.h"\ + ".\closeout.h"\ + ".\error.h"\ + ".\exitfail.h"\ + ".\gettext.h"\ + ".\quotearg.h"\ + ".\timespec.h"\ + + +".\cycle-check.c" : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\stdint.h"\ + "..\windows-NT\woe32.h"\ + ".\cycle-check.h"\ + ".\dev-ino.h"\ + ".\timespec.h"\ + + +.\dirname.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\woe32.h"\ + ".\dirname.h"\ + ".\timespec.h"\ + ".\xalloc.h"\ + + +".\dup-safer.c" : \ + "..\windows-NT\config.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\timespec.h"\ + ".\unistd-safer.h"\ + + +.\exitfail.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\exit.h"\ + ".\exitfail.h"\ + ".\timespec.h"\ + + +".\fd-safer.c" : \ + "..\windows-NT\config.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\timespec.h"\ + ".\unistd-safer.h"\ + + +.\fncase.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\ndir.h"\ + "..\windows-NT\pwd.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\stdint.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\fnmatch.h"\ + ".\getopt.h"\ + ".\gettext.h"\ + ".\system.h"\ + ".\timespec.h"\ + + +.\fnmatch.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\woe32.h"\ + ".\alloca.h"\ + ".\fnmatch.h"\ + ".\fnmatch_loop.c"\ + ".\timespec.h"\ + + +.\fseeko.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\timespec.h"\ + + +.\ftello.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\timespec.h"\ + + +.\gai_strerror.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\netdb.h"\ + "..\windows-NT\sys\socket.h"\ + "..\windows-NT\woe32.h"\ + ".\getaddrinfo.h"\ + ".\gettext.h"\ + ".\timespec.h"\ + + +.\getaddrinfo.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\netdb.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\sys\socket.h"\ + "..\windows-NT\woe32.h"\ + ".\getaddrinfo.h"\ + ".\gettext.h"\ + ".\strdup.h"\ + ".\timespec.h"\ + + +.\getdate.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\woe32.h"\ + ".\getdate.h"\ + ".\setenv.h"\ + ".\timespec.h"\ + ".\xalloc.h"\ + + +.\getdelim.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\getdelim.h"\ + ".\timespec.h"\ + + +.\getline.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\getdelim.h"\ + ".\getline.h"\ + ".\timespec.h"\ + + +.\getlogin_r.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\getlogin_r.h"\ + ".\timespec.h"\ + + +.\getndelim2.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdint.h"\ + "..\windows-NT\woe32.h"\ + ".\getndelim2.h"\ + ".\timespec.h"\ + + +.\getopt.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\getopt.h"\ + ".\getopt_int.h"\ + ".\gettext.h"\ + ".\timespec.h"\ + + +.\getopt1.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\getopt.h"\ + ".\getopt_int.h"\ + ".\timespec.h"\ + + +.\gettime.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\timespec.h"\ + + +.\glob.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\ndir.h"\ + "..\windows-NT\pwd.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\alloca.h"\ + ".\fnmatch.h"\ + ".\getlogin_r.h"\ + ".\glob-libc.h"\ + ".\glob.h"\ + ".\mempcpy.h"\ + ".\stat-macros.h"\ + ".\strdup.h"\ + ".\timespec.h"\ + + +.\lstat.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\stat-macros.h"\ + ".\timespec.h"\ + ".\xalloc.h"\ + + +.\mbchar.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\woe32.h"\ + ".\mbchar.h"\ + ".\timespec.h"\ + + +.\md5.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdint.h"\ + "..\windows-NT\woe32.h"\ + ".\md5.h"\ + ".\timespec.h"\ + + +.\mempcpy.c : \ + ".\mempcpy.h"\ + + +.\mkstemp.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\timespec.h"\ + + +.\pagealign_alloc.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\error.h"\ + ".\exit.h"\ + ".\getpagesize.h"\ + ".\gettext.h"\ + ".\pagealign_alloc.h"\ + ".\timespec.h"\ + ".\xalloc.h"\ + + +".\printf-args.c" : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\printf-args.h"\ + ".\timespec.h"\ + + +".\printf-parse.c" : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdint.h"\ + "..\windows-NT\woe32.h"\ + ".\printf-args.h"\ + ".\printf-parse.h"\ + ".\timespec.h"\ + ".\xsize.h"\ + + +.\quotearg.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\woe32.h"\ + ".\gettext.h"\ + ".\quotearg.h"\ + ".\timespec.h"\ + ".\xalloc.h"\ + + +.\readlink.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\timespec.h"\ + + +.\realloc.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\timespec.h"\ + + +.\regex.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\woe32.h"\ + ".\alloca.h"\ + ".\regcomp.c"\ + ".\regex.h"\ + ".\regex_internal.c"\ + ".\regex_internal.h"\ + ".\regexec.c"\ + ".\timespec.h"\ + + +.\rpmatch.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\timespec.h"\ + + +".\save-cwd.c" : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\chdir-long.h"\ + ".\save-cwd.h"\ + ".\timespec.h"\ + ".\unistd-safer.h"\ + ".\xgetcwd.h"\ + + +.\setenv.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\alloca.h"\ + ".\allocsa.h"\ + ".\timespec.h"\ + + +.\sighandle.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\ndir.h"\ + "..\windows-NT\pwd.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\stdint.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\fnmatch.h"\ + ".\getopt.h"\ + ".\gettext.h"\ + ".\system.h"\ + ".\timespec.h"\ + + +.\strcasecmp.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\woe32.h"\ + ".\mbchar.h"\ + ".\mbuiter.h"\ + ".\strcase.h"\ + ".\strnlen1.h"\ + ".\timespec.h"\ + + +.\strftime.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\woe32.h"\ + ".\time_r.h"\ + ".\timespec.h"\ + + +.\stripslash.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\woe32.h"\ + ".\dirname.h"\ + ".\timespec.h"\ + + +.\strnlen1.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\strnlen1.h"\ + ".\timespec.h"\ + + +.\tempname.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdint.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\stat-macros.h"\ + ".\timespec.h"\ + + +.\time_r.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\time_r.h"\ + ".\timespec.h"\ + + +.\unsetenv.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\timespec.h"\ + + +.\vasnprintf.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdint.h"\ + "..\windows-NT\woe32.h"\ + ".\alloca.h"\ + ".\printf-args.h"\ + ".\printf-parse.h"\ + ".\timespec.h"\ + ".\vasnprintf.h"\ + ".\xsize.h"\ + + +.\vasprintf.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\timespec.h"\ + ".\vasnprintf.h"\ + ".\vasprintf.h"\ + + +".\xalloc-die.c" : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\error.h"\ + ".\exitfail.h"\ + ".\gettext.h"\ + ".\timespec.h"\ + ".\xalloc.h"\ + + +.\xgetcwd.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\getcwd.h"\ + ".\timespec.h"\ + ".\xalloc.h"\ + ".\xgetcwd.h"\ + + +.\xgethostname.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\timespec.h"\ + ".\xalloc.h"\ + ".\xgethostname.h"\ + + +.\xmalloc.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\woe32.h"\ + ".\timespec.h"\ + ".\xalloc.h"\ + + +.\xreadlink.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\unistd.h"\ + "..\windows-NT\woe32.h"\ + ".\timespec.h"\ + ".\xalloc.h"\ + ".\xreadlink.h"\ + + +.\yesno.c : \ + "..\windows-NT\config.h"\ + "..\windows-NT\stdbool.h"\ + "..\windows-NT\woe32.h"\ + ".\getline.h"\ + ".\timespec.h"\ + ".\yesno.h"\ + + +!IF "$(CFG)" == "libcvs - Win32 Release" + +!ELSEIF "$(CFG)" == "libcvs - Win32 Debug" + +!ENDIF + +!IF "$(CFG)" == "libcvs - Win32 Release" + +!ELSEIF "$(CFG)" == "libcvs - Win32 Debug" + +!ENDIF + +!IF "$(CFG)" == "libcvs - Win32 Release" + +!ELSEIF "$(CFG)" == "libcvs - Win32 Debug" + +!ENDIF + +!IF "$(CFG)" == "libcvs - Win32 Release" + +!ELSEIF "$(CFG)" == "libcvs - Win32 Debug" + +!ENDIF + diff --git a/lib/libcvs.dsp b/lib/libcvs.dsp new file mode 100644 index 0000000..b8565bc --- /dev/null +++ b/lib/libcvs.dsp @@ -0,0 +1,704 @@ +# Microsoft Developer Studio Project File - Name="libcvs" - Package Owner=<4> +# Microsoft Developer Studio Generated Build File, Format Version 6.00 +# ** DO NOT EDIT ** + +# TARGTYPE "Win32 (x86) Static Library" 0x0104 + +CFG=libcvs - Win32 Debug +!MESSAGE This is not a valid makefile. To build this project using NMAKE, +!MESSAGE use the Export Makefile command and run +!MESSAGE +!MESSAGE NMAKE /f "libcvs.mak". +!MESSAGE +!MESSAGE You can specify a configuration when running NMAKE +!MESSAGE by defining the macro CFG on the command line. For example: +!MESSAGE +!MESSAGE NMAKE /f "libcvs.mak" CFG="libcvs - Win32 Debug" +!MESSAGE +!MESSAGE Possible choices for configuration are: +!MESSAGE +!MESSAGE "libcvs - Win32 Release" (based on "Win32 (x86) Static Library") +!MESSAGE "libcvs - Win32 Debug" (based on "Win32 (x86) Static Library") +!MESSAGE + +# Begin Project +# PROP AllowPerConfigDependencies 0 +# PROP Scc_ProjName "" +# PROP Scc_LocalPath "" +CPP=cl.exe +RSC=rc.exe + +!IF "$(CFG)" == "libcvs - Win32 Release" + +# PROP BASE Use_MFC 0 +# PROP BASE Use_Debug_Libraries 0 +# PROP BASE Output_Dir "WinRel" +# PROP BASE Intermediate_Dir "WinRel" +# PROP BASE Target_Dir "" +# PROP Use_MFC 0 +# PROP Use_Debug_Libraries 0 +# PROP Output_Dir "WinRel" +# PROP Intermediate_Dir "WinRel" +# PROP Target_Dir "" +# ADD BASE CPP /nologo /W3 /GX /O2 /D "WIN32" /D "NDEBUG" /D "_MBCS" /D "_LIB" /YX /FD /c +# ADD CPP /nologo /W3 /GX /O2 /I "..\windows-NT" /I "." /D "WIN32" /D "NDEBUG" /D "_MBCS" /D "_LIB" /D "HAVE_CONFIG_H" /YX /FD /c +# ADD BASE RSC /l 0x409 /d "NDEBUG" +# ADD RSC /l 0x409 /d "NDEBUG" +BSC32=bscmake.exe +# ADD BASE BSC32 /nologo +# ADD BSC32 /nologo +LIB32=link.exe -lib +# ADD BASE LIB32 /nologo +# ADD LIB32 /nologo + +!ELSEIF "$(CFG)" == "libcvs - Win32 Debug" + +# PROP BASE Use_MFC 0 +# PROP BASE Use_Debug_Libraries 1 +# PROP BASE Output_Dir "WinDebug" +# PROP BASE Intermediate_Dir "WinDebug" +# PROP BASE Target_Dir "" +# PROP Use_MFC 0 +# PROP Use_Debug_Libraries 1 +# PROP Output_Dir "WinDebug" +# PROP Intermediate_Dir "WinDebug" +# PROP Target_Dir "" +# ADD BASE CPP /nologo /W3 /Gm /GX /ZI /Od /D "WIN32" /D "_DEBUG" /D "_MBCS" /D "_LIB" /YX /FD /GZ /c +# ADD CPP /nologo /W3 /Gm /GX /ZI /Od /I "..\windows-NT" /I "." /D "WIN32" /D "_DEBUG" /D "_MBCS" /D "_LIB" /D "HAVE_CONFIG_H" /YX /FD /GZ /c +# ADD BASE RSC /l 0x409 /d "_DEBUG" +# ADD RSC /l 0x409 /d "_DEBUG" +BSC32=bscmake.exe +# ADD BASE BSC32 /nologo +# ADD BSC32 /nologo +LIB32=link.exe -lib +# ADD BASE LIB32 /nologo +# ADD LIB32 /nologo + +!ENDIF + +# Begin Target + +# Name "libcvs - Win32 Release" +# Name "libcvs - Win32 Debug" +# Begin Group "Source Files" + +# PROP Default_Filter "cpp;c;cxx;rc;def;r;odl;idl;hpj;bat" +# Begin Source File + +SOURCE=.\__fpending.c +# End Source File +# Begin Source File + +SOURCE=.\asnprintf.c +# End Source File +# Begin Source File + +SOURCE=.\basename.c +# End Source File +# Begin Source File + +SOURCE=".\canon-host.c" +# End Source File +# Begin Source File + +SOURCE=.\canonicalize.c +# End Source File +# Begin Source File + +SOURCE=.\closeout.c +# End Source File +# Begin Source File + +SOURCE=".\cycle-check.c" +# End Source File +# Begin Source File + +SOURCE=.\dirname.c +# End Source File +# Begin Source File + +SOURCE=".\dup-safer.c" +# End Source File +# Begin Source File + +SOURCE=.\exitfail.c +# End Source File +# Begin Source File + +SOURCE=".\fd-safer.c" +# End Source File +# Begin Source File + +SOURCE=.\fncase.c +# End Source File +# Begin Source File + +SOURCE=.\fnmatch.c +# End Source File +# Begin Source File + +SOURCE=.\fseeko.c +# End Source File +# Begin Source File + +SOURCE=.\ftello.c +# End Source File +# Begin Source File + +SOURCE=.\gai_strerror.c +# End Source File +# Begin Source File + +SOURCE=.\getaddrinfo.c +# End Source File +# Begin Source File + +SOURCE=.\getdate.c +# End Source File +# Begin Source File + +SOURCE=.\getdelim.c +# End Source File +# Begin Source File + +SOURCE=.\getline.c +# End Source File +# Begin Source File + +SOURCE=.\getlogin_r.c +# End Source File +# Begin Source File + +SOURCE=.\getndelim2.c +# End Source File +# Begin Source File + +SOURCE=.\getopt.c +# End Source File +# Begin Source File + +SOURCE=.\getopt1.c +# End Source File +# Begin Source File + +SOURCE=.\gettime.c +# End Source File +# Begin Source File + +SOURCE=.\glob.c +# End Source File +# Begin Source File + +SOURCE=.\lstat.c +# End Source File +# Begin Source File + +SOURCE=.\mbchar.c +# End Source File +# Begin Source File + +SOURCE=.\md5.c +# End Source File +# Begin Source File + +SOURCE=.\mempcpy.c +# End Source File +# Begin Source File + +SOURCE=.\mkstemp.c +# End Source File +# Begin Source File + +SOURCE=.\pagealign_alloc.c +# End Source File +# Begin Source File + +SOURCE=".\printf-args.c" +# End Source File +# Begin Source File + +SOURCE=".\printf-parse.c" +# End Source File +# Begin Source File + +SOURCE=.\quotearg.c +# End Source File +# Begin Source File + +SOURCE=.\readlink.c +# End Source File +# Begin Source File + +SOURCE=.\realloc.c +# End Source File +# Begin Source File + +SOURCE=.\regex.c +# End Source File +# Begin Source File + +SOURCE=.\rpmatch.c +# End Source File +# Begin Source File + +SOURCE=".\save-cwd.c" +# End Source File +# Begin Source File + +SOURCE=.\setenv.c +# End Source File +# Begin Source File + +SOURCE=.\sighandle.c +# End Source File +# Begin Source File + +SOURCE=.\strcasecmp.c +# End Source File +# Begin Source File + +SOURCE=.\strftime.c +# End Source File +# Begin Source File + +SOURCE=.\stripslash.c +# End Source File +# Begin Source File + +SOURCE=.\strnlen1.c +# End Source File +# Begin Source File + +SOURCE=.\tempname.c +# End Source File +# Begin Source File + +SOURCE=.\time_r.c +# End Source File +# Begin Source File + +SOURCE=.\unsetenv.c +# End Source File +# Begin Source File + +SOURCE=.\vasnprintf.c +# End Source File +# Begin Source File + +SOURCE=.\vasprintf.c +# End Source File +# Begin Source File + +SOURCE=".\xalloc-die.c" +# End Source File +# Begin Source File + +SOURCE=.\xgetcwd.c +# End Source File +# Begin Source File + +SOURCE=.\xgethostname.c +# End Source File +# Begin Source File + +SOURCE=.\xmalloc.c +# End Source File +# Begin Source File + +SOURCE=.\xreadlink.c +# End Source File +# Begin Source File + +SOURCE=.\yesno.c +# End Source File +# End Group +# Begin Group "Header Files" + +# PROP Default_Filter "h;hpp;hxx;hm;inl" +# Begin Source File + +SOURCE=.\__fpending.h +# End Source File +# Begin Source File + +SOURCE=.\alloca.h +# End Source File +# Begin Source File + +SOURCE=.\alloca_.h + +!IF "$(CFG)" == "libcvs - Win32 Release" + +# Begin Custom Build +InputPath=.\alloca_.h + +".\alloca.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)" + copy .\alloca_.h .\alloca.h + +# End Custom Build + +!ELSEIF "$(CFG)" == "libcvs - Win32 Debug" + +# Begin Custom Build +InputPath=.\alloca_.h + +".\alloca.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)" + copy .\alloca_.h .\alloca.h + +# End Custom Build + +!ENDIF + +# End Source File +# Begin Source File + +SOURCE=.\allocsa.h +# End Source File +# Begin Source File + +SOURCE=".\canon-host.h" +# End Source File +# Begin Source File + +SOURCE=.\canonicalize.h +# End Source File +# Begin Source File + +SOURCE=".\chdir-long.h" +# End Source File +# Begin Source File + +SOURCE=.\closeout.h +# End Source File +# Begin Source File + +SOURCE="..\windows-NT\config.h" +# End Source File +# Begin Source File + +SOURCE=".\cycle-check.h" +# End Source File +# Begin Source File + +SOURCE=".\dev-ino.h" +# End Source File +# Begin Source File + +SOURCE=.\dirname.h +# End Source File +# Begin Source File + +SOURCE=.\error.h +# End Source File +# Begin Source File + +SOURCE=.\exit.h +# End Source File +# Begin Source File + +SOURCE=.\exitfail.h +# End Source File +# Begin Source File + +SOURCE=.\filenamecat.h +# End Source File +# Begin Source File + +SOURCE=.\fnmatch.h +# End Source File +# Begin Source File + +SOURCE=.\fnmatch_.h + +!IF "$(CFG)" == "libcvs - Win32 Release" + +# Begin Custom Build +InputPath=.\fnmatch_.h + +".\fnmatch.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)" + copy .\fnmatch_.h .\fnmatch.h + +# End Custom Build + +!ELSEIF "$(CFG)" == "libcvs - Win32 Debug" + +# Begin Custom Build +InputPath=.\fnmatch_.h + +".\fnmatch.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)" + copy .\fnmatch_.h .\fnmatch.h + +# End Custom Build + +!ENDIF + +# End Source File +# Begin Source File + +SOURCE=.\getaddrinfo.h +# End Source File +# Begin Source File + +SOURCE=.\getcwd.h +# End Source File +# Begin Source File + +SOURCE=.\getdate.h +# End Source File +# Begin Source File + +SOURCE=.\getdelim.h +# End Source File +# Begin Source File + +SOURCE=.\getline.h +# End Source File +# Begin Source File + +SOURCE=.\getlogin_r.h +# End Source File +# Begin Source File + +SOURCE=.\getndelim2.h +# End Source File +# Begin Source File + +SOURCE=.\getopt.h +# End Source File +# Begin Source File + +SOURCE=.\getopt_.h + +!IF "$(CFG)" == "libcvs - Win32 Release" + +# Begin Custom Build +InputPath=.\getopt_.h + +".\getopt.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)" + copy .\getopt_.h .\getopt.h + +# End Custom Build + +!ELSEIF "$(CFG)" == "libcvs - Win32 Debug" + +# Begin Custom Build +InputPath=.\getopt_.h + +".\getopt.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)" + copy .\getopt_.h .\getopt.h + +# End Custom Build + +!ENDIF + +# End Source File +# Begin Source File + +SOURCE=.\getopt_int.h +# End Source File +# Begin Source File + +SOURCE=.\getpagesize.h +# End Source File +# Begin Source File + +SOURCE=.\gettext.h +# End Source File +# Begin Source File + +SOURCE=".\glob-libc.h" +# End Source File +# Begin Source File + +SOURCE=.\glob.h +# End Source File +# Begin Source File + +SOURCE=.\glob_.h + +!IF "$(CFG)" == "libcvs - Win32 Release" + +# Begin Custom Build +InputPath=.\glob_.h + +".\glob.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)" + copy .\glob_.h .\glob.h + +# End Custom Build + +!ELSEIF "$(CFG)" == "libcvs - Win32 Debug" + +# Begin Custom Build +InputPath=.\glob_.h + +".\glob.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)" + copy .\glob_.h .\glob.h + +# End Custom Build + +!ENDIF + +# End Source File +# Begin Source File + +SOURCE=.\mbchar.h +# End Source File +# Begin Source File + +SOURCE=.\mbuiter.h +# End Source File +# Begin Source File + +SOURCE=.\md5.h +# End Source File +# Begin Source File + +SOURCE=.\mempcpy.h +# End Source File +# Begin Source File + +SOURCE="..\windows-NT\ndir.h" +# End Source File +# Begin Source File + +SOURCE="..\windows-NT\netdb.h" +# End Source File +# Begin Source File + +SOURCE=.\pagealign_alloc.h +# End Source File +# Begin Source File + +SOURCE=".\path-concat.h" +# End Source File +# Begin Source File + +SOURCE=.\pathmax.h +# End Source File +# Begin Source File + +SOURCE=".\printf-args.h" +# End Source File +# Begin Source File + +SOURCE=".\printf-parse.h" +# End Source File +# Begin Source File + +SOURCE="..\windows-NT\pwd.h" +# End Source File +# Begin Source File + +SOURCE=.\quotearg.h +# End Source File +# Begin Source File + +SOURCE=.\regex.h +# End Source File +# Begin Source File + +SOURCE=.\regex_internal.h +# End Source File +# Begin Source File + +SOURCE=".\save-cwd.h" +# End Source File +# Begin Source File + +SOURCE=.\setenv.h +# End Source File +# Begin Source File + +SOURCE="..\windows-NT\sys\socket.h" +# End Source File +# Begin Source File + +SOURCE=".\stat-macros.h" +# End Source File +# Begin Source File + +SOURCE="..\windows-NT\stdbool.h" +# End Source File +# Begin Source File + +SOURCE="..\windows-NT\stdint.h" +# End Source File +# Begin Source File + +SOURCE=.\strcase.h +# End Source File +# Begin Source File + +SOURCE=.\strdup.h +# End Source File +# Begin Source File + +SOURCE=.\strnlen1.h +# End Source File +# Begin Source File + +SOURCE=.\system.h +# End Source File +# Begin Source File + +SOURCE=.\time_r.h +# End Source File +# Begin Source File + +SOURCE=.\timespec.h +# End Source File +# Begin Source File + +SOURCE=".\unistd-safer.h" +# End Source File +# Begin Source File + +SOURCE="..\windows-NT\unistd.h" +# End Source File +# Begin Source File + +SOURCE=".\unlocked-io.h" +# End Source File +# Begin Source File + +SOURCE=.\vasnprintf.h +# End Source File +# Begin Source File + +SOURCE=.\vasprintf.h +# End Source File +# Begin Source File + +SOURCE="..\windows-NT\woe32.h" +# End Source File +# Begin Source File + +SOURCE=.\xalloc.h +# End Source File +# Begin Source File + +SOURCE=.\xgetcwd.h +# End Source File +# Begin Source File + +SOURCE=.\xgethostname.h +# End Source File +# Begin Source File + +SOURCE=.\xreadlink.h +# End Source File +# Begin Source File + +SOURCE=.\xsize.h +# End Source File +# Begin Source File + +SOURCE=.\xtime.h +# End Source File +# Begin Source File + +SOURCE=.\yesno.h +# End Source File +# End Group +# End Target +# End Project diff --git a/lib/libcvs.mak b/lib/libcvs.mak new file mode 100644 index 0000000..1b0d532 --- /dev/null +++ b/lib/libcvs.mak @@ -0,0 +1,795 @@ +# Microsoft Developer Studio Generated NMAKE File, Based on libcvs.dsp +!IF "$(CFG)" == "" +CFG=libcvs - Win32 Debug +!MESSAGE No configuration specified. Defaulting to libcvs - Win32 Debug. +!ENDIF + +!IF "$(CFG)" != "libcvs - Win32 Release" && "$(CFG)" != "libcvs - Win32 Debug" +!MESSAGE Invalid configuration "$(CFG)" specified. +!MESSAGE You can specify a configuration when running NMAKE +!MESSAGE by defining the macro CFG on the command line. For example: +!MESSAGE +!MESSAGE NMAKE /f "libcvs.mak" CFG="libcvs - Win32 Debug" +!MESSAGE +!MESSAGE Possible choices for configuration are: +!MESSAGE +!MESSAGE "libcvs - Win32 Release" (based on "Win32 (x86) Static Library") +!MESSAGE "libcvs - Win32 Debug" (based on "Win32 (x86) Static Library") +!MESSAGE +!ERROR An invalid configuration is specified. +!ENDIF + +!IF "$(OS)" == "Windows_NT" +NULL= +!ELSE +NULL=nul +!ENDIF + +!IF "$(CFG)" == "libcvs - Win32 Release" + +OUTDIR=.\WinRel +INTDIR=.\WinRel +# Begin Custom Macros +OutDir=.\WinRel +# End Custom Macros + +ALL : ".\glob.h" ".\getopt.h" ".\fnmatch.h" ".\alloca.h" "$(OUTDIR)\libcvs.lib" + + +CLEAN : + -@erase "$(INTDIR)\__fpending.obj" + -@erase "$(INTDIR)\asnprintf.obj" + -@erase "$(INTDIR)\basename.obj" + -@erase "$(INTDIR)\canon-host.obj" + -@erase "$(INTDIR)\canonicalize.obj" + -@erase "$(INTDIR)\closeout.obj" + -@erase "$(INTDIR)\cycle-check.obj" + -@erase "$(INTDIR)\dirname.obj" + -@erase "$(INTDIR)\dup-safer.obj" + -@erase "$(INTDIR)\exitfail.obj" + -@erase "$(INTDIR)\fd-safer.obj" + -@erase "$(INTDIR)\fncase.obj" + -@erase "$(INTDIR)\fnmatch.obj" + -@erase "$(INTDIR)\fseeko.obj" + -@erase "$(INTDIR)\ftello.obj" + -@erase "$(INTDIR)\gai_strerror.obj" + -@erase "$(INTDIR)\getaddrinfo.obj" + -@erase "$(INTDIR)\getdate.obj" + -@erase "$(INTDIR)\getdelim.obj" + -@erase "$(INTDIR)\getline.obj" + -@erase "$(INTDIR)\getlogin_r.obj" + -@erase "$(INTDIR)\getndelim2.obj" + -@erase "$(INTDIR)\getopt.obj" + -@erase "$(INTDIR)\getopt1.obj" + -@erase "$(INTDIR)\gettime.obj" + -@erase "$(INTDIR)\glob.obj" + -@erase "$(INTDIR)\lstat.obj" + -@erase "$(INTDIR)\mbchar.obj" + -@erase "$(INTDIR)\md5.obj" + -@erase "$(INTDIR)\mempcpy.obj" + -@erase "$(INTDIR)\mkstemp.obj" + -@erase "$(INTDIR)\pagealign_alloc.obj" + -@erase "$(INTDIR)\printf-args.obj" + -@erase "$(INTDIR)\printf-parse.obj" + -@erase "$(INTDIR)\quotearg.obj" + -@erase "$(INTDIR)\readlink.obj" + -@erase "$(INTDIR)\realloc.obj" + -@erase "$(INTDIR)\regex.obj" + -@erase "$(INTDIR)\rpmatch.obj" + -@erase "$(INTDIR)\save-cwd.obj" + -@erase "$(INTDIR)\setenv.obj" + -@erase "$(INTDIR)\sighandle.obj" + -@erase "$(INTDIR)\strcasecmp.obj" + -@erase "$(INTDIR)\strftime.obj" + -@erase "$(INTDIR)\stripslash.obj" + -@erase "$(INTDIR)\strnlen1.obj" + -@erase "$(INTDIR)\tempname.obj" + -@erase "$(INTDIR)\time_r.obj" + -@erase "$(INTDIR)\unsetenv.obj" + -@erase "$(INTDIR)\vasnprintf.obj" + -@erase "$(INTDIR)\vasprintf.obj" + -@erase "$(INTDIR)\vc60.idb" + -@erase "$(INTDIR)\xalloc-die.obj" + -@erase "$(INTDIR)\xgetcwd.obj" + -@erase "$(INTDIR)\xgethostname.obj" + -@erase "$(INTDIR)\xmalloc.obj" + -@erase "$(INTDIR)\xreadlink.obj" + -@erase "$(INTDIR)\yesno.obj" + -@erase "$(OUTDIR)\libcvs.lib" + -@erase ".\alloca.h" + -@erase ".\fnmatch.h" + -@erase ".\getopt.h" + -@erase ".\glob.h" + +"$(OUTDIR)" : + if not exist "$(OUTDIR)/$(NULL)" mkdir "$(OUTDIR)" + +CPP=cl.exe +CPP_PROJ=/nologo /ML /W3 /GX /O2 /I "..\windows-NT" /I "." /D "WIN32" /D "NDEBUG" /D "_MBCS" /D "_LIB" /D "HAVE_CONFIG_H" /Fp"$(INTDIR)\libcvs.pch" /YX /Fo"$(INTDIR)\\" /Fd"$(INTDIR)\\" /FD /c + +.c{$(INTDIR)}.obj:: + $(CPP) @<< + $(CPP_PROJ) $< +<< + +.cpp{$(INTDIR)}.obj:: + $(CPP) @<< + $(CPP_PROJ) $< +<< + +.cxx{$(INTDIR)}.obj:: + $(CPP) @<< + $(CPP_PROJ) $< +<< + +.c{$(INTDIR)}.sbr:: + $(CPP) @<< + $(CPP_PROJ) $< +<< + +.cpp{$(INTDIR)}.sbr:: + $(CPP) @<< + $(CPP_PROJ) $< +<< + +.cxx{$(INTDIR)}.sbr:: + $(CPP) @<< + $(CPP_PROJ) $< +<< + +RSC=rc.exe +BSC32=bscmake.exe +BSC32_FLAGS=/nologo /o"$(OUTDIR)\libcvs.bsc" +BSC32_SBRS= \ + +LIB32=link.exe -lib +LIB32_FLAGS=/nologo /out:"$(OUTDIR)\libcvs.lib" +LIB32_OBJS= \ + "$(INTDIR)\__fpending.obj" \ + "$(INTDIR)\asnprintf.obj" \ + "$(INTDIR)\basename.obj" \ + "$(INTDIR)\canon-host.obj" \ + "$(INTDIR)\canonicalize.obj" \ + "$(INTDIR)\closeout.obj" \ + "$(INTDIR)\cycle-check.obj" \ + "$(INTDIR)\dirname.obj" \ + "$(INTDIR)\dup-safer.obj" \ + "$(INTDIR)\exitfail.obj" \ + "$(INTDIR)\fd-safer.obj" \ + "$(INTDIR)\fncase.obj" \ + "$(INTDIR)\fnmatch.obj" \ + "$(INTDIR)\fseeko.obj" \ + "$(INTDIR)\ftello.obj" \ + "$(INTDIR)\gai_strerror.obj" \ + "$(INTDIR)\getaddrinfo.obj" \ + "$(INTDIR)\getdate.obj" \ + "$(INTDIR)\getdelim.obj" \ + "$(INTDIR)\getline.obj" \ + "$(INTDIR)\getlogin_r.obj" \ + "$(INTDIR)\getndelim2.obj" \ + "$(INTDIR)\getopt.obj" \ + "$(INTDIR)\getopt1.obj" \ + "$(INTDIR)\gettime.obj" \ + "$(INTDIR)\glob.obj" \ + "$(INTDIR)\lstat.obj" \ + "$(INTDIR)\mbchar.obj" \ + "$(INTDIR)\md5.obj" \ + "$(INTDIR)\mempcpy.obj" \ + "$(INTDIR)\mkstemp.obj" \ + "$(INTDIR)\pagealign_alloc.obj" \ + "$(INTDIR)\printf-args.obj" \ + "$(INTDIR)\printf-parse.obj" \ + "$(INTDIR)\quotearg.obj" \ + "$(INTDIR)\readlink.obj" \ + "$(INTDIR)\realloc.obj" \ + "$(INTDIR)\regex.obj" \ + "$(INTDIR)\rpmatch.obj" \ + "$(INTDIR)\save-cwd.obj" \ + "$(INTDIR)\setenv.obj" \ + "$(INTDIR)\sighandle.obj" \ + "$(INTDIR)\strcasecmp.obj" \ + "$(INTDIR)\strftime.obj" \ + "$(INTDIR)\stripslash.obj" \ + "$(INTDIR)\strnlen1.obj" \ + "$(INTDIR)\tempname.obj" \ + "$(INTDIR)\time_r.obj" \ + "$(INTDIR)\unsetenv.obj" \ + "$(INTDIR)\vasnprintf.obj" \ + "$(INTDIR)\vasprintf.obj" \ + "$(INTDIR)\xalloc-die.obj" \ + "$(INTDIR)\xgetcwd.obj" \ + "$(INTDIR)\xgethostname.obj" \ + "$(INTDIR)\xmalloc.obj" \ + "$(INTDIR)\xreadlink.obj" \ + "$(INTDIR)\yesno.obj" + +"$(OUTDIR)\libcvs.lib" : "$(OUTDIR)" $(DEF_FILE) $(LIB32_OBJS) + $(LIB32) @<< + $(LIB32_FLAGS) $(DEF_FLAGS) $(LIB32_OBJS) +<< + +!ELSEIF "$(CFG)" == "libcvs - Win32 Debug" + +OUTDIR=.\WinDebug +INTDIR=.\WinDebug +# Begin Custom Macros +OutDir=.\WinDebug +# End Custom Macros + +ALL : "$(OUTDIR)\libcvs.lib" + + +CLEAN : + -@erase "$(INTDIR)\__fpending.obj" + -@erase "$(INTDIR)\asnprintf.obj" + -@erase "$(INTDIR)\basename.obj" + -@erase "$(INTDIR)\canon-host.obj" + -@erase "$(INTDIR)\canonicalize.obj" + -@erase "$(INTDIR)\closeout.obj" + -@erase "$(INTDIR)\cycle-check.obj" + -@erase "$(INTDIR)\dirname.obj" + -@erase "$(INTDIR)\dup-safer.obj" + -@erase "$(INTDIR)\exitfail.obj" + -@erase "$(INTDIR)\fd-safer.obj" + -@erase "$(INTDIR)\fncase.obj" + -@erase "$(INTDIR)\fnmatch.obj" + -@erase "$(INTDIR)\fseeko.obj" + -@erase "$(INTDIR)\ftello.obj" + -@erase "$(INTDIR)\gai_strerror.obj" + -@erase "$(INTDIR)\getaddrinfo.obj" + -@erase "$(INTDIR)\getdate.obj" + -@erase "$(INTDIR)\getdelim.obj" + -@erase "$(INTDIR)\getline.obj" + -@erase "$(INTDIR)\getlogin_r.obj" + -@erase "$(INTDIR)\getndelim2.obj" + -@erase "$(INTDIR)\getopt.obj" + -@erase "$(INTDIR)\getopt1.obj" + -@erase "$(INTDIR)\gettime.obj" + -@erase "$(INTDIR)\glob.obj" + -@erase "$(INTDIR)\lstat.obj" + -@erase "$(INTDIR)\mbchar.obj" + -@erase "$(INTDIR)\md5.obj" + -@erase "$(INTDIR)\mempcpy.obj" + -@erase "$(INTDIR)\mkstemp.obj" + -@erase "$(INTDIR)\pagealign_alloc.obj" + -@erase "$(INTDIR)\printf-args.obj" + -@erase "$(INTDIR)\printf-parse.obj" + -@erase "$(INTDIR)\quotearg.obj" + -@erase "$(INTDIR)\readlink.obj" + -@erase "$(INTDIR)\realloc.obj" + -@erase "$(INTDIR)\regex.obj" + -@erase "$(INTDIR)\rpmatch.obj" + -@erase "$(INTDIR)\save-cwd.obj" + -@erase "$(INTDIR)\setenv.obj" + -@erase "$(INTDIR)\sighandle.obj" + -@erase "$(INTDIR)\strcasecmp.obj" + -@erase "$(INTDIR)\strftime.obj" + -@erase "$(INTDIR)\stripslash.obj" + -@erase "$(INTDIR)\strnlen1.obj" + -@erase "$(INTDIR)\tempname.obj" + -@erase "$(INTDIR)\time_r.obj" + -@erase "$(INTDIR)\unsetenv.obj" + -@erase "$(INTDIR)\vasnprintf.obj" + -@erase "$(INTDIR)\vasprintf.obj" + -@erase "$(INTDIR)\vc60.idb" + -@erase "$(INTDIR)\vc60.pdb" + -@erase "$(INTDIR)\xalloc-die.obj" + -@erase "$(INTDIR)\xgetcwd.obj" + -@erase "$(INTDIR)\xgethostname.obj" + -@erase "$(INTDIR)\xmalloc.obj" + -@erase "$(INTDIR)\xreadlink.obj" + -@erase "$(INTDIR)\yesno.obj" + -@erase "$(OUTDIR)\libcvs.lib" + +"$(OUTDIR)" : + if not exist "$(OUTDIR)/$(NULL)" mkdir "$(OUTDIR)" + +CPP=cl.exe +CPP_PROJ=/nologo /MLd /W3 /Gm /GX /ZI /Od /I "..\windows-NT" /I "." /D "WIN32" /D "_DEBUG" /D "_MBCS" /D "_LIB" /D "HAVE_CONFIG_H" /Fp"$(INTDIR)\libcvs.pch" /YX /Fo"$(INTDIR)\\" /Fd"$(INTDIR)\\" /FD /GZ /c + +.c{$(INTDIR)}.obj:: + $(CPP) @<< + $(CPP_PROJ) $< +<< + +.cpp{$(INTDIR)}.obj:: + $(CPP) @<< + $(CPP_PROJ) $< +<< + +.cxx{$(INTDIR)}.obj:: + $(CPP) @<< + $(CPP_PROJ) $< +<< + +.c{$(INTDIR)}.sbr:: + $(CPP) @<< + $(CPP_PROJ) $< +<< + +.cpp{$(INTDIR)}.sbr:: + $(CPP) @<< + $(CPP_PROJ) $< +<< + +.cxx{$(INTDIR)}.sbr:: + $(CPP) @<< + $(CPP_PROJ) $< +<< + +RSC=rc.exe +BSC32=bscmake.exe +BSC32_FLAGS=/nologo /o"$(OUTDIR)\libcvs.bsc" +BSC32_SBRS= \ + +LIB32=link.exe -lib +LIB32_FLAGS=/nologo /out:"$(OUTDIR)\libcvs.lib" +LIB32_OBJS= \ + "$(INTDIR)\__fpending.obj" \ + "$(INTDIR)\asnprintf.obj" \ + "$(INTDIR)\basename.obj" \ + "$(INTDIR)\canon-host.obj" \ + "$(INTDIR)\canonicalize.obj" \ + "$(INTDIR)\closeout.obj" \ + "$(INTDIR)\cycle-check.obj" \ + "$(INTDIR)\dirname.obj" \ + "$(INTDIR)\dup-safer.obj" \ + "$(INTDIR)\exitfail.obj" \ + "$(INTDIR)\fd-safer.obj" \ + "$(INTDIR)\fncase.obj" \ + "$(INTDIR)\fnmatch.obj" \ + "$(INTDIR)\fseeko.obj" \ + "$(INTDIR)\ftello.obj" \ + "$(INTDIR)\gai_strerror.obj" \ + "$(INTDIR)\getaddrinfo.obj" \ + "$(INTDIR)\getdate.obj" \ + "$(INTDIR)\getdelim.obj" \ + "$(INTDIR)\getline.obj" \ + "$(INTDIR)\getlogin_r.obj" \ + "$(INTDIR)\getndelim2.obj" \ + "$(INTDIR)\getopt.obj" \ + "$(INTDIR)\getopt1.obj" \ + "$(INTDIR)\gettime.obj" \ + "$(INTDIR)\glob.obj" \ + "$(INTDIR)\lstat.obj" \ + "$(INTDIR)\mbchar.obj" \ + "$(INTDIR)\md5.obj" \ + "$(INTDIR)\mempcpy.obj" \ + "$(INTDIR)\mkstemp.obj" \ + "$(INTDIR)\pagealign_alloc.obj" \ + "$(INTDIR)\printf-args.obj" \ + "$(INTDIR)\printf-parse.obj" \ + "$(INTDIR)\quotearg.obj" \ + "$(INTDIR)\readlink.obj" \ + "$(INTDIR)\realloc.obj" \ + "$(INTDIR)\regex.obj" \ + "$(INTDIR)\rpmatch.obj" \ + "$(INTDIR)\save-cwd.obj" \ + "$(INTDIR)\setenv.obj" \ + "$(INTDIR)\sighandle.obj" \ + "$(INTDIR)\strcasecmp.obj" \ + "$(INTDIR)\strftime.obj" \ + "$(INTDIR)\stripslash.obj" \ + "$(INTDIR)\strnlen1.obj" \ + "$(INTDIR)\tempname.obj" \ + "$(INTDIR)\time_r.obj" \ + "$(INTDIR)\unsetenv.obj" \ + "$(INTDIR)\vasnprintf.obj" \ + "$(INTDIR)\vasprintf.obj" \ + "$(INTDIR)\xalloc-die.obj" \ + "$(INTDIR)\xgetcwd.obj" \ + "$(INTDIR)\xgethostname.obj" \ + "$(INTDIR)\xmalloc.obj" \ + "$(INTDIR)\xreadlink.obj" \ + "$(INTDIR)\yesno.obj" + +"$(OUTDIR)\libcvs.lib" : "$(OUTDIR)" $(DEF_FILE) $(LIB32_OBJS) + $(LIB32) @<< + $(LIB32_FLAGS) $(DEF_FLAGS) $(LIB32_OBJS) +<< + +!ENDIF + + +!IF "$(NO_EXTERNAL_DEPS)" != "1" +!IF EXISTS("libcvs.dep") +!INCLUDE "libcvs.dep" +!ELSE +!MESSAGE Warning: cannot find "libcvs.dep" +!ENDIF +!ENDIF + + +!IF "$(CFG)" == "libcvs - Win32 Release" || "$(CFG)" == "libcvs - Win32 Debug" +SOURCE=.\__fpending.c + +"$(INTDIR)\__fpending.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\asnprintf.c + +"$(INTDIR)\asnprintf.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\basename.c + +"$(INTDIR)\basename.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=".\canon-host.c" + +"$(INTDIR)\canon-host.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\canonicalize.c + +"$(INTDIR)\canonicalize.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\closeout.c + +"$(INTDIR)\closeout.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=".\cycle-check.c" + +"$(INTDIR)\cycle-check.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\dirname.c + +"$(INTDIR)\dirname.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=".\dup-safer.c" + +"$(INTDIR)\dup-safer.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\exitfail.c + +"$(INTDIR)\exitfail.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=".\fd-safer.c" + +"$(INTDIR)\fd-safer.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\fncase.c + +"$(INTDIR)\fncase.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\fnmatch.c + +"$(INTDIR)\fnmatch.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\fseeko.c + +"$(INTDIR)\fseeko.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\ftello.c + +"$(INTDIR)\ftello.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\gai_strerror.c + +"$(INTDIR)\gai_strerror.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\getaddrinfo.c + +"$(INTDIR)\getaddrinfo.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\getdate.c + +"$(INTDIR)\getdate.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\getdelim.c + +"$(INTDIR)\getdelim.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\getline.c + +"$(INTDIR)\getline.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\getlogin_r.c + +"$(INTDIR)\getlogin_r.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\getndelim2.c + +"$(INTDIR)\getndelim2.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\getopt.c + +"$(INTDIR)\getopt.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\getopt1.c + +"$(INTDIR)\getopt1.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\gettime.c + +"$(INTDIR)\gettime.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\glob.c + +"$(INTDIR)\glob.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\lstat.c + +"$(INTDIR)\lstat.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\mbchar.c + +"$(INTDIR)\mbchar.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\md5.c + +"$(INTDIR)\md5.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\mempcpy.c + +"$(INTDIR)\mempcpy.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\mkstemp.c + +"$(INTDIR)\mkstemp.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\pagealign_alloc.c + +"$(INTDIR)\pagealign_alloc.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=".\printf-args.c" + +"$(INTDIR)\printf-args.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=".\printf-parse.c" + +"$(INTDIR)\printf-parse.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\quotearg.c + +"$(INTDIR)\quotearg.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\readlink.c + +"$(INTDIR)\readlink.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\realloc.c + +"$(INTDIR)\realloc.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\regex.c + +"$(INTDIR)\regex.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\rpmatch.c + +"$(INTDIR)\rpmatch.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=".\save-cwd.c" + +"$(INTDIR)\save-cwd.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\setenv.c + +"$(INTDIR)\setenv.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\sighandle.c + +"$(INTDIR)\sighandle.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\strcasecmp.c + +"$(INTDIR)\strcasecmp.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\strftime.c + +"$(INTDIR)\strftime.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\stripslash.c + +"$(INTDIR)\stripslash.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\strnlen1.c + +"$(INTDIR)\strnlen1.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\tempname.c + +"$(INTDIR)\tempname.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\time_r.c + +"$(INTDIR)\time_r.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\unsetenv.c + +"$(INTDIR)\unsetenv.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\vasnprintf.c + +"$(INTDIR)\vasnprintf.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\vasprintf.c + +"$(INTDIR)\vasprintf.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=".\xalloc-die.c" + +"$(INTDIR)\xalloc-die.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\xgetcwd.c + +"$(INTDIR)\xgetcwd.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\xgethostname.c + +"$(INTDIR)\xgethostname.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\xmalloc.c + +"$(INTDIR)\xmalloc.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\xreadlink.c + +"$(INTDIR)\xreadlink.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\yesno.c + +"$(INTDIR)\yesno.obj" : $(SOURCE) "$(INTDIR)" + + +SOURCE=.\alloca_.h + +!IF "$(CFG)" == "libcvs - Win32 Release" + +InputPath=.\alloca_.h + +".\alloca.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)" + <<tempfile.bat + @echo off + copy .\alloca_.h .\alloca.h +<< + + +!ELSEIF "$(CFG)" == "libcvs - Win32 Debug" + +InputPath=.\alloca_.h + +".\alloca.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)" + <<tempfile.bat + @echo off + copy .\alloca_.h .\alloca.h +<< + + +!ENDIF + +SOURCE=.\fnmatch_.h + +!IF "$(CFG)" == "libcvs - Win32 Release" + +InputPath=.\fnmatch_.h + +".\fnmatch.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)" + <<tempfile.bat + @echo off + copy .\fnmatch_.h .\fnmatch.h +<< + + +!ELSEIF "$(CFG)" == "libcvs - Win32 Debug" + +InputPath=.\fnmatch_.h + +".\fnmatch.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)" + <<tempfile.bat + @echo off + copy .\fnmatch_.h .\fnmatch.h +<< + + +!ENDIF + +SOURCE=.\getopt_.h + +!IF "$(CFG)" == "libcvs - Win32 Release" + +InputPath=.\getopt_.h + +".\getopt.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)" + <<tempfile.bat + @echo off + copy .\getopt_.h .\getopt.h +<< + + +!ELSEIF "$(CFG)" == "libcvs - Win32 Debug" + +InputPath=.\getopt_.h + +".\getopt.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)" + <<tempfile.bat + @echo off + copy .\getopt_.h .\getopt.h +<< + + +!ENDIF + +SOURCE=.\glob_.h + +!IF "$(CFG)" == "libcvs - Win32 Release" + +InputPath=.\glob_.h + +".\glob.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)" + <<tempfile.bat + @echo off + copy .\glob_.h .\glob.h +<< + + +!ELSEIF "$(CFG)" == "libcvs - Win32 Debug" + +InputPath=.\glob_.h + +".\glob.h" : $(SOURCE) "$(INTDIR)" "$(OUTDIR)" + <<tempfile.bat + @echo off + copy .\glob_.h .\glob.h +<< + + +!ENDIF + + +!ENDIF + diff --git a/lib/lstat.c b/lib/lstat.c new file mode 100644 index 0000000..f5a22b0 --- /dev/null +++ b/lib/lstat.c @@ -0,0 +1,77 @@ +/* Work around a bug of lstat on some systems + + Copyright (C) 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005 Free + Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* written by Jim Meyering */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* The specification of these functions is in sys_stat.h. But we cannot + include this include file here, because on some systems, a + "#define lstat lstat64" is being used, and sys_stat.h deletes this + definition. */ + +#include <sys/types.h> +#include <sys/stat.h> +#include <stdlib.h> +#include <string.h> + +#include "stat-macros.h" +#include "xalloc.h" + +/* lstat works differently on Linux and Solaris systems. POSIX (see + `pathname resolution' in the glossary) requires that programs like `ls' + take into consideration the fact that FILE has a trailing slash when + FILE is a symbolic link. On Linux systems, the lstat function already + has the desired semantics (in treating `lstat("symlink/",sbuf)' just like + `lstat("symlink/.",sbuf)', but on Solaris it does not. + + If FILE has a trailing slash and specifies a symbolic link, + then append a `.' to FILE and call lstat a second time. */ + +int +rpl_lstat (const char *file, struct stat *sbuf) +{ + size_t len; + char *new_file; + + int lstat_result = lstat (file, sbuf); + + if (lstat_result != 0 || !S_ISLNK (sbuf->st_mode)) + return lstat_result; + + len = strlen (file); + if (len == 0 || file[len - 1] != '/') + return lstat_result; + + /* FILE refers to a symbolic link and the name ends with a slash. + Append a `.' to FILE and repeat the lstat call. */ + + /* Add one for the `.' we'll append, and one more for the trailing NUL. */ + new_file = xmalloc (len + 1 + 1); + memcpy (new_file, file, len); + new_file[len] = '.'; + new_file[len + 1] = 0; + + lstat_result = lstat (new_file, sbuf); + free (new_file); + + return lstat_result; +} diff --git a/lib/lstat.h b/lib/lstat.h new file mode 100644 index 0000000..6a83fbf --- /dev/null +++ b/lib/lstat.h @@ -0,0 +1,24 @@ +/* Retrieving information about files. + Copyright (C) 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#include <sys/stat.h> + +#if !LSTAT_FOLLOWS_SLASHED_SYMLINK +extern int rpl_lstat (const char *name, struct stat *buf); +# undef lstat +# define lstat rpl_lstat +#endif diff --git a/lib/malloc.c b/lib/malloc.c new file mode 100644 index 0000000..58fa611 --- /dev/null +++ b/lib/malloc.c @@ -0,0 +1,36 @@ +/* malloc() function that is glibc compatible. + Copyright (C) 1997, 1998 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* written by Jim Meyering */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif +#undef malloc + +#include <stdlib.h> + +/* Allocate an N-byte block of memory from the heap. + If N is zero, allocate a 1-byte block. */ + +void * +rpl_malloc (size_t n) +{ + if (n == 0) + n = 1; + return malloc (n); +} diff --git a/lib/mbchar.c b/lib/mbchar.c new file mode 100644 index 0000000..6facc14 --- /dev/null +++ b/lib/mbchar.c @@ -0,0 +1,38 @@ +/* Copyright (C) 2001 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ + + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include <limits.h> + +#include "mbchar.h" + +#if IS_BASIC_ASCII + +/* Bit table of characters in the ISO C "basic character set". */ +unsigned int is_basic_table [UCHAR_MAX / 32 + 1] = +{ + 0x00001a00, /* '\t' '\v' '\f' */ + 0xffffffef, /* ' '...'#' '%'...'?' */ + 0xfffffffe, /* 'A'...'Z' '[' '\\' ']' '^' '_' */ + 0x7ffffffe /* 'a'...'z' '{' '|' '}' '~' */ + /* The remaining bits are 0. */ +}; + +#endif /* IS_BASIC_ASCII */ diff --git a/lib/mbchar.h b/lib/mbchar.h new file mode 100644 index 0000000..4472aa0 --- /dev/null +++ b/lib/mbchar.h @@ -0,0 +1,352 @@ +/* Multibyte character data type. + Copyright (C) 2001, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ + +/* Written by Bruno Haible <bruno@clisp.org>. */ + +/* A multibyte character is a short subsequence of a char* string, + representing a single wide character. + + We use multibyte characters instead of wide characters because of + the following goals: + 1) correct multibyte handling, i.e. operate according to the LC_CTYPE + locale, + 2) ease of maintenance, i.e. the maintainer needs not know all details + of the ISO C 99 standard, + 3) don't fail grossly if the input is not in the encoding set by the + locale, because often different encodings are in use in the same + countries (ISO-8859-1/UTF-8, EUC-JP/Shift_JIS, ...), + 4) fast in the case of ASCII characters, + 5) portability, i.e. don't make unportable assumptions about wchar_t. + + Multibyte characters are only accessed through the mb* macros. + + mb_ptr (mbc) + return a pointer to the beginning of the multibyte sequence. + + mb_len (mbc) + returns the number of bytes occupied by the multibyte sequence. + Always > 0. + + mb_iseq (mbc, sc) + returns true if mbc is the standard ASCII character sc. + + mb_isnul (mbc) + returns true if mbc is the nul character. + + mb_cmp (mbc1, mbc2) + returns a positive, zero, or negative value depending on whether mbc1 + sorts after, same or before mbc2. + + mb_casecmp (mbc1, mbc2) + returns a positive, zero, or negative value depending on whether mbc1 + sorts after, same or before mbc2, modulo upper/lowercase conversion. + + mb_equal (mbc1, mbc2) + returns true if mbc1 and mbc2 are equal. + + mb_caseequal (mbc1, mbc2) + returns true if mbc1 and mbc2 are equal modulo upper/lowercase conversion. + + mb_isalnum (mbc) + returns true if mbc is alphanumeric. + + mb_isalpha (mbc) + returns true if mbc is alphabetic. + + mb_isascii(mbc) + returns true if mbc is plain ASCII. + + mb_isblank (mbc) + returns true if mbc is a blank. + + mb_iscntrl (mbc) + returns true if mbc is a control character. + + mb_isdigit (mbc) + returns true if mbc is a decimal digit. + + mb_isgraph (mbc) + returns true if mbc is a graphic character. + + mb_islower (mbc) + returns true if mbc is lowercase. + + mb_isprint (mbc) + returns true if mbc is a printable character. + + mb_ispunct (mbc) + returns true if mbc is a punctuation character. + + mb_isspace (mbc) + returns true if mbc is a space character. + + mb_isupper (mbc) + returns true if mbc is uppercase. + + mb_isxdigit (mbc) + returns true if mbc is a hexadecimal digit. + + mb_width (mbc) + returns the number of columns on the output device occupied by mbc. + Always >= 0. + + mb_putc (mbc, stream) + outputs mbc on stream, a byte oriented FILE stream opened for output. + + mb_setascii (&mbc, sc) + assigns the standard ASCII character sc to mbc. + + mb_copy (&destmbc, &srcmbc) + copies srcmbc to destmbc. + + Here are the function prototypes of the macros. + + extern const char * mb_ptr (const mbchar_t mbc); + extern size_t mb_len (const mbchar_t mbc); + extern bool mb_iseq (const mbchar_t mbc, char sc); + extern bool mb_isnul (const mbchar_t mbc); + extern int mb_cmp (const mbchar_t mbc1, const mbchar_t mbc2); + extern int mb_casecmp (const mbchar_t mbc1, const mbchar_t mbc2); + extern bool mb_equal (const mbchar_t mbc1, const mbchar_t mbc2); + extern bool mb_caseequal (const mbchar_t mbc1, const mbchar_t mbc2); + extern bool mb_isalnum (const mbchar_t mbc); + extern bool mb_isalpha (const mbchar_t mbc); + extern bool mb_isascii (const mbchar_t mbc); + extern bool mb_isblank (const mbchar_t mbc); + extern bool mb_iscntrl (const mbchar_t mbc); + extern bool mb_isdigit (const mbchar_t mbc); + extern bool mb_isgraph (const mbchar_t mbc); + extern bool mb_islower (const mbchar_t mbc); + extern bool mb_isprint (const mbchar_t mbc); + extern bool mb_ispunct (const mbchar_t mbc); + extern bool mb_isspace (const mbchar_t mbc); + extern bool mb_isupper (const mbchar_t mbc); + extern bool mb_isxdigit (const mbchar_t mbc); + extern int mb_width (const mbchar_t mbc); + extern void mb_putc (const mbchar_t mbc, FILE *stream); + extern void mb_setascii (mbchar_t *new, char sc); + extern void mb_copy (mbchar_t *new, const mbchar_t *old); + */ + +#ifndef _MBCHAR_H +#define _MBCHAR_H 1 + +#include <stdbool.h> +#include <string.h> + +/* Tru64 with Desktop Toolkit C has a bug: <stdio.h> must be included before + <wchar.h>. + BSD/OS 4.1 has a bug: <stdio.h> and <time.h> must be included before + <wchar.h>. */ +#include <stdio.h> +#include <time.h> +#include <wchar.h> + +#include <wctype.h> + +#define MBCHAR_BUF_SIZE 24 + +struct mbchar +{ + const char *ptr; /* pointer to current character */ + size_t bytes; /* number of bytes of current character, > 0 */ + bool wc_valid; /* true if wc is a valid wide character */ + wchar_t wc; /* if wc_valid: the current character */ + char buf[MBCHAR_BUF_SIZE]; /* room for the bytes, used for file input only */ +}; + +/* EOF (not a real character) is represented with bytes = 0 and + wc_valid = false. */ + +typedef struct mbchar mbchar_t; + +/* Access the current character. */ +#define mb_ptr(mbc) ((mbc).ptr) +#define mb_len(mbc) ((mbc).bytes) + +/* Comparison of characters. */ +#define mb_iseq(mbc, sc) ((mbc).wc_valid && (mbc).wc == (sc)) +#define mb_isnul(mbc) ((mbc).wc_valid && (mbc).wc == 0) +#define mb_cmp(mbc1, mbc2) \ + ((mbc1).wc_valid \ + ? ((mbc2).wc_valid \ + ? (int) (mbc1).wc - (int) (mbc2).wc \ + : -1) \ + : ((mbc2).wc_valid \ + ? 1 \ + : (mbc1).bytes == (mbc2).bytes \ + ? memcmp ((mbc1).ptr, (mbc2).ptr, (mbc1).bytes) \ + : (mbc1).bytes < (mbc2).bytes \ + ? (memcmp ((mbc1).ptr, (mbc2).ptr, (mbc1).bytes) > 0 ? 1 : -1) \ + : (memcmp ((mbc1).ptr, (mbc2).ptr, (mbc2).bytes) >= 0 ? 1 : -1))) +#define mb_casecmp(mbc1, mbc2) \ + ((mbc1).wc_valid \ + ? ((mbc2).wc_valid \ + ? (int) towlower ((mbc1).wc) - (int) towlower ((mbc2).wc) \ + : -1) \ + : ((mbc2).wc_valid \ + ? 1 \ + : (mbc1).bytes == (mbc2).bytes \ + ? memcmp ((mbc1).ptr, (mbc2).ptr, (mbc1).bytes) \ + : (mbc1).bytes < (mbc2).bytes \ + ? (memcmp ((mbc1).ptr, (mbc2).ptr, (mbc1).bytes) > 0 ? 1 : -1) \ + : (memcmp ((mbc1).ptr, (mbc2).ptr, (mbc2).bytes) >= 0 ? 1 : -1))) +#define mb_equal(mbc1, mbc2) \ + ((mbc1).wc_valid && (mbc2).wc_valid \ + ? (mbc1).wc == (mbc2).wc \ + : (mbc1).bytes == (mbc2).bytes \ + && memcmp ((mbc1).ptr, (mbc2).ptr, (mbc1).bytes) == 0) +#define mb_caseequal(mbc1, mbc2) \ + ((mbc1).wc_valid && (mbc2).wc_valid \ + ? towlower ((mbc1).wc) == towlower ((mbc2).wc) \ + : (mbc1).bytes == (mbc2).bytes \ + && memcmp ((mbc1).ptr, (mbc2).ptr, (mbc1).bytes) == 0) + +/* <ctype.h>, <wctype.h> classification. */ +#define mb_isascii(mbc) \ + ((mbc).wc_valid && (mbc).wc >= 0 && (mbc).wc <= 127) +#define mb_isalnum(mbc) ((mbc).wc_valid && iswalnum ((mbc).wc)) +#define mb_isalpha(mbc) ((mbc).wc_valid && iswalpha ((mbc).wc)) +#define mb_isblank(mbc) ((mbc).wc_valid && iswblank ((mbc).wc)) +#define mb_iscntrl(mbc) ((mbc).wc_valid && iswcntrl ((mbc).wc)) +#define mb_isdigit(mbc) ((mbc).wc_valid && iswdigit ((mbc).wc)) +#define mb_isgraph(mbc) ((mbc).wc_valid && iswgraph ((mbc).wc)) +#define mb_islower(mbc) ((mbc).wc_valid && iswlower ((mbc).wc)) +#define mb_isprint(mbc) ((mbc).wc_valid && iswprint ((mbc).wc)) +#define mb_ispunct(mbc) ((mbc).wc_valid && iswpunct ((mbc).wc)) +#define mb_isspace(mbc) ((mbc).wc_valid && iswspace ((mbc).wc)) +#define mb_isupper(mbc) ((mbc).wc_valid && iswupper ((mbc).wc)) +#define mb_isxdigit(mbc) ((mbc).wc_valid && iswxdigit ((mbc).wc)) + +/* Extra <wchar.h> function. */ + +/* Unprintable characters appear as a small box of width 1. */ +#define MB_UNPRINTABLE_WIDTH 1 + +static inline int +mb_width_aux (wint_t wc) +{ + int w = wcwidth (wc); + /* For unprintable characters, arbitrarily return 0 for control characters + and MB_UNPRINTABLE_WIDTH otherwise. */ + return (w >= 0 ? w : iswcntrl (wc) ? 0 : MB_UNPRINTABLE_WIDTH); +} + +#define mb_width(mbc) \ + ((mbc).wc_valid ? mb_width_aux ((mbc).wc) : MB_UNPRINTABLE_WIDTH) + +/* Output. */ +#define mb_putc(mbc, stream) fwrite ((mbc).ptr, 1, (mbc).bytes, (stream)) + +/* Assignment. */ +#define mb_setascii(mbc, sc) \ + ((mbc)->ptr = (mbc)->buf, (mbc)->bytes = 1, (mbc)->wc_valid = 1, \ + (mbc)->wc = (mbc)->buf[0] = (sc)) + +/* Copying a character. */ +static inline void +mb_copy (mbchar_t *new, const mbchar_t *old) +{ + if (old->ptr == &old->buf[0]) + { + memcpy (&new->buf[0], &old->buf[0], old->bytes); + new->ptr = &new->buf[0]; + } + else + new->ptr = old->ptr; + new->bytes = old->bytes; + if ((new->wc_valid = old->wc_valid)) + new->wc = old->wc; +} + + +/* is_basic(c) tests whether the single-byte character c is in the + ISO C "basic character set". + This is a convenience function, and is in this file only to share code + between mbiter_multi.h and mbfile_multi.h. */ +#if (' ' == 32) && ('!' == 33) && ('"' == 34) && ('#' == 35) \ + && ('%' == 37) && ('&' == 38) && ('\'' == 39) && ('(' == 40) \ + && (')' == 41) && ('*' == 42) && ('+' == 43) && (',' == 44) \ + && ('-' == 45) && ('.' == 46) && ('/' == 47) && ('0' == 48) \ + && ('1' == 49) && ('2' == 50) && ('3' == 51) && ('4' == 52) \ + && ('5' == 53) && ('6' == 54) && ('7' == 55) && ('8' == 56) \ + && ('9' == 57) && (':' == 58) && (';' == 59) && ('<' == 60) \ + && ('=' == 61) && ('>' == 62) && ('?' == 63) && ('A' == 65) \ + && ('B' == 66) && ('C' == 67) && ('D' == 68) && ('E' == 69) \ + && ('F' == 70) && ('G' == 71) && ('H' == 72) && ('I' == 73) \ + && ('J' == 74) && ('K' == 75) && ('L' == 76) && ('M' == 77) \ + && ('N' == 78) && ('O' == 79) && ('P' == 80) && ('Q' == 81) \ + && ('R' == 82) && ('S' == 83) && ('T' == 84) && ('U' == 85) \ + && ('V' == 86) && ('W' == 87) && ('X' == 88) && ('Y' == 89) \ + && ('Z' == 90) && ('[' == 91) && ('\\' == 92) && (']' == 93) \ + && ('^' == 94) && ('_' == 95) && ('a' == 97) && ('b' == 98) \ + && ('c' == 99) && ('d' == 100) && ('e' == 101) && ('f' == 102) \ + && ('g' == 103) && ('h' == 104) && ('i' == 105) && ('j' == 106) \ + && ('k' == 107) && ('l' == 108) && ('m' == 109) && ('n' == 110) \ + && ('o' == 111) && ('p' == 112) && ('q' == 113) && ('r' == 114) \ + && ('s' == 115) && ('t' == 116) && ('u' == 117) && ('v' == 118) \ + && ('w' == 119) && ('x' == 120) && ('y' == 121) && ('z' == 122) \ + && ('{' == 123) && ('|' == 124) && ('}' == 125) && ('~' == 126) +/* The character set is ISO-646, not EBCDIC. */ +# define IS_BASIC_ASCII 1 + +extern unsigned int is_basic_table[]; + +static inline bool +is_basic (char c) +{ + return (is_basic_table [(unsigned char) c >> 5] >> ((unsigned char) c & 31)) + & 1; +} + +#else + +static inline bool +is_basic (char c) +{ + switch (c) + { + case '\t': case '\v': case '\f': + case ' ': case '!': case '"': case '#': case '%': + case '&': case '\'': case '(': case ')': case '*': + case '+': case ',': case '-': case '.': case '/': + case '0': case '1': case '2': case '3': case '4': + case '5': case '6': case '7': case '8': case '9': + case ':': case ';': case '<': case '=': case '>': + case '?': + case 'A': case 'B': case 'C': case 'D': case 'E': + case 'F': case 'G': case 'H': case 'I': case 'J': + case 'K': case 'L': case 'M': case 'N': case 'O': + case 'P': case 'Q': case 'R': case 'S': case 'T': + case 'U': case 'V': case 'W': case 'X': case 'Y': + case 'Z': + case '[': case '\\': case ']': case '^': case '_': + case 'a': case 'b': case 'c': case 'd': case 'e': + case 'f': case 'g': case 'h': case 'i': case 'j': + case 'k': case 'l': case 'm': case 'n': case 'o': + case 'p': case 'q': case 'r': case 's': case 't': + case 'u': case 'v': case 'w': case 'x': case 'y': + case 'z': case '{': case '|': case '}': case '~': + return 1; + default: + return 0; + } +} + +#endif + +#endif /* _MBCHAR_H */ diff --git a/lib/mbuiter.h b/lib/mbuiter.h new file mode 100644 index 0000000..9da3a6c --- /dev/null +++ b/lib/mbuiter.h @@ -0,0 +1,203 @@ +/* Iterating through multibyte strings: macros for multi-byte encodings. + Copyright (C) 2001, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ + +/* Written by Bruno Haible <bruno@clisp.org>. */ + +/* The macros in this file implement forward iteration through a + multi-byte string, without knowing its length a-priori. + + With these macros, an iteration loop that looks like + + char *iter; + for (iter = buf; *iter != '\0'; iter++) + { + do_something (*iter); + } + + becomes + + mbui_iterator_t iter; + for (mbui_init (iter, buf); mbui_avail (iter); mbui_advance (iter)) + { + do_something (mbui_cur_ptr (iter), mb_len (mbui_cur (iter))); + } + + The benefit of these macros over plain use of mbrtowc is: + - Handling of invalid multibyte sequences is possible without + making the code more complicated, while still preserving the + invalid multibyte sequences. + + Compared to mbiter.h, the macros here don't need to know the string's + length a-priori. The downside is that at each step, the look-ahead + that guards against overrunning the terminating '\0' is more expensive. + The mbui_* macros are therefore suitable when there is a high probability + that only the first few multibyte characters need to be inspected. + Whereas the mbi_* macros are better if usually the iteration runs + through the entire string. + + mbui_iterator_t + is a type usable for variable declarations. + + mbui_init (iter, startptr) + initializes the iterator, starting at startptr. + + mbui_avail (iter) + returns true if there are more multibyte chracters available before + the end of string is reached. In this case, mbui_cur (iter) is + initialized to the next multibyte chracter. + + mbui_advance (iter) + advances the iterator by one multibyte character. + + mbui_cur (iter) + returns the current multibyte character, of type mbchar_t. All the + macros defined in mbchar.h can be used on it. + + mbui_cur_ptr (iter) + return a pointer to the beginning of the current multibyte character. + + mbui_reloc (iter, ptrdiff) + relocates iterator when the string is moved by ptrdiff bytes. + + Here are the function prototypes of the macros. + + extern void mbui_init (mbui_iterator_t iter, const char *startptr); + extern bool mbui_avail (mbui_iterator_t iter); + extern void mbui_advance (mbui_iterator_t iter); + extern mbchar_t mbui_cur (mbui_iterator_t iter); + extern const char * mbui_cur_ptr (mbui_iterator_t iter); + extern void mbui_reloc (mbui_iterator_t iter, ptrdiff_t ptrdiff); + */ + +#ifndef _MBUITER_H +#define _MBUITER_H 1 + +#include <assert.h> +#include <stdbool.h> +#include <stdlib.h> + +/* Tru64 with Desktop Toolkit C has a bug: <stdio.h> must be included before + <wchar.h>. + BSD/OS 4.1 has a bug: <stdio.h> and <time.h> must be included before + <wchar.h>. */ +#include <stdio.h> +#include <time.h> +#include <wchar.h> + +#include "mbchar.h" +#include "strnlen1.h" + +struct mbuiter_multi +{ + bool in_shift; /* true if next byte may not be interpreted as ASCII */ + mbstate_t state; /* if in_shift: current shift state */ + bool next_done; /* true if mbui_avail has already filled the following */ + struct mbchar cur; /* the current character: + const char *cur.ptr pointer to current character + The following are only valid after mbui_avail. + size_t cur.bytes number of bytes of current character + bool cur.wc_valid true if wc is a valid wide character + wchar_t cur.wc if wc_valid: the current character + */ +}; + +static inline void +mbuiter_multi_next (struct mbuiter_multi *iter) +{ + if (iter->next_done) + return; + if (iter->in_shift) + goto with_shift; + /* Handle most ASCII characters quickly, without calling mbrtowc(). */ + if (is_basic (*iter->cur.ptr)) + { + /* These characters are part of the basic character set. ISO C 99 + guarantees that their wide character code is identical to their + char code. */ + iter->cur.bytes = 1; + iter->cur.wc = *iter->cur.ptr; + iter->cur.wc_valid = true; + } + else + { + assert (mbsinit (&iter->state)); + iter->in_shift = true; + with_shift: + iter->cur.bytes = mbrtowc (&iter->cur.wc, iter->cur.ptr, + strnlen1 (iter->cur.ptr, MB_CUR_MAX), + &iter->state); + if (iter->cur.bytes == (size_t) -1) + { + /* An invalid multibyte sequence was encountered. */ + iter->cur.bytes = 1; + iter->cur.wc_valid = false; + /* Whether to set iter->in_shift = false and reset iter->state + or not is not very important; the string is bogus anyway. */ + } + else if (iter->cur.bytes == (size_t) -2) + { + /* An incomplete multibyte character at the end. */ + iter->cur.bytes = strlen (iter->cur.ptr); + iter->cur.wc_valid = false; + /* Whether to set iter->in_shift = false and reset iter->state + or not is not important; the string end is reached anyway. */ + } + else + { + if (iter->cur.bytes == 0) + { + /* A null wide character was encountered. */ + iter->cur.bytes = 1; + assert (*iter->cur.ptr == '\0'); + assert (iter->cur.wc == 0); + } + iter->cur.wc_valid = true; + + /* When in the initial state, we can go back treating ASCII + characters more quickly. */ + if (mbsinit (&iter->state)) + iter->in_shift = false; + } + } + iter->next_done = true; +} + +static inline void +mbuiter_multi_reloc (struct mbuiter_multi *iter, ptrdiff_t ptrdiff) +{ + iter->cur.ptr += ptrdiff; +} + +/* Iteration macros. */ +typedef struct mbuiter_multi mbui_iterator_t; +#define mbui_init(iter, startptr) \ + ((iter).cur.ptr = (startptr), \ + (iter).in_shift = false, memset (&(iter).state, '\0', sizeof (mbstate_t)), \ + (iter).next_done = false) +#define mbui_avail(iter) \ + (mbuiter_multi_next (&(iter)), !mb_isnul ((iter).cur)) +#define mbui_advance(iter) \ + ((iter).cur.ptr += (iter).cur.bytes, (iter).next_done = false) + +/* Access to the current character. */ +#define mbui_cur(iter) (iter).cur +#define mbui_cur_ptr(iter) (iter).cur.ptr + +/* Relocation. */ +#define mbui_reloc(iter, ptrdiff) mbuiter_multi_reloc (&iter, ptrdiff) + +#endif /* _MBUITER_H */ diff --git a/lib/md5.c b/lib/md5.c new file mode 100644 index 0000000..87db8f9 --- /dev/null +++ b/lib/md5.c @@ -0,0 +1,452 @@ +/* md5.c - Functions to compute MD5 message digest of files or memory blocks + according to the definition of MD5 in RFC 1321 from April 1992. + Copyright (C) 1995, 1996, 2001, 2003, 2004, 2005 Free Software Foundation, Inc. + NOTE: The canonical source of this file is maintained with the GNU C + Library. Bugs can be reported to bug-glibc@prep.ai.mit.edu. + + This program is free software; you can redistribute it and/or modify it + under the terms of the GNU General Public License as published by the + Free Software Foundation; either version 2, or (at your option) any + later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Ulrich Drepper <drepper@gnu.ai.mit.edu>, 1995. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "md5.h" + +#include <stddef.h> +#include <string.h> + +#if USE_UNLOCKED_IO +# include "unlocked-io.h" +#endif + +#ifdef _LIBC +# include <endian.h> +# if __BYTE_ORDER == __BIG_ENDIAN +# define WORDS_BIGENDIAN 1 +# endif +/* We need to keep the namespace clean so define the MD5 function + protected using leading __ . */ +# define md5_init_ctx __md5_init_ctx +# define md5_process_block __md5_process_block +# define md5_process_bytes __md5_process_bytes +# define md5_finish_ctx __md5_finish_ctx +# define md5_read_ctx __md5_read_ctx +# define md5_stream __md5_stream +# define md5_buffer __md5_buffer +#endif + +#ifdef WORDS_BIGENDIAN +# define SWAP(n) \ + (((n) << 24) | (((n) & 0xff00) << 8) | (((n) >> 8) & 0xff00) | ((n) >> 24)) +#else +# define SWAP(n) (n) +#endif + +#define BLOCKSIZE 4096 +#if BLOCKSIZE % 64 != 0 +# error "invalid BLOCKSIZE" +#endif + +/* This array contains the bytes used to pad the buffer to the next + 64-byte boundary. (RFC 1321, 3.1: Step 1) */ +static const unsigned char fillbuf[64] = { 0x80, 0 /* , 0, 0, ... */ }; + + +/* Initialize structure containing state of computation. + (RFC 1321, 3.3: Step 3) */ +void +md5_init_ctx (struct md5_ctx *ctx) +{ + ctx->A = 0x67452301; + ctx->B = 0xefcdab89; + ctx->C = 0x98badcfe; + ctx->D = 0x10325476; + + ctx->total[0] = ctx->total[1] = 0; + ctx->buflen = 0; +} + +/* Put result from CTX in first 16 bytes following RESBUF. The result + must be in little endian byte order. + + IMPORTANT: On some systems it is required that RESBUF is correctly + aligned for a 32 bits value. */ +void * +md5_read_ctx (const struct md5_ctx *ctx, void *resbuf) +{ + ((md5_uint32 *) resbuf)[0] = SWAP (ctx->A); + ((md5_uint32 *) resbuf)[1] = SWAP (ctx->B); + ((md5_uint32 *) resbuf)[2] = SWAP (ctx->C); + ((md5_uint32 *) resbuf)[3] = SWAP (ctx->D); + + return resbuf; +} + +/* Process the remaining bytes in the internal buffer and the usual + prolog according to the standard and write the result to RESBUF. + + IMPORTANT: On some systems it is required that RESBUF is correctly + aligned for a 32 bits value. */ +void * +md5_finish_ctx (struct md5_ctx *ctx, void *resbuf) +{ + /* Take yet unprocessed bytes into account. */ + md5_uint32 bytes = ctx->buflen; + size_t pad; + + /* Now count remaining bytes. */ + ctx->total[0] += bytes; + if (ctx->total[0] < bytes) + ++ctx->total[1]; + + pad = bytes >= 56 ? 64 + 56 - bytes : 56 - bytes; + memcpy (&ctx->buffer[bytes], fillbuf, pad); + + /* Put the 64-bit file length in *bits* at the end of the buffer. */ + *(md5_uint32 *) &ctx->buffer[bytes + pad] = SWAP (ctx->total[0] << 3); + *(md5_uint32 *) &ctx->buffer[bytes + pad + 4] = SWAP ((ctx->total[1] << 3) | + (ctx->total[0] >> 29)); + + /* Process last bytes. */ + md5_process_block (ctx->buffer, bytes + pad + 8, ctx); + + return md5_read_ctx (ctx, resbuf); +} + +/* Compute MD5 message digest for bytes read from STREAM. The + resulting message digest number will be written into the 16 bytes + beginning at RESBLOCK. */ +int +md5_stream (FILE *stream, void *resblock) +{ + struct md5_ctx ctx; + char buffer[BLOCKSIZE + 72]; + size_t sum; + + /* Initialize the computation context. */ + md5_init_ctx (&ctx); + + /* Iterate over full file contents. */ + while (1) + { + /* We read the file in blocks of BLOCKSIZE bytes. One call of the + computation function processes the whole buffer so that with the + next round of the loop another block can be read. */ + size_t n; + sum = 0; + + /* Read block. Take care for partial reads. */ + while (1) + { + n = fread (buffer + sum, 1, BLOCKSIZE - sum, stream); + + sum += n; + + if (sum == BLOCKSIZE) + break; + + if (n == 0) + { + /* Check for the error flag IFF N == 0, so that we don't + exit the loop after a partial read due to e.g., EAGAIN + or EWOULDBLOCK. */ + if (ferror (stream)) + return 1; + goto process_partial_block; + } + + /* We've read at least one byte, so ignore errors. But always + check for EOF, since feof may be true even though N > 0. + Otherwise, we could end up calling fread after EOF. */ + if (feof (stream)) + goto process_partial_block; + } + + /* Process buffer with BLOCKSIZE bytes. Note that + BLOCKSIZE % 64 == 0 + */ + md5_process_block (buffer, BLOCKSIZE, &ctx); + } + + process_partial_block:; + + /* Process any remaining bytes. */ + if (sum > 0) + md5_process_bytes (buffer, sum, &ctx); + + /* Construct result in desired memory. */ + md5_finish_ctx (&ctx, resblock); + return 0; +} + +/* Compute MD5 message digest for LEN bytes beginning at BUFFER. The + result is always in little endian byte order, so that a byte-wise + output yields to the wanted ASCII representation of the message + digest. */ +void * +md5_buffer (const char *buffer, size_t len, void *resblock) +{ + struct md5_ctx ctx; + + /* Initialize the computation context. */ + md5_init_ctx (&ctx); + + /* Process whole buffer but last len % 64 bytes. */ + md5_process_bytes (buffer, len, &ctx); + + /* Put result in desired memory area. */ + return md5_finish_ctx (&ctx, resblock); +} + + +void +md5_process_bytes (const void *buffer, size_t len, struct md5_ctx *ctx) +{ + /* When we already have some bits in our internal buffer concatenate + both inputs first. */ + if (ctx->buflen != 0) + { + size_t left_over = ctx->buflen; + size_t add = 128 - left_over > len ? len : 128 - left_over; + + memcpy (&ctx->buffer[left_over], buffer, add); + ctx->buflen += add; + + if (ctx->buflen > 64) + { + md5_process_block (ctx->buffer, ctx->buflen & ~63, ctx); + + ctx->buflen &= 63; + /* The regions in the following copy operation cannot overlap. */ + memcpy (ctx->buffer, &ctx->buffer[(left_over + add) & ~63], + ctx->buflen); + } + + buffer = (const char *) buffer + add; + len -= add; + } + + /* Process available complete blocks. */ + if (len >= 64) + { +#if !_STRING_ARCH_unaligned +# define alignof(type) offsetof (struct { char c; type x; }, x) +# define UNALIGNED_P(p) (((size_t) p) % alignof (md5_uint32) != 0) + if (UNALIGNED_P (buffer)) + while (len > 64) + { + md5_process_block (memcpy (ctx->buffer, buffer, 64), 64, ctx); + buffer = (const char *) buffer + 64; + len -= 64; + } + else +#endif + { + md5_process_block (buffer, len & ~63, ctx); + buffer = (const char *) buffer + (len & ~63); + len &= 63; + } + } + + /* Move remaining bytes in internal buffer. */ + if (len > 0) + { + size_t left_over = ctx->buflen; + + memcpy (&ctx->buffer[left_over], buffer, len); + left_over += len; + if (left_over >= 64) + { + md5_process_block (ctx->buffer, 64, ctx); + left_over -= 64; + memcpy (ctx->buffer, &ctx->buffer[64], left_over); + } + ctx->buflen = left_over; + } +} + + +/* These are the four functions used in the four steps of the MD5 algorithm + and defined in the RFC 1321. The first function is a little bit optimized + (as found in Colin Plumbs public domain implementation). */ +/* #define FF(b, c, d) ((b & c) | (~b & d)) */ +#define FF(b, c, d) (d ^ (b & (c ^ d))) +#define FG(b, c, d) FF (d, b, c) +#define FH(b, c, d) (b ^ c ^ d) +#define FI(b, c, d) (c ^ (b | ~d)) + +/* Process LEN bytes of BUFFER, accumulating context into CTX. + It is assumed that LEN % 64 == 0. */ + +void +md5_process_block (const void *buffer, size_t len, struct md5_ctx *ctx) +{ + md5_uint32 correct_words[16]; + const md5_uint32 *words = buffer; + size_t nwords = len / sizeof (md5_uint32); + const md5_uint32 *endp = words + nwords; + md5_uint32 A = ctx->A; + md5_uint32 B = ctx->B; + md5_uint32 C = ctx->C; + md5_uint32 D = ctx->D; + + /* First increment the byte count. RFC 1321 specifies the possible + length of the file up to 2^64 bits. Here we only compute the + number of bytes. Do a double word increment. */ + ctx->total[0] += len; + if (ctx->total[0] < len) + ++ctx->total[1]; + + /* Process all bytes in the buffer with 64 bytes in each round of + the loop. */ + while (words < endp) + { + md5_uint32 *cwp = correct_words; + md5_uint32 A_save = A; + md5_uint32 B_save = B; + md5_uint32 C_save = C; + md5_uint32 D_save = D; + + /* First round: using the given function, the context and a constant + the next context is computed. Because the algorithms processing + unit is a 32-bit word and it is determined to work on words in + little endian byte order we perhaps have to change the byte order + before the computation. To reduce the work for the next steps + we store the swapped words in the array CORRECT_WORDS. */ + +#define OP(a, b, c, d, s, T) \ + do \ + { \ + a += FF (b, c, d) + (*cwp++ = SWAP (*words)) + T; \ + ++words; \ + CYCLIC (a, s); \ + a += b; \ + } \ + while (0) + + /* It is unfortunate that C does not provide an operator for + cyclic rotation. Hope the C compiler is smart enough. */ +#define CYCLIC(w, s) (w = (w << s) | (w >> (32 - s))) + + /* Before we start, one word to the strange constants. + They are defined in RFC 1321 as + + T[i] = (int) (4294967296.0 * fabs (sin (i))), i=1..64 + + Here is an equivalent invocation using Perl: + + perl -e 'foreach(1..64){printf "0x%08x\n", int (4294967296 * abs (sin $_))}' + */ + + /* Round 1. */ + OP (A, B, C, D, 7, 0xd76aa478); + OP (D, A, B, C, 12, 0xe8c7b756); + OP (C, D, A, B, 17, 0x242070db); + OP (B, C, D, A, 22, 0xc1bdceee); + OP (A, B, C, D, 7, 0xf57c0faf); + OP (D, A, B, C, 12, 0x4787c62a); + OP (C, D, A, B, 17, 0xa8304613); + OP (B, C, D, A, 22, 0xfd469501); + OP (A, B, C, D, 7, 0x698098d8); + OP (D, A, B, C, 12, 0x8b44f7af); + OP (C, D, A, B, 17, 0xffff5bb1); + OP (B, C, D, A, 22, 0x895cd7be); + OP (A, B, C, D, 7, 0x6b901122); + OP (D, A, B, C, 12, 0xfd987193); + OP (C, D, A, B, 17, 0xa679438e); + OP (B, C, D, A, 22, 0x49b40821); + + /* For the second to fourth round we have the possibly swapped words + in CORRECT_WORDS. Redefine the macro to take an additional first + argument specifying the function to use. */ +#undef OP +#define OP(f, a, b, c, d, k, s, T) \ + do \ + { \ + a += f (b, c, d) + correct_words[k] + T; \ + CYCLIC (a, s); \ + a += b; \ + } \ + while (0) + + /* Round 2. */ + OP (FG, A, B, C, D, 1, 5, 0xf61e2562); + OP (FG, D, A, B, C, 6, 9, 0xc040b340); + OP (FG, C, D, A, B, 11, 14, 0x265e5a51); + OP (FG, B, C, D, A, 0, 20, 0xe9b6c7aa); + OP (FG, A, B, C, D, 5, 5, 0xd62f105d); + OP (FG, D, A, B, C, 10, 9, 0x02441453); + OP (FG, C, D, A, B, 15, 14, 0xd8a1e681); + OP (FG, B, C, D, A, 4, 20, 0xe7d3fbc8); + OP (FG, A, B, C, D, 9, 5, 0x21e1cde6); + OP (FG, D, A, B, C, 14, 9, 0xc33707d6); + OP (FG, C, D, A, B, 3, 14, 0xf4d50d87); + OP (FG, B, C, D, A, 8, 20, 0x455a14ed); + OP (FG, A, B, C, D, 13, 5, 0xa9e3e905); + OP (FG, D, A, B, C, 2, 9, 0xfcefa3f8); + OP (FG, C, D, A, B, 7, 14, 0x676f02d9); + OP (FG, B, C, D, A, 12, 20, 0x8d2a4c8a); + + /* Round 3. */ + OP (FH, A, B, C, D, 5, 4, 0xfffa3942); + OP (FH, D, A, B, C, 8, 11, 0x8771f681); + OP (FH, C, D, A, B, 11, 16, 0x6d9d6122); + OP (FH, B, C, D, A, 14, 23, 0xfde5380c); + OP (FH, A, B, C, D, 1, 4, 0xa4beea44); + OP (FH, D, A, B, C, 4, 11, 0x4bdecfa9); + OP (FH, C, D, A, B, 7, 16, 0xf6bb4b60); + OP (FH, B, C, D, A, 10, 23, 0xbebfbc70); + OP (FH, A, B, C, D, 13, 4, 0x289b7ec6); + OP (FH, D, A, B, C, 0, 11, 0xeaa127fa); + OP (FH, C, D, A, B, 3, 16, 0xd4ef3085); + OP (FH, B, C, D, A, 6, 23, 0x04881d05); + OP (FH, A, B, C, D, 9, 4, 0xd9d4d039); + OP (FH, D, A, B, C, 12, 11, 0xe6db99e5); + OP (FH, C, D, A, B, 15, 16, 0x1fa27cf8); + OP (FH, B, C, D, A, 2, 23, 0xc4ac5665); + + /* Round 4. */ + OP (FI, A, B, C, D, 0, 6, 0xf4292244); + OP (FI, D, A, B, C, 7, 10, 0x432aff97); + OP (FI, C, D, A, B, 14, 15, 0xab9423a7); + OP (FI, B, C, D, A, 5, 21, 0xfc93a039); + OP (FI, A, B, C, D, 12, 6, 0x655b59c3); + OP (FI, D, A, B, C, 3, 10, 0x8f0ccc92); + OP (FI, C, D, A, B, 10, 15, 0xffeff47d); + OP (FI, B, C, D, A, 1, 21, 0x85845dd1); + OP (FI, A, B, C, D, 8, 6, 0x6fa87e4f); + OP (FI, D, A, B, C, 15, 10, 0xfe2ce6e0); + OP (FI, C, D, A, B, 6, 15, 0xa3014314); + OP (FI, B, C, D, A, 13, 21, 0x4e0811a1); + OP (FI, A, B, C, D, 4, 6, 0xf7537e82); + OP (FI, D, A, B, C, 11, 10, 0xbd3af235); + OP (FI, C, D, A, B, 2, 15, 0x2ad7d2bb); + OP (FI, B, C, D, A, 9, 21, 0xeb86d391); + + /* Add the starting values of the context. */ + A += A_save; + B += B_save; + C += C_save; + D += D_save; + } + + /* Put checksum in context given as argument. */ + ctx->A = A; + ctx->B = B; + ctx->C = C; + ctx->D = D; +} diff --git a/lib/md5.h b/lib/md5.h new file mode 100644 index 0000000..1c29e67 --- /dev/null +++ b/lib/md5.h @@ -0,0 +1,136 @@ +/* Declaration of functions and data types used for MD5 sum computing + library functions. + Copyright (C) 1995-1997,1999-2005 Free Software Foundation, Inc. + + NOTE: The canonical source of this file is maintained with the GNU C + Library. Bugs can be reported to bug-glibc@prep.ai.mit.edu. + + This program is free software; you can redistribute it and/or modify it + under the terms of the GNU General Public License as published by the + Free Software Foundation; either version 2, or (at your option) any + later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _MD5_H +#define _MD5_H 1 + +#include <stdio.h> + +#if HAVE_INTTYPES_H +# include <inttypes.h> +#endif +#if HAVE_STDINT_H || _LIBC +# include <stdint.h> +#endif + +#ifndef __GNUC_PREREQ +# if defined __GNUC__ && defined __GNUC_MINOR__ +# define __GNUC_PREREQ(maj, min) \ + ((__GNUC__ << 16) + __GNUC_MINOR__ >= ((maj) << 16) + (min)) +# else +# define __GNUC_PREREQ(maj, min) 0 +# endif +#endif + +#ifndef __THROW +# if defined __cplusplus && __GNUC_PREREQ (2,8) +# define __THROW throw () +# else +# define __THROW +# endif +#endif + +#ifndef __attribute__ +# if ! __GNUC_PREREQ (2,8) || __STRICT_ANSI__ +# define __attribute__(x) +# endif +#endif + +#ifndef _LIBC +# define __md5_buffer md5_buffer +# define __md5_finish_ctx md5_finish_ctx +# define __md5_init_ctx md5_init_ctx +# define __md5_process_block md5_process_block +# define __md5_process_bytes md5_process_bytes +# define __md5_read_ctx md5_read_ctx +# define __md5_stream md5_stream +#endif + +typedef uint32_t md5_uint32; + +/* Structure to save state of computation between the single steps. */ +struct md5_ctx +{ + md5_uint32 A; + md5_uint32 B; + md5_uint32 C; + md5_uint32 D; + + md5_uint32 total[2]; + md5_uint32 buflen; + char buffer[128] __attribute__ ((__aligned__ (__alignof__ (md5_uint32)))); +}; + +/* + * The following three functions are build up the low level used in + * the functions `md5_stream' and `md5_buffer'. + */ + +/* Initialize structure containing state of computation. + (RFC 1321, 3.3: Step 3) */ +extern void __md5_init_ctx (struct md5_ctx *ctx) __THROW; + +/* Starting with the result of former calls of this function (or the + initialization function update the context for the next LEN bytes + starting at BUFFER. + It is necessary that LEN is a multiple of 64!!! */ +extern void __md5_process_block (const void *buffer, size_t len, + struct md5_ctx *ctx) __THROW; + +/* Starting with the result of former calls of this function (or the + initialization function update the context for the next LEN bytes + starting at BUFFER. + It is NOT required that LEN is a multiple of 64. */ +extern void __md5_process_bytes (const void *buffer, size_t len, + struct md5_ctx *ctx) __THROW; + +/* Process the remaining bytes in the buffer and put result from CTX + in first 16 bytes following RESBUF. The result is always in little + endian byte order, so that a byte-wise output yields to the wanted + ASCII representation of the message digest. + + IMPORTANT: On some systems it is required that RESBUF be correctly + aligned for a 32 bits value. */ +extern void *__md5_finish_ctx (struct md5_ctx *ctx, void *resbuf) __THROW; + + +/* Put result from CTX in first 16 bytes following RESBUF. The result is + always in little endian byte order, so that a byte-wise output yields + to the wanted ASCII representation of the message digest. + + IMPORTANT: On some systems it is required that RESBUF is correctly + aligned for a 32 bits value. */ +extern void *__md5_read_ctx (const struct md5_ctx *ctx, void *resbuf) __THROW; + + +/* Compute MD5 message digest for bytes read from STREAM. The + resulting message digest number will be written into the 16 bytes + beginning at RESBLOCK. */ +extern int __md5_stream (FILE *stream, void *resblock) __THROW; + +/* Compute MD5 message digest for LEN bytes beginning at BUFFER. The + result is always in little endian byte order, so that a byte-wise + output yields to the wanted ASCII representation of the message + digest. */ +extern void *__md5_buffer (const char *buffer, size_t len, + void *resblock) __THROW; + +#endif /* md5.h */ diff --git a/lib/memchr.c b/lib/memchr.c new file mode 100644 index 0000000..3c5ecfd --- /dev/null +++ b/lib/memchr.c @@ -0,0 +1,201 @@ +/* Copyright (C) 1991, 1993, 1996, 1997, 1999, 2000, 2003, 2004 Free + Software Foundation, Inc. + + Based on strlen implementation by Torbjorn Granlund (tege@sics.se), + with help from Dan Sahlin (dan@sics.se) and + commentary by Jim Blandy (jimb@ai.mit.edu); + adaptation to memchr suggested by Dick Karpinski (dick@cca.ucsf.edu), + and implemented by Roland McGrath (roland@ai.mit.edu). + +NOTE: The canonical source of this file is maintained with the GNU C Library. +Bugs can be reported to bug-glibc@prep.ai.mit.edu. + +This program is free software; you can redistribute it and/or modify it +under the terms of the GNU General Public License as published by the +Free Software Foundation; either version 2, or (at your option) any +later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software Foundation, +Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include <string.h> + +#include <stddef.h> + +#if defined _LIBC +# include <memcopy.h> +#else +# define reg_char char +#endif + +#include <limits.h> + +#if HAVE_BP_SYM_H || defined _LIBC +# include <bp-sym.h> +#else +# define BP_SYM(sym) sym +#endif + +#undef memchr +#undef __memchr + +/* Search no more than N bytes of S for C. */ +void * +__memchr (void const *s, int c_in, size_t n) +{ + const unsigned char *char_ptr; + const unsigned long int *longword_ptr; + unsigned long int longword, magic_bits, charmask; + unsigned reg_char c; + int i; + + c = (unsigned char) c_in; + + /* Handle the first few characters by reading one character at a time. + Do this until CHAR_PTR is aligned on a longword boundary. */ + for (char_ptr = (const unsigned char *) s; + n > 0 && (size_t) char_ptr % sizeof longword != 0; + --n, ++char_ptr) + if (*char_ptr == c) + return (void *) char_ptr; + + /* All these elucidatory comments refer to 4-byte longwords, + but the theory applies equally well to any size longwords. */ + + longword_ptr = (const unsigned long int *) char_ptr; + + /* Bits 31, 24, 16, and 8 of this number are zero. Call these bits + the "holes." Note that there is a hole just to the left of + each byte, with an extra at the end: + + bits: 01111110 11111110 11111110 11111111 + bytes: AAAAAAAA BBBBBBBB CCCCCCCC DDDDDDDD + + The 1-bits make sure that carries propagate to the next 0-bit. + The 0-bits provide holes for carries to fall into. */ + + /* Set MAGIC_BITS to be this pattern of 1 and 0 bits. + Set CHARMASK to be a longword, each of whose bytes is C. */ + + magic_bits = 0xfefefefe; + charmask = c | (c << 8); + charmask |= charmask << 16; +#if 0xffffffffU < ULONG_MAX + magic_bits |= magic_bits << 32; + charmask |= charmask << 32; + if (8 < sizeof longword) + for (i = 64; i < sizeof longword * 8; i *= 2) + { + magic_bits |= magic_bits << i; + charmask |= charmask << i; + } +#endif + magic_bits = (ULONG_MAX >> 1) & (magic_bits | 1); + + /* Instead of the traditional loop which tests each character, + we will test a longword at a time. The tricky part is testing + if *any of the four* bytes in the longword in question are zero. */ + while (n >= sizeof longword) + { + /* We tentatively exit the loop if adding MAGIC_BITS to + LONGWORD fails to change any of the hole bits of LONGWORD. + + 1) Is this safe? Will it catch all the zero bytes? + Suppose there is a byte with all zeros. Any carry bits + propagating from its left will fall into the hole at its + least significant bit and stop. Since there will be no + carry from its most significant bit, the LSB of the + byte to the left will be unchanged, and the zero will be + detected. + + 2) Is this worthwhile? Will it ignore everything except + zero bytes? Suppose every byte of LONGWORD has a bit set + somewhere. There will be a carry into bit 8. If bit 8 + is set, this will carry into bit 16. If bit 8 is clear, + one of bits 9-15 must be set, so there will be a carry + into bit 16. Similarly, there will be a carry into bit + 24. If one of bits 24-30 is set, there will be a carry + into bit 31, so all of the hole bits will be changed. + + The one misfire occurs when bits 24-30 are clear and bit + 31 is set; in this case, the hole at bit 31 is not + changed. If we had access to the processor carry flag, + we could close this loophole by putting the fourth hole + at bit 32! + + So it ignores everything except 128's, when they're aligned + properly. + + 3) But wait! Aren't we looking for C, not zero? + Good point. So what we do is XOR LONGWORD with a longword, + each of whose bytes is C. This turns each byte that is C + into a zero. */ + + longword = *longword_ptr++ ^ charmask; + + /* Add MAGIC_BITS to LONGWORD. */ + if ((((longword + magic_bits) + + /* Set those bits that were unchanged by the addition. */ + ^ ~longword) + + /* Look at only the hole bits. If any of the hole bits + are unchanged, most likely one of the bytes was a + zero. */ + & ~magic_bits) != 0) + { + /* Which of the bytes was C? If none of them were, it was + a misfire; continue the search. */ + + const unsigned char *cp = (const unsigned char *) (longword_ptr - 1); + + if (cp[0] == c) + return (void *) cp; + if (cp[1] == c) + return (void *) &cp[1]; + if (cp[2] == c) + return (void *) &cp[2]; + if (cp[3] == c) + return (void *) &cp[3]; + if (4 < sizeof longword && cp[4] == c) + return (void *) &cp[4]; + if (5 < sizeof longword && cp[5] == c) + return (void *) &cp[5]; + if (6 < sizeof longword && cp[6] == c) + return (void *) &cp[6]; + if (7 < sizeof longword && cp[7] == c) + return (void *) &cp[7]; + if (8 < sizeof longword) + for (i = 8; i < sizeof longword; i++) + if (cp[i] == c) + return (void *) &cp[i]; + } + + n -= sizeof longword; + } + + char_ptr = (const unsigned char *) longword_ptr; + + while (n-- > 0) + { + if (*char_ptr == c) + return (void *) char_ptr; + else + ++char_ptr; + } + + return 0; +} +#ifdef weak_alias +weak_alias (__memchr, BP_SYM (memchr)) +#endif diff --git a/lib/memmove.c b/lib/memmove.c new file mode 100644 index 0000000..753c899 --- /dev/null +++ b/lib/memmove.c @@ -0,0 +1,28 @@ +/* memmove.c -- copy memory. + Copy LENGTH bytes from SOURCE to DEST. Does not null-terminate. + In the public domain. + By David MacKenzie <djm@gnu.ai.mit.edu>. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include <stddef.h> + +void * +memmove (void *dest0, void const *source0, size_t length) +{ + char *dest = dest0; + char const *source = source0; + if (source < dest) + /* Moving from low mem to hi mem; start at end. */ + for (source += length, dest += length; length; --length) + *--dest = *--source; + else if (source != dest) + { + /* Moving from hi mem to low mem; start at beginning. */ + for (; length; --length) + *dest++ = *source++; + } + return dest0; +} diff --git a/lib/mempcpy.c b/lib/mempcpy.c new file mode 100644 index 0000000..3502da2 --- /dev/null +++ b/lib/mempcpy.c @@ -0,0 +1,29 @@ +/* Copy memory area and return pointer after last written byte. + Copyright (C) 2003 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Specification. */ +#include "mempcpy.h" + +#include <string.h> + +/* Copy N bytes of SRC to DEST, return pointer to bytes after the + last written byte. */ +void * +mempcpy (void *dest, const void *src, size_t n) +{ + return (char *) memcpy (dest, src, n) + n; +} diff --git a/lib/mempcpy.h b/lib/mempcpy.h new file mode 100644 index 0000000..fa20321 --- /dev/null +++ b/lib/mempcpy.h @@ -0,0 +1,36 @@ +/* Copy memory area and return pointer after last written byte. + Copyright (C) 2003, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef mempcpy + +# if HAVE_MEMPCPY + +/* Get mempcpy() declaration. */ +# include <string.h> + +# else + +/* Get size_t */ +# include <stddef.h> + +/* Copy N bytes of SRC to DEST, return pointer to bytes after the + last written byte. */ +extern void *mempcpy (void *dest, const void *src, size_t n); + +# endif + +#endif diff --git a/lib/memrchr.c b/lib/memrchr.c new file mode 100644 index 0000000..0b6659f --- /dev/null +++ b/lib/memrchr.c @@ -0,0 +1,194 @@ +/* memrchr -- find the last occurrence of a byte in a memory block + + Copyright (C) 1991, 1993, 1996, 1997, 1999, 2000, 2003, 2004, 2005 + Free Software Foundation, Inc. + + Based on strlen implementation by Torbjorn Granlund (tege@sics.se), + with help from Dan Sahlin (dan@sics.se) and + commentary by Jim Blandy (jimb@ai.mit.edu); + adaptation to memchr suggested by Dick Karpinski (dick@cca.ucsf.edu), + and implemented by Roland McGrath (roland@ai.mit.edu). + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#if defined _LIBC +# include <string.h> +# include <memcopy.h> +#else +# include "memrchr.h" +# define reg_char char +#endif + +#include <limits.h> + +#undef __memrchr +#undef memrchr + +#ifndef weak_alias +# define __memrchr memrchr +#endif + +/* Search no more than N bytes of S for C. */ +void * +__memrchr (void const *s, int c_in, size_t n) +{ + const unsigned char *char_ptr; + const unsigned long int *longword_ptr; + unsigned long int longword, magic_bits, charmask; + unsigned reg_char c; + int i; + + c = (unsigned char) c_in; + + /* Handle the last few characters by reading one character at a time. + Do this until CHAR_PTR is aligned on a longword boundary. */ + for (char_ptr = (const unsigned char *) s + n; + n > 0 && (size_t) char_ptr % sizeof longword != 0; + --n) + if (*--char_ptr == c) + return (void *) char_ptr; + + /* All these elucidatory comments refer to 4-byte longwords, + but the theory applies equally well to any size longwords. */ + + longword_ptr = (const unsigned long int *) char_ptr; + + /* Bits 31, 24, 16, and 8 of this number are zero. Call these bits + the "holes." Note that there is a hole just to the left of + each byte, with an extra at the end: + + bits: 01111110 11111110 11111110 11111111 + bytes: AAAAAAAA BBBBBBBB CCCCCCCC DDDDDDDD + + The 1-bits make sure that carries propagate to the next 0-bit. + The 0-bits provide holes for carries to fall into. */ + + /* Set MAGIC_BITS to be this pattern of 1 and 0 bits. + Set CHARMASK to be a longword, each of whose bytes is C. */ + + magic_bits = 0xfefefefe; + charmask = c | (c << 8); + charmask |= charmask << 16; +#if 0xffffffffU < ULONG_MAX + magic_bits |= magic_bits << 32; + charmask |= charmask << 32; + if (8 < sizeof longword) + for (i = 64; i < sizeof longword * 8; i *= 2) + { + magic_bits |= magic_bits << i; + charmask |= charmask << i; + } +#endif + magic_bits = (ULONG_MAX >> 1) & (magic_bits | 1); + + /* Instead of the traditional loop which tests each character, + we will test a longword at a time. The tricky part is testing + if *any of the four* bytes in the longword in question are zero. */ + while (n >= sizeof longword) + { + /* We tentatively exit the loop if adding MAGIC_BITS to + LONGWORD fails to change any of the hole bits of LONGWORD. + + 1) Is this safe? Will it catch all the zero bytes? + Suppose there is a byte with all zeros. Any carry bits + propagating from its left will fall into the hole at its + least significant bit and stop. Since there will be no + carry from its most significant bit, the LSB of the + byte to the left will be unchanged, and the zero will be + detected. + + 2) Is this worthwhile? Will it ignore everything except + zero bytes? Suppose every byte of LONGWORD has a bit set + somewhere. There will be a carry into bit 8. If bit 8 + is set, this will carry into bit 16. If bit 8 is clear, + one of bits 9-15 must be set, so there will be a carry + into bit 16. Similarly, there will be a carry into bit + 24. If one of bits 24-30 is set, there will be a carry + into bit 31, so all of the hole bits will be changed. + + The one misfire occurs when bits 24-30 are clear and bit + 31 is set; in this case, the hole at bit 31 is not + changed. If we had access to the processor carry flag, + we could close this loophole by putting the fourth hole + at bit 32! + + So it ignores everything except 128's, when they're aligned + properly. + + 3) But wait! Aren't we looking for C, not zero? + Good point. So what we do is XOR LONGWORD with a longword, + each of whose bytes is C. This turns each byte that is C + into a zero. */ + + longword = *--longword_ptr ^ charmask; + + /* Add MAGIC_BITS to LONGWORD. */ + if ((((longword + magic_bits) + + /* Set those bits that were unchanged by the addition. */ + ^ ~longword) + + /* Look at only the hole bits. If any of the hole bits + are unchanged, most likely one of the bytes was a + zero. */ + & ~magic_bits) != 0) + { + /* Which of the bytes was C? If none of them were, it was + a misfire; continue the search. */ + + const unsigned char *cp = (const unsigned char *) longword_ptr; + + if (8 < sizeof longword) + for (i = sizeof longword - 1; 8 <= i; i--) + if (cp[i] == c) + return (void *) &cp[i]; + if (7 < sizeof longword && cp[7] == c) + return (void *) &cp[7]; + if (6 < sizeof longword && cp[6] == c) + return (void *) &cp[6]; + if (5 < sizeof longword && cp[5] == c) + return (void *) &cp[5]; + if (4 < sizeof longword && cp[4] == c) + return (void *) &cp[4]; + if (cp[3] == c) + return (void *) &cp[3]; + if (cp[2] == c) + return (void *) &cp[2]; + if (cp[1] == c) + return (void *) &cp[1]; + if (cp[0] == c) + return (void *) cp; + } + + n -= sizeof longword; + } + + char_ptr = (const unsigned char *) longword_ptr; + + while (n-- > 0) + { + if (*--char_ptr == c) + return (void *) char_ptr; + } + + return 0; +} +#ifdef weak_alias +weak_alias (__memrchr, memrchr) +#endif diff --git a/lib/memrchr.h b/lib/memrchr.h new file mode 100644 index 0000000..d4785c2 --- /dev/null +++ b/lib/memrchr.h @@ -0,0 +1,27 @@ +/* memrchr -- Find the last occurrence of a byte in a memory block. + + Copyright (C) 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#if HAVE_DECL_MEMRCHR +# include <string.h> +#else +# include <stddef.h> + +/* Search backwards through a block for a byte (specified as an int). */ +void *memrchr (void const *, int, size_t); + +#endif diff --git a/lib/minmax.h b/lib/minmax.h new file mode 100644 index 0000000..975ea76 --- /dev/null +++ b/lib/minmax.h @@ -0,0 +1,60 @@ +/* MIN, MAX macros. + Copyright (C) 1995, 1998, 2001, 2003, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _MINMAX_H +#define _MINMAX_H + +/* Note: MIN, MAX are also defined in <sys/param.h> on some systems + (glibc, IRIX, HP-UX, OSF/1). Therefore you might get warnings about + MIN, MAX macro redefinitions on some systems; the workaround is to + #include this file as the last one among the #include list. */ + +/* Before we define the following symbols we get the <limits.h> file + since otherwise we get redefinitions on some systems if <limits.h> is + included after this file. Likewise for <sys/param.h>. + If more than one of these system headers define MIN and MAX, pick just + one of the headers (because the definitions most likely are the same). */ +#if HAVE_MINMAX_IN_LIMITS_H +# include <limits.h> +#elif HAVE_MINMAX_IN_SYS_PARAM_H +# include <sys/param.h> +#endif + +/* Note: MIN and MAX should be used with two arguments of the + same type. They might not return the minimum and maximum of their two + arguments, if the arguments have different types or have unusual + floating-point values. For example, on a typical host with 32-bit 'int', + 64-bit 'long long', and 64-bit IEEE 754 'double' types: + + MAX (-1, 2147483648) returns 4294967295. + MAX (9007199254740992.0, 9007199254740993) returns 9007199254740992.0. + MAX (NaN, 0.0) returns 0.0. + MAX (+0.0, -0.0) returns -0.0. + + and in each case the answer is in some sense bogus. */ + +/* MAX(a,b) returns the maximum of A and B. */ +#ifndef MAX +# define MAX(a,b) ((a) > (b) ? (a) : (b)) +#endif + +/* MIN(a,b) returns the minimum of A and B. */ +#ifndef MIN +# define MIN(a,b) ((a) < (b) ? (a) : (b)) +#endif + +#endif /* _MINMAX_H */ diff --git a/lib/mkdir.c b/lib/mkdir.c new file mode 100644 index 0000000..c921ecf --- /dev/null +++ b/lib/mkdir.c @@ -0,0 +1,65 @@ +/* On some systems, mkdir ("foo/", 0700) fails because of the trailing + slash. On those systems, this wrapper removes the trailing slash. + + Copyright (C) 2001, 2003 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* written by Jim Meyering */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Disable the definition of mkdir to rpl_mkdir (from config.h) in this + file. Otherwise, we'd get conflicting prototypes for rpl_mkdir on + most systems. */ +#undef mkdir + +#include <sys/types.h> +#include <sys/stat.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +#include "dirname.h" +#include "xalloc.h" + +/* This function is required at least for NetBSD 1.5.2. */ + +int +rpl_mkdir (char const *dir, mode_t mode) +{ + int ret_val; + char *tmp_dir; + size_t len = strlen (dir); + + if (len && dir[len - 1] == '/') + { + tmp_dir = xstrdup (dir); + strip_trailing_slashes (tmp_dir); + } + else + { + tmp_dir = (char *) dir; + } + + ret_val = mkdir (tmp_dir, mode); + + if (tmp_dir != dir) + free (tmp_dir); + + return ret_val; +} diff --git a/lib/mkstemp.c b/lib/mkstemp.c new file mode 100644 index 0000000..cec5ba8 --- /dev/null +++ b/lib/mkstemp.c @@ -0,0 +1,44 @@ +/* Copyright (C) 1998, 1999, 2001, 2005 Free Software Foundation, Inc. + This file is derived from the one in the GNU C Library. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Disable the definition of mkstemp to rpl_mkstemp (from config.h) in this + file. Otherwise, we'd get conflicting prototypes for rpl_mkstemp on + most systems. */ +#undef mkstemp + +#include <stdio.h> +#include <stdlib.h> + +#ifndef __GT_FILE +# define __GT_FILE 0 +#endif + +int __gen_tempname (); + +/* Generate a unique temporary file name from TEMPLATE. + The last six characters of TEMPLATE must be "XXXXXX"; + they are replaced with a string that makes the file name unique. + Then open the file and return a fd. */ +int +rpl_mkstemp (char *template) +{ + return __gen_tempname (template, __GT_FILE); +} diff --git a/lib/mktime.c b/lib/mktime.c new file mode 100644 index 0000000..b2d9e52 --- /dev/null +++ b/lib/mktime.c @@ -0,0 +1,664 @@ +/* Convert a `struct tm' to a time_t value. + Copyright (C) 1993-1999, 2002-2004, 2005 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by Paul Eggert (eggert@twinsun.com). + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Define this to have a standalone program to test this implementation of + mktime. */ +/* #define DEBUG 1 */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Assume that leap seconds are possible, unless told otherwise. + If the host has a `zic' command with a `-L leapsecondfilename' option, + then it supports leap seconds; otherwise it probably doesn't. */ +#ifndef LEAP_SECONDS_POSSIBLE +# define LEAP_SECONDS_POSSIBLE 1 +#endif + +#include <sys/types.h> /* Some systems define `time_t' here. */ +#include <time.h> + +#include <limits.h> + +#include <string.h> /* For the real memcpy prototype. */ + +#if DEBUG +# include <stdio.h> +# include <stdlib.h> +/* Make it work even if the system's libc has its own mktime routine. */ +# define mktime my_mktime +#endif /* DEBUG */ + +/* Shift A right by B bits portably, by dividing A by 2**B and + truncating towards minus infinity. A and B should be free of side + effects, and B should be in the range 0 <= B <= INT_BITS - 2, where + INT_BITS is the number of useful bits in an int. GNU code can + assume that INT_BITS is at least 32. + + ISO C99 says that A >> B is implementation-defined if A < 0. Some + implementations (e.g., UNICOS 9.0 on a Cray Y-MP EL) don't shift + right in the usual way when A < 0, so SHR falls back on division if + ordinary A >> B doesn't seem to be the usual signed shift. */ +#define SHR(a, b) \ + (-1 >> 1 == -1 \ + ? (a) >> (b) \ + : (a) / (1 << (b)) - ((a) % (1 << (b)) < 0)) + +/* The extra casts in the following macros work around compiler bugs, + e.g., in Cray C 5.0.3.0. */ + +/* True if the arithmetic type T is an integer type. bool counts as + an integer. */ +#define TYPE_IS_INTEGER(t) ((t) 1.5 == 1) + +/* True if negative values of the signed integer type T use two's + complement, ones' complement, or signed magnitude representation, + respectively. Much GNU code assumes two's complement, but some + people like to be portable to all possible C hosts. */ +#define TYPE_TWOS_COMPLEMENT(t) ((t) ~ (t) 0 == (t) -1) +#define TYPE_ONES_COMPLEMENT(t) ((t) ~ (t) 0 == 0) +#define TYPE_SIGNED_MAGNITUDE(t) ((t) ~ (t) 0 < (t) -1) + +/* True if the arithmetic type T is signed. */ +#define TYPE_SIGNED(t) (! ((t) 0 < (t) -1)) + +/* The maximum and minimum values for the integer type T. These + macros have undefined behavior if T is signed and has padding bits. + If this is a problem for you, please let us know how to fix it for + your host. */ +#define TYPE_MINIMUM(t) \ + ((t) (! TYPE_SIGNED (t) \ + ? (t) 0 \ + : TYPE_SIGNED_MAGNITUDE (t) \ + ? ~ (t) 0 \ + : ~ (t) 0 << (sizeof (t) * CHAR_BIT - 1))) +#define TYPE_MAXIMUM(t) \ + ((t) (! TYPE_SIGNED (t) \ + ? (t) -1 \ + : ~ (~ (t) 0 << (sizeof (t) * CHAR_BIT - 1)))) + +#ifndef TIME_T_MIN +# define TIME_T_MIN TYPE_MINIMUM (time_t) +#endif +#ifndef TIME_T_MAX +# define TIME_T_MAX TYPE_MAXIMUM (time_t) +#endif +#define TIME_T_MIDPOINT (SHR (TIME_T_MIN + TIME_T_MAX, 1) + 1) + +/* Verify a requirement at compile-time (unlike assert, which is runtime). */ +#define verify(name, assertion) struct name { char a[(assertion) ? 1 : -1]; } + +verify (time_t_is_integer, TYPE_IS_INTEGER (time_t)); +verify (twos_complement_arithmetic, TYPE_TWOS_COMPLEMENT (int)); +/* The code also assumes that signed integer overflow silently wraps + around, but this assumption can't be stated without causing a + diagnostic on some hosts. */ + +#define EPOCH_YEAR 1970 +#define TM_YEAR_BASE 1900 +verify (base_year_is_a_multiple_of_100, TM_YEAR_BASE % 100 == 0); + +/* Return 1 if YEAR + TM_YEAR_BASE is a leap year. */ +static inline int +leapyear (long int year) +{ + /* Don't add YEAR to TM_YEAR_BASE, as that might overflow. + Also, work even if YEAR is negative. */ + return + ((year & 3) == 0 + && (year % 100 != 0 + || ((year / 100) & 3) == (- (TM_YEAR_BASE / 100) & 3))); +} + +/* How many days come before each month (0-12). */ +#ifndef _LIBC +static +#endif +const unsigned short int __mon_yday[2][13] = + { + /* Normal years. */ + { 0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334, 365 }, + /* Leap years. */ + { 0, 31, 60, 91, 121, 152, 182, 213, 244, 274, 305, 335, 366 } + }; + + +#ifndef _LIBC +/* Portable standalone applications should supply a "time_r.h" that + declares a POSIX-compliant localtime_r, for the benefit of older + implementations that lack localtime_r or have a nonstandard one. + See the gnulib time_r module for one way to implement this. */ +# include "time_r.h" +# undef __localtime_r +# define __localtime_r localtime_r +# define __mktime_internal mktime_internal +#endif + +/* Return an integer value measuring (YEAR1-YDAY1 HOUR1:MIN1:SEC1) - + (YEAR0-YDAY0 HOUR0:MIN0:SEC0) in seconds, assuming that the clocks + were not adjusted between the time stamps. + + The YEAR values uses the same numbering as TP->tm_year. Values + need not be in the usual range. However, YEAR1 must not be less + than 2 * INT_MIN or greater than 2 * INT_MAX. + + The result may overflow. It is the caller's responsibility to + detect overflow. */ + +static inline time_t +ydhms_diff (long int year1, long int yday1, int hour1, int min1, int sec1, + int year0, int yday0, int hour0, int min0, int sec0) +{ + verify (C99_integer_division, -1 / 2 == 0); + verify (long_int_year_and_yday_are_wide_enough, + INT_MAX <= LONG_MAX / 2 || TIME_T_MAX <= UINT_MAX); + + /* Compute intervening leap days correctly even if year is negative. + Take care to avoid integer overflow here. */ + int a4 = SHR (year1, 2) + SHR (TM_YEAR_BASE, 2) - ! (year1 & 3); + int b4 = SHR (year0, 2) + SHR (TM_YEAR_BASE, 2) - ! (year0 & 3); + int a100 = a4 / 25 - (a4 % 25 < 0); + int b100 = b4 / 25 - (b4 % 25 < 0); + int a400 = SHR (a100, 2); + int b400 = SHR (b100, 2); + int intervening_leap_days = (a4 - b4) - (a100 - b100) + (a400 - b400); + + /* Compute the desired time in time_t precision. Overflow might + occur here. */ + time_t tyear1 = year1; + time_t years = tyear1 - year0; + time_t days = 365 * years + yday1 - yday0 + intervening_leap_days; + time_t hours = 24 * days + hour1 - hour0; + time_t minutes = 60 * hours + min1 - min0; + time_t seconds = 60 * minutes + sec1 - sec0; + return seconds; +} + + +/* Return a time_t value corresponding to (YEAR-YDAY HOUR:MIN:SEC), + assuming that *T corresponds to *TP and that no clock adjustments + occurred between *TP and the desired time. + If TP is null, return a value not equal to *T; this avoids false matches. + If overflow occurs, yield the minimal or maximal value, except do not + yield a value equal to *T. */ +static time_t +guess_time_tm (long int year, long int yday, int hour, int min, int sec, + const time_t *t, const struct tm *tp) +{ + if (tp) + { + time_t d = ydhms_diff (year, yday, hour, min, sec, + tp->tm_year, tp->tm_yday, + tp->tm_hour, tp->tm_min, tp->tm_sec); + time_t t1 = *t + d; + if ((t1 < *t) == (TYPE_SIGNED (time_t) ? d < 0 : TIME_T_MAX / 2 < d)) + return t1; + } + + /* Overflow occurred one way or another. Return the nearest result + that is actually in range, except don't report a zero difference + if the actual difference is nonzero, as that would cause a false + match. */ + return (*t < TIME_T_MIDPOINT + ? TIME_T_MIN + (*t == TIME_T_MIN) + : TIME_T_MAX - (*t == TIME_T_MAX)); +} + +/* Use CONVERT to convert *T to a broken down time in *TP. + If *T is out of range for conversion, adjust it so that + it is the nearest in-range value and then convert that. */ +static struct tm * +ranged_convert (struct tm *(*convert) (const time_t *, struct tm *), + time_t *t, struct tm *tp) +{ + struct tm *r = convert (t, tp); + + if (!r && *t) + { + time_t bad = *t; + time_t ok = 0; + + /* BAD is a known unconvertible time_t, and OK is a known good one. + Use binary search to narrow the range between BAD and OK until + they differ by 1. */ + while (bad != ok + (bad < 0 ? -1 : 1)) + { + time_t mid = *t = (bad < 0 + ? bad + ((ok - bad) >> 1) + : ok + ((bad - ok) >> 1)); + r = convert (t, tp); + if (r) + ok = mid; + else + bad = mid; + } + + if (!r && ok) + { + /* The last conversion attempt failed; + revert to the most recent successful attempt. */ + *t = ok; + r = convert (t, tp); + } + } + + return r; +} + + +/* Convert *TP to a time_t value, inverting + the monotonic and mostly-unit-linear conversion function CONVERT. + Use *OFFSET to keep track of a guess at the offset of the result, + compared to what the result would be for UTC without leap seconds. + If *OFFSET's guess is correct, only one CONVERT call is needed. + This function is external because it is used also by timegm.c. */ +time_t +__mktime_internal (struct tm *tp, + struct tm *(*convert) (const time_t *, struct tm *), + time_t *offset) +{ + time_t t, gt, t0, t1, t2; + struct tm tm; + + /* The maximum number of probes (calls to CONVERT) should be enough + to handle any combinations of time zone rule changes, solar time, + leap seconds, and oscillations around a spring-forward gap. + POSIX.1 prohibits leap seconds, but some hosts have them anyway. */ + int remaining_probes = 6; + + /* Time requested. Copy it in case CONVERT modifies *TP; this can + occur if TP is localtime's returned value and CONVERT is localtime. */ + int sec = tp->tm_sec; + int min = tp->tm_min; + int hour = tp->tm_hour; + int mday = tp->tm_mday; + int mon = tp->tm_mon; + int year_requested = tp->tm_year; + int isdst = tp->tm_isdst; + + /* 1 if the previous probe was DST. */ + int dst2; + + /* Ensure that mon is in range, and set year accordingly. */ + int mon_remainder = mon % 12; + int negative_mon_remainder = mon_remainder < 0; + int mon_years = mon / 12 - negative_mon_remainder; + long int lyear_requested = year_requested; + long int year = lyear_requested + mon_years; + + /* The other values need not be in range: + the remaining code handles minor overflows correctly, + assuming int and time_t arithmetic wraps around. + Major overflows are caught at the end. */ + + /* Calculate day of year from year, month, and day of month. + The result need not be in range. */ + int mon_yday = ((__mon_yday[leapyear (year)] + [mon_remainder + 12 * negative_mon_remainder]) + - 1); + long int lmday = mday; + long int yday = mon_yday + lmday; + + time_t guessed_offset = *offset; + + int sec_requested = sec; + + if (LEAP_SECONDS_POSSIBLE) + { + /* Handle out-of-range seconds specially, + since ydhms_tm_diff assumes every minute has 60 seconds. */ + if (sec < 0) + sec = 0; + if (59 < sec) + sec = 59; + } + + /* Invert CONVERT by probing. First assume the same offset as last + time. */ + + t0 = ydhms_diff (year, yday, hour, min, sec, + EPOCH_YEAR - TM_YEAR_BASE, 0, 0, 0, - guessed_offset); + + if (TIME_T_MAX / INT_MAX / 366 / 24 / 60 / 60 < 3) + { + /* time_t isn't large enough to rule out overflows, so check + for major overflows. A gross check suffices, since if t0 + has overflowed, it is off by a multiple of TIME_T_MAX - + TIME_T_MIN + 1. So ignore any component of the difference + that is bounded by a small value. */ + + /* Approximate log base 2 of the number of time units per + biennium. A biennium is 2 years; use this unit instead of + years to avoid integer overflow. For example, 2 average + Gregorian years are 2 * 365.2425 * 24 * 60 * 60 seconds, + which is 63113904 seconds, and rint (log2 (63113904)) is + 26. */ + int ALOG2_SECONDS_PER_BIENNIUM = 26; + int ALOG2_MINUTES_PER_BIENNIUM = 20; + int ALOG2_HOURS_PER_BIENNIUM = 14; + int ALOG2_DAYS_PER_BIENNIUM = 10; + int LOG2_YEARS_PER_BIENNIUM = 1; + + int approx_requested_biennia = + (SHR (year_requested, LOG2_YEARS_PER_BIENNIUM) + - SHR (EPOCH_YEAR - TM_YEAR_BASE, LOG2_YEARS_PER_BIENNIUM) + + SHR (mday, ALOG2_DAYS_PER_BIENNIUM) + + SHR (hour, ALOG2_HOURS_PER_BIENNIUM) + + SHR (min, ALOG2_MINUTES_PER_BIENNIUM) + + (LEAP_SECONDS_POSSIBLE + ? 0 + : SHR (sec, ALOG2_SECONDS_PER_BIENNIUM))); + + int approx_biennia = SHR (t0, ALOG2_SECONDS_PER_BIENNIUM); + int diff = approx_biennia - approx_requested_biennia; + int abs_diff = diff < 0 ? - diff : diff; + + /* IRIX 4.0.5 cc miscaculates TIME_T_MIN / 3: it erroneously + gives a positive value of 715827882. Setting a variable + first then doing math on it seems to work. + (ghazi@caip.rutgers.edu) */ + time_t time_t_max = TIME_T_MAX; + time_t time_t_min = TIME_T_MIN; + time_t overflow_threshold = + (time_t_max / 3 - time_t_min / 3) >> ALOG2_SECONDS_PER_BIENNIUM; + + if (overflow_threshold < abs_diff) + { + /* Overflow occurred. Try repairing it; this might work if + the time zone offset is enough to undo the overflow. */ + time_t repaired_t0 = -1 - t0; + approx_biennia = SHR (repaired_t0, ALOG2_SECONDS_PER_BIENNIUM); + diff = approx_biennia - approx_requested_biennia; + abs_diff = diff < 0 ? - diff : diff; + if (overflow_threshold < abs_diff) + return -1; + guessed_offset += repaired_t0 - t0; + t0 = repaired_t0; + } + } + + /* Repeatedly use the error to improve the guess. */ + + for (t = t1 = t2 = t0, dst2 = 0; + (gt = guess_time_tm (year, yday, hour, min, sec, &t, + ranged_convert (convert, &t, &tm)), + t != gt); + t1 = t2, t2 = t, t = gt, dst2 = tm.tm_isdst != 0) + if (t == t1 && t != t2 + && (tm.tm_isdst < 0 + || (isdst < 0 + ? dst2 <= (tm.tm_isdst != 0) + : (isdst != 0) != (tm.tm_isdst != 0)))) + /* We can't possibly find a match, as we are oscillating + between two values. The requested time probably falls + within a spring-forward gap of size GT - T. Follow the common + practice in this case, which is to return a time that is GT - T + away from the requested time, preferring a time whose + tm_isdst differs from the requested value. (If no tm_isdst + was requested and only one of the two values has a nonzero + tm_isdst, prefer that value.) In practice, this is more + useful than returning -1. */ + goto offset_found; + else if (--remaining_probes == 0) + return -1; + + /* We have a match. Check whether tm.tm_isdst has the requested + value, if any. */ + if (isdst != tm.tm_isdst && 0 <= isdst && 0 <= tm.tm_isdst) + { + /* tm.tm_isdst has the wrong value. Look for a neighboring + time with the right value, and use its UTC offset. + + Heuristic: probe the adjacent timestamps in both directions, + looking for the desired isdst. This should work for all real + time zone histories in the tz database. */ + + /* Distance between probes when looking for a DST boundary. In + tzdata2003a, the shortest period of DST is 601200 seconds + (e.g., America/Recife starting 2000-10-08 01:00), and the + shortest period of non-DST surrounded by DST is 694800 + seconds (Africa/Tunis starting 1943-04-17 01:00). Use the + minimum of these two values, so we don't miss these short + periods when probing. */ + int stride = 601200; + + /* The longest period of DST in tzdata2003a is 536454000 seconds + (e.g., America/Jujuy starting 1946-10-01 01:00). The longest + period of non-DST is much longer, but it makes no real sense + to search for more than a year of non-DST, so use the DST + max. */ + int duration_max = 536454000; + + /* Search in both directions, so the maximum distance is half + the duration; add the stride to avoid off-by-1 problems. */ + int delta_bound = duration_max / 2 + stride; + + int delta, direction; + + for (delta = stride; delta < delta_bound; delta += stride) + for (direction = -1; direction <= 1; direction += 2) + { + time_t ot = t + delta * direction; + if ((ot < t) == (direction < 0)) + { + struct tm otm; + ranged_convert (convert, &ot, &otm); + if (otm.tm_isdst == isdst) + { + /* We found the desired tm_isdst. + Extrapolate back to the desired time. */ + t = guess_time_tm (year, yday, hour, min, sec, &ot, &otm); + ranged_convert (convert, &t, &tm); + goto offset_found; + } + } + } + } + + offset_found: + *offset = guessed_offset + t - t0; + + if (LEAP_SECONDS_POSSIBLE && sec_requested != tm.tm_sec) + { + /* Adjust time to reflect the tm_sec requested, not the normalized value. + Also, repair any damage from a false match due to a leap second. */ + int sec_adjustment = (sec == 0 && tm.tm_sec == 60) - sec; + t1 = t + sec_requested; + t2 = t1 + sec_adjustment; + if (((t1 < t) != (sec_requested < 0)) + | ((t2 < t1) != (sec_adjustment < 0)) + | ! convert (&t2, &tm)) + return -1; + t = t2; + } + + *tp = tm; + return t; +} + + +/* FIXME: This should use a signed type wide enough to hold any UTC + offset in seconds. 'int' should be good enough for GNU code. We + can't fix this unilaterally though, as other modules invoke + __mktime_internal. */ +static time_t localtime_offset; + +/* Convert *TP to a time_t value. */ +time_t +mktime (struct tm *tp) +{ +#ifdef _LIBC + /* POSIX.1 8.1.1 requires that whenever mktime() is called, the + time zone names contained in the external variable `tzname' shall + be set as if the tzset() function had been called. */ + __tzset (); +#endif + + return __mktime_internal (tp, __localtime_r, &localtime_offset); +} + +#ifdef weak_alias +weak_alias (mktime, timelocal) +#endif + +#ifdef _LIBC +libc_hidden_def (mktime) +libc_hidden_weak (timelocal) +#endif + +#if DEBUG + +static int +not_equal_tm (const struct tm *a, const struct tm *b) +{ + return ((a->tm_sec ^ b->tm_sec) + | (a->tm_min ^ b->tm_min) + | (a->tm_hour ^ b->tm_hour) + | (a->tm_mday ^ b->tm_mday) + | (a->tm_mon ^ b->tm_mon) + | (a->tm_year ^ b->tm_year) + | (a->tm_yday ^ b->tm_yday) + | (a->tm_isdst ^ b->tm_isdst)); +} + +static void +print_tm (const struct tm *tp) +{ + if (tp) + printf ("%04d-%02d-%02d %02d:%02d:%02d yday %03d wday %d isdst %d", + tp->tm_year + TM_YEAR_BASE, tp->tm_mon + 1, tp->tm_mday, + tp->tm_hour, tp->tm_min, tp->tm_sec, + tp->tm_yday, tp->tm_wday, tp->tm_isdst); + else + printf ("0"); +} + +static int +check_result (time_t tk, struct tm tmk, time_t tl, const struct tm *lt) +{ + if (tk != tl || !lt || not_equal_tm (&tmk, lt)) + { + printf ("mktime ("); + print_tm (lt); + printf (")\nyields ("); + print_tm (&tmk); + printf (") == %ld, should be %ld\n", (long int) tk, (long int) tl); + return 1; + } + + return 0; +} + +int +main (int argc, char **argv) +{ + int status = 0; + struct tm tm, tmk, tml; + struct tm *lt; + time_t tk, tl, tl1; + char trailer; + + if ((argc == 3 || argc == 4) + && (sscanf (argv[1], "%d-%d-%d%c", + &tm.tm_year, &tm.tm_mon, &tm.tm_mday, &trailer) + == 3) + && (sscanf (argv[2], "%d:%d:%d%c", + &tm.tm_hour, &tm.tm_min, &tm.tm_sec, &trailer) + == 3)) + { + tm.tm_year -= TM_YEAR_BASE; + tm.tm_mon--; + tm.tm_isdst = argc == 3 ? -1 : atoi (argv[3]); + tmk = tm; + tl = mktime (&tmk); + lt = localtime (&tl); + if (lt) + { + tml = *lt; + lt = &tml; + } + printf ("mktime returns %ld == ", (long int) tl); + print_tm (&tmk); + printf ("\n"); + status = check_result (tl, tmk, tl, lt); + } + else if (argc == 4 || (argc == 5 && strcmp (argv[4], "-") == 0)) + { + time_t from = atol (argv[1]); + time_t by = atol (argv[2]); + time_t to = atol (argv[3]); + + if (argc == 4) + for (tl = from; by < 0 ? to <= tl : tl <= to; tl = tl1) + { + lt = localtime (&tl); + if (lt) + { + tmk = tml = *lt; + tk = mktime (&tmk); + status |= check_result (tk, tmk, tl, &tml); + } + else + { + printf ("localtime (%ld) yields 0\n", (long int) tl); + status = 1; + } + tl1 = tl + by; + if ((tl1 < tl) != (by < 0)) + break; + } + else + for (tl = from; by < 0 ? to <= tl : tl <= to; tl = tl1) + { + /* Null benchmark. */ + lt = localtime (&tl); + if (lt) + { + tmk = tml = *lt; + tk = tl; + status |= check_result (tk, tmk, tl, &tml); + } + else + { + printf ("localtime (%ld) yields 0\n", (long int) tl); + status = 1; + } + tl1 = tl + by; + if ((tl1 < tl) != (by < 0)) + break; + } + } + else + printf ("Usage:\ +\t%s YYYY-MM-DD HH:MM:SS [ISDST] # Test given time.\n\ +\t%s FROM BY TO # Test values FROM, FROM+BY, ..., TO.\n\ +\t%s FROM BY TO - # Do not test those values (for benchmark).\n", + argv[0], argv[0], argv[0]); + + return status; +} + +#endif /* DEBUG */ + +/* +Local Variables: +compile-command: "gcc -DDEBUG -Wall -W -O -g mktime.c -o mktime" +End: +*/ diff --git a/lib/nanosleep.c b/lib/nanosleep.c new file mode 100644 index 0000000..fa20785 --- /dev/null +++ b/lib/nanosleep.c @@ -0,0 +1,124 @@ +/* Provide a replacement for the POSIX nanosleep function. + Copyright (C) 1999, 2000, 2002, 2004, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* written by Jim Meyering */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Undefine nanosleep here so any prototype is not redefined to be a + prototype for rpl_nanosleep. (they'd conflict e.g., on alpha-dec-osf3.2) */ +#undef nanosleep + +#include <stdbool.h> +#include <stdio.h> +#include <sys/types.h> +#include <signal.h> + +#include <errno.h> + +#if HAVE_UNISTD_H +# include <unistd.h> +#endif + +#include "timespec.h" + +/* Some systems (MSDOS) don't have SIGCONT. + Using SIGTERM here turns the signal-handling code below + into a no-op on such systems. */ +#ifndef SIGCONT +# define SIGCONT SIGTERM +#endif + +#if ! HAVE_SIGINTERRUPT +# define siginterrupt(sig, flag) /* empty */ +#endif + +static sig_atomic_t volatile suspended; + +/* Handle SIGCONT. */ + +static void +sighandler (int sig) +{ + suspended = 1; +} + +/* FIXME: comment */ + +static void +my_usleep (const struct timespec *ts_delay) +{ + struct timeval tv_delay; + tv_delay.tv_sec = ts_delay->tv_sec; + tv_delay.tv_usec = (ts_delay->tv_nsec + 999) / 1000; + if (tv_delay.tv_usec == 1000000) + { + tv_delay.tv_sec++; + tv_delay.tv_usec = 0; + } + select (0, NULL, NULL, NULL, &tv_delay); +} + +/* FIXME: comment */ + +int +rpl_nanosleep (const struct timespec *requested_delay, + struct timespec *remaining_delay) +{ + static bool initialized; + + /* set up sig handler */ + if (! initialized) + { +#ifdef SA_NOCLDSTOP + struct sigaction oldact, newact; + newact.sa_handler = sighandler; + sigemptyset (&newact.sa_mask); + newact.sa_flags = 0; + + sigaction (SIGCONT, NULL, &oldact); + if (oldact.sa_handler != SIG_IGN) + sigaction (SIGCONT, &newact, NULL); +#else + if (signal (SIGCONT, SIG_IGN) != SIG_IGN) + { + signal (SIGCONT, sighandler); + siginterrupt (SIGCONT, 1); + } +#endif + initialized = true; + } + + suspended = 0; + + my_usleep (requested_delay); + + if (suspended) + { + /* Calculate time remaining. */ + /* FIXME: the code in sleep doesn't use this, so there's no + rush to implement it. */ + + errno = EINTR; + } + + /* FIXME: Restore sig handler? */ + + return suspended; +} diff --git a/lib/openat.c b/lib/openat.c new file mode 100644 index 0000000..443cb3d --- /dev/null +++ b/lib/openat.c @@ -0,0 +1,181 @@ +/* provide a replacement openat function + Copyright (C) 2004, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* written by Jim Meyering */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "openat.h" + +#include <stdlib.h> +#include <stdarg.h> +#include <unistd.h> +#include <errno.h> +#include <fcntl.h> + +#include "error.h" +#include "exitfail.h" +#include "save-cwd.h" + +#include "gettext.h" +#define _(msgid) gettext (msgid) + +/* Replacement for Solaris' openat function. + <http://www.google.com/search?q=openat+site:docs.sun.com> + Simulate it by doing save_cwd/fchdir/open/restore_cwd. + If either the save_cwd or the restore_cwd fails (relatively unlikely, + and usually indicative of a problem that deserves close attention), + then give a diagnostic and exit nonzero. + Otherwise, upon failure, set errno and return -1, as openat does. + Upon successful completion, return a file descriptor. */ +int +rpl_openat (int fd, char const *file, int flags, ...) +{ + struct saved_cwd saved_cwd; + int saved_errno; + int new_fd; + mode_t mode = 0; + + if (flags & O_CREAT) + { + va_list arg; + va_start (arg, flags); + + /* Assume that mode_t is passed compatibly with mode_t's type + after argument promotion. */ + mode = va_arg (arg, mode_t); + + va_end (arg); + } + + if (fd == AT_FDCWD || *file == '/') + return open (file, flags, mode); + + if (save_cwd (&saved_cwd) != 0) + error (exit_failure, errno, + _("openat: unable to record current working directory")); + + if (fchdir (fd) != 0) + { + saved_errno = errno; + free_cwd (&saved_cwd); + errno = saved_errno; + return -1; + } + + new_fd = open (file, flags, mode); + saved_errno = errno; + + if (restore_cwd (&saved_cwd) != 0) + error (exit_failure, errno, + _("openat: unable to restore working directory")); + + free_cwd (&saved_cwd); + + errno = saved_errno; + return new_fd; +} + +/* Replacement for Solaris' function by the same name. + <http://www.google.com/search?q=fdopendir+site:docs.sun.com> + Simulate it by doing save_cwd/fchdir/opendir(".")/restore_cwd. + If either the save_cwd or the restore_cwd fails (relatively unlikely, + and usually indicative of a problem that deserves close attention), + then give a diagnostic and exit nonzero. + Otherwise, this function works just like Solaris' fdopendir. */ +DIR * +fdopendir (int fd) +{ + struct saved_cwd saved_cwd; + int saved_errno; + DIR *dir; + + if (fd == AT_FDCWD) + return opendir ("."); + + if (save_cwd (&saved_cwd) != 0) + error (exit_failure, errno, + _("fdopendir: unable to record current working directory")); + + if (fchdir (fd) != 0) + { + saved_errno = errno; + free_cwd (&saved_cwd); + errno = saved_errno; + return NULL; + } + + dir = opendir ("."); + saved_errno = errno; + + if (restore_cwd (&saved_cwd) != 0) + error (exit_failure, errno, + _("fdopendir: unable to restore working directory")); + + free_cwd (&saved_cwd); + + errno = saved_errno; + return dir; +} + +/* Replacement for Solaris' function by the same name. + <http://www.google.com/search?q=fstatat+site:docs.sun.com> + Simulate it by doing save_cwd/fchdir/(stat|lstat)/restore_cwd. + If either the save_cwd or the restore_cwd fails (relatively unlikely, + and usually indicative of a problem that deserves close attention), + then give a diagnostic and exit nonzero. + Otherwise, this function works just like Solaris' fstatat. */ +int +fstatat (int fd, char const *file, struct stat *st, int flag) +{ + struct saved_cwd saved_cwd; + int saved_errno; + int err; + + if (fd == AT_FDCWD) + return (flag == AT_SYMLINK_NOFOLLOW + ? lstat (file, st) + : stat (file, st)); + + if (save_cwd (&saved_cwd) != 0) + error (exit_failure, errno, + _("fstatat: unable to record current working directory")); + + if (fchdir (fd) != 0) + { + saved_errno = errno; + free_cwd (&saved_cwd); + errno = saved_errno; + return -1; + } + + err = (flag == AT_SYMLINK_NOFOLLOW + ? lstat (file, st) + : stat (file, st)); + saved_errno = errno; + + if (restore_cwd (&saved_cwd) != 0) + error (exit_failure, errno, + _("fstatat: unable to restore working directory")); + + free_cwd (&saved_cwd); + + errno = saved_errno; + return err; +} diff --git a/lib/openat.h b/lib/openat.h new file mode 100644 index 0000000..76ad46b --- /dev/null +++ b/lib/openat.h @@ -0,0 +1,44 @@ +/* provide a replacement openat function + Copyright (C) 2004, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* written by Jim Meyering */ + +#include <fcntl.h> + +#include <sys/types.h> +#include <sys/stat.h> +#include <dirent.h> +#include <unistd.h> + +#ifndef AT_FDCWD +# define AT_FDCWD (-3041965) /* same value as Solaris 9 */ +# define AT_SYMLINK_NOFOLLOW 4096 /* same value as Solaris 9 */ + +# ifdef __OPENAT_PREFIX +# undef openat +# define __OPENAT_CONCAT(x, y) x ## y +# define __OPENAT_XCONCAT(x, y) __OPENAT_CONCAT (x, y) +# define __OPENAT_ID(y) __OPENAT_XCONCAT (__OPENAT_PREFIX, y) +# define openat __OPENAT_ID (openat) +int openat (int fd, char const *file, int flags, /* mode_t mode */ ...); +# define fdopendir __OPENAT_ID (fdopendir) +DIR *fdopendir (int fd); +# define fstatat __OPENAT_ID (fstatat) +int fstatat (int fd, char const *file, struct stat *st, int flag); +# endif + +#endif diff --git a/lib/pagealign_alloc.c b/lib/pagealign_alloc.c new file mode 100644 index 0000000..28e8209 --- /dev/null +++ b/lib/pagealign_alloc.c @@ -0,0 +1,198 @@ +/* Memory allocation aligned to system page boundaries. + + Copyright (C) 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify it + under the terms of the GNU General Public License as published + by the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + General Public License for more details. + + You should have received a copy of the GNU General Public + License along with this program; if not, write to the Free Software + Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, + USA. */ + +/* Written by Derek R. Price <derek@ximbiot.com>. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "pagealign_alloc.h" + +#include <errno.h> +#include <stdlib.h> + +#include <fcntl.h> + +#if HAVE_UNISTD_H +# include <unistd.h> +#endif + +#if HAVE_MMAP +# include <sys/mman.h> +#endif + +#include "error.h" +#include "exit.h" +#include "getpagesize.h" +#include "xalloc.h" +#include "gettext.h" + +#define _(str) gettext (str) + +#if HAVE_MMAP +/* Define MAP_FILE when it isn't otherwise. */ +# ifndef MAP_FILE +# define MAP_FILE 0 +# endif +/* Define MAP_FAILED for old systems which neglect to. */ +# ifndef MAP_FAILED +# define MAP_FAILED ((void *)-1) +# endif +#endif + + +#if HAVE_MMAP || ! HAVE_POSIX_MEMALIGN + +# if HAVE_MMAP +/* For each memory region, we store its size. */ +typedef size_t info_t; +# else +/* For each memory region, we store the original pointer returned by + malloc(). */ +typedef void * info_t; +# endif + +/* A simple linked list of allocated memory regions. It is probably not the + most efficient way to store these, but anyway... */ +typedef struct memnode_s memnode_t; +struct memnode_s +{ + void *aligned_ptr; + info_t info; + memnode_t *next; +}; + +/* The list of currently allocated memory regions. */ +static memnode_t *memnode_table = NULL; + + +static void +new_memnode (void *aligned_ptr, info_t info) +{ + memnode_t *new_node = (memnode_t *) xmalloc (sizeof (memnode_t)); + new_node->aligned_ptr = aligned_ptr; + new_node->info = info; + new_node->next = memnode_table; + memnode_table = new_node; +} + + +/* Dispose of the memnode containing a map for the ALIGNED_PTR in question + and return the content of the node's INFO field. */ +static info_t +get_memnode (void *aligned_ptr) +{ + info_t ret; + memnode_t *c; + memnode_t **p_next = &memnode_table; + + for (c = *p_next; c != NULL; p_next = &c->next, c = c->next) + if (c->aligned_ptr == aligned_ptr) + break; + + if (c == NULL) + /* An attempt to free untracked memory. A wrong pointer was passed + to pagealign_free(). */ + abort (); + + /* Remove this entry from the list, save the return value, and free it. */ + *p_next = c->next; + ret = c->info; + free (c); + + return ret; +} + +#endif /* HAVE_MMAP || !HAVE_POSIX_MEMALIGN */ + + +void * +pagealign_alloc (size_t size) +{ + void *ret; +#if HAVE_MMAP +# ifdef HAVE_MAP_ANONYMOUS + const int fd = -1; + const int flags = MAP_ANONYMOUS | MAP_PRIVATE; +# else /* !HAVE_MAP_ANONYMOUS */ + static int fd = -1; /* Only open /dev/zero once in order to avoid limiting + the amount of memory we may allocate based on the + number of open file descriptors. */ + const int flags = MAP_FILE | MAP_PRIVATE; + if (fd == -1) + { + fd = open ("/dev/zero", O_RDONLY, 0666); + if (fd < 0) + error (EXIT_FAILURE, errno, _("Failed to open /dev/zero for read")); + } +# endif /* HAVE_MAP_ANONYMOUS */ + ret = mmap (NULL, size, PROT_READ | PROT_WRITE, flags, fd, 0); + if (ret == MAP_FAILED) + return NULL; + new_memnode (ret, size); +#elif HAVE_POSIX_MEMALIGN + int status = posix_memalign (&ret, getpagesize (), size); + if (status) + { + errno = status; + return NULL; + } +#else /* !HAVE_MMAP && !HAVE_POSIX_MEMALIGN */ + size_t pagesize = getpagesize (); + void *unaligned_ptr = malloc (size + pagesize - 1); + if (unaligned_ptr == NULL) + { + /* Set errno. We don't know whether malloc already set errno: some + implementations of malloc do, some don't. */ + errno = ENOMEM; + return NULL; + } + ret = (char *) unaligned_ptr + + ((- (unsigned long) unaligned_ptr) & (pagesize - 1)); + new_memnode (ret, unaligned_ptr); +#endif /* HAVE_MMAP && HAVE_POSIX_MEMALIGN */ + return ret; +} + + +void * +pagealign_xalloc (size_t size) +{ + void *ret; + + ret = pagealign_alloc (size); + if (ret == NULL) + xalloc_die (); + return ret; +} + + +void +pagealign_free (void *aligned_ptr) +{ +#if HAVE_MMAP + if (munmap (aligned_ptr, get_memnode (aligned_ptr)) < 0) + error (EXIT_FAILURE, errno, "Failed to unmap memory"); +#elif HAVE_POSIX_MEMALIGN + free (aligned_ptr); +#else + free (get_memnode (aligned_ptr)); +#endif +} diff --git a/lib/pagealign_alloc.h b/lib/pagealign_alloc.h new file mode 100644 index 0000000..abad1aa --- /dev/null +++ b/lib/pagealign_alloc.h @@ -0,0 +1,42 @@ +/* Memory allocation aligned to system page boundaries. + + Copyright (C) 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify it + under the terms of the GNU General Public License as published + by the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + General Public License for more details. + + You should have received a copy of the GNU General Public + License along with this program; if not, write to the Free Software + Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, + USA. */ + +#ifndef _PAGEALIGN_ALLOC_H +# define _PAGEALIGN_ALLOC_H + +# include <stddef.h> + +/* Allocate a block of memory of SIZE bytes, aligned on a system page + boundary. + If SIZE is not a multiple of the system page size, it will be rounded up + to the next multiple. + Return a pointer to the start of the memory block. Upon allocation failure, + return NULL and set errno. */ +extern void *pagealign_alloc (size_t size); + +/* Like pagealign_alloc, except it exits the program if the allocation + fails. */ +extern void *pagealign_xalloc (size_t size); + +/* Free a memory block. + PTR must be a non-NULL pointer returned by pagealign_alloc or + pagealign_xalloc. */ +extern void pagealign_free (void *ptr); + +#endif /* _PAGEALIGN_ALLOC_H */ diff --git a/lib/pathmax.h b/lib/pathmax.h new file mode 100644 index 0000000..613c56c --- /dev/null +++ b/lib/pathmax.h @@ -0,0 +1,49 @@ +/* Define PATH_MAX somehow. Requires sys/types.h. + Copyright (C) 1992, 1999, 2001, 2003, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _PATHMAX_H +# define _PATHMAX_H + +# if HAVE_UNISTD_H +# include <unistd.h> +# endif + +# include <limits.h> + +# ifndef _POSIX_PATH_MAX +# define _POSIX_PATH_MAX 256 +# endif + +# if !defined PATH_MAX && defined _PC_PATH_MAX +# define PATH_MAX (pathconf ("/", _PC_PATH_MAX) < 1 ? 1024 \ + : pathconf ("/", _PC_PATH_MAX)) +# endif + +/* Don't include sys/param.h if it already has been. */ +# if defined HAVE_SYS_PARAM_H && !defined PATH_MAX && !defined MAXPATHLEN +# include <sys/param.h> +# endif + +# if !defined PATH_MAX && defined MAXPATHLEN +# define PATH_MAX MAXPATHLEN +# endif + +# ifndef PATH_MAX +# define PATH_MAX _POSIX_PATH_MAX +# endif + +#endif /* _PATHMAX_H */ diff --git a/lib/pipe-safer.c b/lib/pipe-safer.c new file mode 100644 index 0000000..fb02d72 --- /dev/null +++ b/lib/pipe-safer.c @@ -0,0 +1,50 @@ +/* Invoke pipe, but avoid some glitches. + Copyright (C) 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Jim Meyering. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "unistd-safer.h" + +#include <unistd.h> + +/* Like pipe, but ensure that neither of the file descriptors is + STDIN_FILENO, STDOUT_FILENO, or STDERR_FILENO. */ + +int +pipe_safer (int fd[2]) +{ + int fail = pipe (fd); + if (fail) + return fail; + + { + int i; + for (i = 0; i < 2; i++) + { + int f = fd_safer (fd[i]); + if (f < 0) + return -1; + fd[i] = f; + } + } + + return 0; +} diff --git a/lib/printf-args.c b/lib/printf-args.c new file mode 100644 index 0000000..0ed1acb --- /dev/null +++ b/lib/printf-args.c @@ -0,0 +1,118 @@ +/* Decomposed printf argument list. + Copyright (C) 1999, 2002-2003 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Specification. */ +#include "printf-args.h" + +#ifdef STATIC +STATIC +#endif +int +printf_fetchargs (va_list args, arguments *a) +{ + size_t i; + argument *ap; + + for (i = 0, ap = &a->arg[0]; i < a->count; i++, ap++) + switch (ap->type) + { + case TYPE_SCHAR: + ap->a.a_schar = va_arg (args, /*signed char*/ int); + break; + case TYPE_UCHAR: + ap->a.a_uchar = va_arg (args, /*unsigned char*/ int); + break; + case TYPE_SHORT: + ap->a.a_short = va_arg (args, /*short*/ int); + break; + case TYPE_USHORT: + ap->a.a_ushort = va_arg (args, /*unsigned short*/ int); + break; + case TYPE_INT: + ap->a.a_int = va_arg (args, int); + break; + case TYPE_UINT: + ap->a.a_uint = va_arg (args, unsigned int); + break; + case TYPE_LONGINT: + ap->a.a_longint = va_arg (args, long int); + break; + case TYPE_ULONGINT: + ap->a.a_ulongint = va_arg (args, unsigned long int); + break; +#ifdef HAVE_LONG_LONG + case TYPE_LONGLONGINT: + ap->a.a_longlongint = va_arg (args, long long int); + break; + case TYPE_ULONGLONGINT: + ap->a.a_ulonglongint = va_arg (args, unsigned long long int); + break; +#endif + case TYPE_DOUBLE: + ap->a.a_double = va_arg (args, double); + break; +#ifdef HAVE_LONG_DOUBLE + case TYPE_LONGDOUBLE: + ap->a.a_longdouble = va_arg (args, long double); + break; +#endif + case TYPE_CHAR: + ap->a.a_char = va_arg (args, int); + break; +#ifdef HAVE_WINT_T + case TYPE_WIDE_CHAR: + ap->a.a_wide_char = va_arg (args, wint_t); + break; +#endif + case TYPE_STRING: + ap->a.a_string = va_arg (args, const char *); + break; +#ifdef HAVE_WCHAR_T + case TYPE_WIDE_STRING: + ap->a.a_wide_string = va_arg (args, const wchar_t *); + break; +#endif + case TYPE_POINTER: + ap->a.a_pointer = va_arg (args, void *); + break; + case TYPE_COUNT_SCHAR_POINTER: + ap->a.a_count_schar_pointer = va_arg (args, signed char *); + break; + case TYPE_COUNT_SHORT_POINTER: + ap->a.a_count_short_pointer = va_arg (args, short *); + break; + case TYPE_COUNT_INT_POINTER: + ap->a.a_count_int_pointer = va_arg (args, int *); + break; + case TYPE_COUNT_LONGINT_POINTER: + ap->a.a_count_longint_pointer = va_arg (args, long int *); + break; +#ifdef HAVE_LONG_LONG + case TYPE_COUNT_LONGLONGINT_POINTER: + ap->a.a_count_longlongint_pointer = va_arg (args, long long int *); + break; +#endif + default: + /* Unknown type. */ + return -1; + } + return 0; +} diff --git a/lib/printf-args.h b/lib/printf-args.h new file mode 100644 index 0000000..cec1cc6 --- /dev/null +++ b/lib/printf-args.h @@ -0,0 +1,136 @@ +/* Decomposed printf argument list. + Copyright (C) 1999, 2002-2003 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _PRINTF_ARGS_H +#define _PRINTF_ARGS_H + +/* Get size_t. */ +#include <stddef.h> + +/* Get wchar_t. */ +#ifdef HAVE_WCHAR_T +# include <stddef.h> +#endif + +/* Get wint_t. */ +#ifdef HAVE_WINT_T +# include <wchar.h> +#endif + +/* Get va_list. */ +#include <stdarg.h> + + +/* Argument types */ +typedef enum +{ + TYPE_NONE, + TYPE_SCHAR, + TYPE_UCHAR, + TYPE_SHORT, + TYPE_USHORT, + TYPE_INT, + TYPE_UINT, + TYPE_LONGINT, + TYPE_ULONGINT, +#ifdef HAVE_LONG_LONG + TYPE_LONGLONGINT, + TYPE_ULONGLONGINT, +#endif + TYPE_DOUBLE, +#ifdef HAVE_LONG_DOUBLE + TYPE_LONGDOUBLE, +#endif + TYPE_CHAR, +#ifdef HAVE_WINT_T + TYPE_WIDE_CHAR, +#endif + TYPE_STRING, +#ifdef HAVE_WCHAR_T + TYPE_WIDE_STRING, +#endif + TYPE_POINTER, + TYPE_COUNT_SCHAR_POINTER, + TYPE_COUNT_SHORT_POINTER, + TYPE_COUNT_INT_POINTER, + TYPE_COUNT_LONGINT_POINTER +#ifdef HAVE_LONG_LONG +, TYPE_COUNT_LONGLONGINT_POINTER +#endif +} arg_type; + +/* Polymorphic argument */ +typedef struct +{ + arg_type type; + union + { + signed char a_schar; + unsigned char a_uchar; + short a_short; + unsigned short a_ushort; + int a_int; + unsigned int a_uint; + long int a_longint; + unsigned long int a_ulongint; +#ifdef HAVE_LONG_LONG + long long int a_longlongint; + unsigned long long int a_ulonglongint; +#endif + float a_float; + double a_double; +#ifdef HAVE_LONG_DOUBLE + long double a_longdouble; +#endif + int a_char; +#ifdef HAVE_WINT_T + wint_t a_wide_char; +#endif + const char* a_string; +#ifdef HAVE_WCHAR_T + const wchar_t* a_wide_string; +#endif + void* a_pointer; + signed char * a_count_schar_pointer; + short * a_count_short_pointer; + int * a_count_int_pointer; + long int * a_count_longint_pointer; +#ifdef HAVE_LONG_LONG + long long int * a_count_longlongint_pointer; +#endif + } + a; +} +argument; + +typedef struct +{ + size_t count; + argument *arg; +} +arguments; + + +/* Fetch the arguments, putting them into a. */ +#ifdef STATIC +STATIC +#else +extern +#endif +int printf_fetchargs (va_list args, arguments *a); + +#endif /* _PRINTF_ARGS_H */ diff --git a/lib/printf-parse.c b/lib/printf-parse.c new file mode 100644 index 0000000..3d2fb17 --- /dev/null +++ b/lib/printf-parse.c @@ -0,0 +1,536 @@ +/* Formatted output to strings. + Copyright (C) 1999-2000, 2002-2003 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Specification. */ +#if WIDE_CHAR_VERSION +# include "wprintf-parse.h" +#else +# include "printf-parse.h" +#endif + +/* Get size_t, NULL. */ +#include <stddef.h> + +/* Get intmax_t. */ +#if HAVE_STDINT_H_WITH_UINTMAX +# include <stdint.h> +#endif +#if HAVE_INTTYPES_H_WITH_UINTMAX +# include <inttypes.h> +#endif + +/* malloc(), realloc(), free(). */ +#include <stdlib.h> + +/* Checked size_t computations. */ +#include "xsize.h" + +#if WIDE_CHAR_VERSION +# define PRINTF_PARSE wprintf_parse +# define CHAR_T wchar_t +# define DIRECTIVE wchar_t_directive +# define DIRECTIVES wchar_t_directives +#else +# define PRINTF_PARSE printf_parse +# define CHAR_T char +# define DIRECTIVE char_directive +# define DIRECTIVES char_directives +#endif + +#ifdef STATIC +STATIC +#endif +int +PRINTF_PARSE (const CHAR_T *format, DIRECTIVES *d, arguments *a) +{ + const CHAR_T *cp = format; /* pointer into format */ + size_t arg_posn = 0; /* number of regular arguments consumed */ + size_t d_allocated; /* allocated elements of d->dir */ + size_t a_allocated; /* allocated elements of a->arg */ + size_t max_width_length = 0; + size_t max_precision_length = 0; + + d->count = 0; + d_allocated = 1; + d->dir = malloc (d_allocated * sizeof (DIRECTIVE)); + if (d->dir == NULL) + /* Out of memory. */ + return -1; + + a->count = 0; + a_allocated = 0; + a->arg = NULL; + +#define REGISTER_ARG(_index_,_type_) \ + { \ + size_t n = (_index_); \ + if (n >= a_allocated) \ + { \ + size_t memory_size; \ + argument *memory; \ + \ + a_allocated = xtimes (a_allocated, 2); \ + if (a_allocated <= n) \ + a_allocated = xsum (n, 1); \ + memory_size = xtimes (a_allocated, sizeof (argument)); \ + if (size_overflow_p (memory_size)) \ + /* Overflow, would lead to out of memory. */ \ + goto error; \ + memory = (a->arg \ + ? realloc (a->arg, memory_size) \ + : malloc (memory_size)); \ + if (memory == NULL) \ + /* Out of memory. */ \ + goto error; \ + a->arg = memory; \ + } \ + while (a->count <= n) \ + a->arg[a->count++].type = TYPE_NONE; \ + if (a->arg[n].type == TYPE_NONE) \ + a->arg[n].type = (_type_); \ + else if (a->arg[n].type != (_type_)) \ + /* Ambiguous type for positional argument. */ \ + goto error; \ + } + + while (*cp != '\0') + { + CHAR_T c = *cp++; + if (c == '%') + { + size_t arg_index = ARG_NONE; + DIRECTIVE *dp = &d->dir[d->count];/* pointer to next directive */ + + /* Initialize the next directive. */ + dp->dir_start = cp - 1; + dp->flags = 0; + dp->width_start = NULL; + dp->width_end = NULL; + dp->width_arg_index = ARG_NONE; + dp->precision_start = NULL; + dp->precision_end = NULL; + dp->precision_arg_index = ARG_NONE; + dp->arg_index = ARG_NONE; + + /* Test for positional argument. */ + if (*cp >= '0' && *cp <= '9') + { + const CHAR_T *np; + + for (np = cp; *np >= '0' && *np <= '9'; np++) + ; + if (*np == '$') + { + size_t n = 0; + + for (np = cp; *np >= '0' && *np <= '9'; np++) + n = xsum (xtimes (n, 10), *np - '0'); + if (n == 0) + /* Positional argument 0. */ + goto error; + if (size_overflow_p (n)) + /* n too large, would lead to out of memory later. */ + goto error; + arg_index = n - 1; + cp = np + 1; + } + } + + /* Read the flags. */ + for (;;) + { + if (*cp == '\'') + { + dp->flags |= FLAG_GROUP; + cp++; + } + else if (*cp == '-') + { + dp->flags |= FLAG_LEFT; + cp++; + } + else if (*cp == '+') + { + dp->flags |= FLAG_SHOWSIGN; + cp++; + } + else if (*cp == ' ') + { + dp->flags |= FLAG_SPACE; + cp++; + } + else if (*cp == '#') + { + dp->flags |= FLAG_ALT; + cp++; + } + else if (*cp == '0') + { + dp->flags |= FLAG_ZERO; + cp++; + } + else + break; + } + + /* Parse the field width. */ + if (*cp == '*') + { + dp->width_start = cp; + cp++; + dp->width_end = cp; + if (max_width_length < 1) + max_width_length = 1; + + /* Test for positional argument. */ + if (*cp >= '0' && *cp <= '9') + { + const CHAR_T *np; + + for (np = cp; *np >= '0' && *np <= '9'; np++) + ; + if (*np == '$') + { + size_t n = 0; + + for (np = cp; *np >= '0' && *np <= '9'; np++) + n = xsum (xtimes (n, 10), *np - '0'); + if (n == 0) + /* Positional argument 0. */ + goto error; + if (size_overflow_p (n)) + /* n too large, would lead to out of memory later. */ + goto error; + dp->width_arg_index = n - 1; + cp = np + 1; + } + } + if (dp->width_arg_index == ARG_NONE) + { + dp->width_arg_index = arg_posn++; + if (dp->width_arg_index == ARG_NONE) + /* arg_posn wrapped around. */ + goto error; + } + REGISTER_ARG (dp->width_arg_index, TYPE_INT); + } + else if (*cp >= '0' && *cp <= '9') + { + size_t width_length; + + dp->width_start = cp; + for (; *cp >= '0' && *cp <= '9'; cp++) + ; + dp->width_end = cp; + width_length = dp->width_end - dp->width_start; + if (max_width_length < width_length) + max_width_length = width_length; + } + + /* Parse the precision. */ + if (*cp == '.') + { + cp++; + if (*cp == '*') + { + dp->precision_start = cp - 1; + cp++; + dp->precision_end = cp; + if (max_precision_length < 2) + max_precision_length = 2; + + /* Test for positional argument. */ + if (*cp >= '0' && *cp <= '9') + { + const CHAR_T *np; + + for (np = cp; *np >= '0' && *np <= '9'; np++) + ; + if (*np == '$') + { + size_t n = 0; + + for (np = cp; *np >= '0' && *np <= '9'; np++) + n = xsum (xtimes (n, 10), *np - '0'); + if (n == 0) + /* Positional argument 0. */ + goto error; + if (size_overflow_p (n)) + /* n too large, would lead to out of memory + later. */ + goto error; + dp->precision_arg_index = n - 1; + cp = np + 1; + } + } + if (dp->precision_arg_index == ARG_NONE) + { + dp->precision_arg_index = arg_posn++; + if (dp->precision_arg_index == ARG_NONE) + /* arg_posn wrapped around. */ + goto error; + } + REGISTER_ARG (dp->precision_arg_index, TYPE_INT); + } + else + { + size_t precision_length; + + dp->precision_start = cp - 1; + for (; *cp >= '0' && *cp <= '9'; cp++) + ; + dp->precision_end = cp; + precision_length = dp->precision_end - dp->precision_start; + if (max_precision_length < precision_length) + max_precision_length = precision_length; + } + } + + { + arg_type type; + + /* Parse argument type/size specifiers. */ + { + int flags = 0; + + for (;;) + { + if (*cp == 'h') + { + flags |= (1 << (flags & 1)); + cp++; + } + else if (*cp == 'L') + { + flags |= 4; + cp++; + } + else if (*cp == 'l') + { + flags += 8; + cp++; + } +#ifdef HAVE_INTMAX_T + else if (*cp == 'j') + { + if (sizeof (intmax_t) > sizeof (long)) + { + /* intmax_t = long long */ + flags += 16; + } + else if (sizeof (intmax_t) > sizeof (int)) + { + /* intmax_t = long */ + flags += 8; + } + cp++; + } +#endif + else if (*cp == 'z' || *cp == 'Z') + { + /* 'z' is standardized in ISO C 99, but glibc uses 'Z' + because the warning facility in gcc-2.95.2 understands + only 'Z' (see gcc-2.95.2/gcc/c-common.c:1784). */ + if (sizeof (size_t) > sizeof (long)) + { + /* size_t = long long */ + flags += 16; + } + else if (sizeof (size_t) > sizeof (int)) + { + /* size_t = long */ + flags += 8; + } + cp++; + } + else if (*cp == 't') + { + if (sizeof (ptrdiff_t) > sizeof (long)) + { + /* ptrdiff_t = long long */ + flags += 16; + } + else if (sizeof (ptrdiff_t) > sizeof (int)) + { + /* ptrdiff_t = long */ + flags += 8; + } + cp++; + } + else + break; + } + + /* Read the conversion character. */ + c = *cp++; + switch (c) + { + case 'd': case 'i': +#ifdef HAVE_LONG_LONG + if (flags >= 16 || (flags & 4)) + type = TYPE_LONGLONGINT; + else +#endif + if (flags >= 8) + type = TYPE_LONGINT; + else if (flags & 2) + type = TYPE_SCHAR; + else if (flags & 1) + type = TYPE_SHORT; + else + type = TYPE_INT; + break; + case 'o': case 'u': case 'x': case 'X': +#ifdef HAVE_LONG_LONG + if (flags >= 16 || (flags & 4)) + type = TYPE_ULONGLONGINT; + else +#endif + if (flags >= 8) + type = TYPE_ULONGINT; + else if (flags & 2) + type = TYPE_UCHAR; + else if (flags & 1) + type = TYPE_USHORT; + else + type = TYPE_UINT; + break; + case 'f': case 'F': case 'e': case 'E': case 'g': case 'G': + case 'a': case 'A': +#ifdef HAVE_LONG_DOUBLE + if (flags >= 16 || (flags & 4)) + type = TYPE_LONGDOUBLE; + else +#endif + type = TYPE_DOUBLE; + break; + case 'c': + if (flags >= 8) +#ifdef HAVE_WINT_T + type = TYPE_WIDE_CHAR; +#else + goto error; +#endif + else + type = TYPE_CHAR; + break; +#ifdef HAVE_WINT_T + case 'C': + type = TYPE_WIDE_CHAR; + c = 'c'; + break; +#endif + case 's': + if (flags >= 8) +#ifdef HAVE_WCHAR_T + type = TYPE_WIDE_STRING; +#else + goto error; +#endif + else + type = TYPE_STRING; + break; +#ifdef HAVE_WCHAR_T + case 'S': + type = TYPE_WIDE_STRING; + c = 's'; + break; +#endif + case 'p': + type = TYPE_POINTER; + break; + case 'n': +#ifdef HAVE_LONG_LONG + if (flags >= 16 || (flags & 4)) + type = TYPE_COUNT_LONGLONGINT_POINTER; + else +#endif + if (flags >= 8) + type = TYPE_COUNT_LONGINT_POINTER; + else if (flags & 2) + type = TYPE_COUNT_SCHAR_POINTER; + else if (flags & 1) + type = TYPE_COUNT_SHORT_POINTER; + else + type = TYPE_COUNT_INT_POINTER; + break; + case '%': + type = TYPE_NONE; + break; + default: + /* Unknown conversion character. */ + goto error; + } + } + + if (type != TYPE_NONE) + { + dp->arg_index = arg_index; + if (dp->arg_index == ARG_NONE) + { + dp->arg_index = arg_posn++; + if (dp->arg_index == ARG_NONE) + /* arg_posn wrapped around. */ + goto error; + } + REGISTER_ARG (dp->arg_index, type); + } + dp->conversion = c; + dp->dir_end = cp; + } + + d->count++; + if (d->count >= d_allocated) + { + size_t memory_size; + DIRECTIVE *memory; + + d_allocated = xtimes (d_allocated, 2); + memory_size = xtimes (d_allocated, sizeof (DIRECTIVE)); + if (size_overflow_p (memory_size)) + /* Overflow, would lead to out of memory. */ + goto error; + memory = realloc (d->dir, memory_size); + if (memory == NULL) + /* Out of memory. */ + goto error; + d->dir = memory; + } + } + } + d->dir[d->count].dir_start = cp; + + d->max_width_length = max_width_length; + d->max_precision_length = max_precision_length; + return 0; + +error: + if (a->arg) + free (a->arg); + if (d->dir) + free (d->dir); + return -1; +} + +#undef DIRECTIVES +#undef DIRECTIVE +#undef CHAR_T +#undef PRINTF_PARSE diff --git a/lib/printf-parse.h b/lib/printf-parse.h new file mode 100644 index 0000000..82a0d37 --- /dev/null +++ b/lib/printf-parse.h @@ -0,0 +1,74 @@ +/* Parse printf format string. + Copyright (C) 1999, 2002-2003 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _PRINTF_PARSE_H +#define _PRINTF_PARSE_H + +#include "printf-args.h" + + +/* Flags */ +#define FLAG_GROUP 1 /* ' flag */ +#define FLAG_LEFT 2 /* - flag */ +#define FLAG_SHOWSIGN 4 /* + flag */ +#define FLAG_SPACE 8 /* space flag */ +#define FLAG_ALT 16 /* # flag */ +#define FLAG_ZERO 32 + +/* arg_index value indicating that no argument is consumed. */ +#define ARG_NONE (~(size_t)0) + +/* A parsed directive. */ +typedef struct +{ + const char* dir_start; + const char* dir_end; + int flags; + const char* width_start; + const char* width_end; + size_t width_arg_index; + const char* precision_start; + const char* precision_end; + size_t precision_arg_index; + char conversion; /* d i o u x X f e E g G c s p n U % but not C S */ + size_t arg_index; +} +char_directive; + +/* A parsed format string. */ +typedef struct +{ + size_t count; + char_directive *dir; + size_t max_width_length; + size_t max_precision_length; +} +char_directives; + + +/* Parses the format string. Fills in the number N of directives, and fills + in directives[0], ..., directives[N-1], and sets directives[N].dir_start + to the end of the format string. Also fills in the arg_type fields of the + arguments and the needed count of arguments. */ +#ifdef STATIC +STATIC +#else +extern +#endif +int printf_parse (const char *format, char_directives *d, arguments *a); + +#endif /* _PRINTF_PARSE_H */ diff --git a/lib/progname.c b/lib/progname.c new file mode 100644 index 0000000..9f1be28 --- /dev/null +++ b/lib/progname.c @@ -0,0 +1,24 @@ +/* Declare program_name for support functions compiled with the getdate.y test + program. + + Copyright (C) 1990-1998, 2000-2002, 2003 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ + +/* Written by Derek R. Price <derek@ximbiot.com>. */ + +/* This gets set to the name of the executing program. */ +char *program_name = "getdate"; diff --git a/lib/quotearg.c b/lib/quotearg.c new file mode 100644 index 0000000..13bd16f --- /dev/null +++ b/lib/quotearg.c @@ -0,0 +1,679 @@ +/* quotearg.c - quote arguments for output + + Copyright (C) 1998, 1999, 2000, 2001, 2002, 2004, 2005 Free Software + Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Paul Eggert <eggert@twinsun.com> */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "quotearg.h" + +#include "xalloc.h" + +#include <ctype.h> +#include <errno.h> +#include <limits.h> +#include <stdbool.h> +#include <stdlib.h> +#include <string.h> + +#include "gettext.h" +#define _(msgid) gettext (msgid) +#define N_(msgid) msgid + +#if HAVE_WCHAR_H + +/* BSD/OS 4.1 wchar.h requires FILE and struct tm to be declared. */ +# include <stdio.h> +# include <time.h> + +# include <wchar.h> +#endif + +#if !HAVE_MBRTOWC +/* Disable multibyte processing entirely. Since MB_CUR_MAX is 1, the + other macros are defined only for documentation and to satisfy C + syntax. */ +# undef MB_CUR_MAX +# define MB_CUR_MAX 1 +# define mbrtowc(pwc, s, n, ps) ((*(pwc) = *(s)) != 0) +# define iswprint(wc) isprint ((unsigned char) (wc)) +# undef HAVE_MBSINIT +#endif + +#if !defined mbsinit && !HAVE_MBSINIT +# define mbsinit(ps) 1 +#endif + +#ifndef iswprint +# if HAVE_WCTYPE_H +# include <wctype.h> +# endif +# if !defined iswprint && !HAVE_ISWPRINT +# define iswprint(wc) 1 +# endif +#endif + +#ifndef SIZE_MAX +# define SIZE_MAX ((size_t) -1) +#endif + +#define INT_BITS (sizeof (int) * CHAR_BIT) + +struct quoting_options +{ + /* Basic quoting style. */ + enum quoting_style style; + + /* Quote the characters indicated by this bit vector even if the + quoting style would not normally require them to be quoted. */ + unsigned int quote_these_too[(UCHAR_MAX / INT_BITS) + 1]; +}; + +/* Names of quoting styles. */ +char const *const quoting_style_args[] = +{ + "literal", + "shell", + "shell-always", + "c", + "escape", + "locale", + "clocale", + 0 +}; + +/* Correspondences to quoting style names. */ +enum quoting_style const quoting_style_vals[] = +{ + literal_quoting_style, + shell_quoting_style, + shell_always_quoting_style, + c_quoting_style, + escape_quoting_style, + locale_quoting_style, + clocale_quoting_style +}; + +/* The default quoting options. */ +static struct quoting_options default_quoting_options; + +/* Allocate a new set of quoting options, with contents initially identical + to O if O is not null, or to the default if O is null. + It is the caller's responsibility to free the result. */ +struct quoting_options * +clone_quoting_options (struct quoting_options *o) +{ + int e = errno; + struct quoting_options *p = xmalloc (sizeof *p); + *p = *(o ? o : &default_quoting_options); + errno = e; + return p; +} + +/* Get the value of O's quoting style. If O is null, use the default. */ +enum quoting_style +get_quoting_style (struct quoting_options *o) +{ + return (o ? o : &default_quoting_options)->style; +} + +/* In O (or in the default if O is null), + set the value of the quoting style to S. */ +void +set_quoting_style (struct quoting_options *o, enum quoting_style s) +{ + (o ? o : &default_quoting_options)->style = s; +} + +/* In O (or in the default if O is null), + set the value of the quoting options for character C to I. + Return the old value. Currently, the only values defined for I are + 0 (the default) and 1 (which means to quote the character even if + it would not otherwise be quoted). */ +int +set_char_quoting (struct quoting_options *o, char c, int i) +{ + unsigned char uc = c; + unsigned int *p = + (o ? o : &default_quoting_options)->quote_these_too + uc / INT_BITS; + int shift = uc % INT_BITS; + int r = (*p >> shift) & 1; + *p ^= ((i & 1) ^ r) << shift; + return r; +} + +/* MSGID approximates a quotation mark. Return its translation if it + has one; otherwise, return either it or "\"", depending on S. */ +static char const * +gettext_quote (char const *msgid, enum quoting_style s) +{ + char const *translation = _(msgid); + if (translation == msgid && s == clocale_quoting_style) + translation = "\""; + return translation; +} + +/* Place into buffer BUFFER (of size BUFFERSIZE) a quoted version of + argument ARG (of size ARGSIZE), using QUOTING_STYLE and the + non-quoting-style part of O to control quoting. + Terminate the output with a null character, and return the written + size of the output, not counting the terminating null. + If BUFFERSIZE is too small to store the output string, return the + value that would have been returned had BUFFERSIZE been large enough. + If ARGSIZE is SIZE_MAX, use the string length of the argument for ARGSIZE. + + This function acts like quotearg_buffer (BUFFER, BUFFERSIZE, ARG, + ARGSIZE, O), except it uses QUOTING_STYLE instead of the quoting + style specified by O, and O may not be null. */ + +static size_t +quotearg_buffer_restyled (char *buffer, size_t buffersize, + char const *arg, size_t argsize, + enum quoting_style quoting_style, + struct quoting_options const *o) +{ + size_t i; + size_t len = 0; + char const *quote_string = 0; + size_t quote_string_len = 0; + bool backslash_escapes = false; + bool unibyte_locale = MB_CUR_MAX == 1; + +#define STORE(c) \ + do \ + { \ + if (len < buffersize) \ + buffer[len] = (c); \ + len++; \ + } \ + while (0) + + switch (quoting_style) + { + case c_quoting_style: + STORE ('"'); + backslash_escapes = true; + quote_string = "\""; + quote_string_len = 1; + break; + + case escape_quoting_style: + backslash_escapes = true; + break; + + case locale_quoting_style: + case clocale_quoting_style: + { + /* TRANSLATORS: + Get translations for open and closing quotation marks. + + The message catalog should translate "`" to a left + quotation mark suitable for the locale, and similarly for + "'". If the catalog has no translation, + locale_quoting_style quotes `like this', and + clocale_quoting_style quotes "like this". + + For example, an American English Unicode locale should + translate "`" to U+201C (LEFT DOUBLE QUOTATION MARK), and + should translate "'" to U+201D (RIGHT DOUBLE QUOTATION + MARK). A British English Unicode locale should instead + translate these to U+2018 (LEFT SINGLE QUOTATION MARK) and + U+2019 (RIGHT SINGLE QUOTATION MARK), respectively. + + If you don't know what to put here, please see + <http://en.wikipedia.org/wiki/Quotation_mark#Glyphs> + and use glyphs suitable for your language. */ + + char const *left = gettext_quote (N_("`"), quoting_style); + char const *right = gettext_quote (N_("'"), quoting_style); + for (quote_string = left; *quote_string; quote_string++) + STORE (*quote_string); + backslash_escapes = true; + quote_string = right; + quote_string_len = strlen (quote_string); + } + break; + + case shell_always_quoting_style: + STORE ('\''); + quote_string = "'"; + quote_string_len = 1; + break; + + default: + break; + } + + for (i = 0; ! (argsize == SIZE_MAX ? arg[i] == '\0' : i == argsize); i++) + { + unsigned char c; + unsigned char esc; + + if (backslash_escapes + && quote_string_len + && i + quote_string_len <= argsize + && memcmp (arg + i, quote_string, quote_string_len) == 0) + STORE ('\\'); + + c = arg[i]; + switch (c) + { + case '\0': + if (backslash_escapes) + { + STORE ('\\'); + STORE ('0'); + STORE ('0'); + c = '0'; + } + break; + + case '?': + switch (quoting_style) + { + case shell_quoting_style: + goto use_shell_always_quoting_style; + + case c_quoting_style: + if (i + 2 < argsize && arg[i + 1] == '?') + switch (arg[i + 2]) + { + case '!': case '\'': + case '(': case ')': case '-': case '/': + case '<': case '=': case '>': + /* Escape the second '?' in what would otherwise be + a trigraph. */ + c = arg[i + 2]; + i += 2; + STORE ('?'); + STORE ('\\'); + STORE ('?'); + break; + } + break; + + default: + break; + } + break; + + case '\a': esc = 'a'; goto c_escape; + case '\b': esc = 'b'; goto c_escape; + case '\f': esc = 'f'; goto c_escape; + case '\n': esc = 'n'; goto c_and_shell_escape; + case '\r': esc = 'r'; goto c_and_shell_escape; + case '\t': esc = 't'; goto c_and_shell_escape; + case '\v': esc = 'v'; goto c_escape; + case '\\': esc = c; goto c_and_shell_escape; + + c_and_shell_escape: + if (quoting_style == shell_quoting_style) + goto use_shell_always_quoting_style; + c_escape: + if (backslash_escapes) + { + c = esc; + goto store_escape; + } + break; + + case '{': case '}': /* sometimes special if isolated */ + if (! (argsize == SIZE_MAX ? arg[1] == '\0' : argsize == 1)) + break; + /* Fall through. */ + case '#': case '~': + if (i != 0) + break; + /* Fall through. */ + case ' ': + case '!': /* special in bash */ + case '"': case '$': case '&': + case '(': case ')': case '*': case ';': + case '<': + case '=': /* sometimes special in 0th or (with "set -k") later args */ + case '>': case '[': + case '^': /* special in old /bin/sh, e.g. SunOS 4.1.4 */ + case '`': case '|': + /* A shell special character. In theory, '$' and '`' could + be the first bytes of multibyte characters, which means + we should check them with mbrtowc, but in practice this + doesn't happen so it's not worth worrying about. */ + if (quoting_style == shell_quoting_style) + goto use_shell_always_quoting_style; + break; + + case '\'': + switch (quoting_style) + { + case shell_quoting_style: + goto use_shell_always_quoting_style; + + case shell_always_quoting_style: + STORE ('\''); + STORE ('\\'); + STORE ('\''); + break; + + default: + break; + } + break; + + case '%': case '+': case ',': case '-': case '.': case '/': + case '0': case '1': case '2': case '3': case '4': case '5': + case '6': case '7': case '8': case '9': case ':': + case 'A': case 'B': case 'C': case 'D': case 'E': case 'F': + case 'G': case 'H': case 'I': case 'J': case 'K': case 'L': + case 'M': case 'N': case 'O': case 'P': case 'Q': case 'R': + case 'S': case 'T': case 'U': case 'V': case 'W': case 'X': + case 'Y': case 'Z': case ']': case '_': case 'a': case 'b': + case 'c': case 'd': case 'e': case 'f': case 'g': case 'h': + case 'i': case 'j': case 'k': case 'l': case 'm': case 'n': + case 'o': case 'p': case 'q': case 'r': case 's': case 't': + case 'u': case 'v': case 'w': case 'x': case 'y': case 'z': + /* These characters don't cause problems, no matter what the + quoting style is. They cannot start multibyte sequences. */ + break; + + default: + /* If we have a multibyte sequence, copy it until we reach + its end, find an error, or come back to the initial shift + state. For C-like styles, if the sequence has + unprintable characters, escape the whole sequence, since + we can't easily escape single characters within it. */ + { + /* Length of multibyte sequence found so far. */ + size_t m; + + bool printable; + + if (unibyte_locale) + { + m = 1; + printable = isprint (c) != 0; + } + else + { + mbstate_t mbstate; + memset (&mbstate, 0, sizeof mbstate); + + m = 0; + printable = true; + if (argsize == SIZE_MAX) + argsize = strlen (arg); + + do + { + wchar_t w; + size_t bytes = mbrtowc (&w, &arg[i + m], + argsize - (i + m), &mbstate); + if (bytes == 0) + break; + else if (bytes == (size_t) -1) + { + printable = false; + break; + } + else if (bytes == (size_t) -2) + { + printable = false; + while (i + m < argsize && arg[i + m]) + m++; + break; + } + else + { + /* Work around a bug with older shells that "see" a '\' + that is really the 2nd byte of a multibyte character. + In practice the problem is limited to ASCII + chars >= '@' that are shell special chars. */ + if ('[' == 0x5b && quoting_style == shell_quoting_style) + { + size_t j; + for (j = 1; j < bytes; j++) + switch (arg[i + m + j]) + { + case '[': case '\\': case '^': + case '`': case '|': + goto use_shell_always_quoting_style; + } + } + + if (! iswprint (w)) + printable = false; + m += bytes; + } + } + while (! mbsinit (&mbstate)); + } + + if (1 < m || (backslash_escapes && ! printable)) + { + /* Output a multibyte sequence, or an escaped + unprintable unibyte character. */ + size_t ilim = i + m; + + for (;;) + { + if (backslash_escapes && ! printable) + { + STORE ('\\'); + STORE ('0' + (c >> 6)); + STORE ('0' + ((c >> 3) & 7)); + c = '0' + (c & 7); + } + if (ilim <= i + 1) + break; + STORE (c); + c = arg[++i]; + } + + goto store_c; + } + } + } + + if (! (backslash_escapes + && o->quote_these_too[c / INT_BITS] & (1 << (c % INT_BITS)))) + goto store_c; + + store_escape: + STORE ('\\'); + + store_c: + STORE (c); + } + + if (i == 0 && quoting_style == shell_quoting_style) + goto use_shell_always_quoting_style; + + if (quote_string) + for (; *quote_string; quote_string++) + STORE (*quote_string); + + if (len < buffersize) + buffer[len] = '\0'; + return len; + + use_shell_always_quoting_style: + return quotearg_buffer_restyled (buffer, buffersize, arg, argsize, + shell_always_quoting_style, o); +} + +/* Place into buffer BUFFER (of size BUFFERSIZE) a quoted version of + argument ARG (of size ARGSIZE), using O to control quoting. + If O is null, use the default. + Terminate the output with a null character, and return the written + size of the output, not counting the terminating null. + If BUFFERSIZE is too small to store the output string, return the + value that would have been returned had BUFFERSIZE been large enough. + If ARGSIZE is SIZE_MAX, use the string length of the argument for + ARGSIZE. */ +size_t +quotearg_buffer (char *buffer, size_t buffersize, + char const *arg, size_t argsize, + struct quoting_options const *o) +{ + struct quoting_options const *p = o ? o : &default_quoting_options; + int e = errno; + size_t r = quotearg_buffer_restyled (buffer, buffersize, arg, argsize, + p->style, p); + errno = e; + return r; +} + +/* Like quotearg_buffer (..., ARG, ARGSIZE, O), except return newly + allocated storage containing the quoted string. */ +char * +quotearg_alloc (char const *arg, size_t argsize, + struct quoting_options const *o) +{ + int e = errno; + size_t bufsize = quotearg_buffer (0, 0, arg, argsize, o) + 1; + char *buf = xmalloc (bufsize); + quotearg_buffer (buf, bufsize, arg, argsize, o); + errno = e; + return buf; +} + +/* Use storage slot N to return a quoted version of argument ARG. + ARG is of size ARGSIZE, but if that is SIZE_MAX, ARG is a + null-terminated string. + OPTIONS specifies the quoting options. + The returned value points to static storage that can be + reused by the next call to this function with the same value of N. + N must be nonnegative. N is deliberately declared with type "int" + to allow for future extensions (using negative values). */ +static char * +quotearg_n_options (int n, char const *arg, size_t argsize, + struct quoting_options const *options) +{ + int e = errno; + + /* Preallocate a slot 0 buffer, so that the caller can always quote + one small component of a "memory exhausted" message in slot 0. */ + static char slot0[256]; + static unsigned int nslots = 1; + unsigned int n0 = n; + struct slotvec + { + size_t size; + char *val; + }; + static struct slotvec slotvec0 = {sizeof slot0, slot0}; + static struct slotvec *slotvec = &slotvec0; + + if (n < 0) + abort (); + + if (nslots <= n0) + { + unsigned int n1 = n0 + 1; + + if (xalloc_oversized (n1, sizeof *slotvec)) + xalloc_die (); + + if (slotvec == &slotvec0) + { + slotvec = xmalloc (sizeof *slotvec); + *slotvec = slotvec0; + } + slotvec = xrealloc (slotvec, n1 * sizeof *slotvec); + memset (slotvec + nslots, 0, (n1 - nslots) * sizeof *slotvec); + nslots = n1; + } + + { + size_t size = slotvec[n].size; + char *val = slotvec[n].val; + size_t qsize = quotearg_buffer (val, size, arg, argsize, options); + + if (size <= qsize) + { + slotvec[n].size = size = qsize + 1; + if (val != slot0) + free (val); + slotvec[n].val = val = xmalloc (size); + quotearg_buffer (val, size, arg, argsize, options); + } + + errno = e; + return val; + } +} + +char * +quotearg_n (int n, char const *arg) +{ + return quotearg_n_options (n, arg, SIZE_MAX, &default_quoting_options); +} + +char * +quotearg (char const *arg) +{ + return quotearg_n (0, arg); +} + +/* Return quoting options for STYLE, with no extra quoting. */ +static struct quoting_options +quoting_options_from_style (enum quoting_style style) +{ + struct quoting_options o; + o.style = style; + memset (o.quote_these_too, 0, sizeof o.quote_these_too); + return o; +} + +char * +quotearg_n_style (int n, enum quoting_style s, char const *arg) +{ + struct quoting_options const o = quoting_options_from_style (s); + return quotearg_n_options (n, arg, SIZE_MAX, &o); +} + +char * +quotearg_n_style_mem (int n, enum quoting_style s, + char const *arg, size_t argsize) +{ + struct quoting_options const o = quoting_options_from_style (s); + return quotearg_n_options (n, arg, argsize, &o); +} + +char * +quotearg_style (enum quoting_style s, char const *arg) +{ + return quotearg_n_style (0, s, arg); +} + +char * +quotearg_char (char const *arg, char ch) +{ + struct quoting_options options; + options = default_quoting_options; + set_char_quoting (&options, ch, 1); + return quotearg_n_options (0, arg, SIZE_MAX, &options); +} + +char * +quotearg_colon (char const *arg) +{ + return quotearg_char (arg, ':'); +} diff --git a/lib/quotearg.h b/lib/quotearg.h new file mode 100644 index 0000000..24f26f7 --- /dev/null +++ b/lib/quotearg.h @@ -0,0 +1,137 @@ +/* quotearg.h - quote arguments for output + + Copyright (C) 1998, 1999, 2000, 2001, 2002, 2004 Free Software + Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Paul Eggert <eggert@twinsun.com> */ + +#ifndef QUOTEARG_H_ +# define QUOTEARG_H_ 1 + +# include <stddef.h> + +/* Basic quoting styles. */ +enum quoting_style + { + /* Output names as-is (ls --quoting-style=literal). */ + literal_quoting_style, + + /* Quote names for the shell if they contain shell metacharacters + or would cause ambiguous output (ls --quoting-style=shell). */ + shell_quoting_style, + + /* Quote names for the shell, even if they would normally not + require quoting (ls --quoting-style=shell-always). */ + shell_always_quoting_style, + + /* Quote names as for a C language string (ls --quoting-style=c). */ + c_quoting_style, + + /* Like c_quoting_style except omit the surrounding double-quote + characters (ls --quoting-style=escape). */ + escape_quoting_style, + + /* Like clocale_quoting_style, but quote `like this' instead of + "like this" in the default C locale (ls --quoting-style=locale). */ + locale_quoting_style, + + /* Like c_quoting_style except use quotation marks appropriate for + the locale (ls --quoting-style=clocale). */ + clocale_quoting_style + }; + +/* For now, --quoting-style=literal is the default, but this may change. */ +# ifndef DEFAULT_QUOTING_STYLE +# define DEFAULT_QUOTING_STYLE literal_quoting_style +# endif + +/* Names of quoting styles and their corresponding values. */ +extern char const *const quoting_style_args[]; +extern enum quoting_style const quoting_style_vals[]; + +struct quoting_options; + +/* The functions listed below set and use a hidden variable + that contains the default quoting style options. */ + +/* Allocate a new set of quoting options, with contents initially identical + to O if O is not null, or to the default if O is null. + It is the caller's responsibility to free the result. */ +struct quoting_options *clone_quoting_options (struct quoting_options *o); + +/* Get the value of O's quoting style. If O is null, use the default. */ +enum quoting_style get_quoting_style (struct quoting_options *o); + +/* In O (or in the default if O is null), + set the value of the quoting style to S. */ +void set_quoting_style (struct quoting_options *o, enum quoting_style s); + +/* In O (or in the default if O is null), + set the value of the quoting options for character C to I. + Return the old value. Currently, the only values defined for I are + 0 (the default) and 1 (which means to quote the character even if + it would not otherwise be quoted). */ +int set_char_quoting (struct quoting_options *o, char c, int i); + +/* Place into buffer BUFFER (of size BUFFERSIZE) a quoted version of + argument ARG (of size ARGSIZE), using O to control quoting. + If O is null, use the default. + Terminate the output with a null character, and return the written + size of the output, not counting the terminating null. + If BUFFERSIZE is too small to store the output string, return the + value that would have been returned had BUFFERSIZE been large enough. + If ARGSIZE is -1, use the string length of the argument for ARGSIZE. */ +size_t quotearg_buffer (char *buffer, size_t buffersize, + char const *arg, size_t argsize, + struct quoting_options const *o); + +/* Like quotearg_buffer, except return the result in a newly allocated + buffer. It is the caller's responsibility to free the result. */ +char *quotearg_alloc (char const *arg, size_t argsize, + struct quoting_options const *o); + +/* Use storage slot N to return a quoted version of the string ARG. + Use the default quoting options. + The returned value points to static storage that can be + reused by the next call to this function with the same value of N. + N must be nonnegative. */ +char *quotearg_n (int n, char const *arg); + +/* Equivalent to quotearg_n (0, ARG). */ +char *quotearg (char const *arg); + +/* Use style S and storage slot N to return a quoted version of the string ARG. + This is like quotearg_n (N, ARG), except that it uses S with no other + options to specify the quoting method. */ +char *quotearg_n_style (int n, enum quoting_style s, char const *arg); + +/* Use style S and storage slot N to return a quoted version of the + argument ARG of size ARGSIZE. This is like quotearg_n_style + (N, S, ARG), except it can quote null bytes. */ +char *quotearg_n_style_mem (int n, enum quoting_style s, + char const *arg, size_t argsize); + +/* Equivalent to quotearg_n_style (0, S, ARG). */ +char *quotearg_style (enum quoting_style s, char const *arg); + +/* Like quotearg (ARG), except also quote any instances of CH. */ +char *quotearg_char (char const *arg, char ch); + +/* Equivalent to quotearg_char (ARG, ':'). */ +char *quotearg_colon (char const *arg); + +#endif /* !QUOTEARG_H_ */ diff --git a/lib/readlink.c b/lib/readlink.c new file mode 100644 index 0000000..3196aec --- /dev/null +++ b/lib/readlink.c @@ -0,0 +1,49 @@ +/* Stub for readlink(). + Copyright (C) 2003-2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include <errno.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <stddef.h> + +#if !HAVE_READLINK + +/* readlink() substitute for systems that don't have a readlink() function, + such as DJGPP 2.03 and mingw32. */ + +/* The official POSIX return type of readlink() is ssize_t, but since here + we have no declaration in a public header file, we use 'int' as return + type. */ + +int +readlink (const char *path, char *buf, size_t bufsize) +{ + struct stat statbuf; + + /* In general we should use lstat() here, not stat(). But on platforms + without symbolic links lstat() - if it exists - would be equivalent to + stat(), therefore we can use stat(). This saves us a configure check. */ + if (stat (path, &statbuf) >= 0) + errno = EINVAL; + return -1; +} + +#endif diff --git a/lib/realloc.c b/lib/realloc.c new file mode 100644 index 0000000..fe94822 --- /dev/null +++ b/lib/realloc.c @@ -0,0 +1,46 @@ +/* realloc() function that is glibc compatible. + Copyright (C) 1997, 2003, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* written by Jim Meyering */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif +#undef realloc + +#include <stdlib.h> + +/* Change the size of an allocated block of memory P to N bytes, + with error checking. If N is zero, change it to 1. If P is NULL, + use malloc. */ + +void * +rpl_realloc (void *p, size_t n) +{ + if (n == 0) + { + n = 1; + + /* In theory realloc might fail, so don't rely on it to free. */ + free (p); + p = NULL; + } + + if (p == NULL) + return malloc (n); + return realloc (p, n); +} diff --git a/lib/regcomp.c b/lib/regcomp.c new file mode 100644 index 0000000..279b20c --- /dev/null +++ b/lib/regcomp.c @@ -0,0 +1,3779 @@ +/* Extended regular expression matching and search library. + Copyright (C) 2002, 2003, 2004, 2005 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +static reg_errcode_t re_compile_internal (regex_t *preg, const char * pattern, + Idx length, reg_syntax_t syntax); +static void re_compile_fastmap_iter (regex_t *bufp, + const re_dfastate_t *init_state, + char *fastmap); +static reg_errcode_t init_dfa (re_dfa_t *dfa, Idx pat_len); +#ifdef RE_ENABLE_I18N +static void free_charset (re_charset_t *cset); +#endif /* RE_ENABLE_I18N */ +static void free_workarea_compile (regex_t *preg); +static reg_errcode_t create_initial_state (re_dfa_t *dfa); +#ifdef RE_ENABLE_I18N +static void optimize_utf8 (re_dfa_t *dfa); +#endif +static reg_errcode_t analyze (regex_t *preg); +static reg_errcode_t preorder (bin_tree_t *root, + reg_errcode_t (fn (void *, bin_tree_t *)), + void *extra); +static reg_errcode_t postorder (bin_tree_t *root, + reg_errcode_t (fn (void *, bin_tree_t *)), + void *extra); +static reg_errcode_t optimize_subexps (void *extra, bin_tree_t *node); +static reg_errcode_t lower_subexps (void *extra, bin_tree_t *node); +static bin_tree_t *lower_subexp (reg_errcode_t *err, regex_t *preg, + bin_tree_t *node); +static reg_errcode_t calc_first (void *extra, bin_tree_t *node); +static reg_errcode_t calc_next (void *extra, bin_tree_t *node); +static reg_errcode_t link_nfa_nodes (void *extra, bin_tree_t *node); +static Idx duplicate_node (re_dfa_t *dfa, Idx org_idx, unsigned int constraint); +static Idx search_duplicated_node (const re_dfa_t *dfa, Idx org_node, + unsigned int constraint); +static reg_errcode_t calc_eclosure (re_dfa_t *dfa); +static reg_errcode_t calc_eclosure_iter (re_node_set *new_set, re_dfa_t *dfa, + Idx node, bool root); +static reg_errcode_t calc_inveclosure (re_dfa_t *dfa); +static Idx fetch_number (re_string_t *input, re_token_t *token, + reg_syntax_t syntax); +static int peek_token (re_token_t *token, re_string_t *input, + reg_syntax_t syntax); +static bin_tree_t *parse (re_string_t *regexp, regex_t *preg, + reg_syntax_t syntax, reg_errcode_t *err); +static bin_tree_t *parse_reg_exp (re_string_t *regexp, regex_t *preg, + re_token_t *token, reg_syntax_t syntax, + Idx nest, reg_errcode_t *err); +static bin_tree_t *parse_branch (re_string_t *regexp, regex_t *preg, + re_token_t *token, reg_syntax_t syntax, + Idx nest, reg_errcode_t *err); +static bin_tree_t *parse_expression (re_string_t *regexp, regex_t *preg, + re_token_t *token, reg_syntax_t syntax, + Idx nest, reg_errcode_t *err); +static bin_tree_t *parse_sub_exp (re_string_t *regexp, regex_t *preg, + re_token_t *token, reg_syntax_t syntax, + Idx nest, reg_errcode_t *err); +static bin_tree_t *parse_dup_op (bin_tree_t *dup_elem, re_string_t *regexp, + re_dfa_t *dfa, re_token_t *token, + reg_syntax_t syntax, reg_errcode_t *err); +static bin_tree_t *parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, + re_token_t *token, reg_syntax_t syntax, + reg_errcode_t *err); +static reg_errcode_t parse_bracket_element (bracket_elem_t *elem, + re_string_t *regexp, + re_token_t *token, int token_len, + re_dfa_t *dfa, + reg_syntax_t syntax, + bool accept_hyphen); +static reg_errcode_t parse_bracket_symbol (bracket_elem_t *elem, + re_string_t *regexp, + re_token_t *token); +#ifdef RE_ENABLE_I18N +static reg_errcode_t build_equiv_class (bitset sbcset, + re_charset_t *mbcset, + Idx *equiv_class_alloc, + const unsigned char *name); +static reg_errcode_t build_charclass (unsigned REG_TRANSLATE_TYPE trans, + bitset sbcset, + re_charset_t *mbcset, + Idx *char_class_alloc, + const unsigned char *class_name, + reg_syntax_t syntax); +#else /* not RE_ENABLE_I18N */ +static reg_errcode_t build_equiv_class (bitset sbcset, + const unsigned char *name); +static reg_errcode_t build_charclass (unsigned REG_TRANSLATE_TYPE trans, + bitset sbcset, + const unsigned char *class_name, + reg_syntax_t syntax); +#endif /* not RE_ENABLE_I18N */ +static bin_tree_t *build_charclass_op (re_dfa_t *dfa, + unsigned REG_TRANSLATE_TYPE trans, + const unsigned char *class_name, + const unsigned char *extra, + bool non_match, reg_errcode_t *err); +static bin_tree_t *create_tree (re_dfa_t *dfa, + bin_tree_t *left, bin_tree_t *right, + re_token_type_t type); +static bin_tree_t *create_token_tree (re_dfa_t *dfa, + bin_tree_t *left, bin_tree_t *right, + const re_token_t *token); +static bin_tree_t *duplicate_tree (const bin_tree_t *src, re_dfa_t *dfa); +static void free_token (re_token_t *node); +static reg_errcode_t free_tree (void *extra, bin_tree_t *node); +static reg_errcode_t mark_opt_subexp (void *extra, bin_tree_t *node); + +/* This table gives an error message for each of the error codes listed + in regex.h. Obviously the order here has to be same as there. + POSIX doesn't require that we do anything for REG_NOERROR, + but why not be nice? */ + +const char __re_error_msgid[] attribute_hidden = + { +#define REG_NOERROR_IDX 0 + gettext_noop ("Success") /* REG_NOERROR */ + "\0" +#define REG_NOMATCH_IDX (REG_NOERROR_IDX + sizeof "Success") + gettext_noop ("No match") /* REG_NOMATCH */ + "\0" +#define REG_BADPAT_IDX (REG_NOMATCH_IDX + sizeof "No match") + gettext_noop ("Invalid regular expression") /* REG_BADPAT */ + "\0" +#define REG_ECOLLATE_IDX (REG_BADPAT_IDX + sizeof "Invalid regular expression") + gettext_noop ("Invalid collation character") /* REG_ECOLLATE */ + "\0" +#define REG_ECTYPE_IDX (REG_ECOLLATE_IDX + sizeof "Invalid collation character") + gettext_noop ("Invalid character class name") /* REG_ECTYPE */ + "\0" +#define REG_EESCAPE_IDX (REG_ECTYPE_IDX + sizeof "Invalid character class name") + gettext_noop ("Trailing backslash") /* REG_EESCAPE */ + "\0" +#define REG_ESUBREG_IDX (REG_EESCAPE_IDX + sizeof "Trailing backslash") + gettext_noop ("Invalid back reference") /* REG_ESUBREG */ + "\0" +#define REG_EBRACK_IDX (REG_ESUBREG_IDX + sizeof "Invalid back reference") + gettext_noop ("Unmatched [ or [^") /* REG_EBRACK */ + "\0" +#define REG_EPAREN_IDX (REG_EBRACK_IDX + sizeof "Unmatched [ or [^") + gettext_noop ("Unmatched ( or \\(") /* REG_EPAREN */ + "\0" +#define REG_EBRACE_IDX (REG_EPAREN_IDX + sizeof "Unmatched ( or \\(") + gettext_noop ("Unmatched \\{") /* REG_EBRACE */ + "\0" +#define REG_BADBR_IDX (REG_EBRACE_IDX + sizeof "Unmatched \\{") + gettext_noop ("Invalid content of \\{\\}") /* REG_BADBR */ + "\0" +#define REG_ERANGE_IDX (REG_BADBR_IDX + sizeof "Invalid content of \\{\\}") + gettext_noop ("Invalid range end") /* REG_ERANGE */ + "\0" +#define REG_ESPACE_IDX (REG_ERANGE_IDX + sizeof "Invalid range end") + gettext_noop ("Memory exhausted") /* REG_ESPACE */ + "\0" +#define REG_BADRPT_IDX (REG_ESPACE_IDX + sizeof "Memory exhausted") + gettext_noop ("Invalid preceding regular expression") /* REG_BADRPT */ + "\0" +#define REG_EEND_IDX (REG_BADRPT_IDX + sizeof "Invalid preceding regular expression") + gettext_noop ("Premature end of regular expression") /* REG_EEND */ + "\0" +#define REG_ESIZE_IDX (REG_EEND_IDX + sizeof "Premature end of regular expression") + gettext_noop ("Regular expression too big") /* REG_ESIZE */ + "\0" +#define REG_ERPAREN_IDX (REG_ESIZE_IDX + sizeof "Regular expression too big") + gettext_noop ("Unmatched ) or \\)") /* REG_ERPAREN */ + }; + +const size_t __re_error_msgid_idx[] attribute_hidden = + { + REG_NOERROR_IDX, + REG_NOMATCH_IDX, + REG_BADPAT_IDX, + REG_ECOLLATE_IDX, + REG_ECTYPE_IDX, + REG_EESCAPE_IDX, + REG_ESUBREG_IDX, + REG_EBRACK_IDX, + REG_EPAREN_IDX, + REG_EBRACE_IDX, + REG_BADBR_IDX, + REG_ERANGE_IDX, + REG_ESPACE_IDX, + REG_BADRPT_IDX, + REG_EEND_IDX, + REG_ESIZE_IDX, + REG_ERPAREN_IDX + }; + +/* Entry points for GNU code. */ + +/* re_compile_pattern is the GNU regular expression compiler: it + compiles PATTERN (of length LENGTH) and puts the result in BUFP. + Returns 0 if the pattern was valid, otherwise an error string. + + Assumes the `re_allocated' (and perhaps `re_buffer') and `translate' fields + are set in BUFP on entry. */ + +const char * +re_compile_pattern (const char *pattern, size_t length, + struct re_pattern_buffer *bufp) +{ + reg_errcode_t ret; + + /* And GNU code determines whether or not to get register information + by passing null for the REGS argument to re_match, etc., not by + setting re_no_sub, unless REG_NO_SUB is set. */ + bufp->re_no_sub = !!(re_syntax_options & REG_NO_SUB); + + /* Match anchors at newline. */ + bufp->re_newline_anchor = 1; + + ret = re_compile_internal (bufp, pattern, length, re_syntax_options); + + if (!ret) + return NULL; + return gettext (__re_error_msgid + __re_error_msgid_idx[(int) ret]); +} +#ifdef _LIBC +weak_alias (__re_compile_pattern, re_compile_pattern) +#endif + +/* Set by `re_set_syntax' to the current regexp syntax to recognize. Can + also be assigned to arbitrarily: each pattern buffer stores its own + syntax, so it can be changed between regex compilations. */ +/* This has no initializer because initialized variables in Emacs + become read-only after dumping. */ +reg_syntax_t re_syntax_options; + + +/* Specify the precise syntax of regexps for compilation. This provides + for compatibility for various utilities which historically have + different, incompatible syntaxes. + + The argument SYNTAX is a bit mask comprised of the various bits + defined in regex.h. We return the old syntax. */ + +reg_syntax_t +re_set_syntax (reg_syntax_t syntax) +{ + reg_syntax_t ret = re_syntax_options; + + re_syntax_options = syntax; + return ret; +} +#ifdef _LIBC +weak_alias (__re_set_syntax, re_set_syntax) +#endif + +int +re_compile_fastmap (struct re_pattern_buffer *bufp) +{ + re_dfa_t *dfa = (re_dfa_t *) bufp->re_buffer; + char *fastmap = bufp->re_fastmap; + + memset (fastmap, '\0', sizeof (char) * SBC_MAX); + re_compile_fastmap_iter (bufp, dfa->init_state, fastmap); + if (dfa->init_state != dfa->init_state_word) + re_compile_fastmap_iter (bufp, dfa->init_state_word, fastmap); + if (dfa->init_state != dfa->init_state_nl) + re_compile_fastmap_iter (bufp, dfa->init_state_nl, fastmap); + if (dfa->init_state != dfa->init_state_begbuf) + re_compile_fastmap_iter (bufp, dfa->init_state_begbuf, fastmap); + bufp->re_fastmap_accurate = 1; + return 0; +} +#ifdef _LIBC +weak_alias (__re_compile_fastmap, re_compile_fastmap) +#endif + +static inline void +__attribute ((always_inline)) +re_set_fastmap (char *fastmap, bool icase, int ch) +{ + fastmap[ch] = 1; + if (icase) + fastmap[tolower (ch)] = 1; +} + +/* Helper function for re_compile_fastmap. + Compile fastmap for the initial_state INIT_STATE. */ + +static void +re_compile_fastmap_iter (regex_t *bufp, const re_dfastate_t *init_state, + char *fastmap) +{ + re_dfa_t *dfa = (re_dfa_t *) bufp->re_buffer; + Idx node_cnt; + bool icase = (dfa->mb_cur_max == 1 && (bufp->re_syntax & REG_IGNORE_CASE)); + for (node_cnt = 0; node_cnt < init_state->nodes.nelem; ++node_cnt) + { + Idx node = init_state->nodes.elems[node_cnt]; + re_token_type_t type = dfa->nodes[node].type; + + if (type == CHARACTER) + { + re_set_fastmap (fastmap, icase, dfa->nodes[node].opr.c); +#ifdef RE_ENABLE_I18N + if ((bufp->re_syntax & REG_IGNORE_CASE) && dfa->mb_cur_max > 1) + { + unsigned char buf[MB_LEN_MAX]; + unsigned char *p; + wchar_t wc; + mbstate_t state; + + p = buf; + *p++ = dfa->nodes[node].opr.c; + while (++node < dfa->nodes_len + && dfa->nodes[node].type == CHARACTER + && dfa->nodes[node].mb_partial) + *p++ = dfa->nodes[node].opr.c; + memset (&state, 0, sizeof (state)); + if (mbrtowc (&wc, (const char *) buf, p - buf, + &state) == p - buf + && (__wcrtomb ((char *) buf, towlower (wc), &state) + != (size_t) -1)) + re_set_fastmap (fastmap, false, buf[0]); + } +#endif + } + else if (type == SIMPLE_BRACKET) + { + int i, j, ch; + for (i = 0, ch = 0; i < BITSET_WORDS; ++i) + for (j = 0; j < BITSET_WORD_BITS; ++j, ++ch) + if (dfa->nodes[node].opr.sbcset[i] & ((bitset_word) 1 << j)) + re_set_fastmap (fastmap, icase, ch); + } +#ifdef RE_ENABLE_I18N + else if (type == COMPLEX_BRACKET) + { + Idx i; + re_charset_t *cset = dfa->nodes[node].opr.mbcset; + if (cset->non_match || cset->ncoll_syms || cset->nequiv_classes + || cset->nranges || cset->nchar_classes) + { +# ifdef _LIBC + if (_NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES) != 0) + { + /* In this case we want to catch the bytes which are + the first byte of any collation elements. + e.g. In da_DK, we want to catch 'a' since "aa" + is a valid collation element, and don't catch + 'b' since 'b' is the only collation element + which starts from 'b'. */ + const int32_t *table = (const int32_t *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB); + for (i = 0; i < SBC_MAX; ++i) + if (table[i] < 0) + re_set_fastmap (fastmap, icase, i); + } +# else + if (dfa->mb_cur_max > 1) + for (i = 0; i < SBC_MAX; ++i) + if (__btowc (i) == WEOF) + re_set_fastmap (fastmap, icase, i); +# endif /* not _LIBC */ + } + for (i = 0; i < cset->nmbchars; ++i) + { + char buf[256]; + mbstate_t state; + memset (&state, '\0', sizeof (state)); + if (__wcrtomb (buf, cset->mbchars[i], &state) != (size_t) -1) + re_set_fastmap (fastmap, icase, *(unsigned char *) buf); + if ((bufp->re_syntax & REG_IGNORE_CASE) && dfa->mb_cur_max > 1) + { + if (__wcrtomb (buf, towlower (cset->mbchars[i]), &state) + != (size_t) -1) + re_set_fastmap (fastmap, false, *(unsigned char *) buf); + } + } + } +#endif /* RE_ENABLE_I18N */ + else if (type == OP_PERIOD +#ifdef RE_ENABLE_I18N + || type == OP_UTF8_PERIOD +#endif /* RE_ENABLE_I18N */ + || type == END_OF_RE) + { + memset (fastmap, '\1', sizeof (char) * SBC_MAX); + if (type == END_OF_RE) + bufp->re_can_be_null = 1; + return; + } + } +} + +/* Entry point for POSIX code. */ +/* regcomp takes a regular expression as a string and compiles it. + + PREG is a regex_t *. We do not expect any fields to be initialized, + since POSIX says we shouldn't. Thus, we set + + `re_buffer' to the compiled pattern; + `re_used' to the length of the compiled pattern; + `re_syntax' to REG_SYNTAX_POSIX_EXTENDED if the + REG_EXTENDED bit in CFLAGS is set; otherwise, to + REG_SYNTAX_POSIX_BASIC; + `re_newline_anchor' to REG_NEWLINE being set in CFLAGS; + `re_fastmap' to an allocated space for the fastmap; + `re_fastmap_accurate' to zero; + `re_nsub' to the number of subexpressions in PATTERN. + + PATTERN is the address of the pattern string. + + CFLAGS is a series of bits which affect compilation. + + If REG_EXTENDED is set, we use POSIX extended syntax; otherwise, we + use POSIX basic syntax. + + If REG_NEWLINE is set, then . and [^...] don't match newline. + Also, regexec will try a match beginning after every newline. + + If REG_ICASE is set, then we considers upper- and lowercase + versions of letters to be equivalent when matching. + + If REG_NOSUB is set, then when PREG is passed to regexec, that + routine will report only success or failure, and nothing about the + registers. + + It returns 0 if it succeeds, nonzero if it doesn't. (See regex.h for + the return codes and their meanings.) */ + +int +regcomp (regex_t *__restrict preg, const char *__restrict pattern, int cflags) +{ + reg_errcode_t ret; + reg_syntax_t syntax = ((cflags & REG_EXTENDED) ? REG_SYNTAX_POSIX_EXTENDED + : REG_SYNTAX_POSIX_BASIC); + + preg->re_buffer = NULL; + preg->re_allocated = 0; + preg->re_used = 0; + + /* Try to allocate space for the fastmap. */ + preg->re_fastmap = re_malloc (char, SBC_MAX); + if (BE (preg->re_fastmap == NULL, 0)) + return REG_ESPACE; + + syntax |= (cflags & REG_ICASE) ? REG_IGNORE_CASE : 0; + + /* If REG_NEWLINE is set, newlines are treated differently. */ + if (cflags & REG_NEWLINE) + { /* REG_NEWLINE implies neither . nor [^...] match newline. */ + syntax &= ~REG_DOT_NEWLINE; + syntax |= REG_HAT_LISTS_NOT_NEWLINE; + /* It also changes the matching behavior. */ + preg->re_newline_anchor = 1; + } + else + preg->re_newline_anchor = 0; + preg->re_no_sub = !!(cflags & REG_NOSUB); + preg->re_translate = NULL; + + ret = re_compile_internal (preg, pattern, strlen (pattern), syntax); + + /* POSIX doesn't distinguish between an unmatched open-group and an + unmatched close-group: both are REG_EPAREN. */ + if (ret == REG_ERPAREN) + ret = REG_EPAREN; + + /* We have already checked preg->re_fastmap != NULL. */ + if (BE (ret == REG_NOERROR, 1)) + /* Compute the fastmap now, since regexec cannot modify the pattern + buffer. This function never fails in this implementation. */ + (void) re_compile_fastmap (preg); + else + { + /* Some error occurred while compiling the expression. */ + re_free (preg->re_fastmap); + preg->re_fastmap = NULL; + } + + return (int) ret; +} +#ifdef _LIBC +weak_alias (__regcomp, regcomp) +#endif + +/* Returns a message corresponding to an error code, ERRCODE, returned + from either regcomp or regexec. We don't use PREG here. */ + +size_t +regerror (int errcode, const regex_t *__restrict preg, + char *__restrict errbuf, size_t errbuf_size) +{ + const char *msg; + size_t msg_size; + + if (BE (errcode < 0 + || errcode >= (int) (sizeof (__re_error_msgid_idx) + / sizeof (__re_error_msgid_idx[0])), 0)) + /* Only error codes returned by the rest of the code should be passed + to this routine. If we are given anything else, or if other regex + code generates an invalid error code, then the program has a bug. + Dump core so we can fix it. */ + abort (); + + msg = gettext (__re_error_msgid + __re_error_msgid_idx[errcode]); + + msg_size = strlen (msg) + 1; /* Includes the null. */ + + if (BE (errbuf_size != 0, 1)) + { + if (BE (msg_size > errbuf_size, 0)) + { +#if defined HAVE_MEMPCPY || defined _LIBC + *((char *) __mempcpy (errbuf, msg, errbuf_size - 1)) = '\0'; +#else + memcpy (errbuf, msg, errbuf_size - 1); + errbuf[errbuf_size - 1] = 0; +#endif + } + else + memcpy (errbuf, msg, msg_size); + } + + return msg_size; +} +#ifdef _LIBC +weak_alias (__regerror, regerror) +#endif + + +#ifdef RE_ENABLE_I18N +/* This static array is used for the map to single-byte characters when + UTF-8 is used. Otherwise we would allocate memory just to initialize + it the same all the time. UTF-8 is the preferred encoding so this is + a worthwhile optimization. */ +static const bitset utf8_sb_map = +{ + /* Set the first 128 bits. */ +# if 2 < BITSET_WORDS + BITSET_WORD_MAX, +# endif +# if 4 < BITSET_WORDS + BITSET_WORD_MAX, +# endif +# if 6 < BITSET_WORDS + BITSET_WORD_MAX, +# endif +# if 8 < BITSET_WORDS +# error "Invalid BITSET_WORDS" +# endif + (BITSET_WORD_MAX + >> (SBC_MAX % BITSET_WORD_BITS == 0 + ? 0 + : BITSET_WORD_BITS - SBC_MAX % BITSET_WORD_BITS)) +}; +#endif + + +static void +free_dfa_content (re_dfa_t *dfa) +{ + Idx i, j; + + if (dfa->nodes) + for (i = 0; i < dfa->nodes_len; ++i) + free_token (dfa->nodes + i); + re_free (dfa->nexts); + for (i = 0; i < dfa->nodes_len; ++i) + { + if (dfa->eclosures != NULL) + re_node_set_free (dfa->eclosures + i); + if (dfa->inveclosures != NULL) + re_node_set_free (dfa->inveclosures + i); + if (dfa->edests != NULL) + re_node_set_free (dfa->edests + i); + } + re_free (dfa->edests); + re_free (dfa->eclosures); + re_free (dfa->inveclosures); + re_free (dfa->nodes); + + if (dfa->state_table) + for (i = 0; i <= dfa->state_hash_mask; ++i) + { + struct re_state_table_entry *entry = dfa->state_table + i; + for (j = 0; j < entry->num; ++j) + { + re_dfastate_t *state = entry->array[j]; + free_state (state); + } + re_free (entry->array); + } + re_free (dfa->state_table); +#ifdef RE_ENABLE_I18N + if (dfa->sb_char != utf8_sb_map) + re_free (dfa->sb_char); +#endif + re_free (dfa->subexp_map); +#ifdef DEBUG + re_free (dfa->re_str); +#endif + + re_free (dfa); +} + + +/* Free dynamically allocated space used by PREG. */ + +void +regfree (regex_t *preg) +{ + re_dfa_t *dfa = (re_dfa_t *) preg->re_buffer; + if (BE (dfa != NULL, 1)) + free_dfa_content (dfa); + preg->re_buffer = NULL; + preg->re_allocated = 0; + + re_free (preg->re_fastmap); + preg->re_fastmap = NULL; + + re_free (preg->re_translate); + preg->re_translate = NULL; +} +#ifdef _LIBC +weak_alias (__regfree, regfree) +#endif + +/* Entry points compatible with 4.2 BSD regex library. We don't define + them unless specifically requested. */ + +#if defined _REGEX_RE_COMP || defined _LIBC + +/* BSD has one and only one pattern buffer. */ +static struct re_pattern_buffer re_comp_buf; + +char * +# ifdef _LIBC +/* Make these definitions weak in libc, so POSIX programs can redefine + these names if they don't use our functions, and still use + regcomp/regexec above without link errors. */ +weak_function +# endif +re_comp (const char *s) +{ + reg_errcode_t ret; + char *fastmap; + + if (!s) + { + if (!re_comp_buf.re_buffer) + return gettext ("No previous regular expression"); + return 0; + } + + if (re_comp_buf.re_buffer) + { + fastmap = re_comp_buf.re_fastmap; + re_comp_buf.re_fastmap = NULL; + __regfree (&re_comp_buf); + memset (&re_comp_buf, '\0', sizeof (re_comp_buf)); + re_comp_buf.re_fastmap = fastmap; + } + + if (re_comp_buf.re_fastmap == NULL) + { + re_comp_buf.re_fastmap = (char *) malloc (SBC_MAX); + if (re_comp_buf.re_fastmap == NULL) + return (char *) gettext (__re_error_msgid + + __re_error_msgid_idx[(int) REG_ESPACE]); + } + + /* Since `re_exec' always passes NULL for the `regs' argument, we + don't need to initialize the pattern buffer fields which affect it. */ + + /* Match anchors at newlines. */ + re_comp_buf.re_newline_anchor = 1; + + ret = re_compile_internal (&re_comp_buf, s, strlen (s), re_syntax_options); + + if (!ret) + return NULL; + + /* Yes, we're discarding `const' here if !HAVE_LIBINTL. */ + return (char *) gettext (__re_error_msgid + __re_error_msgid_idx[(int) ret]); +} + +#ifdef _LIBC +libc_freeres_fn (free_mem) +{ + __regfree (&re_comp_buf); +} +#endif + +#endif /* _REGEX_RE_COMP */ + +/* Internal entry point. + Compile the regular expression PATTERN, whose length is LENGTH. + SYNTAX indicate regular expression's syntax. */ + +static reg_errcode_t +re_compile_internal (regex_t *preg, const char *pattern, Idx length, + reg_syntax_t syntax) +{ + reg_errcode_t err = REG_NOERROR; + re_dfa_t *dfa; + re_string_t regexp; + + /* Initialize the pattern buffer. */ + preg->re_fastmap_accurate = 0; + preg->re_syntax = syntax; + preg->re_not_bol = preg->re_not_eol = 0; + preg->re_used = 0; + preg->re_nsub = 0; + preg->re_can_be_null = 0; + preg->re_regs_allocated = REG_UNALLOCATED; + + /* Initialize the dfa. */ + dfa = (re_dfa_t *) preg->re_buffer; + if (BE (preg->re_allocated < sizeof (re_dfa_t), 0)) + { + /* If zero allocated, but buffer is non-null, try to realloc + enough space. This loses if buffer's address is bogus, but + that is the user's responsibility. If buffer is null this + is a simple allocation. */ + dfa = re_realloc (preg->re_buffer, re_dfa_t, 1); + if (dfa == NULL) + return REG_ESPACE; + preg->re_allocated = sizeof (re_dfa_t); + preg->re_buffer = (unsigned char *) dfa; + } + preg->re_used = sizeof (re_dfa_t); + + __libc_lock_init (dfa->lock); + + err = init_dfa (dfa, length); + if (BE (err != REG_NOERROR, 0)) + { + free_dfa_content (dfa); + preg->re_buffer = NULL; + preg->re_allocated = 0; + return err; + } +#ifdef DEBUG + dfa->re_str = re_malloc (char, length + 1); + strncpy (dfa->re_str, pattern, length + 1); +#endif + + err = re_string_construct (®exp, pattern, length, preg->re_translate, + syntax & REG_IGNORE_CASE, dfa); + if (BE (err != REG_NOERROR, 0)) + { + re_compile_internal_free_return: + free_workarea_compile (preg); + re_string_destruct (®exp); + free_dfa_content (dfa); + preg->re_buffer = NULL; + preg->re_allocated = 0; + return err; + } + + /* Parse the regular expression, and build a structure tree. */ + preg->re_nsub = 0; + dfa->str_tree = parse (®exp, preg, syntax, &err); + if (BE (dfa->str_tree == NULL, 0)) + goto re_compile_internal_free_return; + + /* Analyze the tree and create the nfa. */ + err = analyze (preg); + if (BE (err != REG_NOERROR, 0)) + goto re_compile_internal_free_return; + +#ifdef RE_ENABLE_I18N + /* If possible, do searching in single byte encoding to speed things up. */ + if (dfa->is_utf8 && !(syntax & REG_IGNORE_CASE) && preg->re_translate == NULL) + optimize_utf8 (dfa); +#endif + + /* Then create the initial state of the dfa. */ + err = create_initial_state (dfa); + + /* Release work areas. */ + free_workarea_compile (preg); + re_string_destruct (®exp); + + if (BE (err != REG_NOERROR, 0)) + { + free_dfa_content (dfa); + preg->re_buffer = NULL; + preg->re_allocated = 0; + } + + return err; +} + +/* Initialize DFA. We use the length of the regular expression PAT_LEN + as the initial length of some arrays. */ + +static reg_errcode_t +init_dfa (re_dfa_t *dfa, Idx pat_len) +{ + __re_size_t table_size; +#ifndef _LIBC + char *codeset_name; +#endif + + memset (dfa, '\0', sizeof (re_dfa_t)); + + /* Force allocation of str_tree_storage the first time. */ + dfa->str_tree_storage_idx = BIN_TREE_STORAGE_SIZE; + + dfa->nodes_alloc = pat_len + 1; + dfa->nodes = re_xmalloc (re_token_t, dfa->nodes_alloc); + + /* table_size = 2 ^ ceil(log pat_len) */ + for (table_size = 1; table_size <= pat_len; table_size <<= 1) + if (0 < (Idx) -1 && table_size == 0) + return REG_ESPACE; + + dfa->state_table = re_calloc (struct re_state_table_entry, table_size); + dfa->state_hash_mask = table_size - 1; + + dfa->mb_cur_max = MB_CUR_MAX; +#ifdef _LIBC + if (dfa->mb_cur_max == 6 + && strcmp (_NL_CURRENT (LC_CTYPE, _NL_CTYPE_CODESET_NAME), "UTF-8") == 0) + dfa->is_utf8 = 1; + dfa->map_notascii = (_NL_CURRENT_WORD (LC_CTYPE, _NL_CTYPE_MAP_TO_NONASCII) + != 0); +#else +# ifdef HAVE_LANGINFO_CODESET + codeset_name = nl_langinfo (CODESET); +# else + codeset_name = getenv ("LC_ALL"); + if (codeset_name == NULL || codeset_name[0] == '\0') + codeset_name = getenv ("LC_CTYPE"); + if (codeset_name == NULL || codeset_name[0] == '\0') + codeset_name = getenv ("LANG"); + if (codeset_name == NULL) + codeset_name = ""; + else if (strchr (codeset_name, '.') != NULL) + codeset_name = strchr (codeset_name, '.') + 1; +# endif + + if (strcasecmp (codeset_name, "UTF-8") == 0 + || strcasecmp (codeset_name, "UTF8") == 0) + dfa->is_utf8 = 1; + + /* We check exhaustively in the loop below if this charset is a + superset of ASCII. */ + dfa->map_notascii = 0; +#endif + +#ifdef RE_ENABLE_I18N + if (dfa->mb_cur_max > 1) + { + if (dfa->is_utf8) + dfa->sb_char = (re_bitset_ptr_t) utf8_sb_map; + else + { + int i, j, ch; + + dfa->sb_char = re_calloc (bitset_word, BITSET_WORDS); + if (BE (dfa->sb_char == NULL, 0)) + return REG_ESPACE; + + /* Set the bits corresponding to single byte chars. */ + for (i = 0, ch = 0; i < BITSET_WORDS; ++i) + for (j = 0; j < BITSET_WORD_BITS; ++j, ++ch) + { + wint_t wch = __btowc (ch); + if (wch != WEOF) + dfa->sb_char[i] |= (bitset_word) 1 << j; +# ifndef _LIBC + if (isascii (ch) && wch != ch) + dfa->map_notascii = 1; +# endif + } + } + } +#endif + + if (BE (dfa->nodes == NULL || dfa->state_table == NULL, 0)) + return REG_ESPACE; + return REG_NOERROR; +} + +/* Initialize WORD_CHAR table, which indicate which character is + "word". In this case "word" means that it is the word construction + character used by some operators like "\<", "\>", etc. */ + +static void +init_word_char (re_dfa_t *dfa) +{ + int i, j, ch; + dfa->word_ops_used = 1; + for (i = 0, ch = 0; i < BITSET_WORDS; ++i) + for (j = 0; j < BITSET_WORD_BITS; ++j, ++ch) + if (isalnum (ch) || ch == '_') + dfa->word_char[i] |= (bitset_word) 1 << j; +} + +/* Free the work area which are only used while compiling. */ + +static void +free_workarea_compile (regex_t *preg) +{ + re_dfa_t *dfa = (re_dfa_t *) preg->re_buffer; + bin_tree_storage_t *storage, *next; + for (storage = dfa->str_tree_storage; storage; storage = next) + { + next = storage->next; + re_free (storage); + } + dfa->str_tree_storage = NULL; + dfa->str_tree_storage_idx = BIN_TREE_STORAGE_SIZE; + dfa->str_tree = NULL; + re_free (dfa->org_indices); + dfa->org_indices = NULL; +} + +/* Create initial states for all contexts. */ + +static reg_errcode_t +create_initial_state (re_dfa_t *dfa) +{ + Idx first, i; + reg_errcode_t err; + re_node_set init_nodes; + + /* Initial states have the epsilon closure of the node which is + the first node of the regular expression. */ + first = dfa->str_tree->first->node_idx; + dfa->init_node = first; + err = re_node_set_init_copy (&init_nodes, dfa->eclosures + first); + if (BE (err != REG_NOERROR, 0)) + return err; + + /* The back-references which are in initial states can epsilon transit, + since in this case all of the subexpressions can be null. + Then we add epsilon closures of the nodes which are the next nodes of + the back-references. */ + if (dfa->nbackref > 0) + for (i = 0; i < init_nodes.nelem; ++i) + { + Idx node_idx = init_nodes.elems[i]; + re_token_type_t type = dfa->nodes[node_idx].type; + + Idx clexp_idx; + if (type != OP_BACK_REF) + continue; + for (clexp_idx = 0; clexp_idx < init_nodes.nelem; ++clexp_idx) + { + re_token_t *clexp_node; + clexp_node = dfa->nodes + init_nodes.elems[clexp_idx]; + if (clexp_node->type == OP_CLOSE_SUBEXP + && clexp_node->opr.idx == dfa->nodes[node_idx].opr.idx) + break; + } + if (clexp_idx == init_nodes.nelem) + continue; + + if (type == OP_BACK_REF) + { + Idx dest_idx = dfa->edests[node_idx].elems[0]; + if (!re_node_set_contains (&init_nodes, dest_idx)) + { + re_node_set_merge (&init_nodes, dfa->eclosures + dest_idx); + i = 0; + } + } + } + + /* It must be the first time to invoke acquire_state. */ + dfa->init_state = re_acquire_state_context (&err, dfa, &init_nodes, 0); + /* We don't check ERR here, since the initial state must not be NULL. */ + if (BE (dfa->init_state == NULL, 0)) + return err; + if (dfa->init_state->has_constraint) + { + dfa->init_state_word = re_acquire_state_context (&err, dfa, &init_nodes, + CONTEXT_WORD); + dfa->init_state_nl = re_acquire_state_context (&err, dfa, &init_nodes, + CONTEXT_NEWLINE); + dfa->init_state_begbuf = re_acquire_state_context (&err, dfa, + &init_nodes, + CONTEXT_NEWLINE + | CONTEXT_BEGBUF); + if (BE (dfa->init_state_word == NULL || dfa->init_state_nl == NULL + || dfa->init_state_begbuf == NULL, 0)) + return err; + } + else + dfa->init_state_word = dfa->init_state_nl + = dfa->init_state_begbuf = dfa->init_state; + + re_node_set_free (&init_nodes); + return REG_NOERROR; +} + +#ifdef RE_ENABLE_I18N +/* If it is possible to do searching in single byte encoding instead of UTF-8 + to speed things up, set dfa->mb_cur_max to 1, clear is_utf8 and change + DFA nodes where needed. */ + +static void +optimize_utf8 (re_dfa_t *dfa) +{ + Idx node; + int i; + bool mb_chars = false; + bool has_period = false; + + for (node = 0; node < dfa->nodes_len; ++node) + switch (dfa->nodes[node].type) + { + case CHARACTER: + if (dfa->nodes[node].opr.c >= 0x80) + mb_chars = true; + break; + case ANCHOR: + switch (dfa->nodes[node].opr.idx) + { + case LINE_FIRST: + case LINE_LAST: + case BUF_FIRST: + case BUF_LAST: + break; + default: + /* Word anchors etc. cannot be handled. */ + return; + } + break; + case OP_PERIOD: + has_period = true; + break; + case OP_BACK_REF: + case OP_ALT: + case END_OF_RE: + case OP_DUP_ASTERISK: + case OP_OPEN_SUBEXP: + case OP_CLOSE_SUBEXP: + break; + case COMPLEX_BRACKET: + return; + case SIMPLE_BRACKET: + /* Just double check. */ + { + int rshift = + (SBC_MAX / 2 % BITSET_WORD_BITS == 0 + ? 0 + : BITSET_WORD_BITS - SBC_MAX / 2 % BITSET_WORD_BITS); + for (i = SBC_MAX / 2 / BITSET_WORD_BITS; i < BITSET_WORDS; ++i) + { + if (dfa->nodes[node].opr.sbcset[i] >> rshift != 0) + return; + rshift = 0; + } + } + break; + default: + abort (); + } + + if (mb_chars || has_period) + for (node = 0; node < dfa->nodes_len; ++node) + { + if (dfa->nodes[node].type == CHARACTER + && dfa->nodes[node].opr.c >= 0x80) + dfa->nodes[node].mb_partial = 0; + else if (dfa->nodes[node].type == OP_PERIOD) + dfa->nodes[node].type = OP_UTF8_PERIOD; + } + + /* The search can be in single byte locale. */ + dfa->mb_cur_max = 1; + dfa->is_utf8 = 0; + dfa->has_mb_node = dfa->nbackref > 0 || has_period; +} +#endif + +/* Analyze the structure tree, and calculate "first", "next", "edest", + "eclosure", and "inveclosure". */ + +static reg_errcode_t +analyze (regex_t *preg) +{ + re_dfa_t *dfa = (re_dfa_t *) preg->re_buffer; + reg_errcode_t ret; + + /* Allocate arrays. */ + dfa->nexts = re_malloc (Idx, dfa->nodes_alloc); + dfa->org_indices = re_malloc (Idx, dfa->nodes_alloc); + dfa->edests = re_xmalloc (re_node_set, dfa->nodes_alloc); + dfa->eclosures = re_malloc (re_node_set, dfa->nodes_alloc); + if (BE (dfa->nexts == NULL || dfa->org_indices == NULL || dfa->edests == NULL + || dfa->eclosures == NULL, 0)) + return REG_ESPACE; + + dfa->subexp_map = re_xmalloc (Idx, preg->re_nsub); + if (dfa->subexp_map != NULL) + { + Idx i; + for (i = 0; i < preg->re_nsub; i++) + dfa->subexp_map[i] = i; + preorder (dfa->str_tree, optimize_subexps, dfa); + for (i = 0; i < preg->re_nsub; i++) + if (dfa->subexp_map[i] != i) + break; + if (i == preg->re_nsub) + { + free (dfa->subexp_map); + dfa->subexp_map = NULL; + } + } + + ret = postorder (dfa->str_tree, lower_subexps, preg); + if (BE (ret != REG_NOERROR, 0)) + return ret; + ret = postorder (dfa->str_tree, calc_first, dfa); + if (BE (ret != REG_NOERROR, 0)) + return ret; + preorder (dfa->str_tree, calc_next, dfa); + ret = preorder (dfa->str_tree, link_nfa_nodes, dfa); + if (BE (ret != REG_NOERROR, 0)) + return ret; + ret = calc_eclosure (dfa); + if (BE (ret != REG_NOERROR, 0)) + return ret; + + /* We only need this during the prune_impossible_nodes pass in regexec.c; + skip it if p_i_n will not run, as calc_inveclosure can be quadratic. */ + if ((!preg->re_no_sub && preg->re_nsub > 0 && dfa->has_plural_match) + || dfa->nbackref) + { + dfa->inveclosures = re_xmalloc (re_node_set, dfa->nodes_len); + if (BE (dfa->inveclosures == NULL, 0)) + return REG_ESPACE; + ret = calc_inveclosure (dfa); + } + + return ret; +} + +/* Our parse trees are very unbalanced, so we cannot use a stack to + implement parse tree visits. Instead, we use parent pointers and + some hairy code in these two functions. */ +static reg_errcode_t +postorder (bin_tree_t *root, reg_errcode_t (fn (void *, bin_tree_t *)), + void *extra) +{ + bin_tree_t *node, *prev; + + for (node = root; ; ) + { + /* Descend down the tree, preferably to the left (or to the right + if that's the only child). */ + while (node->left || node->right) + if (node->left) + node = node->left; + else + node = node->right; + + do + { + reg_errcode_t err = fn (extra, node); + if (BE (err != REG_NOERROR, 0)) + return err; + if (node->parent == NULL) + return REG_NOERROR; + prev = node; + node = node->parent; + } + /* Go up while we have a node that is reached from the right. */ + while (node->right == prev || node->right == NULL); + node = node->right; + } +} + +static reg_errcode_t +preorder (bin_tree_t *root, reg_errcode_t (fn (void *, bin_tree_t *)), + void *extra) +{ + bin_tree_t *node; + + for (node = root; ; ) + { + reg_errcode_t err = fn (extra, node); + if (BE (err != REG_NOERROR, 0)) + return err; + + /* Go to the left node, or up and to the right. */ + if (node->left) + node = node->left; + else + { + bin_tree_t *prev = NULL; + while (node->right == prev || node->right == NULL) + { + prev = node; + node = node->parent; + if (!node) + return REG_NOERROR; + } + node = node->right; + } + } +} + +/* Optimization pass: if a SUBEXP is entirely contained, strip it and tell + re_search_internal to map the inner one's opr.idx to this one's. Adjust + backreferences as well. Requires a preorder visit. */ +static reg_errcode_t +optimize_subexps (void *extra, bin_tree_t *node) +{ + re_dfa_t *dfa = (re_dfa_t *) extra; + + if (node->token.type == OP_BACK_REF && dfa->subexp_map) + { + int idx = node->token.opr.idx; + node->token.opr.idx = dfa->subexp_map[idx]; + dfa->used_bkref_map |= 1 << node->token.opr.idx; + } + + else if (node->token.type == SUBEXP + && node->left && node->left->token.type == SUBEXP) + { + Idx other_idx = node->left->token.opr.idx; + + node->left = node->left->left; + if (node->left) + node->left->parent = node; + + dfa->subexp_map[other_idx] = dfa->subexp_map[node->token.opr.idx]; + if (other_idx < BITSET_WORD_BITS) + dfa->used_bkref_map &= ~ ((bitset_word) 1 << other_idx); + } + + return REG_NOERROR; +} + +/* Lowering pass: Turn each SUBEXP node into the appropriate concatenation + of OP_OPEN_SUBEXP, the body of the SUBEXP (if any) and OP_CLOSE_SUBEXP. */ +static reg_errcode_t +lower_subexps (void *extra, bin_tree_t *node) +{ + regex_t *preg = (regex_t *) extra; + reg_errcode_t err = REG_NOERROR; + + if (node->left && node->left->token.type == SUBEXP) + { + node->left = lower_subexp (&err, preg, node->left); + if (node->left) + node->left->parent = node; + } + if (node->right && node->right->token.type == SUBEXP) + { + node->right = lower_subexp (&err, preg, node->right); + if (node->right) + node->right->parent = node; + } + + return err; +} + +static bin_tree_t * +lower_subexp (reg_errcode_t *err, regex_t *preg, bin_tree_t *node) +{ + re_dfa_t *dfa = (re_dfa_t *) preg->re_buffer; + bin_tree_t *body = node->left; + bin_tree_t *op, *cls, *tree1, *tree; + + if (preg->re_no_sub + /* We do not optimize empty subexpressions, because otherwise we may + have bad CONCAT nodes with NULL children. This is obviously not + very common, so we do not lose much. An example that triggers + this case is the sed "script" /\(\)/x. */ + && node->left != NULL + && ! (node->token.opr.idx < BITSET_WORD_BITS + && dfa->used_bkref_map & ((bitset_word) 1 << node->token.opr.idx))) + return node->left; + + /* Convert the SUBEXP node to the concatenation of an + OP_OPEN_SUBEXP, the contents, and an OP_CLOSE_SUBEXP. */ + op = create_tree (dfa, NULL, NULL, OP_OPEN_SUBEXP); + cls = create_tree (dfa, NULL, NULL, OP_CLOSE_SUBEXP); + tree1 = body ? create_tree (dfa, body, cls, CONCAT) : cls; + tree = create_tree (dfa, op, tree1, CONCAT); + if (BE (tree == NULL || tree1 == NULL || op == NULL || cls == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + + op->token.opr.idx = cls->token.opr.idx = node->token.opr.idx; + op->token.opt_subexp = cls->token.opt_subexp = node->token.opt_subexp; + return tree; +} + +/* Pass 1 in building the NFA: compute FIRST and create unlinked automaton + nodes. Requires a postorder visit. */ +static reg_errcode_t +calc_first (void *extra, bin_tree_t *node) +{ + re_dfa_t *dfa = (re_dfa_t *) extra; + if (node->token.type == CONCAT) + { + node->first = node->left->first; + node->node_idx = node->left->node_idx; + } + else + { + node->first = node; + node->node_idx = re_dfa_add_node (dfa, node->token); + if (BE (node->node_idx == REG_MISSING, 0)) + return REG_ESPACE; + } + return REG_NOERROR; +} + +/* Pass 2: compute NEXT on the tree. Preorder visit. */ +static reg_errcode_t +calc_next (void *extra, bin_tree_t *node) +{ + switch (node->token.type) + { + case OP_DUP_ASTERISK: + node->left->next = node; + break; + case CONCAT: + node->left->next = node->right->first; + node->right->next = node->next; + break; + default: + if (node->left) + node->left->next = node->next; + if (node->right) + node->right->next = node->next; + break; + } + return REG_NOERROR; +} + +/* Pass 3: link all DFA nodes to their NEXT node (any order will do). */ +static reg_errcode_t +link_nfa_nodes (void *extra, bin_tree_t *node) +{ + re_dfa_t *dfa = (re_dfa_t *) extra; + Idx idx = node->node_idx; + reg_errcode_t err = REG_NOERROR; + + switch (node->token.type) + { + case CONCAT: + break; + + case END_OF_RE: + assert (node->next == NULL); + break; + + case OP_DUP_ASTERISK: + case OP_ALT: + { + Idx left, right; + dfa->has_plural_match = 1; + if (node->left != NULL) + left = node->left->first->node_idx; + else + left = node->next->node_idx; + if (node->right != NULL) + right = node->right->first->node_idx; + else + right = node->next->node_idx; + assert (REG_VALID_INDEX (left)); + assert (REG_VALID_INDEX (right)); + err = re_node_set_init_2 (dfa->edests + idx, left, right); + } + break; + + case ANCHOR: + case OP_OPEN_SUBEXP: + case OP_CLOSE_SUBEXP: + err = re_node_set_init_1 (dfa->edests + idx, node->next->node_idx); + break; + + case OP_BACK_REF: + dfa->nexts[idx] = node->next->node_idx; + if (node->token.type == OP_BACK_REF) + re_node_set_init_1 (dfa->edests + idx, dfa->nexts[idx]); + break; + + default: + assert (!IS_EPSILON_NODE (node->token.type)); + dfa->nexts[idx] = node->next->node_idx; + break; + } + + return err; +} + +/* Duplicate the epsilon closure of the node ROOT_NODE. + Note that duplicated nodes have constraint INIT_CONSTRAINT in addition + to their own constraint. */ + +static reg_errcode_t +duplicate_node_closure (re_dfa_t *dfa, Idx top_org_node, + Idx top_clone_node, Idx root_node, + unsigned int init_constraint) +{ + Idx org_node, clone_node; + bool ok; + unsigned int constraint = init_constraint; + for (org_node = top_org_node, clone_node = top_clone_node;;) + { + Idx org_dest, clone_dest; + if (dfa->nodes[org_node].type == OP_BACK_REF) + { + /* If the back reference epsilon-transit, its destination must + also have the constraint. Then duplicate the epsilon closure + of the destination of the back reference, and store it in + edests of the back reference. */ + org_dest = dfa->nexts[org_node]; + re_node_set_empty (dfa->edests + clone_node); + clone_dest = duplicate_node (dfa, org_dest, constraint); + if (BE (clone_dest == REG_MISSING, 0)) + return REG_ESPACE; + dfa->nexts[clone_node] = dfa->nexts[org_node]; + ok = re_node_set_insert (dfa->edests + clone_node, clone_dest); + if (BE (! ok, 0)) + return REG_ESPACE; + } + else if (dfa->edests[org_node].nelem == 0) + { + /* In case of the node can't epsilon-transit, don't duplicate the + destination and store the original destination as the + destination of the node. */ + dfa->nexts[clone_node] = dfa->nexts[org_node]; + break; + } + else if (dfa->edests[org_node].nelem == 1) + { + /* In case of the node can epsilon-transit, and it has only one + destination. */ + org_dest = dfa->edests[org_node].elems[0]; + re_node_set_empty (dfa->edests + clone_node); + if (dfa->nodes[org_node].type == ANCHOR) + { + /* In case of the node has another constraint, append it. */ + if (org_node == root_node && clone_node != org_node) + { + /* ...but if the node is root_node itself, it means the + epsilon closure have a loop, then tie it to the + destination of the root_node. */ + ok = re_node_set_insert (dfa->edests + clone_node, + org_dest); + if (BE (! ok, 0)) + return REG_ESPACE; + break; + } + constraint |= dfa->nodes[org_node].opr.ctx_type; + } + clone_dest = duplicate_node (dfa, org_dest, constraint); + if (BE (clone_dest == REG_MISSING, 0)) + return REG_ESPACE; + ok = re_node_set_insert (dfa->edests + clone_node, clone_dest); + if (BE (! ok, 0)) + return REG_ESPACE; + } + else /* dfa->edests[org_node].nelem == 2 */ + { + /* In case of the node can epsilon-transit, and it has two + destinations. In the bin_tree_t and DFA, that's '|' and '*'. */ + org_dest = dfa->edests[org_node].elems[0]; + re_node_set_empty (dfa->edests + clone_node); + /* Search for a duplicated node which satisfies the constraint. */ + clone_dest = search_duplicated_node (dfa, org_dest, constraint); + if (clone_dest == REG_MISSING) + { + /* There are no such a duplicated node, create a new one. */ + reg_errcode_t err; + clone_dest = duplicate_node (dfa, org_dest, constraint); + if (BE (clone_dest == REG_MISSING, 0)) + return REG_ESPACE; + ok = re_node_set_insert (dfa->edests + clone_node, clone_dest); + if (BE (! ok, 0)) + return REG_ESPACE; + err = duplicate_node_closure (dfa, org_dest, clone_dest, + root_node, constraint); + if (BE (err != REG_NOERROR, 0)) + return err; + } + else + { + /* There are a duplicated node which satisfy the constraint, + use it to avoid infinite loop. */ + ok = re_node_set_insert (dfa->edests + clone_node, clone_dest); + if (BE (! ok, 0)) + return REG_ESPACE; + } + + org_dest = dfa->edests[org_node].elems[1]; + clone_dest = duplicate_node (dfa, org_dest, constraint); + if (BE (clone_dest == REG_MISSING, 0)) + return REG_ESPACE; + ok = re_node_set_insert (dfa->edests + clone_node, clone_dest); + if (BE (! ok, 0)) + return REG_ESPACE; + } + org_node = org_dest; + clone_node = clone_dest; + } + return REG_NOERROR; +} + +/* Search for a node which is duplicated from the node ORG_NODE, and + satisfies the constraint CONSTRAINT. */ + +static Idx +search_duplicated_node (const re_dfa_t *dfa, Idx org_node, + unsigned int constraint) +{ + Idx idx; + for (idx = dfa->nodes_len - 1; dfa->nodes[idx].duplicated && idx > 0; --idx) + { + if (org_node == dfa->org_indices[idx] + && constraint == dfa->nodes[idx].constraint) + return idx; /* Found. */ + } + return REG_MISSING; /* Not found. */ +} + +/* Duplicate the node whose index is ORG_IDX and set the constraint CONSTRAINT. + Return the index of the new node, or REG_MISSING if insufficient storage is + available. */ + +static Idx +duplicate_node (re_dfa_t *dfa, Idx org_idx, unsigned int constraint) +{ + Idx dup_idx = re_dfa_add_node (dfa, dfa->nodes[org_idx]); + if (BE (dup_idx != REG_MISSING, 1)) + { + dfa->nodes[dup_idx].constraint = constraint; + if (dfa->nodes[org_idx].type == ANCHOR) + dfa->nodes[dup_idx].constraint |= dfa->nodes[org_idx].opr.ctx_type; + dfa->nodes[dup_idx].duplicated = 1; + + /* Store the index of the original node. */ + dfa->org_indices[dup_idx] = org_idx; + } + return dup_idx; +} + +static reg_errcode_t +calc_inveclosure (re_dfa_t *dfa) +{ + Idx src, idx; + bool ok; + for (idx = 0; idx < dfa->nodes_len; ++idx) + re_node_set_init_empty (dfa->inveclosures + idx); + + for (src = 0; src < dfa->nodes_len; ++src) + { + Idx *elems = dfa->eclosures[src].elems; + for (idx = 0; idx < dfa->eclosures[src].nelem; ++idx) + { + ok = re_node_set_insert_last (dfa->inveclosures + elems[idx], src); + if (BE (! ok, 0)) + return REG_ESPACE; + } + } + + return REG_NOERROR; +} + +/* Calculate "eclosure" for all the node in DFA. */ + +static reg_errcode_t +calc_eclosure (re_dfa_t *dfa) +{ + Idx node_idx; + bool incomplete; +#ifdef DEBUG + assert (dfa->nodes_len > 0); +#endif + incomplete = false; + /* For each nodes, calculate epsilon closure. */ + for (node_idx = 0; ; ++node_idx) + { + reg_errcode_t err; + re_node_set eclosure_elem; + if (node_idx == dfa->nodes_len) + { + if (!incomplete) + break; + incomplete = false; + node_idx = 0; + } + +#ifdef DEBUG + assert (dfa->eclosures[node_idx].nelem != REG_MISSING); +#endif + + /* If we have already calculated, skip it. */ + if (dfa->eclosures[node_idx].nelem != 0) + continue; + /* Calculate epsilon closure of `node_idx'. */ + err = calc_eclosure_iter (&eclosure_elem, dfa, node_idx, true); + if (BE (err != REG_NOERROR, 0)) + return err; + + if (dfa->eclosures[node_idx].nelem == 0) + { + incomplete = true; + re_node_set_free (&eclosure_elem); + } + } + return REG_NOERROR; +} + +/* Calculate epsilon closure of NODE. */ + +static reg_errcode_t +calc_eclosure_iter (re_node_set *new_set, re_dfa_t *dfa, Idx node, bool root) +{ + reg_errcode_t err; + unsigned int constraint; + Idx i; + bool incomplete; + bool ok; + re_node_set eclosure; + incomplete = false; + err = re_node_set_alloc (&eclosure, dfa->edests[node].nelem + 1); + if (BE (err != REG_NOERROR, 0)) + return err; + + /* This indicates that we are calculating this node now. + We reference this value to avoid infinite loop. */ + dfa->eclosures[node].nelem = REG_MISSING; + + constraint = ((dfa->nodes[node].type == ANCHOR) + ? dfa->nodes[node].opr.ctx_type : 0); + /* If the current node has constraints, duplicate all nodes. + Since they must inherit the constraints. */ + if (constraint + && dfa->edests[node].nelem + && !dfa->nodes[dfa->edests[node].elems[0]].duplicated) + { + Idx org_node, cur_node; + org_node = cur_node = node; + err = duplicate_node_closure (dfa, node, node, node, constraint); + if (BE (err != REG_NOERROR, 0)) + return err; + } + + /* Expand each epsilon destination nodes. */ + if (IS_EPSILON_NODE(dfa->nodes[node].type)) + for (i = 0; i < dfa->edests[node].nelem; ++i) + { + re_node_set eclosure_elem; + Idx edest = dfa->edests[node].elems[i]; + /* If calculating the epsilon closure of `edest' is in progress, + return intermediate result. */ + if (dfa->eclosures[edest].nelem == REG_MISSING) + { + incomplete = true; + continue; + } + /* If we haven't calculated the epsilon closure of `edest' yet, + calculate now. Otherwise use calculated epsilon closure. */ + if (dfa->eclosures[edest].nelem == 0) + { + err = calc_eclosure_iter (&eclosure_elem, dfa, edest, false); + if (BE (err != REG_NOERROR, 0)) + return err; + } + else + eclosure_elem = dfa->eclosures[edest]; + /* Merge the epsilon closure of `edest'. */ + re_node_set_merge (&eclosure, &eclosure_elem); + /* If the epsilon closure of `edest' is incomplete, + the epsilon closure of this node is also incomplete. */ + if (dfa->eclosures[edest].nelem == 0) + { + incomplete = true; + re_node_set_free (&eclosure_elem); + } + } + + /* Epsilon closures include itself. */ + ok = re_node_set_insert (&eclosure, node); + if (BE (! ok, 0)) + return REG_ESPACE; + if (incomplete && !root) + dfa->eclosures[node].nelem = 0; + else + dfa->eclosures[node] = eclosure; + *new_set = eclosure; + return REG_NOERROR; +} + +/* Functions for token which are used in the parser. */ + +/* Fetch a token from INPUT. + We must not use this function inside bracket expressions. */ + +static void +fetch_token (re_token_t *result, re_string_t *input, reg_syntax_t syntax) +{ + re_string_skip_bytes (input, peek_token (result, input, syntax)); +} + +/* Peek a token from INPUT, and return the length of the token. + We must not use this function inside bracket expressions. */ + +static int +peek_token (re_token_t *token, re_string_t *input, reg_syntax_t syntax) +{ + unsigned char c; + + if (re_string_eoi (input)) + { + token->type = END_OF_RE; + return 0; + } + + c = re_string_peek_byte (input, 0); + token->opr.c = c; + + token->word_char = 0; +#ifdef RE_ENABLE_I18N + token->mb_partial = 0; + if (input->mb_cur_max > 1 && + !re_string_first_byte (input, re_string_cur_idx (input))) + { + token->type = CHARACTER; + token->mb_partial = 1; + return 1; + } +#endif + if (c == '\\') + { + unsigned char c2; + if (re_string_cur_idx (input) + 1 >= re_string_length (input)) + { + token->type = BACK_SLASH; + return 1; + } + + c2 = re_string_peek_byte_case (input, 1); + token->opr.c = c2; + token->type = CHARACTER; +#ifdef RE_ENABLE_I18N + if (input->mb_cur_max > 1) + { + wint_t wc = re_string_wchar_at (input, + re_string_cur_idx (input) + 1); + token->word_char = IS_WIDE_WORD_CHAR (wc) != 0; + } + else +#endif + token->word_char = IS_WORD_CHAR (c2) != 0; + + switch (c2) + { + case '|': + if (!(syntax & REG_LIMITED_OPS) && !(syntax & REG_NO_BK_VBAR)) + token->type = OP_ALT; + break; + case '1': case '2': case '3': case '4': case '5': + case '6': case '7': case '8': case '9': + if (!(syntax & REG_NO_BK_REFS)) + { + token->type = OP_BACK_REF; + token->opr.idx = c2 - '1'; + } + break; + case '<': + if (!(syntax & REG_NO_GNU_OPS)) + { + token->type = ANCHOR; + token->opr.ctx_type = WORD_FIRST; + } + break; + case '>': + if (!(syntax & REG_NO_GNU_OPS)) + { + token->type = ANCHOR; + token->opr.ctx_type = WORD_LAST; + } + break; + case 'b': + if (!(syntax & REG_NO_GNU_OPS)) + { + token->type = ANCHOR; + token->opr.ctx_type = WORD_DELIM; + } + break; + case 'B': + if (!(syntax & REG_NO_GNU_OPS)) + { + token->type = ANCHOR; + token->opr.ctx_type = NOT_WORD_DELIM; + } + break; + case 'w': + if (!(syntax & REG_NO_GNU_OPS)) + token->type = OP_WORD; + break; + case 'W': + if (!(syntax & REG_NO_GNU_OPS)) + token->type = OP_NOTWORD; + break; + case 's': + if (!(syntax & REG_NO_GNU_OPS)) + token->type = OP_SPACE; + break; + case 'S': + if (!(syntax & REG_NO_GNU_OPS)) + token->type = OP_NOTSPACE; + break; + case '`': + if (!(syntax & REG_NO_GNU_OPS)) + { + token->type = ANCHOR; + token->opr.ctx_type = BUF_FIRST; + } + break; + case '\'': + if (!(syntax & REG_NO_GNU_OPS)) + { + token->type = ANCHOR; + token->opr.ctx_type = BUF_LAST; + } + break; + case '(': + if (!(syntax & REG_NO_BK_PARENS)) + token->type = OP_OPEN_SUBEXP; + break; + case ')': + if (!(syntax & REG_NO_BK_PARENS)) + token->type = OP_CLOSE_SUBEXP; + break; + case '+': + if (!(syntax & REG_LIMITED_OPS) && (syntax & REG_BK_PLUS_QM)) + token->type = OP_DUP_PLUS; + break; + case '?': + if (!(syntax & REG_LIMITED_OPS) && (syntax & REG_BK_PLUS_QM)) + token->type = OP_DUP_QUESTION; + break; + case '{': + if ((syntax & REG_INTERVALS) && (!(syntax & REG_NO_BK_BRACES))) + token->type = OP_OPEN_DUP_NUM; + break; + case '}': + if ((syntax & REG_INTERVALS) && (!(syntax & REG_NO_BK_BRACES))) + token->type = OP_CLOSE_DUP_NUM; + break; + default: + break; + } + return 2; + } + + token->type = CHARACTER; +#ifdef RE_ENABLE_I18N + if (input->mb_cur_max > 1) + { + wint_t wc = re_string_wchar_at (input, re_string_cur_idx (input)); + token->word_char = IS_WIDE_WORD_CHAR (wc) != 0; + } + else +#endif + token->word_char = IS_WORD_CHAR (token->opr.c); + + switch (c) + { + case '\n': + if (syntax & REG_NEWLINE_ALT) + token->type = OP_ALT; + break; + case '|': + if (!(syntax & REG_LIMITED_OPS) && (syntax & REG_NO_BK_VBAR)) + token->type = OP_ALT; + break; + case '*': + token->type = OP_DUP_ASTERISK; + break; + case '+': + if (!(syntax & REG_LIMITED_OPS) && !(syntax & REG_BK_PLUS_QM)) + token->type = OP_DUP_PLUS; + break; + case '?': + if (!(syntax & REG_LIMITED_OPS) && !(syntax & REG_BK_PLUS_QM)) + token->type = OP_DUP_QUESTION; + break; + case '{': + if ((syntax & REG_INTERVALS) && (syntax & REG_NO_BK_BRACES)) + token->type = OP_OPEN_DUP_NUM; + break; + case '}': + if ((syntax & REG_INTERVALS) && (syntax & REG_NO_BK_BRACES)) + token->type = OP_CLOSE_DUP_NUM; + break; + case '(': + if (syntax & REG_NO_BK_PARENS) + token->type = OP_OPEN_SUBEXP; + break; + case ')': + if (syntax & REG_NO_BK_PARENS) + token->type = OP_CLOSE_SUBEXP; + break; + case '[': + token->type = OP_OPEN_BRACKET; + break; + case '.': + token->type = OP_PERIOD; + break; + case '^': + if (!(syntax & (REG_CONTEXT_INDEP_ANCHORS | REG_CARET_ANCHORS_HERE)) && + re_string_cur_idx (input) != 0) + { + char prev = re_string_peek_byte (input, -1); + if (!(syntax & REG_NEWLINE_ALT) || prev != '\n') + break; + } + token->type = ANCHOR; + token->opr.ctx_type = LINE_FIRST; + break; + case '$': + if (!(syntax & REG_CONTEXT_INDEP_ANCHORS) && + re_string_cur_idx (input) + 1 != re_string_length (input)) + { + re_token_t next; + re_string_skip_bytes (input, 1); + peek_token (&next, input, syntax); + re_string_skip_bytes (input, -1); + if (next.type != OP_ALT && next.type != OP_CLOSE_SUBEXP) + break; + } + token->type = ANCHOR; + token->opr.ctx_type = LINE_LAST; + break; + default: + break; + } + return 1; +} + +/* Peek a token from INPUT, and return the length of the token. + We must not use this function out of bracket expressions. */ + +static int +peek_token_bracket (re_token_t *token, re_string_t *input, reg_syntax_t syntax) +{ + unsigned char c; + if (re_string_eoi (input)) + { + token->type = END_OF_RE; + return 0; + } + c = re_string_peek_byte (input, 0); + token->opr.c = c; + +#ifdef RE_ENABLE_I18N + if (input->mb_cur_max > 1 && + !re_string_first_byte (input, re_string_cur_idx (input))) + { + token->type = CHARACTER; + return 1; + } +#endif /* RE_ENABLE_I18N */ + + if (c == '\\' && (syntax & REG_BACKSLASH_ESCAPE_IN_LISTS) + && re_string_cur_idx (input) + 1 < re_string_length (input)) + { + /* In this case, '\' escape a character. */ + unsigned char c2; + re_string_skip_bytes (input, 1); + c2 = re_string_peek_byte (input, 0); + token->opr.c = c2; + token->type = CHARACTER; + return 1; + } + if (c == '[') /* '[' is a special char in a bracket exps. */ + { + unsigned char c2; + int token_len; + if (re_string_cur_idx (input) + 1 < re_string_length (input)) + c2 = re_string_peek_byte (input, 1); + else + c2 = 0; + token->opr.c = c2; + token_len = 2; + switch (c2) + { + case '.': + token->type = OP_OPEN_COLL_ELEM; + break; + case '=': + token->type = OP_OPEN_EQUIV_CLASS; + break; + case ':': + if (syntax & REG_CHAR_CLASSES) + { + token->type = OP_OPEN_CHAR_CLASS; + break; + } + /* else fall through. */ + default: + token->type = CHARACTER; + token->opr.c = c; + token_len = 1; + break; + } + return token_len; + } + switch (c) + { + case '-': + token->type = OP_CHARSET_RANGE; + break; + case ']': + token->type = OP_CLOSE_BRACKET; + break; + case '^': + token->type = OP_NON_MATCH_LIST; + break; + default: + token->type = CHARACTER; + } + return 1; +} + +/* Functions for parser. */ + +/* Entry point of the parser. + Parse the regular expression REGEXP and return the structure tree. + If an error is occured, ERR is set by error code, and return NULL. + This function build the following tree, from regular expression <reg_exp>: + CAT + / \ + / \ + <reg_exp> EOR + + CAT means concatenation. + EOR means end of regular expression. */ + +static bin_tree_t * +parse (re_string_t *regexp, regex_t *preg, reg_syntax_t syntax, + reg_errcode_t *err) +{ + re_dfa_t *dfa = (re_dfa_t *) preg->re_buffer; + bin_tree_t *tree, *eor, *root; + re_token_t current_token; + dfa->syntax = syntax; + fetch_token (¤t_token, regexp, syntax | REG_CARET_ANCHORS_HERE); + tree = parse_reg_exp (regexp, preg, ¤t_token, syntax, 0, err); + if (BE (*err != REG_NOERROR && tree == NULL, 0)) + return NULL; + eor = create_tree (dfa, NULL, NULL, END_OF_RE); + if (tree != NULL) + root = create_tree (dfa, tree, eor, CONCAT); + else + root = eor; + if (BE (eor == NULL || root == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + return root; +} + +/* This function build the following tree, from regular expression + <branch1>|<branch2>: + ALT + / \ + / \ + <branch1> <branch2> + + ALT means alternative, which represents the operator `|'. */ + +static bin_tree_t * +parse_reg_exp (re_string_t *regexp, regex_t *preg, re_token_t *token, + reg_syntax_t syntax, Idx nest, reg_errcode_t *err) +{ + re_dfa_t *dfa = (re_dfa_t *) preg->re_buffer; + bin_tree_t *tree, *branch = NULL; + tree = parse_branch (regexp, preg, token, syntax, nest, err); + if (BE (*err != REG_NOERROR && tree == NULL, 0)) + return NULL; + + while (token->type == OP_ALT) + { + fetch_token (token, regexp, syntax | REG_CARET_ANCHORS_HERE); + if (token->type != OP_ALT && token->type != END_OF_RE + && (nest == 0 || token->type != OP_CLOSE_SUBEXP)) + { + branch = parse_branch (regexp, preg, token, syntax, nest, err); + if (BE (*err != REG_NOERROR && branch == NULL, 0)) + return NULL; + } + else + branch = NULL; + tree = create_tree (dfa, tree, branch, OP_ALT); + if (BE (tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + } + return tree; +} + +/* This function build the following tree, from regular expression + <exp1><exp2>: + CAT + / \ + / \ + <exp1> <exp2> + + CAT means concatenation. */ + +static bin_tree_t * +parse_branch (re_string_t *regexp, regex_t *preg, re_token_t *token, + reg_syntax_t syntax, Idx nest, reg_errcode_t *err) +{ + bin_tree_t *tree, *exp; + re_dfa_t *dfa = (re_dfa_t *) preg->re_buffer; + tree = parse_expression (regexp, preg, token, syntax, nest, err); + if (BE (*err != REG_NOERROR && tree == NULL, 0)) + return NULL; + + while (token->type != OP_ALT && token->type != END_OF_RE + && (nest == 0 || token->type != OP_CLOSE_SUBEXP)) + { + exp = parse_expression (regexp, preg, token, syntax, nest, err); + if (BE (*err != REG_NOERROR && exp == NULL, 0)) + { + return NULL; + } + if (tree != NULL && exp != NULL) + { + tree = create_tree (dfa, tree, exp, CONCAT); + if (tree == NULL) + { + *err = REG_ESPACE; + return NULL; + } + } + else if (tree == NULL) + tree = exp; + /* Otherwise exp == NULL, we don't need to create new tree. */ + } + return tree; +} + +/* This function build the following tree, from regular expression a*: + * + | + a +*/ + +static bin_tree_t * +parse_expression (re_string_t *regexp, regex_t *preg, re_token_t *token, + reg_syntax_t syntax, Idx nest, reg_errcode_t *err) +{ + re_dfa_t *dfa = (re_dfa_t *) preg->re_buffer; + bin_tree_t *tree; + switch (token->type) + { + case CHARACTER: + tree = create_token_tree (dfa, NULL, NULL, token); + if (BE (tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } +#ifdef RE_ENABLE_I18N + if (dfa->mb_cur_max > 1) + { + while (!re_string_eoi (regexp) + && !re_string_first_byte (regexp, re_string_cur_idx (regexp))) + { + bin_tree_t *mbc_remain; + fetch_token (token, regexp, syntax); + mbc_remain = create_token_tree (dfa, NULL, NULL, token); + tree = create_tree (dfa, tree, mbc_remain, CONCAT); + if (BE (mbc_remain == NULL || tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + } + } +#endif + break; + case OP_OPEN_SUBEXP: + tree = parse_sub_exp (regexp, preg, token, syntax, nest + 1, err); + if (BE (*err != REG_NOERROR && tree == NULL, 0)) + return NULL; + break; + case OP_OPEN_BRACKET: + tree = parse_bracket_exp (regexp, dfa, token, syntax, err); + if (BE (*err != REG_NOERROR && tree == NULL, 0)) + return NULL; + break; + case OP_BACK_REF: + if (!BE (dfa->completed_bkref_map & (1 << token->opr.idx), 1)) + { + *err = REG_ESUBREG; + return NULL; + } + dfa->used_bkref_map |= 1 << token->opr.idx; + tree = create_token_tree (dfa, NULL, NULL, token); + if (BE (tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + ++dfa->nbackref; + dfa->has_mb_node = 1; + break; + case OP_OPEN_DUP_NUM: + if (syntax & REG_CONTEXT_INVALID_DUP) + { + *err = REG_BADRPT; + return NULL; + } + /* FALLTHROUGH */ + case OP_DUP_ASTERISK: + case OP_DUP_PLUS: + case OP_DUP_QUESTION: + if (syntax & REG_CONTEXT_INVALID_OPS) + { + *err = REG_BADRPT; + return NULL; + } + else if (syntax & REG_CONTEXT_INDEP_OPS) + { + fetch_token (token, regexp, syntax); + return parse_expression (regexp, preg, token, syntax, nest, err); + } + /* else fall through */ + case OP_CLOSE_SUBEXP: + if ((token->type == OP_CLOSE_SUBEXP) && + !(syntax & REG_UNMATCHED_RIGHT_PAREN_ORD)) + { + *err = REG_ERPAREN; + return NULL; + } + /* else fall through */ + case OP_CLOSE_DUP_NUM: + /* We treat it as a normal character. */ + + /* Then we can these characters as normal characters. */ + token->type = CHARACTER; + /* mb_partial and word_char bits should be initialized already + by peek_token. */ + tree = create_token_tree (dfa, NULL, NULL, token); + if (BE (tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + break; + case ANCHOR: + if ((token->opr.ctx_type + & (WORD_DELIM | NOT_WORD_DELIM | WORD_FIRST | WORD_LAST)) + && dfa->word_ops_used == 0) + init_word_char (dfa); + if (token->opr.ctx_type == WORD_DELIM + || token->opr.ctx_type == NOT_WORD_DELIM) + { + bin_tree_t *tree_first, *tree_last; + if (token->opr.ctx_type == WORD_DELIM) + { + token->opr.ctx_type = WORD_FIRST; + tree_first = create_token_tree (dfa, NULL, NULL, token); + token->opr.ctx_type = WORD_LAST; + } + else + { + token->opr.ctx_type = INSIDE_WORD; + tree_first = create_token_tree (dfa, NULL, NULL, token); + token->opr.ctx_type = INSIDE_NOTWORD; + } + tree_last = create_token_tree (dfa, NULL, NULL, token); + tree = create_tree (dfa, tree_first, tree_last, OP_ALT); + if (BE (tree_first == NULL || tree_last == NULL || tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + } + else + { + tree = create_token_tree (dfa, NULL, NULL, token); + if (BE (tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + } + /* We must return here, since ANCHORs can't be followed + by repetition operators. + eg. RE"^*" is invalid or "<ANCHOR(^)><CHAR(*)>", + it must not be "<ANCHOR(^)><REPEAT(*)>". */ + fetch_token (token, regexp, syntax); + return tree; + case OP_PERIOD: + tree = create_token_tree (dfa, NULL, NULL, token); + if (BE (tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + if (dfa->mb_cur_max > 1) + dfa->has_mb_node = 1; + break; + case OP_WORD: + case OP_NOTWORD: + tree = build_charclass_op (dfa, regexp->trans, + (const unsigned char *) "alnum", + (const unsigned char *) "_", + token->type == OP_NOTWORD, err); + if (BE (*err != REG_NOERROR && tree == NULL, 0)) + return NULL; + break; + case OP_SPACE: + case OP_NOTSPACE: + tree = build_charclass_op (dfa, regexp->trans, + (const unsigned char *) "space", + (const unsigned char *) "", + token->type == OP_NOTSPACE, err); + if (BE (*err != REG_NOERROR && tree == NULL, 0)) + return NULL; + break; + case OP_ALT: + case END_OF_RE: + return NULL; + case BACK_SLASH: + *err = REG_EESCAPE; + return NULL; + default: + /* Must not happen? */ +#ifdef DEBUG + assert (0); +#endif + return NULL; + } + fetch_token (token, regexp, syntax); + + while (token->type == OP_DUP_ASTERISK || token->type == OP_DUP_PLUS + || token->type == OP_DUP_QUESTION || token->type == OP_OPEN_DUP_NUM) + { + tree = parse_dup_op (tree, regexp, dfa, token, syntax, err); + if (BE (*err != REG_NOERROR && tree == NULL, 0)) + return NULL; + /* In BRE consecutive duplications are not allowed. */ + if ((syntax & REG_CONTEXT_INVALID_DUP) + && (token->type == OP_DUP_ASTERISK + || token->type == OP_OPEN_DUP_NUM)) + { + *err = REG_BADRPT; + return NULL; + } + } + + return tree; +} + +/* This function build the following tree, from regular expression + (<reg_exp>): + SUBEXP + | + <reg_exp> +*/ + +static bin_tree_t * +parse_sub_exp (re_string_t *regexp, regex_t *preg, re_token_t *token, + reg_syntax_t syntax, Idx nest, reg_errcode_t *err) +{ + re_dfa_t *dfa = (re_dfa_t *) preg->re_buffer; + bin_tree_t *tree; + size_t cur_nsub; + cur_nsub = preg->re_nsub++; + + fetch_token (token, regexp, syntax | REG_CARET_ANCHORS_HERE); + + /* The subexpression may be a null string. */ + if (token->type == OP_CLOSE_SUBEXP) + tree = NULL; + else + { + tree = parse_reg_exp (regexp, preg, token, syntax, nest, err); + if (BE (*err == REG_NOERROR && token->type != OP_CLOSE_SUBEXP, 0)) + *err = REG_EPAREN; + if (BE (*err != REG_NOERROR, 0)) + return NULL; + } + + if (cur_nsub <= '9' - '1') + dfa->completed_bkref_map |= 1 << cur_nsub; + + tree = create_tree (dfa, tree, NULL, SUBEXP); + if (BE (tree == NULL, 0)) + { + *err = REG_ESPACE; + return NULL; + } + tree->token.opr.idx = cur_nsub; + return tree; +} + +/* This function parse repetition operators like "*", "+", "{1,3}" etc. */ + +static bin_tree_t * +parse_dup_op (bin_tree_t *elem, re_string_t *regexp, re_dfa_t *dfa, + re_token_t *token, reg_syntax_t syntax, reg_errcode_t *err) +{ + bin_tree_t *tree = NULL, *old_tree = NULL; + Idx i, start, end, start_idx = re_string_cur_idx (regexp); + re_token_t start_token = *token; + + if (token->type == OP_OPEN_DUP_NUM) + { + end = 0; + start = fetch_number (regexp, token, syntax); + if (start == REG_MISSING) + { + if (token->type == CHARACTER && token->opr.c == ',') + start = 0; /* We treat "{,m}" as "{0,m}". */ + else + { + *err = REG_BADBR; /* <re>{} is invalid. */ + return NULL; + } + } + if (BE (start != REG_ERROR, 1)) + { + /* We treat "{n}" as "{n,n}". */ + end = ((token->type == OP_CLOSE_DUP_NUM) ? start + : ((token->type == CHARACTER && token->opr.c == ',') + ? fetch_number (regexp, token, syntax) : REG_ERROR)); + } + if (BE (start == REG_ERROR || end == REG_ERROR, 0)) + { + /* Invalid sequence. */ + if (BE (!(syntax & REG_INVALID_INTERVAL_ORD), 0)) + { + if (token->type == END_OF_RE) + *err = REG_EBRACE; + else + *err = REG_BADBR; + + return NULL; + } + + /* If the syntax bit is set, rollback. */ + re_string_set_index (regexp, start_idx); + *token = start_token; + token->type = CHARACTER; + /* mb_partial and word_char bits should be already initialized by + peek_token. */ + return elem; + } + + if (BE (end != REG_MISSING && start > end, 0)) + { + /* First number greater than second. */ + *err = REG_BADBR; + return NULL; + } + } + else + { + start = (token->type == OP_DUP_PLUS) ? 1 : 0; + end = (token->type == OP_DUP_QUESTION) ? 1 : REG_MISSING; + } + + fetch_token (token, regexp, syntax); + + if (BE (elem == NULL, 0)) + return NULL; + if (BE (start == 0 && end == 0, 0)) + { + postorder (elem, free_tree, NULL); + return NULL; + } + + /* Extract "<re>{n,m}" to "<re><re>...<re><re>{0,<m-n>}". */ + if (BE (start > 0, 0)) + { + tree = elem; + for (i = 2; i <= start; ++i) + { + elem = duplicate_tree (elem, dfa); + tree = create_tree (dfa, tree, elem, CONCAT); + if (BE (elem == NULL || tree == NULL, 0)) + goto parse_dup_op_espace; + } + + if (start == end) + return tree; + + /* Duplicate ELEM before it is marked optional. */ + elem = duplicate_tree (elem, dfa); + old_tree = tree; + } + else + old_tree = NULL; + + if (elem->token.type == SUBEXP) + postorder (elem, mark_opt_subexp, (void *) (long) elem->token.opr.idx); + + tree = create_tree (dfa, elem, NULL, + (end == REG_MISSING ? OP_DUP_ASTERISK : OP_ALT)); + if (BE (tree == NULL, 0)) + goto parse_dup_op_espace; + + /* This loop is actually executed only when end != REG_MISSING, + to rewrite <re>{0,n} as (<re>(<re>...<re>?)?)?... We have + already created the start+1-th copy. */ + if ((Idx) -1 < 0 || end != REG_MISSING) + for (i = start + 2; i <= end; ++i) + { + elem = duplicate_tree (elem, dfa); + tree = create_tree (dfa, tree, elem, CONCAT); + if (BE (elem == NULL || tree == NULL, 0)) + goto parse_dup_op_espace; + + tree = create_tree (dfa, tree, NULL, OP_ALT); + if (BE (tree == NULL, 0)) + goto parse_dup_op_espace; + } + + if (old_tree) + tree = create_tree (dfa, old_tree, tree, CONCAT); + + return tree; + + parse_dup_op_espace: + *err = REG_ESPACE; + return NULL; +} + +/* Size of the names for collating symbol/equivalence_class/character_class. + I'm not sure, but maybe enough. */ +#define BRACKET_NAME_BUF_SIZE 32 + +#ifndef _LIBC + /* Local function for parse_bracket_exp only used in case of NOT _LIBC. + Build the range expression which starts from START_ELEM, and ends + at END_ELEM. The result are written to MBCSET and SBCSET. + RANGE_ALLOC is the allocated size of mbcset->range_starts, and + mbcset->range_ends, is a pointer argument sinse we may + update it. */ + +static reg_errcode_t +build_range_exp (bitset sbcset, +# ifdef RE_ENABLE_I18N + re_charset_t *mbcset, Idx *range_alloc, +# endif + bracket_elem_t *start_elem, bracket_elem_t *end_elem) +{ + unsigned int start_ch, end_ch; + /* Equivalence Classes and Character Classes can't be a range start/end. */ + if (BE (start_elem->type == EQUIV_CLASS || start_elem->type == CHAR_CLASS + || end_elem->type == EQUIV_CLASS || end_elem->type == CHAR_CLASS, + 0)) + return REG_ERANGE; + + /* We can handle no multi character collating elements without libc + support. */ + if (BE ((start_elem->type == COLL_SYM + && strlen ((char *) start_elem->opr.name) > 1) + || (end_elem->type == COLL_SYM + && strlen ((char *) end_elem->opr.name) > 1), 0)) + return REG_ECOLLATE; + +# ifdef RE_ENABLE_I18N + { + wchar_t wc; + wint_t start_wc, end_wc; + wchar_t cmp_buf[6] = {L'\0', L'\0', L'\0', L'\0', L'\0', L'\0'}; + + start_ch = ((start_elem->type == SB_CHAR) ? start_elem->opr.ch + : ((start_elem->type == COLL_SYM) ? start_elem->opr.name[0] + : 0)); + end_ch = ((end_elem->type == SB_CHAR) ? end_elem->opr.ch + : ((end_elem->type == COLL_SYM) ? end_elem->opr.name[0] + : 0)); + start_wc = ((start_elem->type == SB_CHAR || start_elem->type == COLL_SYM) + ? __btowc (start_ch) : start_elem->opr.wch); + end_wc = ((end_elem->type == SB_CHAR || end_elem->type == COLL_SYM) + ? __btowc (end_ch) : end_elem->opr.wch); + if (start_wc == WEOF || end_wc == WEOF) + return REG_ECOLLATE; + cmp_buf[0] = start_wc; + cmp_buf[4] = end_wc; + if (wcscoll (cmp_buf, cmp_buf + 4) > 0) + return REG_ERANGE; + + /* Got valid collation sequence values, add them as a new entry. + However, for !_LIBC we have no collation elements: if the + character set is single byte, the single byte character set + that we build below suffices. parse_bracket_exp passes + no MBCSET if dfa->mb_cur_max == 1. */ + if (mbcset) + { + /* Check the space of the arrays. */ + if (BE (*range_alloc == mbcset->nranges, 0)) + { + /* There is not enough space, need realloc. */ + wchar_t *new_array_start, *new_array_end; + Idx new_nranges; + + new_nranges = mbcset->nranges; + /* Use realloc since mbcset->range_starts and mbcset->range_ends + are NULL if *range_alloc == 0. */ + new_array_start = re_x2realloc (mbcset->range_starts, wchar_t, + &new_nranges); + new_array_end = re_realloc (mbcset->range_ends, wchar_t, + new_nranges); + + if (BE (new_array_start == NULL || new_array_end == NULL, 0)) + return REG_ESPACE; + + mbcset->range_starts = new_array_start; + mbcset->range_ends = new_array_end; + *range_alloc = new_nranges; + } + + mbcset->range_starts[mbcset->nranges] = start_wc; + mbcset->range_ends[mbcset->nranges++] = end_wc; + } + + /* Build the table for single byte characters. */ + for (wc = 0; wc < SBC_MAX; ++wc) + { + cmp_buf[2] = wc; + if (wcscoll (cmp_buf, cmp_buf + 2) <= 0 + && wcscoll (cmp_buf + 2, cmp_buf + 4) <= 0) + bitset_set (sbcset, wc); + } + } +# else /* not RE_ENABLE_I18N */ + { + unsigned int ch; + start_ch = ((start_elem->type == SB_CHAR ) ? start_elem->opr.ch + : ((start_elem->type == COLL_SYM) ? start_elem->opr.name[0] + : 0)); + end_ch = ((end_elem->type == SB_CHAR ) ? end_elem->opr.ch + : ((end_elem->type == COLL_SYM) ? end_elem->opr.name[0] + : 0)); + if (start_ch > end_ch) + return REG_ERANGE; + /* Build the table for single byte characters. */ + for (ch = 0; ch < SBC_MAX; ++ch) + if (start_ch <= ch && ch <= end_ch) + bitset_set (sbcset, ch); + } +# endif /* not RE_ENABLE_I18N */ + return REG_NOERROR; +} +#endif /* not _LIBC */ + +#ifndef _LIBC +/* Helper function for parse_bracket_exp only used in case of NOT _LIBC.. + Build the collating element which is represented by NAME. + The result are written to MBCSET and SBCSET. + COLL_SYM_ALLOC is the allocated size of mbcset->coll_sym, is a + pointer argument since we may update it. */ + +static reg_errcode_t +build_collating_symbol (bitset sbcset, +# ifdef RE_ENABLE_I18N + re_charset_t *mbcset, Idx *coll_sym_alloc, +# endif + const unsigned char *name) +{ + size_t name_len = strlen ((const char *) name); + if (BE (name_len != 1, 0)) + return REG_ECOLLATE; + else + { + bitset_set (sbcset, name[0]); + return REG_NOERROR; + } +} +#endif /* not _LIBC */ + +/* This function parse bracket expression like "[abc]", "[a-c]", + "[[.a-a.]]" etc. */ + +static bin_tree_t * +parse_bracket_exp (re_string_t *regexp, re_dfa_t *dfa, re_token_t *token, + reg_syntax_t syntax, reg_errcode_t *err) +{ +#ifdef _LIBC + const unsigned char *collseqmb; + const char *collseqwc; + uint32_t nrules; + int32_t table_size; + const int32_t *symb_table; + const unsigned char *extra; + + /* Local function for parse_bracket_exp used in _LIBC environement. + Seek the collating symbol entry correspondings to NAME. + Return the index of the symbol in the SYMB_TABLE. */ + + auto inline int32_t + __attribute ((always_inline)) + seek_collating_symbol_entry (const unsigned char *name, size_t name_len) + { + int32_t hash = elem_hash ((const char *) name, name_len); + int32_t elem = hash % table_size; + int32_t second = hash % (table_size - 2); + while (symb_table[2 * elem] != 0) + { + /* First compare the hashing value. */ + if (symb_table[2 * elem] == hash + /* Compare the length of the name. */ + && name_len == extra[symb_table[2 * elem + 1]] + /* Compare the name. */ + && memcmp (name, &extra[symb_table[2 * elem + 1] + 1], + name_len) == 0) + { + /* Yep, this is the entry. */ + break; + } + + /* Next entry. */ + elem += second; + } + return elem; + } + + /* Local function for parse_bracket_exp used in _LIBC environement. + Look up the collation sequence value of BR_ELEM. + Return the value if succeeded, UINT_MAX otherwise. */ + + auto inline unsigned int + __attribute ((always_inline)) + lookup_collation_sequence_value (bracket_elem_t *br_elem) + { + if (br_elem->type == SB_CHAR) + { + /* + if (MB_CUR_MAX == 1) + */ + if (nrules == 0) + return collseqmb[br_elem->opr.ch]; + else + { + wint_t wc = __btowc (br_elem->opr.ch); + return __collseq_table_lookup (collseqwc, wc); + } + } + else if (br_elem->type == MB_CHAR) + { + return __collseq_table_lookup (collseqwc, br_elem->opr.wch); + } + else if (br_elem->type == COLL_SYM) + { + size_t sym_name_len = strlen ((char *) br_elem->opr.name); + if (nrules != 0) + { + int32_t elem, idx; + elem = seek_collating_symbol_entry (br_elem->opr.name, + sym_name_len); + if (symb_table[2 * elem] != 0) + { + /* We found the entry. */ + idx = symb_table[2 * elem + 1]; + /* Skip the name of collating element name. */ + idx += 1 + extra[idx]; + /* Skip the byte sequence of the collating element. */ + idx += 1 + extra[idx]; + /* Adjust for the alignment. */ + idx = (idx + 3) & ~3; + /* Skip the multibyte collation sequence value. */ + idx += sizeof (unsigned int); + /* Skip the wide char sequence of the collating element. */ + idx += sizeof (unsigned int) * + (1 + *(unsigned int *) (extra + idx)); + /* Return the collation sequence value. */ + return *(unsigned int *) (extra + idx); + } + else if (symb_table[2 * elem] == 0 && sym_name_len == 1) + { + /* No valid character. Match it as a single byte + character. */ + return collseqmb[br_elem->opr.name[0]]; + } + } + else if (sym_name_len == 1) + return collseqmb[br_elem->opr.name[0]]; + } + return UINT_MAX; + } + + /* Local function for parse_bracket_exp used in _LIBC environement. + Build the range expression which starts from START_ELEM, and ends + at END_ELEM. The result are written to MBCSET and SBCSET. + RANGE_ALLOC is the allocated size of mbcset->range_starts, and + mbcset->range_ends, is a pointer argument sinse we may + update it. */ + + auto inline reg_errcode_t + __attribute ((always_inline)) + build_range_exp (bitset sbcset, re_charset_t *mbcset, + Idx *range_alloc, + bracket_elem_t *start_elem, bracket_elem_t *end_elem) + { + unsigned int ch; + uint32_t start_collseq; + uint32_t end_collseq; + + /* Equivalence Classes and Character Classes can't be a range + start/end. */ + if (BE (start_elem->type == EQUIV_CLASS || start_elem->type == CHAR_CLASS + || end_elem->type == EQUIV_CLASS || end_elem->type == CHAR_CLASS, + 0)) + return REG_ERANGE; + + start_collseq = lookup_collation_sequence_value (start_elem); + end_collseq = lookup_collation_sequence_value (end_elem); + /* Check start/end collation sequence values. */ + if (BE (start_collseq == UINT_MAX || end_collseq == UINT_MAX, 0)) + return REG_ECOLLATE; + if (BE ((syntax & REG_NO_EMPTY_RANGES) && start_collseq > end_collseq, 0)) + return REG_ERANGE; + + /* Got valid collation sequence values, add them as a new entry. + However, if we have no collation elements, and the character set + is single byte, the single byte character set that we + build below suffices. */ + if (nrules > 0 || dfa->mb_cur_max > 1) + { + /* Check the space of the arrays. */ + if (BE (*range_alloc == mbcset->nranges, 0)) + { + /* There is not enough space, need realloc. */ + uint32_t *new_array_start; + uint32_t *new_array_end; + Idx new_nranges; + + new_nranges = mbcset->nranges; + new_array_start = re_x2realloc (mbcset->range_starts, uint32_t, + &new_nranges); + new_array_end = re_realloc (mbcset->range_ends, uint32_t, + new_nranges); + + if (BE (new_array_start == NULL || new_array_end == NULL, 0)) + return REG_ESPACE; + + mbcset->range_starts = new_array_start; + mbcset->range_ends = new_array_end; + *range_alloc = new_nranges; + } + + mbcset->range_starts[mbcset->nranges] = start_collseq; + mbcset->range_ends[mbcset->nranges++] = end_collseq; + } + + /* Build the table for single byte characters. */ + for (ch = 0; ch < SBC_MAX; ch++) + { + uint32_t ch_collseq; + /* + if (MB_CUR_MAX == 1) + */ + if (nrules == 0) + ch_collseq = collseqmb[ch]; + else + ch_collseq = __collseq_table_lookup (collseqwc, __btowc (ch)); + if (start_collseq <= ch_collseq && ch_collseq <= end_collseq) + bitset_set (sbcset, ch); + } + return REG_NOERROR; + } + + /* Local function for parse_bracket_exp used in _LIBC environement. + Build the collating element which is represented by NAME. + The result are written to MBCSET and SBCSET. + COLL_SYM_ALLOC is the allocated size of mbcset->coll_sym, is a + pointer argument sinse we may update it. */ + + auto inline reg_errcode_t + __attribute ((always_inline)) + build_collating_symbol (bitset sbcset, re_charset_t *mbcset, + Idx *coll_sym_alloc, const unsigned char *name) + { + int32_t elem, idx; + size_t name_len = strlen ((const char *) name); + if (nrules != 0) + { + elem = seek_collating_symbol_entry (name, name_len); + if (symb_table[2 * elem] != 0) + { + /* We found the entry. */ + idx = symb_table[2 * elem + 1]; + /* Skip the name of collating element name. */ + idx += 1 + extra[idx]; + } + else if (symb_table[2 * elem] == 0 && name_len == 1) + { + /* No valid character, treat it as a normal + character. */ + bitset_set (sbcset, name[0]); + return REG_NOERROR; + } + else + return REG_ECOLLATE; + + /* Got valid collation sequence, add it as a new entry. */ + /* Check the space of the arrays. */ + if (BE (*coll_sym_alloc == mbcset->ncoll_syms, 0)) + { + /* Not enough, realloc it. */ + Idx new_coll_sym_alloc = mbcset->ncoll_syms; + /* Use realloc since mbcset->coll_syms is NULL + if *alloc == 0. */ + int32_t *new_coll_syms = re_x2realloc (mbcset->coll_syms, int32_t, + &new_coll_sym_alloc); + if (BE (new_coll_syms == NULL, 0)) + return REG_ESPACE; + mbcset->coll_syms = new_coll_syms; + *coll_sym_alloc = new_coll_sym_alloc; + } + mbcset->coll_syms[mbcset->ncoll_syms++] = idx; + return REG_NOERROR; + } + else + { + if (BE (name_len != 1, 0)) + return REG_ECOLLATE; + else + { + bitset_set (sbcset, name[0]); + return REG_NOERROR; + } + } + } +#endif + + re_token_t br_token; + re_bitset_ptr_t sbcset; +#ifdef RE_ENABLE_I18N + re_charset_t *mbcset; + Idx coll_sym_alloc = 0, range_alloc = 0, mbchar_alloc = 0; + Idx equiv_class_alloc = 0, char_class_alloc = 0; +#endif /* not RE_ENABLE_I18N */ + bool non_match = false; + bin_tree_t *work_tree; + int token_len; + bool first_round = true; +#ifdef _LIBC + collseqmb = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_COLLSEQMB); + nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES); + if (nrules) + { + /* + if (MB_CUR_MAX > 1) + */ + collseqwc = _NL_CURRENT (LC_COLLATE, _NL_COLLATE_COLLSEQWC); + table_size = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_SYMB_HASH_SIZEMB); + symb_table = (const int32_t *) _NL_CURRENT (LC_COLLATE, + _NL_COLLATE_SYMB_TABLEMB); + extra = (const unsigned char *) _NL_CURRENT (LC_COLLATE, + _NL_COLLATE_SYMB_EXTRAMB); + } +#endif + sbcset = re_calloc (bitset_word, BITSET_WORDS); +#ifdef RE_ENABLE_I18N + mbcset = re_calloc (re_charset_t, 1); +#endif /* RE_ENABLE_I18N */ +#ifdef RE_ENABLE_I18N + if (BE (sbcset == NULL || mbcset == NULL, 0)) +#else + if (BE (sbcset == NULL, 0)) +#endif /* RE_ENABLE_I18N */ + { + *err = REG_ESPACE; + return NULL; + } + + token_len = peek_token_bracket (token, regexp, syntax); + if (BE (token->type == END_OF_RE, 0)) + { + *err = REG_BADPAT; + goto parse_bracket_exp_free_return; + } + if (token->type == OP_NON_MATCH_LIST) + { +#ifdef RE_ENABLE_I18N + mbcset->non_match = 1; +#endif /* not RE_ENABLE_I18N */ + non_match = true; + if (syntax & REG_HAT_LISTS_NOT_NEWLINE) + bitset_set (sbcset, '\0'); + re_string_skip_bytes (regexp, token_len); /* Skip a token. */ + token_len = peek_token_bracket (token, regexp, syntax); + if (BE (token->type == END_OF_RE, 0)) + { + *err = REG_BADPAT; + goto parse_bracket_exp_free_return; + } + } + + /* We treat the first ']' as a normal character. */ + if (token->type == OP_CLOSE_BRACKET) + token->type = CHARACTER; + + while (1) + { + bracket_elem_t start_elem, end_elem; + unsigned char start_name_buf[BRACKET_NAME_BUF_SIZE]; + unsigned char end_name_buf[BRACKET_NAME_BUF_SIZE]; + reg_errcode_t ret; + int token_len2 = 0; + bool is_range_exp = false; + re_token_t token2; + + start_elem.opr.name = start_name_buf; + ret = parse_bracket_element (&start_elem, regexp, token, token_len, dfa, + syntax, first_round); + if (BE (ret != REG_NOERROR, 0)) + { + *err = ret; + goto parse_bracket_exp_free_return; + } + first_round = false; + + /* Get information about the next token. We need it in any case. */ + token_len = peek_token_bracket (token, regexp, syntax); + + /* Do not check for ranges if we know they are not allowed. */ + if (start_elem.type != CHAR_CLASS && start_elem.type != EQUIV_CLASS) + { + if (BE (token->type == END_OF_RE, 0)) + { + *err = REG_EBRACK; + goto parse_bracket_exp_free_return; + } + if (token->type == OP_CHARSET_RANGE) + { + re_string_skip_bytes (regexp, token_len); /* Skip '-'. */ + token_len2 = peek_token_bracket (&token2, regexp, syntax); + if (BE (token2.type == END_OF_RE, 0)) + { + *err = REG_EBRACK; + goto parse_bracket_exp_free_return; + } + if (token2.type == OP_CLOSE_BRACKET) + { + /* We treat the last '-' as a normal character. */ + re_string_skip_bytes (regexp, -token_len); + token->type = CHARACTER; + } + else + is_range_exp = true; + } + } + + if (is_range_exp == true) + { + end_elem.opr.name = end_name_buf; + ret = parse_bracket_element (&end_elem, regexp, &token2, token_len2, + dfa, syntax, true); + if (BE (ret != REG_NOERROR, 0)) + { + *err = ret; + goto parse_bracket_exp_free_return; + } + + token_len = peek_token_bracket (token, regexp, syntax); + +#ifdef _LIBC + *err = build_range_exp (sbcset, mbcset, &range_alloc, + &start_elem, &end_elem); +#else +# ifdef RE_ENABLE_I18N + *err = build_range_exp (sbcset, + dfa->mb_cur_max > 1 ? mbcset : NULL, + &range_alloc, &start_elem, &end_elem); +# else + *err = build_range_exp (sbcset, &start_elem, &end_elem); +# endif +#endif /* RE_ENABLE_I18N */ + if (BE (*err != REG_NOERROR, 0)) + goto parse_bracket_exp_free_return; + } + else + { + switch (start_elem.type) + { + case SB_CHAR: + bitset_set (sbcset, start_elem.opr.ch); + break; +#ifdef RE_ENABLE_I18N + case MB_CHAR: + /* Check whether the array has enough space. */ + if (BE (mbchar_alloc == mbcset->nmbchars, 0)) + { + wchar_t *new_mbchars; + /* Not enough, realloc it. */ + mbchar_alloc = mbcset->nmbchars; + /* Use realloc since array is NULL if *alloc == 0. */ + new_mbchars = re_x2realloc (mbcset->mbchars, wchar_t, + &mbchar_alloc); + if (BE (new_mbchars == NULL, 0)) + goto parse_bracket_exp_espace; + mbcset->mbchars = new_mbchars; + } + mbcset->mbchars[mbcset->nmbchars++] = start_elem.opr.wch; + break; +#endif /* RE_ENABLE_I18N */ + case EQUIV_CLASS: + *err = build_equiv_class (sbcset, +#ifdef RE_ENABLE_I18N + mbcset, &equiv_class_alloc, +#endif /* RE_ENABLE_I18N */ + start_elem.opr.name); + if (BE (*err != REG_NOERROR, 0)) + goto parse_bracket_exp_free_return; + break; + case COLL_SYM: + *err = build_collating_symbol (sbcset, +#ifdef RE_ENABLE_I18N + mbcset, &coll_sym_alloc, +#endif /* RE_ENABLE_I18N */ + start_elem.opr.name); + if (BE (*err != REG_NOERROR, 0)) + goto parse_bracket_exp_free_return; + break; + case CHAR_CLASS: + *err = build_charclass (regexp->trans, sbcset, +#ifdef RE_ENABLE_I18N + mbcset, &char_class_alloc, +#endif /* RE_ENABLE_I18N */ + start_elem.opr.name, syntax); + if (BE (*err != REG_NOERROR, 0)) + goto parse_bracket_exp_free_return; + break; + default: + assert (0); + break; + } + } + if (BE (token->type == END_OF_RE, 0)) + { + *err = REG_EBRACK; + goto parse_bracket_exp_free_return; + } + if (token->type == OP_CLOSE_BRACKET) + break; + } + + re_string_skip_bytes (regexp, token_len); /* Skip a token. */ + + /* If it is non-matching list. */ + if (non_match) + bitset_not (sbcset); + +#ifdef RE_ENABLE_I18N + /* Ensure only single byte characters are set. */ + if (dfa->mb_cur_max > 1) + bitset_mask (sbcset, dfa->sb_char); + + if (mbcset->nmbchars || mbcset->ncoll_syms || mbcset->nequiv_classes + || mbcset->nranges || (dfa->mb_cur_max > 1 && (mbcset->nchar_classes + || mbcset->non_match))) + { + bin_tree_t *mbc_tree; + int sbc_idx; + /* Build a tree for complex bracket. */ + dfa->has_mb_node = 1; + br_token.type = COMPLEX_BRACKET; + br_token.opr.mbcset = mbcset; + mbc_tree = create_token_tree (dfa, NULL, NULL, &br_token); + if (BE (mbc_tree == NULL, 0)) + goto parse_bracket_exp_espace; + for (sbc_idx = 0; sbc_idx < BITSET_WORDS; ++sbc_idx) + if (sbcset[sbc_idx]) + break; + /* If there are no bits set in sbcset, there is no point + of having both SIMPLE_BRACKET and COMPLEX_BRACKET. */ + if (sbc_idx < BITSET_WORDS) + { + /* Build a tree for simple bracket. */ + br_token.type = SIMPLE_BRACKET; + br_token.opr.sbcset = sbcset; + work_tree = create_token_tree (dfa, NULL, NULL, &br_token); + if (BE (work_tree == NULL, 0)) + goto parse_bracket_exp_espace; + + /* Then join them by ALT node. */ + work_tree = create_tree (dfa, work_tree, mbc_tree, OP_ALT); + if (BE (work_tree == NULL, 0)) + goto parse_bracket_exp_espace; + } + else + { + re_free (sbcset); + work_tree = mbc_tree; + } + } + else +#endif /* not RE_ENABLE_I18N */ + { +#ifdef RE_ENABLE_I18N + free_charset (mbcset); +#endif + /* Build a tree for simple bracket. */ + br_token.type = SIMPLE_BRACKET; + br_token.opr.sbcset = sbcset; + work_tree = create_token_tree (dfa, NULL, NULL, &br_token); + if (BE (work_tree == NULL, 0)) + goto parse_bracket_exp_espace; + } + return work_tree; + + parse_bracket_exp_espace: + *err = REG_ESPACE; + parse_bracket_exp_free_return: + re_free (sbcset); +#ifdef RE_ENABLE_I18N + free_charset (mbcset); +#endif /* RE_ENABLE_I18N */ + return NULL; +} + +/* Parse an element in the bracket expression. */ + +static reg_errcode_t +parse_bracket_element (bracket_elem_t *elem, re_string_t *regexp, + re_token_t *token, int token_len, re_dfa_t *dfa, + reg_syntax_t syntax, bool accept_hyphen) +{ +#ifdef RE_ENABLE_I18N + int cur_char_size; + cur_char_size = re_string_char_size_at (regexp, re_string_cur_idx (regexp)); + if (cur_char_size > 1) + { + elem->type = MB_CHAR; + elem->opr.wch = re_string_wchar_at (regexp, re_string_cur_idx (regexp)); + re_string_skip_bytes (regexp, cur_char_size); + return REG_NOERROR; + } +#endif /* RE_ENABLE_I18N */ + re_string_skip_bytes (regexp, token_len); /* Skip a token. */ + if (token->type == OP_OPEN_COLL_ELEM || token->type == OP_OPEN_CHAR_CLASS + || token->type == OP_OPEN_EQUIV_CLASS) + return parse_bracket_symbol (elem, regexp, token); + if (BE (token->type == OP_CHARSET_RANGE, 0) && !accept_hyphen) + { + /* A '-' must only appear as anything but a range indicator before + the closing bracket. Everything else is an error. */ + re_token_t token2; + (void) peek_token_bracket (&token2, regexp, syntax); + if (token2.type != OP_CLOSE_BRACKET) + /* The actual error value is not standardized since this whole + case is undefined. But ERANGE makes good sense. */ + return REG_ERANGE; + } + elem->type = SB_CHAR; + elem->opr.ch = token->opr.c; + return REG_NOERROR; +} + +/* Parse a bracket symbol in the bracket expression. Bracket symbols are + such as [:<character_class>:], [.<collating_element>.], and + [=<equivalent_class>=]. */ + +static reg_errcode_t +parse_bracket_symbol (bracket_elem_t *elem, re_string_t *regexp, + re_token_t *token) +{ + unsigned char ch, delim = token->opr.c; + int i = 0; + if (re_string_eoi(regexp)) + return REG_EBRACK; + for (;; ++i) + { + if (i >= BRACKET_NAME_BUF_SIZE) + return REG_EBRACK; + if (token->type == OP_OPEN_CHAR_CLASS) + ch = re_string_fetch_byte_case (regexp); + else + ch = re_string_fetch_byte (regexp); + if (re_string_eoi(regexp)) + return REG_EBRACK; + if (ch == delim && re_string_peek_byte (regexp, 0) == ']') + break; + elem->opr.name[i] = ch; + } + re_string_skip_bytes (regexp, 1); + elem->opr.name[i] = '\0'; + switch (token->type) + { + case OP_OPEN_COLL_ELEM: + elem->type = COLL_SYM; + break; + case OP_OPEN_EQUIV_CLASS: + elem->type = EQUIV_CLASS; + break; + case OP_OPEN_CHAR_CLASS: + elem->type = CHAR_CLASS; + break; + default: + break; + } + return REG_NOERROR; +} + + /* Helper function for parse_bracket_exp. + Build the equivalence class which is represented by NAME. + The result are written to MBCSET and SBCSET. + EQUIV_CLASS_ALLOC is the allocated size of mbcset->equiv_classes, + is a pointer argument sinse we may update it. */ + +static reg_errcode_t +build_equiv_class (bitset sbcset, +#ifdef RE_ENABLE_I18N + re_charset_t *mbcset, Idx *equiv_class_alloc, +#endif + const unsigned char *name) +{ +#if defined _LIBC + uint32_t nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES); + if (nrules != 0) + { + const int32_t *table, *indirect; + const unsigned char *weights, *extra, *cp; + unsigned char char_buf[2]; + int32_t idx1, idx2; + unsigned int ch; + size_t len; + /* This #include defines a local function! */ +# include <locale/weight.h> + /* Calculate the index for equivalence class. */ + cp = name; + table = (const int32_t *) _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB); + weights = (const unsigned char *) _NL_CURRENT (LC_COLLATE, + _NL_COLLATE_WEIGHTMB); + extra = (const unsigned char *) _NL_CURRENT (LC_COLLATE, + _NL_COLLATE_EXTRAMB); + indirect = (const int32_t *) _NL_CURRENT (LC_COLLATE, + _NL_COLLATE_INDIRECTMB); + idx1 = findidx (&cp); + if (BE (idx1 == 0 || cp < name + strlen ((const char *) name), 0)) + /* This isn't a valid character. */ + return REG_ECOLLATE; + + /* Build single byte matcing table for this equivalence class. */ + char_buf[1] = (unsigned char) '\0'; + len = weights[idx1]; + for (ch = 0; ch < SBC_MAX; ++ch) + { + char_buf[0] = ch; + cp = char_buf; + idx2 = findidx (&cp); +/* + idx2 = table[ch]; +*/ + if (idx2 == 0) + /* This isn't a valid character. */ + continue; + if (len == weights[idx2]) + { + int cnt = 0; + while (cnt <= len && + weights[idx1 + 1 + cnt] == weights[idx2 + 1 + cnt]) + ++cnt; + + if (cnt > len) + bitset_set (sbcset, ch); + } + } + /* Check whether the array has enough space. */ + if (BE (*equiv_class_alloc == mbcset->nequiv_classes, 0)) + { + /* Not enough, realloc it. */ + Idx new_equiv_class_alloc = mbcset->nequiv_classes; + /* Use realloc since the array is NULL if *alloc == 0. */ + int32_t *new_equiv_classes = re_x2realloc (mbcset->equiv_classes, + int32_t, + &new_equiv_class_alloc); + if (BE (new_equiv_classes == NULL, 0)) + return REG_ESPACE; + mbcset->equiv_classes = new_equiv_classes; + *equiv_class_alloc = new_equiv_class_alloc; + } + mbcset->equiv_classes[mbcset->nequiv_classes++] = idx1; + } + else +#endif /* _LIBC */ + { + if (BE (strlen ((const char *) name) != 1, 0)) + return REG_ECOLLATE; + bitset_set (sbcset, *name); + } + return REG_NOERROR; +} + + /* Helper function for parse_bracket_exp. + Build the character class which is represented by NAME. + The result are written to MBCSET and SBCSET. + CHAR_CLASS_ALLOC is the allocated size of mbcset->char_classes, + is a pointer argument sinse we may update it. */ + +static reg_errcode_t +build_charclass (unsigned REG_TRANSLATE_TYPE trans, bitset sbcset, +#ifdef RE_ENABLE_I18N + re_charset_t *mbcset, Idx *char_class_alloc, +#endif + const unsigned char *class_name, reg_syntax_t syntax) +{ + int i; + const char *name = (const char *) class_name; + + /* In case of REG_ICASE "upper" and "lower" match the both of + upper and lower cases. */ + if ((syntax & REG_IGNORE_CASE) + && (strcmp (name, "upper") == 0 || strcmp (name, "lower") == 0)) + name = "alpha"; + +#ifdef RE_ENABLE_I18N + /* Check the space of the arrays. */ + if (BE (*char_class_alloc == mbcset->nchar_classes, 0)) + { + /* Not enough, realloc it. */ + Idx new_char_class_alloc = mbcset->nchar_classes; + /* Use realloc since array is NULL if *alloc == 0. */ + wctype_t *new_char_classes = re_x2realloc (mbcset->char_classes, wctype_t, + &new_char_class_alloc); + if (BE (new_char_classes == NULL, 0)) + return REG_ESPACE; + mbcset->char_classes = new_char_classes; + *char_class_alloc = new_char_class_alloc; + } + mbcset->char_classes[mbcset->nchar_classes++] = __wctype (name); +#endif /* RE_ENABLE_I18N */ + +#define BUILD_CHARCLASS_LOOP(ctype_func) \ + for (i = 0; i < SBC_MAX; ++i) \ + { \ + if (ctype_func (i)) \ + { \ + int ch = trans ? trans[i] : i; \ + bitset_set (sbcset, ch); \ + } \ + } + + if (strcmp (name, "alnum") == 0) + BUILD_CHARCLASS_LOOP (isalnum) + else if (strcmp (name, "cntrl") == 0) + BUILD_CHARCLASS_LOOP (iscntrl) + else if (strcmp (name, "lower") == 0) + BUILD_CHARCLASS_LOOP (islower) + else if (strcmp (name, "space") == 0) + BUILD_CHARCLASS_LOOP (isspace) + else if (strcmp (name, "alpha") == 0) + BUILD_CHARCLASS_LOOP (isalpha) + else if (strcmp (name, "digit") == 0) + BUILD_CHARCLASS_LOOP (isdigit) + else if (strcmp (name, "print") == 0) + BUILD_CHARCLASS_LOOP (isprint) + else if (strcmp (name, "upper") == 0) + BUILD_CHARCLASS_LOOP (isupper) + else if (strcmp (name, "blank") == 0) + BUILD_CHARCLASS_LOOP (isblank) + else if (strcmp (name, "graph") == 0) + BUILD_CHARCLASS_LOOP (isgraph) + else if (strcmp (name, "punct") == 0) + BUILD_CHARCLASS_LOOP (ispunct) + else if (strcmp (name, "xdigit") == 0) + BUILD_CHARCLASS_LOOP (isxdigit) + else + return REG_ECTYPE; + + return REG_NOERROR; +} + +static bin_tree_t * +build_charclass_op (re_dfa_t *dfa, unsigned REG_TRANSLATE_TYPE trans, + const unsigned char *class_name, + const unsigned char *extra, + bool non_match, reg_errcode_t *err) +{ + re_bitset_ptr_t sbcset; +#ifdef RE_ENABLE_I18N + re_charset_t *mbcset; + Idx alloc = 0; +#endif /* not RE_ENABLE_I18N */ + reg_errcode_t ret; + re_token_t br_token; + bin_tree_t *tree; + + sbcset = re_calloc (bitset_word, BITSET_WORDS); +#ifdef RE_ENABLE_I18N + mbcset = re_calloc (re_charset_t, 1); +#endif /* RE_ENABLE_I18N */ + +#ifdef RE_ENABLE_I18N + if (BE (sbcset == NULL || mbcset == NULL, 0)) +#else /* not RE_ENABLE_I18N */ + if (BE (sbcset == NULL, 0)) +#endif /* not RE_ENABLE_I18N */ + { + *err = REG_ESPACE; + return NULL; + } + + if (non_match) + { +#ifdef RE_ENABLE_I18N + /* + if (syntax & REG_HAT_LISTS_NOT_NEWLINE) + bitset_set(cset->sbcset, '\0'); + */ + mbcset->non_match = 1; +#endif /* not RE_ENABLE_I18N */ + } + + /* We don't care the syntax in this case. */ + ret = build_charclass (trans, sbcset, +#ifdef RE_ENABLE_I18N + mbcset, &alloc, +#endif /* RE_ENABLE_I18N */ + class_name, 0); + + if (BE (ret != REG_NOERROR, 0)) + { + re_free (sbcset); +#ifdef RE_ENABLE_I18N + free_charset (mbcset); +#endif /* RE_ENABLE_I18N */ + *err = ret; + return NULL; + } + /* \w match '_' also. */ + for (; *extra; extra++) + bitset_set (sbcset, *extra); + + /* If it is non-matching list. */ + if (non_match) + bitset_not (sbcset); + +#ifdef RE_ENABLE_I18N + /* Ensure only single byte characters are set. */ + if (dfa->mb_cur_max > 1) + bitset_mask (sbcset, dfa->sb_char); +#endif + + /* Build a tree for simple bracket. */ + br_token.type = SIMPLE_BRACKET; + br_token.opr.sbcset = sbcset; + tree = create_token_tree (dfa, NULL, NULL, &br_token); + if (BE (tree == NULL, 0)) + goto build_word_op_espace; + +#ifdef RE_ENABLE_I18N + if (dfa->mb_cur_max > 1) + { + bin_tree_t *mbc_tree; + /* Build a tree for complex bracket. */ + br_token.type = COMPLEX_BRACKET; + br_token.opr.mbcset = mbcset; + dfa->has_mb_node = 1; + mbc_tree = create_token_tree (dfa, NULL, NULL, &br_token); + if (BE (mbc_tree == NULL, 0)) + goto build_word_op_espace; + /* Then join them by ALT node. */ + tree = create_tree (dfa, tree, mbc_tree, OP_ALT); + if (BE (mbc_tree != NULL, 1)) + return tree; + } + else + { + free_charset (mbcset); + return tree; + } +#else /* not RE_ENABLE_I18N */ + return tree; +#endif /* not RE_ENABLE_I18N */ + + build_word_op_espace: + re_free (sbcset); +#ifdef RE_ENABLE_I18N + free_charset (mbcset); +#endif /* RE_ENABLE_I18N */ + *err = REG_ESPACE; + return NULL; +} + +/* This is intended for the expressions like "a{1,3}". + Fetch a number from `input', and return the number. + Return REG_MISSING if the number field is empty like "{,1}". + Return REG_ERROR if an error occurred. */ + +static Idx +fetch_number (re_string_t *input, re_token_t *token, reg_syntax_t syntax) +{ + Idx num = REG_MISSING; + unsigned char c; + while (1) + { + fetch_token (token, input, syntax); + c = token->opr.c; + if (BE (token->type == END_OF_RE, 0)) + return REG_ERROR; + if (token->type == OP_CLOSE_DUP_NUM || c == ',') + break; + num = ((token->type != CHARACTER || c < '0' || '9' < c + || num == REG_ERROR) + ? REG_ERROR + : ((num == REG_MISSING) ? c - '0' : num * 10 + c - '0')); + num = (num > REG_DUP_MAX) ? REG_ERROR : num; + } + return num; +} + +#ifdef RE_ENABLE_I18N +static void +free_charset (re_charset_t *cset) +{ + re_free (cset->mbchars); +# ifdef _LIBC + re_free (cset->coll_syms); + re_free (cset->equiv_classes); + re_free (cset->range_starts); + re_free (cset->range_ends); +# endif + re_free (cset->char_classes); + re_free (cset); +} +#endif /* RE_ENABLE_I18N */ + +/* Functions for binary tree operation. */ + +/* Create a tree node. */ + +static bin_tree_t * +create_tree (re_dfa_t *dfa, bin_tree_t *left, bin_tree_t *right, + re_token_type_t type) +{ + re_token_t t; + t.type = type; + return create_token_tree (dfa, left, right, &t); +} + +static bin_tree_t * +create_token_tree (re_dfa_t *dfa, bin_tree_t *left, bin_tree_t *right, + const re_token_t *token) +{ + bin_tree_t *tree; + if (BE (dfa->str_tree_storage_idx == BIN_TREE_STORAGE_SIZE, 0)) + { + bin_tree_storage_t *storage = re_malloc (bin_tree_storage_t, 1); + + if (storage == NULL) + return NULL; + storage->next = dfa->str_tree_storage; + dfa->str_tree_storage = storage; + dfa->str_tree_storage_idx = 0; + } + tree = &dfa->str_tree_storage->data[dfa->str_tree_storage_idx++]; + + tree->parent = NULL; + tree->left = left; + tree->right = right; + tree->token = *token; + tree->token.duplicated = 0; + tree->token.opt_subexp = 0; + tree->first = NULL; + tree->next = NULL; + tree->node_idx = REG_MISSING; + + if (left != NULL) + left->parent = tree; + if (right != NULL) + right->parent = tree; + return tree; +} + +/* Mark the tree SRC as an optional subexpression. + To be called from preorder or postorder. */ + +static reg_errcode_t +mark_opt_subexp (void *extra, bin_tree_t *node) +{ + Idx idx = (Idx) (long) extra; + if (node->token.type == SUBEXP && node->token.opr.idx == idx) + node->token.opt_subexp = 1; + + return REG_NOERROR; +} + +/* Free the allocated memory inside NODE. */ + +static void +free_token (re_token_t *node) +{ +#ifdef RE_ENABLE_I18N + if (node->type == COMPLEX_BRACKET && node->duplicated == 0) + free_charset (node->opr.mbcset); + else +#endif /* RE_ENABLE_I18N */ + if (node->type == SIMPLE_BRACKET && node->duplicated == 0) + re_free (node->opr.sbcset); +} + +/* Worker function for tree walking. Free the allocated memory inside NODE + and its children. */ + +static reg_errcode_t +free_tree (void *extra, bin_tree_t *node) +{ + free_token (&node->token); + return REG_NOERROR; +} + + +/* Duplicate the node SRC, and return new node. This is a preorder + visit similar to the one implemented by the generic visitor, but + we need more infrastructure to maintain two parallel trees --- so, + it's easier to duplicate. */ + +static bin_tree_t * +duplicate_tree (const bin_tree_t *root, re_dfa_t *dfa) +{ + const bin_tree_t *node; + bin_tree_t *dup_root; + bin_tree_t **p_new = &dup_root, *dup_node = root->parent; + + for (node = root; ; ) + { + /* Create a new tree and link it back to the current parent. */ + *p_new = create_token_tree (dfa, NULL, NULL, &node->token); + if (*p_new == NULL) + return NULL; + (*p_new)->parent = dup_node; + (*p_new)->token.duplicated = 1; + dup_node = *p_new; + + /* Go to the left node, or up and to the right. */ + if (node->left) + { + node = node->left; + p_new = &dup_node->left; + } + else + { + const bin_tree_t *prev = NULL; + while (node->right == prev || node->right == NULL) + { + prev = node; + node = node->parent; + dup_node = dup_node->parent; + if (!node) + return dup_root; + } + node = node->right; + p_new = &dup_node->right; + } + } +} diff --git a/lib/regex.c b/lib/regex.c new file mode 100644 index 0000000..82e76c0 --- /dev/null +++ b/lib/regex.c @@ -0,0 +1,68 @@ +/* Extended regular expression matching and search library. + Copyright (C) 2002, 2003 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#ifdef _LIBC +/* We have to keep the namespace clean. */ +# define regfree(preg) __regfree (preg) +# define regexec(pr, st, nm, pm, ef) __regexec (pr, st, nm, pm, ef) +# define regcomp(preg, pattern, cflags) __regcomp (preg, pattern, cflags) +# define regerror(errcode, preg, errbuf, errbuf_size) \ + __regerror(errcode, preg, errbuf, errbuf_size) +# define re_set_registers(bu, re, nu, st, en) \ + __re_set_registers (bu, re, nu, st, en) +# define re_match_2(bufp, string1, size1, string2, size2, pos, regs, stop) \ + __re_match_2 (bufp, string1, size1, string2, size2, pos, regs, stop) +# define re_match(bufp, string, size, pos, regs) \ + __re_match (bufp, string, size, pos, regs) +# define re_search(bufp, string, size, startpos, range, regs) \ + __re_search (bufp, string, size, startpos, range, regs) +# define re_compile_pattern(pattern, length, bufp) \ + __re_compile_pattern (pattern, length, bufp) +# define re_set_syntax(syntax) __re_set_syntax (syntax) +# define re_search_2(bufp, st1, s1, st2, s2, startpos, range, regs, stop) \ + __re_search_2 (bufp, st1, s1, st2, s2, startpos, range, regs, stop) +# define re_compile_fastmap(bufp) __re_compile_fastmap (bufp) + +# include "../locale/localeinfo.h" +#endif + +/* On some systems, limits.h sets RE_DUP_MAX to a lower value than + GNU regex allows. Include it before <regex.h>, which correctly + #undefs RE_DUP_MAX and sets it to the right value. */ +#include <limits.h> + +#include <regex.h> +#include "regex_internal.h" + +#include "regex_internal.c" +#include "regcomp.c" +#include "regexec.c" + +/* Binary backward compatibility. */ +#if _LIBC +# include <shlib-compat.h> +# if SHLIB_COMPAT (libc, GLIBC_2_0, GLIBC_2_3) +link_warning (re_max_failures, "the 're_max_failures' variable is obsolete and will go away.") +int re_max_failures = 2000; +# endif +#endif diff --git a/lib/regex.h b/lib/regex.h new file mode 100644 index 0000000..c06a062 --- /dev/null +++ b/lib/regex.h @@ -0,0 +1,769 @@ +/* Definitions for data structures and routines for the regular + expression library. + Copyright (C) 1985,1989-93,1995-98,2000,2001,2002,2003,2005 + Free Software Foundation, Inc. + This file is part of the GNU C Library. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _REGEX_H +#define _REGEX_H 1 + +#include <sys/types.h> + +/* Allow the use in C++ code. */ +#ifdef __cplusplus +extern "C" { +#endif + +/* Define _REGEX_SOURCE to get definitions that are incompatible with + POSIX. */ +#if (!defined _REGEX_SOURCE \ + && (defined _GNU_SOURCE \ + || (!defined _POSIX_C_SOURCE && !defined _POSIX_SOURCE \ + && !defined _XOPEN_SOURCE))) +# define _REGEX_SOURCE 1 +#endif + +#if defined _REGEX_SOURCE && defined VMS +/* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it + should be there. */ +# include <stddef.h> +#endif + +#ifdef _REGEX_LARGE_OFFSETS + +/* Use types and values that are wide enough to represent signed and + unsigned byte offsets in memory. This currently works only when + the regex code is used outside of the GNU C library; it is not yet + supported within glibc itself, and glibc users should not define + _REGEX_LARGE_OFFSETS. */ + +/* The type of the offset of a byte within a string. + For historical reasons POSIX 1003.1-2004 requires that regoff_t be + at least as wide as off_t. This is a bit odd (and many common + POSIX platforms set it to the more-sensible ssize_t) but we might + as well conform. We don't know of any hosts where ssize_t is wider + than off_t, so off_t is safe. */ +typedef off_t regoff_t; + +/* The type of nonnegative object indexes. Traditionally, GNU regex + uses 'int' for these. Code that uses __re_idx_t should work + regardless of whether the type is signed. */ +typedef size_t __re_idx_t; + +/* The type of object sizes. */ +typedef size_t __re_size_t; + +/* The type of object sizes, in places where the traditional code + uses unsigned long int. */ +typedef size_t __re_long_size_t; + +#else + +/* Use types that are binary-compatible with the traditional GNU regex + implementation, which mishandles strings longer than INT_MAX. */ + +typedef int regoff_t; +typedef int __re_idx_t; +typedef unsigned int __re_size_t; +typedef unsigned long int __re_long_size_t; + +#endif + +/* The following two types have to be signed and unsigned integer type + wide enough to hold a value of a pointer. For most ANSI compilers + ptrdiff_t and size_t should be likely OK. Still size of these two + types is 2 for Microsoft C. Ugh... */ +typedef long int s_reg_t; +typedef unsigned long int active_reg_t; + +/* The following bits are used to determine the regexp syntax we + recognize. The set/not-set meanings are chosen so that Emacs syntax + remains the value 0. The bits are given in alphabetical order, and + the definitions shifted by one from the previous bit; thus, when we + add or remove a bit, only one other definition need change. */ +typedef unsigned long int reg_syntax_t; + +/* If this bit is not set, then \ inside a bracket expression is literal. + If set, then such a \ quotes the following character. */ +#define REG_BACKSLASH_ESCAPE_IN_LISTS 1ul + +/* If this bit is not set, then + and ? are operators, and \+ and \? are + literals. + If set, then \+ and \? are operators and + and ? are literals. */ +#define REG_BK_PLUS_QM (1ul << 1) + +/* If this bit is set, then character classes are supported. They are: + [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:], + [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:]. + If not set, then character classes are not supported. */ +#define REG_CHAR_CLASSES (1ul << 2) + +/* If this bit is set, then ^ and $ are always anchors (outside bracket + expressions, of course). + If this bit is not set, then it depends: + ^ is an anchor if it is at the beginning of a regular + expression or after an open-group or an alternation operator; + $ is an anchor if it is at the end of a regular expression, or + before a close-group or an alternation operator. + + This bit could be (re)combined with REG_CONTEXT_INDEP_OPS, because + POSIX draft 11.2 says that * etc. in leading positions is undefined. + We already implemented a previous draft which made those constructs + invalid, though, so we haven't changed the code back. */ +#define REG_CONTEXT_INDEP_ANCHORS (1ul << 3) + +/* If this bit is set, then special characters are always special + regardless of where they are in the pattern. + If this bit is not set, then special characters are special only in + some contexts; otherwise they are ordinary. Specifically, + * + ? and intervals are only special when not after the beginning, + open-group, or alternation operator. */ +#define REG_CONTEXT_INDEP_OPS (1ul << 4) + +/* If this bit is set, then *, +, ?, and { cannot be first in an re or + immediately after an alternation or begin-group operator. */ +#define REG_CONTEXT_INVALID_OPS (1ul << 5) + +/* If this bit is set, then . matches newline. + If not set, then it doesn't. */ +#define REG_DOT_NEWLINE (1ul << 6) + +/* If this bit is set, then . doesn't match NUL. + If not set, then it does. */ +#define REG_DOT_NOT_NULL (1ul << 7) + +/* If this bit is set, nonmatching lists [^...] do not match newline. + If not set, they do. */ +#define REG_HAT_LISTS_NOT_NEWLINE (1ul << 8) + +/* If this bit is set, either \{...\} or {...} defines an + interval, depending on REG_NO_BK_BRACES. + If not set, \{, \}, {, and } are literals. */ +#define REG_INTERVALS (1ul << 9) + +/* If this bit is set, +, ? and | aren't recognized as operators. + If not set, they are. */ +#define REG_LIMITED_OPS (1ul << 10) + +/* If this bit is set, newline is an alternation operator. + If not set, newline is literal. */ +#define REG_NEWLINE_ALT (1ul << 11) + +/* If this bit is set, then `{...}' defines an interval, and \{ and \} + are literals. + If not set, then `\{...\}' defines an interval. */ +#define REG_NO_BK_BRACES (1ul << 12) + +/* If this bit is set, (...) defines a group, and \( and \) are literals. + If not set, \(...\) defines a group, and ( and ) are literals. */ +#define REG_NO_BK_PARENS (1ul << 13) + +/* If this bit is set, then \<digit> matches <digit>. + If not set, then \<digit> is a back-reference. */ +#define REG_NO_BK_REFS (1ul << 14) + +/* If this bit is set, then | is an alternation operator, and \| is literal. + If not set, then \| is an alternation operator, and | is literal. */ +#define REG_NO_BK_VBAR (1ul << 15) + +/* If this bit is set, then an ending range point collating higher + than the starting range point, as in [z-a], is invalid. + If not set, the containing range is empty and does not match any string. */ +#define REG_NO_EMPTY_RANGES (1ul << 16) + +/* If this bit is set, then an unmatched ) is ordinary. + If not set, then an unmatched ) is invalid. */ +#define REG_UNMATCHED_RIGHT_PAREN_ORD (1ul << 17) + +/* If this bit is set, succeed as soon as we match the whole pattern, + without further backtracking. */ +#define REG_NO_POSIX_BACKTRACKING (1ul << 18) + +/* If this bit is set, do not process the GNU regex operators. + If not set, then the GNU regex operators are recognized. */ +#define REG_NO_GNU_OPS (1ul << 19) + +/* If this bit is set, turn on internal regex debugging. + If not set, and debugging was on, turn it off. + This only works if regex.c is compiled -DDEBUG. + We define this bit always, so that all that's needed to turn on + debugging is to recompile regex.c; the calling code can always have + this bit set, and it won't affect anything in the normal case. */ +#define REG_DEBUG (1ul << 20) + +/* If this bit is set, a syntactically invalid interval is treated as + a string of ordinary characters. For example, the ERE 'a{1' is + treated as 'a\{1'. */ +#define REG_INVALID_INTERVAL_ORD (1ul << 21) + +/* If this bit is set, then ignore case when matching. + If not set, then case is significant. */ +#define REG_IGNORE_CASE (1ul << 22) + +/* This bit is used internally like REG_CONTEXT_INDEP_ANCHORS but only + for ^, because it is difficult to scan the regex backwards to find + whether ^ should be special. */ +#define REG_CARET_ANCHORS_HERE (1ul << 23) + +/* If this bit is set, then \{ cannot be first in an bre or + immediately after an alternation or begin-group operator. */ +#define REG_CONTEXT_INVALID_DUP (1ul << 24) + +/* If this bit is set, then no_sub will be set to 1 during + re_compile_pattern. */ +#define REG_NO_SUB (1ul << 25) + +/* This global variable defines the particular regexp syntax to use (for + some interfaces). When a regexp is compiled, the syntax used is + stored in the pattern buffer, so changing this does not affect + already-compiled regexps. */ +extern reg_syntax_t re_syntax_options; + +/* Define combinations of the above bits for the standard possibilities. + (The [[[ comments delimit what gets put into the Texinfo file, so + don't delete them!) */ +/* [[[begin syntaxes]]] */ +#define REG_SYNTAX_EMACS 0 + +#define REG_SYNTAX_AWK \ + (REG_BACKSLASH_ESCAPE_IN_LISTS | REG_DOT_NOT_NULL \ + | REG_NO_BK_PARENS | REG_NO_BK_REFS \ + | REG_NO_BK_VBAR | REG_NO_EMPTY_RANGES \ + | REG_DOT_NEWLINE | REG_CONTEXT_INDEP_ANCHORS \ + | REG_UNMATCHED_RIGHT_PAREN_ORD | REG_NO_GNU_OPS) + +#define REG_SYNTAX_GNU_AWK \ + ((REG_SYNTAX_POSIX_EXTENDED | REG_BACKSLASH_ESCAPE_IN_LISTS \ + | REG_DEBUG) \ + & ~(REG_DOT_NOT_NULL | REG_INTERVALS | REG_CONTEXT_INDEP_OPS \ + | REG_CONTEXT_INVALID_OPS )) + +#define REG_SYNTAX_POSIX_AWK \ + (REG_SYNTAX_POSIX_EXTENDED | REG_BACKSLASH_ESCAPE_IN_LISTS \ + | REG_INTERVALS | REG_NO_GNU_OPS) + +#define REG_SYNTAX_GREP \ + (REG_BK_PLUS_QM | REG_CHAR_CLASSES \ + | REG_HAT_LISTS_NOT_NEWLINE | REG_INTERVALS \ + | REG_NEWLINE_ALT) + +#define REG_SYNTAX_EGREP \ + (REG_CHAR_CLASSES | REG_CONTEXT_INDEP_ANCHORS \ + | REG_CONTEXT_INDEP_OPS | REG_HAT_LISTS_NOT_NEWLINE \ + | REG_NEWLINE_ALT | REG_NO_BK_PARENS \ + | REG_NO_BK_VBAR) + +#define REG_SYNTAX_POSIX_EGREP \ + (REG_SYNTAX_EGREP | REG_INTERVALS | REG_NO_BK_BRACES \ + | REG_INVALID_INTERVAL_ORD) + +/* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */ +#define REG_SYNTAX_ED REG_SYNTAX_POSIX_BASIC + +#define REG_SYNTAX_SED REG_SYNTAX_POSIX_BASIC + +/* Syntax bits common to both basic and extended POSIX regex syntax. */ +#define _REG_SYNTAX_POSIX_COMMON \ + (REG_CHAR_CLASSES | REG_DOT_NEWLINE | REG_DOT_NOT_NULL \ + | REG_INTERVALS | REG_NO_EMPTY_RANGES) + +#define REG_SYNTAX_POSIX_BASIC \ + (_REG_SYNTAX_POSIX_COMMON | REG_BK_PLUS_QM | REG_CONTEXT_INVALID_DUP) + +/* Differs from ..._POSIX_BASIC only in that REG_BK_PLUS_QM becomes + REG_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this + isn't minimal, since other operators, such as \`, aren't disabled. */ +#define REG_SYNTAX_POSIX_MINIMAL_BASIC \ + (_REG_SYNTAX_POSIX_COMMON | REG_LIMITED_OPS) + +#define REG_SYNTAX_POSIX_EXTENDED \ + (_REG_SYNTAX_POSIX_COMMON | REG_CONTEXT_INDEP_ANCHORS \ + | REG_CONTEXT_INDEP_OPS | REG_NO_BK_BRACES \ + | REG_NO_BK_PARENS | REG_NO_BK_VBAR \ + | REG_CONTEXT_INVALID_OPS | REG_UNMATCHED_RIGHT_PAREN_ORD) + +/* Differs from ..._POSIX_EXTENDED in that REG_CONTEXT_INDEP_OPS is + removed and REG_NO_BK_REFS is added. */ +#define REG_SYNTAX_POSIX_MINIMAL_EXTENDED \ + (_REG_SYNTAX_POSIX_COMMON | REG_CONTEXT_INDEP_ANCHORS \ + | REG_CONTEXT_INVALID_OPS | REG_NO_BK_BRACES \ + | REG_NO_BK_PARENS | REG_NO_BK_REFS \ + | REG_NO_BK_VBAR | REG_UNMATCHED_RIGHT_PAREN_ORD) +/* [[[end syntaxes]]] */ + +/* Maximum number of duplicates an interval can allow. This is + distinct from RE_DUP_MAX, to conform to POSIX name space rules and + to avoid collisions with <limits.h>. */ +#define REG_DUP_MAX 32767 + + +/* POSIX `cflags' bits (i.e., information for `regcomp'). */ + +/* If this bit is set, then use extended regular expression syntax. + If not set, then use basic regular expression syntax. */ +#define REG_EXTENDED 1 + +/* If this bit is set, then ignore case when matching. + If not set, then case is significant. */ +#define REG_ICASE (1 << 1) + +/* If this bit is set, then anchors do not match at newline + characters in the string. + If not set, then anchors do match at newlines. */ +#define REG_NEWLINE (1 << 2) + +/* If this bit is set, then report only success or fail in regexec. + If not set, then returns differ between not matching and errors. */ +#define REG_NOSUB (1 << 3) + + +/* POSIX `eflags' bits (i.e., information for regexec). */ + +/* If this bit is set, then the beginning-of-line operator doesn't match + the beginning of the string (presumably because it's not the + beginning of a line). + If not set, then the beginning-of-line operator does match the + beginning of the string. */ +#define REG_NOTBOL 1 + +/* Like REG_NOTBOL, except for the end-of-line. */ +#define REG_NOTEOL (1 << 1) + +/* Use PMATCH[0] to delimit the start and end of the search in the + buffer. */ +#define REG_STARTEND (1 << 2) + + +/* If any error codes are removed, changed, or added, update the + `__re_error_msgid' table in regcomp.c. */ + +typedef enum +{ + _REG_ENOSYS = -1, /* This will never happen for this implementation. */ +#define REG_ENOSYS _REG_ENOSYS + + _REG_NOERROR, /* Success. */ +#define REG_NOERROR _REG_NOERROR + + _REG_NOMATCH, /* Didn't find a match (for regexec). */ +#define REG_NOMATCH _REG_NOMATCH + + /* POSIX regcomp return error codes. (In the order listed in the + standard.) */ + + _REG_BADPAT, /* Invalid pattern. */ +#define REG_BADPAT _REG_BADPAT + + _REG_ECOLLATE, /* Inalid collating element. */ +#define REG_ECOLLATE _REG_ECOLLATE + + _REG_ECTYPE, /* Invalid character class name. */ +#define REG_ECTYPE _REG_ECTYPE + + _REG_EESCAPE, /* Trailing backslash. */ +#define REG_EESCAPE _REG_EESCAPE + + _REG_ESUBREG, /* Invalid back reference. */ +#define REG_ESUBREG _REG_ESUBREG + + _REG_EBRACK, /* Unmatched left bracket. */ +#define REG_EBRACK _REG_EBRACK + + _REG_EPAREN, /* Parenthesis imbalance. */ +#define REG_EPAREN _REG_EPAREN + + _REG_EBRACE, /* Unmatched \{. */ +#define REG_EBRACE _REG_EBRACE + + _REG_BADBR, /* Invalid contents of \{\}. */ +#define REG_BADBR _REG_BADBR + + _REG_ERANGE, /* Invalid range end. */ +#define REG_ERANGE _REG_ERANGE + + _REG_ESPACE, /* Ran out of memory. */ +#define REG_ESPACE _REG_ESPACE + + _REG_BADRPT, /* No preceding re for repetition op. */ +#define REG_BADRPT _REG_BADRPT + + /* Error codes we've added. */ + + _REG_EEND, /* Premature end. */ +#define REG_EEND _REG_EEND + + _REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */ +#define REG_ESIZE _REG_ESIZE + + _REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */ +#define REG_ERPAREN _REG_ERPAREN + +} reg_errcode_t; + +/* In the traditional GNU implementation, regex.h defined member names + like `buffer' that POSIX does not allow. These members now have + names with leading `re_' (e.g., `re_buffer'). Support the old + names only if _REGEX_SOURCE is defined. New programs should use + the new names. */ +#ifdef _REGEX_SOURCE +# define _REG_RE_NAME(id) id +# define _REG_RM_NAME(id) id +#else +# define _REG_RE_NAME(id) re_##id +# define _REG_RM_NAME(id) rm_##id +#endif + +/* The user can specify the type of the re_translate member by + defining the macro REG_TRANSLATE_TYPE. In the traditional GNU + implementation, this macro was named RE_TRANSLATE_TYPE, but POSIX + does not allow this. Support the old name only if _REGEX_SOURCE + and if the new name is not defined. New programs should use the new + name. */ +#ifndef REG_TRANSLATE_TYPE +# if defined _REGEX_SOURCE && defined RE_TRANSLATE_TYPE +# define REG_TRANSLATE_TYPE RE_TRANSLATE_TYPE +# else +# define REG_TRANSLATE_TYPE char * +# endif +#endif + +/* This data structure represents a compiled pattern. Before calling + the pattern compiler), the fields `re_buffer', `re_allocated', `re_fastmap', + `re_translate', and `re_no_sub' can be set. After the pattern has been + compiled, the `re_nsub' field is available. All other fields are + private to the regex routines. */ + +struct re_pattern_buffer +{ +/* [[[begin pattern_buffer]]] */ + /* Space that holds the compiled pattern. It is declared as + `unsigned char *' because its elements are + sometimes used as array indexes. */ + unsigned char *_REG_RE_NAME (buffer); + + /* Number of bytes to which `re_buffer' points. */ + __re_long_size_t _REG_RE_NAME (allocated); + + /* Number of bytes actually used in `re_buffer'. */ + __re_long_size_t _REG_RE_NAME (used); + + /* Syntax setting with which the pattern was compiled. */ + reg_syntax_t _REG_RE_NAME (syntax); + + /* Pointer to a fastmap, if any, otherwise zero. re_search uses + the fastmap, if there is one, to skip over impossible + starting points for matches. */ + char *_REG_RE_NAME (fastmap); + + /* Either a translate table to apply to all characters before + comparing them, or zero for no translation. The translation + is applied to a pattern when it is compiled and to a string + when it is matched. */ + REG_TRANSLATE_TYPE _REG_RE_NAME (translate); + + /* Number of subexpressions found by the compiler. */ + size_t re_nsub; + + /* Zero if this pattern cannot match the empty string, one else. + Well, in truth it's used only in `re_search_2', to see + whether or not we should use the fastmap, so we don't set + this absolutely perfectly; see `re_compile_fastmap' (the + `duplicate' case). */ + unsigned int _REG_RE_NAME (can_be_null) : 1; + + /* If REG_UNALLOCATED, allocate space in the `regs' structure + for `max (REG_NREGS, re_nsub + 1)' groups. + If REG_REALLOCATE, reallocate space if necessary. + If REG_FIXED, use what's there. */ +#define REG_UNALLOCATED 0 +#define REG_REALLOCATE 1 +#define REG_FIXED 2 + unsigned int _REG_RE_NAME (regs_allocated) : 2; + + /* Set to zero when `regex_compile' compiles a pattern; set to one + by `re_compile_fastmap' if it updates the fastmap. */ + unsigned int _REG_RE_NAME (fastmap_accurate) : 1; + + /* If set, `re_match_2' does not return information about + subexpressions. */ + unsigned int _REG_RE_NAME (no_sub) : 1; + + /* If set, a beginning-of-line anchor doesn't match at the + beginning of the string. */ + unsigned int _REG_RE_NAME (not_bol) : 1; + + /* Similarly for an end-of-line anchor. */ + unsigned int _REG_RE_NAME (not_eol) : 1; + + /* If true, an anchor at a newline matches. */ + unsigned int _REG_RE_NAME (newline_anchor) : 1; + +/* [[[end pattern_buffer]]] */ +}; + +typedef struct re_pattern_buffer regex_t; + +/* This is the structure we store register match data in. See + regex.texinfo for a full description of what registers match. */ +struct re_registers +{ + __re_size_t _REG_RM_NAME (num_regs); + regoff_t *_REG_RM_NAME (start); + regoff_t *_REG_RM_NAME (end); +}; + + +/* If `regs_allocated' is REG_UNALLOCATED in the pattern buffer, + `re_match_2' returns information about at least this many registers + the first time a `regs' structure is passed. */ +#ifndef REG_NREGS +# define REG_NREGS 30 +#endif + + +/* POSIX specification for registers. Aside from the different names than + `re_registers', POSIX uses an array of structures, instead of a + structure of arrays. */ +typedef struct +{ + regoff_t rm_so; /* Byte offset from string's start to substring's start. */ + regoff_t rm_eo; /* Byte offset from string's start to substring's end. */ +} regmatch_t; + +/* Declarations for routines. */ + +/* Sets the current default syntax to SYNTAX, and return the old syntax. + You can also simply assign to the `re_syntax_options' variable. */ +extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax); + +/* Compile the regular expression PATTERN, with length LENGTH + and syntax given by the global `re_syntax_options', into the buffer + BUFFER. Return NULL if successful, and an error string if not. */ +extern const char *re_compile_pattern (const char *__pattern, size_t __length, + struct re_pattern_buffer *__buffer); + + +/* Compile a fastmap for the compiled pattern in BUFFER; used to + accelerate searches. Return 0 if successful and -2 if was an + internal error. */ +extern int re_compile_fastmap (struct re_pattern_buffer *__buffer); + + +/* Search in the string STRING (with length LENGTH) for the pattern + compiled into BUFFER. Start searching at position START, for RANGE + characters. Return the starting position of the match, -1 for no + match, or -2 for an internal error. Also return register + information in REGS (if REGS and BUFFER->re_no_sub are nonzero). */ +extern regoff_t re_search (struct re_pattern_buffer *__buffer, + const char *__string, __re_idx_t __length, + __re_idx_t __start, regoff_t __range, + struct re_registers *__regs); + + +/* Like `re_search', but search in the concatenation of STRING1 and + STRING2. Also, stop searching at index START + STOP. */ +extern regoff_t re_search_2 (struct re_pattern_buffer *__buffer, + const char *__string1, __re_idx_t __length1, + const char *__string2, __re_idx_t __length2, + __re_idx_t __start, regoff_t __range, + struct re_registers *__regs, + __re_idx_t __stop); + + +/* Like `re_search', but return how many characters in STRING the regexp + in BUFFER matched, starting at position START. */ +extern regoff_t re_match (struct re_pattern_buffer *__buffer, + const char *__string, __re_idx_t __length, + __re_idx_t __start, struct re_registers *__regs); + + +/* Relates to `re_match' as `re_search_2' relates to `re_search'. */ +extern regoff_t re_match_2 (struct re_pattern_buffer *__buffer, + const char *__string1, __re_idx_t __length1, + const char *__string2, __re_idx_t __length2, + __re_idx_t __start, struct re_registers *__regs, + __re_idx_t __stop); + + +/* Set REGS to hold NUM_REGS registers, storing them in STARTS and + ENDS. Subsequent matches using BUFFER and REGS will use this memory + for recording register information. STARTS and ENDS must be + allocated with malloc, and must each be at least `NUM_REGS * sizeof + (regoff_t)' bytes long. + + If NUM_REGS == 0, then subsequent matches should allocate their own + register data. + + Unless this function is called, the first search or match using + PATTERN_BUFFER will allocate its own register data, without + freeing the old data. */ +extern void re_set_registers (struct re_pattern_buffer *__buffer, + struct re_registers *__regs, + __re_size_t __num_regs, + regoff_t *__starts, regoff_t *__ends); + +#if defined _REGEX_RE_COMP || defined _LIBC +# ifndef _CRAY +/* 4.2 bsd compatibility. */ +extern char *re_comp (const char *); +extern int re_exec (const char *); +# endif +#endif + +/* GCC 2.95 and later have "__restrict"; C99 compilers have + "restrict", and "configure" may have defined "restrict". */ +#ifndef __restrict +# if ! (2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__)) +# if defined restrict || 199901L <= __STDC_VERSION__ +# define __restrict restrict +# else +# define __restrict +# endif +# endif +#endif +/* gcc 3.1 and up support the [restrict] syntax, but g++ doesn't. */ +#ifndef __restrict_arr +# if (__GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1)) && !defined __cplusplus +# define __restrict_arr __restrict +# else +# define __restrict_arr +# endif +#endif + +/* POSIX compatibility. */ +extern int regcomp (regex_t *__restrict __preg, + const char *__restrict __pattern, + int __cflags); + +extern int regexec (const regex_t *__restrict __preg, + const char *__restrict __string, size_t __nmatch, + regmatch_t __pmatch[__restrict_arr], + int __eflags); + +extern size_t regerror (int __errcode, const regex_t *__restrict __preg, + char *__restrict __errbuf, size_t __errbuf_size); + +extern void regfree (regex_t *__preg); + + +#ifdef _REGEX_SOURCE + +/* Define the POSIX-compatible member names in terms of the + incompatible (and deprecated) names established by _REG_RE_NAME. + New programs should use the re_* names. */ + +# define re_allocated allocated +# define re_buffer buffer +# define re_can_be_null can_be_null +# define re_fastmap fastmap +# define re_fastmap_accurate fastmap_accurate +# define re_newline_anchor newline_anchor +# define re_no_sub no_sub +# define re_not_bol not_bol +# define re_not_eol not_eol +# define re_regs_allocated regs_allocated +# define re_syntax syntax +# define re_translate translate +# define re_used used + +/* Similarly for _REG_RM_NAME. */ + +# define rm_end end +# define rm_num_regs num_regs +# define rm_start start + +/* Undef RE_DUP_MAX first, in case the user has already included a + <limits.h> with an incompatible definition. + + On GNU systems, the most common spelling for RE_DUP_MAX's value in + <limits.h> is (0x7ffff), so define RE_DUP_MAX to that, not to + REG_DUP_MAX. This avoid some duplicate-macro-definition warnings + with programs that include <limits.h> after this file. + + New programs should not assume that regex.h defines RE_DUP_MAX; to + get the value of RE_DUP_MAX, they should instead include <limits.h> + and possibly invoke the sysconf function. */ + +# undef RE_DUP_MAX +# define RE_DUP_MAX (0x7fff) + +/* Define the following symbols for backward source compatibility. + These symbols violate the POSIX name space rules, and new programs + should avoid them. */ + +# define REGS_FIXED REG_FIXED +# define REGS_REALLOCATE REG_REALLOCATE +# define REGS_UNALLOCATED REG_UNALLOCATED +# define RE_BACKSLASH_ESCAPE_IN_LISTS REG_BACKSLASH_ESCAPE_IN_LISTS +# define RE_BK_PLUS_QM REG_BK_PLUS_QM +# define RE_CARET_ANCHORS_HERE REG_CARET_ANCHORS_HERE +# define RE_CHAR_CLASSES REG_CHAR_CLASSES +# define RE_CONTEXT_INDEP_ANCHORS REG_CONTEXT_INDEP_ANCHORS +# define RE_CONTEXT_INDEP_OPS REG_CONTEXT_INDEP_OPS +# define RE_CONTEXT_INVALID_DUP REG_CONTEXT_INVALID_DUP +# define RE_CONTEXT_INVALID_OPS REG_CONTEXT_INVALID_OPS +# define RE_DEBUG REG_DEBUG +# define RE_DOT_NEWLINE REG_DOT_NEWLINE +# define RE_DOT_NOT_NULL REG_DOT_NOT_NULL +# define RE_HAT_LISTS_NOT_NEWLINE REG_HAT_LISTS_NOT_NEWLINE +# define RE_ICASE REG_IGNORE_CASE /* avoid collision with REG_ICASE */ +# define RE_INTERVALS REG_INTERVALS +# define RE_INVALID_INTERVAL_ORD REG_INVALID_INTERVAL_ORD +# define RE_LIMITED_OPS REG_LIMITED_OPS +# define RE_NEWLINE_ALT REG_NEWLINE_ALT +# define RE_NO_BK_BRACES REG_NO_BK_BRACES +# define RE_NO_BK_PARENS REG_NO_BK_PARENS +# define RE_NO_BK_REFS REG_NO_BK_REFS +# define RE_NO_BK_VBAR REG_NO_BK_VBAR +# define RE_NO_EMPTY_RANGES REG_NO_EMPTY_RANGES +# define RE_NO_GNU_OPS REG_NO_GNU_OPS +# define RE_NO_POSIX_BACKTRACKING REG_NO_POSIX_BACKTRACKING +# define RE_NO_SUB REG_NO_SUB +# define RE_NREGS REG_NREGS +# define RE_SYNTAX_AWK REG_SYNTAX_AWK +# define RE_SYNTAX_ED REG_SYNTAX_ED +# define RE_SYNTAX_EGREP REG_SYNTAX_EGREP +# define RE_SYNTAX_EMACS REG_SYNTAX_EMACS +# define RE_SYNTAX_GNU_AWK REG_SYNTAX_GNU_AWK +# define RE_SYNTAX_GREP REG_SYNTAX_GREP +# define RE_SYNTAX_POSIX_AWK REG_SYNTAX_POSIX_AWK +# define RE_SYNTAX_POSIX_BASIC REG_SYNTAX_POSIX_BASIC +# define RE_SYNTAX_POSIX_EGREP REG_SYNTAX_POSIX_EGREP +# define RE_SYNTAX_POSIX_EXTENDED REG_SYNTAX_POSIX_EXTENDED +# define RE_SYNTAX_POSIX_MINIMAL_BASIC REG_SYNTAX_POSIX_MINIMAL_BASIC +# define RE_SYNTAX_POSIX_MINIMAL_EXTENDED REG_SYNTAX_POSIX_MINIMAL_EXTENDED +# define RE_SYNTAX_SED REG_SYNTAX_SED +# define RE_UNMATCHED_RIGHT_PAREN_ORD REG_UNMATCHED_RIGHT_PAREN_ORD +# ifndef RE_TRANSLATE_TYPE +# define RE_TRANSLATE_TYPE REG_TRANSLATE_TYPE +# endif + +#endif /* defined _REGEX_SOURCE */ + +#ifdef __cplusplus +} +#endif /* C++ */ + +#endif /* regex.h */ + +/* +Local variables: +make-backup-files: t +version-control: t +trim-versions-without-asking: nil +End: +*/ diff --git a/lib/regex_internal.c b/lib/regex_internal.c new file mode 100644 index 0000000..ad618cf --- /dev/null +++ b/lib/regex_internal.c @@ -0,0 +1,1656 @@ +/* Extended regular expression matching and search library. + Copyright (C) 2002, 2003, 2004, 2005 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +static void re_string_construct_common (const char *str, Idx len, + re_string_t *pstr, + REG_TRANSLATE_TYPE trans, bool icase, + const re_dfa_t *dfa) internal_function; +static re_dfastate_t *create_ci_newstate (const re_dfa_t *dfa, + const re_node_set *nodes, + re_hashval_t hash) internal_function; +static re_dfastate_t *create_cd_newstate (const re_dfa_t *dfa, + const re_node_set *nodes, + unsigned int context, + re_hashval_t hash) internal_function; + +/* Functions for string operation. */ + +/* This function allocate the buffers. It is necessary to call + re_string_reconstruct before using the object. */ + +static reg_errcode_t +internal_function +re_string_allocate (re_string_t *pstr, const char *str, Idx len, Idx init_len, + REG_TRANSLATE_TYPE trans, bool icase, const re_dfa_t *dfa) +{ + reg_errcode_t ret; + Idx init_buf_len; + + /* Ensure at least one character fits into the buffers. */ + if (init_len < dfa->mb_cur_max) + init_len = dfa->mb_cur_max; + init_buf_len = (len + 1 < init_len) ? len + 1: init_len; + re_string_construct_common (str, len, pstr, trans, icase, dfa); + + ret = re_string_realloc_buffers (pstr, init_buf_len); + if (BE (ret != REG_NOERROR, 0)) + return ret; + + pstr->word_char = dfa->word_char; + pstr->word_ops_used = dfa->word_ops_used; + pstr->mbs = pstr->mbs_allocated ? pstr->mbs : (unsigned char *) str; + pstr->valid_len = (pstr->mbs_allocated || dfa->mb_cur_max > 1) ? 0 : len; + pstr->valid_raw_len = pstr->valid_len; + return REG_NOERROR; +} + +/* This function allocate the buffers, and initialize them. */ + +static reg_errcode_t +internal_function +re_string_construct (re_string_t *pstr, const char *str, Idx len, + REG_TRANSLATE_TYPE trans, bool icase, const re_dfa_t *dfa) +{ + reg_errcode_t ret; + memset (pstr, '\0', sizeof (re_string_t)); + re_string_construct_common (str, len, pstr, trans, icase, dfa); + + if (len > 0) + { + ret = re_string_realloc_buffers (pstr, len + 1); + if (BE (ret != REG_NOERROR, 0)) + return ret; + } + pstr->mbs = pstr->mbs_allocated ? pstr->mbs : (unsigned char *) str; + + if (icase) + { +#ifdef RE_ENABLE_I18N + if (dfa->mb_cur_max > 1) + { + while (1) + { + ret = build_wcs_upper_buffer (pstr); + if (BE (ret != REG_NOERROR, 0)) + return ret; + if (pstr->valid_raw_len >= len) + break; + if (pstr->bufs_len > pstr->valid_len + dfa->mb_cur_max) + break; + ret = re_string_realloc_buffers (pstr, pstr->bufs_len * 2); + if (BE (ret != REG_NOERROR, 0)) + return ret; + } + } + else +#endif /* RE_ENABLE_I18N */ + build_upper_buffer (pstr); + } + else + { +#ifdef RE_ENABLE_I18N + if (dfa->mb_cur_max > 1) + build_wcs_buffer (pstr); + else +#endif /* RE_ENABLE_I18N */ + { + if (trans != NULL) + re_string_translate_buffer (pstr); + else + { + pstr->valid_len = pstr->bufs_len; + pstr->valid_raw_len = pstr->bufs_len; + } + } + } + + return REG_NOERROR; +} + +/* Helper functions for re_string_allocate, and re_string_construct. */ + +static reg_errcode_t +internal_function +re_string_realloc_buffers (re_string_t *pstr, Idx new_buf_len) +{ +#ifdef RE_ENABLE_I18N + if (pstr->mb_cur_max > 1) + { + wint_t *new_wcs = re_xrealloc (pstr->wcs, wint_t, new_buf_len); + if (BE (new_wcs == NULL, 0)) + return REG_ESPACE; + pstr->wcs = new_wcs; + if (pstr->offsets != NULL) + { + Idx *new_offsets = re_xrealloc (pstr->offsets, Idx, new_buf_len); + if (BE (new_offsets == NULL, 0)) + return REG_ESPACE; + pstr->offsets = new_offsets; + } + } +#endif /* RE_ENABLE_I18N */ + if (pstr->mbs_allocated) + { + unsigned char *new_mbs = re_realloc (pstr->mbs, unsigned char, + new_buf_len); + if (BE (new_mbs == NULL, 0)) + return REG_ESPACE; + pstr->mbs = new_mbs; + } + pstr->bufs_len = new_buf_len; + return REG_NOERROR; +} + + +static void +internal_function +re_string_construct_common (const char *str, Idx len, re_string_t *pstr, + REG_TRANSLATE_TYPE trans, bool icase, + const re_dfa_t *dfa) +{ + pstr->raw_mbs = (const unsigned char *) str; + pstr->len = len; + pstr->raw_len = len; + pstr->trans = (unsigned REG_TRANSLATE_TYPE) trans; + pstr->icase = icase; + pstr->mbs_allocated = (trans != NULL || icase); + pstr->mb_cur_max = dfa->mb_cur_max; + pstr->is_utf8 = dfa->is_utf8; + pstr->map_notascii = dfa->map_notascii; + pstr->stop = pstr->len; + pstr->raw_stop = pstr->stop; +} + +#ifdef RE_ENABLE_I18N + +/* Build wide character buffer PSTR->WCS. + If the byte sequence of the string are: + <mb1>(0), <mb1>(1), <mb2>(0), <mb2>(1), <sb3> + Then wide character buffer will be: + <wc1> , WEOF , <wc2> , WEOF , <wc3> + We use WEOF for padding, they indicate that the position isn't + a first byte of a multibyte character. + + Note that this function assumes PSTR->VALID_LEN elements are already + built and starts from PSTR->VALID_LEN. */ + +static void +internal_function +build_wcs_buffer (re_string_t *pstr) +{ +#ifdef _LIBC + unsigned char buf[MB_LEN_MAX]; + assert (MB_LEN_MAX >= pstr->mb_cur_max); +#else + unsigned char buf[64]; +#endif + mbstate_t prev_st; + Idx byte_idx, end_idx, remain_len; + size_t mbclen; + + /* Build the buffers from pstr->valid_len to either pstr->len or + pstr->bufs_len. */ + end_idx = (pstr->bufs_len > pstr->len) ? pstr->len : pstr->bufs_len; + for (byte_idx = pstr->valid_len; byte_idx < end_idx;) + { + wchar_t wc; + const char *p; + + remain_len = end_idx - byte_idx; + prev_st = pstr->cur_state; + /* Apply the translation if we need. */ + if (BE (pstr->trans != NULL, 0)) + { + int i, ch; + + for (i = 0; i < pstr->mb_cur_max && i < remain_len; ++i) + { + ch = pstr->raw_mbs [pstr->raw_mbs_idx + byte_idx + i]; + buf[i] = pstr->mbs[byte_idx + i] = pstr->trans[ch]; + } + p = (const char *) buf; + } + else + p = (const char *) pstr->raw_mbs + pstr->raw_mbs_idx + byte_idx; + mbclen = mbrtowc (&wc, p, remain_len, &pstr->cur_state); + if (BE (mbclen == (size_t) -2, 0)) + { + /* The buffer doesn't have enough space, finish to build. */ + pstr->cur_state = prev_st; + break; + } + else if (BE (mbclen == (size_t) -1 || mbclen == 0, 0)) + { + /* We treat these cases as a singlebyte character. */ + mbclen = 1; + wc = (wchar_t) pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx]; + if (BE (pstr->trans != NULL, 0)) + wc = pstr->trans[wc]; + pstr->cur_state = prev_st; + } + + /* Write wide character and padding. */ + pstr->wcs[byte_idx++] = wc; + /* Write paddings. */ + for (remain_len = byte_idx + mbclen - 1; byte_idx < remain_len ;) + pstr->wcs[byte_idx++] = WEOF; + } + pstr->valid_len = byte_idx; + pstr->valid_raw_len = byte_idx; +} + +/* Build wide character buffer PSTR->WCS like build_wcs_buffer, + but for REG_ICASE. */ + +static reg_errcode_t +internal_function +build_wcs_upper_buffer (re_string_t *pstr) +{ + mbstate_t prev_st; + Idx src_idx, byte_idx, end_idx, remain_len; + size_t mbclen; +#ifdef _LIBC + char buf[MB_LEN_MAX]; + assert (MB_LEN_MAX >= pstr->mb_cur_max); +#else + char buf[64]; +#endif + + byte_idx = pstr->valid_len; + end_idx = (pstr->bufs_len > pstr->len) ? pstr->len : pstr->bufs_len; + + /* The following optimization assumes that ASCII characters can be + mapped to wide characters with a simple cast. */ + if (! pstr->map_notascii && pstr->trans == NULL && !pstr->offsets_needed) + { + while (byte_idx < end_idx) + { + wchar_t wc; + + if (isascii (pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx]) + && mbsinit (&pstr->cur_state)) + { + /* In case of a singlebyte character. */ + pstr->mbs[byte_idx] + = toupper (pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx]); + /* The next step uses the assumption that wchar_t is encoded + ASCII-safe: all ASCII values can be converted like this. */ + pstr->wcs[byte_idx] = (wchar_t) pstr->mbs[byte_idx]; + ++byte_idx; + continue; + } + + remain_len = end_idx - byte_idx; + prev_st = pstr->cur_state; + mbclen = mbrtowc (&wc, + ((const char *) pstr->raw_mbs + pstr->raw_mbs_idx + + byte_idx), remain_len, &pstr->cur_state); + if (BE ((size_t) (mbclen + 2) > 2, 1)) + { + wchar_t wcu = wc; + if (iswlower (wc)) + { + size_t mbcdlen; + + wcu = towupper (wc); + mbcdlen = wcrtomb (buf, wcu, &prev_st); + if (BE (mbclen == mbcdlen, 1)) + memcpy (pstr->mbs + byte_idx, buf, mbclen); + else + { + src_idx = byte_idx; + goto offsets_needed; + } + } + else + memcpy (pstr->mbs + byte_idx, + pstr->raw_mbs + pstr->raw_mbs_idx + byte_idx, mbclen); + pstr->wcs[byte_idx++] = wcu; + /* Write paddings. */ + for (remain_len = byte_idx + mbclen - 1; byte_idx < remain_len ;) + pstr->wcs[byte_idx++] = WEOF; + } + else if (mbclen == (size_t) -1 || mbclen == 0) + { + /* It is an invalid character or '\0'. Just use the byte. */ + int ch = pstr->raw_mbs[pstr->raw_mbs_idx + byte_idx]; + pstr->mbs[byte_idx] = ch; + /* And also cast it to wide char. */ + pstr->wcs[byte_idx++] = (wchar_t) ch; + if (BE (mbclen == (size_t) -1, 0)) + pstr->cur_state = prev_st; + } + else + { + /* The buffer doesn't have enough space, finish to build. */ + pstr->cur_state = prev_st; + break; + } + } + pstr->valid_len = byte_idx; + pstr->valid_raw_len = byte_idx; + return REG_NOERROR; + } + else + for (src_idx = pstr->valid_raw_len; byte_idx < end_idx;) + { + wchar_t wc; + const char *p; + offsets_needed: + remain_len = end_idx - byte_idx; + prev_st = pstr->cur_state; + if (BE (pstr->trans != NULL, 0)) + { + int i, ch; + + for (i = 0; i < pstr->mb_cur_max && i < remain_len; ++i) + { + ch = pstr->raw_mbs [pstr->raw_mbs_idx + src_idx + i]; + buf[i] = pstr->trans[ch]; + } + p = (const char *) buf; + } + else + p = (const char *) pstr->raw_mbs + pstr->raw_mbs_idx + src_idx; + mbclen = mbrtowc (&wc, p, remain_len, &pstr->cur_state); + if (BE ((size_t) (mbclen + 2) > 2, 1)) + { + wchar_t wcu = wc; + if (iswlower (wc)) + { + size_t mbcdlen; + + wcu = towupper (wc); + mbcdlen = wcrtomb ((char *) buf, wcu, &prev_st); + if (BE (mbclen == mbcdlen, 1)) + memcpy (pstr->mbs + byte_idx, buf, mbclen); + else if (mbcdlen != (size_t) -1) + { + size_t i; + + if (byte_idx + mbcdlen > pstr->bufs_len) + { + pstr->cur_state = prev_st; + break; + } + + if (pstr->offsets == NULL) + { + pstr->offsets = re_xmalloc (Idx, pstr->bufs_len); + + if (pstr->offsets == NULL) + return REG_ESPACE; + } + if (!pstr->offsets_needed) + { + for (i = 0; i < (size_t) byte_idx; ++i) + pstr->offsets[i] = i; + pstr->offsets_needed = 1; + } + + memcpy (pstr->mbs + byte_idx, buf, mbcdlen); + pstr->wcs[byte_idx] = wcu; + pstr->offsets[byte_idx] = src_idx; + for (i = 1; i < mbcdlen; ++i) + { + pstr->offsets[byte_idx + i] + = src_idx + (i < mbclen ? i : mbclen - 1); + pstr->wcs[byte_idx + i] = WEOF; + } + pstr->len += mbcdlen - mbclen; + if (pstr->raw_stop > src_idx) + pstr->stop += mbcdlen - mbclen; + end_idx = (pstr->bufs_len > pstr->len) + ? pstr->len : pstr->bufs_len; + byte_idx += mbcdlen; + src_idx += mbclen; + continue; + } + else + memcpy (pstr->mbs + byte_idx, p, mbclen); + } + else + memcpy (pstr->mbs + byte_idx, p, mbclen); + + if (BE (pstr->offsets_needed != 0, 0)) + { + size_t i; + for (i = 0; i < mbclen; ++i) + pstr->offsets[byte_idx + i] = src_idx + i; + } + src_idx += mbclen; + + pstr->wcs[byte_idx++] = wcu; + /* Write paddings. */ + for (remain_len = byte_idx + mbclen - 1; byte_idx < remain_len ;) + pstr->wcs[byte_idx++] = WEOF; + } + else if (mbclen == (size_t) -1 || mbclen == 0) + { + /* It is an invalid character or '\0'. Just use the byte. */ + int ch = pstr->raw_mbs[pstr->raw_mbs_idx + src_idx]; + + if (BE (pstr->trans != NULL, 0)) + ch = pstr->trans [ch]; + pstr->mbs[byte_idx] = ch; + + if (BE (pstr->offsets_needed != 0, 0)) + pstr->offsets[byte_idx] = src_idx; + ++src_idx; + + /* And also cast it to wide char. */ + pstr->wcs[byte_idx++] = (wchar_t) ch; + if (BE (mbclen == (size_t) -1, 0)) + pstr->cur_state = prev_st; + } + else + { + /* The buffer doesn't have enough space, finish to build. */ + pstr->cur_state = prev_st; + break; + } + } + pstr->valid_len = byte_idx; + pstr->valid_raw_len = src_idx; + return REG_NOERROR; +} + +/* Skip characters until the index becomes greater than NEW_RAW_IDX. + Return the index. */ + +static Idx +internal_function +re_string_skip_chars (re_string_t *pstr, Idx new_raw_idx, wint_t *last_wc) +{ + mbstate_t prev_st; + Idx rawbuf_idx; + size_t mbclen; + wchar_t wc = 0; + + /* Skip the characters which are not necessary to check. */ + for (rawbuf_idx = pstr->raw_mbs_idx + pstr->valid_raw_len; + rawbuf_idx < new_raw_idx;) + { + Idx remain_len; + remain_len = pstr->len - rawbuf_idx; + prev_st = pstr->cur_state; + mbclen = mbrtowc (&wc, (const char *) pstr->raw_mbs + rawbuf_idx, + remain_len, &pstr->cur_state); + if (BE (mbclen == (size_t) -2 || mbclen == (size_t) -1 || mbclen == 0, 0)) + { + /* We treat these cases as a singlebyte character. */ + mbclen = 1; + pstr->cur_state = prev_st; + } + /* Then proceed the next character. */ + rawbuf_idx += mbclen; + } + *last_wc = (wint_t) wc; + return rawbuf_idx; +} +#endif /* RE_ENABLE_I18N */ + +/* Build the buffer PSTR->MBS, and apply the translation if we need. + This function is used in case of REG_ICASE. */ + +static void +internal_function +build_upper_buffer (re_string_t *pstr) +{ + Idx char_idx, end_idx; + end_idx = (pstr->bufs_len > pstr->len) ? pstr->len : pstr->bufs_len; + + for (char_idx = pstr->valid_len; char_idx < end_idx; ++char_idx) + { + int ch = pstr->raw_mbs[pstr->raw_mbs_idx + char_idx]; + if (BE (pstr->trans != NULL, 0)) + ch = pstr->trans[ch]; + if (islower (ch)) + pstr->mbs[char_idx] = toupper (ch); + else + pstr->mbs[char_idx] = ch; + } + pstr->valid_len = char_idx; + pstr->valid_raw_len = char_idx; +} + +/* Apply TRANS to the buffer in PSTR. */ + +static void +internal_function +re_string_translate_buffer (re_string_t *pstr) +{ + Idx buf_idx, end_idx; + end_idx = (pstr->bufs_len > pstr->len) ? pstr->len : pstr->bufs_len; + + for (buf_idx = pstr->valid_len; buf_idx < end_idx; ++buf_idx) + { + int ch = pstr->raw_mbs[pstr->raw_mbs_idx + buf_idx]; + pstr->mbs[buf_idx] = pstr->trans[ch]; + } + + pstr->valid_len = buf_idx; + pstr->valid_raw_len = buf_idx; +} + +/* This function re-construct the buffers. + Concretely, convert to wide character in case of pstr->mb_cur_max > 1, + convert to upper case in case of REG_ICASE, apply translation. */ + +static reg_errcode_t +internal_function +re_string_reconstruct (re_string_t *pstr, Idx idx, int eflags) +{ + Idx offset; + + if (BE (pstr->raw_mbs_idx <= idx, 0)) + offset = idx - pstr->raw_mbs_idx; + else + { + /* Reset buffer. */ +#ifdef RE_ENABLE_I18N + if (pstr->mb_cur_max > 1) + memset (&pstr->cur_state, '\0', sizeof (mbstate_t)); +#endif /* RE_ENABLE_I18N */ + pstr->len = pstr->raw_len; + pstr->stop = pstr->raw_stop; + pstr->valid_len = 0; + pstr->raw_mbs_idx = 0; + pstr->valid_raw_len = 0; + pstr->offsets_needed = 0; + pstr->tip_context = ((eflags & REG_NOTBOL) ? CONTEXT_BEGBUF + : CONTEXT_NEWLINE | CONTEXT_BEGBUF); + if (!pstr->mbs_allocated) + pstr->mbs = (unsigned char *) pstr->raw_mbs; + offset = idx; + } + + if (BE (offset != 0, 1)) + { + /* Are the characters which are already checked remain? */ + if (BE (offset < pstr->valid_raw_len, 1) +#ifdef RE_ENABLE_I18N + /* Handling this would enlarge the code too much. + Accept a slowdown in that case. */ + && pstr->offsets_needed == 0 +#endif + ) + { + /* Yes, move them to the front of the buffer. */ + pstr->tip_context = re_string_context_at (pstr, offset - 1, eflags); +#ifdef RE_ENABLE_I18N + if (pstr->mb_cur_max > 1) + memmove (pstr->wcs, pstr->wcs + offset, + (pstr->valid_len - offset) * sizeof (wint_t)); +#endif /* RE_ENABLE_I18N */ + if (BE (pstr->mbs_allocated, 0)) + memmove (pstr->mbs, pstr->mbs + offset, + pstr->valid_len - offset); + pstr->valid_len -= offset; + pstr->valid_raw_len -= offset; +#if DEBUG + assert (pstr->valid_len > 0); +#endif + } + else + { + /* No, skip all characters until IDX. */ +#ifdef RE_ENABLE_I18N + if (BE (pstr->offsets_needed, 0)) + { + pstr->len = pstr->raw_len - idx + offset; + pstr->stop = pstr->raw_stop - idx + offset; + pstr->offsets_needed = 0; + } +#endif + pstr->valid_len = 0; + pstr->valid_raw_len = 0; +#ifdef RE_ENABLE_I18N + if (pstr->mb_cur_max > 1) + { + Idx wcs_idx; + wint_t wc = WEOF; + + if (pstr->is_utf8) + { + const unsigned char *raw, *p, *q, *end; + + /* Special case UTF-8. Multi-byte chars start with any + byte other than 0x80 - 0xbf. */ + raw = pstr->raw_mbs + pstr->raw_mbs_idx; + end = raw + (offset - pstr->mb_cur_max); + for (p = raw + offset - 1; p >= end; --p) + if ((*p & 0xc0) != 0x80) + { + mbstate_t cur_state; + wchar_t wc2; + Idx mlen = raw + pstr->len - p; + unsigned char buf[6]; + size_t mbclen; + + q = p; + if (BE (pstr->trans != NULL, 0)) + { + int i = mlen < 6 ? mlen : 6; + while (--i >= 0) + buf[i] = pstr->trans[p[i]]; + q = buf; + } + /* XXX Don't use mbrtowc, we know which conversion + to use (UTF-8 -> UCS4). */ + memset (&cur_state, 0, sizeof (cur_state)); + mbclen = mbrtowc (&wc2, (const char *) p, mlen, + &cur_state); + if (raw + offset - p <= mbclen && mbclen < (size_t) -2) + { + memset (&pstr->cur_state, '\0', + sizeof (mbstate_t)); + pstr->valid_len = mbclen - (raw + offset - p); + wc = wc2; + } + break; + } + } + + if (wc == WEOF) + pstr->valid_len = re_string_skip_chars (pstr, idx, &wc) - idx; + if (BE (pstr->valid_len, 0)) + { + for (wcs_idx = 0; wcs_idx < pstr->valid_len; ++wcs_idx) + pstr->wcs[wcs_idx] = WEOF; + if (pstr->mbs_allocated) + memset (pstr->mbs, -1, pstr->valid_len); + } + pstr->valid_raw_len = pstr->valid_len; + pstr->tip_context = ((BE (pstr->word_ops_used != 0, 0) + && IS_WIDE_WORD_CHAR (wc)) + ? CONTEXT_WORD + : ((IS_WIDE_NEWLINE (wc) + && pstr->newline_anchor) + ? CONTEXT_NEWLINE : 0)); + } + else +#endif /* RE_ENABLE_I18N */ + { + int c = pstr->raw_mbs[pstr->raw_mbs_idx + offset - 1]; + if (pstr->trans) + c = pstr->trans[c]; + pstr->tip_context = (bitset_contain (pstr->word_char, c) + ? CONTEXT_WORD + : ((IS_NEWLINE (c) && pstr->newline_anchor) + ? CONTEXT_NEWLINE : 0)); + } + } + if (!BE (pstr->mbs_allocated, 0)) + pstr->mbs += offset; + } + pstr->raw_mbs_idx = idx; + pstr->len -= offset; + pstr->stop -= offset; + + /* Then build the buffers. */ +#ifdef RE_ENABLE_I18N + if (pstr->mb_cur_max > 1) + { + if (pstr->icase) + { + reg_errcode_t ret = build_wcs_upper_buffer (pstr); + if (BE (ret != REG_NOERROR, 0)) + return ret; + } + else + build_wcs_buffer (pstr); + } + else +#endif /* RE_ENABLE_I18N */ + if (BE (pstr->mbs_allocated, 0)) + { + if (pstr->icase) + build_upper_buffer (pstr); + else if (pstr->trans != NULL) + re_string_translate_buffer (pstr); + } + else + pstr->valid_len = pstr->len; + + pstr->cur_idx = 0; + return REG_NOERROR; +} + +static unsigned char +internal_function __attribute ((pure)) +re_string_peek_byte_case (const re_string_t *pstr, Idx idx) +{ + int ch; + Idx off; + + /* Handle the common (easiest) cases first. */ + if (BE (!pstr->mbs_allocated, 1)) + return re_string_peek_byte (pstr, idx); + +#ifdef RE_ENABLE_I18N + if (pstr->mb_cur_max > 1 + && ! re_string_is_single_byte_char (pstr, pstr->cur_idx + idx)) + return re_string_peek_byte (pstr, idx); +#endif + + off = pstr->cur_idx + idx; +#ifdef RE_ENABLE_I18N + if (pstr->offsets_needed) + off = pstr->offsets[off]; +#endif + + ch = pstr->raw_mbs[pstr->raw_mbs_idx + off]; + +#ifdef RE_ENABLE_I18N + /* Ensure that e.g. for tr_TR.UTF-8 BACKSLASH DOTLESS SMALL LETTER I + this function returns CAPITAL LETTER I instead of first byte of + DOTLESS SMALL LETTER I. The latter would confuse the parser, + since peek_byte_case doesn't advance cur_idx in any way. */ + if (pstr->offsets_needed && !isascii (ch)) + return re_string_peek_byte (pstr, idx); +#endif + + return ch; +} + +static unsigned char +internal_function __attribute ((pure)) +re_string_fetch_byte_case (re_string_t *pstr) +{ + if (BE (!pstr->mbs_allocated, 1)) + return re_string_fetch_byte (pstr); + +#ifdef RE_ENABLE_I18N + if (pstr->offsets_needed) + { + Idx off; + int ch; + + /* For tr_TR.UTF-8 [[:islower:]] there is + [[: CAPITAL LETTER I WITH DOT lower:]] in mbs. Skip + in that case the whole multi-byte character and return + the original letter. On the other side, with + [[: DOTLESS SMALL LETTER I return [[:I, as doing + anything else would complicate things too much. */ + + if (!re_string_first_byte (pstr, pstr->cur_idx)) + return re_string_fetch_byte (pstr); + + off = pstr->offsets[pstr->cur_idx]; + ch = pstr->raw_mbs[pstr->raw_mbs_idx + off]; + + if (! isascii (ch)) + return re_string_fetch_byte (pstr); + + re_string_skip_bytes (pstr, + re_string_char_size_at (pstr, pstr->cur_idx)); + return ch; + } +#endif + + return pstr->raw_mbs[pstr->raw_mbs_idx + pstr->cur_idx++]; +} + +static void +internal_function +re_string_destruct (re_string_t *pstr) +{ +#ifdef RE_ENABLE_I18N + re_free (pstr->wcs); + re_free (pstr->offsets); +#endif /* RE_ENABLE_I18N */ + if (pstr->mbs_allocated) + re_free (pstr->mbs); +} + +/* Return the context at IDX in INPUT. */ + +static unsigned int +internal_function +re_string_context_at (const re_string_t *input, Idx idx, int eflags) +{ + int c; + if (BE (! REG_VALID_INDEX (idx), 0)) + /* In this case, we use the value stored in input->tip_context, + since we can't know the character in input->mbs[-1] here. */ + return input->tip_context; + if (BE (idx == input->len, 0)) + return ((eflags & REG_NOTEOL) ? CONTEXT_ENDBUF + : CONTEXT_NEWLINE | CONTEXT_ENDBUF); +#ifdef RE_ENABLE_I18N + if (input->mb_cur_max > 1) + { + wint_t wc; + Idx wc_idx = idx; + while(input->wcs[wc_idx] == WEOF) + { +#ifdef DEBUG + /* It must not happen. */ + assert (REG_VALID_INDEX (wc_idx)); +#endif + --wc_idx; + if (! REG_VALID_INDEX (wc_idx)) + return input->tip_context; + } + wc = input->wcs[wc_idx]; + if (BE (input->word_ops_used != 0, 0) && IS_WIDE_WORD_CHAR (wc)) + return CONTEXT_WORD; + return (IS_WIDE_NEWLINE (wc) && input->newline_anchor + ? CONTEXT_NEWLINE : 0); + } + else +#endif + { + c = re_string_byte_at (input, idx); + if (bitset_contain (input->word_char, c)) + return CONTEXT_WORD; + return IS_NEWLINE (c) && input->newline_anchor ? CONTEXT_NEWLINE : 0; + } +} + +/* Functions for set operation. */ + +static reg_errcode_t +internal_function +re_node_set_alloc (re_node_set *set, Idx size) +{ + set->alloc = size; + set->nelem = 0; + set->elems = re_xmalloc (Idx, size); + if (BE (set->elems == NULL, 0)) + return REG_ESPACE; + return REG_NOERROR; +} + +static reg_errcode_t +internal_function +re_node_set_init_1 (re_node_set *set, Idx elem) +{ + set->alloc = 1; + set->nelem = 1; + set->elems = re_malloc (Idx, 1); + if (BE (set->elems == NULL, 0)) + { + set->alloc = set->nelem = 0; + return REG_ESPACE; + } + set->elems[0] = elem; + return REG_NOERROR; +} + +static reg_errcode_t +internal_function +re_node_set_init_2 (re_node_set *set, Idx elem1, Idx elem2) +{ + set->alloc = 2; + set->elems = re_malloc (Idx, 2); + if (BE (set->elems == NULL, 0)) + return REG_ESPACE; + if (elem1 == elem2) + { + set->nelem = 1; + set->elems[0] = elem1; + } + else + { + set->nelem = 2; + if (elem1 < elem2) + { + set->elems[0] = elem1; + set->elems[1] = elem2; + } + else + { + set->elems[0] = elem2; + set->elems[1] = elem1; + } + } + return REG_NOERROR; +} + +static reg_errcode_t +internal_function +re_node_set_init_copy (re_node_set *dest, const re_node_set *src) +{ + dest->nelem = src->nelem; + if (src->nelem > 0) + { + dest->alloc = dest->nelem; + dest->elems = re_malloc (Idx, dest->alloc); + if (BE (dest->elems == NULL, 0)) + { + dest->alloc = dest->nelem = 0; + return REG_ESPACE; + } + memcpy (dest->elems, src->elems, src->nelem * sizeof dest->elems[0]); + } + else + re_node_set_init_empty (dest); + return REG_NOERROR; +} + +/* Calculate the intersection of the sets SRC1 and SRC2. And merge it to + DEST. Return value indicate the error code or REG_NOERROR if succeeded. + Note: We assume dest->elems is NULL, when dest->alloc is 0. */ + +static reg_errcode_t +internal_function +re_node_set_add_intersect (re_node_set *dest, const re_node_set *src1, + const re_node_set *src2) +{ + Idx i1, i2, is, id, delta, sbase; + if (src1->nelem == 0 || src2->nelem == 0) + return REG_NOERROR; + + /* We need dest->nelem + 2 * elems_in_intersection; this is a + conservative estimate. */ + if (src1->nelem + src2->nelem + dest->nelem > dest->alloc) + { + Idx new_alloc = src1->nelem + src2->nelem + dest->alloc; + Idx *new_elems; + if (sizeof (Idx) < 3 + && (new_alloc < dest->alloc + || ((Idx) (src1->nelem + src2->nelem) < src1->nelem))) + return REG_ESPACE; + new_elems = re_xrealloc (dest->elems, Idx, new_alloc); + if (BE (new_elems == NULL, 0)) + return REG_ESPACE; + dest->elems = new_elems; + dest->alloc = new_alloc; + } + + /* Find the items in the intersection of SRC1 and SRC2, and copy + into the top of DEST those that are not already in DEST itself. */ + sbase = dest->nelem + src1->nelem + src2->nelem; + i1 = src1->nelem - 1; + i2 = src2->nelem - 1; + id = dest->nelem - 1; + for (;;) + { + if (src1->elems[i1] == src2->elems[i2]) + { + /* Try to find the item in DEST. Maybe we could binary search? */ + while (REG_VALID_INDEX (id) && dest->elems[id] > src1->elems[i1]) + --id; + + if (! REG_VALID_INDEX (id) || dest->elems[id] != src1->elems[i1]) + dest->elems[--sbase] = src1->elems[i1]; + + if (! REG_VALID_INDEX (--i1) || ! REG_VALID_INDEX (--i2)) + break; + } + + /* Lower the highest of the two items. */ + else if (src1->elems[i1] < src2->elems[i2]) + { + if (! REG_VALID_INDEX (--i2)) + break; + } + else + { + if (! REG_VALID_INDEX (--i1)) + break; + } + } + + id = dest->nelem - 1; + is = dest->nelem + src1->nelem + src2->nelem - 1; + delta = is - sbase + 1; + + /* Now copy. When DELTA becomes zero, the remaining + DEST elements are already in place; this is more or + less the same loop that is in re_node_set_merge. */ + dest->nelem += delta; + if (delta > 0 && REG_VALID_INDEX (id)) + for (;;) + { + if (dest->elems[is] > dest->elems[id]) + { + /* Copy from the top. */ + dest->elems[id + delta--] = dest->elems[is--]; + if (delta == 0) + break; + } + else + { + /* Slide from the bottom. */ + dest->elems[id + delta] = dest->elems[id]; + if (! REG_VALID_INDEX (--id)) + break; + } + } + + /* Copy remaining SRC elements. */ + memcpy (dest->elems, dest->elems + sbase, delta * sizeof dest->elems[0]); + + return REG_NOERROR; +} + +/* Calculate the union set of the sets SRC1 and SRC2. And store it to + DEST. Return value indicate the error code or REG_NOERROR if succeeded. */ + +static reg_errcode_t +internal_function +re_node_set_init_union (re_node_set *dest, const re_node_set *src1, + const re_node_set *src2) +{ + Idx i1, i2, id; + if (src1 != NULL && src1->nelem > 0 && src2 != NULL && src2->nelem > 0) + { + dest->alloc = src1->nelem + src2->nelem; + if (sizeof (Idx) < 2 && dest->alloc < src1->nelem) + return REG_ESPACE; + dest->elems = re_xmalloc (Idx, dest->alloc); + if (BE (dest->elems == NULL, 0)) + return REG_ESPACE; + } + else + { + if (src1 != NULL && src1->nelem > 0) + return re_node_set_init_copy (dest, src1); + else if (src2 != NULL && src2->nelem > 0) + return re_node_set_init_copy (dest, src2); + else + re_node_set_init_empty (dest); + return REG_NOERROR; + } + for (i1 = i2 = id = 0 ; i1 < src1->nelem && i2 < src2->nelem ;) + { + if (src1->elems[i1] > src2->elems[i2]) + { + dest->elems[id++] = src2->elems[i2++]; + continue; + } + if (src1->elems[i1] == src2->elems[i2]) + ++i2; + dest->elems[id++] = src1->elems[i1++]; + } + if (i1 < src1->nelem) + { + memcpy (dest->elems + id, src1->elems + i1, + (src1->nelem - i1) * sizeof dest->elems[0]); + id += src1->nelem - i1; + } + else if (i2 < src2->nelem) + { + memcpy (dest->elems + id, src2->elems + i2, + (src2->nelem - i2) * sizeof dest->elems[0]); + id += src2->nelem - i2; + } + dest->nelem = id; + return REG_NOERROR; +} + +/* Calculate the union set of the sets DEST and SRC. And store it to + DEST. Return value indicate the error code or REG_NOERROR if succeeded. */ + +static reg_errcode_t +internal_function +re_node_set_merge (re_node_set *dest, const re_node_set *src) +{ + Idx is, id, sbase, delta; + if (src == NULL || src->nelem == 0) + return REG_NOERROR; + if (sizeof (Idx) < 3 + && ((Idx) (2 * src->nelem) < src->nelem + || (Idx) (2 * src->nelem + dest->nelem) < dest->nelem)) + return REG_ESPACE; + if (dest->alloc < 2 * src->nelem + dest->nelem) + { + Idx new_alloc = src->nelem + dest->alloc; + Idx *new_buffer; + if (sizeof (Idx) < 4 && new_alloc < dest->alloc) + return REG_ESPACE; + new_buffer = re_x2realloc (dest->elems, Idx, &new_alloc); + if (BE (new_buffer == NULL, 0)) + return REG_ESPACE; + dest->elems = new_buffer; + dest->alloc = new_alloc; + } + + if (BE (dest->nelem == 0, 0)) + { + dest->nelem = src->nelem; + memcpy (dest->elems, src->elems, src->nelem * sizeof dest->elems[0]); + return REG_NOERROR; + } + + /* Copy into the top of DEST the items of SRC that are not + found in DEST. Maybe we could binary search in DEST? */ + for (sbase = dest->nelem + 2 * src->nelem, + is = src->nelem - 1, id = dest->nelem - 1; + REG_VALID_INDEX (is) && REG_VALID_INDEX (id); ) + { + if (dest->elems[id] == src->elems[is]) + is--, id--; + else if (dest->elems[id] < src->elems[is]) + dest->elems[--sbase] = src->elems[is--]; + else /* if (dest->elems[id] > src->elems[is]) */ + --id; + } + + if (REG_VALID_INDEX (is)) + { + /* If DEST is exhausted, the remaining items of SRC must be unique. */ + sbase -= is + 1; + memcpy (dest->elems + sbase, src->elems, + (is + 1) * sizeof dest->elems[0]); + } + + id = dest->nelem - 1; + is = dest->nelem + 2 * src->nelem - 1; + delta = is - sbase + 1; + if (delta == 0) + return REG_NOERROR; + + /* Now copy. When DELTA becomes zero, the remaining + DEST elements are already in place. */ + dest->nelem += delta; + for (;;) + { + if (dest->elems[is] > dest->elems[id]) + { + /* Copy from the top. */ + dest->elems[id + delta--] = dest->elems[is--]; + if (delta == 0) + break; + } + else + { + /* Slide from the bottom. */ + dest->elems[id + delta] = dest->elems[id]; + if (! REG_VALID_INDEX (--id)) + { + /* Copy remaining SRC elements. */ + memcpy (dest->elems, dest->elems + sbase, + delta * sizeof dest->elems[0]); + break; + } + } + } + + return REG_NOERROR; +} + +/* Insert the new element ELEM to the re_node_set* SET. + SET should not already have ELEM. + Return true if successful. */ + +static bool +internal_function +re_node_set_insert (re_node_set *set, Idx elem) +{ + Idx idx; + /* In case the set is empty. */ + if (set->alloc == 0) + return re_node_set_init_1 (set, elem) == REG_NOERROR; + + if (BE (set->nelem, 0) == 0) + { + /* We already guaranteed above that set->alloc != 0. */ + set->elems[0] = elem; + ++set->nelem; + return true; + } + + /* Realloc if we need. */ + if (set->alloc == set->nelem) + { + Idx *new_elems = re_x2realloc (set->elems, Idx, &set->alloc); + if (BE (new_elems == NULL, 0)) + return false; + set->elems = new_elems; + } + + /* Move the elements which follows the new element. Test the + first element separately to skip a check in the inner loop. */ + if (elem < set->elems[0]) + { + idx = 0; + for (idx = set->nelem; idx > 0; idx--) + set->elems[idx] = set->elems[idx - 1]; + } + else + { + for (idx = set->nelem; set->elems[idx - 1] > elem; idx--) + set->elems[idx] = set->elems[idx - 1]; + } + + /* Insert the new element. */ + set->elems[idx] = elem; + ++set->nelem; + return true; +} + +/* Insert the new element ELEM to the re_node_set* SET. + SET should not already have any element greater than or equal to ELEM. + Return true if successful. */ + +static bool +internal_function +re_node_set_insert_last (re_node_set *set, Idx elem) +{ + /* Realloc if we need. */ + if (set->alloc == set->nelem) + { + Idx *new_elems; + new_elems = re_x2realloc (set->elems, Idx, &set->alloc); + if (BE (new_elems == NULL, 0)) + return false; + set->elems = new_elems; + } + + /* Insert the new element. */ + set->elems[set->nelem++] = elem; + return true; +} + +/* Compare two node sets SET1 and SET2. + Return true if SET1 and SET2 are equivalent. */ + +static bool +internal_function __attribute ((pure)) +re_node_set_compare (const re_node_set *set1, const re_node_set *set2) +{ + Idx i; + if (set1 == NULL || set2 == NULL || set1->nelem != set2->nelem) + return false; + for (i = set1->nelem ; REG_VALID_INDEX (--i) ; ) + if (set1->elems[i] != set2->elems[i]) + return false; + return true; +} + +/* Return (idx + 1) if SET contains the element ELEM, return 0 otherwise. */ + +static Idx +internal_function __attribute ((pure)) +re_node_set_contains (const re_node_set *set, Idx elem) +{ + __re_size_t idx, right, mid; + if (! REG_VALID_NONZERO_INDEX (set->nelem)) + return 0; + + /* Binary search the element. */ + idx = 0; + right = set->nelem - 1; + while (idx < right) + { + mid = (idx + right) / 2; + if (set->elems[mid] < elem) + idx = mid + 1; + else + right = mid; + } + return set->elems[idx] == elem ? idx + 1 : 0; +} + +static void +internal_function +re_node_set_remove_at (re_node_set *set, Idx idx) +{ + if (idx < 0 || idx >= set->nelem) + return; + --set->nelem; + for (; idx < set->nelem; idx++) + set->elems[idx] = set->elems[idx + 1]; +} + + +/* Add the token TOKEN to dfa->nodes, and return the index of the token. + Or return REG_MISSING if an error occurred. */ + +static Idx +internal_function +re_dfa_add_node (re_dfa_t *dfa, re_token_t token) +{ + int type = token.type; + if (BE (dfa->nodes_len >= dfa->nodes_alloc, 0)) + { + Idx new_nodes_alloc = dfa->nodes_alloc; + Idx *new_nexts, *new_indices; + re_node_set *new_edests, *new_eclosures; + + re_token_t *new_nodes = re_x2realloc (dfa->nodes, re_token_t, + &new_nodes_alloc); + if (BE (new_nodes == NULL, 0)) + return REG_MISSING; + dfa->nodes = new_nodes; + new_nexts = re_realloc (dfa->nexts, Idx, new_nodes_alloc); + new_indices = re_realloc (dfa->org_indices, Idx, new_nodes_alloc); + new_edests = re_xrealloc (dfa->edests, re_node_set, new_nodes_alloc); + new_eclosures = re_realloc (dfa->eclosures, re_node_set, new_nodes_alloc); + if (BE (new_nexts == NULL || new_indices == NULL + || new_edests == NULL || new_eclosures == NULL, 0)) + return REG_MISSING; + dfa->nexts = new_nexts; + dfa->org_indices = new_indices; + dfa->edests = new_edests; + dfa->eclosures = new_eclosures; + dfa->nodes_alloc = new_nodes_alloc; + } + dfa->nodes[dfa->nodes_len] = token; + dfa->nodes[dfa->nodes_len].constraint = 0; +#ifdef RE_ENABLE_I18N + dfa->nodes[dfa->nodes_len].accept_mb = + (type == OP_PERIOD && dfa->mb_cur_max > 1) || type == COMPLEX_BRACKET; +#endif + dfa->nexts[dfa->nodes_len] = REG_MISSING; + re_node_set_init_empty (dfa->edests + dfa->nodes_len); + re_node_set_init_empty (dfa->eclosures + dfa->nodes_len); + return dfa->nodes_len++; +} + +static inline re_hashval_t +internal_function +calc_state_hash (const re_node_set *nodes, unsigned int context) +{ + re_hashval_t hash = nodes->nelem + context; + Idx i; + for (i = 0 ; i < nodes->nelem ; i++) + hash += nodes->elems[i]; + return hash; +} + +/* Search for the state whose node_set is equivalent to NODES. + Return the pointer to the state, if we found it in the DFA. + Otherwise create the new one and return it. In case of an error + return NULL and set the error code in ERR. + Note: - We assume NULL as the invalid state, then it is possible that + return value is NULL and ERR is REG_NOERROR. + - We never return non-NULL value in case of any errors, it is for + optimization. */ + +static re_dfastate_t* +internal_function +re_acquire_state (reg_errcode_t *err, re_dfa_t *dfa, const re_node_set *nodes) +{ + re_hashval_t hash; + re_dfastate_t *new_state; + struct re_state_table_entry *spot; + Idx i; +#ifdef lint + /* Suppress bogus uninitialized-variable warnings. */ + *err = REG_NOERROR; +#endif + if (BE (nodes->nelem == 0, 0)) + { + *err = REG_NOERROR; + return NULL; + } + hash = calc_state_hash (nodes, 0); + spot = dfa->state_table + (hash & dfa->state_hash_mask); + + for (i = 0 ; i < spot->num ; i++) + { + re_dfastate_t *state = spot->array[i]; + if (hash != state->hash) + continue; + if (re_node_set_compare (&state->nodes, nodes)) + return state; + } + + /* There are no appropriate state in the dfa, create the new one. */ + new_state = create_ci_newstate (dfa, nodes, hash); + if (BE (new_state != NULL, 1)) + return new_state; + else + { + *err = REG_ESPACE; + return NULL; + } +} + +/* Search for the state whose node_set is equivalent to NODES and + whose context is equivalent to CONTEXT. + Return the pointer to the state, if we found it in the DFA. + Otherwise create the new one and return it. In case of an error + return NULL and set the error code in ERR. + Note: - We assume NULL as the invalid state, then it is possible that + return value is NULL and ERR is REG_NOERROR. + - We never return non-NULL value in case of any errors, it is for + optimization. */ + +static re_dfastate_t* +internal_function +re_acquire_state_context (reg_errcode_t *err, re_dfa_t *dfa, + const re_node_set *nodes, unsigned int context) +{ + re_hashval_t hash; + re_dfastate_t *new_state; + struct re_state_table_entry *spot; + Idx i; +#ifdef lint + /* Suppress bogus uninitialized-variable warnings. */ + *err = REG_NOERROR; +#endif + if (nodes->nelem == 0) + { + *err = REG_NOERROR; + return NULL; + } + hash = calc_state_hash (nodes, context); + spot = dfa->state_table + (hash & dfa->state_hash_mask); + + for (i = 0 ; i < spot->num ; i++) + { + re_dfastate_t *state = spot->array[i]; + if (state->hash == hash + && state->context == context + && re_node_set_compare (state->entrance_nodes, nodes)) + return state; + } + /* There are no appropriate state in `dfa', create the new one. */ + new_state = create_cd_newstate (dfa, nodes, context, hash); + if (BE (new_state != NULL, 1)) + return new_state; + else + { + *err = REG_ESPACE; + return NULL; + } +} + +/* Finish initialization of the new state NEWSTATE, and using its hash value + HASH put in the appropriate bucket of DFA's state table. Return value + indicates the error code if failed. */ + +static reg_errcode_t +internal_function +register_state (const re_dfa_t *dfa, re_dfastate_t *newstate, re_hashval_t hash) +{ + struct re_state_table_entry *spot; + reg_errcode_t err; + Idx i; + + newstate->hash = hash; + err = re_node_set_alloc (&newstate->non_eps_nodes, newstate->nodes.nelem); + if (BE (err != REG_NOERROR, 0)) + return REG_ESPACE; + for (i = 0; i < newstate->nodes.nelem; i++) + { + Idx elem = newstate->nodes.elems[i]; + if (!IS_EPSILON_NODE (dfa->nodes[elem].type)) + { + bool ok = re_node_set_insert_last (&newstate->non_eps_nodes, elem); + if (BE (! ok, 0)) + return REG_ESPACE; + } + } + + spot = dfa->state_table + (hash & dfa->state_hash_mask); + if (BE (spot->alloc <= spot->num, 0)) + { + Idx new_alloc = spot->num; + re_dfastate_t **new_array = re_x2realloc (spot->array, re_dfastate_t *, + &new_alloc); + if (BE (new_array == NULL, 0)) + return REG_ESPACE; + spot->array = new_array; + spot->alloc = new_alloc; + } + spot->array[spot->num++] = newstate; + return REG_NOERROR; +} + +/* Create the new state which is independ of contexts. + Return the new state if succeeded, otherwise return NULL. */ + +static re_dfastate_t * +internal_function +create_ci_newstate (const re_dfa_t *dfa, const re_node_set *nodes, + re_hashval_t hash) +{ + Idx i; + reg_errcode_t err; + re_dfastate_t *newstate; + + newstate = re_calloc (re_dfastate_t, 1); + if (BE (newstate == NULL, 0)) + return NULL; + err = re_node_set_init_copy (&newstate->nodes, nodes); + if (BE (err != REG_NOERROR, 0)) + { + re_free (newstate); + return NULL; + } + + newstate->entrance_nodes = &newstate->nodes; + for (i = 0 ; i < nodes->nelem ; i++) + { + re_token_t *node = dfa->nodes + nodes->elems[i]; + re_token_type_t type = node->type; + if (type == CHARACTER && !node->constraint) + continue; +#ifdef RE_ENABLE_I18N + newstate->accept_mb |= node->accept_mb; +#endif /* RE_ENABLE_I18N */ + + /* If the state has the halt node, the state is a halt state. */ + if (type == END_OF_RE) + newstate->halt = 1; + else if (type == OP_BACK_REF) + newstate->has_backref = 1; + else if (type == ANCHOR || node->constraint) + newstate->has_constraint = 1; + } + err = register_state (dfa, newstate, hash); + if (BE (err != REG_NOERROR, 0)) + { + free_state (newstate); + newstate = NULL; + } + return newstate; +} + +/* Create the new state which is depend on the context CONTEXT. + Return the new state if succeeded, otherwise return NULL. */ + +static re_dfastate_t * +internal_function +create_cd_newstate (const re_dfa_t *dfa, const re_node_set *nodes, + unsigned int context, re_hashval_t hash) +{ + Idx i, nctx_nodes = 0; + reg_errcode_t err; + re_dfastate_t *newstate; + + newstate = re_calloc (re_dfastate_t, 1); + if (BE (newstate == NULL, 0)) + return NULL; + err = re_node_set_init_copy (&newstate->nodes, nodes); + if (BE (err != REG_NOERROR, 0)) + { + re_free (newstate); + return NULL; + } + + newstate->context = context; + newstate->entrance_nodes = &newstate->nodes; + + for (i = 0 ; i < nodes->nelem ; i++) + { + unsigned int constraint = 0; + re_token_t *node = dfa->nodes + nodes->elems[i]; + re_token_type_t type = node->type; + if (node->constraint) + constraint = node->constraint; + + if (type == CHARACTER && !constraint) + continue; +#ifdef RE_ENABLE_I18N + newstate->accept_mb |= node->accept_mb; +#endif /* RE_ENABLE_I18N */ + + /* If the state has the halt node, the state is a halt state. */ + if (type == END_OF_RE) + newstate->halt = 1; + else if (type == OP_BACK_REF) + newstate->has_backref = 1; + else if (type == ANCHOR) + constraint = node->opr.ctx_type; + + if (constraint) + { + if (newstate->entrance_nodes == &newstate->nodes) + { + newstate->entrance_nodes = re_malloc (re_node_set, 1); + if (BE (newstate->entrance_nodes == NULL, 0)) + { + free_state (newstate); + return NULL; + } + re_node_set_init_copy (newstate->entrance_nodes, nodes); + nctx_nodes = 0; + newstate->has_constraint = 1; + } + + if (NOT_SATISFY_PREV_CONSTRAINT (constraint,context)) + { + re_node_set_remove_at (&newstate->nodes, i - nctx_nodes); + ++nctx_nodes; + } + } + } + err = register_state (dfa, newstate, hash); + if (BE (err != REG_NOERROR, 0)) + { + free_state (newstate); + newstate = NULL; + } + return newstate; +} + +static void +internal_function +free_state (re_dfastate_t *state) +{ + re_node_set_free (&state->non_eps_nodes); + re_node_set_free (&state->inveclosure); + if (state->entrance_nodes != &state->nodes) + { + re_node_set_free (state->entrance_nodes); + re_free (state->entrance_nodes); + } + re_node_set_free (&state->nodes); + re_free (state->word_trtable); + re_free (state->trtable); + re_free (state); +} diff --git a/lib/regex_internal.h b/lib/regex_internal.h new file mode 100644 index 0000000..b94473a --- /dev/null +++ b/lib/regex_internal.h @@ -0,0 +1,907 @@ +/* Extended regular expression matching and search library. + Copyright (C) 2002, 2003, 2004, 2005 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _REGEX_INTERNAL_H +#define _REGEX_INTERNAL_H 1 + +#include <assert.h> +#include <ctype.h> +#include <stdbool.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +#if defined HAVE_LANGINFO_H || defined HAVE_LANGINFO_CODESET || defined _LIBC +# include <langinfo.h> +#endif +#if defined HAVE_LOCALE_H || defined _LIBC +# include <locale.h> +#endif +#if defined HAVE_WCHAR_H || defined _LIBC +# include <wchar.h> +#endif /* HAVE_WCHAR_H || _LIBC */ +#if defined HAVE_WCTYPE_H || defined _LIBC +# include <wctype.h> +#endif /* HAVE_WCTYPE_H || _LIBC */ +#if defined _LIBC +# include <bits/libc-lock.h> +#else +# define __libc_lock_define(CLASS,NAME) +# define __libc_lock_init(NAME) do { } while (0) +# define __libc_lock_lock(NAME) do { } while (0) +# define __libc_lock_unlock(NAME) do { } while (0) +#endif + +/* In case that the system doesn't have isblank(). */ +#if !defined _LIBC && !defined HAVE_ISBLANK && !defined isblank +# define isblank(ch) ((ch) == ' ' || (ch) == '\t') +#endif + +#ifdef _LIBC +# ifndef _RE_DEFINE_LOCALE_FUNCTIONS +# define _RE_DEFINE_LOCALE_FUNCTIONS 1 +# include <locale/localeinfo.h> +# include <locale/elem-hash.h> +# include <locale/coll-lookup.h> +# endif +#endif + +/* This is for other GNU distributions with internationalized messages. */ +#if (HAVE_LIBINTL_H && ENABLE_NLS) || defined _LIBC +# include <libintl.h> +# ifdef _LIBC +# undef gettext +# define gettext(msgid) \ + INTUSE(__dcgettext) (_libc_intl_domainname, msgid, LC_MESSAGES) +# endif +#else +# define gettext(msgid) (msgid) +#endif + +#ifndef gettext_noop +/* This define is so xgettext can find the internationalizable + strings. */ +# define gettext_noop(String) String +#endif + +#if (defined MB_CUR_MAX && HAVE_LOCALE_H && HAVE_WCTYPE_H && HAVE_WCHAR_H && HAVE_WCRTOMB && HAVE_MBRTOWC && HAVE_WCSCOLL) || _LIBC +# define RE_ENABLE_I18N +#endif + +#if __GNUC__ >= 3 +# define BE(expr, val) __builtin_expect (expr, val) +#else +# define BE(expr, val) (expr) +#endif + +/* Number of single byte character. */ +#define SBC_MAX 256 + +#define COLL_ELEM_LEN_MAX 8 + +/* The character which represents newline. */ +#define NEWLINE_CHAR '\n' +#define WIDE_NEWLINE_CHAR L'\n' + +/* Rename to standard API for using out of glibc. */ +#ifndef _LIBC +# define __wctype wctype +# define __iswctype iswctype +# define __btowc btowc +# ifndef __mempcpy +# define __mempcpy mempcpy +# endif +# define __wcrtomb wcrtomb +# define __regfree regfree +# define attribute_hidden +#endif /* not _LIBC */ + +#if __GNUC__ >= 4 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1) +# define __attribute(arg) __attribute__ (arg) +#else +# define __attribute(arg) +#endif + +extern const char __re_error_msgid[] attribute_hidden; +extern const size_t __re_error_msgid_idx[] attribute_hidden; + +typedef __re_idx_t Idx; + +/* Special return value for failure to match. */ +#define REG_MISSING ((Idx) -1) + +/* Special return value for internal error. */ +#define REG_ERROR ((Idx) -2) + +/* Test whether N is a valid index, and is not one of the above. */ +#ifdef _REGEX_LARGE_OFFSETS +# define REG_VALID_INDEX(n) ((Idx) (n) < REG_ERROR) +#else +# define REG_VALID_INDEX(n) (0 <= (n)) +#endif + +/* Test whether N is a valid nonzero index. */ +#ifdef _REGEX_LARGE_OFFSETS +# define REG_VALID_NONZERO_INDEX(n) ((Idx) ((n) - 1) < (Idx) (REG_ERROR - 1)) +#else +# define REG_VALID_NONZERO_INDEX(n) (0 < (n)) +#endif + +/* A hash value, suitable for computing hash tables. */ +typedef __re_size_t re_hashval_t; + +/* An integer used to represent a set of bits. It must be unsigned, + and must be at least as wide as unsigned int. */ +typedef unsigned long int bitset_word; + +/* Maximum value of a bitset word. It must be useful in preprocessor + contexts, and must be consistent with bitset_word. */ +#define BITSET_WORD_MAX ULONG_MAX + +/* Number of bits in a bitset word. Avoid greater-than-32-bit + integers and unconditional shifts by more than 31 bits, as they're + not portable. */ +#if BITSET_WORD_MAX == 0xffffffff +# define BITSET_WORD_BITS 32 +#elif BITSET_WORD_MAX >> 31 >> 5 == 1 +# define BITSET_WORD_BITS 36 +#elif BITSET_WORD_MAX >> 31 >> 16 == 1 +# define BITSET_WORD_BITS 48 +#elif BITSET_WORD_MAX >> 31 >> 28 == 1 +# define BITSET_WORD_BITS 60 +#elif BITSET_WORD_MAX >> 31 >> 31 >> 1 == 1 +# define BITSET_WORD_BITS 64 +#elif BITSET_WORD_MAX >> 31 >> 31 >> 9 == 1 +# define BITSET_WORD_BITS 72 +#elif BITSET_WORD_MAX >> 31 >> 31 >> 31 >> 31 >> 3 == 1 +# define BITSET_WORD_BITS 128 +#elif BITSET_WORD_MAX >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 7 == 1 +# define BITSET_WORD_BITS 256 +#elif BITSET_WORD_MAX >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 7 > 1 +# define BITSET_WORD_BITS 257 /* any value > SBC_MAX will do here */ +# if BITSET_WORD_BITS <= SBC_MAX +# error "Invalid SBC_MAX" +# endif +#else +# error "Add case for new bitset_word size" +#endif + +/* Number of bitset words in a bitset. */ +#define BITSET_WORDS ((SBC_MAX + BITSET_WORD_BITS - 1) / BITSET_WORD_BITS) + +typedef bitset_word bitset[BITSET_WORDS]; +typedef bitset_word *re_bitset_ptr_t; +typedef const bitset_word *re_const_bitset_ptr_t; + +#define PREV_WORD_CONSTRAINT 0x0001 +#define PREV_NOTWORD_CONSTRAINT 0x0002 +#define NEXT_WORD_CONSTRAINT 0x0004 +#define NEXT_NOTWORD_CONSTRAINT 0x0008 +#define PREV_NEWLINE_CONSTRAINT 0x0010 +#define NEXT_NEWLINE_CONSTRAINT 0x0020 +#define PREV_BEGBUF_CONSTRAINT 0x0040 +#define NEXT_ENDBUF_CONSTRAINT 0x0080 +#define WORD_DELIM_CONSTRAINT 0x0100 +#define NOT_WORD_DELIM_CONSTRAINT 0x0200 + +typedef enum +{ + INSIDE_WORD = PREV_WORD_CONSTRAINT | NEXT_WORD_CONSTRAINT, + WORD_FIRST = PREV_NOTWORD_CONSTRAINT | NEXT_WORD_CONSTRAINT, + WORD_LAST = PREV_WORD_CONSTRAINT | NEXT_NOTWORD_CONSTRAINT, + INSIDE_NOTWORD = PREV_NOTWORD_CONSTRAINT | NEXT_NOTWORD_CONSTRAINT, + LINE_FIRST = PREV_NEWLINE_CONSTRAINT, + LINE_LAST = NEXT_NEWLINE_CONSTRAINT, + BUF_FIRST = PREV_BEGBUF_CONSTRAINT, + BUF_LAST = NEXT_ENDBUF_CONSTRAINT, + WORD_DELIM = WORD_DELIM_CONSTRAINT, + NOT_WORD_DELIM = NOT_WORD_DELIM_CONSTRAINT +} re_context_type; + +typedef struct +{ + Idx alloc; + Idx nelem; + Idx *elems; +} re_node_set; + +typedef enum +{ + NON_TYPE = 0, + + /* Node type, These are used by token, node, tree. */ + CHARACTER = 1, + END_OF_RE = 2, + SIMPLE_BRACKET = 3, + OP_BACK_REF = 4, + OP_PERIOD = 5, +#ifdef RE_ENABLE_I18N + COMPLEX_BRACKET = 6, + OP_UTF8_PERIOD = 7, +#endif /* RE_ENABLE_I18N */ + + /* We define EPSILON_BIT as a macro so that OP_OPEN_SUBEXP is used + when the debugger shows values of this enum type. */ +#define EPSILON_BIT 8 + OP_OPEN_SUBEXP = EPSILON_BIT | 0, + OP_CLOSE_SUBEXP = EPSILON_BIT | 1, + OP_ALT = EPSILON_BIT | 2, + OP_DUP_ASTERISK = EPSILON_BIT | 3, + ANCHOR = EPSILON_BIT | 4, + + /* Tree type, these are used only by tree. */ + CONCAT = 16, + SUBEXP = 17, + + /* Token type, these are used only by token. */ + OP_DUP_PLUS = 18, + OP_DUP_QUESTION, + OP_OPEN_BRACKET, + OP_CLOSE_BRACKET, + OP_CHARSET_RANGE, + OP_OPEN_DUP_NUM, + OP_CLOSE_DUP_NUM, + OP_NON_MATCH_LIST, + OP_OPEN_COLL_ELEM, + OP_CLOSE_COLL_ELEM, + OP_OPEN_EQUIV_CLASS, + OP_CLOSE_EQUIV_CLASS, + OP_OPEN_CHAR_CLASS, + OP_CLOSE_CHAR_CLASS, + OP_WORD, + OP_NOTWORD, + OP_SPACE, + OP_NOTSPACE, + BACK_SLASH + +} re_token_type_t; + +#ifdef RE_ENABLE_I18N +typedef struct +{ + /* Multibyte characters. */ + wchar_t *mbchars; + + /* Collating symbols. */ +# ifdef _LIBC + int32_t *coll_syms; +# endif + + /* Equivalence classes. */ +# ifdef _LIBC + int32_t *equiv_classes; +# endif + + /* Range expressions. */ +# ifdef _LIBC + uint32_t *range_starts; + uint32_t *range_ends; +# else /* not _LIBC */ + wchar_t *range_starts; + wchar_t *range_ends; +# endif /* not _LIBC */ + + /* Character classes. */ + wctype_t *char_classes; + + /* If this character set is the non-matching list. */ + unsigned int non_match : 1; + + /* # of multibyte characters. */ + Idx nmbchars; + + /* # of collating symbols. */ + Idx ncoll_syms; + + /* # of equivalence classes. */ + Idx nequiv_classes; + + /* # of range expressions. */ + Idx nranges; + + /* # of character classes. */ + Idx nchar_classes; +} re_charset_t; +#endif /* RE_ENABLE_I18N */ + +typedef struct +{ + union + { + unsigned char c; /* for CHARACTER */ + re_bitset_ptr_t sbcset; /* for SIMPLE_BRACKET */ +#ifdef RE_ENABLE_I18N + re_charset_t *mbcset; /* for COMPLEX_BRACKET */ +#endif /* RE_ENABLE_I18N */ + Idx idx; /* for BACK_REF */ + re_context_type ctx_type; /* for ANCHOR */ + } opr; +#if __GNUC__ >= 2 + re_token_type_t type : 8; +#else + re_token_type_t type; +#endif + unsigned int constraint : 10; /* context constraint */ + unsigned int duplicated : 1; + unsigned int opt_subexp : 1; +#ifdef RE_ENABLE_I18N + unsigned int accept_mb : 1; + /* These 2 bits can be moved into the union if needed (e.g. if running out + of bits; move opr.c to opr.c.c and move the flags to opr.c.flags). */ + unsigned int mb_partial : 1; +#endif + unsigned int word_char : 1; +} re_token_t; + +#define IS_EPSILON_NODE(type) ((type) & EPSILON_BIT) + +struct re_string_t +{ + /* Indicate the raw buffer which is the original string passed as an + argument of regexec(), re_search(), etc.. */ + const unsigned char *raw_mbs; + /* Store the multibyte string. In case of "case insensitive mode" like + REG_ICASE, upper cases of the string are stored, otherwise MBS points + the same address that RAW_MBS points. */ + unsigned char *mbs; +#ifdef RE_ENABLE_I18N + /* Store the wide character string which is corresponding to MBS. */ + wint_t *wcs; + Idx *offsets; + mbstate_t cur_state; +#endif + /* Index in RAW_MBS. Each character mbs[i] corresponds to + raw_mbs[raw_mbs_idx + i]. */ + Idx raw_mbs_idx; + /* The length of the valid characters in the buffers. */ + Idx valid_len; + /* The corresponding number of bytes in raw_mbs array. */ + Idx valid_raw_len; + /* The length of the buffers MBS and WCS. */ + Idx bufs_len; + /* The index in MBS, which is updated by re_string_fetch_byte. */ + Idx cur_idx; + /* length of RAW_MBS array. */ + Idx raw_len; + /* This is RAW_LEN - RAW_MBS_IDX + VALID_LEN - VALID_RAW_LEN. */ + Idx len; + /* End of the buffer may be shorter than its length in the cases such + as re_match_2, re_search_2. Then, we use STOP for end of the buffer + instead of LEN. */ + Idx raw_stop; + /* This is RAW_STOP - RAW_MBS_IDX adjusted through OFFSETS. */ + Idx stop; + + /* The context of mbs[0]. We store the context independently, since + the context of mbs[0] may be different from raw_mbs[0], which is + the beginning of the input string. */ + unsigned int tip_context; + /* The translation passed as a part of an argument of re_compile_pattern. */ + unsigned REG_TRANSLATE_TYPE trans; + /* Copy of re_dfa_t's word_char. */ + re_const_bitset_ptr_t word_char; + /* true if REG_ICASE. */ + unsigned char icase; + unsigned char is_utf8; + unsigned char map_notascii; + unsigned char mbs_allocated; + unsigned char offsets_needed; + unsigned char newline_anchor; + unsigned char word_ops_used; + int mb_cur_max; +}; +typedef struct re_string_t re_string_t; + + +struct re_dfa_t; +typedef struct re_dfa_t re_dfa_t; + +#ifndef _LIBC +# ifdef __i386__ +# define internal_function __attribute ((regparm (3), stdcall)) +# else +# define internal_function +# endif +#endif + +static reg_errcode_t re_string_realloc_buffers (re_string_t *pstr, + Idx new_buf_len) + internal_function; +#ifdef RE_ENABLE_I18N +static void build_wcs_buffer (re_string_t *pstr) internal_function; +static reg_errcode_t build_wcs_upper_buffer (re_string_t *pstr) + internal_function; +#endif /* RE_ENABLE_I18N */ +static void build_upper_buffer (re_string_t *pstr) internal_function; +static void re_string_translate_buffer (re_string_t *pstr) internal_function; +static unsigned int re_string_context_at (const re_string_t *input, + Idx idx, int eflags) + internal_function __attribute ((pure)); + +#define re_string_peek_byte(pstr, offset) \ + ((pstr)->mbs[(pstr)->cur_idx + offset]) +#define re_string_fetch_byte(pstr) \ + ((pstr)->mbs[(pstr)->cur_idx++]) +#define re_string_first_byte(pstr, idx) \ + ((idx) == (pstr)->valid_len || (pstr)->wcs[idx] != WEOF) +#define re_string_is_single_byte_char(pstr, idx) \ + ((pstr)->wcs[idx] != WEOF && ((pstr)->valid_len == (idx) + 1 \ + || (pstr)->wcs[(idx) + 1] != WEOF)) +#define re_string_eoi(pstr) ((pstr)->stop <= (pstr)->cur_idx) +#define re_string_cur_idx(pstr) ((pstr)->cur_idx) +#define re_string_get_buffer(pstr) ((pstr)->mbs) +#define re_string_length(pstr) ((pstr)->len) +#define re_string_byte_at(pstr,idx) ((pstr)->mbs[idx]) +#define re_string_skip_bytes(pstr,idx) ((pstr)->cur_idx += (idx)) +#define re_string_set_index(pstr,idx) ((pstr)->cur_idx = (idx)) + +#include <alloca.h> + +#ifndef _LIBC +# if HAVE_ALLOCA +/* The OS usually guarantees only one guard page at the bottom of the stack, + and a page size can be as small as 4096 bytes. So we cannot safely + allocate anything larger than 4096 bytes. Also care for the possibility + of a few compiler-allocated temporary stack slots. */ +# define __libc_use_alloca(n) ((n) < 4032) +# else +/* alloca is implemented with malloc, so just use malloc. */ +# define __libc_use_alloca(n) 0 +# endif +#endif + +#define re_malloc(t,n) ((t *) malloc ((n) * sizeof (t))) +#define re_xmalloc(t,n) ((t *) re_xnmalloc (n, sizeof (t))) +#define re_calloc(t,n) ((t *) calloc (n, sizeof (t))) +#define re_realloc(p,t,n) ((t *) realloc (p, (n) * sizeof (t))) +#define re_xrealloc(p,t,n) ((t *) re_xnrealloc (p, n, sizeof (t))) +#define re_x2realloc(p,t,pn) ((t *) re_x2nrealloc (p, pn, sizeof (t))) +#define re_free(p) free (p) + +#ifndef SIZE_MAX +# define SIZE_MAX ((size_t) -1) +#endif + +/* Return true if an array of N objects, each of size S, cannot exist + due to size arithmetic overflow. S must be nonzero. */ +static inline bool +re_alloc_oversized (size_t n, size_t s) +{ + return BE (SIZE_MAX / s < n, 0); +} + +/* Return true if an array of (2 * N + 1) objects, each of size S, + cannot exist due to size arithmetic overflow. S must be nonzero. */ +static inline bool +re_x2alloc_oversized (size_t n, size_t s) +{ + return BE ((SIZE_MAX / s - 1) / 2 < n, 0); +} + +/* Allocate an array of N objects, each with S bytes of memory, + dynamically, with error checking. S must be nonzero. */ +static inline void * +re_xnmalloc (size_t n, size_t s) +{ + return re_alloc_oversized (n, s) ? NULL : malloc (n * s); +} + +/* Change the size of an allocated block of memory P to an array of N + objects each of S bytes, with error checking. S must be nonzero. */ +static inline void * +re_xnrealloc (void *p, size_t n, size_t s) +{ + return re_alloc_oversized (n, s) ? NULL : realloc (p, n * s); +} + +/* Reallocate a block of memory P to an array of (2 * (*PN) + 1) + objects each of S bytes, with error checking. S must be nonzero. + If the allocation is successful, set *PN to the new allocation + count and return the resulting pointer. Otherwise, return + NULL. */ +static inline void * +re_x2nrealloc (void *p, size_t *pn, size_t s) +{ + if (re_x2alloc_oversized (*pn, s)) + return NULL; + else + { + /* Add 1 in case *PN is zero. */ + size_t n1 = 2 * *pn + 1; + p = realloc (p, n1 * s); + if (BE (p != NULL, 1)) + *pn = n1; + return p; + } +} + +struct bin_tree_t +{ + struct bin_tree_t *parent; + struct bin_tree_t *left; + struct bin_tree_t *right; + struct bin_tree_t *first; + struct bin_tree_t *next; + + re_token_t token; + + /* `node_idx' is the index in dfa->nodes, if `type' == 0. + Otherwise `type' indicate the type of this node. */ + Idx node_idx; +}; +typedef struct bin_tree_t bin_tree_t; + +#define BIN_TREE_STORAGE_SIZE \ + ((1024 - sizeof (void *)) / sizeof (bin_tree_t)) + +struct bin_tree_storage_t +{ + struct bin_tree_storage_t *next; + bin_tree_t data[BIN_TREE_STORAGE_SIZE]; +}; +typedef struct bin_tree_storage_t bin_tree_storage_t; + +#define CONTEXT_WORD 1 +#define CONTEXT_NEWLINE (CONTEXT_WORD << 1) +#define CONTEXT_BEGBUF (CONTEXT_NEWLINE << 1) +#define CONTEXT_ENDBUF (CONTEXT_BEGBUF << 1) + +#define IS_WORD_CONTEXT(c) ((c) & CONTEXT_WORD) +#define IS_NEWLINE_CONTEXT(c) ((c) & CONTEXT_NEWLINE) +#define IS_BEGBUF_CONTEXT(c) ((c) & CONTEXT_BEGBUF) +#define IS_ENDBUF_CONTEXT(c) ((c) & CONTEXT_ENDBUF) +#define IS_ORDINARY_CONTEXT(c) ((c) == 0) + +#define IS_WORD_CHAR(ch) (isalnum (ch) || (ch) == '_') +#define IS_NEWLINE(ch) ((ch) == NEWLINE_CHAR) +#define IS_WIDE_WORD_CHAR(ch) (iswalnum (ch) || (ch) == L'_') +#define IS_WIDE_NEWLINE(ch) ((ch) == WIDE_NEWLINE_CHAR) + +#define NOT_SATISFY_PREV_CONSTRAINT(constraint,context) \ + ((((constraint) & PREV_WORD_CONSTRAINT) && !IS_WORD_CONTEXT (context)) \ + || ((constraint & PREV_NOTWORD_CONSTRAINT) && IS_WORD_CONTEXT (context)) \ + || ((constraint & PREV_NEWLINE_CONSTRAINT) && !IS_NEWLINE_CONTEXT (context))\ + || ((constraint & PREV_BEGBUF_CONSTRAINT) && !IS_BEGBUF_CONTEXT (context))) + +#define NOT_SATISFY_NEXT_CONSTRAINT(constraint,context) \ + ((((constraint) & NEXT_WORD_CONSTRAINT) && !IS_WORD_CONTEXT (context)) \ + || (((constraint) & NEXT_NOTWORD_CONSTRAINT) && IS_WORD_CONTEXT (context)) \ + || (((constraint) & NEXT_NEWLINE_CONSTRAINT) && !IS_NEWLINE_CONTEXT (context)) \ + || (((constraint) & NEXT_ENDBUF_CONSTRAINT) && !IS_ENDBUF_CONTEXT (context))) + +struct re_dfastate_t +{ + re_hashval_t hash; + re_node_set nodes; + re_node_set non_eps_nodes; + re_node_set inveclosure; + re_node_set *entrance_nodes; + struct re_dfastate_t **trtable, **word_trtable; + unsigned int context : 4; + unsigned int halt : 1; + /* If this state can accept `multi byte'. + Note that we refer to multibyte characters, and multi character + collating elements as `multi byte'. */ + unsigned int accept_mb : 1; + /* If this state has backreference node(s). */ + unsigned int has_backref : 1; + unsigned int has_constraint : 1; +}; +typedef struct re_dfastate_t re_dfastate_t; + +struct re_state_table_entry +{ + Idx num; + Idx alloc; + re_dfastate_t **array; +}; + +/* Array type used in re_sub_match_last_t and re_sub_match_top_t. */ + +typedef struct +{ + Idx next_idx; + Idx alloc; + re_dfastate_t **array; +} state_array_t; + +/* Store information about the node NODE whose type is OP_CLOSE_SUBEXP. */ + +typedef struct +{ + Idx node; + Idx str_idx; /* The position NODE match at. */ + state_array_t path; +} re_sub_match_last_t; + +/* Store information about the node NODE whose type is OP_OPEN_SUBEXP. + And information about the node, whose type is OP_CLOSE_SUBEXP, + corresponding to NODE is stored in LASTS. */ + +typedef struct +{ + Idx str_idx; + Idx node; + state_array_t *path; + Idx alasts; /* Allocation size of LASTS. */ + Idx nlasts; /* The number of LASTS. */ + re_sub_match_last_t **lasts; +} re_sub_match_top_t; + +struct re_backref_cache_entry +{ + Idx node; + Idx str_idx; + Idx subexp_from; + Idx subexp_to; + char more; + char unused; + unsigned short int eps_reachable_subexps_map; +}; + +typedef struct +{ + /* The string object corresponding to the input string. */ + re_string_t input; +#if defined _LIBC || (defined __STDC_VERSION__ && __STDC_VERSION__ >= 199901L) + re_dfa_t *const dfa; +#else + re_dfa_t *dfa; +#endif + /* EFLAGS of the argument of regexec. */ + int eflags; + /* Where the matching ends. */ + Idx match_last; + Idx last_node; + /* The state log used by the matcher. */ + re_dfastate_t **state_log; + Idx state_log_top; + /* Back reference cache. */ + Idx nbkref_ents; + Idx abkref_ents; + struct re_backref_cache_entry *bkref_ents; + int max_mb_elem_len; + Idx nsub_tops; + Idx asub_tops; + re_sub_match_top_t **sub_tops; +} re_match_context_t; + +typedef struct +{ + re_dfastate_t **sifted_states; + re_dfastate_t **limited_states; + Idx last_node; + Idx last_str_idx; + re_node_set limits; +} re_sift_context_t; + +struct re_fail_stack_ent_t +{ + Idx idx; + Idx node; + regmatch_t *regs; + re_node_set eps_via_nodes; +}; + +struct re_fail_stack_t +{ + Idx num; + Idx alloc; + struct re_fail_stack_ent_t *stack; +}; + +struct re_dfa_t +{ + re_token_t *nodes; + Idx nodes_alloc; + Idx nodes_len; + Idx *nexts; + Idx *org_indices; + re_node_set *edests; + re_node_set *eclosures; + re_node_set *inveclosures; + struct re_state_table_entry *state_table; + re_dfastate_t *init_state; + re_dfastate_t *init_state_word; + re_dfastate_t *init_state_nl; + re_dfastate_t *init_state_begbuf; + bin_tree_t *str_tree; + bin_tree_storage_t *str_tree_storage; + re_bitset_ptr_t sb_char; + int str_tree_storage_idx; + + /* number of subexpressions `re_nsub' is in regex_t. */ + re_hashval_t state_hash_mask; + Idx init_node; + Idx nbackref; /* The number of backreference in this dfa. */ + + /* Bitmap expressing which backreference is used. */ + bitset_word used_bkref_map; + bitset_word completed_bkref_map; + + unsigned int has_plural_match : 1; + /* If this dfa has "multibyte node", which is a backreference or + a node which can accept multibyte character or multi character + collating element. */ + unsigned int has_mb_node : 1; + unsigned int is_utf8 : 1; + unsigned int map_notascii : 1; + unsigned int word_ops_used : 1; + int mb_cur_max; + bitset word_char; + reg_syntax_t syntax; + Idx *subexp_map; +#ifdef DEBUG + char* re_str; +#endif + __libc_lock_define (, lock) +}; + +#define re_node_set_init_empty(set) memset (set, '\0', sizeof (re_node_set)) +#define re_node_set_remove(set,id) \ + (re_node_set_remove_at (set, re_node_set_contains (set, id) - 1)) +#define re_node_set_empty(p) ((p)->nelem = 0) +#define re_node_set_free(set) re_free ((set)->elems) + +static void free_state (re_dfastate_t *state) internal_function; + + +typedef enum +{ + SB_CHAR, + MB_CHAR, + EQUIV_CLASS, + COLL_SYM, + CHAR_CLASS +} bracket_elem_type; + +typedef struct +{ + bracket_elem_type type; + union + { + unsigned char ch; + unsigned char *name; + wchar_t wch; + } opr; +} bracket_elem_t; + + +/* Inline functions for bitset operation. */ + +static inline void +bitset_set (bitset set, Idx i) +{ + set[i / BITSET_WORD_BITS] |= (bitset_word) 1 << i % BITSET_WORD_BITS; +} + +static inline void +bitset_clear (bitset set, Idx i) +{ + set[i / BITSET_WORD_BITS] &= ~ ((bitset_word) 1 << i % BITSET_WORD_BITS); +} + +static inline bool +bitset_contain (const bitset set, Idx i) +{ + return (set[i / BITSET_WORD_BITS] >> i % BITSET_WORD_BITS) & 1; +} + +static inline void +bitset_empty (bitset set) +{ + memset (set, 0, sizeof (bitset)); +} + +static inline void +bitset_set_all (bitset set) +{ + memset (set, -1, sizeof (bitset_word) * (SBC_MAX / BITSET_WORD_BITS)); + if (SBC_MAX % BITSET_WORD_BITS != 0) + set[BITSET_WORDS - 1] = + ((bitset_word) 1 << SBC_MAX % BITSET_WORD_BITS) - 1; +} + +static inline void +bitset_copy (bitset dest, const bitset src) +{ + memcpy (dest, src, sizeof (bitset)); +} + +static inline void +bitset_not (bitset set) +{ + int i; + for (i = 0; i < SBC_MAX / BITSET_WORD_BITS; ++i) + set[i] = ~set[i]; + if (SBC_MAX % BITSET_WORD_BITS != 0) + set[BITSET_WORDS - 1] = + ((((bitset_word) 1 << SBC_MAX % BITSET_WORD_BITS) - 1) + & ~set[BITSET_WORDS - 1]); +} + +static inline void +bitset_merge (bitset dest, const bitset src) +{ + int i; + for (i = 0; i < BITSET_WORDS; ++i) + dest[i] |= src[i]; +} + +static inline void +bitset_mask (bitset dest, const bitset src) +{ + int i; + for (i = 0; i < BITSET_WORDS; ++i) + dest[i] &= src[i]; +} + +#if defined RE_ENABLE_I18N +/* Inline functions for re_string. */ +static inline int +internal_function __attribute ((pure)) +re_string_char_size_at (const re_string_t *pstr, Idx idx) +{ + int byte_idx; + if (pstr->mb_cur_max == 1) + return 1; + for (byte_idx = 1; idx + byte_idx < pstr->valid_len; ++byte_idx) + if (pstr->wcs[idx + byte_idx] != WEOF) + break; + return byte_idx; +} + +static inline wint_t +internal_function __attribute ((pure)) +re_string_wchar_at (const re_string_t *pstr, Idx idx) +{ + if (pstr->mb_cur_max == 1) + return (wint_t) pstr->mbs[idx]; + return (wint_t) pstr->wcs[idx]; +} + +static int +internal_function __attribute ((pure)) +re_string_elem_size_at (const re_string_t *pstr, Idx idx) +{ +#ifdef _LIBC + const unsigned char *p, *extra; + const int32_t *table, *indirect; + int32_t tmp; +# include <locale/weight.h> + uint_fast32_t nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES); + + if (nrules != 0) + { + table = (const int32_t *) _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB); + extra = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_EXTRAMB); + indirect = (const int32_t *) _NL_CURRENT (LC_COLLATE, + _NL_COLLATE_INDIRECTMB); + p = pstr->mbs + idx; + tmp = findidx (&p); + return p - pstr->mbs - idx; + } + else +#endif /* _LIBC */ + return 1; +} +#endif /* RE_ENABLE_I18N */ + +#endif /* _REGEX_INTERNAL_H */ diff --git a/lib/regexec.c b/lib/regexec.c new file mode 100644 index 0000000..a85077c --- /dev/null +++ b/lib/regexec.c @@ -0,0 +1,4333 @@ +/* Extended regular expression matching and search library. + Copyright (C) 2002, 2003, 2004, 2005 Free Software Foundation, Inc. + This file is part of the GNU C Library. + Contributed by Isamu Hasegawa <isamu@yamato.ibm.com>. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +static reg_errcode_t match_ctx_init (re_match_context_t *cache, int eflags, + Idx n) internal_function; +static void match_ctx_clean (re_match_context_t *mctx) internal_function; +static void match_ctx_free (re_match_context_t *cache) internal_function; +static reg_errcode_t match_ctx_add_entry (re_match_context_t *cache, Idx node, + Idx str_idx, Idx from, Idx to) + internal_function; +static Idx search_cur_bkref_entry (const re_match_context_t *mctx, Idx str_idx) + internal_function; +static reg_errcode_t match_ctx_add_subtop (re_match_context_t *mctx, Idx node, + Idx str_idx) internal_function; +static re_sub_match_last_t * match_ctx_add_sublast (re_sub_match_top_t *subtop, + Idx node, Idx str_idx) + internal_function; +static void sift_ctx_init (re_sift_context_t *sctx, re_dfastate_t **sifted_sts, + re_dfastate_t **limited_sts, Idx last_node, + Idx last_str_idx) + internal_function; +static reg_errcode_t re_search_internal (const regex_t *preg, + const char *string, Idx length, + Idx start, Idx last_start, Idx stop, + size_t nmatch, regmatch_t pmatch[], + int eflags) internal_function; +static regoff_t re_search_2_stub (struct re_pattern_buffer *bufp, + const char *string1, Idx length1, + const char *string2, Idx length2, + Idx start, regoff_t range, + struct re_registers *regs, + Idx stop, bool ret_len) internal_function; +static regoff_t re_search_stub (struct re_pattern_buffer *bufp, + const char *string, Idx length, Idx start, + regoff_t range, Idx stop, + struct re_registers *regs, + bool ret_len) internal_function; +static unsigned re_copy_regs (struct re_registers *regs, regmatch_t *pmatch, + Idx nregs, int regs_allocated) internal_function; +static reg_errcode_t prune_impossible_nodes (re_match_context_t *mctx) + internal_function; +static Idx check_matching (re_match_context_t *mctx, bool fl_longest_match, + Idx *p_match_first) + internal_function; +static Idx check_halt_state_context (const re_match_context_t *mctx, + const re_dfastate_t *state, Idx idx) + internal_function; +static void update_regs (re_dfa_t *dfa, regmatch_t *pmatch, + regmatch_t *prev_idx_match, Idx cur_node, + Idx cur_idx, Idx nmatch) internal_function; +static reg_errcode_t push_fail_stack (struct re_fail_stack_t *fs, + Idx str_idx, Idx dest_node, Idx nregs, + regmatch_t *regs, + re_node_set *eps_via_nodes) internal_function; +static reg_errcode_t set_regs (const regex_t *preg, + const re_match_context_t *mctx, + size_t nmatch, regmatch_t *pmatch, + bool fl_backtrack) internal_function; +static reg_errcode_t free_fail_stack_return (struct re_fail_stack_t *fs) internal_function; + +#ifdef RE_ENABLE_I18N +static int sift_states_iter_mb (const re_match_context_t *mctx, + re_sift_context_t *sctx, + Idx node_idx, Idx str_idx, Idx max_str_idx) internal_function; +#endif /* RE_ENABLE_I18N */ +static reg_errcode_t sift_states_backward (re_match_context_t *mctx, + re_sift_context_t *sctx) internal_function; +static reg_errcode_t build_sifted_states (re_match_context_t *mctx, + re_sift_context_t *sctx, Idx str_idx, + re_node_set *cur_dest) internal_function; +static reg_errcode_t update_cur_sifted_state (re_match_context_t *mctx, + re_sift_context_t *sctx, + Idx str_idx, + re_node_set *dest_nodes) internal_function; +static reg_errcode_t add_epsilon_src_nodes (re_dfa_t *dfa, + re_node_set *dest_nodes, + const re_node_set *candidates) internal_function; +static bool check_dst_limits (const re_match_context_t *mctx, + const re_node_set *limits, + Idx dst_node, Idx dst_idx, Idx src_node, + Idx src_idx) internal_function; +static int check_dst_limits_calc_pos_1 (const re_match_context_t *mctx, + int boundaries, Idx subexp_idx, + Idx from_node, Idx bkref_idx) internal_function; +static int check_dst_limits_calc_pos (const re_match_context_t *mctx, + Idx limit, Idx subexp_idx, + Idx node, Idx str_idx, + Idx bkref_idx) internal_function; +static reg_errcode_t check_subexp_limits (re_dfa_t *dfa, + re_node_set *dest_nodes, + const re_node_set *candidates, + re_node_set *limits, + struct re_backref_cache_entry *bkref_ents, + Idx str_idx) internal_function; +static reg_errcode_t sift_states_bkref (re_match_context_t *mctx, + re_sift_context_t *sctx, + Idx str_idx, const re_node_set *candidates) internal_function; +static reg_errcode_t merge_state_array (re_dfa_t *dfa, re_dfastate_t **dst, + re_dfastate_t **src, Idx num) internal_function; +static re_dfastate_t *find_recover_state (reg_errcode_t *err, + re_match_context_t *mctx) internal_function; +static re_dfastate_t *transit_state (reg_errcode_t *err, + re_match_context_t *mctx, + re_dfastate_t *state) internal_function; +static re_dfastate_t *merge_state_with_log (reg_errcode_t *err, + re_match_context_t *mctx, + re_dfastate_t *next_state) internal_function; +static reg_errcode_t check_subexp_matching_top (re_match_context_t *mctx, + re_node_set *cur_nodes, + Idx str_idx) internal_function; +#if 0 +static re_dfastate_t *transit_state_sb (reg_errcode_t *err, + re_match_context_t *mctx, + re_dfastate_t *pstate) internal_function; +#endif +#ifdef RE_ENABLE_I18N +static reg_errcode_t transit_state_mb (re_match_context_t *mctx, + re_dfastate_t *pstate) internal_function; +#endif /* RE_ENABLE_I18N */ +static reg_errcode_t transit_state_bkref (re_match_context_t *mctx, + const re_node_set *nodes) internal_function; +static reg_errcode_t get_subexp (re_match_context_t *mctx, + Idx bkref_node, Idx bkref_str_idx) internal_function; +static reg_errcode_t get_subexp_sub (re_match_context_t *mctx, + const re_sub_match_top_t *sub_top, + re_sub_match_last_t *sub_last, + Idx bkref_node, Idx bkref_str) internal_function; +static Idx find_subexp_node (const re_dfa_t *dfa, const re_node_set *nodes, + Idx subexp_idx, int type) internal_function; +static reg_errcode_t check_arrival (re_match_context_t *mctx, + state_array_t *path, Idx top_node, + Idx top_str, Idx last_node, Idx last_str, + int type) internal_function; +static reg_errcode_t check_arrival_add_next_nodes (re_match_context_t *mctx, + Idx str_idx, + re_node_set *cur_nodes, + re_node_set *next_nodes) internal_function; +static reg_errcode_t check_arrival_expand_ecl (re_dfa_t *dfa, + re_node_set *cur_nodes, + Idx ex_subexp, int type) internal_function; +static reg_errcode_t check_arrival_expand_ecl_sub (re_dfa_t *dfa, + re_node_set *dst_nodes, + Idx target, Idx ex_subexp, + int type) internal_function; +static reg_errcode_t expand_bkref_cache (re_match_context_t *mctx, + re_node_set *cur_nodes, Idx cur_str, + Idx subexp_num, int type) internal_function; +static bool build_trtable (re_dfa_t *dfa, + re_dfastate_t *state) internal_function; +#ifdef RE_ENABLE_I18N +static int check_node_accept_bytes (re_dfa_t *dfa, Idx node_idx, + const re_string_t *input, Idx idx) internal_function; +# ifdef _LIBC +static unsigned int find_collation_sequence_value (const unsigned char *mbs, + size_t name_len) internal_function; +# endif /* _LIBC */ +#endif /* RE_ENABLE_I18N */ +static Idx group_nodes_into_DFAstates (const re_dfa_t *dfa, + const re_dfastate_t *state, + re_node_set *states_node, + bitset *states_ch) internal_function; +static bool check_node_accept (const re_match_context_t *mctx, + const re_token_t *node, Idx idx) + internal_function; +static reg_errcode_t extend_buffers (re_match_context_t *mctx) internal_function; + +/* Entry point for POSIX code. */ + +/* regexec searches for a given pattern, specified by PREG, in the + string STRING. + + If NMATCH is zero or REG_NOSUB was set in the cflags argument to + `regcomp', we ignore PMATCH. Otherwise, we assume PMATCH has at + least NMATCH elements, and we set them to the offsets of the + corresponding matched substrings. + + EFLAGS specifies `execution flags' which affect matching: if + REG_NOTBOL is set, then ^ does not match at the beginning of the + string; if REG_NOTEOL is set, then $ does not match at the end. + + We return 0 if we find a match and REG_NOMATCH if not. */ + +int +regexec (const regex_t *__restrict preg, const char *__restrict string, + size_t nmatch, regmatch_t pmatch[], int eflags) +{ + reg_errcode_t err; + Idx start, length; +#ifdef _LIBC + re_dfa_t *dfa = (re_dfa_t *) preg->re_buffer; +#endif + + if (eflags & ~(REG_NOTBOL | REG_NOTEOL | REG_STARTEND)) + return REG_BADPAT; + + if (eflags & REG_STARTEND) + { + start = pmatch[0].rm_so; + length = pmatch[0].rm_eo; + } + else + { + start = 0; + length = strlen (string); + } + + __libc_lock_lock (dfa->lock); + if (preg->re_no_sub) + err = re_search_internal (preg, string, length, start, length, + length, 0, NULL, eflags); + else + err = re_search_internal (preg, string, length, start, length, + length, nmatch, pmatch, eflags); + __libc_lock_unlock (dfa->lock); + return err != REG_NOERROR; +} + +#ifdef _LIBC +# include <shlib-compat.h> +versioned_symbol (libc, __regexec, regexec, GLIBC_2_3_4); + +# if SHLIB_COMPAT (libc, GLIBC_2_0, GLIBC_2_3_4) +__typeof__ (__regexec) __compat_regexec; + +int +attribute_compat_text_section +__compat_regexec (const regex_t *__restrict preg, + const char *__restrict string, size_t nmatch, + regmatch_t pmatch[], int eflags) +{ + return regexec (preg, string, nmatch, pmatch, + eflags & (REG_NOTBOL | REG_NOTEOL)); +} +compat_symbol (libc, __compat_regexec, regexec, GLIBC_2_0); +# endif +#endif + +/* Entry points for GNU code. */ + +/* re_match, re_search, re_match_2, re_search_2 + + The former two functions operate on STRING with length LENGTH, + while the later two operate on concatenation of STRING1 and STRING2 + with lengths LENGTH1 and LENGTH2, respectively. + + re_match() matches the compiled pattern in BUFP against the string, + starting at index START. + + re_search() first tries matching at index START, then it tries to match + starting from index START + 1, and so on. The last start position tried + is START + RANGE. (Thus RANGE = 0 forces re_search to operate the same + way as re_match().) + + The parameter STOP of re_{match,search}_2 specifies that no match exceeding + the first STOP characters of the concatenation of the strings should be + concerned. + + If REGS is not NULL, and BUFP->re_no_sub is not set, the offsets of the match + and all groups is stroed in REGS. (For the "_2" variants, the offsets are + computed relative to the concatenation, not relative to the individual + strings.) + + On success, re_match* functions return the length of the match, re_search* + return the position of the start of the match. Return value -1 means no + match was found and -2 indicates an internal error. */ + +regoff_t +re_match (struct re_pattern_buffer *bufp, const char *string, + Idx length, Idx start, struct re_registers *regs) +{ + return re_search_stub (bufp, string, length, start, 0, length, regs, true); +} +#ifdef _LIBC +weak_alias (__re_match, re_match) +#endif + +regoff_t +re_search (struct re_pattern_buffer *bufp, const char *string, + Idx length, Idx start, regoff_t range, struct re_registers *regs) +{ + return re_search_stub (bufp, string, length, start, range, length, regs, + false); +} +#ifdef _LIBC +weak_alias (__re_search, re_search) +#endif + +regoff_t +re_match_2 (struct re_pattern_buffer *bufp, + const char *string1, Idx length1, + const char *string2, Idx length2, + Idx start, struct re_registers *regs, Idx stop) +{ + return re_search_2_stub (bufp, string1, length1, string2, length2, + start, 0, regs, stop, true); +} +#ifdef _LIBC +weak_alias (__re_match_2, re_match_2) +#endif + +regoff_t +re_search_2 (struct re_pattern_buffer *bufp, + const char *string1, Idx length1, + const char *string2, Idx length2, + Idx start, regoff_t range, struct re_registers *regs, Idx stop) +{ + return re_search_2_stub (bufp, string1, length1, string2, length2, + start, range, regs, stop, false); +} +#ifdef _LIBC +weak_alias (__re_search_2, re_search_2) +#endif + +static regoff_t +internal_function +re_search_2_stub (struct re_pattern_buffer *bufp, + const char *string1, Idx length1, + const char *string2, Idx length2, + Idx start, regoff_t range, struct re_registers *regs, + Idx stop, bool ret_len) +{ + const char *str; + regoff_t rval; + Idx len = length1 + length2; + char *s = NULL; + + if (BE (length1 < 0 || length2 < 0 || stop < 0 || len < length1, 0)) + return -2; + + /* Concatenate the strings. */ + if (length2 > 0) + if (length1 > 0) + { + s = re_malloc (char, len); + + if (BE (s == NULL, 0)) + return -2; + memcpy (s, string1, length1); + memcpy (s + length1, string2, length2); + str = s; + } + else + str = string2; + else + str = string1; + + rval = re_search_stub (bufp, str, len, start, range, stop, regs, + ret_len); + re_free (s); + return rval; +} + +/* The parameters have the same meaning as those of re_search. + Additional parameters: + If RET_LEN is true the length of the match is returned (re_match style); + otherwise the position of the match is returned. */ + +static regoff_t +internal_function +re_search_stub (struct re_pattern_buffer *bufp, + const char *string, Idx length, + Idx start, regoff_t range, Idx stop, struct re_registers *regs, + bool ret_len) +{ + reg_errcode_t result; + regmatch_t *pmatch; + Idx nregs; + regoff_t rval; + int eflags = 0; +#ifdef _LIBC + re_dfa_t *dfa = (re_dfa_t *) bufp->re_buffer; +#endif + Idx last_start = start + range; + + /* Check for out-of-range. */ + if (BE (start < 0 || start > length, 0)) + return -1; + if (sizeof start < sizeof range) + { + regoff_t length_offset = length; + regoff_t start_offset = start; + if (BE (length_offset - start_offset < range, 0)) + last_start = length; + else if (BE (range < - start_offset, 0)) + last_start = 0; + } + else + { + if (BE ((last_start < start) != (range < 0), 0)) + { + /* Overflow occurred when computing last_start; substitute + the extreme value. */ + last_start = range < 0 ? 0 : length; + } + else + { + if (BE (length < last_start, 0)) + last_start = length; + else if (BE (last_start < 0, 0)) + last_start = 0; + } + } + + __libc_lock_lock (dfa->lock); + + eflags |= (bufp->re_not_bol) ? REG_NOTBOL : 0; + eflags |= (bufp->re_not_eol) ? REG_NOTEOL : 0; + + /* Compile fastmap if we haven't yet. */ + if (start < last_start && bufp->re_fastmap != NULL + && !bufp->re_fastmap_accurate) + re_compile_fastmap (bufp); + + if (BE (bufp->re_no_sub, 0)) + regs = NULL; + + /* We need at least 1 register. */ + if (regs == NULL) + nregs = 1; + else if (BE (bufp->re_regs_allocated == REG_FIXED + && regs->rm_num_regs <= bufp->re_nsub, 0)) + { + nregs = regs->rm_num_regs; + if (BE (nregs < 1, 0)) + { + /* Nothing can be copied to regs. */ + regs = NULL; + nregs = 1; + } + } + else + nregs = bufp->re_nsub + 1; + pmatch = re_xmalloc (regmatch_t, nregs); + if (BE (pmatch == NULL, 0)) + { + rval = -2; + goto out; + } + + result = re_search_internal (bufp, string, length, start, last_start, stop, + nregs, pmatch, eflags); + + rval = 0; + + /* I hope we needn't fill ther regs with -1's when no match was found. */ + if (result != REG_NOERROR) + rval = -1; + else if (regs != NULL) + { + /* If caller wants register contents data back, copy them. */ + bufp->re_regs_allocated = re_copy_regs (regs, pmatch, nregs, + bufp->re_regs_allocated); + if (BE (bufp->re_regs_allocated == REG_UNALLOCATED, 0)) + rval = -2; + } + + if (BE (rval == 0, 1)) + { + if (ret_len) + { + assert (pmatch[0].rm_so == start); + rval = pmatch[0].rm_eo - start; + } + else + rval = pmatch[0].rm_so; + } + re_free (pmatch); + out: + __libc_lock_unlock (dfa->lock); + return rval; +} + +static unsigned +internal_function +re_copy_regs (struct re_registers *regs, regmatch_t *pmatch, Idx nregs, + int regs_allocated) +{ + int rval = REG_REALLOCATE; + Idx i; + Idx need_regs = nregs + 1; + /* We need one extra element beyond `rm_num_regs' for the `-1' marker GNU code + uses. */ + + /* Have the register data arrays been allocated? */ + if (regs_allocated == REG_UNALLOCATED) + { /* No. So allocate them with malloc. */ + regs->rm_start = re_xmalloc (regoff_t, need_regs); + regs->rm_end = re_malloc (regoff_t, need_regs); + if (BE (regs->rm_start == NULL, 0) || BE (regs->rm_end == NULL, 0)) + return REG_UNALLOCATED; + regs->rm_num_regs = need_regs; + } + else if (regs_allocated == REG_REALLOCATE) + { /* Yes. If we need more elements than were already + allocated, reallocate them. If we need fewer, just + leave it alone. */ + if (BE (need_regs > regs->rm_num_regs, 0)) + { + regoff_t *new_start = + re_xrealloc (regs->rm_start, regoff_t, need_regs); + regoff_t *new_end = re_realloc (regs->rm_end, regoff_t, need_regs); + if (BE (new_start == NULL, 0) || BE (new_end == NULL, 0)) + return REG_UNALLOCATED; + regs->rm_start = new_start; + regs->rm_end = new_end; + regs->rm_num_regs = need_regs; + } + } + else + { + assert (regs_allocated == REG_FIXED); + /* This function may not be called with REG_FIXED and nregs too big. */ + assert (regs->rm_num_regs >= nregs); + rval = REG_FIXED; + } + + /* Copy the regs. */ + for (i = 0; i < nregs; ++i) + { + regs->rm_start[i] = pmatch[i].rm_so; + regs->rm_end[i] = pmatch[i].rm_eo; + } + for ( ; i < regs->rm_num_regs; ++i) + regs->rm_start[i] = regs->rm_end[i] = -1; + + return rval; +} + +/* Set REGS to hold NUM_REGS registers, storing them in STARTS and + ENDS. Subsequent matches using PATTERN_BUFFER and REGS will use + this memory for recording register information. STARTS and ENDS + must be allocated using the malloc library routine, and must each + be at least NUM_REGS * sizeof (regoff_t) bytes long. + + If NUM_REGS == 0, then subsequent matches should allocate their own + register data. + + Unless this function is called, the first search or match using + PATTERN_BUFFER will allocate its own register data, without + freeing the old data. */ + +void +re_set_registers (struct re_pattern_buffer *bufp, struct re_registers *regs, + __re_size_t num_regs, regoff_t *starts, regoff_t *ends) +{ + if (num_regs) + { + bufp->re_regs_allocated = REG_REALLOCATE; + regs->rm_num_regs = num_regs; + regs->rm_start = starts; + regs->rm_end = ends; + } + else + { + bufp->re_regs_allocated = REG_UNALLOCATED; + regs->rm_num_regs = 0; + regs->rm_start = regs->rm_end = NULL; + } +} +#ifdef _LIBC +weak_alias (__re_set_registers, re_set_registers) +#endif + +/* Entry points compatible with 4.2 BSD regex library. We don't define + them unless specifically requested. */ + +#if defined _REGEX_RE_COMP || defined _LIBC +int +# ifdef _LIBC +weak_function +# endif +re_exec (const char *s) +{ + return 0 == regexec (&re_comp_buf, s, 0, NULL, 0); +} +#endif /* _REGEX_RE_COMP */ + +/* Internal entry point. */ + +/* Searches for a compiled pattern PREG in the string STRING, whose + length is LENGTH. NMATCH, PMATCH, and EFLAGS have the same + meaning as with regexec. LAST_START is START + RANGE, where + START and RANGE have the same meaning as with re_search. + Return REG_NOERROR if we find a match, and REG_NOMATCH if not, + otherwise return the error code. + Note: We assume front end functions already check ranges. + (0 <= LAST_START && LAST_START <= LENGTH) */ + +static reg_errcode_t +internal_function +re_search_internal (const regex_t *preg, + const char *string, Idx length, + Idx start, Idx last_start, Idx stop, + size_t nmatch, regmatch_t pmatch[], + int eflags) +{ + reg_errcode_t err; + re_dfa_t *dfa = (re_dfa_t *) preg->re_buffer; + Idx left_lim, right_lim; + int incr; + bool fl_longest_match; + int match_kind; + Idx match_first, match_last = REG_MISSING; + Idx extra_nmatch; + bool sb; + int ch; +#if defined _LIBC || (defined __STDC_VERSION__ && __STDC_VERSION__ >= 199901L) + re_match_context_t mctx = { .dfa = dfa }; +#else + re_match_context_t mctx; +#endif + char *fastmap = ((preg->re_fastmap != NULL && preg->re_fastmap_accurate + && start != last_start && !preg->re_can_be_null) + ? preg->re_fastmap : NULL); + unsigned REG_TRANSLATE_TYPE t = + (unsigned REG_TRANSLATE_TYPE) preg->re_translate; + +#if !(defined _LIBC || (defined __STDC_VERSION__ && __STDC_VERSION__ >= 199901L)) + memset (&mctx, '\0', sizeof (re_match_context_t)); + mctx.dfa = dfa; +#endif + + extra_nmatch = (nmatch > preg->re_nsub) ? nmatch - (preg->re_nsub + 1) : 0; + nmatch -= extra_nmatch; + + /* Check if the DFA haven't been compiled. */ + if (BE (preg->re_used == 0 || dfa->init_state == NULL + || dfa->init_state_word == NULL || dfa->init_state_nl == NULL + || dfa->init_state_begbuf == NULL, 0)) + return REG_NOMATCH; + +#ifdef DEBUG + /* We assume front-end functions already check them. */ + assert (0 <= last_start && last_start <= length); +#endif + + /* If initial states with non-begbuf contexts have no elements, + the regex must be anchored. If preg->re_newline_anchor is set, + we'll never use init_state_nl, so do not check it. */ + if (dfa->init_state->nodes.nelem == 0 + && dfa->init_state_word->nodes.nelem == 0 + && (dfa->init_state_nl->nodes.nelem == 0 + || !preg->re_newline_anchor)) + { + if (start != 0 && last_start != 0) + return REG_NOMATCH; + start = last_start = 0; + } + + /* We must check the longest matching, if nmatch > 0. */ + fl_longest_match = (nmatch != 0 || dfa->nbackref); + + err = re_string_allocate (&mctx.input, string, length, dfa->nodes_len + 1, + preg->re_translate, + preg->re_syntax & REG_IGNORE_CASE, dfa); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + mctx.input.stop = stop; + mctx.input.raw_stop = stop; + mctx.input.newline_anchor = preg->re_newline_anchor; + + err = match_ctx_init (&mctx, eflags, dfa->nbackref * 2); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + + /* We will log all the DFA states through which the dfa pass, + if nmatch > 1, or this dfa has "multibyte node", which is a + back-reference or a node which can accept multibyte character or + multi character collating element. */ + if (nmatch > 1 || dfa->has_mb_node) + { + mctx.state_log = re_xmalloc (re_dfastate_t *, mctx.input.bufs_len + 1); + if (BE (mctx.state_log == NULL, 0)) + { + err = REG_ESPACE; + goto free_return; + } + } + else + mctx.state_log = NULL; + + match_first = start; + mctx.input.tip_context = (eflags & REG_NOTBOL) ? CONTEXT_BEGBUF + : CONTEXT_NEWLINE | CONTEXT_BEGBUF; + + /* Check incrementally whether of not the input string match. */ + incr = (last_start < start) ? -1 : 1; + left_lim = (last_start < start) ? last_start : start; + right_lim = (last_start < start) ? start : last_start; + sb = dfa->mb_cur_max == 1; + match_kind = + (fastmap + ? ((sb || !(preg->re_syntax & REG_IGNORE_CASE || t) ? 4 : 0) + | (start <= last_start ? 2 : 0) + | (t != NULL ? 1 : 0)) + : 8); + + for (;; match_first += incr) + { + err = REG_NOMATCH; + if (match_first < left_lim || right_lim < match_first) + goto free_return; + + /* Advance as rapidly as possible through the string, until we + find a plausible place to start matching. This may be done + with varying efficiency, so there are various possibilities: + only the most common of them are specialized, in order to + save on code size. We use a switch statement for speed. */ + switch (match_kind) + { + case 8: + /* No fastmap. */ + break; + + case 7: + /* Fastmap with single-byte translation, match forward. */ + while (BE (match_first < right_lim, 1) + && !fastmap[t[(unsigned char) string[match_first]]]) + ++match_first; + goto forward_match_found_start_or_reached_end; + + case 6: + /* Fastmap without translation, match forward. */ + while (BE (match_first < right_lim, 1) + && !fastmap[(unsigned char) string[match_first]]) + ++match_first; + + forward_match_found_start_or_reached_end: + if (BE (match_first == right_lim, 0)) + { + ch = match_first >= length + ? 0 : (unsigned char) string[match_first]; + if (!fastmap[t ? t[ch] : ch]) + goto free_return; + } + break; + + case 4: + case 5: + /* Fastmap without multi-byte translation, match backwards. */ + while (match_first >= left_lim) + { + ch = match_first >= length + ? 0 : (unsigned char) string[match_first]; + if (fastmap[t ? t[ch] : ch]) + break; + --match_first; + } + if (match_first < left_lim) + goto free_return; + break; + + default: + /* In this case, we can't determine easily the current byte, + since it might be a component byte of a multibyte + character. Then we use the constructed buffer instead. */ + for (;;) + { + /* If MATCH_FIRST is out of the valid range, reconstruct the + buffers. */ + __re_size_t offset = match_first - mctx.input.raw_mbs_idx; + if (BE (offset >= (__re_size_t) mctx.input.valid_raw_len, 0)) + { + err = re_string_reconstruct (&mctx.input, match_first, + eflags); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + + offset = match_first - mctx.input.raw_mbs_idx; + } + /* If MATCH_FIRST is out of the buffer, leave it as '\0'. + Note that MATCH_FIRST must not be smaller than 0. */ + ch = (match_first >= length + ? 0 : re_string_byte_at (&mctx.input, offset)); + if (fastmap[ch]) + break; + match_first += incr; + if (match_first < left_lim || match_first > right_lim) + { + err = REG_NOMATCH; + goto free_return; + } + } + break; + } + + /* Reconstruct the buffers so that the matcher can assume that + the matching starts from the beginning of the buffer. */ + err = re_string_reconstruct (&mctx.input, match_first, eflags); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + +#ifdef RE_ENABLE_I18N + /* Don't consider this char as a possible match start if it part, + yet isn't the head, of a multibyte character. */ + if (!sb && !re_string_first_byte (&mctx.input, 0)) + continue; +#endif + + /* It seems to be appropriate one, then use the matcher. */ + /* We assume that the matching starts from 0. */ + mctx.state_log_top = mctx.nbkref_ents = mctx.max_mb_elem_len = 0; + match_last = check_matching (&mctx, fl_longest_match, + start <= last_start ? &match_first : NULL); + if (match_last != REG_MISSING) + { + if (BE (match_last == REG_ERROR, 0)) + { + err = REG_ESPACE; + goto free_return; + } + else + { + mctx.match_last = match_last; + if ((!preg->re_no_sub && nmatch > 1) || dfa->nbackref) + { + re_dfastate_t *pstate = mctx.state_log[match_last]; + mctx.last_node = check_halt_state_context (&mctx, pstate, + match_last); + } + if ((!preg->re_no_sub && nmatch > 1 && dfa->has_plural_match) + || dfa->nbackref) + { + err = prune_impossible_nodes (&mctx); + if (err == REG_NOERROR) + break; + if (BE (err != REG_NOMATCH, 0)) + goto free_return; + match_last = REG_MISSING; + } + else + break; /* We found a match. */ + } + } + + match_ctx_clean (&mctx); + } + +#ifdef DEBUG + assert (match_last != REG_MISSING); + assert (err == REG_NOERROR); +#endif + + /* Set pmatch[] if we need. */ + if (nmatch > 0) + { + Idx reg_idx; + + /* Initialize registers. */ + for (reg_idx = 1; reg_idx < nmatch; ++reg_idx) + pmatch[reg_idx].rm_so = pmatch[reg_idx].rm_eo = -1; + + /* Set the points where matching start/end. */ + pmatch[0].rm_so = 0; + pmatch[0].rm_eo = mctx.match_last; + /* FIXME: This function should fail if mctx.match_last exceeds + the maximum possible regoff_t value. We need a new error + code REG_OVERFLOW. */ + + if (!preg->re_no_sub && nmatch > 1) + { + err = set_regs (preg, &mctx, nmatch, pmatch, + dfa->has_plural_match && dfa->nbackref > 0); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + } + + /* At last, add the offset to the each registers, since we slided + the buffers so that we could assume that the matching starts + from 0. */ + for (reg_idx = 0; reg_idx < nmatch; ++reg_idx) + if (pmatch[reg_idx].rm_so != -1) + { +#ifdef RE_ENABLE_I18N + if (BE (mctx.input.offsets_needed != 0, 0)) + { + pmatch[reg_idx].rm_so = + (pmatch[reg_idx].rm_so == mctx.input.valid_len + ? mctx.input.valid_raw_len + : mctx.input.offsets[pmatch[reg_idx].rm_so]); + pmatch[reg_idx].rm_eo = + (pmatch[reg_idx].rm_eo == mctx.input.valid_len + ? mctx.input.valid_raw_len + : mctx.input.offsets[pmatch[reg_idx].rm_eo]); + } +#else + assert (mctx.input.offsets_needed == 0); +#endif + pmatch[reg_idx].rm_so += match_first; + pmatch[reg_idx].rm_eo += match_first; + } + for (reg_idx = 0; reg_idx < extra_nmatch; ++reg_idx) + { + pmatch[nmatch + reg_idx].rm_so = -1; + pmatch[nmatch + reg_idx].rm_eo = -1; + } + + if (dfa->subexp_map) + for (reg_idx = 0; reg_idx + 1 < nmatch; reg_idx++) + if (dfa->subexp_map[reg_idx] != reg_idx) + { + pmatch[reg_idx + 1].rm_so + = pmatch[dfa->subexp_map[reg_idx] + 1].rm_so; + pmatch[reg_idx + 1].rm_eo + = pmatch[dfa->subexp_map[reg_idx] + 1].rm_eo; + } + } + + free_return: + re_free (mctx.state_log); + if (dfa->nbackref) + match_ctx_free (&mctx); + re_string_destruct (&mctx.input); + return err; +} + +static reg_errcode_t +internal_function +prune_impossible_nodes (re_match_context_t *mctx) +{ + re_dfa_t *const dfa = mctx->dfa; + Idx halt_node, match_last; + reg_errcode_t ret; + re_dfastate_t **sifted_states; + re_dfastate_t **lim_states = NULL; + re_sift_context_t sctx; +#ifdef DEBUG + assert (mctx->state_log != NULL); +#endif + match_last = mctx->match_last; + halt_node = mctx->last_node; + sifted_states = re_xmalloc (re_dfastate_t *, match_last + 1); + if (BE (sifted_states == NULL, 0)) + { + ret = REG_ESPACE; + goto free_return; + } + if (dfa->nbackref) + { + lim_states = re_xmalloc (re_dfastate_t *, match_last + 1); + if (BE (lim_states == NULL, 0)) + { + ret = REG_ESPACE; + goto free_return; + } + while (1) + { + memset (lim_states, '\0', + sizeof (re_dfastate_t *) * (match_last + 1)); + sift_ctx_init (&sctx, sifted_states, lim_states, halt_node, + match_last); + ret = sift_states_backward (mctx, &sctx); + re_node_set_free (&sctx.limits); + if (BE (ret != REG_NOERROR, 0)) + goto free_return; + if (sifted_states[0] != NULL || lim_states[0] != NULL) + break; + do + { + --match_last; + if (! REG_VALID_INDEX (match_last)) + { + ret = REG_NOMATCH; + goto free_return; + } + } while (mctx->state_log[match_last] == NULL + || !mctx->state_log[match_last]->halt); + halt_node = check_halt_state_context (mctx, + mctx->state_log[match_last], + match_last); + } + ret = merge_state_array (dfa, sifted_states, lim_states, + match_last + 1); + re_free (lim_states); + lim_states = NULL; + if (BE (ret != REG_NOERROR, 0)) + goto free_return; + } + else + { + sift_ctx_init (&sctx, sifted_states, lim_states, halt_node, match_last); + ret = sift_states_backward (mctx, &sctx); + re_node_set_free (&sctx.limits); + if (BE (ret != REG_NOERROR, 0)) + goto free_return; + } + re_free (mctx->state_log); + mctx->state_log = sifted_states; + sifted_states = NULL; + mctx->last_node = halt_node; + mctx->match_last = match_last; + ret = REG_NOERROR; + free_return: + re_free (sifted_states); + re_free (lim_states); + return ret; +} + +/* Acquire an initial state and return it. + We must select appropriate initial state depending on the context, + since initial states may have constraints like "\<", "^", etc.. */ + +static inline re_dfastate_t * +__attribute ((always_inline)) internal_function +acquire_init_state_context (reg_errcode_t *err, const re_match_context_t *mctx, + Idx idx) +{ + re_dfa_t *const dfa = mctx->dfa; + if (dfa->init_state->has_constraint) + { + unsigned int context; + context = re_string_context_at (&mctx->input, idx - 1, mctx->eflags); + if (IS_WORD_CONTEXT (context)) + return dfa->init_state_word; + else if (IS_ORDINARY_CONTEXT (context)) + return dfa->init_state; + else if (IS_BEGBUF_CONTEXT (context) && IS_NEWLINE_CONTEXT (context)) + return dfa->init_state_begbuf; + else if (IS_NEWLINE_CONTEXT (context)) + return dfa->init_state_nl; + else if (IS_BEGBUF_CONTEXT (context)) + { + /* It is relatively rare case, then calculate on demand. */ + return re_acquire_state_context (err, dfa, + dfa->init_state->entrance_nodes, + context); + } + else + /* Must not happen? */ + return dfa->init_state; + } + else + return dfa->init_state; +} + +/* Check whether the regular expression match input string INPUT or not, + and return the index where the matching end. Return REG_MISSING if + there is no match, and return REG_ERROR in case of an error. + FL_LONGEST_MATCH means we want the POSIX longest matching. + If P_MATCH_FIRST is not NULL, and the match fails, it is set to the + next place where we may want to try matching. + Note that the matcher assume that the maching starts from the current + index of the buffer. */ + +static Idx +internal_function +check_matching (re_match_context_t *mctx, bool fl_longest_match, + Idx *p_match_first) +{ + re_dfa_t *const dfa = mctx->dfa; + reg_errcode_t err; + Idx match = 0; + Idx match_last = REG_MISSING; + Idx cur_str_idx = re_string_cur_idx (&mctx->input); + re_dfastate_t *cur_state; + bool at_init_state = p_match_first != NULL; + Idx next_start_idx = cur_str_idx; + + err = REG_NOERROR; + cur_state = acquire_init_state_context (&err, mctx, cur_str_idx); + /* An initial state must not be NULL (invalid). */ + if (BE (cur_state == NULL, 0)) + { + assert (err == REG_ESPACE); + return REG_ERROR; + } + + if (mctx->state_log != NULL) + { + mctx->state_log[cur_str_idx] = cur_state; + + /* Check OP_OPEN_SUBEXP in the initial state in case that we use them + later. E.g. Processing back references. */ + if (BE (dfa->nbackref, 0)) + { + at_init_state = false; + err = check_subexp_matching_top (mctx, &cur_state->nodes, 0); + if (BE (err != REG_NOERROR, 0)) + return err; + + if (cur_state->has_backref) + { + err = transit_state_bkref (mctx, &cur_state->nodes); + if (BE (err != REG_NOERROR, 0)) + return err; + } + } + } + + /* If the RE accepts NULL string. */ + if (BE (cur_state->halt, 0)) + { + if (!cur_state->has_constraint + || check_halt_state_context (mctx, cur_state, cur_str_idx)) + { + if (!fl_longest_match) + return cur_str_idx; + else + { + match_last = cur_str_idx; + match = 1; + } + } + } + + while (!re_string_eoi (&mctx->input)) + { + re_dfastate_t *old_state = cur_state; + Idx next_char_idx = re_string_cur_idx (&mctx->input) + 1; + + if (BE (next_char_idx >= mctx->input.bufs_len, 0) + || (BE (next_char_idx >= mctx->input.valid_len, 0) + && mctx->input.valid_len < mctx->input.len)) + { + err = extend_buffers (mctx); + if (BE (err != REG_NOERROR, 0)) + { + assert (err == REG_ESPACE); + return REG_ERROR; + } + } + + cur_state = transit_state (&err, mctx, cur_state); + if (mctx->state_log != NULL) + cur_state = merge_state_with_log (&err, mctx, cur_state); + + if (cur_state == NULL) + { + /* Reached the invalid state or an error. Try to recover a valid + state using the state log, if available and if we have not + already found a valid (even if not the longest) match. */ + if (BE (err != REG_NOERROR, 0)) + return REG_ERROR; + + if (mctx->state_log == NULL + || (match && !fl_longest_match) + || (cur_state = find_recover_state (&err, mctx)) == NULL) + break; + } + + if (BE (at_init_state, 0)) + { + if (old_state == cur_state) + next_start_idx = next_char_idx; + else + at_init_state = false; + } + + if (cur_state->halt) + { + /* Reached a halt state. + Check the halt state can satisfy the current context. */ + if (!cur_state->has_constraint + || check_halt_state_context (mctx, cur_state, + re_string_cur_idx (&mctx->input))) + { + /* We found an appropriate halt state. */ + match_last = re_string_cur_idx (&mctx->input); + match = 1; + + /* We found a match, do not modify match_first below. */ + p_match_first = NULL; + if (!fl_longest_match) + break; + } + } + } + + if (p_match_first) + *p_match_first += next_start_idx; + + return match_last; +} + +/* Check NODE match the current context. */ + +static bool +internal_function +check_halt_node_context (const re_dfa_t *dfa, Idx node, unsigned int context) +{ + re_token_type_t type = dfa->nodes[node].type; + unsigned int constraint = dfa->nodes[node].constraint; + if (type != END_OF_RE) + return false; + if (!constraint) + return true; + if (NOT_SATISFY_NEXT_CONSTRAINT (constraint, context)) + return false; + return true; +} + +/* Check the halt state STATE match the current context. + Return 0 if not match, if the node, STATE has, is a halt node and + match the context, return the node. */ + +static Idx +internal_function +check_halt_state_context (const re_match_context_t *mctx, + const re_dfastate_t *state, Idx idx) +{ + Idx i; + unsigned int context; +#ifdef DEBUG + assert (state->halt); +#endif + context = re_string_context_at (&mctx->input, idx, mctx->eflags); + for (i = 0; i < state->nodes.nelem; ++i) + if (check_halt_node_context (mctx->dfa, state->nodes.elems[i], context)) + return state->nodes.elems[i]; + return 0; +} + +/* Compute the next node to which "NFA" transit from NODE("NFA" is a NFA + corresponding to the DFA). + Return the destination node, and update EPS_VIA_NODES; + return REG_MISSING in case of errors. */ + +static Idx +internal_function +proceed_next_node (const re_match_context_t *mctx, + Idx nregs, regmatch_t *regs, Idx *pidx, Idx node, + re_node_set *eps_via_nodes, struct re_fail_stack_t *fs) +{ + re_dfa_t *const dfa = mctx->dfa; + Idx i; + bool ok; + if (IS_EPSILON_NODE (dfa->nodes[node].type)) + { + re_node_set *cur_nodes = &mctx->state_log[*pidx]->nodes; + re_node_set *edests = &dfa->edests[node]; + Idx dest_node; + ok = re_node_set_insert (eps_via_nodes, node); + if (BE (! ok, 0)) + return REG_ERROR; + /* Pick up a valid destination, or return REG_MISSING if none + is found. */ + for (dest_node = REG_MISSING, i = 0; i < edests->nelem; ++i) + { + Idx candidate = edests->elems[i]; + if (!re_node_set_contains (cur_nodes, candidate)) + continue; + if (dest_node == REG_MISSING) + dest_node = candidate; + + else + { + /* In order to avoid infinite loop like "(a*)*", return the second + epsilon-transition if the first was already considered. */ + if (re_node_set_contains (eps_via_nodes, dest_node)) + return candidate; + + /* Otherwise, push the second epsilon-transition on the fail stack. */ + else if (fs != NULL + && push_fail_stack (fs, *pidx, candidate, nregs, regs, + eps_via_nodes)) + return REG_ERROR; + + /* We know we are going to exit. */ + break; + } + } + return dest_node; + } + else + { + Idx naccepted = 0; + re_token_type_t type = dfa->nodes[node].type; + +#ifdef RE_ENABLE_I18N + if (dfa->nodes[node].accept_mb) + naccepted = check_node_accept_bytes (dfa, node, &mctx->input, *pidx); + else +#endif /* RE_ENABLE_I18N */ + if (type == OP_BACK_REF) + { + Idx subexp_idx = dfa->nodes[node].opr.idx + 1; + naccepted = regs[subexp_idx].rm_eo - regs[subexp_idx].rm_so; + if (fs != NULL) + { + if (regs[subexp_idx].rm_so == -1 || regs[subexp_idx].rm_eo == -1) + return REG_MISSING; + else if (naccepted) + { + char *buf = (char *) re_string_get_buffer (&mctx->input); + if (memcmp (buf + regs[subexp_idx].rm_so, buf + *pidx, + naccepted) != 0) + return REG_MISSING; + } + } + + if (naccepted == 0) + { + Idx dest_node; + ok = re_node_set_insert (eps_via_nodes, node); + if (BE (! ok, 0)) + return REG_ERROR; + dest_node = dfa->edests[node].elems[0]; + if (re_node_set_contains (&mctx->state_log[*pidx]->nodes, + dest_node)) + return dest_node; + } + } + + if (naccepted != 0 + || check_node_accept (mctx, dfa->nodes + node, *pidx)) + { + Idx dest_node = dfa->nexts[node]; + *pidx = (naccepted == 0) ? *pidx + 1 : *pidx + naccepted; + if (fs && (*pidx > mctx->match_last || mctx->state_log[*pidx] == NULL + || !re_node_set_contains (&mctx->state_log[*pidx]->nodes, + dest_node))) + return REG_MISSING; + re_node_set_empty (eps_via_nodes); + return dest_node; + } + } + return REG_MISSING; +} + +static reg_errcode_t +internal_function +push_fail_stack (struct re_fail_stack_t *fs, Idx str_idx, Idx dest_node, + Idx nregs, regmatch_t *regs, re_node_set *eps_via_nodes) +{ + reg_errcode_t err; + Idx num = fs->num++; + if (fs->num == fs->alloc) + { + struct re_fail_stack_ent_t *new_array = + re_x2realloc (fs->stack, struct re_fail_stack_ent_t, &fs->alloc); + if (new_array == NULL) + return REG_ESPACE; + fs->stack = new_array; + } + fs->stack[num].idx = str_idx; + fs->stack[num].node = dest_node; + fs->stack[num].regs = re_xmalloc (regmatch_t, nregs); + if (fs->stack[num].regs == NULL) + return REG_ESPACE; + memcpy (fs->stack[num].regs, regs, sizeof (regmatch_t) * nregs); + err = re_node_set_init_copy (&fs->stack[num].eps_via_nodes, eps_via_nodes); + return err; +} + +static Idx +internal_function +pop_fail_stack (struct re_fail_stack_t *fs, Idx *pidx, + Idx nregs, regmatch_t *regs, re_node_set *eps_via_nodes) +{ + Idx num = --fs->num; + assert (REG_VALID_INDEX (num)); + *pidx = fs->stack[num].idx; + memcpy (regs, fs->stack[num].regs, sizeof (regmatch_t) * nregs); + re_node_set_free (eps_via_nodes); + re_free (fs->stack[num].regs); + *eps_via_nodes = fs->stack[num].eps_via_nodes; + return fs->stack[num].node; +} + +/* Set the positions where the subexpressions are starts/ends to registers + PMATCH. + Note: We assume that pmatch[0] is already set, and + pmatch[i].rm_so == pmatch[i].rm_eo == -1 for 0 < i < nmatch. */ + +static reg_errcode_t +internal_function +set_regs (const regex_t *preg, const re_match_context_t *mctx, + size_t nmatch, regmatch_t *pmatch, bool fl_backtrack) +{ + re_dfa_t *dfa = (re_dfa_t *) preg->re_buffer; + Idx idx, cur_node; + re_node_set eps_via_nodes; + struct re_fail_stack_t *fs; + struct re_fail_stack_t fs_body = { 0, 2, NULL }; + regmatch_t *prev_idx_match; + bool prev_idx_match_malloced = false; + +#ifdef DEBUG + assert (nmatch > 1); + assert (mctx->state_log != NULL); +#endif + if (fl_backtrack) + { + fs = &fs_body; + fs->stack = re_xmalloc (struct re_fail_stack_ent_t, fs->alloc); + if (fs->stack == NULL) + return REG_ESPACE; + } + else + fs = NULL; + + cur_node = dfa->init_node; + re_node_set_init_empty (&eps_via_nodes); + + if (re_alloc_oversized (nmatch, sizeof (regmatch_t))) + { + free_fail_stack_return (fs); + return REG_ESPACE; + } + if (__libc_use_alloca (nmatch * sizeof (regmatch_t))) + prev_idx_match = (regmatch_t *) alloca (nmatch * sizeof (regmatch_t)); + else + { + prev_idx_match = re_malloc (regmatch_t, nmatch); + if (prev_idx_match == NULL) + { + free_fail_stack_return (fs); + return REG_ESPACE; + } + prev_idx_match_malloced = true; + } + memcpy (prev_idx_match, pmatch, sizeof (regmatch_t) * nmatch); + + for (idx = pmatch[0].rm_so; idx <= pmatch[0].rm_eo ;) + { + update_regs (dfa, pmatch, prev_idx_match, cur_node, idx, nmatch); + + if (idx == pmatch[0].rm_eo && cur_node == mctx->last_node) + { + Idx reg_idx; + if (fs) + { + for (reg_idx = 0; reg_idx < nmatch; ++reg_idx) + if (pmatch[reg_idx].rm_so > -1 && pmatch[reg_idx].rm_eo == -1) + break; + if (reg_idx == nmatch) + { + re_node_set_free (&eps_via_nodes); + if (prev_idx_match_malloced) + re_free (prev_idx_match); + return free_fail_stack_return (fs); + } + cur_node = pop_fail_stack (fs, &idx, nmatch, pmatch, + &eps_via_nodes); + } + else + { + re_node_set_free (&eps_via_nodes); + if (prev_idx_match_malloced) + re_free (prev_idx_match); + return REG_NOERROR; + } + } + + /* Proceed to next node. */ + cur_node = proceed_next_node (mctx, nmatch, pmatch, &idx, cur_node, + &eps_via_nodes, fs); + + if (BE (! REG_VALID_INDEX (cur_node), 0)) + { + if (BE (cur_node == REG_ERROR, 0)) + { + re_node_set_free (&eps_via_nodes); + if (prev_idx_match_malloced) + re_free (prev_idx_match); + free_fail_stack_return (fs); + return REG_ESPACE; + } + if (fs) + cur_node = pop_fail_stack (fs, &idx, nmatch, pmatch, + &eps_via_nodes); + else + { + re_node_set_free (&eps_via_nodes); + if (prev_idx_match_malloced) + re_free (prev_idx_match); + return REG_NOMATCH; + } + } + } + re_node_set_free (&eps_via_nodes); + if (prev_idx_match_malloced) + re_free (prev_idx_match); + return free_fail_stack_return (fs); +} + +static reg_errcode_t +internal_function +free_fail_stack_return (struct re_fail_stack_t *fs) +{ + if (fs) + { + Idx fs_idx; + for (fs_idx = 0; fs_idx < fs->num; ++fs_idx) + { + re_node_set_free (&fs->stack[fs_idx].eps_via_nodes); + re_free (fs->stack[fs_idx].regs); + } + re_free (fs->stack); + } + return REG_NOERROR; +} + +static void +internal_function +update_regs (re_dfa_t *dfa, regmatch_t *pmatch, regmatch_t *prev_idx_match, + Idx cur_node, Idx cur_idx, Idx nmatch) +{ + int type = dfa->nodes[cur_node].type; + if (type == OP_OPEN_SUBEXP) + { + Idx reg_num = dfa->nodes[cur_node].opr.idx + 1; + + /* We are at the first node of this sub expression. */ + if (reg_num < nmatch) + { + pmatch[reg_num].rm_so = cur_idx; + pmatch[reg_num].rm_eo = -1; + } + } + else if (type == OP_CLOSE_SUBEXP) + { + Idx reg_num = dfa->nodes[cur_node].opr.idx + 1; + if (reg_num < nmatch) + { + /* We are at the last node of this sub expression. */ + if (pmatch[reg_num].rm_so < cur_idx) + { + pmatch[reg_num].rm_eo = cur_idx; + /* This is a non-empty match or we are not inside an optional + subexpression. Accept this right away. */ + memcpy (prev_idx_match, pmatch, sizeof (regmatch_t) * nmatch); + } + else + { + if (dfa->nodes[cur_node].opt_subexp + && prev_idx_match[reg_num].rm_so != -1) + /* We transited through an empty match for an optional + subexpression, like (a?)*, and this is not the subexp's + first match. Copy back the old content of the registers + so that matches of an inner subexpression are undone as + well, like in ((a?))*. */ + memcpy (pmatch, prev_idx_match, sizeof (regmatch_t) * nmatch); + else + /* We completed a subexpression, but it may be part of + an optional one, so do not update PREV_IDX_MATCH. */ + pmatch[reg_num].rm_eo = cur_idx; + } + } + } +} + +/* This function checks the STATE_LOG from the SCTX->last_str_idx to 0 + and sift the nodes in each states according to the following rules. + Updated state_log will be wrote to STATE_LOG. + + Rules: We throw away the Node `a' in the STATE_LOG[STR_IDX] if... + 1. When STR_IDX == MATCH_LAST(the last index in the state_log): + If `a' isn't the LAST_NODE and `a' can't epsilon transit to + the LAST_NODE, we throw away the node `a'. + 2. When 0 <= STR_IDX < MATCH_LAST and `a' accepts + string `s' and transit to `b': + i. If 'b' isn't in the STATE_LOG[STR_IDX+strlen('s')], we throw + away the node `a'. + ii. If 'b' is in the STATE_LOG[STR_IDX+strlen('s')] but 'b' is + thrown away, we throw away the node `a'. + 3. When 0 <= STR_IDX < MATCH_LAST and 'a' epsilon transit to 'b': + i. If 'b' isn't in the STATE_LOG[STR_IDX], we throw away the + node `a'. + ii. If 'b' is in the STATE_LOG[STR_IDX] but 'b' is thrown away, + we throw away the node `a'. */ + +#define STATE_NODE_CONTAINS(state,node) \ + ((state) != NULL && re_node_set_contains (&(state)->nodes, node)) + +static reg_errcode_t +internal_function +sift_states_backward (re_match_context_t *mctx, re_sift_context_t *sctx) +{ + reg_errcode_t err; + int null_cnt = 0; + Idx str_idx = sctx->last_str_idx; + re_node_set cur_dest; + +#ifdef DEBUG + assert (mctx->state_log != NULL && mctx->state_log[str_idx] != NULL); +#endif + + /* Build sifted state_log[str_idx]. It has the nodes which can epsilon + transit to the last_node and the last_node itself. */ + err = re_node_set_init_1 (&cur_dest, sctx->last_node); + if (BE (err != REG_NOERROR, 0)) + return err; + err = update_cur_sifted_state (mctx, sctx, str_idx, &cur_dest); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + + /* Then check each states in the state_log. */ + while (str_idx > 0) + { + /* Update counters. */ + null_cnt = (sctx->sifted_states[str_idx] == NULL) ? null_cnt + 1 : 0; + if (null_cnt > mctx->max_mb_elem_len) + { + memset (sctx->sifted_states, '\0', + sizeof (re_dfastate_t *) * str_idx); + re_node_set_free (&cur_dest); + return REG_NOERROR; + } + re_node_set_empty (&cur_dest); + --str_idx; + + if (mctx->state_log[str_idx]) + { + err = build_sifted_states (mctx, sctx, str_idx, &cur_dest); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + } + + /* Add all the nodes which satisfy the following conditions: + - It can epsilon transit to a node in CUR_DEST. + - It is in CUR_SRC. + And update state_log. */ + err = update_cur_sifted_state (mctx, sctx, str_idx, &cur_dest); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + } + err = REG_NOERROR; + free_return: + re_node_set_free (&cur_dest); + return err; +} + +static reg_errcode_t +internal_function +build_sifted_states (re_match_context_t *mctx, re_sift_context_t *sctx, + Idx str_idx, re_node_set *cur_dest) +{ + re_dfa_t *const dfa = mctx->dfa; + re_node_set *cur_src = &mctx->state_log[str_idx]->non_eps_nodes; + Idx i; + + /* Then build the next sifted state. + We build the next sifted state on `cur_dest', and update + `sifted_states[str_idx]' with `cur_dest'. + Note: + `cur_dest' is the sifted state from `state_log[str_idx + 1]'. + `cur_src' points the node_set of the old `state_log[str_idx]' + (with the epsilon nodes pre-filtered out). */ + for (i = 0; i < cur_src->nelem; i++) + { + Idx prev_node = cur_src->elems[i]; + int naccepted = 0; + bool ok; + +#ifdef DEBUG + re_token_type_t type = dfa->nodes[prev_node].type; + assert (!IS_EPSILON_NODE (type)); +#endif +#ifdef RE_ENABLE_I18N + /* If the node may accept `multi byte'. */ + if (dfa->nodes[prev_node].accept_mb) + naccepted = sift_states_iter_mb (mctx, sctx, prev_node, + str_idx, sctx->last_str_idx); +#endif /* RE_ENABLE_I18N */ + + /* We don't check backreferences here. + See update_cur_sifted_state(). */ + if (!naccepted + && check_node_accept (mctx, dfa->nodes + prev_node, str_idx) + && STATE_NODE_CONTAINS (sctx->sifted_states[str_idx + 1], + dfa->nexts[prev_node])) + naccepted = 1; + + if (naccepted == 0) + continue; + + if (sctx->limits.nelem) + { + Idx to_idx = str_idx + naccepted; + if (check_dst_limits (mctx, &sctx->limits, + dfa->nexts[prev_node], to_idx, + prev_node, str_idx)) + continue; + } + ok = re_node_set_insert (cur_dest, prev_node); + if (BE (! ok, 0)) + return REG_ESPACE; + } + + return REG_NOERROR; +} + +/* Helper functions. */ + +static reg_errcode_t +internal_function +clean_state_log_if_needed (re_match_context_t *mctx, Idx next_state_log_idx) +{ + Idx top = mctx->state_log_top; + + if (next_state_log_idx >= mctx->input.bufs_len + || (next_state_log_idx >= mctx->input.valid_len + && mctx->input.valid_len < mctx->input.len)) + { + reg_errcode_t err; + err = extend_buffers (mctx); + if (BE (err != REG_NOERROR, 0)) + return err; + } + + if (top < next_state_log_idx) + { + memset (mctx->state_log + top + 1, '\0', + sizeof (re_dfastate_t *) * (next_state_log_idx - top)); + mctx->state_log_top = next_state_log_idx; + } + return REG_NOERROR; +} + +static reg_errcode_t +internal_function +merge_state_array (re_dfa_t *dfa, re_dfastate_t **dst, re_dfastate_t **src, + Idx num) +{ + Idx st_idx; + reg_errcode_t err; + for (st_idx = 0; st_idx < num; ++st_idx) + { + if (dst[st_idx] == NULL) + dst[st_idx] = src[st_idx]; + else if (src[st_idx] != NULL) + { + re_node_set merged_set; + err = re_node_set_init_union (&merged_set, &dst[st_idx]->nodes, + &src[st_idx]->nodes); + if (BE (err != REG_NOERROR, 0)) + return err; + dst[st_idx] = re_acquire_state (&err, dfa, &merged_set); + re_node_set_free (&merged_set); + if (BE (err != REG_NOERROR, 0)) + return err; + } + } + return REG_NOERROR; +} + +static reg_errcode_t +internal_function +update_cur_sifted_state (re_match_context_t *mctx, re_sift_context_t *sctx, + Idx str_idx, re_node_set *dest_nodes) +{ + re_dfa_t *const dfa = mctx->dfa; + reg_errcode_t err; + const re_node_set *candidates; + candidates = ((mctx->state_log[str_idx] == NULL) ? NULL + : &mctx->state_log[str_idx]->nodes); + + if (dest_nodes->nelem == 0) + sctx->sifted_states[str_idx] = NULL; + else + { + if (candidates) + { + /* At first, add the nodes which can epsilon transit to a node in + DEST_NODE. */ + err = add_epsilon_src_nodes (dfa, dest_nodes, candidates); + if (BE (err != REG_NOERROR, 0)) + return err; + + /* Then, check the limitations in the current sift_context. */ + if (sctx->limits.nelem) + { + err = check_subexp_limits (dfa, dest_nodes, candidates, &sctx->limits, + mctx->bkref_ents, str_idx); + if (BE (err != REG_NOERROR, 0)) + return err; + } + } + + sctx->sifted_states[str_idx] = re_acquire_state (&err, dfa, dest_nodes); + if (BE (err != REG_NOERROR, 0)) + return err; + } + + if (candidates && mctx->state_log[str_idx]->has_backref) + { + err = sift_states_bkref (mctx, sctx, str_idx, candidates); + if (BE (err != REG_NOERROR, 0)) + return err; + } + return REG_NOERROR; +} + +static reg_errcode_t +internal_function +add_epsilon_src_nodes (re_dfa_t *dfa, re_node_set *dest_nodes, + const re_node_set *candidates) +{ + reg_errcode_t err = REG_NOERROR; + Idx i; + + re_dfastate_t *state = re_acquire_state (&err, dfa, dest_nodes); + if (BE (err != REG_NOERROR, 0)) + return err; + + if (!state->inveclosure.alloc) + { + err = re_node_set_alloc (&state->inveclosure, dest_nodes->nelem); + if (BE (err != REG_NOERROR, 0)) + return REG_ESPACE; + for (i = 0; i < dest_nodes->nelem; i++) + re_node_set_merge (&state->inveclosure, + dfa->inveclosures + dest_nodes->elems[i]); + } + return re_node_set_add_intersect (dest_nodes, candidates, + &state->inveclosure); +} + +static reg_errcode_t +internal_function +sub_epsilon_src_nodes (re_dfa_t *dfa, Idx node, re_node_set *dest_nodes, + const re_node_set *candidates) +{ + Idx ecl_idx; + reg_errcode_t err; + re_node_set *inv_eclosure = dfa->inveclosures + node; + re_node_set except_nodes; + re_node_set_init_empty (&except_nodes); + for (ecl_idx = 0; ecl_idx < inv_eclosure->nelem; ++ecl_idx) + { + Idx cur_node = inv_eclosure->elems[ecl_idx]; + if (cur_node == node) + continue; + if (IS_EPSILON_NODE (dfa->nodes[cur_node].type)) + { + Idx edst1 = dfa->edests[cur_node].elems[0]; + Idx edst2 = ((dfa->edests[cur_node].nelem > 1) + ? dfa->edests[cur_node].elems[1] : REG_MISSING); + if ((!re_node_set_contains (inv_eclosure, edst1) + && re_node_set_contains (dest_nodes, edst1)) + || (REG_VALID_NONZERO_INDEX (edst2) + && !re_node_set_contains (inv_eclosure, edst2) + && re_node_set_contains (dest_nodes, edst2))) + { + err = re_node_set_add_intersect (&except_nodes, candidates, + dfa->inveclosures + cur_node); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&except_nodes); + return err; + } + } + } + } + for (ecl_idx = 0; ecl_idx < inv_eclosure->nelem; ++ecl_idx) + { + Idx cur_node = inv_eclosure->elems[ecl_idx]; + if (!re_node_set_contains (&except_nodes, cur_node)) + { + Idx idx = re_node_set_contains (dest_nodes, cur_node) - 1; + re_node_set_remove_at (dest_nodes, idx); + } + } + re_node_set_free (&except_nodes); + return REG_NOERROR; +} + +static bool +internal_function +check_dst_limits (const re_match_context_t *mctx, const re_node_set *limits, + Idx dst_node, Idx dst_idx, Idx src_node, Idx src_idx) +{ + re_dfa_t *const dfa = mctx->dfa; + Idx lim_idx, src_pos, dst_pos; + + Idx dst_bkref_idx = search_cur_bkref_entry (mctx, dst_idx); + Idx src_bkref_idx = search_cur_bkref_entry (mctx, src_idx); + for (lim_idx = 0; lim_idx < limits->nelem; ++lim_idx) + { + Idx subexp_idx; + struct re_backref_cache_entry *ent; + ent = mctx->bkref_ents + limits->elems[lim_idx]; + subexp_idx = dfa->nodes[ent->node].opr.idx; + + dst_pos = check_dst_limits_calc_pos (mctx, limits->elems[lim_idx], + subexp_idx, dst_node, dst_idx, + dst_bkref_idx); + src_pos = check_dst_limits_calc_pos (mctx, limits->elems[lim_idx], + subexp_idx, src_node, src_idx, + src_bkref_idx); + + /* In case of: + <src> <dst> ( <subexp> ) + ( <subexp> ) <src> <dst> + ( <subexp1> <src> <subexp2> <dst> <subexp3> ) */ + if (src_pos == dst_pos) + continue; /* This is unrelated limitation. */ + else + return true; + } + return false; +} + +static int +internal_function +check_dst_limits_calc_pos_1 (const re_match_context_t *mctx, int boundaries, + Idx subexp_idx, Idx from_node, Idx bkref_idx) +{ + re_dfa_t *const dfa = mctx->dfa; + re_node_set *eclosures = dfa->eclosures + from_node; + Idx node_idx; + + /* Else, we are on the boundary: examine the nodes on the epsilon + closure. */ + for (node_idx = 0; node_idx < eclosures->nelem; ++node_idx) + { + Idx node = eclosures->elems[node_idx]; + switch (dfa->nodes[node].type) + { + case OP_BACK_REF: + if (bkref_idx != REG_MISSING) + { + struct re_backref_cache_entry *ent = mctx->bkref_ents + bkref_idx; + do + { + Idx dst; + int cpos; + + if (ent->node != node) + continue; + + if (subexp_idx < BITSET_WORD_BITS + && !(ent->eps_reachable_subexps_map + & ((bitset_word) 1 << subexp_idx))) + continue; + + /* Recurse trying to reach the OP_OPEN_SUBEXP and + OP_CLOSE_SUBEXP cases below. But, if the + destination node is the same node as the source + node, don't recurse because it would cause an + infinite loop: a regex that exhibits this behavior + is ()\1*\1* */ + dst = dfa->edests[node].elems[0]; + if (dst == from_node) + { + if (boundaries & 1) + return -1; + else /* if (boundaries & 2) */ + return 0; + } + + cpos = + check_dst_limits_calc_pos_1 (mctx, boundaries, subexp_idx, + dst, bkref_idx); + if (cpos == -1 /* && (boundaries & 1) */) + return -1; + if (cpos == 0 && (boundaries & 2)) + return 0; + + if (subexp_idx < BITSET_WORD_BITS) + ent->eps_reachable_subexps_map &= + ~ ((bitset_word) 1 << subexp_idx); + } + while (ent++->more); + } + break; + + case OP_OPEN_SUBEXP: + if ((boundaries & 1) && subexp_idx == dfa->nodes[node].opr.idx) + return -1; + break; + + case OP_CLOSE_SUBEXP: + if ((boundaries & 2) && subexp_idx == dfa->nodes[node].opr.idx) + return 0; + break; + + default: + break; + } + } + + return (boundaries & 2) ? 1 : 0; +} + +static int +internal_function +check_dst_limits_calc_pos (const re_match_context_t *mctx, + Idx limit, Idx subexp_idx, + Idx from_node, Idx str_idx, Idx bkref_idx) +{ + struct re_backref_cache_entry *lim = mctx->bkref_ents + limit; + int boundaries; + + /* If we are outside the range of the subexpression, return -1 or 1. */ + if (str_idx < lim->subexp_from) + return -1; + + if (lim->subexp_to < str_idx) + return 1; + + /* If we are within the subexpression, return 0. */ + boundaries = (str_idx == lim->subexp_from); + boundaries |= (str_idx == lim->subexp_to) << 1; + if (boundaries == 0) + return 0; + + /* Else, examine epsilon closure. */ + return check_dst_limits_calc_pos_1 (mctx, boundaries, subexp_idx, + from_node, bkref_idx); +} + +/* Check the limitations of sub expressions LIMITS, and remove the nodes + which are against limitations from DEST_NODES. */ + +static reg_errcode_t +internal_function +check_subexp_limits (re_dfa_t *dfa, re_node_set *dest_nodes, + const re_node_set *candidates, re_node_set *limits, + struct re_backref_cache_entry *bkref_ents, Idx str_idx) +{ + reg_errcode_t err; + Idx node_idx, lim_idx; + + for (lim_idx = 0; lim_idx < limits->nelem; ++lim_idx) + { + Idx subexp_idx; + struct re_backref_cache_entry *ent; + ent = bkref_ents + limits->elems[lim_idx]; + + if (str_idx <= ent->subexp_from || ent->str_idx < str_idx) + continue; /* This is unrelated limitation. */ + + subexp_idx = dfa->nodes[ent->node].opr.idx; + if (ent->subexp_to == str_idx) + { + Idx ops_node = REG_MISSING; + Idx cls_node = REG_MISSING; + for (node_idx = 0; node_idx < dest_nodes->nelem; ++node_idx) + { + Idx node = dest_nodes->elems[node_idx]; + re_token_type_t type = dfa->nodes[node].type; + if (type == OP_OPEN_SUBEXP + && subexp_idx == dfa->nodes[node].opr.idx) + ops_node = node; + else if (type == OP_CLOSE_SUBEXP + && subexp_idx == dfa->nodes[node].opr.idx) + cls_node = node; + } + + /* Check the limitation of the open subexpression. */ + /* Note that (ent->subexp_to = str_idx != ent->subexp_from). */ + if (REG_VALID_INDEX (ops_node)) + { + err = sub_epsilon_src_nodes (dfa, ops_node, dest_nodes, + candidates); + if (BE (err != REG_NOERROR, 0)) + return err; + } + + /* Check the limitation of the close subexpression. */ + if (REG_VALID_INDEX (cls_node)) + for (node_idx = 0; node_idx < dest_nodes->nelem; ++node_idx) + { + Idx node = dest_nodes->elems[node_idx]; + if (!re_node_set_contains (dfa->inveclosures + node, + cls_node) + && !re_node_set_contains (dfa->eclosures + node, + cls_node)) + { + /* It is against this limitation. + Remove it form the current sifted state. */ + err = sub_epsilon_src_nodes (dfa, node, dest_nodes, + candidates); + if (BE (err != REG_NOERROR, 0)) + return err; + --node_idx; + } + } + } + else /* (ent->subexp_to != str_idx) */ + { + for (node_idx = 0; node_idx < dest_nodes->nelem; ++node_idx) + { + Idx node = dest_nodes->elems[node_idx]; + re_token_type_t type = dfa->nodes[node].type; + if (type == OP_CLOSE_SUBEXP || type == OP_OPEN_SUBEXP) + { + if (subexp_idx != dfa->nodes[node].opr.idx) + continue; + /* It is against this limitation. + Remove it form the current sifted state. */ + err = sub_epsilon_src_nodes (dfa, node, dest_nodes, + candidates); + if (BE (err != REG_NOERROR, 0)) + return err; + } + } + } + } + return REG_NOERROR; +} + +static reg_errcode_t +internal_function +sift_states_bkref (re_match_context_t *mctx, re_sift_context_t *sctx, + Idx str_idx, const re_node_set *candidates) +{ + re_dfa_t *const dfa = mctx->dfa; + reg_errcode_t err; + Idx node_idx, node; + re_sift_context_t local_sctx; + Idx first_idx = search_cur_bkref_entry (mctx, str_idx); + + if (first_idx == REG_MISSING) + return REG_NOERROR; + + local_sctx.sifted_states = NULL; /* Mark that it hasn't been initialized. */ + + for (node_idx = 0; node_idx < candidates->nelem; ++node_idx) + { + Idx enabled_idx; + re_token_type_t type; + struct re_backref_cache_entry *entry; + node = candidates->elems[node_idx]; + type = dfa->nodes[node].type; + /* Avoid infinite loop for the REs like "()\1+". */ + if (node == sctx->last_node && str_idx == sctx->last_str_idx) + continue; + if (type != OP_BACK_REF) + continue; + + entry = mctx->bkref_ents + first_idx; + enabled_idx = first_idx; + do + { + bool ok; + Idx subexp_len, to_idx, dst_node; + re_dfastate_t *cur_state; + + if (entry->node != node) + continue; + subexp_len = entry->subexp_to - entry->subexp_from; + to_idx = str_idx + subexp_len; + dst_node = (subexp_len ? dfa->nexts[node] + : dfa->edests[node].elems[0]); + + if (to_idx > sctx->last_str_idx + || sctx->sifted_states[to_idx] == NULL + || !STATE_NODE_CONTAINS (sctx->sifted_states[to_idx], dst_node) + || check_dst_limits (mctx, &sctx->limits, node, + str_idx, dst_node, to_idx)) + continue; + + if (local_sctx.sifted_states == NULL) + { + local_sctx = *sctx; + err = re_node_set_init_copy (&local_sctx.limits, &sctx->limits); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + } + local_sctx.last_node = node; + local_sctx.last_str_idx = str_idx; + ok = re_node_set_insert (&local_sctx.limits, enabled_idx); + if (BE (! ok, 0)) + { + err = REG_ESPACE; + goto free_return; + } + cur_state = local_sctx.sifted_states[str_idx]; + err = sift_states_backward (mctx, &local_sctx); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + if (sctx->limited_states != NULL) + { + err = merge_state_array (dfa, sctx->limited_states, + local_sctx.sifted_states, + str_idx + 1); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + } + local_sctx.sifted_states[str_idx] = cur_state; + re_node_set_remove (&local_sctx.limits, enabled_idx); + + /* mctx->bkref_ents may have changed, reload the pointer. */ + entry = mctx->bkref_ents + enabled_idx; + } + while (enabled_idx++, entry++->more); + } + err = REG_NOERROR; + free_return: + if (local_sctx.sifted_states != NULL) + { + re_node_set_free (&local_sctx.limits); + } + + return err; +} + + +#ifdef RE_ENABLE_I18N +static int +internal_function +sift_states_iter_mb (const re_match_context_t *mctx, re_sift_context_t *sctx, + Idx node_idx, Idx str_idx, Idx max_str_idx) +{ + re_dfa_t *const dfa = mctx->dfa; + int naccepted; + /* Check the node can accept `multi byte'. */ + naccepted = check_node_accept_bytes (dfa, node_idx, &mctx->input, str_idx); + if (naccepted > 0 && str_idx + naccepted <= max_str_idx && + !STATE_NODE_CONTAINS (sctx->sifted_states[str_idx + naccepted], + dfa->nexts[node_idx])) + /* The node can't accept the `multi byte', or the + destination was already thrown away, then the node + could't accept the current input `multi byte'. */ + naccepted = 0; + /* Otherwise, it is sure that the node could accept + `naccepted' bytes input. */ + return naccepted; +} +#endif /* RE_ENABLE_I18N */ + + +/* Functions for state transition. */ + +/* Return the next state to which the current state STATE will transit by + accepting the current input byte, and update STATE_LOG if necessary. + If STATE can accept a multibyte char/collating element/back reference + update the destination of STATE_LOG. */ + +static re_dfastate_t * +internal_function +transit_state (reg_errcode_t *err, re_match_context_t *mctx, + re_dfastate_t *state) +{ + re_dfastate_t **trtable; + unsigned char ch; + +#ifdef RE_ENABLE_I18N + /* If the current state can accept multibyte. */ + if (BE (state->accept_mb, 0)) + { + *err = transit_state_mb (mctx, state); + if (BE (*err != REG_NOERROR, 0)) + return NULL; + } +#endif /* RE_ENABLE_I18N */ + + /* Then decide the next state with the single byte. */ +#if 0 + if (0) + /* don't use transition table */ + return transit_state_sb (err, mctx, state); +#endif + + /* Use transition table */ + ch = re_string_fetch_byte (&mctx->input); + for (;;) + { + trtable = state->trtable; + if (BE (trtable != NULL, 1)) + return trtable[ch]; + + trtable = state->word_trtable; + if (BE (trtable != NULL, 1)) + { + unsigned int context; + context + = re_string_context_at (&mctx->input, + re_string_cur_idx (&mctx->input) - 1, + mctx->eflags); + if (IS_WORD_CONTEXT (context)) + return trtable[ch + SBC_MAX]; + else + return trtable[ch]; + } + + if (!build_trtable (mctx->dfa, state)) + { + *err = REG_ESPACE; + return NULL; + } + + /* Retry, we now have a transition table. */ + } +} + +/* Update the state_log if we need */ +re_dfastate_t * +internal_function +merge_state_with_log (reg_errcode_t *err, re_match_context_t *mctx, + re_dfastate_t *next_state) +{ + re_dfa_t *const dfa = mctx->dfa; + Idx cur_idx = re_string_cur_idx (&mctx->input); + + if (cur_idx > mctx->state_log_top) + { + mctx->state_log[cur_idx] = next_state; + mctx->state_log_top = cur_idx; + } + else if (mctx->state_log[cur_idx] == 0) + { + mctx->state_log[cur_idx] = next_state; + } + else + { + re_dfastate_t *pstate; + unsigned int context; + re_node_set next_nodes, *log_nodes, *table_nodes = NULL; + /* If (state_log[cur_idx] != 0), it implies that cur_idx is + the destination of a multibyte char/collating element/ + back reference. Then the next state is the union set of + these destinations and the results of the transition table. */ + pstate = mctx->state_log[cur_idx]; + log_nodes = pstate->entrance_nodes; + if (next_state != NULL) + { + table_nodes = next_state->entrance_nodes; + *err = re_node_set_init_union (&next_nodes, table_nodes, + log_nodes); + if (BE (*err != REG_NOERROR, 0)) + return NULL; + } + else + next_nodes = *log_nodes; + /* Note: We already add the nodes of the initial state, + then we don't need to add them here. */ + + context = re_string_context_at (&mctx->input, + re_string_cur_idx (&mctx->input) - 1, + mctx->eflags); + next_state = mctx->state_log[cur_idx] + = re_acquire_state_context (err, dfa, &next_nodes, context); + /* We don't need to check errors here, since the return value of + this function is next_state and ERR is already set. */ + + if (table_nodes != NULL) + re_node_set_free (&next_nodes); + } + + if (BE (dfa->nbackref, 0) && next_state != NULL) + { + /* Check OP_OPEN_SUBEXP in the current state in case that we use them + later. We must check them here, since the back references in the + next state might use them. */ + *err = check_subexp_matching_top (mctx, &next_state->nodes, + cur_idx); + if (BE (*err != REG_NOERROR, 0)) + return NULL; + + /* If the next state has back references. */ + if (next_state->has_backref) + { + *err = transit_state_bkref (mctx, &next_state->nodes); + if (BE (*err != REG_NOERROR, 0)) + return NULL; + next_state = mctx->state_log[cur_idx]; + } + } + + return next_state; +} + +/* Skip bytes in the input that correspond to part of a + multi-byte match, then look in the log for a state + from which to restart matching. */ +static re_dfastate_t * +internal_function +find_recover_state (reg_errcode_t *err, re_match_context_t *mctx) +{ + re_dfastate_t *cur_state = NULL; + do + { + Idx max = mctx->state_log_top; + Idx cur_str_idx = re_string_cur_idx (&mctx->input); + + do + { + if (++cur_str_idx > max) + return NULL; + re_string_skip_bytes (&mctx->input, 1); + } + while (mctx->state_log[cur_str_idx] == NULL); + + cur_state = merge_state_with_log (err, mctx, NULL); + } + while (*err == REG_NOERROR && cur_state == NULL); + return cur_state; +} + +/* Helper functions for transit_state. */ + +/* From the node set CUR_NODES, pick up the nodes whose types are + OP_OPEN_SUBEXP and which have corresponding back references in the regular + expression. And register them to use them later for evaluating the + correspoding back references. */ + +static reg_errcode_t +internal_function +check_subexp_matching_top (re_match_context_t *mctx, re_node_set *cur_nodes, + Idx str_idx) +{ + re_dfa_t *const dfa = mctx->dfa; + Idx node_idx; + reg_errcode_t err; + + /* TODO: This isn't efficient. + Because there might be more than one nodes whose types are + OP_OPEN_SUBEXP and whose index is SUBEXP_IDX, we must check all + nodes. + E.g. RE: (a){2} */ + for (node_idx = 0; node_idx < cur_nodes->nelem; ++node_idx) + { + Idx node = cur_nodes->elems[node_idx]; + if (dfa->nodes[node].type == OP_OPEN_SUBEXP + && dfa->nodes[node].opr.idx < BITSET_WORD_BITS + && (dfa->used_bkref_map + & ((bitset_word) 1 << dfa->nodes[node].opr.idx))) + { + err = match_ctx_add_subtop (mctx, node, str_idx); + if (BE (err != REG_NOERROR, 0)) + return err; + } + } + return REG_NOERROR; +} + +#if 0 +/* Return the next state to which the current state STATE will transit by + accepting the current input byte. */ + +static re_dfastate_t * +transit_state_sb (reg_errcode_t *err, re_match_context_t *mctx, + re_dfastate_t *state) +{ + re_dfa_t *const dfa = mctx->dfa; + re_node_set next_nodes; + re_dfastate_t *next_state; + Idx node_cnt, cur_str_idx = re_string_cur_idx (&mctx->input); + unsigned int context; + + *err = re_node_set_alloc (&next_nodes, state->nodes.nelem + 1); + if (BE (*err != REG_NOERROR, 0)) + return NULL; + for (node_cnt = 0; node_cnt < state->nodes.nelem; ++node_cnt) + { + Idx cur_node = state->nodes.elems[node_cnt]; + if (check_node_accept (mctx, dfa->nodes + cur_node, cur_str_idx)) + { + *err = re_node_set_merge (&next_nodes, + dfa->eclosures + dfa->nexts[cur_node]); + if (BE (*err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return NULL; + } + } + } + context = re_string_context_at (&mctx->input, cur_str_idx, mctx->eflags); + next_state = re_acquire_state_context (err, dfa, &next_nodes, context); + /* We don't need to check errors here, since the return value of + this function is next_state and ERR is already set. */ + + re_node_set_free (&next_nodes); + re_string_skip_bytes (&mctx->input, 1); + return next_state; +} +#endif + +#ifdef RE_ENABLE_I18N +static reg_errcode_t +internal_function +transit_state_mb (re_match_context_t *mctx, re_dfastate_t *pstate) +{ + re_dfa_t *const dfa = mctx->dfa; + reg_errcode_t err; + Idx i; + + for (i = 0; i < pstate->nodes.nelem; ++i) + { + re_node_set dest_nodes, *new_nodes; + Idx cur_node_idx = pstate->nodes.elems[i]; + int naccepted; + Idx dest_idx; + unsigned int context; + re_dfastate_t *dest_state; + + if (!dfa->nodes[cur_node_idx].accept_mb) + continue; + + if (dfa->nodes[cur_node_idx].constraint) + { + context = re_string_context_at (&mctx->input, + re_string_cur_idx (&mctx->input), + mctx->eflags); + if (NOT_SATISFY_NEXT_CONSTRAINT (dfa->nodes[cur_node_idx].constraint, + context)) + continue; + } + + /* How many bytes the node can accept? */ + naccepted = check_node_accept_bytes (dfa, cur_node_idx, &mctx->input, + re_string_cur_idx (&mctx->input)); + if (naccepted == 0) + continue; + + /* The node can accepts `naccepted' bytes. */ + dest_idx = re_string_cur_idx (&mctx->input) + naccepted; + mctx->max_mb_elem_len = ((mctx->max_mb_elem_len < naccepted) ? naccepted + : mctx->max_mb_elem_len); + err = clean_state_log_if_needed (mctx, dest_idx); + if (BE (err != REG_NOERROR, 0)) + return err; +#ifdef DEBUG + assert (dfa->nexts[cur_node_idx] != REG_MISSING); +#endif + new_nodes = dfa->eclosures + dfa->nexts[cur_node_idx]; + + dest_state = mctx->state_log[dest_idx]; + if (dest_state == NULL) + dest_nodes = *new_nodes; + else + { + err = re_node_set_init_union (&dest_nodes, + dest_state->entrance_nodes, new_nodes); + if (BE (err != REG_NOERROR, 0)) + return err; + } + context = re_string_context_at (&mctx->input, dest_idx - 1, mctx->eflags); + mctx->state_log[dest_idx] + = re_acquire_state_context (&err, dfa, &dest_nodes, context); + if (dest_state != NULL) + re_node_set_free (&dest_nodes); + if (BE (mctx->state_log[dest_idx] == NULL && err != REG_NOERROR, 0)) + return err; + } + return REG_NOERROR; +} +#endif /* RE_ENABLE_I18N */ + +static reg_errcode_t +internal_function +transit_state_bkref (re_match_context_t *mctx, const re_node_set *nodes) +{ + re_dfa_t *const dfa = mctx->dfa; + reg_errcode_t err; + Idx i; + Idx cur_str_idx = re_string_cur_idx (&mctx->input); + + for (i = 0; i < nodes->nelem; ++i) + { + Idx dest_str_idx, prev_nelem, bkc_idx; + Idx node_idx = nodes->elems[i]; + unsigned int context; + const re_token_t *node = dfa->nodes + node_idx; + re_node_set *new_dest_nodes; + + /* Check whether `node' is a backreference or not. */ + if (node->type != OP_BACK_REF) + continue; + + if (node->constraint) + { + context = re_string_context_at (&mctx->input, cur_str_idx, + mctx->eflags); + if (NOT_SATISFY_NEXT_CONSTRAINT (node->constraint, context)) + continue; + } + + /* `node' is a backreference. + Check the substring which the substring matched. */ + bkc_idx = mctx->nbkref_ents; + err = get_subexp (mctx, node_idx, cur_str_idx); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + + /* And add the epsilon closures (which is `new_dest_nodes') of + the backreference to appropriate state_log. */ +#ifdef DEBUG + assert (dfa->nexts[node_idx] != REG_MISSING); +#endif + for (; bkc_idx < mctx->nbkref_ents; ++bkc_idx) + { + Idx subexp_len; + re_dfastate_t *dest_state; + struct re_backref_cache_entry *bkref_ent; + bkref_ent = mctx->bkref_ents + bkc_idx; + if (bkref_ent->node != node_idx || bkref_ent->str_idx != cur_str_idx) + continue; + subexp_len = bkref_ent->subexp_to - bkref_ent->subexp_from; + new_dest_nodes = (subexp_len == 0 + ? dfa->eclosures + dfa->edests[node_idx].elems[0] + : dfa->eclosures + dfa->nexts[node_idx]); + dest_str_idx = (cur_str_idx + bkref_ent->subexp_to + - bkref_ent->subexp_from); + context = re_string_context_at (&mctx->input, dest_str_idx - 1, + mctx->eflags); + dest_state = mctx->state_log[dest_str_idx]; + prev_nelem = ((mctx->state_log[cur_str_idx] == NULL) ? 0 + : mctx->state_log[cur_str_idx]->nodes.nelem); + /* Add `new_dest_node' to state_log. */ + if (dest_state == NULL) + { + mctx->state_log[dest_str_idx] + = re_acquire_state_context (&err, dfa, new_dest_nodes, + context); + if (BE (mctx->state_log[dest_str_idx] == NULL + && err != REG_NOERROR, 0)) + goto free_return; + } + else + { + re_node_set dest_nodes; + err = re_node_set_init_union (&dest_nodes, + dest_state->entrance_nodes, + new_dest_nodes); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&dest_nodes); + goto free_return; + } + mctx->state_log[dest_str_idx] + = re_acquire_state_context (&err, dfa, &dest_nodes, context); + re_node_set_free (&dest_nodes); + if (BE (mctx->state_log[dest_str_idx] == NULL + && err != REG_NOERROR, 0)) + goto free_return; + } + /* We need to check recursively if the backreference can epsilon + transit. */ + if (subexp_len == 0 + && mctx->state_log[cur_str_idx]->nodes.nelem > prev_nelem) + { + err = check_subexp_matching_top (mctx, new_dest_nodes, + cur_str_idx); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + err = transit_state_bkref (mctx, new_dest_nodes); + if (BE (err != REG_NOERROR, 0)) + goto free_return; + } + } + } + err = REG_NOERROR; + free_return: + return err; +} + +/* Enumerate all the candidates which the backreference BKREF_NODE can match + at BKREF_STR_IDX, and register them by match_ctx_add_entry(). + Note that we might collect inappropriate candidates here. + However, the cost of checking them strictly here is too high, then we + delay these checking for prune_impossible_nodes(). */ + +static reg_errcode_t +internal_function +get_subexp (re_match_context_t *mctx, Idx bkref_node, Idx bkref_str_idx) +{ + re_dfa_t *const dfa = mctx->dfa; + Idx subexp_num, sub_top_idx; + const char *buf = (const char *) re_string_get_buffer (&mctx->input); + /* Return if we have already checked BKREF_NODE at BKREF_STR_IDX. */ + Idx cache_idx = search_cur_bkref_entry (mctx, bkref_str_idx); + if (cache_idx != REG_MISSING) + { + const struct re_backref_cache_entry *entry = mctx->bkref_ents + cache_idx; + do + if (entry->node == bkref_node) + return REG_NOERROR; /* We already checked it. */ + while (entry++->more); + } + + subexp_num = dfa->nodes[bkref_node].opr.idx; + + /* For each sub expression */ + for (sub_top_idx = 0; sub_top_idx < mctx->nsub_tops; ++sub_top_idx) + { + reg_errcode_t err; + re_sub_match_top_t *sub_top = mctx->sub_tops[sub_top_idx]; + re_sub_match_last_t *sub_last; + Idx sub_last_idx, sl_str, bkref_str_off; + + if (dfa->nodes[sub_top->node].opr.idx != subexp_num) + continue; /* It isn't related. */ + + sl_str = sub_top->str_idx; + bkref_str_off = bkref_str_idx; + /* At first, check the last node of sub expressions we already + evaluated. */ + for (sub_last_idx = 0; sub_last_idx < sub_top->nlasts; ++sub_last_idx) + { + regoff_t sl_str_diff; + sub_last = sub_top->lasts[sub_last_idx]; + sl_str_diff = sub_last->str_idx - sl_str; + /* The matched string by the sub expression match with the substring + at the back reference? */ + if (sl_str_diff > 0) + { + if (BE (bkref_str_off + sl_str_diff > mctx->input.valid_len, 0)) + { + /* Not enough chars for a successful match. */ + if (bkref_str_off + sl_str_diff > mctx->input.len) + break; + + err = clean_state_log_if_needed (mctx, + bkref_str_off + + sl_str_diff); + if (BE (err != REG_NOERROR, 0)) + return err; + buf = (const char *) re_string_get_buffer (&mctx->input); + } + if (memcmp (buf + bkref_str_off, buf + sl_str, sl_str_diff) != 0) + break; /* We don't need to search this sub expression any more. */ + } + bkref_str_off += sl_str_diff; + sl_str += sl_str_diff; + err = get_subexp_sub (mctx, sub_top, sub_last, bkref_node, + bkref_str_idx); + + /* Reload buf, since the preceding call might have reallocated + the buffer. */ + buf = (const char *) re_string_get_buffer (&mctx->input); + + if (err == REG_NOMATCH) + continue; + if (BE (err != REG_NOERROR, 0)) + return err; + } + + if (sub_last_idx < sub_top->nlasts) + continue; + if (sub_last_idx > 0) + ++sl_str; + /* Then, search for the other last nodes of the sub expression. */ + for (; sl_str <= bkref_str_idx; ++sl_str) + { + Idx cls_node; + regoff_t sl_str_off; + const re_node_set *nodes; + sl_str_off = sl_str - sub_top->str_idx; + /* The matched string by the sub expression match with the substring + at the back reference? */ + if (sl_str_off > 0) + { + if (BE (bkref_str_off >= mctx->input.valid_len, 0)) + { + /* If we are at the end of the input, we cannot match. */ + if (bkref_str_off >= mctx->input.len) + break; + + err = extend_buffers (mctx); + if (BE (err != REG_NOERROR, 0)) + return err; + + buf = (const char *) re_string_get_buffer (&mctx->input); + } + if (buf [bkref_str_off++] != buf[sl_str - 1]) + break; /* We don't need to search this sub expression + any more. */ + } + if (mctx->state_log[sl_str] == NULL) + continue; + /* Does this state have a ')' of the sub expression? */ + nodes = &mctx->state_log[sl_str]->nodes; + cls_node = find_subexp_node (dfa, nodes, subexp_num, OP_CLOSE_SUBEXP); + if (cls_node == REG_MISSING) + continue; /* No. */ + if (sub_top->path == NULL) + { + sub_top->path = re_calloc (state_array_t, + sl_str - sub_top->str_idx + 1); + if (sub_top->path == NULL) + return REG_ESPACE; + } + /* Can the OP_OPEN_SUBEXP node arrive the OP_CLOSE_SUBEXP node + in the current context? */ + err = check_arrival (mctx, sub_top->path, sub_top->node, + sub_top->str_idx, cls_node, sl_str, OP_CLOSE_SUBEXP); + if (err == REG_NOMATCH) + continue; + if (BE (err != REG_NOERROR, 0)) + return err; + sub_last = match_ctx_add_sublast (sub_top, cls_node, sl_str); + if (BE (sub_last == NULL, 0)) + return REG_ESPACE; + err = get_subexp_sub (mctx, sub_top, sub_last, bkref_node, + bkref_str_idx); + if (err == REG_NOMATCH) + continue; + } + } + return REG_NOERROR; +} + +/* Helper functions for get_subexp(). */ + +/* Check SUB_LAST can arrive to the back reference BKREF_NODE at BKREF_STR. + If it can arrive, register the sub expression expressed with SUB_TOP + and SUB_LAST. */ + +static reg_errcode_t +internal_function +get_subexp_sub (re_match_context_t *mctx, const re_sub_match_top_t *sub_top, + re_sub_match_last_t *sub_last, Idx bkref_node, Idx bkref_str) +{ + reg_errcode_t err; + Idx to_idx; + /* Can the subexpression arrive the back reference? */ + err = check_arrival (mctx, &sub_last->path, sub_last->node, + sub_last->str_idx, bkref_node, bkref_str, OP_OPEN_SUBEXP); + if (err != REG_NOERROR) + return err; + err = match_ctx_add_entry (mctx, bkref_node, bkref_str, sub_top->str_idx, + sub_last->str_idx); + if (BE (err != REG_NOERROR, 0)) + return err; + to_idx = bkref_str + sub_last->str_idx - sub_top->str_idx; + return clean_state_log_if_needed (mctx, to_idx); +} + +/* Find the first node which is '(' or ')' and whose index is SUBEXP_IDX. + Search '(' if FL_OPEN, or search ')' otherwise. + TODO: This function isn't efficient... + Because there might be more than one nodes whose types are + OP_OPEN_SUBEXP and whose index is SUBEXP_IDX, we must check all + nodes. + E.g. RE: (a){2} */ + +static Idx +internal_function +find_subexp_node (const re_dfa_t *dfa, const re_node_set *nodes, + Idx subexp_idx, int type) +{ + Idx cls_idx; + for (cls_idx = 0; cls_idx < nodes->nelem; ++cls_idx) + { + Idx cls_node = nodes->elems[cls_idx]; + const re_token_t *node = dfa->nodes + cls_node; + if (node->type == type + && node->opr.idx == subexp_idx) + return cls_node; + } + return REG_MISSING; +} + +/* Check whether the node TOP_NODE at TOP_STR can arrive to the node + LAST_NODE at LAST_STR. We record the path onto PATH since it will be + heavily reused. + Return REG_NOERROR if it can arrive, or REG_NOMATCH otherwise. */ + +static reg_errcode_t +internal_function +check_arrival (re_match_context_t *mctx, state_array_t *path, + Idx top_node, Idx top_str, Idx last_node, Idx last_str, + int type) +{ + re_dfa_t *const dfa = mctx->dfa; + reg_errcode_t err; + Idx subexp_num, backup_cur_idx, str_idx, null_cnt; + re_dfastate_t *cur_state = NULL; + re_node_set *cur_nodes, next_nodes; + re_dfastate_t **backup_state_log; + unsigned int context; + + subexp_num = dfa->nodes[top_node].opr.idx; + /* Extend the buffer if we need. */ + if (BE (path->alloc < last_str + mctx->max_mb_elem_len + 1, 0)) + { + re_dfastate_t **new_array; + Idx old_alloc = path->alloc; + Idx new_alloc = old_alloc + last_str + mctx->max_mb_elem_len + 1; + if (BE (new_alloc < old_alloc, 0)) + return REG_ESPACE; + new_array = re_xrealloc (path->array, re_dfastate_t *, new_alloc); + if (BE (new_array == NULL, 0)) + return REG_ESPACE; + path->array = new_array; + path->alloc = new_alloc; + memset (new_array + old_alloc, '\0', + sizeof (re_dfastate_t *) * (new_alloc - old_alloc)); + } + + str_idx = path->next_idx == 0 ? top_str : path->next_idx; + + /* Temporary modify MCTX. */ + backup_state_log = mctx->state_log; + backup_cur_idx = mctx->input.cur_idx; + mctx->state_log = path->array; + mctx->input.cur_idx = str_idx; + + /* Setup initial node set. */ + context = re_string_context_at (&mctx->input, str_idx - 1, mctx->eflags); + if (str_idx == top_str) + { + err = re_node_set_init_1 (&next_nodes, top_node); + if (BE (err != REG_NOERROR, 0)) + return err; + err = check_arrival_expand_ecl (dfa, &next_nodes, subexp_num, type); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return err; + } + } + else + { + cur_state = mctx->state_log[str_idx]; + if (cur_state && cur_state->has_backref) + { + err = re_node_set_init_copy (&next_nodes, &cur_state->nodes); + if (BE ( err != REG_NOERROR, 0)) + return err; + } + else + re_node_set_init_empty (&next_nodes); + } + if (str_idx == top_str || (cur_state && cur_state->has_backref)) + { + if (next_nodes.nelem) + { + err = expand_bkref_cache (mctx, &next_nodes, str_idx, + subexp_num, type); + if (BE ( err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return err; + } + } + cur_state = re_acquire_state_context (&err, dfa, &next_nodes, context); + if (BE (cur_state == NULL && err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return err; + } + mctx->state_log[str_idx] = cur_state; + } + + for (null_cnt = 0; str_idx < last_str && null_cnt <= mctx->max_mb_elem_len;) + { + re_node_set_empty (&next_nodes); + if (mctx->state_log[str_idx + 1]) + { + err = re_node_set_merge (&next_nodes, + &mctx->state_log[str_idx + 1]->nodes); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return err; + } + } + if (cur_state) + { + err = check_arrival_add_next_nodes (mctx, str_idx, + &cur_state->non_eps_nodes, &next_nodes); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return err; + } + } + ++str_idx; + if (next_nodes.nelem) + { + err = check_arrival_expand_ecl (dfa, &next_nodes, subexp_num, type); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return err; + } + err = expand_bkref_cache (mctx, &next_nodes, str_idx, + subexp_num, type); + if (BE ( err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return err; + } + } + context = re_string_context_at (&mctx->input, str_idx - 1, mctx->eflags); + cur_state = re_acquire_state_context (&err, dfa, &next_nodes, context); + if (BE (cur_state == NULL && err != REG_NOERROR, 0)) + { + re_node_set_free (&next_nodes); + return err; + } + mctx->state_log[str_idx] = cur_state; + null_cnt = cur_state == NULL ? null_cnt + 1 : 0; + } + re_node_set_free (&next_nodes); + cur_nodes = (mctx->state_log[last_str] == NULL ? NULL + : &mctx->state_log[last_str]->nodes); + path->next_idx = str_idx; + + /* Fix MCTX. */ + mctx->state_log = backup_state_log; + mctx->input.cur_idx = backup_cur_idx; + + /* Then check the current node set has the node LAST_NODE. */ + if (cur_nodes != NULL && re_node_set_contains (cur_nodes, last_node)) + return REG_NOERROR; + + return REG_NOMATCH; +} + +/* Helper functions for check_arrival. */ + +/* Calculate the destination nodes of CUR_NODES at STR_IDX, and append them + to NEXT_NODES. + TODO: This function is similar to the functions transit_state*(), + however this function has many additional works. + Can't we unify them? */ + +static reg_errcode_t +internal_function +check_arrival_add_next_nodes (re_match_context_t *mctx, Idx str_idx, + re_node_set *cur_nodes, + re_node_set *next_nodes) +{ + re_dfa_t *const dfa = mctx->dfa; + bool ok; + Idx cur_idx; + reg_errcode_t err; + re_node_set union_set; + re_node_set_init_empty (&union_set); + for (cur_idx = 0; cur_idx < cur_nodes->nelem; ++cur_idx) + { + int naccepted = 0; + Idx cur_node = cur_nodes->elems[cur_idx]; +#ifdef DEBUG + re_token_type_t type = dfa->nodes[cur_node].type; + assert (!IS_EPSILON_NODE (type)); +#endif +#ifdef RE_ENABLE_I18N + /* If the node may accept `multi byte'. */ + if (dfa->nodes[cur_node].accept_mb) + { + naccepted = check_node_accept_bytes (dfa, cur_node, &mctx->input, + str_idx); + if (naccepted > 1) + { + re_dfastate_t *dest_state; + Idx next_node = dfa->nexts[cur_node]; + Idx next_idx = str_idx + naccepted; + dest_state = mctx->state_log[next_idx]; + re_node_set_empty (&union_set); + if (dest_state) + { + err = re_node_set_merge (&union_set, &dest_state->nodes); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&union_set); + return err; + } + } + ok = re_node_set_insert (&union_set, next_node); + if (BE (! ok, 0)) + { + re_node_set_free (&union_set); + return REG_ESPACE; + } + mctx->state_log[next_idx] = re_acquire_state (&err, dfa, + &union_set); + if (BE (mctx->state_log[next_idx] == NULL + && err != REG_NOERROR, 0)) + { + re_node_set_free (&union_set); + return err; + } + } + } +#endif /* RE_ENABLE_I18N */ + if (naccepted + || check_node_accept (mctx, dfa->nodes + cur_node, str_idx)) + { + ok = re_node_set_insert (next_nodes, dfa->nexts[cur_node]); + if (BE (! ok, 0)) + { + re_node_set_free (&union_set); + return REG_ESPACE; + } + } + } + re_node_set_free (&union_set); + return REG_NOERROR; +} + +/* For all the nodes in CUR_NODES, add the epsilon closures of them to + CUR_NODES, however exclude the nodes which are: + - inside the sub expression whose number is EX_SUBEXP, if FL_OPEN. + - out of the sub expression whose number is EX_SUBEXP, if !FL_OPEN. +*/ + +static reg_errcode_t +internal_function +check_arrival_expand_ecl (re_dfa_t *dfa, re_node_set *cur_nodes, + Idx ex_subexp, int type) +{ + reg_errcode_t err; + Idx idx, outside_node; + re_node_set new_nodes; +#ifdef DEBUG + assert (cur_nodes->nelem); +#endif + err = re_node_set_alloc (&new_nodes, cur_nodes->nelem); + if (BE (err != REG_NOERROR, 0)) + return err; + /* Create a new node set NEW_NODES with the nodes which are epsilon + closures of the node in CUR_NODES. */ + + for (idx = 0; idx < cur_nodes->nelem; ++idx) + { + Idx cur_node = cur_nodes->elems[idx]; + re_node_set *eclosure = dfa->eclosures + cur_node; + outside_node = find_subexp_node (dfa, eclosure, ex_subexp, type); + if (outside_node == REG_MISSING) + { + /* There are no problematic nodes, just merge them. */ + err = re_node_set_merge (&new_nodes, eclosure); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&new_nodes); + return err; + } + } + else + { + /* There are problematic nodes, re-calculate incrementally. */ + err = check_arrival_expand_ecl_sub (dfa, &new_nodes, cur_node, + ex_subexp, type); + if (BE (err != REG_NOERROR, 0)) + { + re_node_set_free (&new_nodes); + return err; + } + } + } + re_node_set_free (cur_nodes); + *cur_nodes = new_nodes; + return REG_NOERROR; +} + +/* Helper function for check_arrival_expand_ecl. + Check incrementally the epsilon closure of TARGET, and if it isn't + problematic append it to DST_NODES. */ + +static reg_errcode_t +internal_function +check_arrival_expand_ecl_sub (re_dfa_t *dfa, re_node_set *dst_nodes, + Idx target, Idx ex_subexp, int type) +{ + Idx cur_node; + for (cur_node = target; !re_node_set_contains (dst_nodes, cur_node);) + { + bool ok; + + if (dfa->nodes[cur_node].type == type + && dfa->nodes[cur_node].opr.idx == ex_subexp) + { + if (type == OP_CLOSE_SUBEXP) + { + ok = re_node_set_insert (dst_nodes, cur_node); + if (BE (! ok, 0)) + return REG_ESPACE; + } + break; + } + ok = re_node_set_insert (dst_nodes, cur_node); + if (BE (! ok, 0)) + return REG_ESPACE; + if (dfa->edests[cur_node].nelem == 0) + break; + if (dfa->edests[cur_node].nelem == 2) + { + reg_errcode_t ret = + check_arrival_expand_ecl_sub (dfa, dst_nodes, + dfa->edests[cur_node].elems[1], + ex_subexp, type); + if (BE (ret != REG_NOERROR, 0)) + return ret; + } + cur_node = dfa->edests[cur_node].elems[0]; + } + return REG_NOERROR; +} + + +/* For all the back references in the current state, calculate the + destination of the back references by the appropriate entry + in MCTX->BKREF_ENTS. */ + +static reg_errcode_t +internal_function +expand_bkref_cache (re_match_context_t *mctx, re_node_set *cur_nodes, + Idx cur_str, Idx subexp_num, int type) +{ + re_dfa_t *const dfa = mctx->dfa; + reg_errcode_t err; + Idx cache_idx_start = search_cur_bkref_entry (mctx, cur_str); + struct re_backref_cache_entry *ent; + + if (cache_idx_start == REG_MISSING) + return REG_NOERROR; + + restart: + ent = mctx->bkref_ents + cache_idx_start; + do + { + Idx to_idx, next_node; + + /* Is this entry ENT is appropriate? */ + if (!re_node_set_contains (cur_nodes, ent->node)) + continue; /* No. */ + + to_idx = cur_str + ent->subexp_to - ent->subexp_from; + /* Calculate the destination of the back reference, and append it + to MCTX->STATE_LOG. */ + if (to_idx == cur_str) + { + /* The backreference did epsilon transit, we must re-check all the + node in the current state. */ + re_node_set new_dests; + reg_errcode_t err2, err3; + next_node = dfa->edests[ent->node].elems[0]; + if (re_node_set_contains (cur_nodes, next_node)) + continue; + err = re_node_set_init_1 (&new_dests, next_node); + err2 = check_arrival_expand_ecl (dfa, &new_dests, subexp_num, type); + err3 = re_node_set_merge (cur_nodes, &new_dests); + re_node_set_free (&new_dests); + if (BE (err != REG_NOERROR || err2 != REG_NOERROR + || err3 != REG_NOERROR, 0)) + { + err = (err != REG_NOERROR ? err + : (err2 != REG_NOERROR ? err2 : err3)); + return err; + } + /* TODO: It is still inefficient... */ + goto restart; + } + else + { + re_node_set union_set; + next_node = dfa->nexts[ent->node]; + if (mctx->state_log[to_idx]) + { + bool ok; + if (re_node_set_contains (&mctx->state_log[to_idx]->nodes, + next_node)) + continue; + err = re_node_set_init_copy (&union_set, + &mctx->state_log[to_idx]->nodes); + ok = re_node_set_insert (&union_set, next_node); + if (BE (err != REG_NOERROR || ! ok, 0)) + { + re_node_set_free (&union_set); + err = err != REG_NOERROR ? err : REG_ESPACE; + return err; + } + } + else + { + err = re_node_set_init_1 (&union_set, next_node); + if (BE (err != REG_NOERROR, 0)) + return err; + } + mctx->state_log[to_idx] = re_acquire_state (&err, dfa, &union_set); + re_node_set_free (&union_set); + if (BE (mctx->state_log[to_idx] == NULL + && err != REG_NOERROR, 0)) + return err; + } + } + while (ent++->more); + return REG_NOERROR; +} + +/* Build transition table for the state. + Return true if successful. */ + +static bool +internal_function +build_trtable (re_dfa_t *dfa, re_dfastate_t *state) +{ + reg_errcode_t err; + Idx i, j; + int ch; + bool need_word_trtable = false; + bitset_word elem, mask; + bool dests_node_malloced = false, dest_states_malloced = false; + Idx ndests; /* Number of the destination states from `state'. */ + re_dfastate_t **trtable; + re_dfastate_t **dest_states = NULL, **dest_states_word, **dest_states_nl; + re_node_set follows, *dests_node; + bitset *dests_ch; + bitset acceptable; + + struct dests_alloc + { + re_node_set dests_node[SBC_MAX]; + bitset dests_ch[SBC_MAX]; + } *dests_alloc; + + /* We build DFA states which corresponds to the destination nodes + from `state'. `dests_node[i]' represents the nodes which i-th + destination state contains, and `dests_ch[i]' represents the + characters which i-th destination state accepts. */ + if (__libc_use_alloca (sizeof (struct dests_alloc))) + dests_alloc = (struct dests_alloc *) alloca (sizeof dests_alloc[0]); + else + { + dests_alloc = re_malloc (struct dests_alloc, 1); + if (BE (dests_alloc == NULL, 0)) + return false; + dests_node_malloced = true; + } + dests_node = dests_alloc->dests_node; + dests_ch = dests_alloc->dests_ch; + + /* Initialize transiton table. */ + state->word_trtable = state->trtable = NULL; + + /* At first, group all nodes belonging to `state' into several + destinations. */ + ndests = group_nodes_into_DFAstates (dfa, state, dests_node, dests_ch); + if (BE (! REG_VALID_NONZERO_INDEX (ndests), 0)) + { + if (dests_node_malloced) + free (dests_alloc); + if (ndests == 0) + { + state->trtable = re_calloc (re_dfastate_t *, SBC_MAX); + return true; + } + return false; + } + + err = re_node_set_alloc (&follows, ndests + 1); + if (BE (err != REG_NOERROR, 0)) + goto out_free; + + /* Avoid arithmetic overflow in size calculation. */ + if (BE (((SIZE_MAX - (sizeof (re_node_set) + sizeof (bitset)) * SBC_MAX) + / (3 * sizeof (re_dfastate_t *))) + < ndests, 0)) + goto out_free; + + if (__libc_use_alloca ((sizeof (re_node_set) + sizeof (bitset)) * SBC_MAX + + ndests * 3 * sizeof (re_dfastate_t *))) + dest_states = (re_dfastate_t **) + alloca (ndests * 3 * sizeof (re_dfastate_t *)); + else + { + dest_states = (re_dfastate_t **) + malloc (ndests * 3 * sizeof (re_dfastate_t *)); + if (BE (dest_states == NULL, 0)) + { +out_free: + if (dest_states_malloced) + free (dest_states); + re_node_set_free (&follows); + for (i = 0; i < ndests; ++i) + re_node_set_free (dests_node + i); + if (dests_node_malloced) + free (dests_alloc); + return false; + } + dest_states_malloced = true; + } + dest_states_word = dest_states + ndests; + dest_states_nl = dest_states_word + ndests; + bitset_empty (acceptable); + + /* Then build the states for all destinations. */ + for (i = 0; i < ndests; ++i) + { + Idx next_node; + re_node_set_empty (&follows); + /* Merge the follows of this destination states. */ + for (j = 0; j < dests_node[i].nelem; ++j) + { + next_node = dfa->nexts[dests_node[i].elems[j]]; + if (next_node != REG_MISSING) + { + err = re_node_set_merge (&follows, dfa->eclosures + next_node); + if (BE (err != REG_NOERROR, 0)) + goto out_free; + } + } + dest_states[i] = re_acquire_state_context (&err, dfa, &follows, 0); + if (BE (dest_states[i] == NULL && err != REG_NOERROR, 0)) + goto out_free; + /* If the new state has context constraint, + build appropriate states for these contexts. */ + if (dest_states[i]->has_constraint) + { + dest_states_word[i] = re_acquire_state_context (&err, dfa, &follows, + CONTEXT_WORD); + if (BE (dest_states_word[i] == NULL && err != REG_NOERROR, 0)) + goto out_free; + + if (dest_states[i] != dest_states_word[i] && dfa->mb_cur_max > 1) + need_word_trtable = true; + + dest_states_nl[i] = re_acquire_state_context (&err, dfa, &follows, + CONTEXT_NEWLINE); + if (BE (dest_states_nl[i] == NULL && err != REG_NOERROR, 0)) + goto out_free; + } + else + { + dest_states_word[i] = dest_states[i]; + dest_states_nl[i] = dest_states[i]; + } + bitset_merge (acceptable, dests_ch[i]); + } + + if (!BE (need_word_trtable, 0)) + { + /* We don't care about whether the following character is a word + character, or we are in a single-byte character set so we can + discern by looking at the character code: allocate a + 256-entry transition table. */ + trtable = state->trtable = re_calloc (re_dfastate_t *, SBC_MAX); + if (BE (trtable == NULL, 0)) + goto out_free; + + /* For all characters ch...: */ + for (i = 0; i < BITSET_WORDS; ++i) + for (ch = i * BITSET_WORD_BITS, elem = acceptable[i], mask = 1; + elem; + mask <<= 1, elem >>= 1, ++ch) + if (BE (elem & 1, 0)) + { + /* There must be exactly one destination which accepts + character ch. See group_nodes_into_DFAstates. */ + for (j = 0; (dests_ch[j][i] & mask) == 0; ++j) + ; + + /* j-th destination accepts the word character ch. */ + if (dfa->word_char[i] & mask) + trtable[ch] = dest_states_word[j]; + else + trtable[ch] = dest_states[j]; + } + } + else + { + /* We care about whether the following character is a word + character, and we are in a multi-byte character set: discern + by looking at the character code: build two 256-entry + transition tables, one starting at trtable[0] and one + starting at trtable[SBC_MAX]. */ + trtable = state->word_trtable = re_calloc (re_dfastate_t *, 2 * SBC_MAX); + if (BE (trtable == NULL, 0)) + goto out_free; + + /* For all characters ch...: */ + for (i = 0; i < BITSET_WORDS; ++i) + for (ch = i * BITSET_WORD_BITS, elem = acceptable[i], mask = 1; + elem; + mask <<= 1, elem >>= 1, ++ch) + if (BE (elem & 1, 0)) + { + /* There must be exactly one destination which accepts + character ch. See group_nodes_into_DFAstates. */ + for (j = 0; (dests_ch[j][i] & mask) == 0; ++j) + ; + + /* j-th destination accepts the word character ch. */ + trtable[ch] = dest_states[j]; + trtable[ch + SBC_MAX] = dest_states_word[j]; + } + } + + /* new line */ + if (bitset_contain (acceptable, NEWLINE_CHAR)) + { + /* The current state accepts newline character. */ + for (j = 0; j < ndests; ++j) + if (bitset_contain (dests_ch[j], NEWLINE_CHAR)) + { + /* k-th destination accepts newline character. */ + trtable[NEWLINE_CHAR] = dest_states_nl[j]; + if (need_word_trtable) + trtable[NEWLINE_CHAR + SBC_MAX] = dest_states_nl[j]; + /* There must be only one destination which accepts + newline. See group_nodes_into_DFAstates. */ + break; + } + } + + if (dest_states_malloced) + free (dest_states); + + re_node_set_free (&follows); + for (i = 0; i < ndests; ++i) + re_node_set_free (dests_node + i); + + if (dests_node_malloced) + free (dests_alloc); + + return true; +} + +/* Group all nodes belonging to STATE into several destinations. + Then for all destinations, set the nodes belonging to the destination + to DESTS_NODE[i] and set the characters accepted by the destination + to DEST_CH[i]. This function return the number of destinations. */ + +static Idx +internal_function +group_nodes_into_DFAstates (const re_dfa_t *dfa, const re_dfastate_t *state, + re_node_set *dests_node, bitset *dests_ch) +{ + reg_errcode_t err; + bool ok; + Idx i, j, k; + Idx ndests; /* Number of the destinations from `state'. */ + bitset accepts; /* Characters a node can accept. */ + const re_node_set *cur_nodes = &state->nodes; + bitset_empty (accepts); + ndests = 0; + + /* For all the nodes belonging to `state', */ + for (i = 0; i < cur_nodes->nelem; ++i) + { + re_token_t *node = &dfa->nodes[cur_nodes->elems[i]]; + re_token_type_t type = node->type; + unsigned int constraint = node->constraint; + + /* Enumerate all single byte character this node can accept. */ + if (type == CHARACTER) + bitset_set (accepts, node->opr.c); + else if (type == SIMPLE_BRACKET) + { + bitset_merge (accepts, node->opr.sbcset); + } + else if (type == OP_PERIOD) + { +#ifdef RE_ENABLE_I18N + if (dfa->mb_cur_max > 1) + bitset_merge (accepts, dfa->sb_char); + else +#endif + bitset_set_all (accepts); + if (!(dfa->syntax & REG_DOT_NEWLINE)) + bitset_clear (accepts, '\n'); + if (dfa->syntax & REG_DOT_NOT_NULL) + bitset_clear (accepts, '\0'); + } +#ifdef RE_ENABLE_I18N + else if (type == OP_UTF8_PERIOD) + { + if (SBC_MAX / 2 % BITSET_WORD_BITS == 0) + memset (accepts, -1, sizeof accepts / 2); + else + bitset_merge (accepts, utf8_sb_map); + if (!(dfa->syntax & REG_DOT_NEWLINE)) + bitset_clear (accepts, '\n'); + if (dfa->syntax & REG_DOT_NOT_NULL) + bitset_clear (accepts, '\0'); + } +#endif + else + continue; + + /* Check the `accepts' and sift the characters which are not + match it the context. */ + if (constraint) + { + if (constraint & NEXT_NEWLINE_CONSTRAINT) + { + bool accepts_newline = bitset_contain (accepts, NEWLINE_CHAR); + bitset_empty (accepts); + if (accepts_newline) + bitset_set (accepts, NEWLINE_CHAR); + else + continue; + } + if (constraint & NEXT_ENDBUF_CONSTRAINT) + { + bitset_empty (accepts); + continue; + } + + if (constraint & NEXT_WORD_CONSTRAINT) + { + bitset_word any_set = 0; + if (type == CHARACTER && !node->word_char) + { + bitset_empty (accepts); + continue; + } +#ifdef RE_ENABLE_I18N + if (dfa->mb_cur_max > 1) + for (j = 0; j < BITSET_WORDS; ++j) + any_set |= (accepts[j] &= (dfa->word_char[j] | ~dfa->sb_char[j])); + else +#endif + for (j = 0; j < BITSET_WORDS; ++j) + any_set |= (accepts[j] &= dfa->word_char[j]); + if (!any_set) + continue; + } + if (constraint & NEXT_NOTWORD_CONSTRAINT) + { + bitset_word any_set = 0; + if (type == CHARACTER && node->word_char) + { + bitset_empty (accepts); + continue; + } +#ifdef RE_ENABLE_I18N + if (dfa->mb_cur_max > 1) + for (j = 0; j < BITSET_WORDS; ++j) + any_set |= (accepts[j] &= ~(dfa->word_char[j] & dfa->sb_char[j])); + else +#endif + for (j = 0; j < BITSET_WORDS; ++j) + any_set |= (accepts[j] &= ~dfa->word_char[j]); + if (!any_set) + continue; + } + } + + /* Then divide `accepts' into DFA states, or create a new + state. Above, we make sure that accepts is not empty. */ + for (j = 0; j < ndests; ++j) + { + bitset intersec; /* Intersection sets, see below. */ + bitset remains; + /* Flags, see below. */ + bitset_word has_intersec, not_subset, not_consumed; + + /* Optimization, skip if this state doesn't accept the character. */ + if (type == CHARACTER && !bitset_contain (dests_ch[j], node->opr.c)) + continue; + + /* Enumerate the intersection set of this state and `accepts'. */ + has_intersec = 0; + for (k = 0; k < BITSET_WORDS; ++k) + has_intersec |= intersec[k] = accepts[k] & dests_ch[j][k]; + /* And skip if the intersection set is empty. */ + if (!has_intersec) + continue; + + /* Then check if this state is a subset of `accepts'. */ + not_subset = not_consumed = 0; + for (k = 0; k < BITSET_WORDS; ++k) + { + not_subset |= remains[k] = ~accepts[k] & dests_ch[j][k]; + not_consumed |= accepts[k] = accepts[k] & ~dests_ch[j][k]; + } + + /* If this state isn't a subset of `accepts', create a + new group state, which has the `remains'. */ + if (not_subset) + { + bitset_copy (dests_ch[ndests], remains); + bitset_copy (dests_ch[j], intersec); + err = re_node_set_init_copy (dests_node + ndests, &dests_node[j]); + if (BE (err != REG_NOERROR, 0)) + goto error_return; + ++ndests; + } + + /* Put the position in the current group. */ + ok = re_node_set_insert (&dests_node[j], cur_nodes->elems[i]); + if (BE (! ok, 0)) + goto error_return; + + /* If all characters are consumed, go to next node. */ + if (!not_consumed) + break; + } + /* Some characters remain, create a new group. */ + if (j == ndests) + { + bitset_copy (dests_ch[ndests], accepts); + err = re_node_set_init_1 (dests_node + ndests, cur_nodes->elems[i]); + if (BE (err != REG_NOERROR, 0)) + goto error_return; + ++ndests; + bitset_empty (accepts); + } + } + return ndests; + error_return: + for (j = 0; j < ndests; ++j) + re_node_set_free (dests_node + j); + return REG_MISSING; +} + +#ifdef RE_ENABLE_I18N +/* Check how many bytes the node `dfa->nodes[node_idx]' accepts. + Return the number of the bytes the node accepts. + STR_IDX is the current index of the input string. + + This function handles the nodes which can accept one character, or + one collating element like '.', '[a-z]', opposite to the other nodes + can only accept one byte. */ + +static int +internal_function +check_node_accept_bytes (re_dfa_t *dfa, Idx node_idx, + const re_string_t *input, Idx str_idx) +{ + const re_token_t *node = dfa->nodes + node_idx; + int char_len, elem_len; + Idx i; + + if (BE (node->type == OP_UTF8_PERIOD, 0)) + { + unsigned char c = re_string_byte_at (input, str_idx), d; + if (BE (c < 0xc2, 1)) + return 0; + + if (str_idx + 2 > input->len) + return 0; + + d = re_string_byte_at (input, str_idx + 1); + if (c < 0xe0) + return (d < 0x80 || d > 0xbf) ? 0 : 2; + else if (c < 0xf0) + { + char_len = 3; + if (c == 0xe0 && d < 0xa0) + return 0; + } + else if (c < 0xf8) + { + char_len = 4; + if (c == 0xf0 && d < 0x90) + return 0; + } + else if (c < 0xfc) + { + char_len = 5; + if (c == 0xf8 && d < 0x88) + return 0; + } + else if (c < 0xfe) + { + char_len = 6; + if (c == 0xfc && d < 0x84) + return 0; + } + else + return 0; + + if (str_idx + char_len > input->len) + return 0; + + for (i = 1; i < char_len; ++i) + { + d = re_string_byte_at (input, str_idx + i); + if (d < 0x80 || d > 0xbf) + return 0; + } + return char_len; + } + + char_len = re_string_char_size_at (input, str_idx); + if (node->type == OP_PERIOD) + { + if (char_len <= 1) + return 0; + /* FIXME: I don't think this if is needed, as both '\n' + and '\0' are char_len == 1. */ + /* '.' accepts any one character except the following two cases. */ + if ((!(dfa->syntax & REG_DOT_NEWLINE) && + re_string_byte_at (input, str_idx) == '\n') || + ((dfa->syntax & REG_DOT_NOT_NULL) && + re_string_byte_at (input, str_idx) == '\0')) + return 0; + return char_len; + } + + elem_len = re_string_elem_size_at (input, str_idx); + if ((elem_len <= 1 && char_len <= 1) || char_len == 0) + return 0; + + if (node->type == COMPLEX_BRACKET) + { + const re_charset_t *cset = node->opr.mbcset; +# ifdef _LIBC + const unsigned char *pin + = ((const unsigned char *) re_string_get_buffer (input) + str_idx); + Idx j; + uint32_t nrules; +# endif /* _LIBC */ + int match_len = 0; + wchar_t wc = ((cset->nranges || cset->nchar_classes || cset->nmbchars) + ? re_string_wchar_at (input, str_idx) : 0); + + /* match with multibyte character? */ + for (i = 0; i < cset->nmbchars; ++i) + if (wc == cset->mbchars[i]) + { + match_len = char_len; + goto check_node_accept_bytes_match; + } + /* match with character_class? */ + for (i = 0; i < cset->nchar_classes; ++i) + { + wctype_t wt = cset->char_classes[i]; + if (__iswctype (wc, wt)) + { + match_len = char_len; + goto check_node_accept_bytes_match; + } + } + +# ifdef _LIBC + nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES); + if (nrules != 0) + { + unsigned int in_collseq = 0; + const int32_t *table, *indirect; + const unsigned char *weights, *extra; + const char *collseqwc; + int32_t idx; + /* This #include defines a local function! */ +# include <locale/weight.h> + + /* match with collating_symbol? */ + if (cset->ncoll_syms) + extra = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_SYMB_EXTRAMB); + for (i = 0; i < cset->ncoll_syms; ++i) + { + const unsigned char *coll_sym = extra + cset->coll_syms[i]; + /* Compare the length of input collating element and + the length of current collating element. */ + if (*coll_sym != elem_len) + continue; + /* Compare each bytes. */ + for (j = 0; j < *coll_sym; j++) + if (pin[j] != coll_sym[1 + j]) + break; + if (j == *coll_sym) + { + /* Match if every bytes is equal. */ + match_len = j; + goto check_node_accept_bytes_match; + } + } + + if (cset->nranges) + { + if (elem_len <= char_len) + { + collseqwc = _NL_CURRENT (LC_COLLATE, _NL_COLLATE_COLLSEQWC); + in_collseq = __collseq_table_lookup (collseqwc, wc); + } + else + in_collseq = find_collation_sequence_value (pin, elem_len); + } + /* match with range expression? */ + for (i = 0; i < cset->nranges; ++i) + if (cset->range_starts[i] <= in_collseq + && in_collseq <= cset->range_ends[i]) + { + match_len = elem_len; + goto check_node_accept_bytes_match; + } + + /* match with equivalence_class? */ + if (cset->nequiv_classes) + { + const unsigned char *cp = pin; + table = (const int32_t *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB); + weights = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_WEIGHTMB); + extra = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_EXTRAMB); + indirect = (const int32_t *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_INDIRECTMB); + idx = findidx (&cp); + if (idx > 0) + for (i = 0; i < cset->nequiv_classes; ++i) + { + int32_t equiv_class_idx = cset->equiv_classes[i]; + size_t weight_len = weights[idx]; + if (weight_len == weights[equiv_class_idx]) + { + Idx cnt = 0; + while (cnt <= weight_len + && (weights[equiv_class_idx + 1 + cnt] + == weights[idx + 1 + cnt])) + ++cnt; + if (cnt > weight_len) + { + match_len = elem_len; + goto check_node_accept_bytes_match; + } + } + } + } + } + else +# endif /* _LIBC */ + { + /* match with range expression? */ +#if __GNUC__ >= 2 + wchar_t cmp_buf[] = {L'\0', L'\0', wc, L'\0', L'\0', L'\0'}; +#else + wchar_t cmp_buf[] = {L'\0', L'\0', L'\0', L'\0', L'\0', L'\0'}; + cmp_buf[2] = wc; +#endif + for (i = 0; i < cset->nranges; ++i) + { + cmp_buf[0] = cset->range_starts[i]; + cmp_buf[4] = cset->range_ends[i]; + if (wcscoll (cmp_buf, cmp_buf + 2) <= 0 + && wcscoll (cmp_buf + 2, cmp_buf + 4) <= 0) + { + match_len = char_len; + goto check_node_accept_bytes_match; + } + } + } + check_node_accept_bytes_match: + if (!cset->non_match) + return match_len; + else + { + if (match_len > 0) + return 0; + else + return (elem_len > char_len) ? elem_len : char_len; + } + } + return 0; +} + +# ifdef _LIBC +static unsigned int +find_collation_sequence_value (const unsigned char *mbs, size_t mbs_len) +{ + uint32_t nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES); + if (nrules == 0) + { + if (mbs_len == 1) + { + /* No valid character. Match it as a single byte character. */ + const unsigned char *collseq = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_COLLSEQMB); + return collseq[mbs[0]]; + } + return UINT_MAX; + } + else + { + int32_t idx; + const unsigned char *extra = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_SYMB_EXTRAMB); + int32_t extrasize = (const unsigned char *) + _NL_CURRENT (LC_COLLATE, _NL_COLLATE_SYMB_EXTRAMB + 1) - extra; + + for (idx = 0; idx < extrasize;) + { + int mbs_cnt; + bool found = false; + int32_t elem_mbs_len; + /* Skip the name of collating element name. */ + idx = idx + extra[idx] + 1; + elem_mbs_len = extra[idx++]; + if (mbs_len == elem_mbs_len) + { + for (mbs_cnt = 0; mbs_cnt < elem_mbs_len; ++mbs_cnt) + if (extra[idx + mbs_cnt] != mbs[mbs_cnt]) + break; + if (mbs_cnt == elem_mbs_len) + /* Found the entry. */ + found = true; + } + /* Skip the byte sequence of the collating element. */ + idx += elem_mbs_len; + /* Adjust for the alignment. */ + idx = (idx + 3) & ~3; + /* Skip the collation sequence value. */ + idx += sizeof (uint32_t); + /* Skip the wide char sequence of the collating element. */ + idx = idx + sizeof (uint32_t) * (extra[idx] + 1); + /* If we found the entry, return the sequence value. */ + if (found) + return *(uint32_t *) (extra + idx); + /* Skip the collation sequence value. */ + idx += sizeof (uint32_t); + } + return UINT_MAX; + } +} +# endif /* _LIBC */ +#endif /* RE_ENABLE_I18N */ + +/* Check whether the node accepts the byte which is IDX-th + byte of the INPUT. */ + +static bool +internal_function +check_node_accept (const re_match_context_t *mctx, const re_token_t *node, + Idx idx) +{ + unsigned char ch; + ch = re_string_byte_at (&mctx->input, idx); + switch (node->type) + { + case CHARACTER: + if (node->opr.c != ch) + return false; + break; + + case SIMPLE_BRACKET: + if (!bitset_contain (node->opr.sbcset, ch)) + return false; + break; + +#ifdef RE_ENABLE_I18N + case OP_UTF8_PERIOD: + if (ch >= 0x80) + return false; + /* FALLTHROUGH */ +#endif + case OP_PERIOD: + if ((ch == '\n' && !(mctx->dfa->syntax & REG_DOT_NEWLINE)) + || (ch == '\0' && (mctx->dfa->syntax & REG_DOT_NOT_NULL))) + return false; + break; + + default: + return false; + } + + if (node->constraint) + { + /* The node has constraints. Check whether the current context + satisfies the constraints. */ + unsigned int context = re_string_context_at (&mctx->input, idx, + mctx->eflags); + if (NOT_SATISFY_NEXT_CONSTRAINT (node->constraint, context)) + return false; + } + + return true; +} + +/* Extend the buffers, if the buffers have run out. */ + +static reg_errcode_t +internal_function +extend_buffers (re_match_context_t *mctx) +{ + reg_errcode_t ret; + re_string_t *pstr = &mctx->input; + + /* Double the lengthes of the buffers. */ + ret = re_string_realloc_buffers (pstr, pstr->bufs_len * 2); + if (BE (ret != REG_NOERROR, 0)) + return ret; + + if (mctx->state_log != NULL) + { + /* And double the length of state_log. */ + /* XXX We have no indication of the size of this buffer. If this + allocation fail we have no indication that the state_log array + does not have the right size. */ + re_dfastate_t **new_array = re_xrealloc (mctx->state_log, re_dfastate_t *, + pstr->bufs_len + 1); + if (BE (new_array == NULL, 0)) + return REG_ESPACE; + mctx->state_log = new_array; + } + + /* Then reconstruct the buffers. */ + if (pstr->icase) + { +#ifdef RE_ENABLE_I18N + if (pstr->mb_cur_max > 1) + { + ret = build_wcs_upper_buffer (pstr); + if (BE (ret != REG_NOERROR, 0)) + return ret; + } + else +#endif /* RE_ENABLE_I18N */ + build_upper_buffer (pstr); + } + else + { +#ifdef RE_ENABLE_I18N + if (pstr->mb_cur_max > 1) + build_wcs_buffer (pstr); + else +#endif /* RE_ENABLE_I18N */ + { + if (pstr->trans != NULL) + re_string_translate_buffer (pstr); + } + } + return REG_NOERROR; +} + + +/* Functions for matching context. */ + +/* Initialize MCTX. */ + +static reg_errcode_t +internal_function +match_ctx_init (re_match_context_t *mctx, int eflags, Idx n) +{ + mctx->eflags = eflags; + mctx->match_last = REG_MISSING; + if (n > 0) + { + mctx->bkref_ents = re_xmalloc (struct re_backref_cache_entry, n); + mctx->sub_tops = re_xmalloc (re_sub_match_top_t *, n); + if (BE (mctx->bkref_ents == NULL || mctx->sub_tops == NULL, 0)) + return REG_ESPACE; + } + /* Already zero-ed by the caller. + else + mctx->bkref_ents = NULL; + mctx->nbkref_ents = 0; + mctx->nsub_tops = 0; */ + mctx->abkref_ents = n; + mctx->max_mb_elem_len = 1; + mctx->asub_tops = n; + return REG_NOERROR; +} + +/* Clean the entries which depend on the current input in MCTX. + This function must be invoked when the matcher changes the start index + of the input, or changes the input string. */ + +static void +internal_function +match_ctx_clean (re_match_context_t *mctx) +{ + Idx st_idx; + for (st_idx = 0; st_idx < mctx->nsub_tops; ++st_idx) + { + Idx sl_idx; + re_sub_match_top_t *top = mctx->sub_tops[st_idx]; + for (sl_idx = 0; sl_idx < top->nlasts; ++sl_idx) + { + re_sub_match_last_t *last = top->lasts[sl_idx]; + re_free (last->path.array); + re_free (last); + } + re_free (top->lasts); + if (top->path) + { + re_free (top->path->array); + re_free (top->path); + } + free (top); + } + + mctx->nsub_tops = 0; + mctx->nbkref_ents = 0; +} + +/* Free all the memory associated with MCTX. */ + +static void +internal_function +match_ctx_free (re_match_context_t *mctx) +{ + /* First, free all the memory associated with MCTX->SUB_TOPS. */ + match_ctx_clean (mctx); + re_free (mctx->sub_tops); + re_free (mctx->bkref_ents); +} + +/* Add a new backreference entry to MCTX. + Note that we assume that caller never call this function with duplicate + entry, and call with STR_IDX which isn't smaller than any existing entry. +*/ + +static reg_errcode_t +internal_function +match_ctx_add_entry (re_match_context_t *mctx, Idx node, Idx str_idx, + Idx from, Idx to) +{ + if (mctx->nbkref_ents >= mctx->abkref_ents) + { + struct re_backref_cache_entry* new_entry; + new_entry = re_x2realloc (mctx->bkref_ents, struct re_backref_cache_entry, + &mctx->abkref_ents); + if (BE (new_entry == NULL, 0)) + { + re_free (mctx->bkref_ents); + return REG_ESPACE; + } + mctx->bkref_ents = new_entry; + memset (mctx->bkref_ents + mctx->nbkref_ents, '\0', + (sizeof (struct re_backref_cache_entry) + * (mctx->abkref_ents - mctx->nbkref_ents))); + } + if (mctx->nbkref_ents > 0 + && mctx->bkref_ents[mctx->nbkref_ents - 1].str_idx == str_idx) + mctx->bkref_ents[mctx->nbkref_ents - 1].more = 1; + + mctx->bkref_ents[mctx->nbkref_ents].node = node; + mctx->bkref_ents[mctx->nbkref_ents].str_idx = str_idx; + mctx->bkref_ents[mctx->nbkref_ents].subexp_from = from; + mctx->bkref_ents[mctx->nbkref_ents].subexp_to = to; + + /* This is a cache that saves negative results of check_dst_limits_calc_pos. + If bit N is clear, means that this entry won't epsilon-transition to + an OP_OPEN_SUBEXP or OP_CLOSE_SUBEXP for the N+1-th subexpression. If + it is set, check_dst_limits_calc_pos_1 will recurse and try to find one + such node. + + A backreference does not epsilon-transition unless it is empty, so set + to all zeros if FROM != TO. */ + mctx->bkref_ents[mctx->nbkref_ents].eps_reachable_subexps_map + = (from == to ? -1 : 0); + + mctx->bkref_ents[mctx->nbkref_ents++].more = 0; + if (mctx->max_mb_elem_len < to - from) + mctx->max_mb_elem_len = to - from; + return REG_NOERROR; +} + +/* Return the first entry with the same str_idx, or REG_MISSING if none is + found. Note that MCTX->BKREF_ENTS is already sorted by MCTX->STR_IDX. */ + +static Idx +internal_function +search_cur_bkref_entry (const re_match_context_t *mctx, Idx str_idx) +{ + Idx left, right, mid, last; + last = right = mctx->nbkref_ents; + for (left = 0; left < right;) + { + mid = (left + right) / 2; + if (mctx->bkref_ents[mid].str_idx < str_idx) + left = mid + 1; + else + right = mid; + } + if (left < last && mctx->bkref_ents[left].str_idx == str_idx) + return left; + else + return REG_MISSING; +} + +/* Register the node NODE, whose type is OP_OPEN_SUBEXP, and which matches + at STR_IDX. */ + +static reg_errcode_t +internal_function +match_ctx_add_subtop (re_match_context_t *mctx, Idx node, Idx str_idx) +{ +#ifdef DEBUG + assert (mctx->sub_tops != NULL); + assert (mctx->asub_tops > 0); +#endif + if (BE (mctx->nsub_tops == mctx->asub_tops, 0)) + { + Idx new_asub_tops = mctx->asub_tops; + re_sub_match_top_t **new_array = re_x2realloc (mctx->sub_tops, + re_sub_match_top_t *, + &new_asub_tops); + if (BE (new_array == NULL, 0)) + return REG_ESPACE; + mctx->sub_tops = new_array; + mctx->asub_tops = new_asub_tops; + } + mctx->sub_tops[mctx->nsub_tops] = re_calloc (re_sub_match_top_t, 1); + if (BE (mctx->sub_tops[mctx->nsub_tops] == NULL, 0)) + return REG_ESPACE; + mctx->sub_tops[mctx->nsub_tops]->node = node; + mctx->sub_tops[mctx->nsub_tops++]->str_idx = str_idx; + return REG_NOERROR; +} + +/* Register the node NODE, whose type is OP_CLOSE_SUBEXP, and which matches + at STR_IDX, whose corresponding OP_OPEN_SUBEXP is SUB_TOP. */ + +static re_sub_match_last_t * +internal_function +match_ctx_add_sublast (re_sub_match_top_t *subtop, Idx node, Idx str_idx) +{ + re_sub_match_last_t *new_entry; + if (BE (subtop->nlasts == subtop->alasts, 0)) + { + Idx new_alasts = subtop->alasts; + re_sub_match_last_t **new_array = re_x2realloc (subtop->lasts, + re_sub_match_last_t *, + &new_alasts); + if (BE (new_array == NULL, 0)) + return NULL; + subtop->lasts = new_array; + subtop->alasts = new_alasts; + } + new_entry = re_calloc (re_sub_match_last_t, 1); + if (BE (new_entry != NULL, 1)) + { + subtop->lasts[subtop->nlasts] = new_entry; + new_entry->node = node; + new_entry->str_idx = str_idx; + ++subtop->nlasts; + } + return new_entry; +} + +static void +internal_function +sift_ctx_init (re_sift_context_t *sctx, + re_dfastate_t **sifted_sts, + re_dfastate_t **limited_sts, + Idx last_node, Idx last_str_idx) +{ + sctx->sifted_states = sifted_sts; + sctx->limited_states = limited_sts; + sctx->last_node = last_node; + sctx->last_str_idx = last_str_idx; + re_node_set_init_empty (&sctx->limits); +} diff --git a/lib/rename.c b/lib/rename.c new file mode 100644 index 0000000..9c9e7f4 --- /dev/null +++ b/lib/rename.c @@ -0,0 +1,58 @@ +/* Work around the bug in some systems whereby rename fails when the source + file has a trailing slash. The rename functions of SunOS 4.1.1_U1 and + mips-dec-ultrix4.4 have this bug. + Copyright (C) 2001, 2002, 2003, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* written by Volker Borchert */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif +#undef rename + +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +#include "dirname.h" +#include "xalloc.h" + +/* Rename the file SRC to DST, removing any trailing + slashes from SRC. Needed for SunOS 4.1.1_U1. */ + +int +rpl_rename (char const *src, char const *dst) +{ + char *src_temp; + int ret_val; + size_t s_len = strlen (src); + + if (s_len && src[s_len - 1] == '/') + { + src_temp = xstrdup (src); + strip_trailing_slashes (src_temp); + } + else + src_temp = (char *) src; + + ret_val = rename (src_temp, dst); + + if (src_temp != src) + free (src_temp); + + return ret_val; +} diff --git a/lib/rpmatch.c b/lib/rpmatch.c new file mode 100644 index 0000000..ba8bc4c --- /dev/null +++ b/lib/rpmatch.c @@ -0,0 +1,79 @@ +/* Determine whether string value is affirmation or negative response + according to current locale's data. + Copyright (C) 1996, 1998, 2000, 2002, 2003 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include <stddef.h> +#include <stdlib.h> + +#if ENABLE_NLS +# include <sys/types.h> +# include <limits.h> +# include <regex.h> +# include "gettext.h" +# define _(msgid) gettext (msgid) + +static int +try (const char *response, const char *pattern, const int match, + const int nomatch, const char **lastp, regex_t *re) +{ + if (pattern != *lastp) + { + /* The pattern has changed. */ + if (*lastp) + { + /* Free the old compiled pattern. */ + regfree (re); + *lastp = NULL; + } + /* Compile the pattern and cache it for future runs. */ + if (regcomp (re, pattern, REG_EXTENDED) != 0) + return -1; + *lastp = pattern; + } + + /* See if the regular expression matches RESPONSE. */ + return regexec (re, response, 0, NULL, 0) == 0 ? match : nomatch; +} +#endif + + +int +rpmatch (const char *response) +{ +#if ENABLE_NLS + /* Match against one of the response patterns, compiling the pattern + first if necessary. */ + + /* We cache the response patterns and compiled regexps here. */ + static const char *yesexpr, *noexpr; + static regex_t yesre, nore; + int result; + + return ((result = try (response, _("^[yY]"), 1, 0, + &yesexpr, &yesre)) + ? result + : try (response, _("^[nN]"), 0, -1, &noexpr, &nore)); +#else + /* Test against "^[yY]" and "^[nN]", hardcoded to avoid requiring regex */ + return (*response == 'y' || *response == 'Y' ? 1 + : *response == 'n' || *response == 'N' ? 0 : -1); +#endif +} diff --git a/lib/save-cwd.c b/lib/save-cwd.c new file mode 100644 index 0000000..7f1b30b --- /dev/null +++ b/lib/save-cwd.c @@ -0,0 +1,115 @@ +/* save-cwd.c -- Save and restore current working directory. + + Copyright (C) 1995, 1997, 1998, 2003, 2004, 2005 Free Software + Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Jim Meyering. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "save-cwd.h" + +#include <stdbool.h> +#include <stdio.h> +#include <stdlib.h> + +#if HAVE_UNISTD_H +# include <unistd.h> +#endif + +#include <fcntl.h> + +#include <errno.h> + +#include "chdir-long.h" +#include "unistd-safer.h" +#include "xgetcwd.h" + +/* On systems without the fchdir function (WOE), pretend that open + always returns -1 so that save_cwd resorts to using xgetcwd. + Since chdir_long requires fchdir, use chdir instead. */ +#if !HAVE_FCHDIR +# undef open +# define open(File, Flags) (-1) +# undef fchdir +# define fchdir(Fd) (abort (), -1) +# undef chdir_long +# define chdir_long(Dir) chdir (Dir) +#endif + +/* Record the location of the current working directory in CWD so that + the program may change to other directories and later use restore_cwd + to return to the recorded location. This function may allocate + space using malloc (via xgetcwd) or leave a file descriptor open; + use free_cwd to perform the necessary free or close. Upon failure, + no memory is allocated, any locally opened file descriptors are + closed; return non-zero -- in that case, free_cwd need not be + called, but doing so is ok. Otherwise, return zero. + + The `raison d'etre' for this interface is that the working directory + is sometimes inaccessible, and getcwd is not robust or as efficient. + So, we prefer to use the open/fchdir approach, but fall back on + getcwd if necessary. + + Some systems lack fchdir altogether: e.g., OS/2, pre-2001 Cygwin, + SCO Xenix. Also, SunOS 4 and Irix 5.3 provide the function, yet it + doesn't work for partitions on which auditing is enabled. If + you're still using an obsolete system with these problems, please + send email to the maintainer of this code. */ + +int +save_cwd (struct saved_cwd *cwd) +{ + cwd->name = NULL; + + cwd->desc = fd_safer (open (".", O_RDONLY)); + if (cwd->desc < 0) + { + cwd->desc = fd_safer (open (".", O_WRONLY)); + if (cwd->desc < 0) + { + cwd->name = xgetcwd (); + return cwd->name ? 0 : -1; + } + } + + return 0; +} + +/* Change to recorded location, CWD, in directory hierarchy. + Upon failure, return -1 (errno is set by chdir or fchdir). + Upon success, return zero. */ + +int +restore_cwd (const struct saved_cwd *cwd) +{ + if (0 <= cwd->desc) + return fchdir (cwd->desc); + else + return chdir_long (cwd->name); +} + +void +free_cwd (struct saved_cwd *cwd) +{ + if (cwd->desc >= 0) + close (cwd->desc); + if (cwd->name) + free (cwd->name); +} diff --git a/lib/save-cwd.h b/lib/save-cwd.h new file mode 100644 index 0000000..d646b55 --- /dev/null +++ b/lib/save-cwd.h @@ -0,0 +1,34 @@ +/* Save and restore current working directory. + + Copyright (C) 1995, 1997, 1998, 2003 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Jim Meyering. */ + +#ifndef SAVE_CWD_H +# define SAVE_CWD_H 1 + +struct saved_cwd + { + int desc; + char *name; + }; + +int save_cwd (struct saved_cwd *cwd); +int restore_cwd (const struct saved_cwd *cwd); +void free_cwd (struct saved_cwd *cwd); + +#endif /* SAVE_CWD_H */ diff --git a/lib/setenv.c b/lib/setenv.c new file mode 100644 index 0000000..33dbeb7 --- /dev/null +++ b/lib/setenv.c @@ -0,0 +1,328 @@ +/* Copyright (C) 1992,1995-1999,2000-2003 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#if HAVE_CONFIG_H +# include <config.h> +#endif +#include <alloca.h> + +#include <errno.h> +#ifndef __set_errno +# define __set_errno(ev) ((errno) = (ev)) +#endif + +#include <stdlib.h> +#include <string.h> +#if _LIBC || HAVE_UNISTD_H +# include <unistd.h> +#endif + +#if !_LIBC +# include "allocsa.h" +#endif + +#if !_LIBC +# define __environ environ +# ifndef HAVE_ENVIRON_DECL +extern char **environ; +# endif +#endif + +#if _LIBC +/* This lock protects against simultaneous modifications of `environ'. */ +# include <bits/libc-lock.h> +__libc_lock_define_initialized (static, envlock) +# define LOCK __libc_lock_lock (envlock) +# define UNLOCK __libc_lock_unlock (envlock) +#else +# define LOCK +# define UNLOCK +#endif + +/* In the GNU C library we must keep the namespace clean. */ +#ifdef _LIBC +# define setenv __setenv +# define clearenv __clearenv +# define tfind __tfind +# define tsearch __tsearch +#endif + +/* In the GNU C library implementation we try to be more clever and + allow arbitrarily many changes of the environment given that the used + values are from a small set. Outside glibc this will eat up all + memory after a while. */ +#if defined _LIBC || (defined HAVE_SEARCH_H && defined HAVE_TSEARCH \ + && defined __GNUC__) +# define USE_TSEARCH 1 +# include <search.h> +typedef int (*compar_fn_t) (const void *, const void *); + +/* This is a pointer to the root of the search tree with the known + values. */ +static void *known_values; + +# define KNOWN_VALUE(Str) \ + ({ \ + void *value = tfind (Str, &known_values, (compar_fn_t) strcmp); \ + value != NULL ? *(char **) value : NULL; \ + }) +# define STORE_VALUE(Str) \ + tsearch (Str, &known_values, (compar_fn_t) strcmp) + +#else +# undef USE_TSEARCH + +# define KNOWN_VALUE(Str) NULL +# define STORE_VALUE(Str) do { } while (0) + +#endif + + +/* If this variable is not a null pointer we allocated the current + environment. */ +static char **last_environ; + + +/* This function is used by `setenv' and `putenv'. The difference between + the two functions is that for the former must create a new string which + is then placed in the environment, while the argument of `putenv' + must be used directly. This is all complicated by the fact that we try + to reuse values once generated for a `setenv' call since we can never + free the strings. */ +int +__add_to_environ (const char *name, const char *value, const char *combined, + int replace) +{ + register char **ep; + register size_t size; + const size_t namelen = strlen (name); + const size_t vallen = value != NULL ? strlen (value) + 1 : 0; + + LOCK; + + /* We have to get the pointer now that we have the lock and not earlier + since another thread might have created a new environment. */ + ep = __environ; + + size = 0; + if (ep != NULL) + { + for (; *ep != NULL; ++ep) + if (!strncmp (*ep, name, namelen) && (*ep)[namelen] == '=') + break; + else + ++size; + } + + if (ep == NULL || *ep == NULL) + { + char **new_environ; +#ifdef USE_TSEARCH + char *new_value; +#endif + + /* We allocated this space; we can extend it. */ + new_environ = + (char **) (last_environ == NULL + ? malloc ((size + 2) * sizeof (char *)) + : realloc (last_environ, (size + 2) * sizeof (char *))); + if (new_environ == NULL) + { + UNLOCK; + return -1; + } + + /* If the whole entry is given add it. */ + if (combined != NULL) + /* We must not add the string to the search tree since it belongs + to the user. */ + new_environ[size] = (char *) combined; + else + { + /* See whether the value is already known. */ +#ifdef USE_TSEARCH +# ifdef _LIBC + new_value = (char *) alloca (namelen + 1 + vallen); + __mempcpy (__mempcpy (__mempcpy (new_value, name, namelen), "=", 1), + value, vallen); +# else + new_value = (char *) allocsa (namelen + 1 + vallen); + if (new_value == NULL) + { + __set_errno (ENOMEM); + UNLOCK; + return -1; + } + memcpy (new_value, name, namelen); + new_value[namelen] = '='; + memcpy (&new_value[namelen + 1], value, vallen); +# endif + + new_environ[size] = KNOWN_VALUE (new_value); + if (new_environ[size] == NULL) +#endif + { + new_environ[size] = (char *) malloc (namelen + 1 + vallen); + if (new_environ[size] == NULL) + { +#if defined USE_TSEARCH && !defined _LIBC + freesa (new_value); +#endif + __set_errno (ENOMEM); + UNLOCK; + return -1; + } + +#ifdef USE_TSEARCH + memcpy (new_environ[size], new_value, namelen + 1 + vallen); +#else + memcpy (new_environ[size], name, namelen); + new_environ[size][namelen] = '='; + memcpy (&new_environ[size][namelen + 1], value, vallen); +#endif + /* And save the value now. We cannot do this when we remove + the string since then we cannot decide whether it is a + user string or not. */ + STORE_VALUE (new_environ[size]); + } +#if defined USE_TSEARCH && !defined _LIBC + freesa (new_value); +#endif + } + + if (__environ != last_environ) + memcpy ((char *) new_environ, (char *) __environ, + size * sizeof (char *)); + + new_environ[size + 1] = NULL; + + last_environ = __environ = new_environ; + } + else if (replace) + { + char *np; + + /* Use the user string if given. */ + if (combined != NULL) + np = (char *) combined; + else + { +#ifdef USE_TSEARCH + char *new_value; +# ifdef _LIBC + new_value = alloca (namelen + 1 + vallen); + __mempcpy (__mempcpy (__mempcpy (new_value, name, namelen), "=", 1), + value, vallen); +# else + new_value = allocsa (namelen + 1 + vallen); + if (new_value == NULL) + { + __set_errno (ENOMEM); + UNLOCK; + return -1; + } + memcpy (new_value, name, namelen); + new_value[namelen] = '='; + memcpy (&new_value[namelen + 1], value, vallen); +# endif + + np = KNOWN_VALUE (new_value); + if (np == NULL) +#endif + { + np = malloc (namelen + 1 + vallen); + if (np == NULL) + { +#if defined USE_TSEARCH && !defined _LIBC + freesa (new_value); +#endif + __set_errno (ENOMEM); + UNLOCK; + return -1; + } + +#ifdef USE_TSEARCH + memcpy (np, new_value, namelen + 1 + vallen); +#else + memcpy (np, name, namelen); + np[namelen] = '='; + memcpy (&np[namelen + 1], value, vallen); +#endif + /* And remember the value. */ + STORE_VALUE (np); + } +#if defined USE_TSEARCH && !defined _LIBC + freesa (new_value); +#endif + } + + *ep = np; + } + + UNLOCK; + + return 0; +} + +int +setenv (const char *name, const char *value, int replace) +{ + return __add_to_environ (name, value, NULL, replace); +} + +/* The `clearenv' was planned to be added to POSIX.1 but probably + never made it. Nevertheless the POSIX.9 standard (POSIX bindings + for Fortran 77) requires this function. */ +int +clearenv (void) +{ + LOCK; + + if (__environ == last_environ && __environ != NULL) + { + /* We allocated this environment so we can free it. */ + free (__environ); + last_environ = NULL; + } + + /* Clear the environment pointer removes the whole environment. */ + __environ = NULL; + + UNLOCK; + + return 0; +} + +#ifdef _LIBC +static void +free_mem (void) +{ + /* Remove all traces. */ + clearenv (); + + /* Now remove the search tree. */ + __tdestroy (known_values, free); + known_values = NULL; +} +text_set_element (__libc_subfreeres, free_mem); + + +# undef setenv +# undef clearenv +weak_alias (__setenv, setenv) +weak_alias (__clearenv, clearenv) +#endif diff --git a/lib/setenv.h b/lib/setenv.h new file mode 100644 index 0000000..c89e7da --- /dev/null +++ b/lib/setenv.h @@ -0,0 +1,54 @@ +/* Setting environment variables. + Copyright (C) 2001-2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#if HAVE_SETENV || HAVE_UNSETENV + +/* Get setenv(), unsetenv() declarations. */ +# include <stdlib.h> + +#endif + +#ifdef __cplusplus +extern "C" { +#endif + +#if !HAVE_SETENV + +/* Set NAME to VALUE in the environment. + If REPLACE is nonzero, overwrite an existing value. */ +extern int setenv (const char *name, const char *value, int replace); + +#endif + +#if HAVE_UNSETENV + +# if VOID_UNSETENV +/* On some systems, unsetenv() returns void. + This is the case for FreeBSD 4.8, NetBSD 1.6, OpenBSD 3.4. */ +# define unsetenv(name) ((unsetenv)(name), 0) +# endif + +#else + +/* Remove the variable NAME from the environment. */ +extern int unsetenv (const char *name); + +#endif + +#ifdef __cplusplus +} +#endif diff --git a/lib/sighandle.c b/lib/sighandle.c new file mode 100644 index 0000000..7934461 --- /dev/null +++ b/lib/sighandle.c @@ -0,0 +1,416 @@ +/* sighandle.c -- Library routines for manipulating chains of signal handlers + Copyright (C) 1992 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. */ + +/* Written by Paul Sander, HaL Computer Systems, Inc. <paul@hal.com> + Brian Berliner <berliner@Sun.COM> added POSIX support */ + +/************************************************************************* + * + * signal.c -- This file contains code that manipulates chains of signal + * handlers. + * + * Facilities are provided to register a signal handler for + * any specific signal. When a signal is received, all of the + * registered signal handlers are invoked in the reverse order + * in which they are registered. Note that the signal handlers + * must not themselves make calls to the signal handling + * facilities. + * + *************************************************************************/ + +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif +#include "system.h" + +#include <sys/types.h> +#include <stdio.h> +#include <signal.h> + +#ifdef STDC_HEADERS +#include <stdlib.h> +#else +#if __STDC__ +char *calloc(unsigned nelem, unsigned size); +char *malloc(unsigned size); +#else +char *calloc(); +char *malloc(); +#endif /* __STDC__ */ +#endif /* STDC_HEADERS */ + +/* Define the highest signal number (usually) */ +#ifndef SIGMAX +#define SIGMAX 64 +#endif + +/* Define linked list of signal handlers structure */ +struct SIG_hlist { + RETSIGTYPE (*handler)(); + struct SIG_hlist *next; +}; + +/* + * Define array of lists of signal handlers. Note that this depends on + * the implementation to initialize each element to a null pointer. + */ + +static struct SIG_hlist **SIG_handlers; + +/* Define array of default signal vectors */ + +#ifdef POSIX_SIGNALS +static struct sigaction *SIG_defaults; +#else +#ifdef BSD_SIGNALS +static struct sigvec *SIG_defaults; +#else +static RETSIGTYPE (**SIG_defaults) (int); +#endif +#endif + +/* Critical section housekeeping */ +static int SIG_crSectNest = 0; /* Nesting level */ +#ifdef POSIX_SIGNALS +static sigset_t SIG_crSectMask; /* Signal mask */ +#else +static int SIG_crSectMask; /* Signal mask */ +#endif + +/* + * Initialize the signal handler arrays + */ + +static int SIG_init() +{ + int i; +#ifdef POSIX_SIGNALS + sigset_t sigset_test; +#endif + + if (SIG_defaults && SIG_handlers) /* already allocated */ + return (0); + +#ifdef POSIX_SIGNALS + (void) sigfillset(&sigset_test); + for (i = 1; i < SIGMAX && sigismember(&sigset_test, i) == 1; i++) + ; + if (i < SIGMAX) + i = SIGMAX; + i++; + if (!SIG_defaults) + SIG_defaults = (struct sigaction *) + calloc(i, sizeof(struct sigaction)); + (void) sigemptyset(&SIG_crSectMask); +#else + i = SIGMAX+1; +#ifdef BSD_SIGNALS + if (!SIG_defaults) + SIG_defaults = (struct sigvec *) + calloc(i, sizeof(struct sigvec)); +#else + if (!SIG_defaults) + SIG_defaults = ( RETSIGTYPE (**) (int) ) + calloc( i, sizeof( RETSIGTYPE (**) (int) ) ); +#endif + SIG_crSectMask = 0; +#endif + if (!SIG_handlers) + SIG_handlers = (struct SIG_hlist **) + calloc(i, sizeof(struct SIG_hlist *)); + return (!SIG_defaults || !SIG_handlers); +} + + + +/* + * The following begins a critical section. + */ +void SIG_beginCrSect (void) +{ + if (SIG_init() == 0) + { + if (SIG_crSectNest == 0) + { +#ifdef POSIX_SIGNALS + sigset_t sigset_mask; + + (void) sigfillset(&sigset_mask); + (void) sigprocmask(SIG_SETMASK, + &sigset_mask, &SIG_crSectMask); +#else +#ifdef BSD_SIGNALS + SIG_crSectMask = sigblock(~0); +#else + /* TBD */ +#endif +#endif + } + SIG_crSectNest++; + } +} + + + +/* + * The following ends a critical section. + */ +void SIG_endCrSect (void) +{ + if (SIG_init() == 0) + { + SIG_crSectNest--; + if (SIG_crSectNest == 0) + { +#ifdef POSIX_SIGNALS + (void) sigprocmask(SIG_SETMASK, &SIG_crSectMask, NULL); +#else +#ifdef BSD_SIGNALS + (void) sigsetmask(SIG_crSectMask); +#else + /* TBD */ +#endif +#endif + } + } +} + + + +/* + * The following invokes each signal handler in the reverse order in which + * they were registered. + */ +static RETSIGTYPE SIG_handle (int sig) +{ + struct SIG_hlist *this; + + /* Dispatch signal handlers */ + /* This crit section stuff is a CVSism - we know that our interrupt + * handlers will always end up exiting and we don't want them to be + * interrupted themselves. + */ + SIG_beginCrSect(); + this = SIG_handlers[sig]; + while (this != (struct SIG_hlist *) NULL) + { + (*this->handler)(sig); + this = this->next; + } + SIG_endCrSect(); + + return; +} + +/* + * The following registers a signal handler. If the handler is already + * registered, it is not registered twice, nor is the order in which signal + * handlers are invoked changed. If this is the first signal handler + * registered for a given signal, the old sigvec structure is saved for + * restoration later. + */ + +int SIG_register(int sig, RETSIGTYPE (*fn)()) +{ + int val; + struct SIG_hlist *this; +#ifdef POSIX_SIGNALS + struct sigaction act; + sigset_t sigset_mask, sigset_omask; +#else +#ifdef BSD_SIGNALS + struct sigvec vec; + int mask; +#endif +#endif + + /* Initialize */ + if (SIG_init() != 0) + return (-1); + val = 0; + + /* Block this signal while we look at handler chain */ +#ifdef POSIX_SIGNALS + (void) sigemptyset(&sigset_mask); + (void) sigaddset(&sigset_mask, sig); + (void) sigprocmask(SIG_BLOCK, &sigset_mask, &sigset_omask); +#else +#ifdef BSD_SIGNALS + mask = sigblock(sigmask(sig)); +#endif +#endif + + /* See if this handler was already registered */ + this = SIG_handlers[sig]; + while (this != (struct SIG_hlist *) NULL) + { + if (this->handler == fn) break; + this = this->next; + } + + /* Register the new handler only if it is not already registered. */ + if (this == (struct SIG_hlist *) NULL) + { + + /* + * If this is the first handler registered for this signal, + * set up the signal handler dispatcher + */ + + if (SIG_handlers[sig] == (struct SIG_hlist *) NULL) + { +#ifdef POSIX_SIGNALS + act.sa_handler = SIG_handle; + (void) sigemptyset(&act.sa_mask); + act.sa_flags = 0; + val = sigaction(sig, &act, &SIG_defaults[sig]); +#else +#ifdef BSD_SIGNALS + memset (&vec, 0, sizeof (vec)); + vec.sv_handler = SIG_handle; + val = sigvec(sig, &vec, &SIG_defaults[sig]); +#else + if ((SIG_defaults[sig] = signal(sig, SIG_handle)) == SIG_ERR) + val = -1; +#endif +#endif + } + + /* If not, register it */ + if ((val == 0) && (this == (struct SIG_hlist *) NULL)) + { + this = (struct SIG_hlist *) + malloc(sizeof(struct SIG_hlist)); + if (this == NULL) + { + val = -1; + } + else + { + this->handler = fn; + this->next = SIG_handlers[sig]; + SIG_handlers[sig] = this; + } + } + } + + /* Unblock the signal */ +#ifdef POSIX_SIGNALS + (void) sigprocmask(SIG_SETMASK, &sigset_omask, NULL); +#else +#ifdef BSD_SIGNALS + (void) sigsetmask(mask); +#endif +#endif + + return val; +} + + + +/* + * The following deregisters a signal handler. If the last signal handler for + * a given signal is deregistered, the default sigvec information is restored. + */ + +int SIG_deregister(int sig, RETSIGTYPE (*fn)()) +{ + int val; + struct SIG_hlist *this; + struct SIG_hlist *last; +#ifdef POSIX_SIGNALS + sigset_t sigset_mask, sigset_omask; +#else +#ifdef BSD_SIGNALS + int mask; +#endif +#endif + + /* Initialize */ + if (SIG_init() != 0) + return (-1); + val = 0; + last = (struct SIG_hlist *) NULL; + + /* Block this signal while we look at handler chain */ +#ifdef POSIX_SIGNALS + (void) sigemptyset(&sigset_mask); + (void) sigaddset(&sigset_mask, sig); + (void) sigprocmask(SIG_BLOCK, &sigset_mask, &sigset_omask); +#else +#ifdef BSD_SIGNALS + mask = sigblock(sigmask(sig)); +#endif +#endif + + /* Search for the signal handler */ + this = SIG_handlers[sig]; + while ((this != (struct SIG_hlist *) NULL) && (this->handler != fn)) + { + last = this; + this = this->next; + } + + /* If it was registered, remove it */ + if (this != (struct SIG_hlist *) NULL) + { + if (last == (struct SIG_hlist *) NULL) + { + SIG_handlers[sig] = this->next; + } + else + { + last->next = this->next; + } + free((char *) this); + } + + /* Restore default behavior if there are no registered handlers */ + if (SIG_handlers[sig] == (struct SIG_hlist *) NULL) + { +#ifdef POSIX_SIGNALS + val = sigaction(sig, &SIG_defaults[sig], + (struct sigaction *) NULL); +#else +#ifdef BSD_SIGNALS + val = sigvec(sig, &SIG_defaults[sig], (struct sigvec *) NULL); +#else + if (signal(sig, SIG_defaults[sig]) == SIG_ERR) + val = -1; +#endif +#endif + } + + /* Unblock the signal */ +#ifdef POSIX_SIGNALS + (void) sigprocmask(SIG_SETMASK, &sigset_omask, NULL); +#else +#ifdef BSD_SIGNALS + (void) sigsetmask(mask); +#endif +#endif + + return val; +} + + + +/* + * Return nonzero if currently in a critical section. + * Otherwise return zero. + */ + +int SIG_inCrSect (void) +{ + return SIG_crSectNest > 0; +} diff --git a/lib/size_max.h b/lib/size_max.h new file mode 100644 index 0000000..ce2377b --- /dev/null +++ b/lib/size_max.h @@ -0,0 +1,27 @@ +/* size_max.h -- declare SIZE_MAX through system headers + Copyright (C) 2005 Free Software Foundation, Inc. + Written by Simon Josefsson. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef GNULIB_SIZE_MAX_H +#define GNULIB_SIZE_MAX_H + +# include <limits.h> +# if HAVE_STDINT_H +# include <stdint.h> +# endif + +#endif /* GNULIB_SIZE_MAX_H */ diff --git a/lib/stat-macros.h b/lib/stat-macros.h new file mode 100644 index 0000000..0957a22 --- /dev/null +++ b/lib/stat-macros.h @@ -0,0 +1,255 @@ +/* stat-related macros + + Copyright (C) 1993, 1994, 2001, 2002, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Paul Eggert and Jim Meyering. */ + +#ifndef STAT_MACROS_H +# define STAT_MACROS_H 1 + +# if ! defined S_ISREG && ! defined S_IFREG +# error "you must include <sys/stat.h> before including this file" +# endif + +# ifndef S_IFMT +# define S_IFMT 0170000 +# endif + +# if STAT_MACROS_BROKEN +# undef S_ISBLK +# undef S_ISCHR +# undef S_ISDIR +# undef S_ISDOOR +# undef S_ISFIFO +# undef S_ISLNK +# undef S_ISNAM +# undef S_ISMPB +# undef S_ISMPC +# undef S_ISNWK +# undef S_ISREG +# undef S_ISSOCK +# endif + + +# ifndef S_ISBLK +# ifdef S_IFBLK +# define S_ISBLK(m) (((m) & S_IFMT) == S_IFBLK) +# else +# define S_ISBLK(m) 0 +# endif +# endif + +# ifndef S_ISCHR +# ifdef S_IFCHR +# define S_ISCHR(m) (((m) & S_IFMT) == S_IFCHR) +# else +# define S_ISCHR(m) 0 +# endif +# endif + +# ifndef S_ISDIR +# ifdef S_IFDIR +# define S_ISDIR(m) (((m) & S_IFMT) == S_IFDIR) +# else +# define S_ISDIR(m) 0 +# endif +# endif + +# ifndef S_ISDOOR /* Solaris 2.5 and up */ +# ifdef S_IFDOOR +# define S_ISDOOR(m) (((m) & S_IFMT) == S_IFDOOR) +# else +# define S_ISDOOR(m) 0 +# endif +# endif + +# ifndef S_ISFIFO +# ifdef S_IFIFO +# define S_ISFIFO(m) (((m) & S_IFMT) == S_IFIFO) +# else +# define S_ISFIFO(m) 0 +# endif +# endif + +# ifndef S_ISLNK +# ifdef S_IFLNK +# define S_ISLNK(m) (((m) & S_IFMT) == S_IFLNK) +# else +# define S_ISLNK(m) 0 +# endif +# endif + +# ifndef S_ISMPB /* V7 */ +# ifdef S_IFMPB +# define S_ISMPB(m) (((m) & S_IFMT) == S_IFMPB) +# define S_ISMPC(m) (((m) & S_IFMT) == S_IFMPC) +# else +# define S_ISMPB(m) 0 +# define S_ISMPC(m) 0 +# endif +# endif + +# ifndef S_ISNAM /* Xenix */ +# ifdef S_IFNAM +# define S_ISNAM(m) (((m) & S_IFMT) == S_IFNAM) +# else +# define S_ISNAM(m) 0 +# endif +# endif + +# ifndef S_ISNWK /* HP/UX */ +# ifdef S_IFNWK +# define S_ISNWK(m) (((m) & S_IFMT) == S_IFNWK) +# else +# define S_ISNWK(m) 0 +# endif +# endif + +# ifndef S_ISREG +# ifdef S_IFREG +# define S_ISREG(m) (((m) & S_IFMT) == S_IFREG) +# else +# define S_ISREG(m) 0 +# endif +# endif + +# ifndef S_ISSOCK +# ifdef S_IFSOCK +# define S_ISSOCK(m) (((m) & S_IFMT) == S_IFSOCK) +# else +# define S_ISSOCK(m) 0 +# endif +# endif + + +# ifndef S_TYPEISMQ +# define S_TYPEISMQ(p) 0 +# endif + +# ifndef S_TYPEISTMO +# define S_TYPEISTMO(p) 0 +# endif + + +# ifndef S_TYPEISSEM +# ifdef S_INSEM +# define S_TYPEISSEM(p) (S_ISNAM ((p)->st_mode) && (p)->st_rdev == S_INSEM) +# else +# define S_TYPEISSEM(p) 0 +# endif +# endif + +# ifndef S_TYPEISSHM +# ifdef S_INSHD +# define S_TYPEISSHM(p) (S_ISNAM ((p)->st_mode) && (p)->st_rdev == S_INSHD) +# else +# define S_TYPEISSHM(p) 0 +# endif +# endif + +/* contiguous */ +# ifndef S_ISCTG +# define S_ISCTG(p) 0 +# endif + +/* Cray DMF (data migration facility): off line, with data */ +# ifndef S_ISOFD +# define S_ISOFD(p) 0 +# endif + +/* Cray DMF (data migration facility): off line, with no data */ +# ifndef S_ISOFL +# define S_ISOFL(p) 0 +# endif + +/* If any of the following are undefined, + define them to their de facto standard values. */ +# if !S_ISUID +# define S_ISUID 04000 +# endif +# if !S_ISGID +# define S_ISGID 02000 +# endif + +/* S_ISVTX is a common extension to POSIX. */ +# ifndef S_ISVTX +# define S_ISVTX 01000 +# endif + +# if !S_IRUSR && S_IREAD +# define S_IRUSR S_IREAD +# endif +# if !S_IRUSR +# define S_IRUSR 00400 +# endif +# if !S_IRGRP +# define S_IRGRP (S_IRUSR >> 3) +# endif +# if !S_IROTH +# define S_IROTH (S_IRUSR >> 6) +# endif + +# if !S_IWUSR && S_IWRITE +# define S_IWUSR S_IWRITE +# endif +# if !S_IWUSR +# define S_IWUSR 00200 +# endif +# if !S_IWGRP +# define S_IWGRP (S_IWUSR >> 3) +# endif +# if !S_IWOTH +# define S_IWOTH (S_IWUSR >> 6) +# endif + +# if !S_IXUSR && S_IEXEC +# define S_IXUSR S_IEXEC +# endif +# if !S_IXUSR +# define S_IXUSR 00100 +# endif +# if !S_IXGRP +# define S_IXGRP (S_IXUSR >> 3) +# endif +# if !S_IXOTH +# define S_IXOTH (S_IXUSR >> 6) +# endif + +# if !S_IRWXU +# define S_IRWXU (S_IRUSR | S_IWUSR | S_IXUSR) +# endif +# if !S_IRWXG +# define S_IRWXG (S_IRGRP | S_IWGRP | S_IXGRP) +# endif +# if !S_IRWXO +# define S_IRWXO (S_IROTH | S_IWOTH | S_IXOTH) +# endif + +/* S_IXUGO is a common extension to POSIX. */ +# if !S_IXUGO +# define S_IXUGO (S_IXUSR | S_IXGRP | S_IXOTH) +# endif + +# ifndef S_IRWXUGO +# define S_IRWXUGO (S_IRWXU | S_IRWXG | S_IRWXO) +# endif + +/* All the mode bits that can be affected by chmod. */ +# define CHMOD_MODE_BITS \ + (S_ISUID | S_ISGID | S_ISVTX | S_IRWXU | S_IRWXG | S_IRWXO) + +#endif /* STAT_MACROS_H */ diff --git a/lib/stdbool_.h b/lib/stdbool_.h new file mode 100644 index 0000000..3177ae0 --- /dev/null +++ b/lib/stdbool_.h @@ -0,0 +1,93 @@ +/* Copyright (C) 2001, 2002, 2003 Free Software Foundation, Inc. + Written by Bruno Haible <haible@clisp.cons.org>, 2001. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _STDBOOL_H +#define _STDBOOL_H + +/* ISO C 99 <stdbool.h> for platforms that lack it. */ + +/* Usage suggestions: + + Programs that use <stdbool.h> should be aware of some limitations + and standards compliance issues. + + Standards compliance: + + - <stdbool.h> must be #included before 'bool', 'false', 'true' + can be used. + + - You cannot assume that sizeof (bool) == 1. + + - Programs should not undefine the macros bool, true, and false, + as C99 lists that as an "obsolescent feature". + + Limitations of this substitute, when used in a C89 environment: + + - <stdbool.h> must be #included before the '_Bool' type can be used. + + - You cannot assume that _Bool is a typedef; it might be a macro. + + - In C99, casts and automatic conversions to '_Bool' or 'bool' are + performed in such a way that every nonzero value gets converted + to 'true', and zero gets converted to 'false'. This doesn't work + with this substitute. With this substitute, only the values 0 and 1 + give the expected result when converted to _Bool' or 'bool'. + + Also, it is suggested that programs use 'bool' rather than '_Bool'; + this isn't required, but 'bool' is more common. */ + + +/* 7.16. Boolean type and values */ + +/* BeOS <sys/socket.h> already #defines false 0, true 1. We use the same + definitions below, but temporarily we have to #undef them. */ +#ifdef __BEOS__ +# include <OS.h> /* defines bool but not _Bool */ +# undef false +# undef true +#endif + +/* For the sake of symbolic names in gdb, we define true and false as + enum constants, not only as macros. + It is tempting to write + typedef enum { false = 0, true = 1 } _Bool; + so that gdb prints values of type 'bool' symbolically. But if we do + this, values of type '_Bool' may promote to 'int' or 'unsigned int' + (see ISO C 99 6.7.2.2.(4)); however, '_Bool' must promote to 'int' + (see ISO C 99 6.3.1.1.(2)). So we add a negative value to the + enum; this ensures that '_Bool' promotes to 'int'. */ +#if !(defined __cplusplus || defined __BEOS__) +# if !@HAVE__BOOL@ +# if defined __SUNPRO_C && (__SUNPRO_C < 0x550 || __STDC__ == 1) + /* Avoid stupid "warning: _Bool is a keyword in ISO C99". */ +# define _Bool signed char +enum { false = 0, true = 1 }; +# else +typedef enum { _Bool_must_promote_to_int = -1, false = 0, true = 1 } _Bool; +# endif +# endif +#else +typedef bool _Bool; +#endif +#define bool _Bool + +/* The other macros must be usable in preprocessor directives. */ +#define false 0 +#define true 1 +#define __bool_true_false_are_defined 1 + +#endif /* _STDBOOL_H */ diff --git a/lib/stdint_.h b/lib/stdint_.h new file mode 100644 index 0000000..ea3619d --- /dev/null +++ b/lib/stdint_.h @@ -0,0 +1,283 @@ +/* Copyright (C) 2001-2002, 2004-2005 Free Software Foundation, Inc. + Written by Bruno Haible, Sam Steingold, Peter Burwood. + This file is part of gnulib. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _STDINT_H +#define _STDINT_H + +/* + * ISO C 99 <stdint.h> for platforms that lack it. + * <http://www.opengroup.org/onlinepubs/007904975/basedefs/stdint.h.html> + */ + +/* Get wchar_t, WCHAR_MIN, WCHAR_MAX. */ +#include <stddef.h> +/* Get CHAR_BIT, LONG_MIN, LONG_MAX, ULONG_MAX. */ +#include <limits.h> + +/* Get those types that are already defined in other system include files. */ +#if defined(__FreeBSD__) +# include <sys/inttypes.h> +#endif +#if defined(__sun) && HAVE_SYS_INTTYPES_H +# include <sys/inttypes.h> + /* Solaris 7 <sys/inttypes.h> has the types except the *_fast*_t types, and + the macros except for *_FAST*_*, INTPTR_MIN, PTRDIFF_MIN, PTRDIFF_MAX. + But note that <sys/int_types.h> contains only the type definitions! */ +# define _STDINT_H_HAVE_SYSTEM_INTTYPES +#endif +#if (defined(__hpux) || defined(_AIX)) && HAVE_INTTYPES_H +# include <inttypes.h> + /* HP-UX 10 <inttypes.h> has nearly everything, except UINT_LEAST8_MAX, + UINT_FAST8_MAX, PTRDIFF_MIN, PTRDIFF_MAX. */ + /* AIX 4 <inttypes.h> has nearly everything, except INTPTR_MIN, INTPTR_MAX, + UINTPTR_MAX, PTRDIFF_MIN, PTRDIFF_MAX. */ +# define _STDINT_H_HAVE_SYSTEM_INTTYPES +#endif +#if !(defined(UNIX_CYGWIN32) && defined(__BIT_TYPES_DEFINED__)) +# define _STDINT_H_NEED_SIGNED_INT_TYPES +#endif + +#if !defined(_STDINT_H_HAVE_SYSTEM_INTTYPES) + +/* 7.18.1.1. Exact-width integer types */ + +#if !defined(__FreeBSD__) + +#ifdef _STDINT_H_NEED_SIGNED_INT_TYPES +typedef signed char int8_t; +#endif +typedef unsigned char uint8_t; + +#ifdef _STDINT_H_NEED_SIGNED_INT_TYPES +typedef short int16_t; +#endif +typedef unsigned short uint16_t; + +#ifdef _STDINT_H_NEED_SIGNED_INT_TYPES +typedef int int32_t; +#endif +typedef unsigned int uint32_t; + +#if @HAVE_LONG_64BIT@ +#ifdef _STDINT_H_NEED_SIGNED_INT_TYPES +typedef long int64_t; +#endif +typedef unsigned long uint64_t; +#define _STDINT_H_HAVE_INT64 +#elif @HAVE_LONG_LONG_64BIT@ +#ifdef _STDINT_H_NEED_SIGNED_INT_TYPES +typedef long long int64_t; +#endif +typedef unsigned long long uint64_t; +#define _STDINT_H_HAVE_INT64 +#elif defined(_MSC_VER) +typedef __int64 int64_t; +typedef unsigned __int64 uint64_t; +#define _STDINT_H_HAVE_INT64 +#endif + +#endif /* !FreeBSD */ + +/* 7.18.1.2. Minimum-width integer types */ + +typedef int8_t int_least8_t; +typedef uint8_t uint_least8_t; +typedef int16_t int_least16_t; +typedef uint16_t uint_least16_t; +typedef int32_t int_least32_t; +typedef uint32_t uint_least32_t; +#ifdef _STDINT_H_HAVE_INT64 +typedef int64_t int_least64_t; +typedef uint64_t uint_least64_t; +#endif + +/* 7.18.1.3. Fastest minimum-width integer types */ + +typedef int32_t int_fast8_t; +typedef uint32_t uint_fast8_t; +typedef int32_t int_fast16_t; +typedef uint32_t uint_fast16_t; +typedef int32_t int_fast32_t; +typedef uint32_t uint_fast32_t; +#ifdef _STDINT_H_HAVE_INT64 +typedef int64_t int_fast64_t; +typedef uint64_t uint_fast64_t; +#endif + +/* 7.18.1.4. Integer types capable of holding object pointers */ + +#if !defined(__FreeBSD__) + +/* On some platforms (like IRIX6 MIPS with -n32) sizeof(void*) < sizeof(long), + but this doesn't matter here. */ +typedef long intptr_t; +typedef unsigned long uintptr_t; + +#endif /* !FreeBSD */ + +/* 7.18.1.5. Greatest-width integer types */ + +#ifdef _STDINT_H_HAVE_INT64 +typedef int64_t intmax_t; +typedef uint64_t uintmax_t; +#else +typedef int32_t intmax_t; +typedef uint32_t uintmax_t; +#endif + +/* 7.18.2. Limits of specified-width integer types */ + +#if !defined(__cplusplus) || defined(__STDC_LIMIT_MACROS) + +/* 7.18.2.1. Limits of exact-width integer types */ + +#define INT8_MIN -128 +#define INT8_MAX 127 +#define UINT8_MAX 255U +#define INT16_MIN -32768 +#define INT16_MAX 32767 +#define UINT16_MAX 65535U +#define INT32_MIN (~INT32_MAX) +#define INT32_MAX 2147483647 +#define UINT32_MAX 4294967295U +#ifdef _STDINT_H_HAVE_INT64 +#define INT64_MIN (~INT64_MAX) +#if @HAVE_LONG_64BIT@ +#define INT64_MAX 9223372036854775807L +#define UINT64_MAX 18446744073709551615UL +#elif @HAVE_LONG_LONG_64BIT@ +#define INT64_MAX 9223372036854775807LL +#define UINT64_MAX 18446744073709551615ULL +#elif defined(_MSC_VER) +#define INT64_MAX 9223372036854775807i64 +#define UINT64_MAX 18446744073709551615ui64 +#endif +#endif + +/* 7.18.2.2. Limits of minimum-width integer types */ + +#define INT_LEAST8_MIN INT8_MIN +#define INT_LEAST8_MAX INT8_MAX +#define UINT_LEAST8_MAX UINT8_MAX +#define INT_LEAST16_MIN INT16_MIN +#define INT_LEAST16_MAX INT16_MAX +#define UINT_LEAST16_MAX UINT16_MAX +#define INT_LEAST32_MIN INT32_MIN +#define INT_LEAST32_MAX INT32_MAX +#define UINT_LEAST32_MAX UINT32_MAX +#ifdef _STDINT_H_HAVE_INT64 +#define INT_LEAST64_MIN INT64_MIN +#define INT_LEAST64_MAX INT64_MAX +#define UINT_LEAST64_MAX UINT64_MAX +#endif + +/* 7.18.2.3. Limits of fastest minimum-width integer types */ + +#define INT_FAST8_MIN INT32_MIN +#define INT_FAST8_MAX INT32_MAX +#define UINT_FAST8_MAX UINT32_MAX +#define INT_FAST16_MIN INT32_MIN +#define INT_FAST16_MAX INT32_MAX +#define UINT_FAST16_MAX UINT32_MAX +#define INT_FAST32_MIN INT32_MIN +#define INT_FAST32_MAX INT32_MAX +#define UINT_FAST32_MAX UINT32_MAX +#ifdef _STDINT_H_HAVE_INT64 +#define INT_FAST64_MIN INT64_MIN +#define INT_FAST64_MAX INT64_MAX +#define UINT_FAST64_MAX UINT64_MAX +#endif + +/* 7.18.2.4. Limits of integer types capable of holding object pointers */ + +#define INTPTR_MIN LONG_MIN +#define INTPTR_MAX LONG_MAX +#define UINTPTR_MAX ULONG_MAX + +/* 7.18.2.5. Limits of greatest-width integer types */ + +#ifdef _STDINT_H_HAVE_INT64 +#define INTMAX_MIN INT64_MIN +#define INTMAX_MAX INT64_MAX +#define UINTMAX_MAX UINT64_MAX +#else +#define INTMAX_MIN INT32_MIN +#define INTMAX_MAX INT32_MAX +#define UINTMAX_MAX UINT32_MAX +#endif + +/* 7.18.3. Limits of other integer types */ + +#define PTRDIFF_MIN (~(ptrdiff_t)0 << (sizeof(ptrdiff_t)*CHAR_BIT-1)) +#define PTRDIFF_MAX (~PTRDIFF_MIN) + +/* This may be wrong... */ +#define SIG_ATOMIC_MIN 0 +#define SIG_ATOMIC_MAX 127 + +#define SIZE_MAX (~(size_t)0) + +/* wchar_t limits already defined in <stddef.h>. */ +/* wint_t limits already defined in <wchar.h>. */ + +#endif + +/* 7.18.4. Macros for integer constants */ + +#if !defined(__cplusplus) || defined(__STDC_CONSTANT_MACROS) + +/* 7.18.4.1. Macros for minimum-width integer constants */ + +#define INT8_C(x) x +#define UINT8_C(x) x##U +#define INT16_C(x) x +#define UINT16_C(x) x##U +#define INT32_C(x) x +#define UINT32_C(x) x##U +#if @HAVE_LONG_64BIT@ +#define INT64_C(x) x##L +#define UINT64_C(x) x##UL +#elif @HAVE_LONG_LONG_64BIT@ +#define INT64_C(x) x##LL +#define UINT64_C(x) x##ULL +#elif defined(_MSC_VER) +#define INT64_C(x) x##i64 +#define UINT64_C(x) x##ui64 +#endif + +/* 7.18.4.2. Macros for greatest-width integer constants */ + +#if @HAVE_LONG_64BIT@ +#define INTMAX_C(x) x##L +#define UINTMAX_C(x) x##UL +#elif @HAVE_LONG_LONG_64BIT@ +#define INTMAX_C(x) x##LL +#define UINTMAX_C(x) x##ULL +#elif defined(_MSC_VER) +#define INTMAX_C(x) x##i64 +#define UINTMAX_C(x) x##ui64 +#else +#define INTMAX_C(x) x +#define UINTMAX_C(x) x##U +#endif + +#endif + +#endif /* !_STDINT_H_HAVE_SYSTEM_INTTYPES */ + +#endif /* _STDINT_H */ diff --git a/lib/strcase.h b/lib/strcase.h new file mode 100644 index 0000000..e420798 --- /dev/null +++ b/lib/strcase.h @@ -0,0 +1,48 @@ +/* Case-insensitive string comparison functions. + Copyright (C) 1995-1996, 2001, 2003, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _STRCASE_H +#define _STRCASE_H + +#include <stddef.h> + + +#ifdef __cplusplus +extern "C" { +#endif + + +/* Compare strings S1 and S2, ignoring case, returning less than, equal to or + greater than zero if S1 is lexicographically less than, equal to or greater + than S2. + Note: This function may, in multibyte locales, return 0 for strings of + different lengths! */ +extern int strcasecmp (const char *s1, const char *s2); + +/* Compare no more than N characters of strings S1 and S2, ignoring case, + returning less than, equal to or greater than zero if S1 is + lexicographically less than, equal to or greater than S2. + Note: This function can not work correctly in multibyte locales. */ +extern int strncasecmp (const char *s1, const char *s2, size_t n); + + +#ifdef __cplusplus +} +#endif + + +#endif /* _STRCASE_H */ diff --git a/lib/strcasecmp.c b/lib/strcasecmp.c new file mode 100644 index 0000000..71f2eca --- /dev/null +++ b/lib/strcasecmp.c @@ -0,0 +1,98 @@ +/* Case-insensitive string comparison function. + Copyright (C) 1998, 1999, 2005 Free Software Foundation, Inc. + Written by Bruno Haible <bruno@clisp.org>, 2005, + based on earlier glibc code. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Specification. */ +#include "strcase.h" + +#include <ctype.h> + +#if HAVE_MBRTOWC +# include "mbuiter.h" +#endif + +#define TOLOWER(Ch) (isupper (Ch) ? tolower (Ch) : (Ch)) + +/* Compare strings S1 and S2, ignoring case, returning less than, equal to or + greater than zero if S1 is lexicographically less than, equal to or greater + than S2. + Note: This function may, in multibyte locales, return 0 for strings of + different lengths! */ +int +strcasecmp (const char *s1, const char *s2) +{ + if (s1 == s2) + return 0; + + /* Be careful not to look at the entire extent of s1 or s2 until needed. + This is useful because when two strings differ, the difference is + most often already in the very few first characters. */ +#if HAVE_MBRTOWC + if (MB_CUR_MAX > 1) + { + mbui_iterator_t iter1; + mbui_iterator_t iter2; + + mbui_init (iter1, s1); + mbui_init (iter2, s2); + + while (mbui_avail (iter1) && mbui_avail (iter2)) + { + int cmp = mb_casecmp (mbui_cur (iter1), mbui_cur (iter2)); + + if (cmp != 0) + return cmp; + + mbui_advance (iter1); + mbui_advance (iter2); + } + if (mbui_avail (iter1)) + /* s2 terminated before s1. */ + return 1; + if (mbui_avail (iter2)) + /* s1 terminated before s2. */ + return -1; + return 0; + } + else +#endif + { + const unsigned char *p1 = (const unsigned char *) s1; + const unsigned char *p2 = (const unsigned char *) s2; + unsigned char c1, c2; + + do + { + c1 = TOLOWER (*p1); + c2 = TOLOWER (*p2); + + if (c1 == '\0') + break; + + ++p1; + ++p2; + } + while (c1 == c2); + + return c1 - c2; + } +} diff --git a/lib/strdup.c b/lib/strdup.c new file mode 100644 index 0000000..d6d0116 --- /dev/null +++ b/lib/strdup.c @@ -0,0 +1,56 @@ +/* Copyright (C) 1991, 1996, 1997, 1998, 2002, 2003, 2004 Free Software + Foundation, Inc. + + This file is part of the GNU C Library. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#ifndef _LIBC +/* Get specification. */ +# include "strdup.h" +#endif + +#include <stdlib.h> +#include <string.h> + +#undef __strdup +#undef strdup + +#ifndef weak_alias +# define __strdup strdup +#endif + +/* Duplicate S, returning an identical malloc'd string. */ +char * +__strdup (const char *s) +{ + size_t len = strlen (s) + 1; + void *new = malloc (len); + + if (new == NULL) + return NULL; + + return (char *) memcpy (new, s, len); +} +#ifdef libc_hidden_def +libc_hidden_def (__strdup) +#endif +#ifdef weak_alias +weak_alias (__strdup, strdup) +#endif diff --git a/lib/strdup.h b/lib/strdup.h new file mode 100644 index 0000000..a0d5fb9 --- /dev/null +++ b/lib/strdup.h @@ -0,0 +1,29 @@ +/* strdup.h -- duplicate a string + Copyright (C) 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef STRDUP_H_ +#define STRDUP_H_ + +/* Get strdup declaration, if available. */ +#include <string.h> + +#if defined HAVE_DECL_STRDUP && !HAVE_DECL_STRDUP && !defined strdup +/* Duplicate S, returning an identical malloc'd string. */ +extern char *strdup (const char *s); +#endif + +#endif /* STRDUP_H_ */ diff --git a/lib/strerror.c b/lib/strerror.c new file mode 100644 index 0000000..623821d --- /dev/null +++ b/lib/strerror.c @@ -0,0 +1,49 @@ +/* strerror.c --- ANSI C compatible system error routine + + Copyright (C) 1986, 1988, 1989, 1991, 2002, 2003 Free Software + Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include <limits.h> + +/* Don't include <stdio.h>, since it may or may not declare + sys_errlist and its declarations may collide with ours. Just + declare the stuff that we need directly. Standard hosted C89 + implementations define strerror and they don't need this strerror + function, so take some liberties with the standard to cater to + ancient or limited freestanding implementations. */ +int sprintf (char *, char const *, ...); +extern int sys_nerr; +extern char *sys_errlist[]; + +char * +strerror (int n) +{ + static char const fmt[] = "Unknown error (%d)"; + static char mesg[sizeof fmt + sizeof n * CHAR_BIT / 3]; + + if (n < 0 || n >= sys_nerr) + { + sprintf (mesg, fmt, n); + return mesg; + } + else + return sys_errlist[n]; +} diff --git a/lib/strftime.c b/lib/strftime.c new file mode 100644 index 0000000..b68b4cf --- /dev/null +++ b/lib/strftime.c @@ -0,0 +1,1375 @@ +/* Copyright (C) 1991-1999, 2000, 2001, 2003, 2004, 2005 Free Software + Foundation, Inc. + + NOTE: The canonical source of this file is maintained with the GNU C Library. + Bugs can be reported to bug-glibc@prep.ai.mit.edu. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#ifdef _LIBC +# define HAVE_MBLEN 1 +# define HAVE_MBRLEN 1 +# define HAVE_STRUCT_ERA_ENTRY 1 +# define HAVE_TM_GMTOFF 1 +# define HAVE_TM_ZONE 1 +# define HAVE_TZNAME 1 +# define HAVE_TZSET 1 +# define MULTIBYTE_IS_FORMAT_SAFE 1 +# include "../locale/localeinfo.h" +#endif + +#include <ctype.h> +#include <sys/types.h> /* Some systems define `time_t' here. */ + +#ifdef TIME_WITH_SYS_TIME +# include <sys/time.h> +# include <time.h> +#else +# ifdef HAVE_SYS_TIME_H +# include <sys/time.h> +# else +# include <time.h> +# endif +#endif +#if HAVE_TZNAME +extern char *tzname[]; +#endif + +/* Do multibyte processing if multibytes are supported, unless + multibyte sequences are safe in formats. Multibyte sequences are + safe if they cannot contain byte sequences that look like format + conversion specifications. The GNU C Library uses UTF8 multibyte + encoding, which is safe for formats, but strftime.c can be used + with other C libraries that use unsafe encodings. */ +#define DO_MULTIBYTE (HAVE_MBLEN && HAVE_WCHAR_H && ! MULTIBYTE_IS_FORMAT_SAFE) + +#if DO_MULTIBYTE +# if HAVE_MBRLEN +# include <wchar.h> +# else + /* Simulate mbrlen with mblen as best we can. */ +# define mbstate_t int +# define mbrlen(s, n, ps) mblen (s, n) +# define mbsinit(ps) (*(ps) == 0) +# endif + static const mbstate_t mbstate_zero; +#endif + +#include <limits.h> +#include <stdbool.h> +#include <stddef.h> +#include <stdlib.h> +#include <string.h> + +#ifdef COMPILE_WIDE +# include <endian.h> +# define CHAR_T wchar_t +# define UCHAR_T unsigned int +# define L_(Str) L##Str +# define NLW(Sym) _NL_W##Sym + +# define MEMCPY(d, s, n) __wmemcpy (d, s, n) +# define STRLEN(s) __wcslen (s) + +#else +# define CHAR_T char +# define UCHAR_T unsigned char +# define L_(Str) Str +# define NLW(Sym) Sym + +# define MEMCPY(d, s, n) memcpy (d, s, n) +# define STRLEN(s) strlen (s) + +# ifdef _LIBC +# define MEMPCPY(d, s, n) __mempcpy (d, s, n) +# else +# ifndef HAVE_MEMPCPY +# define MEMPCPY(d, s, n) ((void *) ((char *) memcpy (d, s, n) + (n))) +# endif +# endif +#endif + +/* Shift A right by B bits portably, by dividing A by 2**B and + truncating towards minus infinity. A and B should be free of side + effects, and B should be in the range 0 <= B <= INT_BITS - 2, where + INT_BITS is the number of useful bits in an int. GNU code can + assume that INT_BITS is at least 32. + + ISO C99 says that A >> B is implementation-defined if A < 0. Some + implementations (e.g., UNICOS 9.0 on a Cray Y-MP EL) don't shift + right in the usual way when A < 0, so SHR falls back on division if + ordinary A >> B doesn't seem to be the usual signed shift. */ +#define SHR(a, b) \ + (-1 >> 1 == -1 \ + ? (a) >> (b) \ + : (a) / (1 << (b)) - ((a) % (1 << (b)) < 0)) + +/* Bound on length of the string representing an integer type or expression T. + Subtract 1 for the sign bit if t is signed; log10 (2.0) < 146/485; + add 1 for integer division truncation; add 1 more for a minus sign + if needed. */ +#define INT_STRLEN_BOUND(t) \ + ((sizeof (t) * CHAR_BIT - 1) * 146 / 485 + 2) + +#define TM_YEAR_BASE 1900 + +#ifndef __isleap +/* Nonzero if YEAR is a leap year (every 4 years, + except every 100th isn't, and every 400th is). */ +# define __isleap(year) \ + ((year) % 4 == 0 && ((year) % 100 != 0 || (year) % 400 == 0)) +#endif + + +#ifdef _LIBC +# define tzname __tzname +# define tzset __tzset +#endif + +#if !HAVE_TM_GMTOFF +/* Portable standalone applications should supply a "time_r.h" that + declares a POSIX-compliant localtime_r, for the benefit of older + implementations that lack localtime_r or have a nonstandard one. + See the gnulib time_r module for one way to implement this. */ +# include "time_r.h" +# undef __gmtime_r +# undef __localtime_r +# define __gmtime_r gmtime_r +# define __localtime_r localtime_r +#endif + + +#ifdef COMPILE_WIDE +# define memset_space(P, Len) (wmemset (P, L' ', Len), (P) += (Len)) +# define memset_zero(P, Len) (wmemset (P, L'0', Len), (P) += (Len)) +#else +# define memset_space(P, Len) (memset (P, ' ', Len), (P) += (Len)) +# define memset_zero(P, Len) (memset (P, '0', Len), (P) += (Len)) +#endif + +#define add(n, f) \ + do \ + { \ + int _n = (n); \ + int _delta = width - _n; \ + int _incr = _n + (_delta > 0 ? _delta : 0); \ + if ((size_t) _incr >= maxsize - i) \ + return 0; \ + if (p) \ + { \ + if (_delta > 0) \ + { \ + if (pad == L_('0')) \ + memset_zero (p, _delta); \ + else \ + memset_space (p, _delta); \ + } \ + f; \ + p += _n; \ + } \ + i += _incr; \ + } while (0) + +#define cpy(n, s) \ + add ((n), \ + if (to_lowcase) \ + memcpy_lowcase (p, (s), _n LOCALE_ARG); \ + else if (to_uppcase) \ + memcpy_uppcase (p, (s), _n LOCALE_ARG); \ + else \ + MEMCPY ((void *) p, (void const *) (s), _n)) + +#ifdef COMPILE_WIDE +# ifndef USE_IN_EXTENDED_LOCALE_MODEL +# undef __mbsrtowcs_l +# define __mbsrtowcs_l(d, s, l, st, loc) __mbsrtowcs (d, s, l, st) +# endif +# define widen(os, ws, l) \ + { \ + mbstate_t __st; \ + const char *__s = os; \ + memset (&__st, '\0', sizeof (__st)); \ + l = __mbsrtowcs_l (NULL, &__s, 0, &__st, loc); \ + ws = (wchar_t *) alloca ((l + 1) * sizeof (wchar_t)); \ + (void) __mbsrtowcs_l (ws, &__s, l, &__st, loc); \ + } +#endif + + +#if defined _LIBC && defined USE_IN_EXTENDED_LOCALE_MODEL +/* We use this code also for the extended locale handling where the + function gets as an additional argument the locale which has to be + used. To access the values we have to redefine the _NL_CURRENT + macro. */ +# define strftime __strftime_l +# define wcsftime __wcsftime_l +# undef _NL_CURRENT +# define _NL_CURRENT(category, item) \ + (current->values[_NL_ITEM_INDEX (item)].string) +# define LOCALE_ARG , loc +# define LOCALE_PARAM_PROTO , __locale_t loc +# define HELPER_LOCALE_ARG , current +#else +# define LOCALE_PARAM_PROTO +# define LOCALE_ARG +# ifdef _LIBC +# define HELPER_LOCALE_ARG , _NL_CURRENT_DATA (LC_TIME) +# else +# define HELPER_LOCALE_ARG +# endif +#endif + +#ifdef COMPILE_WIDE +# ifdef USE_IN_EXTENDED_LOCALE_MODEL +# define TOUPPER(Ch, L) __towupper_l (Ch, L) +# define TOLOWER(Ch, L) __towlower_l (Ch, L) +# else +# define TOUPPER(Ch, L) towupper (Ch) +# define TOLOWER(Ch, L) towlower (Ch) +# endif +#else +# ifdef _LIBC +# ifdef USE_IN_EXTENDED_LOCALE_MODEL +# define TOUPPER(Ch, L) __toupper_l (Ch, L) +# define TOLOWER(Ch, L) __tolower_l (Ch, L) +# else +# define TOUPPER(Ch, L) toupper (Ch) +# define TOLOWER(Ch, L) tolower (Ch) +# endif +# else +# define TOUPPER(Ch, L) (islower (Ch) ? toupper (Ch) : (Ch)) +# define TOLOWER(Ch, L) (isupper (Ch) ? tolower (Ch) : (Ch)) +# endif +#endif +/* We don't use `isdigit' here since the locale dependent + interpretation is not what we want here. We only need to accept + the arabic digits in the ASCII range. One day there is perhaps a + more reliable way to accept other sets of digits. */ +#define ISDIGIT(Ch) ((unsigned int) (Ch) - L_('0') <= 9) + +static CHAR_T * +memcpy_lowcase (CHAR_T *dest, const CHAR_T *src, + size_t len LOCALE_PARAM_PROTO) +{ + while (len-- > 0) + dest[len] = TOLOWER ((UCHAR_T) src[len], loc); + return dest; +} + +static CHAR_T * +memcpy_uppcase (CHAR_T *dest, const CHAR_T *src, + size_t len LOCALE_PARAM_PROTO) +{ + while (len-- > 0) + dest[len] = TOUPPER ((UCHAR_T) src[len], loc); + return dest; +} + + +#if ! HAVE_TM_GMTOFF +/* Yield the difference between *A and *B, + measured in seconds, ignoring leap seconds. */ +# define tm_diff ftime_tm_diff +static int +tm_diff (const struct tm *a, const struct tm *b) +{ + /* Compute intervening leap days correctly even if year is negative. + Take care to avoid int overflow in leap day calculations, + but it's OK to assume that A and B are close to each other. */ + int a4 = SHR (a->tm_year, 2) + SHR (TM_YEAR_BASE, 2) - ! (a->tm_year & 3); + int b4 = SHR (b->tm_year, 2) + SHR (TM_YEAR_BASE, 2) - ! (b->tm_year & 3); + int a100 = a4 / 25 - (a4 % 25 < 0); + int b100 = b4 / 25 - (b4 % 25 < 0); + int a400 = SHR (a100, 2); + int b400 = SHR (b100, 2); + int intervening_leap_days = (a4 - b4) - (a100 - b100) + (a400 - b400); + int years = a->tm_year - b->tm_year; + int days = (365 * years + intervening_leap_days + + (a->tm_yday - b->tm_yday)); + return (60 * (60 * (24 * days + (a->tm_hour - b->tm_hour)) + + (a->tm_min - b->tm_min)) + + (a->tm_sec - b->tm_sec)); +} +#endif /* ! HAVE_TM_GMTOFF */ + + + +/* The number of days from the first day of the first ISO week of this + year to the year day YDAY with week day WDAY. ISO weeks start on + Monday; the first ISO week has the year's first Thursday. YDAY may + be as small as YDAY_MINIMUM. */ +#define ISO_WEEK_START_WDAY 1 /* Monday */ +#define ISO_WEEK1_WDAY 4 /* Thursday */ +#define YDAY_MINIMUM (-366) +#ifdef __GNUC__ +__inline__ +#endif +static int +iso_week_days (int yday, int wday) +{ + /* Add enough to the first operand of % to make it nonnegative. */ + int big_enough_multiple_of_7 = (-YDAY_MINIMUM / 7 + 2) * 7; + return (yday + - (yday - wday + ISO_WEEK1_WDAY + big_enough_multiple_of_7) % 7 + + ISO_WEEK1_WDAY - ISO_WEEK_START_WDAY); +} + + +#if !(defined _NL_CURRENT || HAVE_STRFTIME) +static CHAR_T const weekday_name[][10] = + { + L_("Sunday"), L_("Monday"), L_("Tuesday"), L_("Wednesday"), + L_("Thursday"), L_("Friday"), L_("Saturday") + }; +static CHAR_T const month_name[][10] = + { + L_("January"), L_("February"), L_("March"), L_("April"), L_("May"), + L_("June"), L_("July"), L_("August"), L_("September"), L_("October"), + L_("November"), L_("December") + }; +#endif + + +/* When compiling this file, GNU applications can #define my_strftime + to a symbol (typically nstrftime) to get an extended strftime with + extra arguments UT and NS. Emacs is a special case for now, but + this Emacs-specific code can be removed once Emacs's config.h + defines my_strftime. */ +#if defined emacs && !defined my_strftime +# define my_strftime nstrftime +#endif + +#ifdef my_strftime +# define extra_args , ut, ns +# define extra_args_spec , int ut, int ns +#else +# ifdef COMPILE_WIDE +# define my_strftime wcsftime +# define nl_get_alt_digit _nl_get_walt_digit +# else +# define my_strftime strftime +# define nl_get_alt_digit _nl_get_alt_digit +# endif +# define extra_args +# define extra_args_spec +/* We don't have this information in general. */ +# define ut 0 +# define ns 0 +#endif + + +/* Write information from TP into S according to the format + string FORMAT, writing no more that MAXSIZE characters + (including the terminating '\0') and returning number of + characters written. If S is NULL, nothing will be written + anywhere, so to determine how many characters would be + written, use NULL for S and (size_t) -1 for MAXSIZE. */ +size_t +my_strftime (CHAR_T *s, size_t maxsize, const CHAR_T *format, + const struct tm *tp extra_args_spec LOCALE_PARAM_PROTO) +{ +#if defined _LIBC && defined USE_IN_EXTENDED_LOCALE_MODEL + struct locale_data *const current = loc->__locales[LC_TIME]; +#endif + + int hour12 = tp->tm_hour; +#ifdef _NL_CURRENT + /* We cannot make the following values variables since we must delay + the evaluation of these values until really needed since some + expressions might not be valid in every situation. The `struct tm' + might be generated by a strptime() call that initialized + only a few elements. Dereference the pointers only if the format + requires this. Then it is ok to fail if the pointers are invalid. */ +# define a_wkday \ + ((const CHAR_T *) _NL_CURRENT (LC_TIME, NLW(ABDAY_1) + tp->tm_wday)) +# define f_wkday \ + ((const CHAR_T *) _NL_CURRENT (LC_TIME, NLW(DAY_1) + tp->tm_wday)) +# define a_month \ + ((const CHAR_T *) _NL_CURRENT (LC_TIME, NLW(ABMON_1) + tp->tm_mon)) +# define f_month \ + ((const CHAR_T *) _NL_CURRENT (LC_TIME, NLW(MON_1) + tp->tm_mon)) +# define ampm \ + ((const CHAR_T *) _NL_CURRENT (LC_TIME, tp->tm_hour > 11 \ + ? NLW(PM_STR) : NLW(AM_STR))) + +# define aw_len STRLEN (a_wkday) +# define am_len STRLEN (a_month) +# define ap_len STRLEN (ampm) +#else +# if !HAVE_STRFTIME +# define f_wkday (weekday_name[tp->tm_wday]) +# define f_month (month_name[tp->tm_mon]) +# define a_wkday f_wkday +# define a_month f_month +# define ampm (L_("AMPM") + 2 * (tp->tm_hour > 11)) + + size_t aw_len = 3; + size_t am_len = 3; + size_t ap_len = 2; +# endif +#endif + const char *zone; + size_t i = 0; + CHAR_T *p = s; + const CHAR_T *f; +#if DO_MULTIBYTE && !defined COMPILE_WIDE + const char *format_end = NULL; +#endif + +#if ! defined _LIBC && ! HAVE_RUN_TZSET_TEST + /* Solaris 2.5.x and 2.6 tzset sometimes modify the storage returned + by localtime. On such systems, we must either use the tzset and + localtime wrappers to work around the bug (which sets + HAVE_RUN_TZSET_TEST) or make a copy of the structure. */ + struct tm copy = *tp; + tp = © +#endif + + zone = NULL; +#if HAVE_TM_ZONE + /* The POSIX test suite assumes that setting + the environment variable TZ to a new value before calling strftime() + will influence the result (the %Z format) even if the information in + TP is computed with a totally different time zone. + This is bogus: though POSIX allows bad behavior like this, + POSIX does not require it. Do the right thing instead. */ + zone = (const char *) tp->tm_zone; +#endif +#if HAVE_TZNAME + if (ut) + { + if (! (zone && *zone)) + zone = "GMT"; + } + else + { + /* POSIX.1 requires that local time zone information be used as + though strftime called tzset. */ +# if HAVE_TZSET + tzset (); +# endif + } +#endif + + if (hour12 > 12) + hour12 -= 12; + else + if (hour12 == 0) + hour12 = 12; + + for (f = format; *f != '\0'; ++f) + { + int pad = 0; /* Padding for number ('-', '_', or 0). */ + int modifier; /* Field modifier ('E', 'O', or 0). */ + int digits; /* Max digits for numeric format. */ + int number_value; /* Numeric value to be printed. */ + unsigned int u_number_value; /* (unsigned int) number_value. */ + bool negative_number; /* 1 if the number is negative. */ + const CHAR_T *subfmt; + CHAR_T *bufp; + CHAR_T buf[1 + (sizeof (int) < sizeof (time_t) + ? INT_STRLEN_BOUND (time_t) + : INT_STRLEN_BOUND (int))]; + int width = -1; + bool to_lowcase = false; + bool to_uppcase = false; + bool change_case = false; + int format_char; + +#if DO_MULTIBYTE && !defined COMPILE_WIDE + switch (*f) + { + case L_('%'): + break; + + case L_('\b'): case L_('\t'): case L_('\n'): + case L_('\v'): case L_('\f'): case L_('\r'): + case L_(' '): case L_('!'): case L_('"'): case L_('#'): case L_('&'): + case L_('\''): case L_('('): case L_(')'): case L_('*'): case L_('+'): + case L_(','): case L_('-'): case L_('.'): case L_('/'): case L_('0'): + case L_('1'): case L_('2'): case L_('3'): case L_('4'): case L_('5'): + case L_('6'): case L_('7'): case L_('8'): case L_('9'): case L_(':'): + case L_(';'): case L_('<'): case L_('='): case L_('>'): case L_('?'): + case L_('A'): case L_('B'): case L_('C'): case L_('D'): case L_('E'): + case L_('F'): case L_('G'): case L_('H'): case L_('I'): case L_('J'): + case L_('K'): case L_('L'): case L_('M'): case L_('N'): case L_('O'): + case L_('P'): case L_('Q'): case L_('R'): case L_('S'): case L_('T'): + case L_('U'): case L_('V'): case L_('W'): case L_('X'): case L_('Y'): + case L_('Z'): case L_('['): case L_('\\'): case L_(']'): case L_('^'): + case L_('_'): case L_('a'): case L_('b'): case L_('c'): case L_('d'): + case L_('e'): case L_('f'): case L_('g'): case L_('h'): case L_('i'): + case L_('j'): case L_('k'): case L_('l'): case L_('m'): case L_('n'): + case L_('o'): case L_('p'): case L_('q'): case L_('r'): case L_('s'): + case L_('t'): case L_('u'): case L_('v'): case L_('w'): case L_('x'): + case L_('y'): case L_('z'): case L_('{'): case L_('|'): case L_('}'): + case L_('~'): + /* The C Standard requires these 98 characters (plus '%') to + be in the basic execution character set. None of these + characters can start a multibyte sequence, so they need + not be analyzed further. */ + add (1, *p = *f); + continue; + + default: + /* Copy this multibyte sequence until we reach its end, find + an error, or come back to the initial shift state. */ + { + mbstate_t mbstate = mbstate_zero; + size_t len = 0; + size_t fsize; + + if (! format_end) + format_end = f + strlen (f) + 1; + fsize = format_end - f; + + do + { + size_t bytes = mbrlen (f + len, fsize - len, &mbstate); + + if (bytes == 0) + break; + + if (bytes == (size_t) -2) + { + len += strlen (f + len); + break; + } + + if (bytes == (size_t) -1) + { + len++; + break; + } + + len += bytes; + } + while (! mbsinit (&mbstate)); + + cpy (len, f); + f += len - 1; + continue; + } + } + +#else /* ! DO_MULTIBYTE */ + + /* Either multibyte encodings are not supported, they are + safe for formats, so any non-'%' byte can be copied through, + or this is the wide character version. */ + if (*f != L_('%')) + { + add (1, *p = *f); + continue; + } + +#endif /* ! DO_MULTIBYTE */ + + /* Check for flags that can modify a format. */ + while (1) + { + switch (*++f) + { + /* This influences the number formats. */ + case L_('_'): + case L_('-'): + case L_('0'): + pad = *f; + continue; + + /* This changes textual output. */ + case L_('^'): + to_uppcase = true; + continue; + case L_('#'): + change_case = true; + continue; + + default: + break; + } + break; + } + + /* As a GNU extension we allow to specify the field width. */ + if (ISDIGIT (*f)) + { + width = 0; + do + { + if (width > INT_MAX / 10 + || (width == INT_MAX / 10 && *f - L_('0') > INT_MAX % 10)) + /* Avoid overflow. */ + width = INT_MAX; + else + { + width *= 10; + width += *f - L_('0'); + } + ++f; + } + while (ISDIGIT (*f)); + } + + /* Check for modifiers. */ + switch (*f) + { + case L_('E'): + case L_('O'): + modifier = *f++; + break; + + default: + modifier = 0; + break; + } + + /* Now do the specified format. */ + format_char = *f; + switch (format_char) + { +#define DO_NUMBER(d, v) \ + digits = d; \ + number_value = v; goto do_number +#define DO_SIGNED_NUMBER(d, negative, v) \ + digits = d; \ + negative_number = negative; \ + u_number_value = v; goto do_signed_number +#define DO_NUMBER_SPACEPAD(d, v) \ + digits = d; \ + number_value = v; goto do_number_spacepad + + case L_('%'): + if (modifier != 0) + goto bad_format; + add (1, *p = *f); + break; + + case L_('a'): + if (modifier != 0) + goto bad_format; + if (change_case) + { + to_uppcase = true; + to_lowcase = false; + } +#if defined _NL_CURRENT || !HAVE_STRFTIME + cpy (aw_len, a_wkday); + break; +#else + goto underlying_strftime; +#endif + + case 'A': + if (modifier != 0) + goto bad_format; + if (change_case) + { + to_uppcase = true; + to_lowcase = false; + } +#if defined _NL_CURRENT || !HAVE_STRFTIME + cpy (STRLEN (f_wkday), f_wkday); + break; +#else + goto underlying_strftime; +#endif + + case L_('b'): + case L_('h'): + if (change_case) + { + to_uppcase = true; + to_lowcase = false; + } + if (modifier != 0) + goto bad_format; +#if defined _NL_CURRENT || !HAVE_STRFTIME + cpy (am_len, a_month); + break; +#else + goto underlying_strftime; +#endif + + case L_('B'): + if (modifier != 0) + goto bad_format; + if (change_case) + { + to_uppcase = true; + to_lowcase = false; + } +#if defined _NL_CURRENT || !HAVE_STRFTIME + cpy (STRLEN (f_month), f_month); + break; +#else + goto underlying_strftime; +#endif + + case L_('c'): + if (modifier == L_('O')) + goto bad_format; +#ifdef _NL_CURRENT + if (! (modifier == 'E' + && (*(subfmt = + (const CHAR_T *) _NL_CURRENT (LC_TIME, + NLW(ERA_D_T_FMT))) + != '\0'))) + subfmt = (const CHAR_T *) _NL_CURRENT (LC_TIME, NLW(D_T_FMT)); +#else +# if HAVE_STRFTIME + goto underlying_strftime; +# else + subfmt = L_("%a %b %e %H:%M:%S %Y"); +# endif +#endif + + subformat: + { + CHAR_T *old_start = p; + size_t len = my_strftime (NULL, (size_t) -1, subfmt, + tp extra_args LOCALE_ARG); + add (len, my_strftime (p, maxsize - i, subfmt, + tp extra_args LOCALE_ARG)); + + if (to_uppcase) + while (old_start < p) + { + *old_start = TOUPPER ((UCHAR_T) *old_start, loc); + ++old_start; + } + } + break; + +#if HAVE_STRFTIME && ! (defined _NL_CURRENT && HAVE_STRUCT_ERA_ENTRY) + underlying_strftime: + { + /* The relevant information is available only via the + underlying strftime implementation, so use that. */ + char ufmt[5]; + char *u = ufmt; + char ubuf[1024]; /* enough for any single format in practice */ + size_t len; + /* Make sure we're calling the actual underlying strftime. + In some cases, config.h contains something like + "#define strftime rpl_strftime". */ +# ifdef strftime +# undef strftime + size_t strftime (); +# endif + + /* The space helps distinguish strftime failure from empty + output. */ + *u++ = ' '; + *u++ = '%'; + if (modifier != 0) + *u++ = modifier; + *u++ = format_char; + *u = '\0'; + len = strftime (ubuf, sizeof ubuf, ufmt, tp); + if (len != 0) + cpy (len - 1, ubuf + 1); + } + break; +#endif + + case L_('C'): + if (modifier == L_('O')) + goto bad_format; + if (modifier == L_('E')) + { +#if HAVE_STRUCT_ERA_ENTRY + struct era_entry *era = _nl_get_era_entry (tp HELPER_LOCALE_ARG); + if (era) + { +# ifdef COMPILE_WIDE + size_t len = __wcslen (era->era_wname); + cpy (len, era->era_wname); +# else + size_t len = strlen (era->era_name); + cpy (len, era->era_name); +# endif + break; + } +#else +# if HAVE_STRFTIME + goto underlying_strftime; +# endif +#endif + } + + { + int century = tp->tm_year / 100 + TM_YEAR_BASE / 100; + century -= tp->tm_year % 100 < 0 && 0 < century; + DO_SIGNED_NUMBER (2, tp->tm_year < - TM_YEAR_BASE, century); + } + + case L_('x'): + if (modifier == L_('O')) + goto bad_format; +#ifdef _NL_CURRENT + if (! (modifier == L_('E') + && (*(subfmt = + (const CHAR_T *)_NL_CURRENT (LC_TIME, NLW(ERA_D_FMT))) + != L_('\0')))) + subfmt = (const CHAR_T *) _NL_CURRENT (LC_TIME, NLW(D_FMT)); + goto subformat; +#else +# if HAVE_STRFTIME + goto underlying_strftime; +# else + /* Fall through. */ +# endif +#endif + case L_('D'): + if (modifier != 0) + goto bad_format; + subfmt = L_("%m/%d/%y"); + goto subformat; + + case L_('d'): + if (modifier == L_('E')) + goto bad_format; + + DO_NUMBER (2, tp->tm_mday); + + case L_('e'): + if (modifier == L_('E')) + goto bad_format; + + DO_NUMBER_SPACEPAD (2, tp->tm_mday); + + /* All numeric formats set DIGITS and NUMBER_VALUE (or U_NUMBER_VALUE) + and then jump to one of these three labels. */ + + do_number_spacepad: + /* Force `_' flag unless overridden by `0' or `-' flag. */ + if (pad != L_('0') && pad != L_('-')) + pad = L_('_'); + + do_number: + /* Format NUMBER_VALUE according to the MODIFIER flag. */ + negative_number = number_value < 0; + u_number_value = number_value; + + do_signed_number: + /* Format U_NUMBER_VALUE according to the MODIFIER flag. + NEGATIVE_NUMBER is nonzero if the original number was + negative; in this case it was converted directly to + unsigned int (i.e., modulo (UINT_MAX + 1)) without + negating it. */ + if (modifier == L_('O') && !negative_number) + { +#ifdef _NL_CURRENT + /* Get the locale specific alternate representation of + the number. If none exist NULL is returned. */ + const CHAR_T *cp = nl_get_alt_digit (u_number_value + HELPER_LOCALE_ARG); + + if (cp != NULL) + { + size_t digitlen = STRLEN (cp); + if (digitlen != 0) + { + cpy (digitlen, cp); + break; + } + } +#else +# if HAVE_STRFTIME + goto underlying_strftime; +# endif +#endif + } + + bufp = buf + sizeof (buf) / sizeof (buf[0]); + + if (negative_number) + u_number_value = - u_number_value; + + do + { + *--bufp = u_number_value % 10 + L_('0'); + u_number_value /= 10; + } + while (u_number_value != 0); + + do_number_sign_and_padding: + if (digits < width) + digits = width; + + if (negative_number) + *--bufp = L_('-'); + + if (pad != L_('-')) + { + int padding = digits - (buf + (sizeof (buf) / sizeof (buf[0])) + - bufp); + + if (padding > 0) + { + if (pad == L_('_')) + { + if ((size_t) padding >= maxsize - i) + return 0; + + if (p) + memset_space (p, padding); + i += padding; + width = width > padding ? width - padding : 0; + } + else + { + if ((size_t) digits >= maxsize - i) + return 0; + + if (negative_number) + { + ++bufp; + + if (p) + *p++ = L_('-'); + ++i; + } + + if (p) + memset_zero (p, padding); + i += padding; + width = 0; + } + } + } + + cpy (buf + sizeof (buf) / sizeof (buf[0]) - bufp, bufp); + break; + + case L_('F'): + if (modifier != 0) + goto bad_format; + subfmt = L_("%Y-%m-%d"); + goto subformat; + + case L_('H'): + if (modifier == L_('E')) + goto bad_format; + + DO_NUMBER (2, tp->tm_hour); + + case L_('I'): + if (modifier == L_('E')) + goto bad_format; + + DO_NUMBER (2, hour12); + + case L_('k'): /* GNU extension. */ + if (modifier == L_('E')) + goto bad_format; + + DO_NUMBER_SPACEPAD (2, tp->tm_hour); + + case L_('l'): /* GNU extension. */ + if (modifier == L_('E')) + goto bad_format; + + DO_NUMBER_SPACEPAD (2, hour12); + + case L_('j'): + if (modifier == L_('E')) + goto bad_format; + + DO_SIGNED_NUMBER (3, tp->tm_yday < -1, tp->tm_yday + 1U); + + case L_('M'): + if (modifier == L_('E')) + goto bad_format; + + DO_NUMBER (2, tp->tm_min); + + case L_('m'): + if (modifier == L_('E')) + goto bad_format; + + DO_SIGNED_NUMBER (2, tp->tm_mon < -1, tp->tm_mon + 1U); + +#ifndef _LIBC + case L_('N'): /* GNU extension. */ + if (modifier == L_('E')) + goto bad_format; + + number_value = ns; + if (width != -1) + { + /* Take an explicit width less than 9 as a precision. */ + int j; + for (j = width; j < 9; j++) + number_value /= 10; + } + + DO_NUMBER (9, number_value); +#endif + + case L_('n'): + add (1, *p = L_('\n')); + break; + + case L_('P'): + to_lowcase = true; +#if !defined _NL_CURRENT && HAVE_STRFTIME + format_char = L_('p'); +#endif + /* FALLTHROUGH */ + + case L_('p'): + if (change_case) + { + to_uppcase = false; + to_lowcase = true; + } +#if defined _NL_CURRENT || !HAVE_STRFTIME + cpy (ap_len, ampm); + break; +#else + goto underlying_strftime; +#endif + + case L_('R'): + subfmt = L_("%H:%M"); + goto subformat; + + case L_('r'): +#if !defined _NL_CURRENT && HAVE_STRFTIME + goto underlying_strftime; +#else +# ifdef _NL_CURRENT + if (*(subfmt = (const CHAR_T *) _NL_CURRENT (LC_TIME, + NLW(T_FMT_AMPM))) + == L_('\0')) +# endif + subfmt = L_("%I:%M:%S %p"); + goto subformat; +#endif + + case L_('S'): + if (modifier == L_('E')) + goto bad_format; + + DO_NUMBER (2, tp->tm_sec); + + case L_('s'): /* GNU extension. */ + { + struct tm ltm; + time_t t; + + ltm = *tp; + t = mktime (<m); + + /* Generate string value for T using time_t arithmetic; + this works even if sizeof (long) < sizeof (time_t). */ + + bufp = buf + sizeof (buf) / sizeof (buf[0]); + negative_number = t < 0; + + do + { + int d = t % 10; + t /= 10; + *--bufp = (negative_number ? -d : d) + L_('0'); + } + while (t != 0); + + digits = 1; + goto do_number_sign_and_padding; + } + + case L_('X'): + if (modifier == L_('O')) + goto bad_format; +#ifdef _NL_CURRENT + if (! (modifier == L_('E') + && (*(subfmt = + (const CHAR_T *) _NL_CURRENT (LC_TIME, NLW(ERA_T_FMT))) + != L_('\0')))) + subfmt = (const CHAR_T *) _NL_CURRENT (LC_TIME, NLW(T_FMT)); + goto subformat; +#else +# if HAVE_STRFTIME + goto underlying_strftime; +# else + /* Fall through. */ +# endif +#endif + case L_('T'): + subfmt = L_("%H:%M:%S"); + goto subformat; + + case L_('t'): + add (1, *p = L_('\t')); + break; + + case L_('u'): + DO_NUMBER (1, (tp->tm_wday - 1 + 7) % 7 + 1); + + case L_('U'): + if (modifier == L_('E')) + goto bad_format; + + DO_NUMBER (2, (tp->tm_yday - tp->tm_wday + 7) / 7); + + case L_('V'): + case L_('g'): + case L_('G'): + if (modifier == L_('E')) + goto bad_format; + { + /* YEAR is a leap year if and only if (tp->tm_year + TM_YEAR_BASE) + is a leap year, except that YEAR and YEAR - 1 both work + correctly even when (tp->tm_year + TM_YEAR_BASE) would + overflow. */ + int year = (tp->tm_year + + (tp->tm_year < 0 + ? TM_YEAR_BASE % 400 + : TM_YEAR_BASE % 400 - 400)); + int year_adjust = 0; + int days = iso_week_days (tp->tm_yday, tp->tm_wday); + + if (days < 0) + { + /* This ISO week belongs to the previous year. */ + year_adjust = -1; + days = iso_week_days (tp->tm_yday + (365 + __isleap (year - 1)), + tp->tm_wday); + } + else + { + int d = iso_week_days (tp->tm_yday - (365 + __isleap (year)), + tp->tm_wday); + if (0 <= d) + { + /* This ISO week belongs to the next year. */ + year_adjust = 1; + days = d; + } + } + + switch (*f) + { + case L_('g'): + { + int yy = (tp->tm_year % 100 + year_adjust) % 100; + DO_NUMBER (2, (0 <= yy + ? yy + : tp->tm_year < -TM_YEAR_BASE - year_adjust + ? -yy + : yy + 100)); + } + + case L_('G'): + DO_SIGNED_NUMBER (4, tp->tm_year < -TM_YEAR_BASE - year_adjust, + (tp->tm_year + (unsigned int) TM_YEAR_BASE + + year_adjust)); + + default: + DO_NUMBER (2, days / 7 + 1); + } + } + + case L_('W'): + if (modifier == L_('E')) + goto bad_format; + + DO_NUMBER (2, (tp->tm_yday - (tp->tm_wday - 1 + 7) % 7 + 7) / 7); + + case L_('w'): + if (modifier == L_('E')) + goto bad_format; + + DO_NUMBER (1, tp->tm_wday); + + case L_('Y'): + if (modifier == 'E') + { +#if HAVE_STRUCT_ERA_ENTRY + struct era_entry *era = _nl_get_era_entry (tp HELPER_LOCALE_ARG); + if (era) + { +# ifdef COMPILE_WIDE + subfmt = era->era_wformat; +# else + subfmt = era->era_format; +# endif + goto subformat; + } +#else +# if HAVE_STRFTIME + goto underlying_strftime; +# endif +#endif + } + if (modifier == L_('O')) + goto bad_format; + else + DO_SIGNED_NUMBER (4, tp->tm_year < -TM_YEAR_BASE, + tp->tm_year + (unsigned int) TM_YEAR_BASE); + + case L_('y'): + if (modifier == L_('E')) + { +#if HAVE_STRUCT_ERA_ENTRY + struct era_entry *era = _nl_get_era_entry (tp HELPER_LOCALE_ARG); + if (era) + { + int delta = tp->tm_year - era->start_date[0]; + DO_NUMBER (1, (era->offset + + delta * era->absolute_direction)); + } +#else +# if HAVE_STRFTIME + goto underlying_strftime; +# endif +#endif + } + + { + int yy = tp->tm_year % 100; + if (yy < 0) + yy = tp->tm_year < - TM_YEAR_BASE ? -yy : yy + 100; + DO_NUMBER (2, yy); + } + + case L_('Z'): + if (change_case) + { + to_uppcase = false; + to_lowcase = true; + } + +#if HAVE_TZNAME + /* The tzset() call might have changed the value. */ + if (!(zone && *zone) && tp->tm_isdst >= 0) + zone = tzname[tp->tm_isdst != 0]; +#endif + if (! zone) + zone = ""; + +#ifdef COMPILE_WIDE + { + /* The zone string is always given in multibyte form. We have + to transform it first. */ + wchar_t *wczone; + size_t len; + widen (zone, wczone, len); + cpy (len, wczone); + } +#else + cpy (strlen (zone), zone); +#endif + break; + + case L_('z'): + if (tp->tm_isdst < 0) + break; + + { + int diff; +#if HAVE_TM_GMTOFF + diff = tp->tm_gmtoff; +#else + if (ut) + diff = 0; + else + { + struct tm gtm; + struct tm ltm; + time_t lt; + + ltm = *tp; + lt = mktime (<m); + + if (lt == (time_t) -1) + { + /* mktime returns -1 for errors, but -1 is also a + valid time_t value. Check whether an error really + occurred. */ + struct tm tm; + + if (! __localtime_r (<, &tm) + || ((ltm.tm_sec ^ tm.tm_sec) + | (ltm.tm_min ^ tm.tm_min) + | (ltm.tm_hour ^ tm.tm_hour) + | (ltm.tm_mday ^ tm.tm_mday) + | (ltm.tm_mon ^ tm.tm_mon) + | (ltm.tm_year ^ tm.tm_year))) + break; + } + + if (! __gmtime_r (<, >m)) + break; + + diff = tm_diff (<m, >m); + } +#endif + + if (diff < 0) + { + add (1, *p = L_('-')); + diff = -diff; + } + else + add (1, *p = L_('+')); + + diff /= 60; + DO_NUMBER (4, (diff / 60) * 100 + diff % 60); + } + + case L_('\0'): /* GNU extension: % at end of format. */ + --f; + /* Fall through. */ + default: + /* Unknown format; output the format, including the '%', + since this is most likely the right thing to do if a + multibyte string has been misparsed. */ + bad_format: + { + int flen; + for (flen = 1; f[1 - flen] != L_('%'); flen++) + continue; + cpy (flen, &f[1 - flen]); + } + break; + } + } + + if (p && maxsize != 0) + *p = L_('\0'); + return i; +} +#ifdef _LIBC +libc_hidden_def (my_strftime) +#endif + + +#ifdef emacs +/* For Emacs we have a separate interface which corresponds to the normal + strftime function plus the ut argument, but without the ns argument. */ +size_t +emacs_strftimeu (char *s, size_t maxsize, const char *format, + const struct tm *tp, int ut) +{ + return my_strftime (s, maxsize, format, tp, ut, 0); +} +#endif diff --git a/lib/strftime.h b/lib/strftime.h new file mode 100644 index 0000000..16b996e --- /dev/null +++ b/lib/strftime.h @@ -0,0 +1,21 @@ +/* declarations for strftime.c + + Copyright (C) 2002, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#include <time.h> + +size_t nstrftime (char *, size_t, char const *, struct tm const *, int, int); diff --git a/lib/stripslash.c b/lib/stripslash.c new file mode 100644 index 0000000..9b55da4 --- /dev/null +++ b/lib/stripslash.c @@ -0,0 +1,40 @@ +/* stripslash.c -- remove redundant trailing slashes from a file name + + Copyright (C) 1990, 2001, 2003, 2004, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "dirname.h" + +/* Remove trailing slashes from FILE. + Return true if a trailing slash was removed. + This is useful when using file name completion from a shell that + adds a "/" after directory names (such as tcsh and bash), because + the Unix rename and rmdir system calls return an "Invalid argument" error + when given a file that ends in "/" (except for the root directory). */ + +bool +strip_trailing_slashes (char *file) +{ + char *base = base_name (file); + char *base_lim = base + base_len (base); + bool had_slash = (*base_lim != '\0'); + *base_lim = '\0'; + return had_slash; +} diff --git a/lib/strncasecmp.c b/lib/strncasecmp.c new file mode 100644 index 0000000..72ead01 --- /dev/null +++ b/lib/strncasecmp.c @@ -0,0 +1,58 @@ +/* strncasecmp.c -- case insensitive string comparator + Copyright (C) 1998, 1999 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Specification. */ +#include "strcase.h" + +#include <ctype.h> + +#define TOLOWER(Ch) (isupper (Ch) ? tolower (Ch) : (Ch)) + +/* Compare no more than N bytes of strings S1 and S2, + ignoring case, returning less than, equal to or + greater than zero if S1 is lexicographically less + than, equal to or greater than S2. */ + +int +strncasecmp (const char *s1, const char *s2, size_t n) +{ + register const unsigned char *p1 = (const unsigned char *) s1; + register const unsigned char *p2 = (const unsigned char *) s2; + unsigned char c1, c2; + + if (p1 == p2 || n == 0) + return 0; + + do + { + c1 = TOLOWER (*p1); + c2 = TOLOWER (*p2); + + if (--n == 0 || c1 == '\0') + break; + + ++p1; + ++p2; + } + while (c1 == c2); + + return c1 - c2; +} diff --git a/lib/strnlen1.c b/lib/strnlen1.c new file mode 100644 index 0000000..f3338e7 --- /dev/null +++ b/lib/strnlen1.c @@ -0,0 +1,39 @@ +/* Find the length of STRING + 1, but scan at most MAXLEN bytes. + Copyright (C) 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify it + under the terms of the GNU Library General Public License as published + by the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Library General Public License for more details. + + You should have received a copy of the GNU Library General Public + License along with this program; if not, write to the Free Software + Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, + USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Specification. */ +#include "strnlen1.h" + +#include <string.h> + +/* Find the length of STRING + 1, but scan at most MAXLEN bytes. + If no '\0' terminator is found in that many characters, return MAXLEN. */ +/* This is the same as strnlen (string, maxlen - 1) + 1. */ +size_t +strnlen1 (const char *string, size_t maxlen) +{ + const char *end = memchr (string, '\0', maxlen); + if (end != NULL) + return end - string + 1; + else + return maxlen; +} diff --git a/lib/strnlen1.h b/lib/strnlen1.h new file mode 100644 index 0000000..4f913ff --- /dev/null +++ b/lib/strnlen1.h @@ -0,0 +1,41 @@ +/* Find the length of STRING + 1, but scan at most MAXLEN bytes. + Copyright (C) 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify it + under the terms of the GNU Library General Public License as published + by the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Library General Public License for more details. + + You should have received a copy of the GNU Library General Public + License along with this program; if not, write to the Free Software + Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, + USA. */ + +#ifndef _STRNLEN1_H +#define _STRNLEN1_H + +#include <stddef.h> + + +#ifdef __cplusplus +extern "C" { +#endif + + +/* Find the length of STRING + 1, but scan at most MAXLEN bytes. + If no '\0' terminator is found in that many characters, return MAXLEN. */ +/* This is the same as strnlen (string, maxlen - 1) + 1. */ +extern size_t strnlen1 (const char *string, size_t maxlen); + + +#ifdef __cplusplus +} +#endif + + +#endif /* _STRNLEN1_H */ diff --git a/lib/strstr.c b/lib/strstr.c new file mode 100644 index 0000000..e94cef7 --- /dev/null +++ b/lib/strstr.c @@ -0,0 +1,128 @@ +/* Searching in a string. + Copyright (C) 2005 Free Software Foundation, Inc. + Written by Bruno Haible <bruno@clisp.org>, 2005. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Specification. */ +#include "strstr.h" + +#include <stddef.h> /* for NULL */ + +#if HAVE_MBRTOWC +# include "mbuiter.h" +#endif + +/* Find the first occurrence of NEEDLE in HAYSTACK. */ +char * +strstr (const char *haystack, const char *needle) +{ + /* Be careful not to look at the entire extent of haystack or needle + until needed. This is useful because of these two cases: + - haystack may be very long, and a match of needle found early, + - needle may be very long, and not even a short initial segment of + needle may be found in haystack. */ +#if HAVE_MBRTOWC + if (MB_CUR_MAX > 1) + { + mbui_iterator_t iter_needle; + + mbui_init (iter_needle, needle); + if (mbui_avail (iter_needle)) + { + mbui_iterator_t iter_haystack; + + mbui_init (iter_haystack, haystack); + for (;; mbui_advance (iter_haystack)) + { + if (!mbui_avail (iter_haystack)) + /* No match. */ + return NULL; + + if (mb_equal (mbui_cur (iter_haystack), mbui_cur (iter_needle))) + /* The first character matches. */ + { + mbui_iterator_t rhaystack; + mbui_iterator_t rneedle; + + memcpy (&rhaystack, &iter_haystack, sizeof (mbui_iterator_t)); + mbui_advance (rhaystack); + + mbui_init (rneedle, needle); + if (!mbui_avail (rneedle)) + abort (); + mbui_advance (rneedle); + + for (;; mbui_advance (rhaystack), mbui_advance (rneedle)) + { + if (!mbui_avail (rneedle)) + /* Found a match. */ + return (char *) mbui_cur_ptr (iter_haystack); + if (!mbui_avail (rhaystack)) + /* No match. */ + return NULL; + if (!mb_equal (mbui_cur (rhaystack), mbui_cur (rneedle))) + /* Nothing in this round. */ + break; + } + } + } + } + else + return (char *) haystack; + } + else +#endif + { + if (*needle != '\0') + { + /* Speed up the following searches of needle by caching its first + character. */ + char b = *needle++; + + for (;; haystack++) + { + if (*haystack == '\0') + /* No match. */ + return NULL; + if (*haystack == b) + /* The first character matches. */ + { + const char *rhaystack = haystack + 1; + const char *rneedle = needle; + + for (;; rhaystack++, rneedle++) + { + if (*rneedle == '\0') + /* Found a match. */ + return (char *) haystack; + if (*rhaystack == '\0') + /* No match. */ + return NULL; + if (*rhaystack != *rneedle) + /* Nothing in this round. */ + break; + } + } + } + } + else + return (char *) haystack; + } +} diff --git a/lib/strstr.h b/lib/strstr.h new file mode 100644 index 0000000..a28b140 --- /dev/null +++ b/lib/strstr.h @@ -0,0 +1,37 @@ +/* Searching in a string. + Copyright (C) 2001-2003, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + + +/* Include string.h: on glibc systems, it contains a macro definition of + strstr() that would collide with our definition if included afterwards. */ +#include <string.h> + +#ifdef __cplusplus +extern "C" { +#endif + +/* No known system has a strstr() function that works correctly in + multibyte locales. Therefore we use our version always. */ +#undef strstr +#define strstr rpl_strstr + +/* Find the first occurrence of NEEDLE in HAYSTACK. */ +extern char *strstr (const char *haystack, const char *needle); + +#ifdef __cplusplus +} +#endif diff --git a/lib/strtol.c b/lib/strtol.c new file mode 100644 index 0000000..9bfbe3c --- /dev/null +++ b/lib/strtol.c @@ -0,0 +1,447 @@ +/* Convert string representation of a number into an integer value. + + Copyright (C) 1991, 1992, 1994, 1995, 1996, 1997, 1998, 1999, 2003, 2005 + Free Software Foundation, Inc. + + NOTE: The canonical source of this file is maintained with the GNU C + Library. Bugs can be reported to bug-glibc@gnu.org. + + This program is free software; you can redistribute it and/or modify it + under the terms of the GNU General Public License as published by the + Free Software Foundation; either version 2, or (at your option) any + later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#ifdef _LIBC +# define USE_NUMBER_GROUPING +#endif + +#include <ctype.h> +#include <errno.h> +#ifndef errno +extern int errno; +#endif +#ifndef __set_errno +# define __set_errno(Val) errno = (Val) +#endif + +#include <limits.h> +#include <stddef.h> +#include <stdlib.h> +#include <string.h> + +#ifdef USE_NUMBER_GROUPING +# include "../locale/localeinfo.h" +#endif + +/* Nonzero if we are defining `strtoul' or `strtoull', operating on + unsigned integers. */ +#ifndef UNSIGNED +# define UNSIGNED 0 +# define INT LONG int +#else +# define INT unsigned LONG int +#endif + +/* Determine the name. */ +#ifdef USE_IN_EXTENDED_LOCALE_MODEL +# if UNSIGNED +# ifdef USE_WIDE_CHAR +# ifdef QUAD +# define strtol __wcstoull_l +# else +# define strtol __wcstoul_l +# endif +# else +# ifdef QUAD +# define strtol __strtoull_l +# else +# define strtol __strtoul_l +# endif +# endif +# else +# ifdef USE_WIDE_CHAR +# ifdef QUAD +# define strtol __wcstoll_l +# else +# define strtol __wcstol_l +# endif +# else +# ifdef QUAD +# define strtol __strtoll_l +# else +# define strtol __strtol_l +# endif +# endif +# endif +#else +# if UNSIGNED +# ifdef USE_WIDE_CHAR +# ifdef QUAD +# define strtol wcstoull +# else +# define strtol wcstoul +# endif +# else +# ifdef QUAD +# define strtol strtoull +# else +# define strtol strtoul +# endif +# endif +# else +# ifdef USE_WIDE_CHAR +# ifdef QUAD +# define strtol wcstoll +# else +# define strtol wcstol +# endif +# else +# ifdef QUAD +# define strtol strtoll +# endif +# endif +# endif +#endif + +/* If QUAD is defined, we are defining `strtoll' or `strtoull', + operating on `long long int's. */ +#ifdef QUAD +# define LONG long long +# define STRTOL_LONG_MIN LONG_LONG_MIN +# define STRTOL_LONG_MAX LONG_LONG_MAX +# define STRTOL_ULONG_MAX ULONG_LONG_MAX + +/* The extra casts in the following macros work around compiler bugs, + e.g., in Cray C 5.0.3.0. */ + +/* True if negative values of the signed integer type T use two's + complement, ones' complement, or signed magnitude representation, + respectively. Much GNU code assumes two's complement, but some + people like to be portable to all possible C hosts. */ +# define TYPE_TWOS_COMPLEMENT(t) ((t) ~ (t) 0 == (t) -1) +# define TYPE_ONES_COMPLEMENT(t) ((t) ~ (t) 0 == 0) +# define TYPE_SIGNED_MAGNITUDE(t) ((t) ~ (t) 0 < (t) -1) + +/* True if the arithmetic type T is signed. */ +# define TYPE_SIGNED(t) (! ((t) 0 < (t) -1)) + +/* The maximum and minimum values for the integer type T. These + macros have undefined behavior if T is signed and has padding bits. + If this is a problem for you, please let us know how to fix it for + your host. */ +# define TYPE_MINIMUM(t) \ + ((t) (! TYPE_SIGNED (t) \ + ? (t) 0 \ + : TYPE_SIGNED_MAGNITUDE (t) \ + ? ~ (t) 0 \ + : ~ (t) 0 << (sizeof (t) * CHAR_BIT - 1))) +# define TYPE_MAXIMUM(t) \ + ((t) (! TYPE_SIGNED (t) \ + ? (t) -1 \ + : ~ (~ (t) 0 << (sizeof (t) * CHAR_BIT - 1)))) + +# ifndef ULONG_LONG_MAX +# define ULONG_LONG_MAX TYPE_MAXIMUM (unsigned long long) +# endif +# ifndef LONG_LONG_MAX +# define LONG_LONG_MAX TYPE_MAXIMUM (long long int) +# endif +# ifndef LONG_LONG_MIN +# define LONG_LONG_MIN TYPE_MINIMUM (long long int) +# endif + +# if __GNUC__ == 2 && __GNUC_MINOR__ < 7 + /* Work around gcc bug with using this constant. */ + static const unsigned long long int maxquad = ULONG_LONG_MAX; +# undef STRTOL_ULONG_MAX +# define STRTOL_ULONG_MAX maxquad +# endif +#else +# define LONG long +# define STRTOL_LONG_MIN LONG_MIN +# define STRTOL_LONG_MAX LONG_MAX +# define STRTOL_ULONG_MAX ULONG_MAX +#endif + + +/* We use this code also for the extended locale handling where the + function gets as an additional argument the locale which has to be + used. To access the values we have to redefine the _NL_CURRENT + macro. */ +#ifdef USE_IN_EXTENDED_LOCALE_MODEL +# undef _NL_CURRENT +# define _NL_CURRENT(category, item) \ + (current->values[_NL_ITEM_INDEX (item)].string) +# define LOCALE_PARAM , loc +# define LOCALE_PARAM_PROTO , __locale_t loc +#else +# define LOCALE_PARAM +# define LOCALE_PARAM_PROTO +#endif + +#if defined _LIBC || defined HAVE_WCHAR_H +# include <wchar.h> +#endif + +#ifdef USE_WIDE_CHAR +# include <wctype.h> +# define L_(Ch) L##Ch +# define UCHAR_TYPE wint_t +# define STRING_TYPE wchar_t +# ifdef USE_IN_EXTENDED_LOCALE_MODEL +# define ISSPACE(Ch) __iswspace_l ((Ch), loc) +# define ISALPHA(Ch) __iswalpha_l ((Ch), loc) +# define TOUPPER(Ch) __towupper_l ((Ch), loc) +# else +# define ISSPACE(Ch) iswspace (Ch) +# define ISALPHA(Ch) iswalpha (Ch) +# define TOUPPER(Ch) towupper (Ch) +# endif +#else +# if defined STDC_HEADERS || (!defined isascii && !defined HAVE_ISASCII) +# define IN_CTYPE_DOMAIN(c) 1 +# else +# define IN_CTYPE_DOMAIN(c) isascii(c) +# endif +# define L_(Ch) Ch +# define UCHAR_TYPE unsigned char +# define STRING_TYPE char +# ifdef USE_IN_EXTENDED_LOCALE_MODEL +# define ISSPACE(Ch) __isspace_l ((Ch), loc) +# define ISALPHA(Ch) __isalpha_l ((Ch), loc) +# define TOUPPER(Ch) __toupper_l ((Ch), loc) +# else +# define ISSPACE(Ch) (IN_CTYPE_DOMAIN (Ch) && isspace (Ch)) +# define ISALPHA(Ch) (IN_CTYPE_DOMAIN (Ch) && isalpha (Ch)) +# define TOUPPER(Ch) (IN_CTYPE_DOMAIN (Ch) ? toupper (Ch) : (Ch)) +# endif +#endif + +#define INTERNAL(X) INTERNAL1(X) +#define INTERNAL1(X) __##X##_internal +#define WEAKNAME(X) WEAKNAME1(X) + +#ifdef USE_NUMBER_GROUPING +/* This file defines a function to check for correct grouping. */ +# include "grouping.h" +#endif + + + +/* Convert NPTR to an `unsigned long int' or `long int' in base BASE. + If BASE is 0 the base is determined by the presence of a leading + zero, indicating octal or a leading "0x" or "0X", indicating hexadecimal. + If BASE is < 2 or > 36, it is reset to 10. + If ENDPTR is not NULL, a pointer to the character after the last + one converted is stored in *ENDPTR. */ + +INT +INTERNAL (strtol) (const STRING_TYPE *nptr, STRING_TYPE **endptr, + int base, int group LOCALE_PARAM_PROTO) +{ + int negative; + register unsigned LONG int cutoff; + register unsigned int cutlim; + register unsigned LONG int i; + register const STRING_TYPE *s; + register UCHAR_TYPE c; + const STRING_TYPE *save, *end; + int overflow; + +#ifdef USE_NUMBER_GROUPING +# ifdef USE_IN_EXTENDED_LOCALE_MODEL + struct locale_data *current = loc->__locales[LC_NUMERIC]; +# endif + /* The thousands character of the current locale. */ + wchar_t thousands = L'\0'; + /* The numeric grouping specification of the current locale, + in the format described in <locale.h>. */ + const char *grouping; + + if (group) + { + grouping = _NL_CURRENT (LC_NUMERIC, GROUPING); + if (*grouping <= 0 || *grouping == CHAR_MAX) + grouping = NULL; + else + { + /* Figure out the thousands separator character. */ +# if defined _LIBC || defined _HAVE_BTOWC + thousands = __btowc (*_NL_CURRENT (LC_NUMERIC, THOUSANDS_SEP)); + if (thousands == WEOF) + thousands = L'\0'; +# endif + if (thousands == L'\0') + grouping = NULL; + } + } + else + grouping = NULL; +#endif + + if (base < 0 || base == 1 || base > 36) + { + __set_errno (EINVAL); + return 0; + } + + save = s = nptr; + + /* Skip white space. */ + while (ISSPACE (*s)) + ++s; + if (*s == L_('\0')) + goto noconv; + + /* Check for a sign. */ + if (*s == L_('-')) + { + negative = 1; + ++s; + } + else if (*s == L_('+')) + { + negative = 0; + ++s; + } + else + negative = 0; + + /* Recognize number prefix and if BASE is zero, figure it out ourselves. */ + if (*s == L_('0')) + { + if ((base == 0 || base == 16) && TOUPPER (s[1]) == L_('X')) + { + s += 2; + base = 16; + } + else if (base == 0) + base = 8; + } + else if (base == 0) + base = 10; + + /* Save the pointer so we can check later if anything happened. */ + save = s; + +#ifdef USE_NUMBER_GROUPING + if (group) + { + /* Find the end of the digit string and check its grouping. */ + end = s; + for (c = *end; c != L_('\0'); c = *++end) + if ((wchar_t) c != thousands + && ((wchar_t) c < L_('0') || (wchar_t) c > L_('9')) + && (!ISALPHA (c) || (int) (TOUPPER (c) - L_('A') + 10) >= base)) + break; + if (*s == thousands) + end = s; + else + end = correctly_grouped_prefix (s, end, thousands, grouping); + } + else +#endif + end = NULL; + + cutoff = STRTOL_ULONG_MAX / (unsigned LONG int) base; + cutlim = STRTOL_ULONG_MAX % (unsigned LONG int) base; + + overflow = 0; + i = 0; + for (c = *s; c != L_('\0'); c = *++s) + { + if (s == end) + break; + if (c >= L_('0') && c <= L_('9')) + c -= L_('0'); + else if (ISALPHA (c)) + c = TOUPPER (c) - L_('A') + 10; + else + break; + if ((int) c >= base) + break; + /* Check for overflow. */ + if (i > cutoff || (i == cutoff && c > cutlim)) + overflow = 1; + else + { + i *= (unsigned LONG int) base; + i += c; + } + } + + /* Check if anything actually happened. */ + if (s == save) + goto noconv; + + /* Store in ENDPTR the address of one character + past the last character we converted. */ + if (endptr != NULL) + *endptr = (STRING_TYPE *) s; + +#if !UNSIGNED + /* Check for a value that is within the range of + `unsigned LONG int', but outside the range of `LONG int'. */ + if (overflow == 0 + && i > (negative + ? -((unsigned LONG int) (STRTOL_LONG_MIN + 1)) + 1 + : (unsigned LONG int) STRTOL_LONG_MAX)) + overflow = 1; +#endif + + if (overflow) + { + __set_errno (ERANGE); +#if UNSIGNED + return STRTOL_ULONG_MAX; +#else + return negative ? STRTOL_LONG_MIN : STRTOL_LONG_MAX; +#endif + } + + /* Return the result of the appropriate sign. */ + return negative ? -i : i; + +noconv: + /* We must handle a special case here: the base is 0 or 16 and the + first two characters are '0' and 'x', but the rest are no + hexadecimal digits. This is no error case. We return 0 and + ENDPTR points to the `x`. */ + if (endptr != NULL) + { + if (save - nptr >= 2 && TOUPPER (save[-1]) == L_('X') + && save[-2] == L_('0')) + *endptr = (STRING_TYPE *) &save[-1]; + else + /* There was no number to convert. */ + *endptr = (STRING_TYPE *) nptr; + } + + return 0L; +} + +/* External user entry point. */ + + +INT +#ifdef weak_function +weak_function +#endif +strtol (const STRING_TYPE *nptr, STRING_TYPE **endptr, + int base LOCALE_PARAM_PROTO) +{ + return INTERNAL (strtol) (nptr, endptr, base, 0 LOCALE_PARAM); +} diff --git a/lib/strtoul.c b/lib/strtoul.c new file mode 100644 index 0000000..79ceed2 --- /dev/null +++ b/lib/strtoul.c @@ -0,0 +1,20 @@ +/* Copyright (C) 1991, 1997 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#define UNSIGNED 1 + +#include "strtol.c" diff --git a/lib/sunos57-select.c b/lib/sunos57-select.c new file mode 100644 index 0000000..9ee5b6f --- /dev/null +++ b/lib/sunos57-select.c @@ -0,0 +1,225 @@ +/* Work around the bug in Solaris 7 whereby a fd that is opened on + /dev/null will cause select/poll to hang when given a NULL timeout. + + Copyright (C) 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ + +/* written by Mark D. Baushke */ + +/* + * Observed on Solaris 7: + * If /dev/null is in the readfds set, it will never be marked as + * ready by the OS. In the case of a /dev/null fd being the only fd + * in the select set and timeout == NULL, the select will hang. + * If /dev/null is in the exceptfds set, it will not be set on + * return from select(). + */ +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif /* HAVE_CONFIG_H */ + +/* The rpl_select function calls the real select. */ +#undef select + +#include <stdbool.h> +#include <stdio.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <errno.h> + +#ifdef HAVE_UNISTD_H +# include <unistd.h> +#endif /* HAVE_UNISTD_H */ + +#include "minmax.h" +#include "xtime.h" + +static struct stat devnull; +static int devnull_set = -1; +int +rpl_select (int nfds, fd_set *readfds, fd_set *writefds, fd_set *exceptfds, + struct timeval *timeout) +{ + int ret = 0; + + /* Argument checking */ + if (nfds < 1 || nfds > FD_SETSIZE) + { + errno = EINVAL; + return -1; + } + + /* Perform the initial stat on /dev/null */ + if (devnull_set == -1) + devnull_set = stat ("/dev/null", &devnull); + + if (devnull_set >= 0) + { + int fd; + int maxfd = -1; + fd_set null_rfds, null_wfds; + bool altered = false; /* Whether we have altered the caller's args. + */ + + FD_ZERO (&null_rfds); + FD_ZERO (&null_wfds); + + for (fd = 0; fd < nfds; fd++) + { + /* Check the callers bits for interesting fds */ + bool isread = (readfds && FD_ISSET (fd, readfds)); + bool isexcept = (exceptfds && FD_ISSET (fd, exceptfds)); + bool iswrite = (writefds && FD_ISSET (fd, writefds)); + + /* Check interesting fds against /dev/null */ + if (isread || iswrite || isexcept) + { + struct stat sb; + + /* Equivalent to /dev/null ? */ + if (fstat (fd, &sb) >= 0 + && sb.st_dev == devnull.st_dev + && sb.st_ino == devnull.st_ino + && sb.st_mode == devnull.st_mode + && sb.st_uid == devnull.st_uid + && sb.st_gid == devnull.st_gid + && sb.st_size == devnull.st_size + && sb.st_blocks == devnull.st_blocks + && sb.st_blksize == devnull.st_blksize) + { + /* Save the interesting bits for later use. */ + if (isread) + { + FD_SET (fd, &null_rfds); + FD_CLR (fd, readfds); + altered = true; + } + if (isexcept) + /* Pass exception bits through. + * + * At the moment, we only know that this bug + * exists in Solaris 7 and so this file should + * only be compiled on Solaris 7. Since Solaris 7 + * never returns ready for exceptions on + * /dev/null, we probably could assume this too, + * but since Solaris 9 is known to always return + * ready for exceptions on /dev/null, pass this + * through in case any other systems turn out to + * do the same. Besides, this will cause the + * timeout to be processed as it would have been + * otherwise. + */ + maxfd = MAX (maxfd, fd); + if (iswrite) + { + /* We know of no bugs involving selecting /dev/null + * writefds, but we also know that /dev/null is always + * ready for write. Therefore, since we have already + * performed all the necessary processing, avoid calling + * the system select for this case. + */ + FD_SET (fd, &null_wfds); + FD_CLR (fd, writefds); + altered = true; + } + } + else + /* A non-/dev/null fd is present. */ + maxfd = MAX (maxfd, fd); + } + } + + if (maxfd >= 0) + { + /* we need to call select, one way or another. */ + if (altered) + { + /* We already have some ready bits set, so timeout immediately + * if no bits are set. + */ + struct timeval ztime; + ztime.tv_sec = 0; + ztime.tv_usec = 0; + ret = select (maxfd + 1, readfds, writefds, exceptfds, &ztime); + if (ret == 0) + { + /* Timeout. Zero the sets since the system select might + * not have. + */ + if (readfds) + FD_ZERO (readfds); + if (exceptfds) + FD_ZERO (exceptfds); + if (writefds) + FD_ZERO (writefds); + } + } + else + /* No /dev/null fds. Call select just as the user specified. */ + ret = select (maxfd + 1, readfds, writefds, exceptfds, timeout); + } + + /* + * Borrowed from the Solaris 7 man page for select(3c): + * + * On successful completion, the objects pointed to by the + * readfds, writefds, and exceptfds arguments are modified to + * indicate which file descriptors are ready for reading, + * ready for writing, or have an error condition pending, + * respectively. For each file descriptor less than nfds, the + * corresponding bit will be set on successful completion if + * it was set on input and the associated condition is true + * for that file descriptor. + * + * On failure, the objects pointed to by the readfds, + * writefds, and exceptfds arguments are not modified. If the + * timeout interval expires without the specified condition + * being true for any of the specified file descriptors, the + * objects pointed to by the readfs, writefs, and errorfds + * arguments have all bits set to 0. + * + * On successful completion, select() returns the total number + * of bits set in the bit masks. Otherwise, -1 is returned, + * and errno is set to indicate the error. + */ + + /* Fix up the fd sets for any changes we may have made. */ + if (altered) + { + /* Tell the caller that nothing is blocking the /dev/null fds */ + for (fd = 0; fd < nfds; fd++) + { + /* If ret < 0, then we still need to restore the fd sets. */ + if (FD_ISSET (fd, &null_rfds)) + { + FD_SET (fd, readfds); + if (ret >= 0) + ret++; + } + if (FD_ISSET (fd, &null_wfds)) + { + FD_SET (fd, writefds); + if (ret >= 0) + ret++; + } + } + } + } + else + ret = select (nfds, readfds, writefds, exceptfds, timeout); + + return ret; +} diff --git a/lib/system.h b/lib/system.h new file mode 100644 index 0000000..8ea4a95 --- /dev/null +++ b/lib/system.h @@ -0,0 +1,349 @@ +/* system-dependent definitions for CVS. + Copyright (C) 1989-1992 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. */ + +/*** + *** Begin the default set of autoconf includes. + ***/ + +/* Headers assumed for C89 freestanding compilers. See HACKING for more. */ +#include <limits.h> +#include <stdarg.h> +#include <stddef.h> + +/* C89 hosted headers assumed since they were included in UNIX version 7. + * See HACKING for more. + */ +#include <assert.h> +#include <ctype.h> +#include <errno.h> +#include <signal.h> +#include <stdio.h> + +/* C89 hosted headers we _think_ GCC supplies even on freestanding systems. + * If we find any systems which do not have them, a replacement header should + * be discussed with the GNULIB folks. + * + * For more information, please see the `Portability' section of the `HACKING' + * file. + */ +#include <stdlib.h> +#include <string.h> + +/* We assume this because it has been around forever despite not being a part + * of any of the other standards we assume conformance to. So far this hasn't + * been a problem. + * + * For more information, please see the `Portability' section of the `HACKING' + * file. + */ +#include <sys/types.h> + +/* A GNULIB replacement for this C99 header is supplied when it is missing. + * See the comments in stdbool_.h for its limitations. + */ +#include <stdbool.h> + +/* Ditto for these POSIX.2 headers. */ +#include <fnmatch.h> +#include <getopt.h> /* Has GNU extensions, */ + +/* We assume <sys/stat.h> because GNULIB does. */ +#include <sys/stat.h> + +#if !STDC_HEADERS && HAVE_MEMORY_H +# include <memory.h> +#endif /* !STDC_HEADERS && HAVE_MEMORY_H */ +#if HAVE_INTTYPES_H +# include <inttypes.h> +#else /* ! HAVE_INTTYPES_H */ +# if HAVE_STDINT_H +# include <stdint.h> +# endif /* HAVE_STDINT_H */ +#endif /* HAVE_INTTYPES_H */ +#if HAVE_UNISTD_H +# include <unistd.h> +#endif /* HAVE_UNISTD_H */ +/* End the default set of autoconf includes */ + +/* Assume these headers. */ +#include <pwd.h> + +/* There is a replacement stub for gettext provided by GNULIB when gettext is + * not available. + */ +#include <gettext.h> + +#ifndef DEVNULL +# define DEVNULL "/dev/null" +#endif + +#ifdef HAVE_IO_H +#include <io.h> +#endif + +#ifdef HAVE_DIRECT_H +#include <direct.h> +#endif + +/* The NeXT (without _POSIX_SOURCE, which we don't want) has a utime.h + which doesn't define anything. It would be cleaner to have configure + check for struct utimbuf, but for now I'm checking NeXT here (so I don't + have to debug the configure check across all the machines). */ +#if defined (HAVE_UTIME_H) && !defined (NeXT) +# include <utime.h> +#else +# if defined (HAVE_SYS_UTIME_H) +# include <sys/utime.h> +# else +# ifndef ALTOS +struct utimbuf +{ + long actime; + long modtime; +}; +# endif +int utime (); +# endif +#endif + +/* errno.h variations: + * + * Not all systems set the same error code on a non-existent-file + * error. This tries to ask the question somewhat portably. + * On systems that don't have ENOTEXIST, this should behave just like + * x == ENOENT. "x" is probably errno, of course. + */ +#ifdef ENOTEXIST +# ifdef EOS2ERR +# define existence_error(x) \ + (((x) == ENOTEXIST) || ((x) == ENOENT) || ((x) == EOS2ERR)) +# else +# define existence_error(x) \ + (((x) == ENOTEXIST) || ((x) == ENOENT)) +# endif +#else +# ifdef EVMSERR +# define existence_error(x) \ +((x) == ENOENT || (x) == EINVAL || (x) == EVMSERR) +# else +# define existence_error(x) ((x) == ENOENT) +# endif +#endif + +/* check for POSIX signals */ +#if defined(HAVE_SIGACTION) && defined(HAVE_SIGPROCMASK) +# define POSIX_SIGNALS +#endif + +/* MINIX 1.6 doesn't properly support sigaction */ +#if defined(_MINIX) +# undef POSIX_SIGNALS +#endif + +/* If !POSIX, try for BSD.. Reason: 4.4BSD implements these as wrappers */ +#if !defined(POSIX_SIGNALS) +# if defined(HAVE_SIGVEC) && defined(HAVE_SIGSETMASK) && defined(HAVE_SIGBLOCK) +# define BSD_SIGNALS +# endif +#endif + +/* Under OS/2, this must be included _after_ stdio.h; that's why we do + it here. */ +#ifdef USE_OWN_TCPIP_H +# include "tcpip.h" +#endif + +#ifdef HAVE_FCNTL_H +# include <fcntl.h> +#else +# include <sys/file.h> +#endif + +#ifndef SEEK_SET +# define SEEK_SET 0 +# define SEEK_CUR 1 +# define SEEK_END 2 +#endif + +#ifndef F_OK +# define F_OK 0 +# define X_OK 1 +# define W_OK 2 +# define R_OK 4 +#endif + +#if HAVE_DIRENT_H +# include <dirent.h> +# define NAMLEN(dirent) strlen((dirent)->d_name) +#else +# define dirent direct +# define NAMLEN(dirent) (dirent)->d_namlen +# if HAVE_SYS_NDIR_H +# include <sys/ndir.h> +# endif +# if HAVE_SYS_DIR_H +# include <sys/dir.h> +# endif +# if HAVE_NDIR_H +# include <ndir.h> +# endif +#endif + +/* Convert B 512-byte blocks to kilobytes if K is nonzero, + otherwise return it unchanged. */ +#define convert_blocks(b, k) ((k) ? ((b) + 1) / 2 : (b)) + +/* Under non-UNIX operating systems (MS-DOS, WinNT, MacOS), many filesystem + calls take only one argument; permission is handled very differently on + those systems than in Unix. So we leave such systems a hook on which they + can hang their own definitions. */ + +#ifndef CVS_ACCESS +# define CVS_ACCESS access +#endif + +#ifndef CVS_CHDIR +# define CVS_CHDIR chdir +#endif + +#ifndef CVS_CREAT +# define CVS_CREAT creat +#endif + +#ifndef CVS_FOPEN +# define CVS_FOPEN fopen +#endif + +#ifndef CVS_FDOPEN +# define CVS_FDOPEN fdopen +#endif + +#ifndef CVS_MKDIR +# define CVS_MKDIR mkdir +#endif + +#ifndef CVS_OPEN +# define CVS_OPEN open +#endif + +#ifndef CVS_READDIR +# define CVS_READDIR readdir +#endif + +#ifndef CVS_CLOSEDIR +# define CVS_CLOSEDIR closedir +#endif + +#ifndef CVS_OPENDIR +# define CVS_OPENDIR opendir +#endif + +#ifndef CVS_RENAME +# define CVS_RENAME rename +#endif + +#ifndef CVS_RMDIR +# define CVS_RMDIR rmdir +#endif + +#ifndef CVS_UNLINK +# define CVS_UNLINK unlink +#endif + +/* Wildcard matcher. Should be case-insensitive if the system is. */ +#ifndef CVS_FNMATCH +# define CVS_FNMATCH fnmatch +#endif + +#ifndef HAVE_FSEEKO +off_t ftello (FILE *); +int fseeko (FILE *, off_t, int); +#endif /* HAVE_FSEEKO */ + +#ifdef FILENAMES_CASE_INSENSITIVE + +# if defined (__CYGWIN32__) || defined (WOE32) + /* Under Windows, filenames are case-insensitive, and both / and \ + are path component separators. */ +# define FOLD_FN_CHAR(c) (WNT_filename_classes[(unsigned char) (c)]) +extern unsigned char WNT_filename_classes[]; +# else /* !__CYGWIN32__ && !WOE32 */ + /* As far as I know, only Macintosh OS X & VMS make it here, but any + * platform defining FILENAMES_CASE_INSENSITIVE which isn't WOE32 or + * piggy-backing the same could, in theory. Since the OS X fold just folds + * A-Z into a-z, I'm just allowing it to be used for any case insensitive + * system which we aren't yet making other specific folds or exceptions for. + * WOE32 needs its own class since \ and C:\ style absolute paths also need + * to be accounted for. + */ +# define FOLD_FN_CHAR(c) (OSX_filename_classes[(unsigned char) (c)]) +extern unsigned char OSX_filename_classes[]; +# endif /* __CYGWIN32__ || WOE32 */ + +/* The following need to be declared for all case insensitive filesystems. + * When not FOLD_FN_CHAR is not #defined, a default definition for these + * functions is provided later in this header file. */ + +/* Like strcmp, but with the appropriate tweaks for file names. */ +extern int fncmp (const char *n1, const char *n2); + +/* Fold characters in FILENAME to their canonical forms. */ +extern void fnfold (char *FILENAME); + +#endif /* FILENAMES_CASE_INSENSITIVE */ + + + +/* Some file systems are case-insensitive. If FOLD_FN_CHAR is + #defined, it maps the character C onto its "canonical" form. In a + case-insensitive system, it would map all alphanumeric characters + to lower case. Under Windows NT, / and \ are both path component + separators, so FOLD_FN_CHAR would map them both to /. */ +#ifndef FOLD_FN_CHAR +# define FOLD_FN_CHAR(c) (c) +# define fnfold(filename) (filename) +# define fncmp strcmp +#endif + +/* Different file systems can have different naming patterns which designate + * a path as absolute. + */ +#ifndef ISABSOLUTE +# define ISABSOLUTE(s) ISSLASH(s[FILE_SYSTEM_PREFIX_LEN(s)]) +#endif + + +/* On some systems, we have to be careful about writing/reading files + in text or binary mode (so in text mode the system can handle CRLF + vs. LF, VMS text file conventions, &c). We decide to just always + be careful. That way we don't have to worry about whether text and + binary differ on this system. We just have to worry about whether + the system has O_BINARY and "rb". The latter is easy; all ANSI C + libraries have it, SunOS4 has it, and CVS has used it unguarded + some places for a while now without complaints (e.g. "rb" in + server.c (server_updated), since CVS 1.8). The former is just an + #ifdef. */ + +#define FOPEN_BINARY_READ ("rb") +#define FOPEN_BINARY_WRITE ("wb") +#define FOPEN_BINARY_READWRITE ("r+b") + +#ifdef O_BINARY +#define OPEN_BINARY (O_BINARY) +#else +#define OPEN_BINARY (0) +#endif + +#ifndef fd_select +# define fd_select select +#endif diff --git a/lib/tempname.c b/lib/tempname.c new file mode 100644 index 0000000..cae42e1 --- /dev/null +++ b/lib/tempname.c @@ -0,0 +1,319 @@ +/* tempname.c - generate the name of a temporary file. + + Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, + 2000, 2001, 2002, 2003, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include <sys/types.h> +#include <assert.h> + +#include <errno.h> +#ifndef __set_errno +# define __set_errno(Val) errno = (Val) +#endif + +#include <stdio.h> +#ifndef P_tmpdir +# define P_tmpdir "/tmp" +#endif +#ifndef TMP_MAX +# define TMP_MAX 238328 +#endif +#ifndef __GT_FILE +# define __GT_FILE 0 +# define __GT_BIGFILE 1 +# define __GT_DIR 2 +# define __GT_NOCREATE 3 +#endif + +#include <stddef.h> +#include <stdlib.h> +#include <string.h> + +#include <fcntl.h> + +#if HAVE_SYS_TIME_H || _LIBC +# include <sys/time.h> +#endif + +#if HAVE_STDINT_H || _LIBC +# include <stdint.h> +#endif +#if HAVE_INTTYPES_H +# include <inttypes.h> +#endif + +#if HAVE_UNISTD_H || _LIBC +# include <unistd.h> +#endif + +#include <sys/stat.h> + +#if _LIBC +# define struct_stat64 struct stat64 +#else +# include "stat-macros.h" +# define struct_stat64 struct stat +# define __getpid getpid +# define __gettimeofday gettimeofday +# define __mkdir mkdir +# define __open open +# define __open64 open +# define __lxstat64(version, file, buf) lstat (file, buf) +# define __xstat64(version, file, buf) stat (file, buf) +#endif + +#if ! (HAVE___SECURE_GETENV || _LIBC) +# define __secure_getenv getenv +#endif + +#ifdef _LIBC +# include <hp-timing.h> +# if HP_TIMING_AVAIL +# define RANDOM_BITS(Var) \ + if (__builtin_expect (value == UINT64_C (0), 0)) \ + { \ + /* If this is the first time this function is used initialize \ + the variable we accumulate the value in to some somewhat \ + random value. If we'd not do this programs at startup time \ + might have a reduced set of possible names, at least on slow \ + machines. */ \ + struct timeval tv; \ + __gettimeofday (&tv, NULL); \ + value = ((uint64_t) tv.tv_usec << 16) ^ tv.tv_sec; \ + } \ + HP_TIMING_NOW (Var) +# endif +#endif + +/* Use the widest available unsigned type if uint64_t is not + available. The algorithm below extracts a number less than 62**6 + (approximately 2**35.725) from uint64_t, so ancient hosts where + uintmax_t is only 32 bits lose about 3.725 bits of randomness, + which is better than not having mkstemp at all. */ +#if !defined UINT64_MAX && !defined uint64_t +# define uint64_t uintmax_t +#endif + +/* Return nonzero if DIR is an existent directory. */ +static int +direxists (const char *dir) +{ + struct_stat64 buf; + return __xstat64 (_STAT_VER, dir, &buf) == 0 && S_ISDIR (buf.st_mode); +} + +/* Path search algorithm, for tmpnam, tmpfile, etc. If DIR is + non-null and exists, uses it; otherwise uses the first of $TMPDIR, + P_tmpdir, /tmp that exists. Copies into TMPL a template suitable + for use with mk[s]temp. Will fail (-1) if DIR is non-null and + doesn't exist, none of the searched dirs exists, or there's not + enough space in TMPL. */ +int +__path_search (char *tmpl, size_t tmpl_len, const char *dir, const char *pfx, + int try_tmpdir) +{ + const char *d; + size_t dlen, plen; + + if (!pfx || !pfx[0]) + { + pfx = "file"; + plen = 4; + } + else + { + plen = strlen (pfx); + if (plen > 5) + plen = 5; + } + + if (try_tmpdir) + { + d = __secure_getenv ("TMPDIR"); + if (d != NULL && direxists (d)) + dir = d; + else if (dir != NULL && direxists (dir)) + /* nothing */ ; + else + dir = NULL; + } + if (dir == NULL) + { + if (direxists (P_tmpdir)) + dir = P_tmpdir; + else if (strcmp (P_tmpdir, "/tmp") != 0 && direxists ("/tmp")) + dir = "/tmp"; + else + { + __set_errno (ENOENT); + return -1; + } + } + + dlen = strlen (dir); + while (dlen > 1 && dir[dlen - 1] == '/') + dlen--; /* remove trailing slashes */ + + /* check we have room for "${dir}/${pfx}XXXXXX\0" */ + if (tmpl_len < dlen + 1 + plen + 6 + 1) + { + __set_errno (EINVAL); + return -1; + } + + sprintf (tmpl, "%.*s/%.*sXXXXXX", (int) dlen, dir, (int) plen, pfx); + return 0; +} + +/* These are the characters used in temporary file names. */ +static const char letters[] = +"abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789"; + +/* Generate a temporary file name based on TMPL. TMPL must match the + rules for mk[s]temp (i.e. end in "XXXXXX"). The name constructed + does not exist at the time of the call to __gen_tempname. TMPL is + overwritten with the result. + + KIND may be one of: + __GT_NOCREATE: simply verify that the name does not exist + at the time of the call. + __GT_FILE: create the file using open(O_CREAT|O_EXCL) + and return a read-write fd. The file is mode 0600. + __GT_BIGFILE: same as __GT_FILE but use open64(). + __GT_DIR: create a directory, which will be mode 0700. + + We use a clever algorithm to get hard-to-predict names. */ +int +__gen_tempname (char *tmpl, int kind) +{ + int len; + char *XXXXXX; + static uint64_t value; + uint64_t random_time_bits; + unsigned int count; + int fd = -1; + int save_errno = errno; + struct_stat64 st; + + /* A lower bound on the number of temporary files to attempt to + generate. The maximum total number of temporary file names that + can exist for a given template is 62**6. It should never be + necessary to try all these combinations. Instead if a reasonable + number of names is tried (we define reasonable as 62**3) fail to + give the system administrator the chance to remove the problems. */ + unsigned int attempts_min = 62 * 62 * 62; + + /* The number of times to attempt to generate a temporary file. To + conform to POSIX, this must be no smaller than TMP_MAX. */ + unsigned int attempts = attempts_min < TMP_MAX ? TMP_MAX : attempts_min; + + len = strlen (tmpl); + if (len < 6 || strcmp (&tmpl[len - 6], "XXXXXX")) + { + __set_errno (EINVAL); + return -1; + } + + /* This is where the Xs start. */ + XXXXXX = &tmpl[len - 6]; + + /* Get some more or less random data. */ +#ifdef RANDOM_BITS + RANDOM_BITS (random_time_bits); +#else +# if HAVE_GETTIMEOFDAY || _LIBC + { + struct timeval tv; + __gettimeofday (&tv, NULL); + random_time_bits = ((uint64_t) tv.tv_usec << 16) ^ tv.tv_sec; + } +# else + random_time_bits = time (NULL); +# endif +#endif + value += random_time_bits ^ __getpid (); + + for (count = 0; count < attempts; value += 7777, ++count) + { + uint64_t v = value; + + /* Fill in the random bits. */ + XXXXXX[0] = letters[v % 62]; + v /= 62; + XXXXXX[1] = letters[v % 62]; + v /= 62; + XXXXXX[2] = letters[v % 62]; + v /= 62; + XXXXXX[3] = letters[v % 62]; + v /= 62; + XXXXXX[4] = letters[v % 62]; + v /= 62; + XXXXXX[5] = letters[v % 62]; + + switch (kind) + { + case __GT_FILE: + fd = __open (tmpl, O_RDWR | O_CREAT | O_EXCL, S_IRUSR | S_IWUSR); + break; + + case __GT_BIGFILE: + fd = __open64 (tmpl, O_RDWR | O_CREAT | O_EXCL, S_IRUSR | S_IWUSR); + break; + + case __GT_DIR: + fd = __mkdir (tmpl, S_IRUSR | S_IWUSR | S_IXUSR); + break; + + case __GT_NOCREATE: + /* This case is backward from the other three. __gen_tempname + succeeds if __xstat fails because the name does not exist. + Note the continue to bypass the common logic at the bottom + of the loop. */ + if (__lxstat64 (_STAT_VER, tmpl, &st) < 0) + { + if (errno == ENOENT) + { + __set_errno (save_errno); + return 0; + } + else + /* Give up now. */ + return -1; + } + continue; + + default: + assert (! "invalid KIND in __gen_tempname"); + } + + if (fd >= 0) + { + __set_errno (save_errno); + return fd; + } + else if (errno != EEXIST) + return -1; + } + + /* We got out of the loop because we ran out of combinations to try. */ + __set_errno (EEXIST); + return -1; +} diff --git a/lib/test-getdate.sh b/lib/test-getdate.sh new file mode 100755 index 0000000..128def2 --- /dev/null +++ b/lib/test-getdate.sh @@ -0,0 +1,340 @@ +#! /bin/sh + +# Test that a getdate executable meets its specification. +# +# Copyright (C) 2004 Free Software Foundation, Inc. +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2, or (at your option) +# any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program; if not, write to the Free Software Foundation, +# Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ + + + +### +### Globals +### +LOGFILE=`pwd`/getdate.log +if test -f "$LOGFILE"; then + mv $LOGFILE $LOGFILE~ +fi + + + +### +### Functions +### +verify () +{ + echo >>getdate-got + if cmp getdate-expected getdate-got >getdate.cmp; then + echo "PASS: $1" >>$LOGFILE + else + cat getdate.cmp >>$LOGFILE + echo "** expected: " >>$LOGFILE + cat getdate-expected >>$LOGFILE + echo "** got: " >>$LOGFILE + cat getdate-got >>$LOGFILE + echo "FAIL: $1" | tee -a $LOGFILE >&2 + echo "Failed! See $LOGFILE for more!" >&2 + exit 1 + fi +} + + + +skip () +{ + echo "SKIP: $1"${2+" ($2)"} >>$LOGFILE +} + + + +# Prep for future calls to valid_timezone(). +# +# This should set $UTZ to three spaces, `GMT', `Unrecognized/Unrecognized', or +# possibly the empty string, depending on what system we are running on. With +# any luck, this will catch any other existing variations as well. The way it +# is used later does have the disadvantage of rejecting at least the +# `Europe/London' timezone for half the year when $UTZ gets set to `GMT', like +# happens on NetBSD, but, since I haven't come up with any better ideas and +# since rejecting a timezone just causes a few tests to be skipped, this will +# have to do for now. +# +# UTZ stands for Unrecognized Time Zone. +UTZ=`TZ=Unrecognized/Unrecognized date +%Z` + +# The following function will return true if $1 is a valid timezone. It will +# return false and set $skipreason, otherwise. +# +# Clobbers $NTZ & $skipreason. +# +# SUS2 says `date +%Z' will return `no characters' if `no timezone is +# determinable'. It is, unfortunately, not very specific about what +# `determinable' means. On GNU/Linux, `date +%Z' returns $TZ when $TZ is not +# recognized. NetBSD 1.6.1 "determines" that an unrecognizable value in $TZ +# really means `GMT'. On Cray, the standard is ignored and `date +%Z' returns +# three spaces when $TZ is not recognized. We test for all three cases, plus +# the empty string for good measure, though I know of no set of conditions +# which will actually cause `date +%Z' to return the empty string SUS2 +# specifies. +# +# Due to the current nature of this test, this will not work for the +# three-letter zone codes on some systems. e.g.: +# +# test `TZ=EST date +%Z` = "EST" +# +# should, quite correctly, evaluate to true on most systems, but: +# +# TZ=Asia/Calcutta date +%Z +# +# would return `IST' on GNU/Linux, and hopefully any system which understands +# the `Asia/Calcutta' timezone, and ` ' on Cray. Similarly: +# +# TZ=Doesnt_Exist/Doesnt_Exist date +%Z +# +# returns `Doesnt_Exist/Doesnt_Exist' on GNU/Linux and ` ' on Cray. +# +# Unfortunately, the %z date format string (-HHMM format time zone) supported +# by the GNU `date' command is not part of any standard I know of and, +# therefore, is probably not portable. +# +valid_timezone () +{ + NTZ=`TZ=$1 date +%Z` + if test "$NTZ" = "$UTZ" || test "$NTZ" = "$1"; then + skipreason="$1 is not a recognized timezone on this system" + return `false` + else + return `:` + fi +} + + + +### +### Tests +### + +# Why are these dates tested? +# +# February 29, 2003 +# Is not a leap year - should be invalid. +# +# 2004-12-40 +# Make sure get_date does not "roll" date forward to January 9th. Some +# versions have been known to do this. +# +# Dec-5-1972 +# This is my birthday. :) +# +# 3/29/1974 +# 1996/05/12 13:57:45 +# Because. +# +# 12-05-12 +# This will be my 40th birthday. Ouch. :) +# +# 05/12/96 +# Because. +# +# third tuesday in March, 2078 +# Wanted this to work. +# +# 1969-12-32 2:00:00 UTC +# 1970-01-01 2:00:00 UTC +# 1969-12-32 2:00:00 +0400 +# 1970-01-01 2:00:00 +0400 +# 1969-12-32 2:00:00 -0400 +# 1970-01-01 2:00:00 -0400 +# Playing near the UNIX Epoch boundry condition to make sure date rolling +# is also disabled there. +# +# 1996-12-12 1 month +# Test a relative date. + + + +# The following tests are currently being skipped for being unportable: +# +# Tue Jan 19 03:14:07 2038 +0000 +# For machines with 31-bit time_t, any date past this date will be an +# invalid date. So, any test date with a value greater than this +# time is not portable. +# +# Feb. 29, 2096 4 years +# 4 years from this date is _not_ a leap year, so Feb. 29th does not exist. +# +# Feb. 29, 2096 8 years +# 8 years from this date is a leap year, so Feb. 29th does exist, +# but on many hosts with 32-bit time_t types time, this test will +# fail. So, this is not a portable test. +# + +TZ=UTC0; export TZ + +cat >getdate-expected <<EOF +Enter date, or blank line to exit. + > Bad format - couldn't convert. + > Bad format - couldn't convert. + > 1972-12-05 00:00:00.000000000 + > 1974-03-29 00:00:00.000000000 + > 1996-05-12 13:57:45.000000000 + > 2012-05-12 00:00:00.000000000 + > 1996-05-12 00:00:00.000000000 + > Bad format - couldn't convert. + > Bad format - couldn't convert. + > 1970-01-01 02:00:00.000000000 + > Bad format - couldn't convert. + > 1969-12-31 22:00:00.000000000 + > Bad format - couldn't convert. + > 1970-01-01 06:00:00.000000000 + > 1997-01-12 00:00:00.000000000 + > +EOF + +./getdate >getdate-got <<EOF +February 29, 2003 +2004-12-40 +Dec-5-1972 +3/29/1974 +1996/05/12 13:57:45 +12-05-12 +05/12/96 +third tuesday in March, 2078 +1969-12-32 2:00:00 UTC +1970-01-01 2:00:00 UTC +1969-12-32 2:00:00 +0400 +1970-01-01 2:00:00 +0400 +1969-12-32 2:00:00 -0400 +1970-01-01 2:00:00 -0400 +1996-12-12 1 month +EOF + +verify getdate-1 + + + +# Why are these dates tested? +# +# Ian Abbot reported these odd boundry cases. After daylight savings time went +# into effect, non-daylight time zones would cause +# "Bad format - couldn't convert." errors, even when the non-daylight zone +# happened to be a universal one, like GMT. + +TZ=Europe/London; export TZ +if valid_timezone $TZ; then + cat >getdate-expected <<EOF +Enter date, or blank line to exit. + > 2005-03-01 00:00:00.000000000 + > 2005-03-27 00:00:00.000000000 + > 2005-03-28 01:00:00.000000000 + > 2005-03-28 01:00:00.000000000 + > 2005-03-29 01:00:00.000000000 + > 2005-03-29 01:00:00.000000000 + > 2005-03-30 01:00:00.000000000 + > 2005-03-30 01:00:00.000000000 + > 2005-03-31 01:00:00.000000000 + > 2005-03-31 01:00:00.000000000 + > 2005-04-01 01:00:00.000000000 + > 2005-04-01 01:00:00.000000000 + > 2005-04-10 01:00:00.000000000 + > 2005-04-10 01:00:00.000000000 + > 2005-04-01 00:00:00.000000000 + > +EOF + + ./getdate >getdate-got <<EOF +2005-3-1 GMT +2005-3-27 GMT +2005-3-28 GMT +2005-3-28 UTC0 +2005-3-29 GMT +2005-3-29 UTC0 +2005-3-30 GMT +2005-3-30 UTC0 +2005-3-31 GMT +2005-3-31 UTC0 +2005-4-1 GMT +2005-4-1 UTC0 +2005-4-10 GMT +2005-4-10 UTC0 +2005-4-1 BST +EOF + + verify getdate-2 +else + skip getdate-2 "$skipreason" +fi + + + +# Many of the following cases were also submitted by Ian Abbott, but the same +# errors are not exhibited. The original problem had a similar root, but +# managed to produce errors with GMT, which is considered a "Universal Zone". +# This was fixed. +# +# The deeper problem has to do with "local zone" processing in getdate.y +# that causes local daylight zones to be excluded when local standard time is +# in effect and vice versa. This used to cause trouble with GMT in Britian +# when British Summer Time was in effect, but this was overridden for the +# "Universal Timezones" (GMT, UTC, & UT), that might double as a local zone in +# some locales. We still see in these tests the local daylight/standard zone +# exclusion in EST/EDT. According to Paul Eggert in a message to +# bug-gnulib@gnu.org on 2005-04-12, this is considered a bug but may not be +# fixed soon due to its complexity. + +TZ=America/New_York; export TZ +if valid_timezone $TZ; then + cat >getdate-expected <<EOF +Enter date, or blank line to exit. + > 2005-03-01 00:00:00.000000000 + > 2005-02-28 18:00:00.000000000 + > 2005-04-01 00:00:00.000000000 + > Bad format - couldn't convert. + > 2005-04-30 19:00:00.000000000 + > 2005-04-30 20:00:00.000000000 + > 2005-05-01 00:00:00.000000000 + > 2005-04-30 20:00:00.000000000 + > Bad format - couldn't convert. + > 2005-05-31 19:00:00.000000000 + > 2005-05-31 20:00:00.000000000 + > 2005-06-01 00:00:00.000000000 + > 2005-05-31 20:00:00.000000000 + > +EOF + + ./getdate >getdate-got <<EOF +2005-3-1 EST +2005-3-1 BST +2005-4-1 EST +2005-5-1 EST +2005-5-1 BST +2005-5-1 GMT +2005-5-1 EDT +2005-5-1 UTC0 +2005-6-1 EST +2005-6-1 BST +2005-6-1 GMT +2005-6-1 EDT +2005-6-1 UTC0 +EOF + + verify getdate-3 +else + skip getdate-3 "$skipreason" +fi + + + +rm getdate-expected getdate-got getdate.cmp +exit 0 diff --git a/lib/time_r.c b/lib/time_r.c new file mode 100644 index 0000000..245ac92 --- /dev/null +++ b/lib/time_r.c @@ -0,0 +1,69 @@ +/* Reentrant time functions like localtime_r. + + Copyright (C) 2003 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Paul Eggert. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "time_r.h" + +#include <string.h> + +static char * +copy_string_result (char *dest, char const *src) +{ + if (! src) + return 0; + return strcpy (dest, src); +} + +static struct tm * +copy_tm_result (struct tm *dest, struct tm const *src) +{ + if (! src) + return 0; + *dest = *src; + return dest; +} + + +char * +asctime_r (struct tm const * restrict tm, char * restrict buf) +{ + return copy_string_result (buf, asctime (tm)); +} + +char * +ctime_r (time_t const *t, char *buf) +{ + return copy_string_result (buf, ctime (t)); +} + +struct tm * +gmtime_r (time_t const * restrict t, struct tm * restrict tp) +{ + return copy_tm_result (tp, gmtime (t)); +} + +struct tm * +localtime_r (time_t const * restrict t, struct tm * restrict tp) +{ + return copy_tm_result (tp, localtime (t)); +} diff --git a/lib/time_r.h b/lib/time_r.h new file mode 100644 index 0000000..1f2e94a --- /dev/null +++ b/lib/time_r.h @@ -0,0 +1,57 @@ +/* Reentrant time functions like localtime_r. + + Copyright (C) 2003, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Paul Eggert. */ + +#ifndef _TIME_R_H +#define _TIME_R_H + +/* Include <time.h> first, since it may declare these functions with + signatures that disagree with POSIX, and we don't want to rename + those declarations. */ +#include <time.h> + +#if !HAVE_TIME_R_POSIX +# undef asctime_r +# undef ctime_r +# undef gmtime_r +# undef localtime_r + +# define asctime_r rpl_asctime_r +# define ctime_r rpl_ctime_r +# define gmtime_r rpl_gmtime_r +# define localtime_r rpl_localtime_r + +/* See the POSIX:2001 specification + <http://www.opengroup.org/susv3xsh/asctime.html>. */ +char *asctime_r (struct tm const * restrict, char * restrict); + +/* See the POSIX:2001 specification + <http://www.opengroup.org/susv3xsh/ctime.html>. */ +char *ctime_r (time_t const *, char *); + +/* See the POSIX:2001 specification + <http://www.opengroup.org/susv3xsh/gmtime.html>. */ +struct tm *gmtime_r (time_t const * restrict, struct tm * restrict); + +/* See the POSIX:2001 specification + <http://www.opengroup.org/susv3xsh/localtime.html>. */ +struct tm *localtime_r (time_t const * restrict, struct tm * restrict); +#endif + +#endif diff --git a/lib/timespec.h b/lib/timespec.h new file mode 100644 index 0000000..8188c20 --- /dev/null +++ b/lib/timespec.h @@ -0,0 +1,62 @@ +/* timespec -- System time interface + + Copyright (C) 2000, 2002, 2004, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#if ! defined TIMESPEC_H +# define TIMESPEC_H + +# include <sys/types.h> +# if TIME_WITH_SYS_TIME +# include <sys/time.h> +# include <time.h> +# else +# if HAVE_SYS_TIME_H +# include <sys/time.h> +# else +# include <time.h> +# endif +# endif + +# if ! HAVE_STRUCT_TIMESPEC +/* Some systems don't define this struct, e.g., AIX 4.1, Ultrix 4.3. */ +struct timespec +{ + time_t tv_sec; + long tv_nsec; +}; +# endif + +/* Return negative, zero, positive if A < B, A == B, A > B, respectively. + Assume the nanosecond components are in range, or close to it. */ +static inline int +timespec_cmp (struct timespec a, struct timespec b) +{ + return (a.tv_sec < b.tv_sec ? -1 + : a.tv_sec > b.tv_sec ? 1 + : a.tv_nsec - b.tv_nsec); +} + +# if ! HAVE_DECL_NANOSLEEP +/* Don't specify a prototype here. Some systems (e.g., OSF) declare + nanosleep with a conflicting one (const-less first parameter). */ +int nanosleep (); +# endif + +void gettime (struct timespec *); +int settime (struct timespec const *); + +#endif diff --git a/lib/unistd--.h b/lib/unistd--.h new file mode 100644 index 0000000..1fe6ce8 --- /dev/null +++ b/lib/unistd--.h @@ -0,0 +1,28 @@ +/* Like unistd.h, but redefine some names to avoid glitches. + + Copyright (C) 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Paul Eggert. */ + +#include <unistd.h> +#include "unistd-safer.h" + +#undef dup +#define dup dup_safer + +#undef pipe +#define pipe pipe_safer diff --git a/lib/unistd-safer.h b/lib/unistd-safer.h new file mode 100644 index 0000000..f95999d --- /dev/null +++ b/lib/unistd-safer.h @@ -0,0 +1,23 @@ +/* Invoke unistd-like functions, but avoid some glitches. + + Copyright (C) 2001, 2003, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Paul Eggert. */ + +int dup_safer (int); +int fd_safer (int); +int pipe_safer (int[2]); diff --git a/lib/unlocked-io.h b/lib/unlocked-io.h new file mode 100644 index 0000000..d009303 --- /dev/null +++ b/lib/unlocked-io.h @@ -0,0 +1,137 @@ +/* Prefer faster, non-thread-safe stdio functions if available. + + Copyright (C) 2001, 2002, 2003, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Jim Meyering. */ + +#ifndef UNLOCKED_IO_H +# define UNLOCKED_IO_H 1 + +/* These are wrappers for functions/macros from the GNU C library, and + from other C libraries supporting POSIX's optional thread-safe functions. + + The standard I/O functions are thread-safe. These *_unlocked ones are + more efficient but not thread-safe. That they're not thread-safe is + fine since all of the applications in this package are single threaded. + + Also, some code that is shared with the GNU C library may invoke + the *_unlocked functions directly. On hosts that lack those + functions, invoke the non-thread-safe versions instead. */ + +# include <stdio.h> + +# if HAVE_DECL_CLEARERR_UNLOCKED +# undef clearerr +# define clearerr(x) clearerr_unlocked (x) +# else +# define clearerr_unlocked(x) clearerr (x) +# endif + +# if HAVE_DECL_FEOF_UNLOCKED +# undef feof +# define feof(x) feof_unlocked (x) +# else +# define feof_unlocked(x) feof (x) +# endif + +# if HAVE_DECL_FERROR_UNLOCKED +# undef ferror +# define ferror(x) ferror_unlocked (x) +# else +# define ferror_unlocked(x) ferror (x) +# endif + +# if HAVE_DECL_FFLUSH_UNLOCKED +# undef fflush +# define fflush(x) fflush_unlocked (x) +# else +# define fflush_unlocked(x) fflush (x) +# endif + +# if HAVE_DECL_FGETS_UNLOCKED +# undef fgets +# define fgets(x,y,z) fgets_unlocked (x,y,z) +# else +# define fgets_unlocked(x,y,z) fgets (x,y,z) +# endif + +# if HAVE_DECL_FPUTC_UNLOCKED +# undef fputc +# define fputc(x,y) fputc_unlocked (x,y) +# else +# define fputc_unlocked(x,y) fputc (x,y) +# endif + +# if HAVE_DECL_FPUTS_UNLOCKED +# undef fputs +# define fputs(x,y) fputs_unlocked (x,y) +# else +# define fputs_unlocked(x,y) fputs (x,y) +# endif + +# if HAVE_DECL_FREAD_UNLOCKED +# undef fread +# define fread(w,x,y,z) fread_unlocked (w,x,y,z) +# else +# define fread_unlocked(w,x,y,z) fread (w,x,y,z) +# endif + +# if HAVE_DECL_FWRITE_UNLOCKED +# undef fwrite +# define fwrite(w,x,y,z) fwrite_unlocked (w,x,y,z) +# else +# define fwrite_unlocked(w,x,y,z) fwrite (w,x,y,z) +# endif + +# if HAVE_DECL_GETC_UNLOCKED +# undef getc +# define getc(x) getc_unlocked (x) +# else +# define getc_unlocked(x) getc (x) +# endif + +# if HAVE_DECL_GETCHAR_UNLOCKED +# undef getchar +# define getchar() getchar_unlocked () +# else +# define getchar_unlocked() getchar () +# endif + +# if HAVE_DECL_PUTC_UNLOCKED +# undef putc +# define putc(x,y) putc_unlocked (x,y) +# else +# define putc_unlocked(x,y) putc (x,y) +# endif + +# if HAVE_DECL_PUTCHAR_UNLOCKED +# undef putchar +# define putchar(x) putchar_unlocked (x) +# else +# define putchar_unlocked(x) putchar (x) +# endif + +# undef flockfile +# define flockfile(x) ((void) 0) + +# undef ftrylockfile +# define ftrylockfile(x) 0 + +# undef funlockfile +# define funlockfile(x) ((void) 0) + +#endif /* UNLOCKED_IO_H */ diff --git a/lib/unsetenv.c b/lib/unsetenv.c new file mode 100644 index 0000000..4b53fc9 --- /dev/null +++ b/lib/unsetenv.c @@ -0,0 +1,99 @@ +/* Copyright (C) 1992,1995-1999,2000-2002 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#if HAVE_CONFIG_H +# include <config.h> +#endif + +#include <errno.h> +#if !_LIBC +# if !defined errno && !defined HAVE_ERRNO_DECL +extern int errno; +# endif +# define __set_errno(ev) ((errno) = (ev)) +#endif + +#include <stdlib.h> +#include <string.h> +#if _LIBC || HAVE_UNISTD_H +# include <unistd.h> +#endif + +#if !_LIBC +# define __environ environ +# ifndef HAVE_ENVIRON_DECL +extern char **environ; +# endif +#endif + +#if _LIBC +/* This lock protects against simultaneous modifications of `environ'. */ +# include <bits/libc-lock.h> +__libc_lock_define_initialized (static, envlock) +# define LOCK __libc_lock_lock (envlock) +# define UNLOCK __libc_lock_unlock (envlock) +#else +# define LOCK +# define UNLOCK +#endif + +/* In the GNU C library we must keep the namespace clean. */ +#ifdef _LIBC +# define unsetenv __unsetenv +#endif + + +int +unsetenv (const char *name) +{ + size_t len; + char **ep; + + if (name == NULL || *name == '\0' || strchr (name, '=') != NULL) + { + __set_errno (EINVAL); + return -1; + } + + len = strlen (name); + + LOCK; + + ep = __environ; + while (*ep != NULL) + if (!strncmp (*ep, name, len) && (*ep)[len] == '=') + { + /* Found it. Remove this pointer by moving later ones back. */ + char **dp = ep; + + do + dp[0] = dp[1]; + while (*dp++); + /* Continue the loop in case NAME appears again. */ + } + else + ++ep; + + UNLOCK; + + return 0; +} + +#ifdef _LIBC +# undef unsetenv +weak_alias (__unsetenv, unsetenv) +#endif diff --git a/lib/vasnprintf.c b/lib/vasnprintf.c new file mode 100644 index 0000000..324d62e --- /dev/null +++ b/lib/vasnprintf.c @@ -0,0 +1,901 @@ +/* vsprintf with automatic memory allocation. + Copyright (C) 1999, 2002-2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Tell glibc's <stdio.h> to provide a prototype for snprintf(). + This must come before <config.h> because <config.h> may include + <features.h>, and once <features.h> has been included, it's too late. */ +#ifndef _GNU_SOURCE +# define _GNU_SOURCE 1 +#endif + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif +#ifndef IN_LIBINTL +# include <alloca.h> +#endif + +/* Specification. */ +#if WIDE_CHAR_VERSION +# include "vasnwprintf.h" +#else +# include "vasnprintf.h" +#endif + +#include <stdio.h> /* snprintf(), sprintf() */ +#include <stdlib.h> /* abort(), malloc(), realloc(), free() */ +#include <string.h> /* memcpy(), strlen() */ +#include <errno.h> /* errno */ +#include <limits.h> /* CHAR_BIT, INT_MAX */ +#include <float.h> /* DBL_MAX_EXP, LDBL_MAX_EXP */ +#if WIDE_CHAR_VERSION +# include "wprintf-parse.h" +#else +# include "printf-parse.h" +#endif + +/* Checked size_t computations. */ +#include "xsize.h" + +/* Some systems, like OSF/1 4.0 and Woe32, don't have EOVERFLOW. */ +#ifndef EOVERFLOW +# define EOVERFLOW E2BIG +#endif + +#ifdef HAVE_WCHAR_T +# ifdef HAVE_WCSLEN +# define local_wcslen wcslen +# else + /* Solaris 2.5.1 has wcslen() in a separate library libw.so. To avoid + a dependency towards this library, here is a local substitute. + Define this substitute only once, even if this file is included + twice in the same compilation unit. */ +# ifndef local_wcslen_defined +# define local_wcslen_defined 1 +static size_t +local_wcslen (const wchar_t *s) +{ + const wchar_t *ptr; + + for (ptr = s; *ptr != (wchar_t) 0; ptr++) + ; + return ptr - s; +} +# endif +# endif +#endif + +#if WIDE_CHAR_VERSION +# define VASNPRINTF vasnwprintf +# define CHAR_T wchar_t +# define DIRECTIVE wchar_t_directive +# define DIRECTIVES wchar_t_directives +# define PRINTF_PARSE wprintf_parse +# define USE_SNPRINTF 1 +# if HAVE_DECL__SNWPRINTF + /* On Windows, the function swprintf() has a different signature than + on Unix; we use the _snwprintf() function instead. */ +# define SNPRINTF _snwprintf +# else + /* Unix. */ +# define SNPRINTF swprintf +# endif +#else +# define VASNPRINTF vasnprintf +# define CHAR_T char +# define DIRECTIVE char_directive +# define DIRECTIVES char_directives +# define PRINTF_PARSE printf_parse +# define USE_SNPRINTF (HAVE_DECL__SNPRINTF || HAVE_SNPRINTF) +# if HAVE_DECL__SNPRINTF + /* Windows. */ +# define SNPRINTF _snprintf +# else + /* Unix. */ +# define SNPRINTF snprintf +# endif +#endif + +CHAR_T * +VASNPRINTF (CHAR_T *resultbuf, size_t *lengthp, const CHAR_T *format, va_list args) +{ + DIRECTIVES d; + arguments a; + + if (PRINTF_PARSE (format, &d, &a) < 0) + { + errno = EINVAL; + return NULL; + } + +#define CLEANUP() \ + free (d.dir); \ + if (a.arg) \ + free (a.arg); + + if (printf_fetchargs (args, &a) < 0) + { + CLEANUP (); + errno = EINVAL; + return NULL; + } + + { + size_t buf_neededlength; + CHAR_T *buf; + CHAR_T *buf_malloced; + const CHAR_T *cp; + size_t i; + DIRECTIVE *dp; + /* Output string accumulator. */ + CHAR_T *result; + size_t allocated; + size_t length; + + /* Allocate a small buffer that will hold a directive passed to + sprintf or snprintf. */ + buf_neededlength = + xsum4 (7, d.max_width_length, d.max_precision_length, 6); +#if HAVE_ALLOCA + if (buf_neededlength < 4000 / sizeof (CHAR_T)) + { + buf = (CHAR_T *) alloca (buf_neededlength * sizeof (CHAR_T)); + buf_malloced = NULL; + } + else +#endif + { + size_t buf_memsize = xtimes (buf_neededlength, sizeof (CHAR_T)); + if (size_overflow_p (buf_memsize)) + goto out_of_memory_1; + buf = (CHAR_T *) malloc (buf_memsize); + if (buf == NULL) + goto out_of_memory_1; + buf_malloced = buf; + } + + if (resultbuf != NULL) + { + result = resultbuf; + allocated = *lengthp; + } + else + { + result = NULL; + allocated = 0; + } + length = 0; + /* Invariants: + result is either == resultbuf or == NULL or malloc-allocated. + If length > 0, then result != NULL. */ + + /* Ensures that allocated >= needed. Aborts through a jump to + out_of_memory if needed is SIZE_MAX or otherwise too big. */ +#define ENSURE_ALLOCATION(needed) \ + if ((needed) > allocated) \ + { \ + size_t memory_size; \ + CHAR_T *memory; \ + \ + allocated = (allocated > 0 ? xtimes (allocated, 2) : 12); \ + if ((needed) > allocated) \ + allocated = (needed); \ + memory_size = xtimes (allocated, sizeof (CHAR_T)); \ + if (size_overflow_p (memory_size)) \ + goto out_of_memory; \ + if (result == resultbuf || result == NULL) \ + memory = (CHAR_T *) malloc (memory_size); \ + else \ + memory = (CHAR_T *) realloc (result, memory_size); \ + if (memory == NULL) \ + goto out_of_memory; \ + if (result == resultbuf && length > 0) \ + memcpy (memory, result, length * sizeof (CHAR_T)); \ + result = memory; \ + } + + for (cp = format, i = 0, dp = &d.dir[0]; ; cp = dp->dir_end, i++, dp++) + { + if (cp != dp->dir_start) + { + size_t n = dp->dir_start - cp; + size_t augmented_length = xsum (length, n); + + ENSURE_ALLOCATION (augmented_length); + memcpy (result + length, cp, n * sizeof (CHAR_T)); + length = augmented_length; + } + if (i == d.count) + break; + + /* Execute a single directive. */ + if (dp->conversion == '%') + { + size_t augmented_length; + + if (!(dp->arg_index == ARG_NONE)) + abort (); + augmented_length = xsum (length, 1); + ENSURE_ALLOCATION (augmented_length); + result[length] = '%'; + length = augmented_length; + } + else + { + if (!(dp->arg_index != ARG_NONE)) + abort (); + + if (dp->conversion == 'n') + { + switch (a.arg[dp->arg_index].type) + { + case TYPE_COUNT_SCHAR_POINTER: + *a.arg[dp->arg_index].a.a_count_schar_pointer = length; + break; + case TYPE_COUNT_SHORT_POINTER: + *a.arg[dp->arg_index].a.a_count_short_pointer = length; + break; + case TYPE_COUNT_INT_POINTER: + *a.arg[dp->arg_index].a.a_count_int_pointer = length; + break; + case TYPE_COUNT_LONGINT_POINTER: + *a.arg[dp->arg_index].a.a_count_longint_pointer = length; + break; +#ifdef HAVE_LONG_LONG + case TYPE_COUNT_LONGLONGINT_POINTER: + *a.arg[dp->arg_index].a.a_count_longlongint_pointer = length; + break; +#endif + default: + abort (); + } + } + else + { + arg_type type = a.arg[dp->arg_index].type; + CHAR_T *p; + unsigned int prefix_count; + int prefixes[2]; +#if !USE_SNPRINTF + size_t tmp_length; + CHAR_T tmpbuf[700]; + CHAR_T *tmp; + + /* Allocate a temporary buffer of sufficient size for calling + sprintf. */ + { + size_t width; + size_t precision; + + width = 0; + if (dp->width_start != dp->width_end) + { + if (dp->width_arg_index != ARG_NONE) + { + int arg; + + if (!(a.arg[dp->width_arg_index].type == TYPE_INT)) + abort (); + arg = a.arg[dp->width_arg_index].a.a_int; + width = (arg < 0 ? (unsigned int) (-arg) : arg); + } + else + { + const CHAR_T *digitp = dp->width_start; + + do + width = xsum (xtimes (width, 10), *digitp++ - '0'); + while (digitp != dp->width_end); + } + } + + precision = 6; + if (dp->precision_start != dp->precision_end) + { + if (dp->precision_arg_index != ARG_NONE) + { + int arg; + + if (!(a.arg[dp->precision_arg_index].type == TYPE_INT)) + abort (); + arg = a.arg[dp->precision_arg_index].a.a_int; + precision = (arg < 0 ? 0 : arg); + } + else + { + const CHAR_T *digitp = dp->precision_start + 1; + + precision = 0; + while (digitp != dp->precision_end) + precision = xsum (xtimes (precision, 10), *digitp++ - '0'); + } + } + + switch (dp->conversion) + { + + case 'd': case 'i': case 'u': +# ifdef HAVE_LONG_LONG + if (type == TYPE_LONGLONGINT || type == TYPE_ULONGLONGINT) + tmp_length = + (unsigned int) (sizeof (unsigned long long) * CHAR_BIT + * 0.30103 /* binary -> decimal */ + * 2 /* estimate for FLAG_GROUP */ + ) + + 1 /* turn floor into ceil */ + + 1; /* account for leading sign */ + else +# endif + if (type == TYPE_LONGINT || type == TYPE_ULONGINT) + tmp_length = + (unsigned int) (sizeof (unsigned long) * CHAR_BIT + * 0.30103 /* binary -> decimal */ + * 2 /* estimate for FLAG_GROUP */ + ) + + 1 /* turn floor into ceil */ + + 1; /* account for leading sign */ + else + tmp_length = + (unsigned int) (sizeof (unsigned int) * CHAR_BIT + * 0.30103 /* binary -> decimal */ + * 2 /* estimate for FLAG_GROUP */ + ) + + 1 /* turn floor into ceil */ + + 1; /* account for leading sign */ + break; + + case 'o': +# ifdef HAVE_LONG_LONG + if (type == TYPE_LONGLONGINT || type == TYPE_ULONGLONGINT) + tmp_length = + (unsigned int) (sizeof (unsigned long long) * CHAR_BIT + * 0.333334 /* binary -> octal */ + ) + + 1 /* turn floor into ceil */ + + 1; /* account for leading sign */ + else +# endif + if (type == TYPE_LONGINT || type == TYPE_ULONGINT) + tmp_length = + (unsigned int) (sizeof (unsigned long) * CHAR_BIT + * 0.333334 /* binary -> octal */ + ) + + 1 /* turn floor into ceil */ + + 1; /* account for leading sign */ + else + tmp_length = + (unsigned int) (sizeof (unsigned int) * CHAR_BIT + * 0.333334 /* binary -> octal */ + ) + + 1 /* turn floor into ceil */ + + 1; /* account for leading sign */ + break; + + case 'x': case 'X': +# ifdef HAVE_LONG_LONG + if (type == TYPE_LONGLONGINT || type == TYPE_ULONGLONGINT) + tmp_length = + (unsigned int) (sizeof (unsigned long long) * CHAR_BIT + * 0.25 /* binary -> hexadecimal */ + ) + + 1 /* turn floor into ceil */ + + 2; /* account for leading sign or alternate form */ + else +# endif + if (type == TYPE_LONGINT || type == TYPE_ULONGINT) + tmp_length = + (unsigned int) (sizeof (unsigned long) * CHAR_BIT + * 0.25 /* binary -> hexadecimal */ + ) + + 1 /* turn floor into ceil */ + + 2; /* account for leading sign or alternate form */ + else + tmp_length = + (unsigned int) (sizeof (unsigned int) * CHAR_BIT + * 0.25 /* binary -> hexadecimal */ + ) + + 1 /* turn floor into ceil */ + + 2; /* account for leading sign or alternate form */ + break; + + case 'f': case 'F': +# ifdef HAVE_LONG_DOUBLE + if (type == TYPE_LONGDOUBLE) + tmp_length = + (unsigned int) (LDBL_MAX_EXP + * 0.30103 /* binary -> decimal */ + * 2 /* estimate for FLAG_GROUP */ + ) + + 1 /* turn floor into ceil */ + + 10; /* sign, decimal point etc. */ + else +# endif + tmp_length = + (unsigned int) (DBL_MAX_EXP + * 0.30103 /* binary -> decimal */ + * 2 /* estimate for FLAG_GROUP */ + ) + + 1 /* turn floor into ceil */ + + 10; /* sign, decimal point etc. */ + tmp_length = xsum (tmp_length, precision); + break; + + case 'e': case 'E': case 'g': case 'G': + case 'a': case 'A': + tmp_length = + 12; /* sign, decimal point, exponent etc. */ + tmp_length = xsum (tmp_length, precision); + break; + + case 'c': +# if defined HAVE_WINT_T && !WIDE_CHAR_VERSION + if (type == TYPE_WIDE_CHAR) + tmp_length = MB_CUR_MAX; + else +# endif + tmp_length = 1; + break; + + case 's': +# ifdef HAVE_WCHAR_T + if (type == TYPE_WIDE_STRING) + { + tmp_length = + local_wcslen (a.arg[dp->arg_index].a.a_wide_string); + +# if !WIDE_CHAR_VERSION + tmp_length = xtimes (tmp_length, MB_CUR_MAX); +# endif + } + else +# endif + tmp_length = strlen (a.arg[dp->arg_index].a.a_string); + break; + + case 'p': + tmp_length = + (unsigned int) (sizeof (void *) * CHAR_BIT + * 0.25 /* binary -> hexadecimal */ + ) + + 1 /* turn floor into ceil */ + + 2; /* account for leading 0x */ + break; + + default: + abort (); + } + + if (tmp_length < width) + tmp_length = width; + + tmp_length = xsum (tmp_length, 1); /* account for trailing NUL */ + } + + if (tmp_length <= sizeof (tmpbuf) / sizeof (CHAR_T)) + tmp = tmpbuf; + else + { + size_t tmp_memsize = xtimes (tmp_length, sizeof (CHAR_T)); + + if (size_overflow_p (tmp_memsize)) + /* Overflow, would lead to out of memory. */ + goto out_of_memory; + tmp = (CHAR_T *) malloc (tmp_memsize); + if (tmp == NULL) + /* Out of memory. */ + goto out_of_memory; + } +#endif + + /* Construct the format string for calling snprintf or + sprintf. */ + p = buf; + *p++ = '%'; + if (dp->flags & FLAG_GROUP) + *p++ = '\''; + if (dp->flags & FLAG_LEFT) + *p++ = '-'; + if (dp->flags & FLAG_SHOWSIGN) + *p++ = '+'; + if (dp->flags & FLAG_SPACE) + *p++ = ' '; + if (dp->flags & FLAG_ALT) + *p++ = '#'; + if (dp->flags & FLAG_ZERO) + *p++ = '0'; + if (dp->width_start != dp->width_end) + { + size_t n = dp->width_end - dp->width_start; + memcpy (p, dp->width_start, n * sizeof (CHAR_T)); + p += n; + } + if (dp->precision_start != dp->precision_end) + { + size_t n = dp->precision_end - dp->precision_start; + memcpy (p, dp->precision_start, n * sizeof (CHAR_T)); + p += n; + } + + switch (type) + { +#ifdef HAVE_LONG_LONG + case TYPE_LONGLONGINT: + case TYPE_ULONGLONGINT: + *p++ = 'l'; + /*FALLTHROUGH*/ +#endif + case TYPE_LONGINT: + case TYPE_ULONGINT: +#ifdef HAVE_WINT_T + case TYPE_WIDE_CHAR: +#endif +#ifdef HAVE_WCHAR_T + case TYPE_WIDE_STRING: +#endif + *p++ = 'l'; + break; +#ifdef HAVE_LONG_DOUBLE + case TYPE_LONGDOUBLE: + *p++ = 'L'; + break; +#endif + default: + break; + } + *p = dp->conversion; +#if USE_SNPRINTF + p[1] = '%'; + p[2] = 'n'; + p[3] = '\0'; +#else + p[1] = '\0'; +#endif + + /* Construct the arguments for calling snprintf or sprintf. */ + prefix_count = 0; + if (dp->width_arg_index != ARG_NONE) + { + if (!(a.arg[dp->width_arg_index].type == TYPE_INT)) + abort (); + prefixes[prefix_count++] = a.arg[dp->width_arg_index].a.a_int; + } + if (dp->precision_arg_index != ARG_NONE) + { + if (!(a.arg[dp->precision_arg_index].type == TYPE_INT)) + abort (); + prefixes[prefix_count++] = a.arg[dp->precision_arg_index].a.a_int; + } + +#if USE_SNPRINTF + /* Prepare checking whether snprintf returns the count + via %n. */ + ENSURE_ALLOCATION (xsum (length, 1)); + result[length] = '\0'; +#endif + + for (;;) + { + size_t maxlen; + int count; + int retcount; + + maxlen = allocated - length; + count = -1; + retcount = 0; + +#if USE_SNPRINTF +# define SNPRINTF_BUF(arg) \ + switch (prefix_count) \ + { \ + case 0: \ + retcount = SNPRINTF (result + length, maxlen, buf, \ + arg, &count); \ + break; \ + case 1: \ + retcount = SNPRINTF (result + length, maxlen, buf, \ + prefixes[0], arg, &count); \ + break; \ + case 2: \ + retcount = SNPRINTF (result + length, maxlen, buf, \ + prefixes[0], prefixes[1], arg, \ + &count); \ + break; \ + default: \ + abort (); \ + } +#else +# define SNPRINTF_BUF(arg) \ + switch (prefix_count) \ + { \ + case 0: \ + count = sprintf (tmp, buf, arg); \ + break; \ + case 1: \ + count = sprintf (tmp, buf, prefixes[0], arg); \ + break; \ + case 2: \ + count = sprintf (tmp, buf, prefixes[0], prefixes[1],\ + arg); \ + break; \ + default: \ + abort (); \ + } +#endif + + switch (type) + { + case TYPE_SCHAR: + { + int arg = a.arg[dp->arg_index].a.a_schar; + SNPRINTF_BUF (arg); + } + break; + case TYPE_UCHAR: + { + unsigned int arg = a.arg[dp->arg_index].a.a_uchar; + SNPRINTF_BUF (arg); + } + break; + case TYPE_SHORT: + { + int arg = a.arg[dp->arg_index].a.a_short; + SNPRINTF_BUF (arg); + } + break; + case TYPE_USHORT: + { + unsigned int arg = a.arg[dp->arg_index].a.a_ushort; + SNPRINTF_BUF (arg); + } + break; + case TYPE_INT: + { + int arg = a.arg[dp->arg_index].a.a_int; + SNPRINTF_BUF (arg); + } + break; + case TYPE_UINT: + { + unsigned int arg = a.arg[dp->arg_index].a.a_uint; + SNPRINTF_BUF (arg); + } + break; + case TYPE_LONGINT: + { + long int arg = a.arg[dp->arg_index].a.a_longint; + SNPRINTF_BUF (arg); + } + break; + case TYPE_ULONGINT: + { + unsigned long int arg = a.arg[dp->arg_index].a.a_ulongint; + SNPRINTF_BUF (arg); + } + break; +#ifdef HAVE_LONG_LONG + case TYPE_LONGLONGINT: + { + long long int arg = a.arg[dp->arg_index].a.a_longlongint; + SNPRINTF_BUF (arg); + } + break; + case TYPE_ULONGLONGINT: + { + unsigned long long int arg = a.arg[dp->arg_index].a.a_ulonglongint; + SNPRINTF_BUF (arg); + } + break; +#endif + case TYPE_DOUBLE: + { + double arg = a.arg[dp->arg_index].a.a_double; + SNPRINTF_BUF (arg); + } + break; +#ifdef HAVE_LONG_DOUBLE + case TYPE_LONGDOUBLE: + { + long double arg = a.arg[dp->arg_index].a.a_longdouble; + SNPRINTF_BUF (arg); + } + break; +#endif + case TYPE_CHAR: + { + int arg = a.arg[dp->arg_index].a.a_char; + SNPRINTF_BUF (arg); + } + break; +#ifdef HAVE_WINT_T + case TYPE_WIDE_CHAR: + { + wint_t arg = a.arg[dp->arg_index].a.a_wide_char; + SNPRINTF_BUF (arg); + } + break; +#endif + case TYPE_STRING: + { + const char *arg = a.arg[dp->arg_index].a.a_string; + SNPRINTF_BUF (arg); + } + break; +#ifdef HAVE_WCHAR_T + case TYPE_WIDE_STRING: + { + const wchar_t *arg = a.arg[dp->arg_index].a.a_wide_string; + SNPRINTF_BUF (arg); + } + break; +#endif + case TYPE_POINTER: + { + void *arg = a.arg[dp->arg_index].a.a_pointer; + SNPRINTF_BUF (arg); + } + break; + default: + abort (); + } + +#if USE_SNPRINTF + /* Portability: Not all implementations of snprintf() + are ISO C 99 compliant. Determine the number of + bytes that snprintf() has produced or would have + produced. */ + if (count >= 0) + { + /* Verify that snprintf() has NUL-terminated its + result. */ + if (count < maxlen && result[length + count] != '\0') + abort (); + /* Portability hack. */ + if (retcount > count) + count = retcount; + } + else + { + /* snprintf() doesn't understand the '%n' + directive. */ + if (p[1] != '\0') + { + /* Don't use the '%n' directive; instead, look + at the snprintf() return value. */ + p[1] = '\0'; + continue; + } + else + { + /* Look at the snprintf() return value. */ + if (retcount < 0) + { + /* HP-UX 10.20 snprintf() is doubly deficient: + It doesn't understand the '%n' directive, + *and* it returns -1 (rather than the length + that would have been required) when the + buffer is too small. */ + size_t bigger_need = + xsum (xtimes (allocated, 2), 12); + ENSURE_ALLOCATION (bigger_need); + continue; + } + else + count = retcount; + } + } +#endif + + /* Attempt to handle failure. */ + if (count < 0) + { + if (!(result == resultbuf || result == NULL)) + free (result); + if (buf_malloced != NULL) + free (buf_malloced); + CLEANUP (); + errno = EINVAL; + return NULL; + } + +#if !USE_SNPRINTF + if (count >= tmp_length) + /* tmp_length was incorrectly calculated - fix the + code above! */ + abort (); +#endif + + /* Make room for the result. */ + if (count >= maxlen) + { + /* Need at least count bytes. But allocate + proportionally, to avoid looping eternally if + snprintf() reports a too small count. */ + size_t n = + xmax (xsum (length, count), xtimes (allocated, 2)); + + ENSURE_ALLOCATION (n); +#if USE_SNPRINTF + continue; +#endif + } + +#if USE_SNPRINTF + /* The snprintf() result did fit. */ +#else + /* Append the sprintf() result. */ + memcpy (result + length, tmp, count * sizeof (CHAR_T)); + if (tmp != tmpbuf) + free (tmp); +#endif + + length += count; + break; + } + } + } + } + + /* Add the final NUL. */ + ENSURE_ALLOCATION (xsum (length, 1)); + result[length] = '\0'; + + if (result != resultbuf && length + 1 < allocated) + { + /* Shrink the allocated memory if possible. */ + CHAR_T *memory; + + memory = (CHAR_T *) realloc (result, (length + 1) * sizeof (CHAR_T)); + if (memory != NULL) + result = memory; + } + + if (buf_malloced != NULL) + free (buf_malloced); + CLEANUP (); + *lengthp = length; + if (length > INT_MAX) + goto length_overflow; + return result; + + length_overflow: + /* We could produce such a big string, but its length doesn't fit into + an 'int'. POSIX says that snprintf() fails with errno = EOVERFLOW in + this case. */ + if (result != resultbuf) + free (result); + errno = EOVERFLOW; + return NULL; + + out_of_memory: + if (!(result == resultbuf || result == NULL)) + free (result); + if (buf_malloced != NULL) + free (buf_malloced); + out_of_memory_1: + CLEANUP (); + errno = ENOMEM; + return NULL; + } +} + +#undef SNPRINTF +#undef USE_SNPRINTF +#undef PRINTF_PARSE +#undef DIRECTIVES +#undef DIRECTIVE +#undef CHAR_T +#undef VASNPRINTF diff --git a/lib/vasnprintf.h b/lib/vasnprintf.h new file mode 100644 index 0000000..894008c --- /dev/null +++ b/lib/vasnprintf.h @@ -0,0 +1,77 @@ +/* vsprintf with automatic memory allocation. + Copyright (C) 2002-2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _VASNPRINTF_H +#define _VASNPRINTF_H + +/* Get va_list. */ +#include <stdarg.h> + +/* Get size_t. */ +#include <stddef.h> + +#ifndef __attribute__ +/* This feature is available in gcc versions 2.5 and later. */ +# if __GNUC__ < 2 || (__GNUC__ == 2 && __GNUC_MINOR__ < 5) || __STRICT_ANSI__ +# define __attribute__(Spec) /* empty */ +# endif +/* The __-protected variants of `format' and `printf' attributes + are accepted by gcc versions 2.6.4 (effectively 2.7) and later. */ +# if __GNUC__ < 2 || (__GNUC__ == 2 && __GNUC_MINOR__ < 7) +# define __format__ format +# define __printf__ printf +# endif +#endif + +#ifdef __cplusplus +extern "C" { +#endif + +/* Write formatted output to a string dynamically allocated with malloc(). + You can pass a preallocated buffer for the result in RESULTBUF and its + size in *LENGTHP; otherwise you pass RESULTBUF = NULL. + If successful, return the address of the string (this may be = RESULTBUF + if no dynamic memory allocation was necessary) and set *LENGTHP to the + number of resulting bytes, excluding the trailing NUL. Upon error, set + errno and return NULL. + + When dynamic memory allocation occurs, the preallocated buffer is left + alone (with possibly modified contents). This makes it possible to use + a statically allocated or stack-allocated buffer, like this: + + char buf[100]; + size_t len = sizeof (buf); + char *output = vasnprintf (buf, &len, format, args); + if (output == NULL) + ... error handling ...; + else + { + ... use the output string ...; + if (output != buf) + free (output); + } + */ +extern char * asnprintf (char *resultbuf, size_t *lengthp, const char *format, ...) + __attribute__ ((__format__ (__printf__, 3, 4))); +extern char * vasnprintf (char *resultbuf, size_t *lengthp, const char *format, va_list args) + __attribute__ ((__format__ (__printf__, 3, 0))); + +#ifdef __cplusplus +} +#endif + +#endif /* _VASNPRINTF_H */ diff --git a/lib/vasprintf.c b/lib/vasprintf.c new file mode 100644 index 0000000..149c292 --- /dev/null +++ b/lib/vasprintf.c @@ -0,0 +1,42 @@ +/* Formatted output to strings. + Copyright (C) 1999, 2002 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Specification. */ +#include "vasprintf.h" + +#include <stdlib.h> + +#include "vasnprintf.h" + +int +vasprintf (char **resultp, const char *format, va_list args) +{ + size_t length; + char *result = vasnprintf (NULL, &length, format, args); + if (result == NULL) + return -1; + + *resultp = result; + /* Return the number of resulting bytes, excluding the trailing NUL. + If it wouldn't fit in an 'int', vasnprintf() would have returned NULL + and set errno to EOVERFLOW. */ + return length; +} diff --git a/lib/vasprintf.h b/lib/vasprintf.h new file mode 100644 index 0000000..d02f154 --- /dev/null +++ b/lib/vasprintf.h @@ -0,0 +1,63 @@ +/* vsprintf with automatic memory allocation. + Copyright (C) 2002-2003 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License along + with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _VASPRINTF_H +#define _VASPRINTF_H + +#if HAVE_VASPRINTF + +/* Get asprintf(), vasprintf() declarations. */ +#include <stdio.h> + +#else + +/* Get va_list. */ +#include <stdarg.h> + +#ifndef __attribute__ +/* This feature is available in gcc versions 2.5 and later. */ +# if __GNUC__ < 2 || (__GNUC__ == 2 && __GNUC_MINOR__ < 5) || __STRICT_ANSI__ +# define __attribute__(Spec) /* empty */ +# endif +/* The __-protected variants of `format' and `printf' attributes + are accepted by gcc versions 2.6.4 (effectively 2.7) and later. */ +# if __GNUC__ < 2 || (__GNUC__ == 2 && __GNUC_MINOR__ < 7) +# define __format__ format +# define __printf__ printf +# endif +#endif + +#ifdef __cplusplus +extern "C" { +#endif + +/* Write formatted output to a string dynamically allocated with malloc(). + If the memory allocation succeeds, store the address of the string in + *RESULT and return the number of resulting bytes, excluding the trailing + NUL. Upon memory allocation error, or some other error, return -1. */ +extern int asprintf (char **result, const char *format, ...) + __attribute__ ((__format__ (__printf__, 2, 3))); +extern int vasprintf (char **result, const char *format, va_list args) + __attribute__ ((__format__ (__printf__, 2, 0))); + +#ifdef __cplusplus +} +#endif + +#endif + +#endif /* _VASPRINTF_H */ diff --git a/lib/wait.h b/lib/wait.h new file mode 100644 index 0000000..81df938 --- /dev/null +++ b/lib/wait.h @@ -0,0 +1,42 @@ +/* wait.h -- POSIX macros for evaluating exit statuses + Copyright (C) 1990 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. */ + +#ifdef HAVE_SYS_WAIT_H +#include <sys/types.h> /* For pid_t. */ +#ifdef HAVE_SYS_RESOURCE_H +#include <sys/resource.h> /* for rusage */ +#endif +#include <sys/wait.h> +#endif +#ifndef WIFSTOPPED +#define WIFSTOPPED(w) (((w) & 0xff) == 0x7f) +#endif +#ifndef WIFSIGNALED +#define WIFSIGNALED(w) (((w) & 0xff) != 0x7f && ((w) & 0xff) != 0) +#endif +#ifndef WIFEXITED +#define WIFEXITED(w) (((w) & 0xff) == 0) +#endif +#ifndef WCOREDUMP /* not POSIX, but common and useful */ +#define WCOREDUMP(w) (((w) & 0x80) != 0) +#endif + +#ifndef WSTOPSIG +#define WSTOPSIG(w) (((w) >> 8) & 0xff) +#endif +#ifndef WTERMSIG +#define WTERMSIG(w) ((w) & 0x7f) +#endif +#ifndef WEXITSTATUS +#define WEXITSTATUS(w) (((w) >> 8) & 0xff) +#endif diff --git a/lib/waitpid.c b/lib/waitpid.c new file mode 100644 index 0000000..02d6acb --- /dev/null +++ b/lib/waitpid.c @@ -0,0 +1,80 @@ +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif + +#include "system.h" +#include "wait.h" + +#include <stdio.h> + +struct unreaped { + pid_t pid; + int status; +}; +static struct unreaped *unreaped; +static int n; + +static struct unreaped *ualloc (oldptr, n) + struct unreaped *oldptr; + int n; +{ + n *= sizeof (struct unreaped); + if (n == 0) + n = 1; + if (oldptr) + oldptr = (struct unreaped *) realloc ((char *) oldptr, n); + else + oldptr = (struct unreaped *) malloc (n); + if (oldptr == 0) + { + fprintf (stderr, "cannot allocate %d bytes\n", n); + exit (1); + } + return oldptr; +} + +pid_t waitpid (pid, status, options) + pid_t pid; + int *status; + int options; +{ + int i; + + /* initialize */ + if (unreaped == 0) + { + unreaped = ualloc (unreaped, 1); + unreaped[0].pid = 0; + n = 1; + } + + for (i = 0; unreaped[i].pid; i++) + if (unreaped[i].pid == pid) + { + *status = unreaped[i].status; + while (unreaped[i].pid) + { + unreaped[i] = unreaped[i+1]; + i++; + } + n--; + return pid; + } + + while (1) + { +#ifdef HAVE_WAIT3 + pid_t p = wait3 (status, options, (struct rusage *) 0); +#else + pid_t p = wait (status); +#endif + + if (p == 0 || p == -1 || p == pid) + return p; + + n++; + unreaped = ualloc (unreaped, n); + unreaped[n-1].pid = p; + unreaped[n-1].status = *status; + } +} diff --git a/lib/xalloc-die.c b/lib/xalloc-die.c new file mode 100644 index 0000000..ff5ac9e --- /dev/null +++ b/lib/xalloc-die.c @@ -0,0 +1,45 @@ +/* Report a memory allocation failure and exit. + + Copyright (C) 1997, 1998, 1999, 2000, 2002, 2003, 2004 Free + Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "xalloc.h" + +#include <stdlib.h> + +#include "error.h" +#include "exitfail.h" + +#include "gettext.h" +#define _(msgid) gettext (msgid) +#define N_(msgid) msgid + +void +xalloc_die (void) +{ + error (exit_failure, 0, "%s", _("memory exhausted")); + + /* The `noreturn' cannot be given to error, since it may return if + its first argument is 0. To help compilers understand the + xalloc_die does not return, call abort. Also, the abort is a + safety feature if exit_failure is 0 (which shouldn't happen). */ + abort (); +} diff --git a/lib/xalloc.h b/lib/xalloc.h new file mode 100644 index 0000000..f80977e --- /dev/null +++ b/lib/xalloc.h @@ -0,0 +1,79 @@ +/* xalloc.h -- malloc with out-of-memory checking + + Copyright (C) 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, + 1999, 2000, 2003, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef XALLOC_H_ +# define XALLOC_H_ + +# include <stddef.h> + + +# ifdef __cplusplus +extern "C" { +# endif + + +# ifndef __attribute__ +# if __GNUC__ < 2 || (__GNUC__ == 2 && __GNUC_MINOR__ < 8) || __STRICT_ANSI__ +# define __attribute__(x) +# endif +# endif + +# ifndef ATTRIBUTE_NORETURN +# define ATTRIBUTE_NORETURN __attribute__ ((__noreturn__)) +# endif + +/* This function is always triggered when memory is exhausted. + It must be defined by the application, either explicitly + or by using gnulib's xalloc-die module. This is the + function to call when one wants the program to die because of a + memory allocation failure. */ +extern void xalloc_die (void) ATTRIBUTE_NORETURN; + +void *xmalloc (size_t s); +void *xnmalloc (size_t n, size_t s); +void *xzalloc (size_t s); +void *xcalloc (size_t n, size_t s); +void *xrealloc (void *p, size_t s); +void *xnrealloc (void *p, size_t n, size_t s); +void *x2realloc (void *p, size_t *pn); +void *x2nrealloc (void *p, size_t *pn, size_t s); +void *xmemdup (void const *p, size_t s); +char *xstrdup (char const *str); + +/* Return 1 if an array of N objects, each of size S, cannot exist due + to size arithmetic overflow. S must be positive and N must be + nonnegative. This is a macro, not an inline function, so that it + works correctly even when SIZE_MAX < N. + + By gnulib convention, SIZE_MAX represents overflow in size + calculations, so the conservative dividend to use here is + SIZE_MAX - 1, since SIZE_MAX might represent an overflowed value. + However, malloc (SIZE_MAX) fails on all known hosts where + sizeof (ptrdiff_t) <= sizeof (size_t), so do not bother to test for + exactly-SIZE_MAX allocations on such hosts; this avoids a test and + branch when S is known to be 1. */ +# define xalloc_oversized(n, s) \ + ((size_t) (sizeof (ptrdiff_t) <= sizeof (size_t) ? -1 : -2) / (s) < (n)) + +# ifdef __cplusplus +} +# endif + + +#endif /* !XALLOC_H_ */ diff --git a/lib/xgetcwd.c b/lib/xgetcwd.c new file mode 100644 index 0000000..0f798da --- /dev/null +++ b/lib/xgetcwd.c @@ -0,0 +1,43 @@ +/* xgetcwd.c -- return current directory with unlimited length + + Copyright (C) 2001, 2003, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Jim Meyering. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "xgetcwd.h" + +#include <errno.h> + +#include "getcwd.h" +#include "xalloc.h" + +/* Return the current directory, newly allocated. + Upon an out-of-memory error, call xalloc_die. + Upon any other type of error, return NULL. */ + +char * +xgetcwd (void) +{ + char *cwd = getcwd (NULL, 0); + if (! cwd && errno == ENOMEM) + xalloc_die (); + return cwd; +} diff --git a/lib/xgetcwd.h b/lib/xgetcwd.h new file mode 100644 index 0000000..70afe35 --- /dev/null +++ b/lib/xgetcwd.h @@ -0,0 +1,18 @@ +/* prototype for xgetcwd + Copyright (C) 1995, 2001, 2003 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +extern char *xgetcwd (void); diff --git a/lib/xgethostname.c b/lib/xgethostname.c new file mode 100644 index 0000000..5b8b9b2 --- /dev/null +++ b/lib/xgethostname.c @@ -0,0 +1,84 @@ +/* xgethostname.c -- return current hostname with unlimited length + + Copyright (C) 1992, 1996, 2000, 2001, 2003, 2004, 2005 Free Software + Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* written by Jim Meyering */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +/* Specification. */ +#include "xgethostname.h" + +#include <stdlib.h> +#include <errno.h> + +#if HAVE_UNISTD_H +# include <unistd.h> +#endif + +#include "xalloc.h" + +#ifndef ENAMETOOLONG +# define ENAMETOOLONG 0 +#endif + +#ifndef INITIAL_HOSTNAME_LENGTH +# define INITIAL_HOSTNAME_LENGTH 34 +#endif + +/* Return the current hostname in malloc'd storage. + If malloc fails, exit. + Upon any other failure, return NULL and set errno. */ +char * +xgethostname (void) +{ + char *hostname = NULL; + size_t size = INITIAL_HOSTNAME_LENGTH; + + while (1) + { + /* Use SIZE_1 here rather than SIZE to work around the bug in + SunOS 5.5's gethostname whereby it NUL-terminates HOSTNAME + even when the name is as long as the supplied buffer. */ + size_t size_1; + + hostname = x2realloc (hostname, &size); + size_1 = size - 1; + hostname[size_1 - 1] = '\0'; + errno = 0; + + if (gethostname (hostname, size_1) == 0) + { + if (! hostname[size_1 - 1]) + break; + } + else if (errno != 0 && errno != ENAMETOOLONG && errno != EINVAL + /* OSX/Darwin does this when the buffer is not large enough */ + && errno != ENOMEM) + { + int saved_errno = errno; + free (hostname); + errno = saved_errno; + return NULL; + } + } + + return hostname; +} diff --git a/lib/xgethostname.h b/lib/xgethostname.h new file mode 100644 index 0000000..0177a40 --- /dev/null +++ b/lib/xgethostname.h @@ -0,0 +1 @@ +char *xgethostname (void); diff --git a/lib/xmalloc.c b/lib/xmalloc.c new file mode 100644 index 0000000..687633c --- /dev/null +++ b/lib/xmalloc.c @@ -0,0 +1,241 @@ +/* xmalloc.c -- malloc with out of memory checking + + Copyright (C) 1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, + 1999, 2000, 2002, 2003, 2004, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "xalloc.h" + +#include <stdlib.h> +#include <string.h> + +#ifndef SIZE_MAX +# define SIZE_MAX ((size_t) -1) +#endif + +/* 1 if calloc is known to be compatible with GNU calloc. This + matters if we are not also using the calloc module, which defines + HAVE_CALLOC and supports the GNU API even on non-GNU platforms. */ +#if defined HAVE_CALLOC || defined __GLIBC__ +enum { HAVE_GNU_CALLOC = 1 }; +#else +enum { HAVE_GNU_CALLOC = 0 }; +#endif + +/* Allocate an array of N objects, each with S bytes of memory, + dynamically, with error checking. S must be nonzero. */ + +static inline void * +xnmalloc_inline (size_t n, size_t s) +{ + void *p; + if (xalloc_oversized (n, s) || (! (p = malloc (n * s)) && n != 0)) + xalloc_die (); + return p; +} + +void * +xnmalloc (size_t n, size_t s) +{ + return xnmalloc_inline (n, s); +} + +/* Allocate N bytes of memory dynamically, with error checking. */ + +void * +xmalloc (size_t n) +{ + return xnmalloc_inline (n, 1); +} + +/* Change the size of an allocated block of memory P to an array of N + objects each of S bytes, with error checking. S must be nonzero. */ + +static inline void * +xnrealloc_inline (void *p, size_t n, size_t s) +{ + if (xalloc_oversized (n, s) || (! (p = realloc (p, n * s)) && n != 0)) + xalloc_die (); + return p; +} + +void * +xnrealloc (void *p, size_t n, size_t s) +{ + return xnrealloc_inline (p, n, s); +} + +/* Change the size of an allocated block of memory P to N bytes, + with error checking. */ + +void * +xrealloc (void *p, size_t n) +{ + return xnrealloc_inline (p, n, 1); +} + + +/* If P is null, allocate a block of at least *PN such objects; + otherwise, reallocate P so that it contains more than *PN objects + each of S bytes. *PN must be nonzero unless P is null, and S must + be nonzero. Set *PN to the new number of objects, and return the + pointer to the new block. *PN is never set to zero, and the + returned pointer is never null. + + Repeated reallocations are guaranteed to make progress, either by + allocating an initial block with a nonzero size, or by allocating a + larger block. + + In the following implementation, nonzero sizes are doubled so that + repeated reallocations have O(N log N) overall cost rather than + O(N**2) cost, but the specification for this function does not + guarantee that sizes are doubled. + + Here is an example of use: + + int *p = NULL; + size_t used = 0; + size_t allocated = 0; + + void + append_int (int value) + { + if (used == allocated) + p = x2nrealloc (p, &allocated, sizeof *p); + p[used++] = value; + } + + This causes x2nrealloc to allocate a block of some nonzero size the + first time it is called. + + To have finer-grained control over the initial size, set *PN to a + nonzero value before calling this function with P == NULL. For + example: + + int *p = NULL; + size_t used = 0; + size_t allocated = 0; + size_t allocated1 = 1000; + + void + append_int (int value) + { + if (used == allocated) + { + p = x2nrealloc (p, &allocated1, sizeof *p); + allocated = allocated1; + } + p[used++] = value; + } + + */ + +static inline void * +x2nrealloc_inline (void *p, size_t *pn, size_t s) +{ + size_t n = *pn; + + if (! p) + { + if (! n) + { + /* The approximate size to use for initial small allocation + requests, when the invoking code specifies an old size of + zero. 64 bytes is the largest "small" request for the + GNU C library malloc. */ + enum { DEFAULT_MXFAST = 64 }; + + n = DEFAULT_MXFAST / s; + n += !n; + } + } + else + { + if (SIZE_MAX / 2 / s < n) + xalloc_die (); + n *= 2; + } + + *pn = n; + return xrealloc (p, n * s); +} + +void * +x2nrealloc (void *p, size_t *pn, size_t s) +{ + return x2nrealloc_inline (p, pn, s); +} + +/* If P is null, allocate a block of at least *PN bytes; otherwise, + reallocate P so that it contains more than *PN bytes. *PN must be + nonzero unless P is null. Set *PN to the new block's size, and + return the pointer to the new block. *PN is never set to zero, and + the returned pointer is never null. */ + +void * +x2realloc (void *p, size_t *pn) +{ + return x2nrealloc_inline (p, pn, 1); +} + +/* Allocate S bytes of zeroed memory dynamically, with error checking. + There's no need for xnzalloc (N, S), since it would be equivalent + to xcalloc (N, S). */ + +void * +xzalloc (size_t s) +{ + return memset (xmalloc (s), 0, s); +} + +/* Allocate zeroed memory for N elements of S bytes, with error + checking. S must be nonzero. */ + +void * +xcalloc (size_t n, size_t s) +{ + void *p; + /* Test for overflow, since some calloc implementations don't have + proper overflow checks. But omit overflow and size-zero tests if + HAVE_GNU_CALLOC, since GNU calloc catches overflow and never + returns NULL if successful. */ + if ((! HAVE_GNU_CALLOC && xalloc_oversized (n, s)) + || (! (p = calloc (n, s)) && (HAVE_GNU_CALLOC || n != 0))) + xalloc_die (); + return p; +} + +/* Clone an object P of size S, with error checking. There's no need + for xnmemdup (P, N, S), since xmemdup (P, N * S) works without any + need for an arithmetic overflow check. */ + +void * +xmemdup (void const *p, size_t s) +{ + return memcpy (xmalloc (s), p, s); +} + +/* Clone STRING. */ + +char * +xstrdup (char const *string) +{ + return xmemdup (string, strlen (string) + 1); +} diff --git a/lib/xreadlink.c b/lib/xreadlink.c new file mode 100644 index 0000000..f21812d --- /dev/null +++ b/lib/xreadlink.c @@ -0,0 +1,93 @@ +/* xreadlink.c -- readlink wrapper to return the link name in malloc'd storage + + Copyright (C) 2001, 2003, 2004, 2005 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; see the file COPYING. + If not, write to the Free Software Foundation, + 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Jim Meyering <jim@meyering.net> */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "xreadlink.h" + +#include <stdio.h> +#include <errno.h> +#include <limits.h> +#include <sys/types.h> +#include <stdlib.h> +#if HAVE_UNISTD_H +# include <unistd.h> +#endif + +#ifndef SIZE_MAX +# define SIZE_MAX ((size_t) -1) +#endif +#ifndef SSIZE_MAX +# define SSIZE_MAX ((ssize_t) (SIZE_MAX / 2)) +#endif + +#define MAXSIZE (SIZE_MAX < SSIZE_MAX ? SIZE_MAX : SSIZE_MAX) + +#include "xalloc.h" + +/* Call readlink to get the symbolic link value of FILE. + SIZE is a hint as to how long the link is expected to be; + typically it is taken from st_size. It need not be correct. + Return a pointer to that NUL-terminated string in malloc'd storage. + If readlink fails, return NULL (caller may use errno to diagnose). + If malloc fails, or if the link value is longer than SSIZE_MAX :-), + give a diagnostic and exit. */ + +char * +xreadlink (char const *file, size_t size) +{ + /* The initial buffer size for the link value. A power of 2 + detects arithmetic overflow earlier, but is not required. */ + size_t buf_size = size < MAXSIZE ? size + 1 : MAXSIZE; + + while (1) + { + char *buffer = xmalloc (buf_size); + ssize_t r = readlink (file, buffer, buf_size); + size_t link_length = r; + + /* On AIX 5L v5.3 and HP-UX 11i v2 04/09, readlink returns -1 + with errno == ERANGE if the buffer is too small. */ + if (r < 0 && errno != ERANGE) + { + int saved_errno = errno; + free (buffer); + errno = saved_errno; + return NULL; + } + + if (link_length < buf_size) + { + buffer[link_length] = 0; + return buffer; + } + + free (buffer); + if (buf_size <= MAXSIZE / 2) + buf_size *= 2; + else if (buf_size < MAXSIZE) + buf_size = MAXSIZE; + else + xalloc_die (); + } +} diff --git a/lib/xreadlink.h b/lib/xreadlink.h new file mode 100644 index 0000000..9fcf836 --- /dev/null +++ b/lib/xreadlink.h @@ -0,0 +1,23 @@ +/* readlink wrapper to return the link name in malloc'd storage + + Copyright (C) 2001, 2003, 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; see the file COPYING. + If not, write to the Free Software Foundation, + 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +/* Written by Jim Meyering <jim@meyering.net> */ + +#include <stddef.h> +char *xreadlink (char const *, size_t); diff --git a/lib/xselect.h b/lib/xselect.h new file mode 100644 index 0000000..5605cbd --- /dev/null +++ b/lib/xselect.h @@ -0,0 +1,21 @@ +/* This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. */ + +/* This file simply performs the include magic necessary for using select */ + +/* select also requires <sys/types.h>, "xtime.h", and <unistd.h> */ + +#ifdef HAVE_SYS_BSDTYPES_H +# include <sys/bsdtypes.h> +#endif + +#if HAVE_SYS_SELECT_H +# include <sys/select.h> +#endif diff --git a/lib/xsize.h b/lib/xsize.h new file mode 100644 index 0000000..341fb16 --- /dev/null +++ b/lib/xsize.h @@ -0,0 +1,108 @@ +/* xsize.h -- Checked size_t computations. + + Copyright (C) 2003 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef _XSIZE_H +#define _XSIZE_H + +/* Get size_t. */ +#include <stddef.h> + +/* Get SIZE_MAX. */ +#include <limits.h> +#if HAVE_STDINT_H +# include <stdint.h> +#endif + +/* The size of memory objects is often computed through expressions of + type size_t. Example: + void* p = malloc (header_size + n * element_size). + These computations can lead to overflow. When this happens, malloc() + returns a piece of memory that is way too small, and the program then + crashes while attempting to fill the memory. + To avoid this, the functions and macros in this file check for overflow. + The convention is that SIZE_MAX represents overflow. + malloc (SIZE_MAX) is not guaranteed to fail -- think of a malloc + implementation that uses mmap --, it's recommended to use size_overflow_p() + or size_in_bounds_p() before invoking malloc(). + The example thus becomes: + size_t size = xsum (header_size, xtimes (n, element_size)); + void *p = (size_in_bounds_p (size) ? malloc (size) : NULL); +*/ + +/* Convert an arbitrary value >= 0 to type size_t. */ +#define xcast_size_t(N) \ + ((N) <= SIZE_MAX ? (size_t) (N) : SIZE_MAX) + +/* Sum of two sizes, with overflow check. */ +static inline size_t +#if __GNUC__ >= 3 +__attribute__ ((__pure__)) +#endif +xsum (size_t size1, size_t size2) +{ + size_t sum = size1 + size2; + return (sum >= size1 ? sum : SIZE_MAX); +} + +/* Sum of three sizes, with overflow check. */ +static inline size_t +#if __GNUC__ >= 3 +__attribute__ ((__pure__)) +#endif +xsum3 (size_t size1, size_t size2, size_t size3) +{ + return xsum (xsum (size1, size2), size3); +} + +/* Sum of four sizes, with overflow check. */ +static inline size_t +#if __GNUC__ >= 3 +__attribute__ ((__pure__)) +#endif +xsum4 (size_t size1, size_t size2, size_t size3, size_t size4) +{ + return xsum (xsum (xsum (size1, size2), size3), size4); +} + +/* Maximum of two sizes, with overflow check. */ +static inline size_t +#if __GNUC__ >= 3 +__attribute__ ((__pure__)) +#endif +xmax (size_t size1, size_t size2) +{ + /* No explicit check is needed here, because for any n: + max (SIZE_MAX, n) == SIZE_MAX and max (n, SIZE_MAX) == SIZE_MAX. */ + return (size1 >= size2 ? size1 : size2); +} + +/* Multiplication of a count with an element size, with overflow check. + The count must be >= 0 and the element size must be > 0. + This is a macro, not an inline function, so that it works correctly even + when N is of a wider tupe and N > SIZE_MAX. */ +#define xtimes(N, ELSIZE) \ + ((N) <= SIZE_MAX / (ELSIZE) ? (size_t) (N) * (ELSIZE) : SIZE_MAX) + +/* Check for overflow. */ +#define size_overflow_p(SIZE) \ + ((SIZE) == SIZE_MAX) +/* Check against overflow. */ +#define size_in_bounds_p(SIZE) \ + ((SIZE) != SIZE_MAX) + +#endif /* _XSIZE_H */ diff --git a/lib/xtime.h b/lib/xtime.h new file mode 100644 index 0000000..f9fd46e --- /dev/null +++ b/lib/xtime.h @@ -0,0 +1,42 @@ +/* This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. */ + +/* This file simply performs the include magic necessary for using time + * functions + */ +#ifndef XTIME_HEADER_INCLUDED +#define XTIME_HEADER_INCLUDED + +#ifdef vms +# include <time.h> +#else /* vms */ + +# if TIME_WITH_SYS_TIME +# include <sys/time.h> +# include <time.h> +# else /* TIME_WITH_SYS_TIME */ +# if HAVE_SYS_TIME_H +# include <sys/time.h> +# else /* HAVE_SYS_TIME_H */ +# include <time.h> +# endif /* !HAVE_SYS_TIME_H */ +# endif /* !TIME_WITH_SYS_TIME */ + +# ifdef timezone +# undef timezone /* needed for sgi */ +# endif /* timezone */ + +# if !defined(HAVE_FTIME) && !defined(HAVE_TIMEZONE) +extern long timezone; +# endif /* !defined(HAVE_FTIME) && !defined(HAVE_TIMEZONE) */ + +#endif /* !vms */ + +#endif /* !XTIME_HEADER_INCLUDED */ diff --git a/lib/yesno.c b/lib/yesno.c new file mode 100644 index 0000000..a364a1b --- /dev/null +++ b/lib/yesno.c @@ -0,0 +1,53 @@ +/* yesno.c -- read a yes/no response from stdin + + Copyright (C) 1990, 1998, 2001, 2003, 2004, 2005 Free Software + Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software Foundation, + Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifdef HAVE_CONFIG_H +# include <config.h> +#endif + +#include "yesno.h" + +#include <stdlib.h> +#include <stdio.h> + +#include "getline.h" + +/* Return true if we read an affirmative line from standard input. */ + +extern int rpmatch (char const *response); + +bool +yesno (void) +{ + char *response = NULL; + size_t response_size = 0; + ssize_t response_len = getline (&response, &response_size, stdin); + bool yes; + + if (response_len <= 0) + yes = false; + else + { + response[response_len - 1] = '\0'; + yes = (0 < rpmatch (response)); + } + + free (response); + return yes; +} diff --git a/lib/yesno.h b/lib/yesno.h new file mode 100644 index 0000000..dfa70bc --- /dev/null +++ b/lib/yesno.h @@ -0,0 +1,26 @@ +/* declare yesno + Copyright (C) 2004 Free Software Foundation, Inc. + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2, or (at your option) + any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; see the file COPYING. + If not, write to the Free Software Foundation, + 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#ifndef YESNO_H_ +# define YESNO_H_ + +# include <stdbool.h> + +bool yesno (void); + +#endif |