Tidies to allow easier embedded compilation; avoid (double) where possible.

git-svn-id: svn://vcs.exim.org/pcre/code/trunk@475 2f5784b3-3f2a-0410-8824-cb99058d5e15

9 files changed, 97 insertions, 46 deletions

diff --git a/CMakeLists.txt b/CMakeLists.txt
index 344fb0f..44e119f 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -35,6 +35,7 @@
 #            to disable the final configuration report.
 # 2009-04-11 PH applied Christian Ehrlicher's patch to show compiler flags that
 #            are set by specifying a release type.
+# 2010-01-02 PH added test for stdint.h
 
 PROJECT(PCRE C CXX)
 
@@ -55,6 +56,7 @@ INCLUDE(CheckFunctionExists)
 INCLUDE(CheckTypeSize)
 
 CHECK_INCLUDE_FILE(dirent.h     HAVE_DIRENT_H)
+CHECK_INCLUDE_FILE(stdint.h     HAVE_STDINT_H)
 CHECK_INCLUDE_FILE(sys/stat.h   HAVE_SYS_STAT_H)
 CHECK_INCLUDE_FILE(sys/types.h  HAVE_SYS_TYPES_H)
 CHECK_INCLUDE_FILE(unistd.h     HAVE_UNISTD_H)
diff --git a/ChangeLog b/ChangeLog
index bb10913..64588db 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -30,6 +30,30 @@ Version 8.01 11-Dec-09
 5.  The C++ GlobalReplace function was not working like Perl for the special
     situation when an empty string is matched. It now does the fancy magic
     stuff that is necessary. 
+    
+6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been 
+    removed. (These were left over from very, very early versions of PCRE.)
+    
+7.  Some cosmetic changes to the code to make life easier when compiling it
+    as part of something else:
+    
+    (a) Change DEBUG to PCRE_DEBUG. 
+    
+    (b) In pcre_compile(), rename the member of the "branch_chain" structure 
+        called "current" as "current_branch", to prevent a collision with the 
+        Linux macro when compiled as a kernel module.
+        
+    (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to 
+        prevent a collision with the Linux macro when compiled as a kernel 
+        module.
+        
+8.  In pcre_compile() there are some checks for integer overflows that used to
+    cast potentially large values to (double). This has been changed to that
+    when building, a check for int64_t is made, and if it is found, it is used 
+    instead, thus avoiding the use of floating point arithmetic. (There is no 
+    other use of FP in PCRE.) If int64_t is not found, the fallback is to 
+    double. 
+       
 
 
 Version 8.00 19-Oct-09
diff --git a/configure.ac b/configure.ac
index fc96ec4..1765441 100644
--- a/configure.ac
+++ b/configure.ac
@@ -9,7 +9,7 @@ dnl empty.
 m4_define(pcre_major, [8])
 m4_define(pcre_minor, [01])
 m4_define(pcre_prerelease, [])
-m4_define(pcre_date, [2009-12-11])
+m4_define(pcre_date, [2010-01-02])
 
 # Libtool shared library interface versions (current:revision:age)
 m4_define(libpcre_version, [0:1:0])
@@ -66,6 +66,9 @@ AC_LANG_PUSH(C++)
 AC_COMPILE_IFELSE(AC_LANG_PROGRAM([],[]),, CXX=""; CXXCP=""; CXXFLAGS="")
 AC_LANG_POP
 
+# Check for a 64-bit integer type
+AC_TYPE_INT64_T
+
 AC_PROG_INSTALL
 AC_LIBTOOL_WIN32_DLL
 AC_PROG_LIBTOOL
diff --git a/pcre_compile.c b/pcre_compile.c
index 0605a6e..2065829 100644
--- a/pcre_compile.c
+++ b/pcre_compile.c
@@ -6,7 +6,7 @@
 and semantics are as close as possible to those of the Perl 5 language.
 
                        Written by Philip Hazel
-           Copyright (c) 1997-2009 University of Cambridge
+           Copyright (c) 1997-2010 University of Cambridge
 
 -----------------------------------------------------------------------------
 Redistribution and use in source and binary forms, with or without
@@ -53,10 +53,11 @@ supporting internal functions that are not used by other modules. */
 #include "pcre_internal.h"
 
 
-/* When DEBUG is defined, we need the pcre_printint() function, which is also
-used by pcretest. DEBUG is not defined when building a production library. */
+/* When PCRE_DEBUG is defined, we need the pcre_printint() function, which is
+also used by pcretest. PCRE_DEBUG is not defined when building a production
+library. */
 
-#ifdef DEBUG
+#ifdef PCRE_DEBUG
 #include "pcre_printint.src"
 #endif
 
@@ -1994,9 +1995,10 @@ static BOOL
 could_be_empty(const uschar *code, const uschar *endcode, branch_chain *bcptr,
   BOOL utf8)
 {
-while (bcptr != NULL && bcptr->current >= code)
+while (bcptr != NULL && bcptr->current_branch >= code)
   {
-  if (!could_be_empty_branch(bcptr->current, endcode, utf8)) return FALSE;
+  if (!could_be_empty_branch(bcptr->current_branch, endcode, utf8)) 
+    return FALSE;
   bcptr = bcptr->outer;
   }
 return TRUE;
@@ -2658,7 +2660,7 @@ BOOL utf8 = FALSE;
 uschar *utf8_char = NULL;
 #endif
 
-#ifdef DEBUG
+#ifdef PCRE_DEBUG
 if (lengthptr != NULL) DPRINTF((">> start branch\n"));
 #endif
 
@@ -2717,7 +2719,7 @@ for (;; ptr++)
 
   if (lengthptr != NULL)
     {
-#ifdef DEBUG
+#ifdef PCRE_DEBUG
     if (code > cd->hwm) cd->hwm = code;                 /* High water info */
 #endif
     if (code > cd->start_workspace + COMPILE_WORK_SIZE) /* Check for overrun */
@@ -4213,13 +4215,15 @@ we set the flag only if there is a literal "\r" or "\n" in the class. */
           {
           /* In the pre-compile phase, we don't actually do the replication. We
           just adjust the length as if we had. Do some paranoid checks for
-          potential integer overflow. */
+          potential integer overflow. The INT64_OR_DOUBLE type is a 64-bit
+          integer type when available, otherwise double. */
 
           if (lengthptr != NULL)
             {
             int delta = (repeat_min - 1)*length_prevgroup;
-            if ((double)(repeat_min - 1)*(double)length_prevgroup >
-                                                            (double)INT_MAX ||
+            if ((INT64_OR_DOUBLE)(repeat_min - 1)*
+                  (INT64_OR_DOUBLE)length_prevgroup > 
+                    (INT64_OR_DOUBLE)INT_MAX ||
                 OFLOW_MAX - *lengthptr < delta)
               {
               *errorcodeptr = ERR20;
@@ -4265,15 +4269,16 @@ we set the flag only if there is a literal "\r" or "\n" in the class. */
         just adjust the length as if we had. For each repetition we must add 1
         to the length for BRAZERO and for all but the last repetition we must
         add 2 + 2*LINKSIZE to allow for the nesting that occurs. Do some
-        paranoid checks to avoid integer overflow. */
+        paranoid checks to avoid integer overflow. The INT64_OR_DOUBLE type is 
+        a 64-bit integer type when available, otherwise double. */
 
         if (lengthptr != NULL && repeat_max > 0)
           {
           int delta = repeat_max * (length_prevgroup + 1 + 2 + 2*LINK_SIZE) -
                       2 - 2*LINK_SIZE;   /* Last one doesn't nest */
-          if ((double)repeat_max *
-                (double)(length_prevgroup + 1 + 2 + 2*LINK_SIZE)
-                  > (double)INT_MAX ||
+          if ((INT64_OR_DOUBLE)repeat_max *
+                (INT64_OR_DOUBLE)(length_prevgroup + 1 + 2 + 2*LINK_SIZE)
+                  > (INT64_OR_DOUBLE)INT_MAX ||
               OFLOW_MAX - *lengthptr < delta)
             {
             *errorcodeptr = ERR20;
@@ -5787,7 +5792,7 @@ int old_external_options = cd->external_options;
 branch_chain bc;
 
 bc.outer = bcptr;
-bc.current = code;
+bc.current_branch = code;
 
 firstbyte = reqbyte = REQ_UNSET;
 
@@ -6028,7 +6033,7 @@ for (;;)
     {
     *code = OP_ALT;
     PUT(code, 1, code - last_branch);
-    bc.current = last_branch = code;
+    bc.current_branch = last_branch = code;
     code += 1 + LINK_SIZE;
     }
 
@@ -6641,7 +6646,7 @@ if debugging, leave the test till after things are printed out. */
 
 *code++ = OP_END;
 
-#ifndef DEBUG
+#ifndef PCRE_DEBUG
 if (code - codestart > length) errorcode = ERR23;
 #endif
 
@@ -6765,7 +6770,7 @@ if (reqbyte >= 0 &&
 /* Print out the compiled data if debugging is enabled. This is never the
 case when building a production library. */
 
-#ifdef DEBUG
+#ifdef PCRE_DEBUG
 
 printf("Length = %d top_bracket = %d top_backref = %d\n",
   length, re->top_bracket, re->top_backref);
@@ -6803,7 +6808,7 @@ if (code - codestart > length)
   if (errorcodeptr != NULL) *errorcodeptr = ERR23;
   return NULL;
   }
-#endif   /* DEBUG */
+#endif   /* PCRE_DEBUG */
 
 return (pcre *)re;
 }
diff --git a/pcre_dfa_exec.c b/pcre_dfa_exec.c
index 5c1875a..11a3fba 100644
--- a/pcre_dfa_exec.c
+++ b/pcre_dfa_exec.c
@@ -255,7 +255,7 @@ typedef struct stateblock {
 #define INTS_PER_STATEBLOCK  (sizeof(stateblock)/sizeof(int))
 
 
-#ifdef DEBUG
+#ifdef PCRE_DEBUG
 /*************************************************
 *             Print character string             *
 *************************************************/
@@ -559,7 +559,7 @@ for (;;)
   workspace[0] ^= 1;              /* Remember for the restarting feature */
   workspace[1] = active_count;
 
-#ifdef DEBUG
+#ifdef PCRE_DEBUG
   printf("%.*sNext character: rest of subject = \"", rlevel*2-2, SP);
   pchars((uschar *)ptr, strlen((char *)ptr), stdout);
   printf("\"\n");
@@ -605,7 +605,7 @@ for (;;)
     int state_offset = current_state->offset;
     int count, codevalue, rrc;
 
-#ifdef DEBUG
+#ifdef PCRE_DEBUG
     printf ("%.*sProcessing state %d c=", rlevel*2-2, SP, state_offset);
     if (clen == 0) printf("EOL\n");
       else if (c > 32 && c < 127) printf("'%c'\n", c);
diff --git a/pcre_exec.c b/pcre_exec.c
index fd6866b..899b92a 100644
--- a/pcre_exec.c
+++ b/pcre_exec.c
@@ -89,7 +89,7 @@ static const char rep_max[] = { 0, 0, 0, 0, 1, 1 };
 
 
 
-#ifdef DEBUG
+#ifdef PCRE_DEBUG
 /*************************************************
 *        Debugging function to print chars       *
 *************************************************/
@@ -141,7 +141,7 @@ match_ref(int offset, register USPTR eptr, int length, match_data *md,
 {
 USPTR p = md->start_subject + md->offset_vector[offset];
 
-#ifdef DEBUG
+#ifdef PCRE_DEBUG
 if (eptr >= md->end_subject)
   printf("matching subject <null>");
 else
@@ -254,7 +254,7 @@ actuall used in this definition. */
 #ifndef NO_RECURSE
 #define REGISTER register
 
-#ifdef DEBUG
+#ifdef PCRE_DEBUG
 #define RMATCH(ra,rb,rc,rd,re,rf,rg,rw) \
   { \
   printf("match() called in line %d\n", __LINE__); \
@@ -622,7 +622,7 @@ TAIL_RECURSE:
 /* OK, now we can get on with the real code of the function. Recursive calls
 are specified by the macro RMATCH and RRETURN is used to return. When
 NO_RECURSE is *not* defined, these just turn into a recursive call to match()
-and a "return", respectively (possibly with some debugging if DEBUG is
+and a "return", respectively (possibly with some debugging if PCRE_DEBUG is
 defined). However, RMATCH isn't like a function call because it's quite a
 complicated macro. It has to be used in one particular way. This shouldn't,
 however, impact performance when true recursion is being used. */
@@ -713,7 +713,7 @@ for (;;)
     number = GET2(ecode, 1+LINK_SIZE);
     offset = number << 1;
 
-#ifdef DEBUG
+#ifdef PCRE_DEBUG
     printf("start bracket %d\n", number);
     printf("subject=");
     pchars(eptr, 16, TRUE, md);
@@ -1039,7 +1039,7 @@ for (;;)
     number = GET2(ecode, 1);
     offset = number << 1;
 
-#ifdef DEBUG
+#ifdef PCRE_DEBUG
       printf("end bracket %d at *ACCEPT", number);
       printf("\n");
 #endif
@@ -1468,7 +1468,7 @@ for (;;)
       number = GET2(prev, 1+LINK_SIZE);
       offset = number << 1;
 
-#ifdef DEBUG
+#ifdef PCRE_DEBUG
       printf("end bracket %d", number);
       printf("\n");
 #endif
@@ -5635,7 +5635,7 @@ for(;;)
       }
     }
 
-#ifdef DEBUG  /* Sigh. Some compilers never learn. */
+#ifdef PCRE_DEBUG  /* Sigh. Some compilers never learn. */
   printf(">>>> Match against: ");
   pchars(start_match, end_subject - start_match, TRUE, md);
   printf("\n");
diff --git a/pcre_internal.h b/pcre_internal.h
index de09614..28cbab2 100644
--- a/pcre_internal.h
+++ b/pcre_internal.h
@@ -7,7 +7,7 @@
 and semantics are as close as possible to those of the Perl 5 language.
 
                        Written by Philip Hazel
-           Copyright (c) 1997-2009 University of Cambridge
+           Copyright (c) 1997-2010 University of Cambridge
 
 -----------------------------------------------------------------------------
 Redistribution and use in source and binary forms, with or without
@@ -45,10 +45,10 @@ functions whose names all begin with "_pcre_". */
 #ifndef PCRE_INTERNAL_H
 #define PCRE_INTERNAL_H
 
-/* Define DEBUG to get debugging output on stdout. */
+/* Define PCRE_DEBUG to get debugging output on stdout. */
 
 #if 0
-#define DEBUG
+#define PCRE_DEBUG
 #endif
 
 /* We do not support both EBCDIC and UTF-8 at the same time. The "configure"
@@ -74,7 +74,7 @@ It turns out that the Mac Debugging.h header also defines the macro DPRINTF, so
 be absolutely sure we get our version. */
 
 #undef DPRINTF
-#ifdef DEBUG
+#ifdef PCRE_DEBUG
 #define DPRINTF(p) printf p
 #else
 #define DPRINTF(p) /* Nothing */
@@ -86,8 +86,6 @@ setjmp and stdarg are used is when NO_RECURSE is set. */
 
 #include <ctype.h>
 #include <limits.h>
-#include <setjmp.h>
-#include <stdarg.h>
 #include <stddef.h>
 #include <stdio.h>
 #include <stdlib.h>
@@ -186,6 +184,23 @@ preprocessor time in standard C environments. */
   #error Cannot determine a type for 32-bit unsigned integers
 #endif
 
+/* When checking for integer overflow in pcre_compile(), we need to handle 
+large integers. If a 64-bit integer type is available, we can use that. 
+Otherwise we have to cast to double, which of course requires floating point 
+arithmetic. Handle this by defining a macro for the appropriate type. If 
+stdint.h is available, include it; it may define INT64_MAX. The macro int64_t 
+may be set by "configure". */
+
+#if HAVE_STDINT_H
+#include <stdint.h>
+#endif
+
+#if defined INT64_MAX || defined int64_t
+#define INT64_OR_DOUBLE int64_t
+#else
+#define INT64_OR_DOUBLE double
+#endif
+
 /* All character handling must be done as unsigned characters. Otherwise there
 are problems with top-bit-set characters and functions such as isspace().
 However, we leave the interface to the outside world as char *, because that
@@ -1579,7 +1594,7 @@ branches, for testing for left recursion. */
 
 typedef struct branch_chain {
   struct branch_chain *outer;
-  uschar *current;
+  uschar *current_branch;
 } branch_chain;
 
 /* Structure for items in a linked list that represents an explicit recursive
diff --git a/pcre_printint.src b/pcre_printint.src
index acfc4ca..9b2def1 100644
--- a/pcre_printint.src
+++ b/pcre_printint.src
@@ -6,7 +6,7 @@
 and semantics are as close as possible to those of the Perl 5 language.
 
                        Written by Philip Hazel
-           Copyright (c) 1997-2009 University of Cambridge
+           Copyright (c) 1997-2010 University of Cambridge
 
 -----------------------------------------------------------------------------
 Redistribution and use in source and binary forms, with or without
@@ -43,7 +43,8 @@ internal form of a compiled regular expression, along with some supporting
 local functions. This source file is used in two places:
 
 (1) It is #included by pcre_compile.c when it is compiled in debugging mode
-(DEBUG defined in pcre_internal.h). It is not included in production compiles.
+(PCRE_DEBUG defined in pcre_internal.h). It is not included in production
+compiles.
 
 (2) It is always #included by pcretest.c, which can be asked to print out a
 compiled regex for debugging purposes. */
diff --git a/pcre_study.c b/pcre_study.c
index 957a1fe..d937b95 100644
--- a/pcre_study.c
+++ b/pcre_study.c
@@ -6,7 +6,7 @@
 and semantics are as close as possible to those of the Perl 5 language.
 
                        Written by Philip Hazel
-           Copyright (c) 1997-2009 University of Cambridge
+           Copyright (c) 1997-2010 University of Cambridge
 
 -----------------------------------------------------------------------------
 Redistribution and use in source and binary forms, with or without
@@ -444,7 +444,8 @@ Returns:        nothing
 */
 
 static void
-set_bit(uschar *start_bits, unsigned int c, BOOL caseless, compile_data *cd)
+set_table_bit(uschar *start_bits, unsigned int c, BOOL caseless, 
+  compile_data *cd)
 {
 start_bits[c/8] |= (1 << (c&7));
 if (caseless && (cd->ctypes[c] & ctype_letter) != 0)
@@ -606,7 +607,7 @@ do
       case OP_QUERY:
       case OP_MINQUERY:
       case OP_POSQUERY:
-      set_bit(start_bits, tcode[1], caseless, cd);
+      set_table_bit(start_bits, tcode[1], caseless, cd);
       tcode += 2;
 #ifdef SUPPORT_UTF8
       if (utf8 && tcode[-1] >= 0xc0)
@@ -619,7 +620,7 @@ do
       case OP_UPTO:
       case OP_MINUPTO:
       case OP_POSUPTO:
-      set_bit(start_bits, tcode[3], caseless, cd);
+      set_table_bit(start_bits, tcode[3], caseless, cd);
       tcode += 4;
 #ifdef SUPPORT_UTF8
       if (utf8 && tcode[-1] >= 0xc0)
@@ -637,7 +638,7 @@ do
       case OP_PLUS:
       case OP_MINPLUS:
       case OP_POSPLUS:
-      set_bit(start_bits, tcode[1], caseless, cd);
+      set_table_bit(start_bits, tcode[1], caseless, cd);
       try_next = FALSE;
       break;