summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* POD spelling patchesDaniel S. Lewart1996-09-111-3/+3
|
* perl 5.003_06: pod/perl.podPerl 5 Porters1996-10-071-3/+7
| | | | | | | | | | | | | | | | | | | Date: Fri, 20 Sep 1996 15:08:33 +0100 (BST) From: "Joseph S. Myers" <jsm28@hermes.cam.ac.uk> Subject: Pod typos, pod2man bugs, and miscellaneous installation comments Here is a patch for various typos and other defects in the Perl 5.003_05 pods, including the pods embedded in library modules. Changed Larry's address to larry@wall.org. Date: Mon, 7 Oct 1996 22:03:00 +0300 From: Jarkko Hietaniemi <jhi@alpha.hut.fi> Subject: LC_COLLATE. Big patch to add, document, and test LC_COLLATE support. perli18n advertised.
* LC_COLLATE.Jarkko Hietaniemi1996-10-071-0/+8
| | | | | | Big patch to add, document, and test LC_COLLATE support. perli18n.pod (and perlapio.pod, btw) added.
* perl 5.003_06: perlsdio.hPerl 5 Porters1996-10-031-1/+13
| | | | | | | | | | | | Date: Fri, 13 Sep 1996 17:24:01 -0400 From: John Stoffel <jfs@jfs.fluent.com> Subject: Re: 5.003_05 is available. Undef Irix getc_unlocked and putc_unlocked #defines. Date: Thu, 03 Oct 1996 16:31:46 -0400 (EDT) From: Charles Bailey <bailey@HMIVAX.HUMGEN.UPENN.EDU> Subject: VMS patches to 5.003_05
* Re: 5.003_05 is available.Hunter Kelly1996-09-121-4/+3
| | | | Fix PerlIO_reopen parameters.
* Add new function perl_init_fold. (I'm not sure it goes here.)Perl 5 Porters1996-10-071-1/+2
|
* perl 5.003_06: perl.hPerl 5 Porters1996-09-301-71/+88
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Date: Sat, 21 Sep 1996 21:33:15 -0400 (EDT) From: Kenneth Albanowski <kjahds@kjahds.com> Subject: Full LONG_MAX & co. patch over 5.003_05 This patch contains the changes I've collected for the various _MAX issues since 5.003_05. No patches issued between 5.003_05 and this one should be applied, use this one instead. The effect is to remove the CHAR_* and I8_* constants (which are ambiguous) and to explicitly cast all of the constants. Date: Mon, 30 Sep 1996 01:13:28 -0400 From: Spider Boardman <spider@Orb.Nashua.NH.US> Subject: Re: pre extending hash? - need speed The patch below (which is relative to perl5.001l) implements "keys %hash = 50_000;" (or other integer-evaluable sizes) for pre-sizing hashes. I've only moved the patch forward from when I first did it. I'm sure the code in hv_ksplit could be improved. Revert from Perl_malloc to Mymalloc for binary compatibility with 5.003. Date: Mon, 7 Oct 1996 22:03:00 +0300 From: Jarkko Hietaniemi <jhi@alpha.hut.fi> Subject: LC_COLLATE. Big patch to add, document, and test LC_COLLATE support.
* perl 5.003_06: perl.cPerl 5 Porters1996-10-011-1/+19
| | | | | | | | | | | | | | | | | | From: Roderick Schertler <roderick@gate.net> Subject: Re: -T flag and removal of `.' from @INC support C<perl -e'attached code'> Date: Tue, 01 Oct 1996 19:02:17 -0400 From: Gurusamy Sarathy <gsar@engin.umich.edu> Subject: Re: 2 core dumps (patch) Message-Id: <199610012302.TAA08395@aatma.engin.umich.edu> The problem is an uninitialized SV slot in errgv. Here's a patch. Date: Thu, 03 Oct 1996 16:31:46 -0400 (EDT) From: Charles Bailey <bailey@HMIVAX.HUMGEN.UPENN.EDU> Subject: VMS patches to 5.003_05
* Perl 5.003 bug when embedding in C++ programJonathan Biggar1996-09-162-4/+4
| | | | | | The following patch is necessary in order to embed the Perl5.003 interpreter into a C++ program without getting prototype mismatch errors from the C++ compiler.
* Re: pre extending hash? - need speedSpider Boardman1996-09-301-2/+7
| | | | | | | | | | | Restore the 5.003 gv_fullname() and gv_efullname() functions. Provide new 3-arg forms gv_fullname3() and gv_efullname3(). The patch below (which is relative to perl5.001l) implements "keys %hash = 50_000;" (or other integer-evaluable sizes) for pre-sizing hashes. I've only moved the patch forward from when I first did it. I'm sure the code in hv_ksplit could be improved.
* perl 5.003_06: mg.cPerl 5 Porters1996-09-301-10/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Restore the 5.003 gv_fullname() and gv_efullname() functions. Provide new 3-arg forms gv_fullname3() and gv_efullname3(). Date: Sun, 29 Sep 1996 22:18:19 -0400 (EDT) From: Chip Salzenberg <salzench@nielsenmedia.com> Subject: 5.003_05: Fix numeric value of $! This patch undoes a bit of over-zealous integerization in mg.c, related to the numeric value of $!. Date: Mon, 30 Sep 1996 01:13:28 -0400 From: Spider Boardman <spider@Orb.Nashua.NH.US> Subject: Re: pre extending hash? - need speed The patch below (which is relative to perl5.001l) implements "keys %hash = 50_000;" (or other integer-evaluable sizes) for pre-sizing hashes. I've only moved the patch forward from when I first did it. I'm sure the code in hv_ksplit could be improved. Date: Thu, 03 Oct 1996 16:31:46 -0400 (EDT) From: Charles Bailey <bailey@HMIVAX.HUMGEN.UPENN.EDU> Subject: VMS patches to 5.003_05 Date: Fri, 4 Oct 1996 12:38:31 -0400 (EDT) From: Chip Salzenberg <salzench@nielsenmedia.com> Subject: 5.003_05: Fix numeric $! and $^E This patch undoes a bit of over-zealous integerization in mg.c, related to the numeric values of $! and $^E. This patch *REPLACES* the one I posted earlier, which was only effective for $!. [Some of this is superceded by similar stuff in the VMS patches.]
* Not all sbrks return zeroed memory.Perl 5 Porters1996-10-031-3/+1
|
* perl 5.003_06: makedepend.SHPerl 5 Porters1996-10-031-0/+6
| | | | | | | Add explicit $touch $firstmakefile for QNX which apparently preserves modification times for a 'cp' command. I worry, though, that touch might not be portable to OS/2. If it is, then I'll remove the fancy case statement.
* Re: dereferencing a hash from the debugger won't workIlya Zakharevich1996-09-301-3/+3
|
* Add in stub file.Perl 5 Porters1996-10-031-0/+81
|
* Text::Abbrev (Re: More standard library test scripts)Ulrich Pfeifer1996-09-231-10/+37
| | | | | This patch merges the Text::Abbrev related patches/tests from Gisle and my previous patch (i.e. replaces both).
* Patch for Term::CapUlrich Pfeifer1996-09-231-1/+5
| | | | | | | | 'use Term::Cap' produces a warning when diagnosics are active. The patch below avoids the warning. [The $entry .= $_ usage is idiomatic enough that it ought to be ok, I would think, but the patch certainly is ok too.]
* perl 5.003_06: lib/Symbol.pmPerl 5 Porters1996-09-201-17/+2
| | | | | | | | | | | | | | | | | | | Date: Fri, 20 Sep 1996 12:38:14 +0200 From: Gisle Aas <aas@bergen.sn.no> Subject: Symbol.pm clobbers $_ at startup perl -le 'BEGIN {$_="foo";} use Symbol; print qualify($_)' I don't understand why the module want to initialize %global from <DATA> in the first place. Perhaps we want to apply this patch instead. Date: Fri, 20 Sep 1996 15:08:33 +0100 (BST) From: "Joseph S. Myers" <jsm28@hermes.cam.ac.uk> Subject: Pod typos, pod2man bugs, and miscellaneous installation comments Here is a patch for various typos and other defects in the Perl 5.003_05 pods, including the pods embedded in library modules.
* look() in Search::Dict should use lc() istead of tr/A-Z/a-z/Gisle Aas1996-09-211-3/+3
| | | | | | The Search::Dict look() function should use the lc() function instead of tr/A-Z/a-z/. This will make folding of non-english letters work if the locale is set up correctly.
* perl 5.003_06: lib/Math/Complex.pmRaphael Manfredi1996-10-031-117/+1047
| | | | | | | | | Date: Thu, 03 Oct 96 18:38:08 +0200 From: Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> # Complex numbers and associated mathematical functions # -- Raphael Manfredi, Sept 1996 # New version. Should be backwards compatible, but please # check it out if you use it.
* LC_COLLATE.Jarkko Hietaniemi1996-10-072-14/+34
| | | | | | Big patch to add, document, and test LC_COLLATE support. ord() is a dangerous thing.
* LC_COLLATE.Jarkko Hietaniemi1996-10-071-2/+37
| | | | | | Big patch to add, document, and test LC_COLLATE support. deprecated and trapped (will whine if called and tell to migrate away)
* File::Find assumes $_ remains unchanged; bugMichael De La Rue1996-09-071-0/+3
| | | | | | | | The File::Find perl module assumes that the $_ variable remains unchanged through the user defined function which is callbacked from find. It carries out a stat operation Simplest fix is merely to document this
* perl 5.003_06: lib/File/Basename.pmPerl 5 Porters1996-09-201-7/+7
| | | | | | | | | | | | | | | | | | | | | | | Date: Fri, 20 Sep 1996 14:11:05 +0200 From: Gisle Aas <aas@bergen.sn.no> Subject: File::BaseName: "/" is legal path separator for MSDOS The File::BaseName module should allow "/" as path separator when fileparse_set_fstype("MSDOS") is in effect: Date: Fri, 20 Sep 1996 13:58:52 +0200 From: Gisle Aas <aas@bergen.sn.no> Subject: File::Basename documentation patch Date: Mon, 30 Sep 1996 00:54:37 -0400 From: Spider Boardman <spider@Orb.Nashua.NH.US> For t/lib/basename.t, though, the associated patch for File::Basename was also wrong: Date: Thu, 03 Oct 1996 16:31:46 -0400 (EDT) From: Charles Bailey <bailey@HMIVAX.HUMGEN.UPENN.EDU> Subject: VMS patches to 5.003_05
* Change a reference from perlapi(1) to perlxs(1).Perl 5 Porters1996-09-301-1/+1
|
* perl 5.003_06: lib/ExtUtils/MM_VMS.pm lib/ExtUtils/Manifest.pm ↵Perl 5 Porters1996-10-033-25/+53
| | | | | | | | | | | | | | | lib/Test/Harness.pm Date: Fri, 20 Sep 1996 15:08:33 +0100 (BST) From: "Joseph S. Myers" <jsm28@hermes.cam.ac.uk> Subject: Pod typos, pod2man bugs, and miscellaneous installation comments Here is a patch for various typos and other defects in the Perl 5.003_05 pods, including the pods embedded in library modules. Date: Thu, 03 Oct 1996 16:31:46 -0400 (EDT) From: Charles Bailey <bailey@HMIVAX.HUMGEN.UPENN.EDU> Subject: VMS patches to 5.003_05
* Pod typos, pod2man bugs, and miscellaneous installation commentsJoseph S. Myers1996-09-201-16/+16
| | | | | | | Remove unwantd space after the I in -I$Config[archlib} Here is a patch for various typos and other defects in the Perl 5.003_05 pods, including the pods embedded in library modules.
* perl 5.003_06: lib/Benchmark.pmPerl 5 Porters1996-09-201-4/+4
| | | | | | | | | | | | | Date: Sat, 28 Sep 1996 17:01:22 +0300 (EET DST) From: Jarkko Hietaniemi <jhi@cc.hut.fi> Subject: a really really tiny typo Date: Fri, 20 Sep 1996 15:08:33 +0100 (BST) From: "Joseph S. Myers" <jsm28@hermes.cam.ac.uk> Subject: Pod typos, pod2man bugs, and miscellaneous installation comments Here is a patch for various typos and other defects in the Perl 5.003_05 pods, including the pods embedded in library modules.
* perl 5.003_06: lib/AutoLoader.pmPerl 5 Porters1996-09-301-17/+50
| | | | | | | | | | | | | | | | | | | | | | | | | | | Date: Mon Sep 9 09:29:44 1996 From: Gisle Aas <aas@bergen.sn.no> Subject: Re: problem with 'die' and UserAgent > This is a patch to the AutoLoader.pm (from 5.003) that fixes the problem: This is a better patch (no need to test for /::DESTROY$/ twice): Date: Mon, 30 Sep 1996 00:54:37 -0400 From: Spider Boardman <spider@Orb.Nashua.NH.US> The test and patches for AutoLoader were also non-functional, since the regexp context (curpm) was still being clobbered by the filename manipulations: Date: Sun, 06 Oct 1996 16:15:07 +0200 From: Gisle Aas <aas@bergen.sn.no> Subject: Re: Can't locate auto/U/autosplit.ix It would IMHO be much better if the AutoLoader exported the AUTOLOAD() function. With an exported AUTOLOAD() we would not have to inherit from AutoLoader, and we would avoid these problems. This patch tries to explain the behavior of AutoLoader instead by updating its documentation.
* Re: "Attempt to free non-existent shared string"? (with patch)Gurusamy Sarathy1996-09-201-4/+6
| | | | | | | | | | | | I found a subtle problem with the lazydelete mechanism (which is used to postpone the delete of a entry that may be getting iterated over). I was using the HeKLEN slot to hold the hint, but the real HeKLEN is needed later to call unsharepvn(). This means that only magical hash entries can use the HeKLEN slot to hold flags. Here's a tested patch against 5.00305 that fixes the problem. The patch simply moves the LAZYDEL hint to become a SV-level private flag.
* perl 5.003_06: hv.cPerl 5 Porters1996-09-301-4/+86
| | | | | | | | | | | | | | | | | | | | | | | | | | Date: Fri, 20 Sep 1996 15:38:57 -0400 From: Gurusamy Sarathy <gsar@engin.umich.edu> Subject: Re: "Attempt to free non-existent shared string"? (with patch) I found a subtle problem with the lazydelete mechanism (which is used to postpone the delete of a entry that may be getting iterated over). I was using the HeKLEN slot to hold the hint, but the real HeKLEN is needed later to call unsharepvn(). This means that only magical hash entries can use the HeKLEN slot to hold flags. Here's a tested patch against 5.00305 that fixes the problem. The patch simply moves the LAZYDEL hint to become a SV-level private flag. Date: Mon, 30 Sep 1996 01:13:28 -0400 From: Spider Boardman <spider@Orb.Nashua.NH.US> Subject: Re: pre extending hash? - need speed The patch below (which is relative to perl5.001l) implements "keys %hash = 50_000;" (or other integer-evaluable sizes) for pre-sizing hashes. I've only moved the patch forward from when I first did it. I'm sure the code in hv_ksplit could be improved.
* Dale's posting as patch (Was: Perl 5.003_5 make fails on NS3.2 - CURED)Andreas Koenig1996-09-281-3/+9
| | | | | | | Replace optimize="-g" by optimize="" since we're just trying to turn off the optimizier. Handle NeXT, POSIX, and setpgid in pp_sys.c and POSIX.
* Add notes about MachTen 4.0.3 SYSV IPC.Perl 5 Porters1996-09-301-0/+13
|
* Full LONG_MAX & co. patch over 5.003_05Kenneth Albanowski1996-09-211-2/+2
| | | | | | | | | This patch contains the changes I've collected for the various _MAX issues since 5.003_05. No patches issued between 5.003_05 and this one should be applied, use this one instead. The effect is to remove the CHAR_* and I8_* constants (which are ambiguous) and to explicitly cast all of the constants.
* perl 5.003_06: global.symPerl 5 Porters1996-10-071-0/+7
| | | | | | | | | | | | | | | | | | | | | | | Restore the 5.003 gv_fullname() and gv_efullname() functions. Provide new 3-arg forms gv_fullname3() and gv_efullname3(). Date: Mon, 30 Sep 1996 01:13:28 -0400 From: Spider Boardman <spider@Orb.Nashua.NH.US> Subject: Re: pre extending hash? - need speed The patch below (which is relative to perl5.001l) implements "keys %hash = 50_000;" (or other integer-evaluable sizes) for pre-sizing hashes. I've only moved the patch forward from when I first did it. I'm sure the code in hv_ksplit could be improved. Date: Mon, 7 Oct 1996 22:03:00 +0300 From: Jarkko Hietaniemi <jhi@alpha.hut.fi> Subject: LC_COLLATE. Big patch to add, document, and test LC_COLLATE support. added var lc_collate_active and func mem_collxfrm.
* perl 5.003_06: ext/SDBM_File/sdbm/sdbm.hPerl 5 Porters1996-10-081-4/+4
| | | | | Revert from Perl_malloc to Mymalloc for binary compatibility with 5.003.
* Dale's posting as patch (Was: Perl 5.003_5 make fails on NS3.2 - CURED)Andreas Koenig1996-09-281-0/+5
| | | | Handle NeXT, POSIX, and setpgid in pp_sys.c and POSIX.
* LC_COLLATE.Jarkko Hietaniemi1996-10-071-1/+8
| | | | | | | Big patch to add, document, and test LC_COLLATE support. setlocale() allowed one argument only, call to perl_init_fold() (in util.c) if setlocale() succeeded.
* LC_COLLATE.Jarkko Hietaniemi1996-10-071-1/+21
| | | | | | Big patch to add, document, and test LC_COLLATE support. enhanced setlocale() docs and introduced the one-argument variant doc.
* perl 5.003_06: ext/Opcode/Opcode.pmPerl 5 Porters1996-09-201-1/+2
| | | | | | | | | | | | | | | Date: Fri, 20 Sep 1996 12:59:21 +0200 From: Gisle Aas <aas@bergen.sn.no> Subject: Re: Symbol.pm clobbers $_ at startup The same kind of problem seem to be present in Opcode.pm: Date: Fri, 20 Sep 1996 15:08:33 +0100 (BST) From: "Joseph S. Myers" <jsm28@hermes.cam.ac.uk> Subject: Pod typos, pod2man bugs, and miscellaneous installation comments Here is a patch for various typos and other defects in the Perl 5.003_05 pods, including the pods embedded in library modules.
* perl 5.003_06: dev/nullPerl 5 Porters1996-10-011-0/+3
| | | | | | | | Perl 5.003_05 compiles on DYNIX/ptx 4.0 (v4.1.3), and passes all tests. The only change needed is in "ext/NDBM_File/Makefile.PL" - on this system, ndbm is actually contained in the libc library, and must be linked against -lc when compiling. (this is for dynamic ELF executables, I didn't compile statically)
* Pod typos, pod2man bugs, and miscellaneous installation commentsJoseph S. Myers1996-09-206-141/+364
| | | | | | | Here is a patch for various typos and other defects in the Perl 5.003_05 pods, including the pods embedded in library modules. Updated to IO-1.12.
* Updated to IO-1.12.Perl 5 Porters1996-10-062-13/+64
|
* perl 5.003_06: dump.c embed.h gv.c run.cPerl 5 Porters1996-10-064-9/+32
| | | | | Restore the 5.003 gv_fullname() and gv_efullname() functions. Provide new 3-arg forms gv_fullname3() and gv_efullname3().
* Re: pre extending hash? - need speedSpider Boardman1996-09-301-1/+19
| | | | | | | | The patch below (which is relative to perl5.001l) implements "keys %hash = 50_000;" (or other integer-evaluable sizes) for pre-sizing hashes. I've only moved the patch forward from when I first did it. I'm sure the code in hv_ksplit could be improved.
* VMS patches to 5.003_05Charles Bailey1996-10-036-57/+128
|
* perl 5.003_06: configureSven Verdoolaege1996-09-301-0/+6
| | | | | | | | Date: Mon, 30 Sep 1996 14:01:05 +0100 From: Sven Verdoolaege <skimo@breughel.ufsia.ac.be> Warn the user of case-insensitive file systems that they may have accidentally gotten 'configure' instead of 'Configure'.
* Pod typos, pod2man bugs, and miscellaneous installation commentsJoseph S. Myers1996-09-2043-240/+246
| | | | | Here is a patch for various typos and other defects in the Perl 5.003_05 pods, including the pods embedded in library modules.
* Changed Larry's address to larry@wall.org.Perl 5 Porters1996-10-071-1/+1
|
* perl 5.003_06: MANIFESTPerl 5 Porters1996-09-281-2/+26
| | | | | | | | | | | | | | | | | | | | | | | | Date: Sun, 22 Sep 1996 00:59:56 +0200 From: Gisle Aas <aas@aas.no> Subject: More standard library test scripts This is a collection of test scripts for the standard library modules. Some of the tests does not pass unless some of the patches I have sent out are applied. Date: Sat, 28 Sep 1996 15:11:06 +0200 From: Andreas Koenig <k@anna.in-berlin.de> Subject: Dale's posting as patch (Was: Perl 5.003_5 make fails on NS3.2 - CURED) Handle NeXT, POSIX, and setpgid in pp_sys.c and POSIX. Date: Mon, 7 Oct 1996 22:03:00 +0300 From: Jarkko Hietaniemi <jhi@alpha.hut.fi> Subject: LC_COLLATE. Big patch to add, document, and test LC_COLLATE support. added perli18n.pod.
View Lorry Controller Status