15 years agoFold win32/buildext.pl into make_ext.pl
Nicholas Clark [Mon, 2 Feb 2009 21:42:33 +0000]
Fold win32/buildext.pl into make_ext.pl

15 years agoMake t/pod/pod2usage2.t work on case insensitive file systems.
Jerry D. Hedden [Mon, 2 Feb 2009 21:25:38 +0000]
Make t/pod/pod2usage2.t work on case insensitive file systems.

pod_where() checks the current dir when looking for PODs
when the -dirs option is not specified.  Because of
case-insensitivity under Cygwin, this causes t/pod/usage.pod
to get found before lib/Pod/Usage.pm.

The fix is to set -dir=>[] in pod_where() when testing in
CORE under Cygwin.  The attached patch does this.

15 years agoext/DB_File needs a ppport.h
Nicholas Clark [Mon, 2 Feb 2009 20:31:27 +0000]
ext/DB_File needs a ppport.h

15 years agoAdd missed file ext/DB_File/config.in and update ext/DB_File/Changes, both
Nicholas Clark [Mon, 2 Feb 2009 20:17:59 +0000]
Add missed file ext/DB_File/config.in and update ext/DB_File/Changes, both
accidentally omitted from d8bf0b8cc0654f404eb597a952c50e72976fd4fa.

15 years agoBump version of PerlIO::via after previous change
Rafael Garcia-Suarez [Mon, 2 Feb 2009 16:01:10 +0000]
Bump version of PerlIO::via after previous change

15 years agoWhen loading a PerlIO::via layer, first look in the PerlIO::via namespace
Rafael Garcia-Suarez [Mon, 2 Feb 2009 14:56:13 +0000]
When loading a PerlIO::via layer, first look in the PerlIO::via namespace

Previously an open "via(Foo)" was first looking for the Foo namespace,
then for PerlIO::via::Foo. That could cause loading problems if both Foo
and PerlIO::via::Foo namespaces existed, and when the latter was
required. The workaround of loading "via(PerlIO::via::Foo)" being rather
ugly, it's preferable to look first for the fully qualified namespace
in order to avoid side-effects depending of what namespaces are
instanciated in the interpreter.

15 years agoRestore broken clean target in win32/makefile.mk
Steve Hay [Mon, 2 Feb 2009 13:18:34 +0000]
Restore broken clean target in win32/makefile.mk

Commit aaaf44d72fce5980564533b9072da6e594d91d81 accidentally missed a --all
argument for buildext.pl, which resulted in the extensions not getting
"dmake realclean" run in them.

15 years agoUpdate CPAN.pm to 1.93_03
Andreas J. Koenig [Mon, 2 Feb 2009 09:11:30 +0000]
Update CPAN.pm to 1.93_03

Message-ID: <87vdrumeds.fsf@k75.linux.bogus>

15 years agoUpgrade to Encode-2.29.
Steve Hay [Mon, 2 Feb 2009 08:59:11 +0000]
Upgrade to Encode-2.29.

15 years agoUpgrade DB_File to 1.818.
Steve Hay [Mon, 2 Feb 2009 08:54:00 +0000]
Upgrade DB_File to 1.818.

15 years agoPut parentheses around macro arguments
Rafael Garcia-Suarez [Sun, 1 Feb 2009 21:52:13 +0000]
Put parentheses around macro arguments

15 years agoCast result to character size before array indexed
Karl [Sat, 31 Jan 2009 19:03:42 +0000]
Cast result to character size before array indexed

15 years agovms stat patches (take 2)
John Malmberg [Sun, 1 Feb 2009 02:36:55 +0000]
vms stat patches (take 2)

Conversion of Perl_flex_stat_int() to tolerate a null thread context,
and updates to better handle Unix Format or Extended File System
character sets.

vms_fid_to_name() now understands stat/lstat for properly handling
symbolic links, and uses a similar algorithm as Perl_flex_stat_int() for
consistency.  Will now optionally return the mode bits from the stat
operation.

(The parts of the patch that worked around null thread context when
checking vmsish hints are omitted since that's been handled somewhat
differently by a520a2c43dd549df0ebed1f073e13cc4564380db).

Message-id: <49850AC7.1020906@gmail.com>

15 years agoFor the vmsish pragma, make sure there is a thread context if needed.
Craig A. Berry [Sun, 1 Feb 2009 21:09:25 +0000]
For the vmsish pragma, make sure there is a thread context if needed.

The hints are interpreter-specific, and we check them in some
pretty low-level routines, such as time(), gmtime(), and stat(),
where it's possible thread context may not have been initialized
yet (or may have already been torn down).

15 years agoMake Time/HiRes/Makefile.PL produce a Makefile if one is needed
Max Maischein [Sat, 31 Jan 2009 13:50:34 +0000]
Make Time/HiRes/Makefile.PL produce a Makefile if one is needed

15 years agoReplacing system $scalar with system @list requires splitting $MAKE on spaces.
Nicholas Clark [Sun, 1 Feb 2009 14:37:21 +0000]
Replacing system $scalar with system @list requires splitting $MAKE on spaces.

15 years agoLoop over @extspec, calling build_extension() on each.
Nicholas Clark [Sun, 1 Feb 2009 11:38:31 +0000]
Loop over @extspec, calling build_extension() on each.

15 years agoThe $@ was actually a botched conversion from $make $targ, rather than an
Nicholas Clark [Sun, 1 Feb 2009 10:23:54 +0000]
The $@ was actually a botched conversion from $make $targ, rather than an
attempt to output a literal $

15 years agoCorrectly escape $ in debug statement
Rafael Garcia-Suarez [Sun, 1 Feb 2009 09:53:25 +0000]
Correctly escape $ in debug statement

15 years agoUpdate Makefile-cross-SH with 07f3cc2a611daebb9d83f869d36ef63cf4fa2565 and
Nicholas Clark [Sat, 31 Jan 2009 21:34:49 +0000]
Update Makefile-cross-SH with 07f3cc2a611daebb9d83f869d36ef63cf4fa2565 and
61edc68382f612a884a3181266d6220ea3a727eb

15 years agoMove the static/dynamic/nonxs LINKTYPE logic from make_ext.pl to Makefile.SH
Nicholas Clark [Sat, 31 Jan 2009 20:16:43 +0000]
Move the static/dynamic/nonxs LINKTYPE logic from make_ext.pl to Makefile.SH
This simplifies make_ext.pl considerably. Given that the first use of 'passthru'
was in 40000a8c37c35cc00114329ca0add46bca36a0ec in 1995, possibly we should have
done this particular refactoring a little while ago.

15 years agoUse the build_extension() routine from win32/buildext.pl
Nicholas Clark [Sat, 31 Jan 2009 17:16:21 +0000]
Use the build_extension() routine from win32/buildext.pl
Swap from $passthru to @pass_through, and $run to @run.
Use the list form of system everywhere.
Hard code 'Makefile' as it does not vary.
Correct the perlism 'chdir' in the *clean.sh writing code.

15 years agoCreate build_extension() to encapsulate the per-extension logic.
Nicholas Clark [Sat, 31 Jan 2009 17:12:40 +0000]
Create build_extension() to encapsulate the per-extension logic.
Use @pass_through to hold extra VARIABLE=value assignments.
Add @run to hold optional build-time run prefix.
Use the list form of system for everything.
Always call a target - default to 'all' if not is passed in.
Merge the alternate logic for selecting -MCross from make_ext.pl

15 years agoMake dir and target --options to buildext.pl. Update the documentation.
Nicholas Clark [Sat, 31 Jan 2009 13:56:28 +0000]
Make dir and target --options to buildext.pl. Update the documentation.

15 years agoMake lib/Config_heavy.pl part of $(CONFIGPM) as the build ordering is now sane.
Nicholas Clark [Sat, 31 Jan 2009 13:39:04 +0000]
Make lib/Config_heavy.pl part of $(CONFIGPM) as the build ordering is now sane.
Effectively $(CONFIGPM_EXTRA) becomes $(CONFIGPM) and $(CONFIGPM) becomes
$(CONFIGPM_FROM_CONFIG_SH). Mention lib/Config_heavy.pl as something that
is built by running make_patchnum.pl
Hopefully this avoids warnings from Config.pm during the build.

15 years agoThe hack to write out realclean.sh and veryclean.sh is for Unix only.
Nicholas Clark [Sat, 31 Jan 2009 12:20:26 +0000]
The hack to write out realclean.sh and veryclean.sh is for Unix only.

15 years agoMove the target generic logic above the extension specific logic.
Nicholas Clark [Fri, 30 Jan 2009 23:39:15 +0000]
Move the target generic logic above the extension specific logic.
This will make it easier to refactor to converge on the Win32 (and VMS)
equivalents, which set things up, then loop around all appropriate extensions.

15 years agoThe logic to set $makefile to 'Makefile.old' can't work, so remove it.
Nicholas Clark [Fri, 30 Jan 2009 23:12:24 +0000]
The logic to set $makefile to 'Makefile.old' can't work, so remove it.
(Makefile.old is created by running make clean in an extension, and make clean
at the top level removes ./miniperl, so there won't be a miniperl to run
make_ext.pl. The Makefile.old logic was valid for the shell script version.
Remove $makeopts, which was missed from the make command in the transcription
from shell to Perl.

15 years agoEliminate $makeargs by prepending its contents to $passthru.
Nicholas Clark [Fri, 30 Jan 2009 22:52:50 +0000]
Eliminate $makeargs by prepending its contents to $passthru.

15 years agoUse Config; rather than parsing config.sh and pushing it into %ENV, a literal
Nicholas Clark [Fri, 30 Jan 2009 20:45:48 +0000]
Use Config; rather than parsing config.sh and pushing it into %ENV, a literal
transcription of the make_ext shell script. (We'll soon see if anything we
called was using its environment rather than Config.pm)

15 years agoFix paths on new pod2usage2 tests to work in the core.
Nicholas Clark [Fri, 30 Jan 2009 19:20:08 +0000]
Fix paths on new pod2usage2 tests to work in the core.

15 years agoReplace the one file timestamp dependency check with a call to 'make config'
Nicholas Clark [Fri, 30 Jan 2009 17:53:45 +0000]
Replace the one file timestamp dependency check with a call to 'make config'
The timestamp check was from the original checkin of buidlext.pl in 2001
(17af6fb0658e86e2f4a472e3392fc688e0c25b43) by Nick I-S.
The 'make config' approach of make_ext was in the 5.000 release.

15 years agoPod correction for ExtUtils::XSSymSet.
Craig A. Berry [Fri, 30 Jan 2009 16:08:19 +0000]
Pod correction for ExtUtils::XSSymSet.

15 years agoUpgrade to Pod-Parser-1.36.
Steve Hay [Fri, 30 Jan 2009 10:27:25 +0000]
Upgrade to Pod-Parser-1.36.

Three local changes remain in blead:

Blank lines "between" verbatim sections are now acceptible:
http://perl5.git.perl.org/perl.git/commitdiff/caa547d

Be less picky about what constitutes "numeric lists" in Pod:
http://perl5.git.perl.org/perl.git/commitdiff/4df4f5d

Changes made to contains_pod.t when upgrading to 1.34:
http://perl5.git.perl.org/perl.git/commitdiff/fb59f97

15 years agoTweaks to pod/perlrepository.pod
Gisle Aas [Thu, 29 Jan 2009 16:23:50 +0000]
Tweaks to pod/perlrepository.pod

Mention the github mirror early.
Document the perl5.git.perl.org:/perl.git repository url form.
Use 'git commit -a -m ...' throughout.
Mention that 'git commit -a --amend' is very useful.
The easiest way to inspect the last commit is 'git show HEAD'.
Mention 'git checkout -f' as the way to revert all your edits.

15 years agowin32/ext was removed by change 30379 in 2007, so no need to check for it in the
Nicholas Clark [Thu, 29 Jan 2009 20:00:48 +0000]
win32/ext was removed by change 30379 in 2007, so no need to check for it in the
Makefiles.

15 years agoAdd --options with arguments to make_ext.pl. Pass $target as --target=...
Nicholas Clark [Thu, 29 Jan 2009 21:13:01 +0000]
Add --options with arguments to make_ext.pl. Pass $target as --target=...
Treat all arguments that /=/ as passthrough, the rest being extensions.
This will converge the argument passing conventions for make_ext.pl and
win32/buildext.pl

15 years agovmspath without thread context.
John Malmberg [Thu, 29 Jan 2009 04:29:32 +0000]
vmspath without thread context.

This adds a vmspath routine that does not use a thread context.

Message-id: <498130AC.7000102@gmail.com>

15 years agoRevert "do_exec() is a mathom, so call its replacement directly"
Gisle Aas [Thu, 29 Jan 2009 20:00:50 +0000]
Revert "do_exec() is a mathom, so call its replacement directly"

The change breaks the build on Windows as this function is really
only a mathom when PERL_DEFAULT_DO_EXEC3_IMPLEMENTATION is defined.

This reverts commit 06ba78de6df7f363272e855cc63710a634eb6100.

15 years agoUpgrade to Win32-0.39.
Steve Hay [Thu, 29 Jan 2009 18:05:53 +0000]
Upgrade to Win32-0.39.

15 years agoAdd an explicit --all option to buildext.pl, rather than defaulting to it if
Nicholas Clark [Thu, 29 Jan 2009 16:57:34 +0000]
Add an explicit --all option to buildext.pl, rather than defaulting to it if
--static and --dynamic are not present. Make it run under strict.

15 years agodo_exec() is a mathom, so call its replacement directly
Gisle Aas [Thu, 29 Jan 2009 15:21:48 +0000]
do_exec() is a mathom, so call its replacement directly

15 years agoPass $(MAKE) as one quoted commandline argument, to converge with make_ext.pl
Nicholas Clark [Wed, 28 Jan 2009 22:55:26 +0000]
Pass $(MAKE) as one quoted commandline argument, to converge with make_ext.pl

15 years agoMove ext/util/make_ext.pl to the top level.
Nicholas Clark [Wed, 28 Jan 2009 15:10:42 +0000]
Move ext/util/make_ext.pl to the top level.

15 years agoFold ext/util/make_ext_cross into ext/util/make_ext.pl, and delete it.
Nicholas Clark [Wed, 28 Jan 2009 14:33:34 +0000]
Fold ext/util/make_ext_cross into ext/util/make_ext.pl, and delete it.
Add --option argument parsing to make_ext.pl (swiped from buildext.pl).
Add a --cross option to add -MCross to the Makefile.PL line.
Change Cross/Makefile-cross-SH to call "ext/util/make_ext.pl --cross ..."

15 years agoIn the code for printing -v, replace longhand newSVpvn() with newSVpvs(), and
Nicholas Clark [Sun, 25 Jan 2009 16:45:04 +0000]
In the code for printing -v, replace longhand newSVpvn() with newSVpvs(), and
avoid needing a sv_catpvs().

15 years agoIn buildext.pl, refactor the @ARGV parsing into a single loop.
Nicholas Clark [Tue, 27 Jan 2009 21:59:02 +0000]
In buildext.pl, refactor the @ARGV parsing into a single loop.

15 years agoChange to use 5.1 Unicode file versions
Karl [Tue, 27 Jan 2009 02:31:42 +0000]
Change to use 5.1 Unicode file versions

15 years agounicore Makefile didn't clean sub-sub directories
Karl [Mon, 26 Jan 2009 23:43:57 +0000]
unicore Makefile didn't clean sub-sub directories

15 years agoSilence verbosity in lib/File/Copy.t during 'make test' in core
Jerry D. Hedden [Mon, 26 Jan 2009 19:25:32 +0000]
Silence verbosity in lib/File/Copy.t during 'make test' in core

15 years agoBreak the --create-perllibst-h and --list-static-libs options out of buildext.pl
Nicholas Clark [Mon, 26 Jan 2009 22:49:19 +0000]
Break the --create-perllibst-h and --list-static-libs options out of buildext.pl
into separate scripts.

15 years agoMake t/win32/system.t use the more common compiler check
Max Maischein [Mon, 26 Jan 2009 20:49:22 +0000]
Make t/win32/system.t use the more common compiler check

Hello again,

much of the Perl test suite uses $Config{cc} =~ /^gcc/i to determine
whether the compiler looks/acts like gcc. This patch enables that
behaviour in t/win32/system.t too.

The reason behind this is that I have a batch file named gcc-ccache.cmd,
which I give to dmake as my compiler, and having things like this work
everywhere in Perl is nice.

Possibly, a more sane way of keeping the kind of compiler handy would be
to make $Config{CCTYPE} come into existence and have it store "gcc",
"msvc" and what other cc types we support.

-max

From f2e73cb5249433631fd4db5b46f62d05ff35526c Mon Sep 17 00:00:00 2001
From: Max Maischein <corion@corion.net>
Date: Mon, 26 Jan 2009 21:44:36 +0100
Subject: [PATCH] Unbreak the test if your compiler looks like gcc

15 years agoAdd a section on how to submit a patch using GitHub
Leon Brocard [Mon, 26 Jan 2009 15:07:11 +0000]
Add a section on how to submit a patch using GitHub

15 years agoRun podtidy on pod/perlrepository.pod and document how to do so
Leon Brocard [Mon, 26 Jan 2009 13:55:57 +0000]
Run podtidy on pod/perlrepository.pod and document how to do so

15 years agoTests to check cp() doesn't drop set[eu]id bits.
Abigail [Mon, 26 Jan 2009 13:19:29 +0000]
Tests to check cp() doesn't drop set[eu]id bits.
Also, calculate the number of skipped tests instead of having a hard
coded number.

15 years agoCheck for the group entry returned by getpwuid as well when testing
Abigail [Mon, 26 Jan 2009 12:54:48 +0000]
Check for the group entry returned by getpwuid as well when testing
if a file belongs to the group the user is in.

15 years agoFix memory leak
Karl [Mon, 26 Jan 2009 02:02:13 +0000]
Fix memory leak

15 years agovms fileify_dirspec refactor / Unix mode fixes
John Malmberg [Sun, 25 Jan 2009 01:31:39 +0000]
vms fileify_dirspec refactor / Unix mode fixes

This patch refactors the fileify_dirspec routine to not need a thread
context, and also fixes some issue with Unix compatibility mode.

Message-id: <497BC0FB.5000506@gmail.com>

15 years agoAccording to Yuval, Tie::RefHash is maintained at http://github.com/nothingmuch/tie...
Steffen Mueller [Sun, 25 Jan 2009 22:49:43 +0000]
According to Yuval, Tie::RefHash is maintained at github.com/nothingmuch/tie-refhash/, so blead can't be upstream

15 years agovms glob patches
John Malmberg [Fri, 23 Jan 2009 04:45:19 +0000]
vms glob patches

This updates the vms Perl_vms_start_glob routine to behave more like
Unix when the decc$filename_unix_report is active.

It also fixes the behavior of Unix directory syntax when either the
decc$filename_unix_report or decc$efs_charset options are active.

Message-id: <49794B5F.8030401@gmail.com>

15 years ago[perl #57176] Including XSUB.h before perl.h gives nasty error
fawaka@gmail.com (via RT) [Sun, 25 Jan 2009 18:24:16 +0000]
[perl #57176] Including XSUB.h before perl.h gives nasty error

15 years agoAdd perlmini.c to .gitignore
Rafael Garcia-Suarez [Sun, 25 Jan 2009 17:56:42 +0000]
Add perlmini.c to .gitignore

15 years agoSome POD patches for PerlIO and PerlIO::var
Lanning, Scott [Sun, 25 Jan 2009 17:55:25 +0000]
Some POD patches for PerlIO and PerlIO::var

Mostly just grammar.

15 years agoUnbreak Makefile too (replicate fcd65f8d4322911242887c00dd8c419b973735a6)
Nicholas Clark [Sun, 25 Jan 2009 16:42:51 +0000]
Unbreak Makefile too (replicate fcd65f8d4322911242887c00dd8c419b973735a6)

15 years agoUnbreak build on Windows
Max Maischein [Sun, 25 Jan 2009 16:38:10 +0000]
Unbreak build on Windows

Hello p5p,

the recent shakearound of make_patchnum.pl broke the build on Windows,
because make_patchnum.pl lives in the build root, but the Makefile lives
in ./win32, so all targets need to be relative to ./win32.

I've only tested the patch for makefile.mk, as I don't have nmake.exe.
The fix is so simple that it should work on Makefile as well. These are
famous last words.

-max

>From f4d9e594a2a3ea743e9c381d1836a85ada15af85 Mon Sep 17 00:00:00 2001
From: Max Maischein <corion@corion.net>
Date: Sun, 25 Jan 2009 17:34:39 +0100
Subject: [PATCH] Unbreak build on Windows

15 years agoMerge branch 'Eefix' into blead
Rafael Garcia-Suarez [Sun, 25 Jan 2009 15:59:40 +0000]
Merge branch 'Eefix' into blead

15 years agoDocument that $0 is set to "-e" when perl is run with -e or -E
Rafael Garcia-Suarez [Sun, 25 Jan 2009 15:57:19 +0000]
Document that $0 is set to "-e" when perl is run with -e or -E

15 years agoRevert "show -E in error message when called with -E"
Rafael Garcia-Suarez [Sun, 25 Jan 2009 15:54:16 +0000]
Revert "show -E in error message when called with -E"

This reverts commit efdc43f5015438920ce11f05539e55cb02940377.

15 years agoMove cp(1)-like permission changes from copy to cp,
Charles Bailey [Sat, 24 Jan 2009 00:11:45 +0000]
Move cp(1)-like permission changes from copy to cp,

and add setid bit handling

- Moves the permission-changing code to File::Copy::cp(), making it
non-identical to File::Copy::copy().  I think this is a decent
compromise, retaining backwards compatibility in the "primary"
routine, while allowing people who chose cp() to get more POSIXy
semantics.  It does break backwards compatibility of cp() wrt
permissions.  I can find a handful of examples in CPAN, for which I
hope smokers will give some feedback.  As for the DARKpan, who knows .
. .

- Expands the permission changing in cp() a bit to cover the rest of
what cp(1) claims on the local POSIX box.

- Removes Fcntl from the mix, to eliminate some build-time hoops.

- Shifts Abigail's permission tests to cp(), and adds parallel tests for copy().

- Documents the permission-related behavior in more detail.

15 years agoCode refs aren't lockable
Jerry D. Hedden [Wed, 21 Jan 2009 16:03:38 +0000]
Code refs aren't lockable

Remove the check for code refs in pp_lock.
For debugging, assert that the ref is not a code ref
(per Dave Mitchell's suggestion).

15 years agoRemove stock_git_version.h as nothing uses it any more.
Nicholas Clark [Sun, 25 Jan 2009 11:50:01 +0000]
Remove stock_git_version.h as nothing uses it any more.

15 years agoRegenerate uconfig.h
Nicholas Clark [Sun, 25 Jan 2009 11:22:41 +0000]
Regenerate uconfig.h

15 years agomicroperl has no idea about 64 bit types, so don't use the y2038 code for it.
Nicholas Clark [Sun, 25 Jan 2009 11:22:21 +0000]
microperl has no idea about 64 bit types, so don't use the y2038 code for it.

15 years agomicroperl also needs a canned git config.
Nicholas Clark [Sun, 25 Jan 2009 11:13:38 +0000]
microperl also needs a canned git config.

15 years agoPort 486cd780047ff22471c5cbe417911a042ae23962 to win32/ Makefiles.
Nicholas Clark [Sun, 25 Jan 2009 10:33:41 +0000]
Port 486cd780047ff22471c5cbe417911a042ae23962 to win32/ Makefiles.
(miniperl no longer needs git_version.h, perl.o explictly depends on it)

15 years agoAdd -DPERL_IS_MINIPERL when building the object files for miniperl.exe
Nicholas Clark [Sat, 24 Jan 2009 17:45:57 +0000]
Add -DPERL_IS_MINIPERL when building the object files for miniperl.exe

15 years agoThe exit code of make_patchnum.pl is no longer used, so make it return 0.
Nicholas Clark [Sat, 24 Jan 2009 17:10:08 +0000]
The exit code of make_patchnum.pl is no longer used, so make it return 0.
Remove special-casing from all the makefiles that coped with it returning non-0.

15 years agoVMS has a mode where it preserved case, and also has a mode where it
John Malmberg [Sat, 24 Jan 2009 16:28:58 +0000]
VMS has a mode where it preserved case, and also has a mode where it
returns filenames in UNIX syntax.

This updates this test to handle these modes.

15 years agoBuild miniperl with a special version of perl.c on VMS as well.
Craig A. Berry [Sat, 24 Jan 2009 03:07:49 +0000]
Build miniperl with a special version of perl.c on VMS as well.

15 years agosuidperl goes.
Nicholas Clark [Fri, 23 Jan 2009 22:01:26 +0000]
suidperl goes.

15 years agosperl.o also needs to depend on git_version.h
Nicholas Clark [Fri, 23 Jan 2009 20:05:07 +0000]
sperl.o also needs to depend on git_version.h

15 years agoRemove redundant #ifndef SETUID_SCRIPTS_ARE_SECURE_NOW (and #endif).
Nicholas Clark [Fri, 23 Jan 2009 22:40:18 +0000]
Remove redundant #ifndef SETUID_SCRIPTS_ARE_SECURE_NOW (and #endif).
Redundant because it was always true, as it was inside the #else of
an #ifdef SETUID_SCRIPTS_ARE_SECURE_NOW
(Another bit of cleanup following change 32997)

15 years agoEnsure that C++ modules can be built against a perl configured with -DPERL_GLOBAL_STR...
Marcus Holland-Moritz [Fri, 23 Jan 2009 21:11:38 +0000]
Ensure that C++ modules can be built against a perl configured with -DPERL_GLOBAL_STRUCT_PRIVATE.

Perl_GetVarsPrivate() could not be seen from within a module built
with a C++ compiler if the perl binary was built with a C compiler,
as the prototype was not 'extern "C"' for C++.
The change moves the EXTERN_C defines so they can be used for the
prototype of Perl_GetVarsPrivate().

15 years agoUpgrade to Devel::PPPort 3.16
Marcus Holland-Moritz [Fri, 23 Jan 2009 17:48:37 +0000]
Upgrade to Devel::PPPort 3.16

15 years agoAmend patchlevel.h so that git_version.h is only #include""ed for perl.
Nicholas Clark [Fri, 23 Jan 2009 16:41:31 +0000]
Amend patchlevel.h so that git_version.h is only #include""ed for perl.
miniperl builds with canned data equivalent to stock_git_version.h.
git_version.h becomes an explicit dependency for perl.o, make_patchnum.pl is
explicitly run to create it, and no sub-makes are spawned with their inherent
race conditions and consequent strange build failures.

15 years agoAdd a perlmini.o and perlmini.c akin to opmini.o and opmini.c, for ./miniperl
Nicholas Clark [Thu, 22 Jan 2009 17:19:50 +0000]
Add a perlmini.o and perlmini.c akin to opmini.o and opmini.c, for ./miniperl
This will allow a defined order for generating git_version.h and then perl.o

15 years agoFix deparsing of constant subroutines
Rafael Garcia-Suarez [Fri, 23 Jan 2009 13:56:36 +0000]
Fix deparsing of constant subroutines

This was broken in commit 2990415a4519bc3988d7224ae15100c3e9e901ee
where the inlined constants were deparsed also _inside_ the subroutine
that defined them: that is, before they were available to the parser

15 years agoUpgrade to Encode-2.27
Steve Hay [Thu, 22 Jan 2009 09:51:13 +0000]
Upgrade to Encode-2.27

15 years agoNote that nothing in core uses Perl_get_cv any more, and why it is hard to move
Nicholas Clark [Wed, 21 Jan 2009 17:08:36 +0000]
Note that nothing in core uses Perl_get_cv any more, and why it is hard to move
it to mathoms.c

15 years agoConvert db_get_cv() to use get_cvn_flags() as it already knows the length of
Nicholas Clark [Wed, 21 Jan 2009 17:06:14 +0000]
Convert db_get_cv() to use get_cvn_flags() as it already knows the length of
the string that it is passing.

15 years agoAdd get_cvs() as a shortcut for STR_WITH_LEN() and Perl_get_cvn_flags(), and
Nicholas Clark [Wed, 21 Jan 2009 16:41:58 +0000]
Add get_cvs() as a shortcut for STR_WITH_LEN() and Perl_get_cvn_flags(), and
use it where possible.

15 years agoIn attrs::get(), use Perl_get_cvn_flags() rather than perl_get_cv().
Nicholas Clark [Wed, 21 Jan 2009 16:12:06 +0000]
In attrs::get(), use Perl_get_cvn_flags() rather than perl_get_cv().

15 years agoCorrect the form of the flags passed to perl_get_cv().
Nicholas Clark [Wed, 21 Jan 2009 15:55:25 +0000]
Correct the form of the flags passed to perl_get_cv().

15 years agoCorrect the form of the flags passed to perl_get_sv() in threads.xs
Nicholas Clark [Wed, 21 Jan 2009 15:30:05 +0000]
Correct the form of the flags passed to perl_get_sv() in threads.xs

15 years agoCorrect the form of the flags passed to perl_get_sv() in Storable.xs
Nicholas Clark [Wed, 21 Jan 2009 15:14:57 +0000]
Correct the form of the flags passed to perl_get_sv() in Storable.xs

15 years agoUpdate the documentation of get_sv() to note that it calls Perl_gv_fetchpv(),
Nicholas Clark [Wed, 21 Jan 2009 15:03:03 +0000]
Update the documentation of get_sv() to note that it calls Perl_gv_fetchpv(),
and hence the 'create' argument is actually 'flags'. Fix core code and
documentation that used TRUE or FALSE to use 0 or GV_ADD.

15 years agoUpdate the documentation of get_av() to note that it calls Perl_gv_fetchpv(),
Nicholas Clark [Wed, 21 Jan 2009 13:36:08 +0000]
Update the documentation of get_av() to note that it calls Perl_gv_fetchpv(),
and hence the 'create' argument is actually 'flags'. Fix code and documentation
that used TRUE or FALSE to use 0 or GV_ADD.

15 years agoUpdate the documentation of get_hv() to note that it calls Perl_gv_fetchpv(),
Nicholas Clark [Wed, 21 Jan 2009 12:12:12 +0000]
Update the documentation of get_hv() to note that it calls Perl_gv_fetchpv(),
and hence the 'create' argument is actually 'flags'. Fix code and documentation
that used TRUE or FALSE to use 0 or GV_ADD.

15 years agonow get non-dual lived code
Ricardo SIGNES [Tue, 20 Jan 2009 23:32:17 +0000]
now get non-dual lived code

15 years agoremove some indirect method calls, add a caveat
Ricardo SIGNES [Tue, 20 Jan 2009 23:21:34 +0000]
remove some indirect method calls, add a caveat