Rafael Garcia-Suarez [Thu, 16 Apr 2009 22:50:39 +0000]
Ignore some autogenerated makefiles
Nicholas Clark [Thu, 16 Apr 2009 21:57:05 +0000]
Avoid loading Config, as we don't need to distinguish between NetWare and Win32.
Eliminate $Is_W32 $Is_OS2 $Is_Cygwin $Is_NetWare, and simply set $Needs_Write
directly from $^O.
As far as I can tell, these were all added in
36bb303b6ac55df9c2780b48d374c505
but never actually used for anything.
Rafael Garcia-Suarez [Thu, 16 Apr 2009 22:00:43 +0000]
Ignore attributes.pm in lib since it's now an XS module
Rafael Garcia-Suarez [Thu, 16 Apr 2009 21:41:53 +0000]
Bump VERSION of Data::Dumper
Alex Vandiver [Wed, 15 Apr 2009 17:04:23 +0000]
Data::Dumper hash iterator needs to be reset on all hashrefs (fixes #64744)
Niko Tyni [Tue, 14 Apr 2009 19:55:34 +0000]
Squelch 'Constant subroutine ... undefined' warnings from .ph files
As reported by Christopher Zimmermann in <http://bugs.debian.org/379757>,
code generated from simple #undef directives by h2ph can cause
'Constant subroutine ... undefined' warnings if the undefined
function was eligible for inlining.
Niko Tyni [Tue, 14 Apr 2009 19:55:33 +0000]
Add tests to verify that h2ph output compiles and is warning free
The #include directives are #ifdef'd out so that running the
resulting code does not actually need the headers. We still
get the same effect from comparing with the expected h2ph output.
Paul Marquess [Thu, 16 Apr 2009 08:55:53 +0000]
Build bzip2 witha C++ compiler
Chris 'BinGOs' Williams [Thu, 16 Apr 2009 20:55:52 +0000]
Update CPANPLUS::Dist::Build to 0.22
Tom Hukins [Tue, 14 Apr 2009 15:21:24 +0000]
Documentation Patch: perlxs.pod
Porters,
Please find attached a small patch to the perlxstut.pod document.
The first part of the patch makes a link more precise. The
second, more useful, part of the patch fixes some confused English
that doesn't make sense.
Thanks for your efforts,
Tom
Signed-off-by: H.Merijn Brand <h.m.brand@xs4all.nl>
Steve Peters [Tue, 14 Apr 2009 05:21:05 +0000]
Add file missed in previous commit.
Andreas J. Koenig [Mon, 13 Apr 2009 21:35:16 +0000]
[PATCH] Update CPAN.pm to 1.93_52
Steve Peters [Mon, 13 Apr 2009 13:57:14 +0000]
Merge branch 'blead' of ssh://stevep@perl5.git.perl.org/gitroot/perl into blead
Steve Peters [Mon, 13 Apr 2009 13:56:36 +0000]
Upgrade to CGI.pm-3.43
Nicholas Clark [Mon, 13 Apr 2009 13:38:42 +0000]
Remove an unnecessary use Config;
Nicholas Clark [Mon, 13 Apr 2009 13:37:19 +0000]
Remove Makefile.PLs that make_ext.pl can automatically generate.
(For core modules, or where the dual life module on CPAN has a Makefile.PL that
is not the same. AUTHOR is only used for the ppm target, which the core doesn't
need.)
Paul Marquess [Mon, 13 Apr 2009 13:23:28 +0000]
[PATCH] Compress-Raw-Bzip2 2.018 (was RE: Compress-Raw-Bzip2 fails for g++)
Nicholas Clark [Mon, 13 Apr 2009 12:09:43 +0000]
Restore prototypes to XS code in attributes, accidentally lost by change 48462a7
Nicholas Clark [Mon, 13 Apr 2009 11:13:19 +0000]
TODO tests for deparsing strict, warnings as strings, and my $x if 0;
Nicholas Clark [Mon, 13 Apr 2009 10:47:15 +0000]
our $TODO should be local $::TODO, revealing a bug in the de-commenting regexp.
Remove some orphaned trailing quotes from TODO message that are no longer
conditional code.
Nicholas Clark [Mon, 13 Apr 2009 10:22:11 +0000]
Correct two other tests that apparently should be TODOs rather than SKIPs.
Nicholas Clark [Mon, 13 Apr 2009 09:51:20 +0000]
A perl-version portable fix for B coping with CVf_LOCKED being removed in 5.11.
Nicholas Clark [Mon, 13 Apr 2009 09:31:49 +0000]
Tweak t/lib/common.pl to use test.pl's (better) tempfile() function.
Nicholas Clark [Mon, 13 Apr 2009 09:26:00 +0000]
Propagate the upgraded TODO/SKIP parser from ext/B/t/deparse.t
Nicholas Clark [Mon, 13 Apr 2009 09:25:01 +0000]
Refactor the SKIP parser to also handle TODOs. TODO 3 tests that were wrongly
SKIPs.
Nicholas Clark [Mon, 13 Apr 2009 08:00:51 +0000]
Remove CVf_LOCKED and CvLOCKED*(), now that nothing sets that flag bit.
Nicholas Clark [Mon, 13 Apr 2009 07:43:30 +0000]
GvUNIQUE* have been defined as 0 since 2005/06/30 - high time to remove them.
Nicholas Clark [Sun, 12 Apr 2009 16:23:32 +0000]
Remove a link to the attrs pragma, which has now been removed.
Nicholas Clark [Sun, 12 Apr 2009 16:06:31 +0000]
s/locked/lvalue/ in the examples, and remove mention of the deprecated :unique.
(Which was actually a dangling cross reference, as what it pointed to in
perlfunc had already been deleted.)
Nicholas Clark [Sun, 12 Apr 2009 15:28:49 +0000]
Deprecate using "unique" with the attributes pragma.
Nicholas Clark [Sun, 12 Apr 2009 14:50:16 +0000]
Deprecate using "locked" with the attributes pragma.
Nicholas Clark [Sun, 12 Apr 2009 14:05:04 +0000]
Convert xsutils.c and lib/attributes.pm to a regular XS extension.
Nicholas Clark [Sun, 12 Apr 2009 13:05:51 +0000]
Remove #ifdef CVf_LVALUE/#endif - LVALUE subs aren't conditionally compiled.
Nicholas Clark [Sun, 12 Apr 2009 11:47:27 +0000]
Deprecate use of the attribute :locked on subroutines.
Nicholas Clark [Sun, 12 Apr 2009 09:56:37 +0000]
Eliminate "Old Perl threads tutorial", which describes 5005 threads.
pod/perlthrtut.pod is the same material reworked for ithreads.
Nicholas Clark [Sun, 12 Apr 2009 09:47:47 +0000]
Rebuild pod/perltoc.pod.
Nicholas Clark [Sun, 12 Apr 2009 09:47:36 +0000]
Sort the MANIFEST.
Nicholas Clark [Sun, 12 Apr 2009 09:17:24 +0000]
Remove attrs, which has been deprecated since 1999/10/02.
Florian Ragwitz [Fri, 10 Apr 2009 15:29:05 +0000]
Upgrade to NEXT-0.63.
Nicholas Clark [Thu, 9 Apr 2009 18:25:37 +0000]
Don't SEGV while warning about redefining the format STDOUT. Fixes RT #64562.
Andy Dougherty [Thu, 9 Apr 2009 15:41:34 +0000]
Fix declaration-after-statement in ext/Socket/Socket.xs
This trivial patch lets C89-ish compilers compile Socket.c again.
Signed-off-by: H.Merijn Brand <h.m.brand@xs4all.nl>
Chris Williams [Thu, 9 Apr 2009 12:01:41 +0000]
Update CPANPLUS::Dist::Build to 0.20
Rafael Garcia-Suarez [Thu, 9 Apr 2009 07:56:39 +0000]
Merge commit 'rafl/blead' into blead
David Cantrell [Wed, 8 Apr 2009 12:49:14 +0000]
Net::hostent fails make test if can't resolve localhost
This happens on boxes which otherwise work Just Fine and whose resolvers
work Just Fine if they don't have /etc/hosts. This patch fixes the test
file by making it skip if localhost can't be resolved and /etc/hosts
doesn't exist.
Message-ID: <
20090330205715.GA29690@bytemark.barnyard.co.uk>
Marcus Holland-Moritz [Wed, 8 Apr 2009 07:49:19 +0000]
Use of freed comppad array during clear_yystack()
Message-ID: <
20081026231720.34258457@r2d2>
Patch description from the original email :
I tried to make tests pass on a perl built with -DPERL_POISON,
as some tests were dying with segfaults. They all originated
from the same source: clear_yystack() after a compile error. [...]
As far as I can see, after croaking the newly
created CV is destroyed and its pad is undef'd. [...]
This will SvREFCNT_dec PL_comppad and set PL_comppad to NULL.
However, later, in clear_yystack(), when the ops are freed, the
old PL_comppad is restored by PAD_RESTORE_LOCAL, as a reference
is still in ps->comppad. But now the pad AV is already dead.
Normally (i.e. without PERL_POISON), the dead AV will have
AvARRAY(av) set to NULL by av_undef(). So PAD_RESTORE_LOCAL will
actually set PL_curpad to NULL, and thus pad_free() will not
attempt to do anything.
But with PERL_POISON, the storage for AvARRAY(av) (i.e. sv_u)
will be reused for chaining the free SV heads in the arena
(as opposed to SvANY(sv) in case of !PERL_POISON). This means
that PAD_RESTORE_LOCAL will find AvARRAY(av) non-NULL and will
set PL_curpad to that value, finally causing the segfault in
pad_free().
While I think I understand what's going on, I don't have the
slightest clue how to properly fix this. Given that it's not
a problem only under PERL_POISON, but always (as dead SV heads
are being used), I think it should ultimately be fixed.
The only thing I can offer right now is a patch to make it
work with PERL_POISON as good (or as bad) as without by
making PAD_RESTORE_LOCAL explicitly check if the pad passed
in is already dead and refusing to use it if it is.
Niko Tyni [Sun, 19 Oct 2008 20:25:10 +0000]
h2xs incorrectly treats enum values like macros
Patch and description by Rainer Weikusat, forwarded from
http://bugs.debian.org/502297 :
The h2xs program scans C headers for 'constants' either defined
as preprocessor macros or via enum and builds a sorted list of names
containing the macros and enum values found in this way. This list
is then passed to ExtUtils::Constant::WriteConstants, which generates
the corresponding const-c.inc and const-xs.inc files when perl Makefile.PL
is executed. By default, ie when just processing a constant name, this
function assumes that the name refers to a preprocessor macro and
the generated C-code in const-c.inc contains conditional compilation
directives to either return the macro value or Perl_constant_NOTDEF,
depending on the defined'ness of the macro. This causes constants
defined as enumeration values to never be available to module users,
because they are not visible to the preprocessor.
Instead of passing just a name to the WriteConstants-routine, a
hashref (members documented in ExtUtils::Constant::Base(3perl))
can be used to specify more details regarding what code
should be generated for a particular constant. For an enumeration
value, a hashref constructed as
{ name => <name of the constant>, macro => 1 }
could be used to get rid of the inappropriate preprocessor
directives.
( Another version of the same fix was also submitted by Daniel Burr in
http://bugs.debian.org/320286 )
Florian Ragwitz [Wed, 8 Apr 2009 01:37:43 +0000]
Upgrade to NEXT-0.62.
Michael Witten [Tue, 7 Apr 2009 22:46:00 +0000]
Docs: Fixed a couple of [my] typos
I read through each my patches again and came across a typo,
a slight incorrectness, and a repeated word. Sorry.
Signed-off-by: Michael Witten <mfwitten@gmail.com>
Michael Witten [Tue, 7 Apr 2009 19:59:32 +0000]
Docs: Update summary and add copyright
Yeah, well, I wrote some new stuff.
Michael Witten [Tue, 7 Apr 2009 19:59:31 +0000]
Docs: Expanded treatment of adding instance variables
Signed-off-by: Michael Witten <mfwitten@gmail.com>
Michael Witten [Tue, 7 Apr 2009 19:59:30 +0000]
Docs: Clarify that a class is not an instance
Signed-off-by: Michael Witten <mfwitten@gmail.com>
Michael Witten [Tue, 7 Apr 2009 19:59:29 +0000]
Docs: Minor modifications to discussion of constructor
Signed-off-by: Michael Witten <mfwitten@gmail.com>
Michael Witten [Tue, 7 Apr 2009 19:59:28 +0000]
Docs: Make use of $self and $class clear
Signed-off-by: Michael Witten <mfwitten@gmail.com>
Michael Witten [Tue, 7 Apr 2009 19:59:27 +0000]
Docs: Better orgnization of instance discussion
Signed-off-by: Michael Witten <mfwitten@gmail.com>
Michael Witten [Tue, 7 Apr 2009 19:59:26 +0000]
Docs: Cleanup lead-in to instance variables
Signed-off-by: Michael Witten <mfwitten@gmail.com>
Michael Witten [Tue, 7 Apr 2009 19:59:25 +0000]
Docs: Better[?] development of SUPER
Signed-off-by: Michael Witten <mfwitten@gmail.com>
Michael Witten [Tue, 7 Apr 2009 19:59:24 +0000]
Docs: Get rid of those awful tabs!
Signed-off-by: Michael Witten <mfwitten@gmail.com>
Michael Witten [Tue, 7 Apr 2009 19:59:23 +0000]
Docs: Quick explanation of duck typing and a warning
Signed-off-by: Michael Witten <mfwitten@gmail.com>
Michael Witten [Tue, 7 Apr 2009 19:59:22 +0000]
Docs: Make the "use base" purpose more obvious
Signed-off-by: Michael Witten <mfwitten@gmail.com>
Michael Witten [Tue, 7 Apr 2009 19:59:21 +0000]
Docs: Minor reorganization.
Signed-off-by: Michael Witten <mfwitten@gmail.com>
Michael Witten [Tue, 7 Apr 2009 19:59:20 +0000]
Docs: Typo: 'no strict subs' -> 'no strict refs'
Signed-off-by: Michael Witten <mfwitten@gmail.com>
Michael Witten [Tue, 7 Apr 2009 19:59:19 +0000]
Docs: Typo: \{-1} -> \g{-1}
Signed-off-by: Michael Witten <mfwitten@gmail.com>
Niko Tyni [Sun, 5 Apr 2009 19:29:38 +0000]
Make h2ph look in "include-fixed", needed by at least gcc 4.3
As seen in <http://bugs.debian.org/522673>, since 4.2.0 or so
gcc's search path has a new "include-fixed" directory,
which currently contains <syslimits.h> and a few other headers.
Converting <syslimits.h> therefore fails unless h2ph knows about the new
directory too.
See http://gcc.gnu.org/ml/gcc-patches/2007-02/msg02038.html for some
background to the gcc change.
Rafael Garcia-Suarez [Tue, 7 Apr 2009 10:31:16 +0000]
Deprecate assignment to $[
Based on a patch by James Mastros :
Subject: Deprecating $[
Message-ID: <
abc933c50904020726x31776ab5m192036429af16f03@mail.gmail.com>
H.Merijn Brand [Mon, 6 Apr 2009 10:27:02 +0000]
Remove libbsd/libBSD from wanted libs on AIX
Subject: libbsd on AIX
From: Rainer Tammer <tammer@tammer.net>
Date: Fri, 03 Apr 2009 11:59:18 +0200
Message-ID: <
49D5DDF6.5050804@tammer.net>
Nicholas Clark [Fri, 3 Apr 2009 19:41:48 +0000]
In sdbm_open(), replace strcat()s and strcpy()s with memcpy()s.
Nicholas Clark [Fri, 3 Apr 2009 17:27:30 +0000]
Don't needlessly assign to pagname (which seems to make clang even happier).
Nicholas Clark [Fri, 3 Apr 2009 17:10:12 +0000]
Don't assign needlessly to dirname. (which seems to make clang happier.)
David Mitchell [Thu, 2 Apr 2009 22:09:01 +0000]
document PERL_SYS_INIT, PERL_SYS_TERM and that they should only be used once
Craig A. Berry [Thu, 2 Apr 2009 17:30:54 +0000]
Skip bug in new File::stat test.
-- index returns -1, not false when substring not found.
-- 2 tests in the block, not 1.
Rafael Garcia-Suarez [Wed, 1 Apr 2009 08:06:26 +0000]
Sadahiro says that "first-come" should be preferred for Unicode::Normalize and Unicode::Collate
Nicholas Clark [Tue, 31 Mar 2009 21:06:14 +0000]
Remove the 5.005 threads details from the documentation for :locked and :method
Jerry D. Hedden [Tue, 31 Mar 2009 20:52:20 +0000]
Can't use inet_ntop and inet_pton under Cygwin because it doesn't fully implement IPv6 yet.
Nicholas Clark [Tue, 31 Mar 2009 20:15:53 +0000]
Close the file before renaming it. Problem spotted and fix supplied by corion.
Rafael Garcia-Suarez [Tue, 31 Mar 2009 13:37:52 +0000]
Getopt::Long is maintained on CPAN
Rafael Garcia-Suarez [Tue, 31 Mar 2009 13:37:03 +0000]
Upgrade to Getopt::Long 2.38
Nicholas Clark [Tue, 31 Mar 2009 13:23:58 +0000]
Don't create an empty Errno.pm if there is an error whilst running Errno_pm.PL
Implemented by initially writing to an temporary file name, and renaming as the
last act.
Nicholas Clark [Tue, 31 Mar 2009 09:33:22 +0000]
Note why we need a seemingly useless empty &Regexp::DESTROY
Nicholas Clark [Tue, 31 Mar 2009 09:32:28 +0000]
Avoid "Constant subroutine ... redefined" warning, spotted by Jerry D. Hedden.
Steve Peters [Tue, 31 Mar 2009 05:54:39 +0000]
Upgrade to CPANPLUS-Dist-Build-0.18.
Craig A. Berry [Mon, 30 Mar 2009 12:27:34 +0000]
TZ setting must contain an offset.
Per the standard at:
http://www.opengroup.org/onlinepubs/
7990989775/xbd/envvar.html
H.Merijn Brand [Mon, 30 Mar 2009 10:19:22 +0000]
Make references to the Glossary in Config.pm and Config.pod
Paul Marquess [Mon, 30 Mar 2009 07:17:04 +0000]
Updates for ext/compress*
Vincent Pit [Sun, 29 Mar 2009 22:57:33 +0000]
Document when as a syntax modifier
David Mitchell [Sun, 29 Mar 2009 21:03:13 +0000]
provide minimal documentation for the MUTABLE_*() macros
Rafael Garcia-Suarez [Sun, 29 Mar 2009 20:48:10 +0000]
Merge commit 'origin/postfixwhen' into blead
Nicholas Clark [Sun, 29 Mar 2009 20:39:05 +0000]
Document that DESTROY is subject to AUTOLOAD, and the related optimisations.
Nicholas Clark [Sun, 29 Mar 2009 19:05:06 +0000]
Change Regexp::DESTROY to be a constant subroutine, removing XS_Regexp_DESTROY.
A constant subroutine is now optimised away at DESTROY time.
Nicholas Clark [Sun, 29 Mar 2009 19:03:18 +0000]
Change Perl_newCONSTSUB() so that a NULL sv generates an empty list return.
Don't call DESTROY if it's a constant subroutine.
Nicholas Clark [Sun, 29 Mar 2009 11:30:25 +0000]
In Perl_newCONSTSUB(), sv should not be NULL.
SADAHIRO Tomoyuki [Sun, 29 Mar 2009 14:50:35 +0000]
Update to Unicode::Normalize 1.03
1.03 Sun Mar 29 12:56:23 2009
- mkheader: check if no composition needs growing the string.
- Makefile.PL: a tweak
Paul Marquess [Sat, 28 Mar 2009 13:46:58 +0000]
DB_File 1.820: Remove MAN3PODS => {}
Craig A. Berry [Sun, 29 Mar 2009 00:12:11 +0000]
Make Time::Piece use the init_tm() in util.c, bump to 1.13_03.
Jarkko put it there for that purpose back in
e72cf795050cdfe9905e00270c38ba2547626581,
so we seem to be a little slow noticing.
The old broken method for determining whether we need to initialize a tm
struct in the absence of HAS_TM_TM_ZONE being available is preserved for
Perls prior to 5.8.0.
Vincent Pit [Sun, 29 Mar 2009 00:02:55 +0000]
Don't call SvPV_const() on the method name at the top of method_common(), as it's not needed when the method is in the current stash (which happens quite often)
Nicholas Clark [Sat, 28 Mar 2009 23:00:29 +0000]
Don't COW PVFMs, as the IVX is now used for FmLINES. (since change 34134)
Nicholas Clark [Sat, 28 Mar 2009 23:00:15 +0000]
In Perl_sv_setsv_cow(), dstr can be NULL.
Vincent Pit [Tue, 24 Mar 2009 20:49:14 +0000]
Allow when to be used as a statement modifier
Florian Ragwitz [Sat, 28 Mar 2009 10:10:08 +0000]
Improve mro merging error messages.
This makes them very similar to those produced by Algorithm::C3.
Vincent Pit [Fri, 27 Mar 2009 16:55:46 +0000]
Also skip test 20 in t/Socket.t when no IPv6 is available
Craig A. Berry [Fri, 27 Mar 2009 16:38:58 +0000]