David Mitchell [Wed, 10 Jun 2009 15:41:25 +0000]
update comment in overload.pl following
813720522f
David Mitchell [Wed, 10 Jun 2009 15:06:08 +0000]
remove from blead, some files already removed from Makemaker 6.52
Steve Peters [Wed, 10 Jun 2009 13:32:55 +0000]
Merge branch 'blead' of ssh://stevep@perl5.git.perl.org/gitroot/perl into blead
Niko Tyni [Wed, 10 Jun 2009 13:32:42 +0000]
[perl #66452] TMPDIR not honored when opening an anonymous temporary file
Steffen Mueller [Wed, 10 Jun 2009 10:25:48 +0000]
Upgrade Attribute::Handlers to 0.84
Paul Marquess [Wed, 10 Jun 2009 03:26:21 +0000]
[PATCH] Filter-Util-Call (was RE: pmqs: dual life modules)
Gisle Aas [Tue, 9 Jun 2009 21:15:54 +0000]
Upgrade to MIME-Base64-3.08
Gisle Aas [Tue, 9 Jun 2009 21:13:26 +0000]
Upgrade to Digest-MD5-2.39
Gisle Aas [Tue, 9 Jun 2009 21:08:52 +0000]
Upgrade to Digest-1.16
Steffen Mueller [Tue, 9 Jun 2009 15:52:55 +0000]
Upgrade to Data::Dumper stable release 2.122
Steve Hay [Tue, 9 Jun 2009 13:47:06 +0000]
Fix threads build on Win32
Follow-up to
8264cf327251f9724c99b125cc3eea11be9b97c5.
Nicholas Clark [Tue, 9 Jun 2009 12:17:54 +0000]
Add lib/NEXT/t/dynamically_scoped_regex_vars.t to MANIFEST.
Jerry D. Hedden [Tue, 9 Jun 2009 11:31:26 +0000]
Mask thread signal handling fix on Win32
Ian Goodacre [Mon, 8 Jun 2009 09:06:13 +0000]
Revised description of => operator
Provided more explicit description of the words that are handled specially
by the => operator and removed the incorrect description of how other
expressions are handled.
Rafael Garcia-Suarez [Tue, 9 Jun 2009 08:25:51 +0000]
Fix regression test for RT 63110 when running with PERL_UNICODE set
This is because block_start already does a copy of the hints in this case.
Jerry D. Hedden [Mon, 8 Jun 2009 18:00:11 +0000]
Upgrade to threads 1.73
Steffen Mueller [Mon, 8 Jun 2009 19:41:14 +0000]
Make h2xs add a dependency on Test.pm for < 5.6.2
Also replaces lots of 5.00702 (note missing 0) with 5.006002. This is to
allow us to jettison Test.pm from core eventually.
Steffen Mueller [Mon, 8 Jun 2009 18:37:11 +0000]
t/op/my_stash.t should use test.pl instead of Test.pm
Jesse Vincent [Mon, 8 Jun 2009 17:56:06 +0000]
Update Test.pm to CPAN's 1.25_02.
Steffen Mueller [Mon, 8 Jun 2009 17:50:43 +0000]
Update Maintainers.pl: Pod-Escapes and Test
Update Maintainers.pl to reflect new maintainer of Pod-Escapes. Correct
Jesse's email address (Test maintainer update).
Steve Peters [Mon, 8 Jun 2009 16:43:56 +0000]
Sync bleadperl and Net-Ping with the release of Net-Ping-2.36
Steffen Mueller [Mon, 8 Jun 2009 16:20:58 +0000]
Jesse Vincent now maintains Test.pm
Steffen Mueller [Mon, 8 Jun 2009 15:53:48 +0000]
Errno is no longer dual-lived
(cf. mail from Graham to p5p). Move to _PERLLIB section of
Porting/Maintainers.pl.
Steffen Mueller [Mon, 8 Jun 2009 15:49:10 +0000]
Bump lib.pm version to match CPAN release
David Mitchell [Mon, 8 Jun 2009 14:59:33 +0000]
add remaining Maintainers.pl EXCLUDED and MAP fields
Florian Ragwitz [Mon, 8 Jun 2009 12:57:15 +0000]
Upgrade NEXT to 0.64.
Stepan Kasal [Wed, 3 Jun 2009 10:03:55 +0000]
fix RT 39060, errno incorrectly set in perlio
Rafael Garcia-Suarez [Mon, 8 Jun 2009 08:35:48 +0000]
Add missing test boilerplates to new MakeMaker tests
Rafael Garcia-Suarez [Mon, 8 Jun 2009 08:03:10 +0000]
Upgrade to MakeMaker 6.53_02
Craig A. Berry [Sun, 7 Jun 2009 21:31:04 +0000]
There is no pthread_sigmask on VMS, so use sigprocmask instead.
According to the standard, use of sigprocmask is undefined in a
multi-threaded context, but in the absence of a per-thread signal
blocking mechanism (likely impossible until and unless per-thread
ASTs become available), this may be the best we can do.
This is a follow-up to:
b762d8667351cb765bc1d6419d30acff085da502.
Steve Peters [Sun, 7 Jun 2009 18:16:22 +0000]
Upgrade to Test-Harness-3.17
Rafael Garcia-Suarez [Sun, 7 Jun 2009 13:55:31 +0000]
Bump version of threads.pm
John Wright [Wed, 6 May 2009 21:48:12 +0000]
Mask signals in thread creation and destruction to avoid a segfault
If our signal handler gets called before a thread got a chance to run
PERL_SET_CONTEXT(), the call to get the thread-specific interpreter will
fail, and we'll end up with aTHX == NULL. Prevent this from happening
by blocking most signals right before creating the new thread. This
way, the new thread starts out with most signals blocked, and it
unblocks them when it's ready to handle them.
This required saving the original signal mask somewhere - I put it in
the thread struct.
In several places, PERL_SET_CONTEXT() is called before the target perl
interpreter struct has been fully initialized, so this patch adds code
to block signals around those blocks of code.
John Wright [Wed, 6 May 2009 06:47:15 +0000]
main: Unregister signal handler before destroying my_perl
If the signal handler runs after perl_destruct() has been called, it
will get an invalid (or NULL) my_perl when it asks for the
thread-specific interpreter struct. This patch resets the signal
handler for any signal previously handled by PL_csighandlerp to SIG_DFL
before calling perl_destruct().
David Mitchell [Sat, 6 Jun 2009 23:51:03 +0000]
add more Maintianers.pl EXCLUDED and MAP fields
David Mitchell [Sat, 6 Jun 2009 16:10:50 +0000]
sync a bunch of files with Test::Simple 0.86
(the $Id$ changes didn't get picked up when blead was last synced with
CPAN)
Steffen Mueller [Sat, 6 Jun 2009 15:00:48 +0000]
Synchronize with Data::Dumper 2.121_20 on CPAN
Rafael Garcia-Suarez [Sat, 6 Jun 2009 12:50:16 +0000]
Mark all .t and .pm files as non executable
Rafael Garcia-Suarez [Sat, 6 Jun 2009 12:18:56 +0000]
Merge commit 'leto/debugger_symbols' into blead
Steffen Mueller [Sat, 6 Jun 2009 11:44:18 +0000]
Data::Dumper on 5.6.* needs to use old IDs
This is the change to Dumper.xs that goes hand-in-hand with change
c2231af7a39be4217e07eedae6069566071d1682 in Dumper.pm.
Steffen Mueller [Sat, 6 Jun 2009 11:06:32 +0000]
For the Data::Dumper CPAN version on 5.6, we need PPPort's sv_2pv_flags
Steffen Mueller [Sat, 6 Jun 2009 10:58:26 +0000]
Increment Data::Dumper version to 2.121_20
Steffen Mueller [Sat, 6 Jun 2009 10:57:21 +0000]
Require at least 5.008 for pack('J'...) in Data::Dumper
Steffen Mueller [Sat, 6 Jun 2009 09:49:23 +0000]
Update Parse::CPAN::Meta to 1.39
Alex Vandiver [Sat, 30 May 2009 16:38:28 +0000]
Faster utf8_length method -- fixes [RT#50250]
UTF8SKIP appears to be a rather slow call; use UTF8_IS_INVARIANT to
skip it whenever possible. We also move the malformed utf8 check
until after the loop, since it can be checked after the termination
condition, instead of at every pass through the loop.
David Mitchell [Sat, 6 Jun 2009 00:52:42 +0000]
add yet more EXCLUDED and MAP keys to Maintainers.pl
(83 done, 30 to go...)
Paul Marquess [Sat, 6 Jun 2009 00:42:38 +0000]
[PATCH] *-Compress-* 2.020
Jim Cromie [Sat, 30 May 2009 17:07:57 +0000]
overload.pl shouldnt update its output unconditionally
fix overload.pl to use rename_if_different, imitating reentr.pl
(including commenting safer_unlink rather than scrubbing it).
Leave lib/overload/number.pm unconditionally generated,
since it is not a dependency for make perl
With this, "make regen perl; make regen perl" will recompile at most 1x.
In turn, this improves utility of regen as an added default make target.
Andy Dougherty [Fri, 5 Jun 2009 16:20:48 +0000]
[perl #9328] Update INSTALL example for BerkeleyDB install
Duke Leto [Thu, 4 Jun 2009 23:11:42 +0000]
Fix my comment about DB::sub and make it DB::DB
Craig A. Berry [Thu, 4 Jun 2009 15:29:57 +0000]
Test case to go with
30e682852bf42358156ed62e06e91f75e9f5b807.
We should be able to depend on SYS$SCRATCH being a non-rooted
logical name.
Craig A. Berry [Thu, 4 Jun 2009 13:10:50 +0000]
MakeMaker must handle an empty $self->{LIBS} array.
6.52 broke the build on VMS because the Makefile.PL for Time::HiRes
sends C<'LIBS' => []> to WriteMakefile() and the code to handle that
case was removed in a refactor, so we ended up not linking against
the main perl library (perlshr.exe on VMS). More details and an
upstream submission of this patch are at:
https://rt.cpan.org/Ticket/Display.html?id=46633
John E. Malmberg [Thu, 4 Jun 2009 13:46:50 +0000]
vms.c EFS logical name fix.
This patch fixes an issue that Craig Berry found in the handling of
logical names of the form foo = "device:[dir]" when the EFS character
set is enabled.
Regards,
-John
wb8tyw@qsl.net
Personal Opinion Only
Signed-off-by: H.Merijn Brand <h.m.brand@xs4all.nl>
Nicholas Clark [Thu, 4 Jun 2009 13:34:57 +0000]
Correctly interpolate the attempted permissions as octal in the error message.
Nicholas Clark [Thu, 4 Jun 2009 13:06:58 +0000]
Assuming that $SIG{PIPE} is not set can be bogus when running in a cron job.
Use $SIG{KILL} instead, as no-one sane would set an untrappable signal.
This should address failures some smokers are reporting.
Duke Leto [Thu, 4 Jun 2009 08:22:46 +0000]
Bump version of perl5db.pl to 1.33, add changelog comments and fix a POD typo
H.Merijn Brand [Thu, 4 Jun 2009 07:37:13 +0000]
Merge branch 'blead' of camel.booking.com:/gitroot/perl into blead
H.Merijn Brand [Thu, 4 Jun 2009 07:36:50 +0000]
Remember if binaries are ELF
Duke Leto [Thu, 4 Jun 2009 06:33:24 +0000]
Merge branch 'debugger_symbols' of github.com:leto/perl into debugger_symbols
Duke Leto [Wed, 3 Jun 2009 20:29:14 +0000]
Add lib/perl5db/t/symbol-table-bug to the MANIFEST
Duke Leto [Wed, 3 Jun 2009 19:28:08 +0000]
Fix shebang line and permissions of lib/perl5db.t
Duke Leto [Wed, 3 Jun 2009 19:02:50 +0000]
Add the debugger script which tests for undefs in the symbol table
Duke Leto [Wed, 3 Jun 2009 18:40:35 +0000]
Fix bug introduced in 67924fd which put the key _< with an undefined value in the symbol table
This patch makes sure that $filename is defined before setting the _<$filename symbol. It also
adds a test which makes sure that *no* symbols have undefined values after loading perl5db.pl .
David Mitchell [Wed, 3 Jun 2009 00:42:20 +0000]
add more EXCLUDED and MAP keys to Maintainers.pl
Chris Williams [Mon, 1 Jun 2009 22:28:45 +0000]
Update CPANPLUS::Dist::Build to CPAN version 0.32
Stepan Kasal [Tue, 2 Jun 2009 14:20:27 +0000]
parenthesize check for a flag
Hello,
gcc warned me about a suspicious combination of ! and &
Please consinder the following one-line patch.
Stepan Kasal
From
4b449da2aedf3b40b8f6dfa4c410d9becb33330a Mon Sep 17 00:00:00 2001
From: Stepan Kasal <skasal@redhat.com>
Date: Tue, 2 Jun 2009 16:06:10 +0200
Subject: [PATCH] dump.c: parenthesize flag check
Signed-off-by: H.Merijn Brand <h.m.brand@xs4all.nl>
Nicholas Clark [Tue, 2 Jun 2009 12:53:19 +0000]
Disentangling -g and -DDEBUGGING is a TODO.
Nicholas Clark [Tue, 2 Jun 2009 10:36:35 +0000]
In Perl_hv_iternext_flags(), clarify and generalise the empty hash bailout code.
Rafael Garcia-Suarez [Tue, 2 Jun 2009 06:51:22 +0000]
Adjust MANIFEST
Rafael Garcia-Suarez [Tue, 2 Jun 2009 06:47:54 +0000]
Merge branch 'smartmatch' into blead
Rafael Garcia-Suarez [Tue, 2 Jun 2009 06:46:41 +0000]
Merge branch 'eumm652' into blead
David Mitchell [Tue, 2 Jun 2009 00:13:51 +0000]
stop git whining about trailing whitespace (picky picky..)
David Mitchell [Tue, 2 Jun 2009 00:10:01 +0000]
start adding new EXCLUDED and MAP entries to Maintainers.pl.
David Mitchell [Tue, 2 Jun 2009 00:08:19 +0000]
resync some files with their CPAN equivalents (just the trivial errors for
now, like whitespace)
Max Maischein [Mon, 1 Jun 2009 12:18:42 +0000]
Add benchmark test for keys() on empty hashes (RT26188)
Max Maischein [Sun, 31 May 2009 21:50:12 +0000]
Fix RT26188, speed up keys() on empty hash
Jerry D. Hedden [Mon, 1 Jun 2009 14:09:24 +0000]
Correct test output for t/op/eval.t (missing newline)
Andy Dougherty [Mon, 1 Jun 2009 16:34:36 +0000]
Protect against spaces in build directory name when extracting .PL files
Rafael Garcia-Suarez [Mon, 1 Jun 2009 15:32:56 +0000]
More ~~ tests
Also, add a plan to those tests, for extra safety
Rafael Garcia-Suarez [Mon, 1 Jun 2009 15:16:40 +0000]
Smart-match perldelta adjustments
Rafael Garcia-Suarez [Mon, 1 Jun 2009 14:46:42 +0000]
Minor doc fix for the smart table legend (clarify the "Object" entry).
Rafael Garcia-Suarez [Mon, 1 Jun 2009 13:52:31 +0000]
Add test boilerplate to the new MakeMaker tests
Rafael Garcia-Suarez [Mon, 1 Jun 2009 13:44:14 +0000]
Upgrade to ExtUtils::MakeMaker 6.52
Steffen Mueller [Sun, 31 May 2009 20:02:33 +0000]
Mention Devel::NYTProf in perldebug
In the short section about profiling with Devel::DProf, we now mention
that other profilers (such as Devel::NYTProf) are available from CPAN.
Steffen Mueller [Sun, 31 May 2009 19:57:24 +0000]
Auto-complete lexicals in the debugger shell
When typing the name of a lexical variable in the debugger shell, its
name can now be tab auto-completed just like package variables. Requires
PadWalker to be available. Silently skips lexicals if it's not.
Rafael Garcia-Suarez [Sun, 31 May 2009 20:08:41 +0000]
Fix casting warning
(that was introduced by
9ef5ed94af)
David Mitchell [Sun, 31 May 2009 19:39:33 +0000]
wrap some long FILES fields in Maintainers.pl
(just whitepsace/newline changes)
Basic philosophy is now: if it all fits on one ( < 80 char) line, fine;
otherwise have only one entry per line
David Mitchell [Sun, 31 May 2009 19:22:55 +0000]
fix Math::BigInt, PathTools entries in Maintainers.pl
(botched by
002904c4e2f6cd5caebddf970e6be562946af502)
David Mitchell [Sun, 31 May 2009 19:00:54 +0000]
convert Maintainers.pl from an 8 to a 4 char indent
(to give more room for extra field I'm going to add)
David Mitchell [Sun, 31 May 2009 18:19:58 +0000]
Clarify the meaning of the %Modules keys in Maintainers.pl,
and fix a couple of keys that don't correspond to an actual module name
that you can 'use'.
David Mitchell [Sun, 31 May 2009 18:00:23 +0000]
add DISTRIBUTION field to %Modules in Maintainers.pl
(note that in this initial attempt, the file version numbers are derived
from the newest verions currently available on CPAN, which may not actually
reflect what's bundled with blead. Later on, I'll be fixing this.)
David Mitchell [Sun, 31 May 2009 13:43:56 +0000]
Maintainers.pl: rationalise distributions:
Cwd and File::Spec are both part of PathTools, and
Math::BigFloat is part of Math-BigInt
David Mitchell [Sun, 31 May 2009 13:17:26 +0000]
Maintainers.pl: there is no DB::File, and describe Text::Tabs/Wrap better
Bram [Sun, 31 May 2009 15:20:45 +0000]
Clarify perldebug documentation
Mention what happens without Readline support.
Rafael Garcia-Suarez [Sun, 31 May 2009 14:30:59 +0000]
Silence a warning emitted by test added by last commit
George Greer [Wed, 27 May 2009 01:05:40 +0000]
Add test case from RT#54758.
Rafael Garcia-Suarez [Sun, 31 May 2009 13:11:40 +0000]
Merge branch 'memleaktest' into blead
Rafael Garcia-Suarez [Sun, 31 May 2009 13:09:56 +0000]
Add regression test for [RT#63110]
This is based on a stand-alone test by Bram
Bram [Sun, 31 May 2009 12:47:49 +0000]
Document test structure
(was Re: [perl #56194] Regex: (((??{1 + $^N}))) behaves differently in
5.10.0 than in blead)
Adds some basic documentation about the test structure.
David Mitchell [Sat, 30 May 2009 16:17:45 +0000]
Fix Filter::Simple entry in Maintainers.pl