This is a live mirror of the Perl 5 development currently hosted at
12 years agoIn strftime(), save a malloc()/free() by using sv_usepvn_flags().
Nicholas Clark [Thu, 15 Oct 2009 15:26:51 +0000 (16:26 +0100)]
In strftime(), save a malloc()/free() by using sv_usepvn_flags().

12 years agoPOSIX::strftime() should be able to handle Unicode characters in the format
Nicholas Clark [Thu, 15 Oct 2009 16:09:12 +0000 (17:09 +0100)]
POSIX::strftime() should be able to handle Unicode characters in the format

(Restore the intent of 9e8c01f558a03902ff2f54935fd7e6dcc7ec656c, but with non-
buggy tests. Improve the implementation so that it doesn't always upgrade the
format string to UTF-8.)

12 years agoFix 2 pod errors in perlvms.pod.
Craig A. Berry [Thu, 15 Oct 2009 16:01:33 +0000 (11:01 -0500)]
Fix 2 pod errors in perlvms.pod.

12 years agoMake p_tainted.t find its tests on VMS.
Craig A. Berry [Thu, 15 Oct 2009 15:59:09 +0000 (10:59 -0500)]
Make p_tainted.t find its tests on VMS.

Reported upstream at:

12 years agoHmmm... this may be a heisenbug. Works on some terminals but not
Steve Peters [Thu, 15 Oct 2009 14:50:42 +0000 (09:50 -0500)]
Hmmm... this may be a heisenbug.  Works on some terminals but not

This reverts commit 9e8c01f558a03902ff2f54935fd7e6dcc7ec656c.

12 years agoMerge branch 'blead' of ssh:// into blead
Steve Peters [Thu, 15 Oct 2009 14:14:47 +0000 (09:14 -0500)]
Merge branch 'blead' of ssh:// into blead

12 years agoPOSIX::strftime() should be able to handle Unicode characters in the
Steve Peters [Thu, 15 Oct 2009 14:12:57 +0000 (09:12 -0500)]
POSIX::strftime() should be able to handle Unicode characters in the
format string the same as ASCII ones.

12 years agoMove the boolkeys op to the group of hash ops.
Nicholas Clark [Thu, 15 Oct 2009 14:12:49 +0000 (15:12 +0100)]
Move the boolkeys op to the group of hash ops.

This breaks binary compatibility.

12 years agoOptimise if (%foo) to be faster than if(keys %foo)
demerphq [Thu, 15 Oct 2009 13:27:30 +0000 (14:27 +0100)]
Optimise if (%foo) to be faster than if(keys %foo)

Thread was "[PATCH] Make if (%hash) {} act the same as if (keys %hash) {}"
but the implementation evolved from the approach described in the subject, to
instead add a new opcode pp_boolkeys, to exactly preserve the existing

Various conflicts with the passage of time resolved, 'register' removed, and a
$VERSION bump.

12 years agoSupport for pp_boolkeys in B::Deparse.
demerphq [Thu, 15 Oct 2009 13:22:47 +0000 (14:22 +0100)]
Support for pp_boolkeys in B::Deparse.

Part of "[PATCH] Make if (%hash) {} act the same as if (keys %hash) {}"
which evolved from the approach described in the subject, to instead add a new
opcode pp_boolkeys, to exactly preserve the existing behaviour.

Plus a $VERSION bump.

12 years agoUpdated a comment in Porting/makerel
Jesse Vincent [Wed, 14 Oct 2009 19:24:54 +0000 (15:24 -0400)]
Updated a comment in Porting/makerel

12 years agoAdd mingw64 support for win32
Sisyphus [Wed, 14 Oct 2009 14:02:16 +0000 (10:02 -0400)]
Add mingw64 support for win32

12 years agoRemove old code dealing with installing 5.003's perldiag
Rafael Garcia-Suarez [Wed, 14 Oct 2009 09:40:35 +0000 (11:40 +0200)]
Remove old code dealing with installing 5.003's perldiag

12 years agoIgnore files installed by Parse::CPAN::Meta
Rafael Garcia-Suarez [Wed, 14 Oct 2009 09:36:44 +0000 (11:36 +0200)]
Ignore files installed by Parse::CPAN::Meta

12 years agoDon't install
Rafael Garcia-Suarez [Wed, 14 Oct 2009 09:35:10 +0000 (11:35 +0200)]
Don't install

12 years agoPatch t/win32/system.t for mingw32/64
Sisyphus [Wed, 14 Oct 2009 10:58:49 +0000 (06:58 -0400)]
Patch t/win32/system.t for mingw32/64

12 years agoFix {$^MATCH} typo in perlre
brian d foy [Wed, 14 Oct 2009 08:08:25 +0000 (03:08 -0500)]
Fix {$^MATCH} typo in perlre

12 years agoThe winsock select() implementation doesn't support all empty 'fd_set's.
Jan Dubois [Tue, 13 Oct 2009 23:46:58 +0000 (16:46 -0700)]
The winsock select() implementation doesn't support all empty 'fd_set's.

The code already contained a workaround for the special case

    select(undef, undef, undef, $sleep);

but didn't handle the case when actual bit vectors were passed in
that didn't have any bits set.


12 years agoSort MANIFEST following commit 9bc662c9c4c134e4.
Nicholas Clark [Tue, 13 Oct 2009 17:05:09 +0000 (18:05 +0100)]
Sort MANIFEST following commit 9bc662c9c4c134e4.

12 years agoNote that a test that was run is TODO.
Nicholas Clark [Tue, 13 Oct 2009 17:02:46 +0000 (18:02 +0100)]
Note that a test that was run is TODO.

12 years agoMinor podification nit. *2 legs bad* B<4 legs good>.
Nicholas Clark [Tue, 13 Oct 2009 15:51:09 +0000 (16:51 +0100)]
Minor podification nit. *2 legs bad* B<4 legs good>.

12 years agoIncorrect warning category case
Eric Brine [Tue, 13 Oct 2009 16:14:51 +0000 (12:14 -0400)]
Incorrect warning category case

12 years agoFix minor typo in perlpolicy.pod
Steve Hay [Tue, 13 Oct 2009 16:44:41 +0000 (17:44 +0100)]
Fix minor typo in perlpolicy.pod

12 years agoWork around Parse-CPAN-Meta's desire to run a command at 'make test' time. - Commit...
Jesse Vincent [Tue, 13 Oct 2009 15:14:03 +0000 (11:14 -0400)]
Work around Parse-CPAN-Meta's desire to run a command at 'make test' time.  - Commit the generated file

12 years agoMove the new "perl policy" document into pod/
Jesse Vincent [Tue, 13 Oct 2009 14:36:34 +0000 (10:36 -0400)]
Move the new "perl policy" document into pod/

12 years agoPodify the social contract about contributed modules
Jesse Vincent [Tue, 13 Oct 2009 14:32:38 +0000 (10:32 -0400)]
Podify the social contract about contributed modules

12 years agoperlvar $1 clarification
Smylers [Tue, 13 Oct 2009 12:14:46 +0000 (14:14 +0200)]
perlvar $1 clarification

$1 is currently documented as being set by the "last pattern match".
But it is left alone by unsuccessful pattern match attempts (continuing
to hold a value from an earlier successful match).

Saying "last successful pattern match" clarifies this; it's also the
phrase used to document $&.

Second, the entry for $1 in perlvar doesn't actually contain the text
"$1" anywhere.  As such, doing man perlvar then using /\$1 to search for
it in less doesn't locate it (though does match other places in that
file where $1 happens to be used).

12 years agoChange S_ckwarn_common() to looping over the packed warning value.
Nicholas Clark [Tue, 13 Oct 2009 12:01:52 +0000 (13:01 +0100)]
Change S_ckwarn_common() to looping over the packed warning value.

The core never uses WARN3() or WARN4(), and rarely uses WARN2(), so the
previous code, effectively an unwrapped loop, wasn't a speed up. Functionally
equivalent smaller code fits better into CPU caches.

12 years agoMigrate common code in Perl_ckwarn() and Perl_ckwarn_d() to S_ckwarn_common()
Nicholas Clark [Tue, 13 Oct 2009 10:41:36 +0000 (11:41 +0100)]
Migrate common code in Perl_ckwarn() and Perl_ckwarn_d() to S_ckwarn_common()

12 years agoIn Perl_ckwarn() and Perl_ckwarn_d() disentangle the complex conditionals.
Nicholas Clark [Tue, 13 Oct 2009 10:30:14 +0000 (11:30 +0100)]
In Perl_ckwarn() and Perl_ckwarn_d() disentangle the complex conditionals.

It's much easier to see what is going on, if we use multiple return statements.

12 years agoReplace Perl_deprecate() with a macro that calls Perl_ck_warner()
Nicholas Clark [Tue, 13 Oct 2009 10:05:49 +0000 (11:05 +0100)]
Replace Perl_deprecate() with a macro that calls Perl_ck_warner()

Perl_deprecate was not part of the public API, and did not have a deprecate()
shortcut macro defined without -DPERL_CORE. Neither nor
CPAN::Unpack show any users outside the core.

12 years agoMove the code related to deprecate("comma-less variable list") into a function.
Nicholas Clark [Tue, 13 Oct 2009 09:27:36 +0000 (10:27 +0100)]
Move the code related to deprecate("comma-less variable list") into a function.

This makes all calls to deprecate() string constants.

12 years agoRemove category 'syntax' from 5 warnings that should just be in 'deprecated'.
Nicholas Clark [Tue, 13 Oct 2009 06:37:35 +0000 (07:37 +0100)]
Remove category 'syntax' from 5 warnings that should just be in 'deprecated'.

None were documented as also being in 'syntax'. Effectively, this completes the
reorganisation of commits 12bcd1a617c74d6e and 9014280dc8264580. See and

12 years agoUse deprecate() in place of deprecate_old(), and remove S_deprecate_old().
Nicholas Clark [Tue, 13 Oct 2009 06:12:04 +0000 (07:12 +0100)]
Use deprecate() in place of deprecate_old(), and remove S_deprecate_old().

12 years agockWARNreg() and ckWARNregdep() now use string concatenation, like their brethren
Nicholas Clark [Mon, 12 Oct 2009 17:48:38 +0000 (18:48 +0100)]
ckWARNreg() and ckWARNregdep() now use string concatenation, like their brethren

They had been concatenating "%s" REPORT_LOCATION, as they weren't passing in a
format string, which wasn't consistent with the 2-5 argument versions. None of
the strings passed in have % characters in them, so this is safe (and any static
analyser will be able to see this).

12 years agoReplace "%s"/SvPVX(sv_name) with "%"SVf/SVfARG(sv_name), which is UTF-8 safe.
Nicholas Clark [Mon, 12 Oct 2009 17:32:12 +0000 (18:32 +0100)]
Replace "%s"/SvPVX(sv_name) with "%"SVf/SVfARG(sv_name), which is UTF-8 safe.

12 years agoIn regcomp.c, where possible replace vWARN* with ckWARN*, using Perl_ck_warner.
Nicholas Clark [Mon, 12 Oct 2009 16:59:49 +0000 (17:59 +0100)]
In regcomp.c, where possible replace vWARN* with ckWARN*, using Perl_ck_warner.

This folds many pairs of ckWARN*() && Perl_warner() calls into singles call to
Perl_ck_warner(). vWARN(), vWARNdep() and vWARN2() are no longer used, so are

12 years agooverload no longer implicitly unsets fallback on repeated 'use overload' lines -...
Geoffrey T. Dairiki [Wed, 5 Aug 2009 00:54:34 +0000 (17:54 -0700)]
overload no longer implicitly unsets fallback on repeated 'use overload' lines - Fix for RT#68916

Subject: overload::import resets the setting of 'fallback'
Date: Tue, 04 Aug 2009 17:54:34 -0700
From: "Geoffrey T. Dairiki" <dairiki at>

12 years agoRemove Test-Simple's 00compile.t from core
Steve Hay [Mon, 12 Oct 2009 16:21:05 +0000 (17:21 +0100)]
Remove Test-Simple's 00compile.t from core

It required a hand-edited version of Test-Simple's MANIFEST due to the
exclusion of a .pm file from core, but we shouldn't have edited versions
of CPAN files in core. The presence of the MANIFEST file also caused
warnings from the Makefile.PL due to other files being excluded from core,
so rather than further editing of the MANIFEST it is simplest to just
remove it and the test that requires it.

12 years agoAdd Perl_ck_warner_d(), which combines Perl_ckwarn_d() and Perl_warner().
Nicholas Clark [Mon, 12 Oct 2009 15:39:02 +0000 (16:39 +0100)]
Add Perl_ck_warner_d(), which combines Perl_ckwarn_d() and Perl_warner().

Replace ckWARN_d{,2,3,4}() && Perl_warner() with it, which trades reduced code
size for 1 more function call if warnings are not enabled.

12 years agoAdd Perl_ck_warner(), which combines Perl_ckwarn() and Perl_warner().
Nicholas Clark [Mon, 12 Oct 2009 15:06:20 +0000 (16:06 +0100)]
Add Perl_ck_warner(), which combines Perl_ckwarn() and Perl_warner().

Replace ckWARN{,2,3,4}() && Perl_warner() with it, which trades reduced code
size (about 0.2%), for 1 more function call if warnings are not enabled.
However, if we're now in the L1 or L2 cache when we weren't previously, that's
still going to be a speed win.

12 years agoNormalise 3 fatal pack/unpack error messages to "panic: %s"
Nicholas Clark [Mon, 12 Oct 2009 14:54:30 +0000 (15:54 +0100)]
Normalise 3 fatal pack/unpack error messages to "panic: %s"

12 years agoIn uni_to_bytes(), fix ckWARN() anomaly.
Nicholas Clark [Mon, 12 Oct 2009 14:28:29 +0000 (15:28 +0100)]
In uni_to_bytes(), fix ckWARN() anomaly.

12 years agoRe-order a conditional with ckWARN() last, as it is a function call.
Nicholas Clark [Mon, 12 Oct 2009 13:53:52 +0000 (14:53 +0100)]
Re-order a conditional with ckWARN() last, as it is a function call.

12 years agoversion's t/survey_locales is not in core so mark it as EXCLUDED
Steve Hay [Mon, 12 Oct 2009 10:43:08 +0000 (11:43 +0100)]
version's t/survey_locales is not in core so mark it as EXCLUDED

12 years agodeprecate_old() is not public, and only used within toke.c, so can be static.
Nicholas Clark [Mon, 12 Oct 2009 10:24:15 +0000 (11:24 +0100)]
deprecate_old() is not public, and only used within toke.c, so can be static.

12 years agoIn Perl_ckwarn(), simplify the check of isLEXWARN_on/isLEXWARN_off.
Nicholas Clark [Mon, 12 Oct 2009 10:03:42 +0000 (11:03 +0100)]
In Perl_ckwarn(), simplify the check of isLEXWARN_on/isLEXWARN_off.

12 years agoDon't install
Rafael Garcia-Suarez [Mon, 12 Oct 2009 08:30:42 +0000 (10:30 +0200)]
Don't install

This file was getting incorrectly installed with perl since commit
2adbc9b6919cad1240a834325b355e9b2d5efd67, that moved it from t/ to ext/.

12 years agoImplement TODO support locally in t/comp/opsubs.t
Steve Hay [Sun, 11 Oct 2009 23:30:06 +0000 (00:30 +0100)]
Implement TODO support locally in t/comp/opsubs.t

Removes the FAILED for test 22 on Win32.

12 years agoFix failed() calls in t/comp
Steve Hay [Sun, 11 Oct 2009 23:09:51 +0000 (00:09 +0100)]
Fix failed() calls in t/comp

12 years agoTry for smarter (more portable) skip criteria in Encode's piconv.t.
Craig A. Berry [Sun, 11 Oct 2009 22:56:23 +0000 (17:56 -0500)]
Try for smarter (more portable) skip criteria in Encode's piconv.t.

We already differ from upstream, which skips when PERL_CORE is set.
Since we apparently think we can run it in core now, it seems like
we ought to sort out the proper skip criteria before sending it

12 years agoNo longer need to skip t/comp/opsubs.t #22 on VMS.
Craig A. Berry [Sun, 11 Oct 2009 22:45:02 +0000 (17:45 -0500)]
No longer need to skip t/comp/opsubs.t #22 on VMS.

And it's a good thing since the skip infrastructure was pulled
out from under us with 6b077bebea000af2a5477d50d7604bab33ee75c2.

The Win32 TODO probably still doesn't work and will need to be
implemented locally without the require of

12 years agoporting/manifest.t now tests that all files listed in MANIFEST are present.
Nicholas Clark [Sun, 11 Oct 2009 17:56:01 +0000 (18:56 +0100)]
porting/manifest.t now tests that all files listed in MANIFEST are present.

12 years agoDefault to running tests in cpan/ with relative paths for perl and @INC
Nicholas Clark [Sun, 11 Oct 2009 16:28:02 +0000 (17:28 +0100)]
Default to running tests in cpan/ with relative paths for perl and @INC

Explicitly turn paths absolute for the 33 extensions in cpan/ that fail tests
with relative paths.

12 years agoRun tests in ext/ and dist/ with relative paths for perl and @INC.
Nicholas Clark [Sun, 11 Oct 2009 14:12:15 +0000 (15:12 +0100)]
Run tests in ext/ and dist/ with relative paths for perl and @INC.

Also, as only tests in cpan/ are using %no_abs and %temp_no_core, only consult
these look-up hashes for tests in cpan/

12 years agoSwap the chdir and the call to perl_lib(), to work with relative paths in @INC.
Nicholas Clark [Sun, 11 Oct 2009 14:11:06 +0000 (15:11 +0100)]
Swap the chdir and the call to perl_lib(), to work with relative paths in @INC.

12 years agoMakeMaker::Test::Utils::perl_lib now copes with relative paths for core testing.
Nicholas Clark [Sun, 11 Oct 2009 14:05:58 +0000 (15:05 +0100)]
MakeMaker::Test::Utils::perl_lib now copes with relative paths for core testing.

In the core, @INC already contains the moral equivalent of blib/lib. However,
it's a relative path (by default), so make it absolute. It's easier to KISS if
this is done *before* any change of directory, so document this, and change the
non-core case to add the absolute path of 'blib/lib' to @INC, rather than the
absolute path of '../blib/lib'.

12 years agoRemove core @INC setting and chdir boilerplate from Thread::Semaphore's tests.
Nicholas Clark [Sun, 11 Oct 2009 13:09:02 +0000 (14:09 +0100)]
Remove core @INC setting and chdir boilerplate from Thread::Semaphore's tests.

12 years agoRemove core @INC setting and chdir boilerplate from Thread::Queue's tests.
Nicholas Clark [Sun, 11 Oct 2009 13:07:03 +0000 (14:07 +0100)]
Remove core @INC setting and chdir boilerplate from Thread::Queue's tests.

12 years agoRemove core @INC setting and chdir boilerplate from SelfLoader's tests.
Nicholas Clark [Sun, 11 Oct 2009 13:06:22 +0000 (14:06 +0100)]
Remove core @INC setting and chdir boilerplate from SelfLoader's tests.

12 years agoRemove core @INC setting boilerplate from lib's test.
Nicholas Clark [Sun, 11 Oct 2009 13:05:13 +0000 (14:05 +0100)]
Remove core @INC setting boilerplate from lib's test.

12 years agoCorrect the location of the DISTRIBUTION tarball for Parse::CPAN::Meta
Nicholas Clark [Sun, 11 Oct 2009 10:05:30 +0000 (11:05 +0100)]
Correct the location of the DISTRIBUTION tarball for Parse::CPAN::Meta

Possibly we shouldn't store this, if it can be determined unambiguously from the
PAUSE index files.

12 years agoExclude CGI/t/fast.t from the core, as it relies on FCGI, which is not in core.
Nicholas Clark [Sun, 11 Oct 2009 09:19:03 +0000 (10:19 +0100)]
Exclude CGI/t/fast.t from the core, as it relies on FCGI, which is not in core.

Whilst the test is designed to skip if FCGI is not present, that logic is now
broken, and even if it were fixed, the as-implemented skip count is now wrong.
We're never going to run this test, as core tests run with just the core's
libraries in @INC, so I see no reason to ship it.

12 years agoCorrect the name of subtest.t, added in 2c4d5b9bac4cc261
Nicholas Clark [Sun, 11 Oct 2009 07:58:13 +0000 (08:58 +0100)]
Correct the name of subtest.t, added in 2c4d5b9bac4cc261

There isn't actually an "all files in MANIFEST are present" test in t/porting.
Right now we rely on ./Configure on *nix doing it, which doesn't work if you
aren't on *nix, or are, but don't re-run Configure. Maybe we should add one.

12 years agoCorrect the listed names of the two files uudecoded in commit 7b742a041dc3ba8c
Nicholas Clark [Sun, 11 Oct 2009 07:41:53 +0000 (08:41 +0100)]
Correct the listed names of the two files uudecoded in commit 7b742a041dc3ba8c

It removed the '.uu' from the end of the name, but didn't change the preceding
'_' to '.'. They'd been changed from '.' to avoid having multiple '.'s in the
name, as some VMS filesystems don't like multiple '.'s. It's not easy being

12 years agoDarwin's groups can have spaces in them
Josh ben Jore [Sun, 11 Oct 2009 05:36:52 +0000 (22:36 -0700)]
Darwin's groups can have spaces in them

12 years agoFix Test-Simple's new 00compile.t test in core
Steve Hay [Sun, 11 Oct 2009 02:11:15 +0000 (03:11 +0100)]
Fix Test-Simple's new 00compile.t test in core

It was failing even with its MANIFEST file now in place because that
MANIFEST listed an IO::Scalar file that is EXCLUDED from core.

12 years agoAdd Test-Simple's MANIFEST file
Steve Hay [Sat, 10 Oct 2009 13:46:56 +0000 (14:46 +0100)]
Add Test-Simple's MANIFEST file

It is needed by the new 00compile.t test script. We don't normally include
CPAN distributions' MANIFEST files in core, but there is an existing
precedent for this in ExtUtils-MakeMaker.

12 years agoRemove test boilerpates from ExtUtils-CBuilder tests
Steve Hay [Sat, 10 Oct 2009 13:40:26 +0000 (14:40 +0100)]
Remove test boilerpates from ExtUtils-CBuilder tests

I was obviously wrong to restore in 1051afdbc0a9e60b8f65ed22fefafd8ecf771171
what Nicholas had removed in 4d984b75f1859fc27f556ee42f31a04df42a2004 because
the tests don't work in core with the boilerplates in place.
Don't undo that which has been done by Nicholas!

12 years agoCorrect MANIFEST, broken by Test-Simple upgrade
Steve Hay [Sat, 10 Oct 2009 13:38:16 +0000 (14:38 +0100)]
Correct MANIFEST, broken by Test-Simple upgrade

12 years agoUpgrade to Test-Simple-0.94
Steve Hay [Sat, 10 Oct 2009 12:11:04 +0000 (13:11 +0100)]
Upgrade to Test-Simple-0.94

12 years agoUpgrade to Storable-2.21
Steve Hay [Sat, 10 Oct 2009 11:57:52 +0000 (12:57 +0100)]
Upgrade to Storable-2.21

Retained local changes to *.t (mostly test boilerplates) since blead
is canonical for this distribution

12 years agoUpgrade to Pod-Simple-3.08
Steve Hay [Sat, 10 Oct 2009 11:47:30 +0000 (12:47 +0100)]
Upgrade to Pod-Simple-3.08

12 years agoSync Pod-Simple test scripts with CPAN versions
Steve Hay [Sat, 10 Oct 2009 11:43:02 +0000 (12:43 +0100)]
Sync Pod-Simple test scripts with CPAN versions

Retained local changes in:
corpus.t, reinit.t, search20.t, search22.t, search50.t
because they look more significant

12 years agoUpgrade to Parse-CPAN-Meta-1.40
Steve Hay [Sat, 10 Oct 2009 11:31:12 +0000 (12:31 +0100)]
Upgrade to Parse-CPAN-Meta-1.40

12 years agoUpgrade to ExtUtils-Manifest-1.57
Steve Hay [Sat, 10 Oct 2009 11:24:52 +0000 (12:24 +0100)]
Upgrade to ExtUtils-Manifest-1.57

12 years agoUpgrade to ExtUtils-CBuilder-0.260301
Steve Hay [Sat, 10 Oct 2009 11:22:44 +0000 (12:22 +0100)]
Upgrade to ExtUtils-CBuilder-0.260301

12 years agoSync ExtUtils-CBuilder test scripts with CPAN versions
Steve Hay [Sat, 10 Oct 2009 11:20:43 +0000 (12:20 +0100)]
Sync ExtUtils-CBuilder test scripts with CPAN versions

This restores the test boilerplates that were removed when ExtUtils-CBuilder
was moved from lib/ to ext/ (4d984b75f1859fc27f556ee42f31a04df42a2004)

12 years agoUpgrade to
Steve Hay [Sat, 10 Oct 2009 11:05:09 +0000 (12:05 +0100)]
Upgrade to

12 years agoReplace UU encoded files in CGI with their binary originals
Steve Hay [Sat, 10 Oct 2009 10:55:42 +0000 (11:55 +0100)]
Replace UU encoded files in CGI with their binary originals

There is no longer any need to avoid having binary files in the perl
distribution, and these files are not UU encoded in CGI on CPAN.

12 years agoMove CGI's Changes file to its proper place
Steve Hay [Sat, 10 Oct 2009 10:50:41 +0000 (11:50 +0100)]
Move CGI's Changes file to its proper place

12 years agoApplied a patch from David Fifield to fix an error message in perlipc.pod
Jesse Vincent [Sat, 10 Oct 2009 20:48:38 +0000 (16:48 -0400)]
Applied a patch from David Fifield to fix an error message in perlipc.pod

Subject: Comment doesn't match code in perlipc.pod.
Date: Thu, 30 Jul 2009 07:31:48 -0600
From: David Fifield <>

12 years agoAdded a TODO test for RT#69332 - PerlIO returns undef instead of 0 on EOL
Jesse Vincent [Sat, 10 Oct 2009 20:15:35 +0000 (16:15 -0400)]
Added a TODO test for RT#69332 - PerlIO returns undef instead of 0 on EOL

12 years ago[perl #69686] Fix DragonFly thread lib selection
Simon 'corecode' Schubert [Sat, 10 Oct 2009 16:55:26 +0000 (17:55 +0100)]
[perl #69686] Fix DragonFly thread lib selection


  attached patch fixes the thread library detection in DragonFly.  All
  versions of DragonFly can be used threaded with -pthread.  libc_r must
  not be used directly.

  Please let me know if you have any questions.


12 years agoMake _quote_args() return a result, rather than passing it a scalar reference.
Nicholas Clark [Fri, 9 Oct 2009 20:12:16 +0000 (22:12 +0200)]
Make _quote_args() return a result, rather than passing it a scalar reference.

This was the only use of scalar references in, and the only use of \
to create a reference to a scalar. We are now more robust in the face of parser
or runtime bugs.

12 years agoAvoid using ++, op= and anon hash constructors in the testing code.
Nicholas Clark [Fri, 9 Oct 2009 20:03:17 +0000 (22:03 +0200)]
Avoid using ++, op= and anon hash constructors in the testing code.

{} could be misparsed, ++ has a lot of internal implementation "magic" that we
don't need, but don't want to trip us up if it isn't working, and op= isn't
necessary when we already rely on the more general $a = $b op $c working.

12 years agoRefactoring tests to use is a TODO.
Nicholas Clark [Fri, 9 Oct 2009 19:21:35 +0000 (21:21 +0200)]
Refactoring tests to use is a TODO.

12 years agoAdd a test for the bootstrap rules for tests in t/
Nicholas Clark [Fri, 9 Oct 2009 18:24:56 +0000 (20:24 +0200)]
Add a test for the bootstrap rules for tests in t/

12 years agoDon't use require in comp/fold.t, as require isn't tested yet.
Nicholas Clark [Fri, 9 Oct 2009 17:18:52 +0000 (19:18 +0200)]
Don't use require in comp/fold.t, as require isn't tested yet.

Emit TAP directly.

12 years agoGive names to all tests in t/comp/fold.t
Nicholas Clark [Fri, 9 Oct 2009 17:16:35 +0000 (19:16 +0200)]
Give names to all tests in t/comp/fold.t

12 years agoMove the require './' to the end of t/comp/hints.t
Nicholas Clark [Fri, 9 Oct 2009 16:57:12 +0000 (18:57 +0200)]
Move the require './' to the end of t/comp/hints.t

Ideally tests in t/comp wouldn't use require, as require isn't tested yet, but
this test really needs runperl(), and really wants to live in t/comp/hints.t,
so place it at the end, so that any catestrophic failure only fails the last
test. We don't use any other functionality of t/

This test uses hard-coded test numbers, but I'm not convinced that it would be
correct to re-write it to use an automatically incrementing counter, as that
wouldn't fail in an obvious fashion if some compile-time blocks ran out of
order. What we have *will* fail in an informative fashion if compile time
blocks do not run correctly.

12 years agoDon't use require in comp/multiline.t, as require isn't tested yet.
Nicholas Clark [Fri, 9 Oct 2009 16:35:59 +0000 (18:35 +0200)]
Don't use require in comp/multiline.t, as require isn't tested yet.

Emit TAP directly.

12 years agoDon't use require in comp/opsubs.t, as require isn't tested yet.
Nicholas Clark [Fri, 9 Oct 2009 14:07:22 +0000 (16:07 +0200)]
Don't use require in comp/opsubs.t, as require isn't tested yet.

Emit TAP directly.

12 years agoDon't use require in comp/our.t, as require isn't tested yet.
Nicholas Clark [Fri, 9 Oct 2009 12:17:29 +0000 (14:17 +0200)]
Don't use require in comp/our.t, as require isn't tested yet.

Emit TAP directly.

12 years agoDon't use require in comp/parser.t, as require isn't tested yet.
Nicholas Clark [Fri, 9 Oct 2009 12:13:55 +0000 (14:13 +0200)]
Don't use require in comp/parser.t, as require isn't tested yet.

Emit TAP directly.

12 years agoDon't use require in comp/retainedlines.t, as require isn't tested yet.
Nicholas Clark [Fri, 9 Oct 2009 12:06:07 +0000 (14:06 +0200)]
Don't use require in comp/retainedlines.t, as require isn't tested yet.

Emit TAP directly.

12 years agoAvoid relying on prototypes working for tests to pass. They aren't tested yet.
Nicholas Clark [Fri, 9 Oct 2009 11:54:30 +0000 (13:54 +0200)]
Avoid relying on prototypes working for tests to pass. They aren't tested yet.

12 years agoDon't use require in comp/uproto.t, as require isn't tested yet.
Nicholas Clark [Fri, 9 Oct 2009 11:48:43 +0000 (13:48 +0200)]
Don't use require in comp/uproto.t, as require isn't tested yet.

Emit TAP directly.

12 years agoIn opt(), use is(..., undef) rather than ok(!defined ...)
Nicholas Clark [Fri, 9 Oct 2009 11:44:10 +0000 (13:44 +0200)]
In opt(), use is(..., undef) rather than ok(!defined ...)

12 years agoMove the test for require 5.11.0 not loading strictures to require.t from use.t
Nicholas Clark [Fri, 9 Oct 2009 10:54:29 +0000 (12:54 +0200)]
Move the test for require 5.11.0 not loading strictures to require.t from use.t