15 years agoResolve perl #60344: Regex lookbehind failure after an (if)then|else in perl 5.10
Yves Orton [Thu, 6 Nov 2008 10:44:13 +0000]
Resolve perl #60344: Regex lookbehind failure after an (if)then|else in perl 5.10

During the de-recursivization it looks like Dave M forgot to reset the 'logical'
flag after using it, which in turn causes UNLESSM/IFTHEN when used after a LOGICAL operator to
be incorrectly intrepreted. This change resets the logical flag after each time it is stored
in ST.logical.

p4raw-id: //depot/perl@34746

15 years agoBring ExtUtils::Install up to date with the latest on CPAN 1.50_04 and related test...
Yves Orton [Wed, 5 Nov 2008 22:47:58 +0000]
Bring ExtUtils::Install up to date with the latest on CPAN 1.50_04 and related test changes

This involves changes to shared test files used by ExtUtils::MakeMaker and ExtUtils::Install, in particular updating the core/MakeMaker version of ExtUtils/t/basic.t so that it reflects the 1.50_03 version, and then splitting it in half into basic.pl and basic_finish.pl, which are then evalled (together) by basic.t. ExtUtils::Install does somethign similar but only with basic.pl.

This also silences an odd warning from MM_Any.pm that comes up once in a while, which I believe is to do with a deeper core failure as it occurs in code utterly unrelated to these changes.

This patch needs to be pushed upstream to the ExtUtils::MakeMaker distribution on CPAN under Schwern's supervision.

In this way it will be possible to simply copy the t/basic.pl from core into ExtUtils::Install or into ExtUtils::MakeMaker or whatever.

Includes the following changes from ExtUtils::Install:

1.50_04

SVN revision 41.

Restructure tests to make it easier to maintain given it is distributed in various
ways in three different packages.

1.50_03

SVN revision 40.

Sigh, just after i released 1.50_02 I noticed that a test modified in it will fail
under VMS. So this is a fixup release for that alone.

1.50_02

SVN revision 39.

Synchronize with the changes that were made in blead perl
patch #33567. VMS changes by Craig Berry. See

http://public.activestate.com/cgi-bin/perlbrowse/p/33567

This was marked in the pod as 1.51 but not actually version bumped.

So I've marked it as 1.50_02 as a test release prior to putting it out
as the real 1.51

This release also restores the missing installed.t which was accidentally
missed by the MANIFEST having a duplicate entry for install.t instead.
Probably something should have warned about this, but I haven't worked out
what.

Includes changes from Activestate/ActivePerl:

- To make installation less chatty when not under verbose mode. See

http://rt.cpan.org/Public/Bug/Display.html?id=5903

- To install HTML documentation files under builds that set $Config{installhtmldir}
(and presumably also create HTML versions of the pod -- which is quite nice actually :-)

http://rt.cpan.org/Ticket/Display.html?id=37727

p4raw-id: //depot/perl@34742

15 years agoUpgrade to File::Path 2.06_08.
Nicholas Clark [Wed, 5 Nov 2008 00:29:57 +0000]
Upgrade to File::Path 2.06_08.

p4raw-id: //depot/perl@34733

15 years ago${^CHILD_ERROR_NATIVE} isn't mentioned in perlrun.pod, but is in
Nicholas Clark [Tue, 4 Nov 2008 08:03:04 +0000]
${^CHILD_ERROR_NATIVE} isn't mentioned in perlrun.pod, but is in
perlvar.pod.

p4raw-id: //depot/perl@34714

15 years agoThe breakage caused by #34653 (sh|c)ould have been detected
Marcus Holland-Moritz [Mon, 3 Nov 2008 22:25:09 +0000]
The breakage caused by #34653 (sh|c)ould have been detected
by the compiler. So further improve the STR_WITH_LEN() macro.

p4raw-id: //depot/perl@34712

15 years agoFix DBM module handling broken by #34653.
Marcus Holland-Moritz [Mon, 3 Nov 2008 22:19:47 +0000]
Fix DBM module handling broken by #34653.

p4raw-id: //depot/perl@34710

15 years agoRe: @{"_<$filename"} is unreasonably tied to use of DB::DB ($^P & 0x2)
Tim Bunce [Mon, 3 Nov 2008 16:25:37 +0000]
Re: @{"_<$filename"} is unreasonably tied to use of DB::DB ($^P & 0x2)
Message-ID: <20081103162537.GB45762@timac.local>
Date: Mon, 3 Nov 2008 16:25:37 +0000

p4raw-id: //depot/perl@34706

15 years agoRe: @{"_<$filename"} is unreasonably tied to use of DB::DB ($^P & 0x2)
Tim Bunce [Mon, 3 Nov 2008 16:01:31 +0000]
Re: @{"_<$filename"} is unreasonably tied to use of DB::DB ($^P & 0x2)
Message-ID: <20081103160130.GA45762@timac.local>
Date: Mon, 3 Nov 2008 16:01:31 +0000

p4raw-id: //depot/perl@34705

15 years ago[perl #60300] doc Carp $Carp::Internal{__PACKAGE__} example
Kevin Ryde [Sat, 1 Nov 2008 16:27:09 +0000]
[perl #60300] doc Carp $Carp::Internal{__PACKAGE__} example
From: Kevin Ryde (via RT) <perlbug-followup@perl.org>
Message-ID: <rt-3.6.HEAD-29762-1225582029-1073.60300-75-0@perl.org>

p4raw-id: //depot/perl@34703

15 years agoRe: CPAN-1.9301 can't clean bootstrap .cpan
Andreas J. Koenig [Sun, 2 Nov 2008 11:22:29 +0000]
Re: CPAN-1.9301 can't clean bootstrap .cpan
Message-ID: <87abci5swq.fsf@k75.linux.bogus>
Date: Sun, 02 Nov 2008 11:22:29 +0100

p4raw-id: //depot/perl@34702

15 years agoSilence one more format warning
Rafael Garcia-Suarez [Mon, 3 Nov 2008 09:13:16 +0000]
Silence one more format warning

p4raw-id: //depot/perl@34700

15 years agoUpgrade to Time::Local 1.1901.
Nicholas Clark [Sun, 2 Nov 2008 21:41:07 +0000]
Upgrade to Time::Local 1.1901.

p4raw-id: //depot/perl@34699

15 years agoPATCH: Large omnibus patch to clean up the JRRT quotes
Tom Christiansen [Sun, 2 Nov 2008 01:43:39 +0000]
PATCH: Large omnibus patch to clean up the JRRT quotes
Message-ID: <25940.1225611819@chthon>
Date: Sun, 02 Nov 2008 01:43:39 -0600

p4raw-id: //depot/perl@34698

15 years ago[perl #56526] m/a{1,0}/ compiles but doesn't match a literal string
Michael Cartmell [Wed, 2 Jul 2008 05:17:58 +0000]
[perl #56526] m/a{1,0}/ compiles but doesn't match a literal string
From: Michael Cartmell (via RT) <perlbug-followup@perl.org>
Message-ID: <rt-3.6.HEAD-27577-1215001078-1211.56526-75-0@perl.org>

p4raw-id: //depot/perl@34697

15 years ago[perl #59652][DOC PATCH] "Too late for "-CS" option"
Renee Baecker [Tue, 7 Oct 2008 08:17:22 +0000]
[perl #59652][DOC PATCH] "Too late for "-CS" option"
Message-ID: <48EAFEF2.8030403@smart-websolutions.de>

(tweaked)

p4raw-id: //depot/perl@34696

15 years agoexplicit empty while loops
Robin Barker [Wed, 29 Oct 2008 13:22:04 +0000]
explicit empty while loops
From: "Robin Barker" <Robin.Barker@npl.co.uk>
Message-ID: <46A0F33545E63740BC7563DE59CA9C6D4E2FD9@exchsvr2.npl.ad.local>

p4raw-id: //depot/perl@34695

15 years agoExplicitly specify some printf formats for constant strings.
Rafael Garcia-Suarez [Sun, 2 Nov 2008 16:50:12 +0000]
Explicitly specify some printf formats for constant strings.
This is mostly to silence gcc's warning, "format not a string
literal and no format arguments".

p4raw-id: //depot/perl@34694

15 years agoAdd a flag PERLDBf_SAVESRC, which enables the saved lines part of
Nicholas Clark [Sat, 1 Nov 2008 14:51:05 +0000]
Add a flag PERLDBf_SAVESRC, which enables the saved lines part of
PERLDBf_LINE, so that profilers (such as NYTProf) have access to the
lines of the eval, without the speed impact of other parts of the
debugger infrastructure. PERLDBf_LINE is unchanged. Based largely on a
patch by Tim Bunce in <20081028152749.GA12500@timac.local>

p4raw-id: //depot/perl@34693

15 years agoUpgrade to Text::ParseWords 3.27
Nicholas Clark [Fri, 31 Oct 2008 20:23:32 +0000]
Upgrade to Text::ParseWords 3.27

p4raw-id: //depot/perl@34691

15 years agoAdd MUTABLE_GV(), and eliminate (V *) casts in *.c.
Nicholas Clark [Fri, 31 Oct 2008 08:55:20 +0000]
Add MUTABLE_GV(), and eliminate (V *) casts in *.c.
Can't easily do gv.h, as GvGP() (at least) needs to split into two
macros - one const for reading, one non-const for writing.

p4raw-id: //depot/perl@34679

15 years agoChange 34672 should have removed a const from re.xs. I thought I'd
Nicholas Clark [Fri, 31 Oct 2008 07:37:44 +0000]
Change 34672 should have removed a const from re.xs. I thought I'd
checked for warnings here - not sure whether it was me or gcc that
didn't notice anythign wrong. g++ certainly did.

p4raw-id: //depot/perl@34678

15 years agosizeof(const SV *) is the same as sizeof(SV *), except that it doesn't
Nicholas Clark [Fri, 31 Oct 2008 07:34:02 +0000]
sizeof(const SV *) is the same as sizeof(SV *), except that it doesn't
match my regexp for non-const casts.

p4raw-id: //depot/perl@34677

15 years agoUpgrade to Devel::PPPort 3.14_05
Marcus Holland-Moritz [Fri, 31 Oct 2008 07:23:54 +0000]
Upgrade to Devel::PPPort 3.14_05

p4raw-id: //depot/perl@34676

15 years agoEliminate (SV *) casts from the rest of *.c, picking up one (further)
Nicholas Clark [Thu, 30 Oct 2008 23:22:31 +0000]
Eliminate (SV *) casts from the rest of *.c, picking up one (further)
erroneous const in dump.c.

p4raw-id: //depot/perl@34675

15 years agoChange 34627 didn't realise that if S_find_uninit_var() takes a
Nicholas Clark [Thu, 30 Oct 2008 22:27:16 +0000]
Change 34627 didn't realise that if S_find_uninit_var() takes a
const SV *, then so can Perl_report_uninit().

p4raw-id: //depot/perl@34673

15 years agoSvPV() does not take a const SV*, which means that the pattern argument
Nicholas Clark [Thu, 30 Oct 2008 22:08:41 +0000]
SvPV() does not take a const SV*, which means that the pattern argument
to Perl_re_compile() can't be const, which means that the pattern
argument to Perl_pregcomp() can't be const, as can't the argument in
the function in the regexp engine structure.

It's a shame that no-one spotted this earlier.
(Again) I may have rendered the documentation inaccurate.

p4raw-id: //depot/perl@34672

15 years agoEliminate (SV *) casts from sv.c and [tu]*.c, except for the cast in
Nicholas Clark [Thu, 30 Oct 2008 21:39:14 +0000]
Eliminate (SV *) casts from sv.c and [tu]*.c, except for the cast in
Perl_sv_magicext(), which is documented.

p4raw-id: //depot/perl@34671

15 years agoRe: [PATCH] [doc] be more explicit about magic argv
Moritz Lenz [Thu, 30 Oct 2008 20:30:00 +0000]
Re: [PATCH] [doc] be more explicit about magic argv
Message-ID: <20081030182902.25398.qmail@lists.develooper.com>

p4raw-id: //depot/perl@34670

15 years agoUpgrade to Devel::PPPort 3.14_04
Marcus Holland-Moritz [Thu, 30 Oct 2008 18:55:04 +0000]
Upgrade to Devel::PPPort 3.14_04

p4raw-id: //depot/perl@34669

15 years agoTry to demangle the mess created by 34667 (the "resubmittal" was actually
Craig A. Berry [Thu, 30 Oct 2008 18:17:04 +0000]
Try to demangle the mess created by 34667 (the "resubmittal" was actually
not at all like the original patch I tested).

p4raw-id: //depot/perl@34668

15 years agoRe: patch@34561 VMS exec handling / cwd realpath fixes
John E. Malmberg [Wed, 29 Oct 2008 22:21:38 +0000]
Re: patch@34561 VMS exec handling / cwd realpath fixes
From: "John E. Malmberg" <wb8tyw@qsl.net>
Message-id: <49092842.8090805@qsl.net>

Convert symlink target to UNIX format on VMS. (Cwd changes not
included here.)

p4raw-id: //depot/perl@34667

15 years ago[patch@34600] vms.c setup_cmddsc
John E. Malmberg [Wed, 29 Oct 2008 22:02:43 +0000]
[patch@34600] vms.c setup_cmddsc
From: "John E. Malmberg" <wb8tyw@qsl.net>
Message-id: <490923D3.90407@qsl.net>

Remove trailing dot when efs_charset is in effect.

p4raw-id: //depot/perl@34666

15 years agoAdd a test for bug #60214
Rafael Garcia-Suarez [Thu, 30 Oct 2008 11:34:05 +0000]
Add a test for bug #60214

p4raw-id: //depot/perl@34661

15 years agoThe file test operators -s, -A, -C and -M can return false values
Rafael Garcia-Suarez [Thu, 30 Oct 2008 11:29:50 +0000]
The file test operators -s, -A, -C and -M can return false values
that are not undef or "no". Take that into account when stacking them.
This fixes bug #60214.

p4raw-id: //depot/perl@34660

15 years agoAdd MUTABLE_IO(), and eliminate (IO *) casts in *.c.
Nicholas Clark [Wed, 29 Oct 2008 23:20:01 +0000]
Add MUTABLE_IO(), and eliminate (IO *) casts in *.c.

p4raw-id: //depot/perl@34654

15 years agoUse pvs macros instead of pvn where possible.
Marcus Holland-Moritz [Wed, 29 Oct 2008 23:04:34 +0000]
Use pvs macros instead of pvn where possible.

p4raw-id: //depot/perl@34653

15 years agoEliminate (AV *) casts in *.c.
Nicholas Clark [Wed, 29 Oct 2008 22:45:15 +0000]
Eliminate (AV *) casts in *.c.

p4raw-id: //depot/perl@34650

15 years agoRemove redundant (and wrong) prototype for pv_display()
Marcus Holland-Moritz [Wed, 29 Oct 2008 22:15:33 +0000]
Remove redundant (and wrong) prototype for pv_display()
from its apidoc section.

p4raw-id: //depot/perl@34649

15 years agoFwd: CPAN Upload: S/SA/SAPER/constant-1.17.tar.gz
Sébastien Aperghis-Tramoni [Wed, 29 Oct 2008 22:58:52 +0000]
Fwd: CPAN Upload: S/SA/SAPER/constant-1.17.tar.gz
Message-Id: <93213472-1B76-43A2-AACB-2BBFF901C5A7@free.fr>
Date: Wed, 29 Oct 2008 22:58:52 +0100

p4raw-id: //depot/perl@34648

15 years agoAdd MUTABLE_CV(), and eliminate (CV *) casts in *.c.
Nicholas Clark [Wed, 29 Oct 2008 21:57:34 +0000]
Add MUTABLE_CV(), and eliminate (CV *) casts in *.c.

p4raw-id: //depot/perl@34647

15 years agoAs Perl_cv_const_sv() now takes a const CV *, use that as the cast for
Nicholas Clark [Wed, 29 Oct 2008 21:36:17 +0000]
As Perl_cv_const_sv() now takes a const CV *, use that as the cast for
its arguments.

p4raw-id: //depot/perl@34646

15 years agoPerl_cv_const_sv() takes a const CV *const cv argument.
Nicholas Clark [Wed, 29 Oct 2008 21:24:54 +0000]
Perl_cv_const_sv() takes a const CV *const cv argument.

p4raw-id: //depot/perl@34645

15 years agoUpgrade to PathTools 3.29.
Nicholas Clark [Wed, 29 Oct 2008 20:49:30 +0000]
Upgrade to PathTools 3.29.

p4raw-id: //depot/perl@34643

15 years agoRemove redundant API definitions from '=for apidoc' sections.
Marcus Holland-Moritz [Wed, 29 Oct 2008 20:39:43 +0000]
Remove redundant API definitions from '=for apidoc' sections.
Those are already in embed.fnc, and most of them were already
outdated. This also fixes the docs for pv_escape and pv_pretty.

p4raw-id: //depot/perl@34642

15 years agomake distclean patch from Andy Armstrong.
Nicholas Clark [Wed, 29 Oct 2008 19:41:54 +0000]
make distclean patch from Andy Armstrong.
(Distclean isn't perfect yet because there's a spurious directory tree
created but not cleaned up by ext/Test/Harness/t/harness.t, but Andy is
working on fixing the cause)
No idea about the state of play of disclean on VMS - I couldn't see how
or where to add the three extra directories to it.

p4raw-id: //depot/perl@34640

15 years agoAvoid leaving a VAR.txt after testing terminates on any platform where
Nicholas Clark [Wed, 29 Oct 2008 19:34:05 +0000]
Avoid leaving a VAR.txt after testing terminates on any platform where
you can't unlink an open file by instead opening $^X for read to get a
file handle for test purposes.

p4raw-id: //depot/perl@34639

15 years agoUpgrade to CPAN-1.9301.
Steve Peters [Wed, 29 Oct 2008 19:21:49 +0000]
Upgrade to CPAN-1.9301.

p4raw-id: //depot/perl@34638

15 years agoUpgrade to File::Path 2.06_07. (an e-mail from David)
Nicholas Clark [Wed, 29 Oct 2008 16:50:40 +0000]
Upgrade to File::Path 2.06_07. (an e-mail from David)

p4raw-id: //depot/perl@34637

15 years agoOne more for Haiku
H.Merijn Brand [Wed, 29 Oct 2008 08:31:39 +0000]
One more for Haiku

p4raw-id: //depot/perl@34634

15 years agoTwo haiku files to add
H.Merijn Brand [Wed, 29 Oct 2008 08:18:00 +0000]
Two haiku files to add

p4raw-id: //depot/perl@34631

15 years agoHaiku Port
Ingo Weinhold [Wed, 29 Oct 2008 02:25:44 +0000]
Haiku Port
Message-Id: <20081029022544.413.1@knochen-vm.localdomain>

p4raw-id: //depot/perl@34630

15 years agoEvery remaining (HV *) cast in *.c
Nicholas Clark [Tue, 28 Oct 2008 22:14:26 +0000]
Every remaining (HV *) cast in *.c

p4raw-id: //depot/perl@34629

15 years agoss_dup wrapper const casting analagous to change 34626.
Nicholas Clark [Tue, 28 Oct 2008 21:59:11 +0000]
ss_dup wrapper const casting analagous to change 34626.

p4raw-id: //depot/perl@34628

15 years agoConsting in S_find_uninit_var() and the routines that it calls.
Nicholas Clark [Tue, 28 Oct 2008 21:12:09 +0000]
Consting in S_find_uninit_var() and the routines that it calls.

p4raw-id: //depot/perl@34627

15 years agoSome consting in Perl_ss_dup(), and a few (HV*) casts that can be
Nicholas Clark [Tue, 28 Oct 2008 20:56:50 +0000]
Some consting in Perl_ss_dup(), and a few (HV*) casts that can be
(const HV *).

p4raw-id: //depot/perl@34626

15 years agoEliminate (HV *) casts in u*.c.
Nicholas Clark [Tue, 28 Oct 2008 20:13:55 +0000]
Eliminate (HV *) casts in u*.c.

p4raw-id: //depot/perl@34624

15 years agoA gaggle of casts in Perl_sv_magicext() that are (const ?V *).
Nicholas Clark [Tue, 28 Oct 2008 20:13:13 +0000]
A gaggle of casts in Perl_sv_magicext() that are (const ?V *).

p4raw-id: //depot/perl@34623

15 years agoTwo casts that are MUTABLE_?V.
Nicholas Clark [Tue, 28 Oct 2008 20:12:16 +0000]
Two casts that are MUTABLE_?V.

p4raw-id: //depot/perl@34622

15 years agoA couple of (COP*)s that can actually be (const COP *)s.
Nicholas Clark [Tue, 28 Oct 2008 19:51:07 +0000]
A couple of (COP*)s that can actually be (const COP *)s.

p4raw-id: //depot/perl@34621

15 years agoinstallperl patchlet
Andreas J. Koenig [Tue, 28 Oct 2008 05:11:20 +0000]
installperl patchlet
Message-ID: <87k5btpdef.fsf@k75.linux.bogus>
Date: Tue, 28 Oct 2008 05:11:20 +0100

p4raw-id: //depot/perl@34620

15 years agoAdd MUTABLE_HV(), and remove (HV *) casts from headers.
Nicholas Clark [Tue, 28 Oct 2008 18:30:54 +0000]
Add MUTABLE_HV(), and remove (HV *) casts from headers.

p4raw-id: //depot/perl@34619

15 years agoPerl_hv_placeholders_get() actually takes a const HV *hv.
Nicholas Clark [Tue, 28 Oct 2008 18:29:36 +0000]
Perl_hv_placeholders_get() actually takes a const HV *hv.

p4raw-id: //depot/perl@34618

15 years agofix for failed Gconvert detection under C++
Tony Cook [Tue, 28 Oct 2008 22:24:05 +0000]
fix for failed Gconvert detection under C++
Message-ID: <20081028112405.GA14163@mars.tony.develop-help.com>

p4raw-id: //depot/perl@34617

15 years agoUpgrade to File::Path 2.06_06. (a diff from David via http)
Nicholas Clark [Tue, 28 Oct 2008 11:27:58 +0000]
Upgrade to File::Path 2.06_06. (a diff from David via http)

p4raw-id: //depot/perl@34615

15 years agoCorrect SvVALID() and SvTAIL() - they are actually const.
Nicholas Clark [Mon, 27 Oct 2008 22:39:45 +0000]
Correct SvVALID() and SvTAIL() - they are actually const.

p4raw-id: //depot/perl@34613

15 years agoMHX notes that I have a superfluous void cast.
Nicholas Clark [Mon, 27 Oct 2008 22:22:50 +0000]
MHX notes that I have a superfluous void cast.

p4raw-id: //depot/perl@34612

15 years agoAfter the upgrade to 3.28_03 we need to revert change 34543.
Nicholas Clark [Mon, 27 Oct 2008 22:10:26 +0000]
After the upgrade to 3.28_03 we need to revert change 34543.

p4raw-id: //depot/perl@34611

15 years agoConvert all (CV *) casts to (const CV *). Convert (XPVCV*) casts to add
Nicholas Clark [Mon, 27 Oct 2008 21:43:01 +0000]
Convert all (CV *) casts to (const CV *). Convert (XPVCV*) casts to add
MUTABLE_PTR(), to validate that there is no casting away of const.

p4raw-id: //depot/perl@34610

15 years agoUpgrade to PathTools 3.28_03.
Nicholas Clark [Mon, 27 Oct 2008 21:34:55 +0000]
Upgrade to PathTools 3.28_03.

p4raw-id: //depot/perl@34609

15 years agoAdd MUTABLE_AV(), and remove (AV *) casts from headers.
Nicholas Clark [Mon, 27 Oct 2008 21:23:04 +0000]
Add MUTABLE_AV(), and remove (AV *) casts from headers.

p4raw-id: //depot/perl@34608

15 years agoGrab the second half of r10084 for cwd.t, which wasn't in change 33042.
Nicholas Clark [Mon, 27 Oct 2008 21:01:18 +0000]
Grab the second half of r10084 for cwd.t, which wasn't in change 33042.
Update the Changes file with changes from 3.26 to 3.28_01 inclusive.

p4raw-id: //depot/perl@34607

15 years agoUpgrade Cwd.xs to the Cwd.xs in PathTools 3.26_01:
Nicholas Clark [Mon, 27 Oct 2008 20:29:54 +0000]
Upgrade Cwd.xs to the Cwd.xs in PathTools 3.26_01:

- abs_path() on Unix(ish) platforms has been upgraded to a much later
  version of the underlying C code from BSD. [Michael Schwern]

- If strlcpy() and strlcat() aren't available on the user's system,
  we now use ppport.h to provide them, so our C code works. [Steve
  Peters]

This unwinds change 32710 (consting from Robin Barker of the old code)

p4raw-id: //depot/perl@34606

15 years agoAdd a macro MUTABLE_PTR(p), which on (non-pedantic) gcc will not cast
Nicholas Clark [Mon, 27 Oct 2008 20:25:46 +0000]
Add a macro MUTABLE_PTR(p), which on (non-pedantic) gcc will not cast
away const, returning a void *. Add MUTABLE_SV(sv) which uses this, and
replace all (SV *) casts either with MUTABLE_SV(sv), or (const SV *).

This probably still needs some work - assigning to SvPVX() and SvRV()
is now likely to generate a casting error. The core doesn't do this.
But as-is it's finding bugs that can be fixed.

p4raw-id: //depot/perl@34605

15 years agoBugs revealed by replacing (SV *) casts with something that doesn't
Nicholas Clark [Mon, 27 Oct 2008 20:22:36 +0000]
Bugs revealed by replacing (SV *) casts with something that doesn't
cast away const - AvFILL() doesn't guarantee that it won't modify the
AV * passed to it. So the prototype for Perl_av_len() needs to change,
and a const needs to go in Perl_magic_setarraylen().

p4raw-id: //depot/perl@34604

15 years agoFix x2p, which was broken by #34598 with -DPERL_MEM_LOG.
Marcus Holland-Moritz [Mon, 27 Oct 2008 19:26:50 +0000]
Fix x2p, which was broken by #34598 with -DPERL_MEM_LOG.

p4raw-id: //depot/perl@34603

15 years agoFix prototype detection for C++
Marcus Holland-Moritz [Sun, 26 Oct 2008 21:41:13 +0000]
Fix prototype detection for C++
Message-ID: <20081026214113.0a5dc1a3@r2d2>

p4raw-id: //depot/perl@34602

15 years agoNo, it's not a const SV* if we subsequently call
Marcus Holland-Moritz [Sun, 26 Oct 2008 22:20:48 +0000]
No, it's not a const SV* if we subsequently call
SvREFCNT_dec() on it.

p4raw-id: //depot/perl@34600

15 years agoMake sure Perl_mem_log_(?:alloc|realloc|free) have C linkage,
Marcus Holland-Moritz [Sun, 26 Oct 2008 20:54:43 +0000]
Make sure Perl_mem_log_(?:alloc|realloc|free) have C linkage,
so modules like Digest::MD5, that are including perl.h from
within an 'extern "C"' block, will actually see them when
building with a C++ compiler.

Also make sure that Perl_mem_log_(?:new|del)_sv are only seen
by sv.c.

p4raw-id: //depot/perl@34598

15 years agoCope with brain damage in PerlIO::via, which will let you fclose() the
Nicholas Clark [Sun, 26 Oct 2008 18:37:35 +0000]
Cope with brain damage in PerlIO::via, which will let you fclose() the
same FILE * twice, thanks to it calling out to Perl space inside the
close call tree, with the underlying PerlIO * already closed, but not
unlinked.

p4raw-id: //depot/perl@34596

15 years agoUSE_FAST_STDIO can affect behaviour, so list it in -V
Nicholas Clark [Sun, 26 Oct 2008 13:32:20 +0000]
USE_FAST_STDIO can affect behaviour, so list it in -V

p4raw-id: //depot/perl@34595

15 years agoUpgrade to ExtUtils::Manifest 1.55.
Nicholas Clark [Sat, 25 Oct 2008 15:23:48 +0000]
Upgrade to ExtUtils::Manifest 1.55.

p4raw-id: //depot/perl@34591

15 years agoUpgrade to ExtUtils::Command 1.15
Nicholas Clark [Sat, 25 Oct 2008 15:06:56 +0000]
Upgrade to ExtUtils::Command 1.15

p4raw-id: //depot/perl@34590

15 years agolib/ExtUtils/t/eu_command.t belongs to ExtUtils-Command, and so does
Nicholas Clark [Sat, 25 Oct 2008 14:58:32 +0000]
lib/ExtUtils/t/eu_command.t belongs to ExtUtils-Command, and so does
the about to be committed lib/ExtUtils/t/cp.t

p4raw-id: //depot/perl@34589

15 years agoUpdate copyright year in opcode.pl to reflect change 33364.
Nicholas Clark [Sat, 25 Oct 2008 13:05:07 +0000]
Update copyright year in opcode.pl to reflect change 33364.

p4raw-id: //depot/perl@34587

15 years agoUpdate copyright year in embed.pl, and everything that it builds.
Nicholas Clark [Sat, 25 Oct 2008 12:47:01 +0000]
Update copyright year in embed.pl, and everything that it builds.

p4raw-id: //depot/perl@34586

15 years agoUpdate copyright years.
Nicholas Clark [Sat, 25 Oct 2008 12:23:01 +0000]
Update copyright years.

p4raw-id: //depot/perl@34585

15 years agoRe: [Fwd: CPAN Upload: S/SM/SMUELLER/Attribute-Handlers-0.80.tar.gz]
Steffen Mueller [Sat, 25 Oct 2008 12:04:35 +0000]
Re: [Fwd: CPAN Upload: S/SM/SMUELLER/Attribute-Handlers-0.80.tar.gz]
Message-ID: <4902EF33.7000707@sneakemail.com>
Date: Sat, 25 Oct 2008 12:04:35 +0200

p4raw-id: //depot/perl@34584

15 years agoUpgrade to lib.pm 0.61.
Nicholas Clark [Sat, 25 Oct 2008 11:23:39 +0000]
Upgrade to lib.pm 0.61.

p4raw-id: //depot/perl@34583

15 years agoIntegrate:
Vincent Pit [Thu, 23 Oct 2008 00:40:36 +0000]
Integrate:
[ 34580]
Subject: Re: [perl #60034] (?>) causes wrongness on long string
Message-ID: <48FFABE4.2030507@profvince.com>

[just the changes, as TODOs]
p4raw-link: @34580 on //depot/maint-5.8/perl: 7e8969c0412490151f26944243e2767cf32aca95

p4raw-id: //depot/perl@34581
p4raw-integrated: from //depot/maint-5.8/perl@34579 'edit in'
t/op/pat.t (@34396..)

15 years agoMake C++ compilers happy #5: jump to 'do_write' crossed
Marcus Holland-Moritz [Sat, 25 Oct 2008 06:38:00 +0000]
Make C++ compilers happy #5: jump to 'do_write' crossed
initialization in dTHX.

Plus some consting, which also makes C++ happier.

p4raw-id: //depot/perl@34579

15 years agoMake C++ compilers happy #4: move cast so Newc() and
Marcus Holland-Moritz [Fri, 24 Oct 2008 21:49:21 +0000]
Make C++ compilers happy #4: move cast so Newc() and
Renewc() will return the correct type under PERL_MEM_LOG.

p4raw-id: //depot/perl@34577

15 years agoMake C++ compilers happy #3: make implicit cast
Marcus Holland-Moritz [Fri, 24 Oct 2008 21:45:44 +0000]
Make C++ compilers happy #3: make implicit cast
from 'void *' explicit.

p4raw-id: //depot/perl@34576

15 years agoMake C++ compilers happy #2: const POD without initializer
Marcus Holland-Moritz [Fri, 24 Oct 2008 21:41:36 +0000]
Make C++ compilers happy #2: const POD without initializer
is an error for g++, even when static.

p4raw-id: //depot/perl@34575

15 years agoMake C++ compilers happy #1: "typename" is a C++ keyword.
Marcus Holland-Moritz [Fri, 24 Oct 2008 21:38:48 +0000]
Make C++ compilers happy #1: "typename" is a C++ keyword.

p4raw-id: //depot/perl@34574

15 years agoUpgrade to SelfLoader 1.17.
Nicholas Clark [Fri, 24 Oct 2008 21:00:44 +0000]
Upgrade to SelfLoader 1.17.

p4raw-id: //depot/perl@34573

15 years agoAdd PL_sv_serial to makedef.pl, so that it is skipped approriately
Nicholas Clark [Fri, 24 Oct 2008 20:43:36 +0000]
Add PL_sv_serial to makedef.pl, so that it is skipped approriately
on AIX, OS/2, Win32 etc

p4raw-id: //depot/perl@34571

15 years agoAdd PERL_ARGS_ASSERT_MEM_LOG_COMMON to S_mem_log_common().
Marcus Holland-Moritz [Fri, 24 Oct 2008 18:33:26 +0000]
Add PERL_ARGS_ASSERT_MEM_LOG_COMMON to S_mem_log_common().

p4raw-id: //depot/perl@34570

15 years agoRun 'make regen' for #34567 and #34568.
Marcus Holland-Moritz [Fri, 24 Oct 2008 16:37:49 +0000]
Run 'make regen' for #34567 and #34568.

p4raw-id: //depot/perl@34569

15 years agoAdd SV allocation tracing to -Dm and PERL_MEM_LOG
Marcus Holland-Moritz [Wed, 22 Oct 2008 01:37:31 +0000]
Add SV allocation tracing to -Dm and PERL_MEM_LOG
Message-ID: <20081022013731.23b5a2e5@r2d2>

p4raw-id: //depot/perl@34568

15 years agoRefactor Perl_mem_log_ functions
Marcus Holland-Moritz [Wed, 22 Oct 2008 01:37:21 +0000]
Refactor Perl_mem_log_ functions
Message-ID: <20081022013721.374a490c@r2d2>

p4raw-id: //depot/perl@34567