18 years agoCorrect the macros in the comments in hv.c. Given the improvements in
Nicholas Clark [Thu, 16 Mar 2006 21:52:49 +0000]
Correct the macros in the comments in hv.c. Given the improvements in
compiler optimisation, and the almost irrelevance of instructions
executed over cache misses, maybe we should just remove the comments
by putting their code back in place of the o{ptimis,bfuscat}ed version.

p4raw-id: //depot/perl@27523

18 years agoAvoid hardcoded $(OBJ_EXT)
Gisle Aas [Thu, 16 Mar 2006 21:27:00 +0000]
Avoid hardcoded $(OBJ_EXT)

p4raw-id: //depot/perl@27522

18 years agoChange 27520 was a bit too aggressive. Let's keep the warning.
Gisle Aas [Thu, 16 Mar 2006 21:20:46 +0000]
Change 27520 was a bit too aggressive.  Let's keep the warning.

p4raw-id: //depot/perl@27521

18 years agoRemove unused code. No extension use Makefile.SH any more.
Gisle Aas [Thu, 16 Mar 2006 21:14:26 +0000]
Remove unused code. No extension use Makefile.SH any more.

p4raw-id: //depot/perl@27520

18 years agoFix up miniperl dependency.
Gisle Aas [Thu, 16 Mar 2006 20:32:27 +0000]
Fix up miniperl dependency.
Depend on $(obj) when we use that to build miniperl.

p4raw-id: //depot/perl@27519

18 years agoSimplify; make perl.o included in $(obj)
Gisle Aas [Thu, 16 Mar 2006 20:24:08 +0000]
Simplify; make perl.o included in $(obj)

p4raw-id: //depot/perl@27518

18 years agoAdd a couple of regression tests for the untested
Rafael Garcia-Suarez [Thu, 16 Mar 2006 16:17:38 +0000]
Add a couple of regression tests for the untested
"Bad name after %s" parsing error

p4raw-id: //depot/perl@27517

18 years agoCompress::Raw::Zlib, Compress::Zlib, IO::Compress::Zlib 2.000_10
Paul Marquess [Thu, 16 Mar 2006 14:38:52 +0000]
Compress::Raw::Zlib, Compress::Zlib, IO::Compress::Zlib 2.000_10
From: "Paul Marquess" <paul.marquess@ntlworld.com>
Message-ID: <009401c64907$5a6ed710$4c05140a@myopwv.com>

p4raw-id: //depot/perl@27516

18 years agorequire should ignore directories found when searching @INC not just
Gisle Aas [Thu, 16 Mar 2006 12:01:10 +0000]
require should ignore directories found when searching @INC not just
die as soon as it finds one.  It should for instance be possible to
for require "File" to read the file "./File" even if there happens to
be a "File" directory in perl's standard library.

This fixes the RT #24404 fix in change 26373.

p4raw-id: //depot/perl@27515

18 years agoMake the threads tests silent with regard to the new :unique deprecation warning
Rafael Garcia-Suarez [Thu, 16 Mar 2006 08:32:28 +0000]
Make the threads tests silent with regard to the new :unique deprecation warning

p4raw-id: //depot/perl@27514

18 years agoMake the original versions of relocated paths available from
Nicholas Clark [Wed, 15 Mar 2006 23:02:49 +0000]
Make the original versions of relocated paths available from
%Config::Config with the prefix "raw_". Not sure if "raw_" is the
best choice.

p4raw-id: //depot/perl@27513

18 years agoCorrect the non gcc-brace-group definition of SvAMAGIC_off
Nicholas Clark [Wed, 15 Mar 2006 19:47:02 +0000]
Correct the non gcc-brace-group definition of SvAMAGIC_off

p4raw-id: //depot/perl@27511

18 years agoD'oh! 27508 wasn't quite testing what I wanted it to test. This does.
Nicholas Clark [Wed, 15 Mar 2006 18:47:32 +0000]
D'oh! 27508 wasn't quite testing what I wanted it to test. This does.

p4raw-id: //depot/perl@27510

18 years agoTest for reblessing objects with weak references.
Nicholas Clark [Wed, 15 Mar 2006 17:57:14 +0000]
Test for reblessing objects with weak references.

p4raw-id: //depot/perl@27508

18 years agoExpunge references to HV_AMAGIC, HV_AMAGICbad and HV_AMAGICmb which
Nicholas Clark [Wed, 15 Mar 2006 17:21:00 +0000]
Expunge references to HV_AMAGIC, HV_AMAGICbad and HV_AMAGICmb which
have been unused since revision 1.

p4raw-id: //depot/perl@27507

18 years agoMoving the overloading flag from the reference to the referant allows
Nicholas Clark [Wed, 15 Mar 2006 15:08:49 +0000]
Moving the overloading flag from the reference to the referant allows
(re)?blessing of overloaded objects to work correctly.

p4raw-id: //depot/perl@27506

18 years agoFinish converting overload.t to Test::More
Nicholas Clark [Wed, 15 Mar 2006 13:13:19 +0000]
Finish converting overload.t to Test::More

p4raw-id: //depot/perl@27505

18 years agoFurther conversion of overload.t
Nicholas Clark [Wed, 15 Mar 2006 12:43:15 +0000]
Further conversion of overload.t

p4raw-id: //depot/perl@27504

18 years agoAdditions to README.win32 and win32/Makefile based on comments from
Steve Peters [Wed, 15 Mar 2006 12:34:10 +0000]
Additions to README.win32 and win32/Makefile based on comments from
Steve Hay.

p4raw-id: //depot/perl@27503

18 years agoFurther conversion of overload.t
Nicholas Clark [Wed, 15 Mar 2006 12:29:01 +0000]
Further conversion of overload.t

p4raw-id: //depot/perl@27502

18 years agoPartially convert overload.t to Test::More
Nicholas Clark [Wed, 15 Mar 2006 12:05:27 +0000]
Partially convert overload.t to Test::More

p4raw-id: //depot/perl@27501

18 years agoThe first step in supporting VC++ 8.0. Added macros for the free .NET 2.0 SDK and...
Steve Peters [Wed, 15 Mar 2006 11:06:34 +0000]
The first step in supporting VC++ 8.0.  Added macros for the free .NET 2.0 SDK and Visual Studio
2005.  Added these macros to "if" conditions that were appropriate.  Added a fix to prevent
the .NET 1.1 and 2.0 SDK compilers from attempting to build perlglob.exe (they are both missing
the setargv.obj file necessary to do it).

p4raw-id: //depot/perl@27500

18 years agoDeprecate :unique even with threads
Rafael Garcia-Suarez [Tue, 14 Mar 2006 17:32:48 +0000]
Deprecate :unique even with threads

p4raw-id: //depot/perl@27499

18 years agoMake usage of :unique emit a deprecation warning.
Rafael Garcia-Suarez [Tue, 14 Mar 2006 17:02:38 +0000]
Make usage of :unique emit a deprecation warning.
Remove its outdated documentation.

p4raw-id: //depot/perl@27498

18 years agoperl @ 27483 fix lib/Pod/Simple/t/corpustest.t to work on VMS ODS-2
Peter Prymmer [Tue, 14 Mar 2006 10:38:00 +0000]
perl @ 27483 fix lib/Pod/Simple/t/corpustest.t to work on VMS ODS-2
Message-ID: <OF85EC8E5E.4DF3A439-ON85257131.0055870A-85257131.0055E07C@factset.com>

p4raw-id: //depot/perl@27497

18 years agoSVphv_REHASH can use the same flag bit as SVf_FAKE. This allows
Nicholas Clark [Tue, 14 Mar 2006 15:21:54 +0000]
SVphv_REHASH can use the same flag bit as SVf_FAKE. This allows
0x10000000 (SVf_AMAGIC) to be set on all 16 SV types, which seems to
be necessary to fix subtle bugs involving reblessing and overloading.

p4raw-id: //depot/perl@27496

18 years ago[perl #38710] localised stash slice
Hugo van der Sanden [Sat, 11 Mar 2006 02:47:56 +0000]
[perl #38710] localised stash slice
From: Hugo van der Sanden (via RT) <perlbug-followup@perl.org>
Message-ID: <rt-3.0.11-38710-131000.18.015529928953@perl.org>

(new TODO tests)

p4raw-id: //depot/perl@27495

18 years agothreads and threads::shared are now dual-lived modules
Rafael Garcia-Suarez [Tue, 14 Mar 2006 08:53:07 +0000]
threads and threads::shared are now dual-lived modules

p4raw-id: //depot/perl@27494

18 years agoDo nothing in a GCC-friendly way
Andy Lester [Mon, 13 Mar 2006 12:48:50 +0000]
Do nothing in a GCC-friendly way
Message-ID: <20060313184850.GA16905@petdance.com>

p4raw-id: //depot/perl@27493

18 years agolooks_like_bool can be const
Andy Lester [Mon, 13 Mar 2006 09:31:52 +0000]
looks_like_bool can be const
Message-ID: <20060313153152.GA13254@petdance.com>

p4raw-id: //depot/perl@27492

18 years agoSomewhat less shaky relocatable @INC support. You can install things
Nicholas Clark [Mon, 13 Mar 2006 23:13:23 +0000]
Somewhat less shaky relocatable @INC support. You can install things
to the current location of the perl tree, move it, install more things,
move it again, lather rinse repeat.
Configure with -Duserelocatableinc

p4raw-id: //depot/perl@27491

18 years agoFirst stab at rather shaky Configure support for relocatable @INC.
Nicholas Clark [Mon, 13 Mar 2006 20:49:55 +0000]
First stab at rather shaky Configure support for relocatable @INC.
Seems to be installing rather more that it's supposed to be.

p4raw-id: //depot/perl@27490

18 years agoSplit break out the installprefix code from setprefixvar, which allows
Nicholas Clark [Mon, 13 Mar 2006 18:59:27 +0000]
Split break out the installprefix code from setprefixvar, which allows
the remaining 8 places to call the common prefix conversion code.

p4raw-id: //depot/perl@27489

18 years agoRefactor most of the cut & paste for setting prefix/prefixexp variables
Nicholas Clark [Mon, 13 Mar 2006 18:01:11 +0000]
Refactor most of the cut & paste for setting prefix/prefixexp variables
and the "Change installation prefix, if necessary" into a new
"function" setprefixvar (which I assume will become Setprefixvar.U)
How many platforms still have Bourne shells that can't do shell
functions? Someone's going to say Irix?

p4raw-id: //depot/perl@27488

18 years agoForgot the ending \. Why did it still compile ?
Rafael Garcia-Suarez [Mon, 13 Mar 2006 17:35:59 +0000]
Forgot the ending \. Why did it still compile ?

p4raw-id: //depot/perl@27487

18 years agoAutomatically SPAGAIN in the poor-man's-multicall API
Robin Houston [Mon, 13 Mar 2006 12:48:21 +0000]
Automatically SPAGAIN in the poor-man's-multicall API
Message-ID: <20060313124821.GA11203@rpc142.cs.man.ac.uk>

p4raw-id: //depot/perl@27486

18 years agoThe existing order of the checks for whether man1dir and man3dir were
Nicholas Clark [Mon, 13 Mar 2006 17:10:14 +0000]
The existing order of the checks for whether man1dir and man3dir were
' ' were obviously buggy, because they set installman[13]dir only for
the next if block to set it again. Reordering removes that daftness.

p4raw-id: //depot/perl@27485

18 years agoThe "Change installation prefix, if necessary." code needs to use
Nicholas Clark [Mon, 13 Mar 2006 15:50:04 +0000]
The "Change installation prefix, if necessary." code needs to use
the ~ expanded prefix and installprefix in the substitution.

p4raw-id: //depot/perl@27484

18 years agoLoad .pmc always, even if they are older than a matching .pm file.
Rafael Garcia-Suarez [Sun, 12 Mar 2006 15:12:29 +0000]
Load .pmc always, even if they are older than a matching .pm file.
(This trick is going to be used by pugs.)

p4raw-id: //depot/perl@27483

18 years agoRemoved unnecessary NULL check in Digest::SHA
Andy Lester [Tue, 7 Mar 2006 12:12:35 +0000]
Removed unnecessary NULL check in Digest::SHA
Message-ID: <20060307181235.GA25688@petdance.com>

p4raw-id: //depot/perl@27482

18 years agoUpgrade to Module-Build-0.27_09
Steve Peters [Sun, 12 Mar 2006 13:48:10 +0000]
Upgrade to Module-Build-0.27_09

p4raw-id: //depot/perl@27481

18 years agoisSPACE('\0') is false, so can simplify the condition in the for
Nicholas Clark [Sat, 11 Mar 2006 23:40:34 +0000]
isSPACE('\0') is false, so can simplify the condition in the for
loop. As the for loop is basically a while loop, write it as one, to
make its intentions clearer.

p4raw-id: //depot/perl@27480

18 years agoFix the last 2 perly.y specific parts of regen_perly.pl, and make
Nicholas Clark [Sat, 11 Mar 2006 22:51:26 +0000]
Fix the last 2 perly.y specific parts of regen_perly.pl, and make
the regen_perly target call it for both perly.y and madly.y.

p4raw-id: //depot/perl@27479

18 years agoNNs and NULLOKs for MAD's new functions.
Nicholas Clark [Sat, 11 Mar 2006 22:29:56 +0000]
NNs and NULLOKs for MAD's new functions.

p4raw-id: //depot/perl@27478

18 years agoPerl_newWHILEOP() had exactly the same lack of a NULL pointer check
Steve Peters [Sat, 11 Mar 2006 22:08:34 +0000]
Perl_newWHILEOP() had exactly the same lack of a NULL pointer check
that Perl_newLOOPOP() had and fixed in change #27476.  Maybe some
refactoring is needed?
p4raw-link: @27476 on //depot/perl: 06dc7ac6fa597f9446b4a27a32d667bbcbde0453

p4raw-id: //depot/perl@27477

18 years agoPossible NULL pointer reference found by Coverity checks.
Steve Peters [Sat, 11 Mar 2006 21:41:20 +0000]
Possible NULL pointer reference found by Coverity checks.

p4raw-id: //depot/perl@27476

18 years agoKeep from using a variable in S_my_kid() before it is initialized. This was...
Steve Peters [Sat, 11 Mar 2006 20:57:26 +0000]
Keep from using a variable in S_my_kid() before it is initialized.  This        was introduced with the MAD changes.

p4raw-id: //depot/perl@27475

18 years agoperl@27437 fix File::Path::mkpath so that perl installs on VMS
Peter Prymmer [Fri, 10 Mar 2006 15:24:23 +0000]
perl@27437 fix File::Path::mkpath so that perl installs on VMS
Message-id: <OFBE8F84F1.A708DA4A-ON8525712D.006E4BF7-8525712D.007018C6@factset.com>

p4raw-id: //depot/perl@27474

18 years agoReapply change 24432
Jan Dubois [Fri, 10 Mar 2006 18:21:39 +0000]
Reapply change 24432
From: "Jan Dubois" <jand@ActiveState.com>
Message-ID: <01b001c644b2$879a1f50$6062a8c0@candy>

p4raw-id: //depot/perl@27473

18 years agoMissed the hunk that adds regen_madly.pl to the regen_perl target.
Nicholas Clark [Sat, 11 Mar 2006 10:21:27 +0000]
Missed the hunk that adds regen_madly.pl to the regen_perl target.

p4raw-id: //depot/perl@27471

18 years agoSync up embed.fnc and doio.c
Andy Lester [Sat, 11 Mar 2006 01:13:38 +0000]
Sync up embed.fnc and doio.c
Message-ID: <20060311071338.GA12490@petdance.com>
Date: Sat, 11 Mar 2006 01:13:38 -0600

p4raw-id: //depot/perl@27470

18 years agoLocalizing some vars in S_checkcomma
Andy Lester [Fri, 10 Mar 2006 16:31:00 +0000]
Localizing some vars in S_checkcomma
Message-ID: <20060310223100.GA13455@petdance.com>
Date: Fri, 10 Mar 2006 16:31:00 -0600

p4raw-id: //depot/perl@27469

18 years agoRemove the obsolete mad/P5RE.pm
Nicholas Clark [Sat, 11 Mar 2006 09:32:18 +0000]
Remove the obsolete mad/P5RE.pm

p4raw-id: //depot/perl@27468

18 years agonull terminate command string -- broken in #27438
Craig A. Berry [Sat, 11 Mar 2006 01:52:37 +0000]
null terminate command string -- broken in #27438

p4raw-id: //depot/perl@27467

18 years agoMove the 12 static MAD variables from toke.c into the interpeter
Nicholas Clark [Fri, 10 Mar 2006 21:54:18 +0000]
Move the 12 static MAD variables from toke.c into the interpeter
structure.

p4raw-id: //depot/perl@27466

18 years agoRename the 12 static variables in toke.c to PL_*
Nicholas Clark [Fri, 10 Mar 2006 20:58:23 +0000]
Rename the 12 static variables in toke.c to PL_*

p4raw-id: //depot/perl@27465

18 years agogozer took over from gsar (some time ago).
Nicholas Clark [Fri, 10 Mar 2006 19:47:11 +0000]
gozer took over from gsar (some time ago).

p4raw-id: //depot/perl@27464

18 years agoAll the autogenerated files from change 27462.
Nicholas Clark [Fri, 10 Mar 2006 18:53:01 +0000]
All the autogenerated files from change 27462.

p4raw-id: //depot/perl@27463

18 years agoMerge perly.y changes from 5.9.2 to now into madly.y.
Nicholas Clark [Fri, 10 Mar 2006 18:52:40 +0000]
Merge perly.y changes from 5.9.2 to now into madly.y.

p4raw-id: //depot/perl@27462

18 years agoTeaching embed.pl about conditional symbols is a TODO.
Nicholas Clark [Fri, 10 Mar 2006 12:57:18 +0000]
Teaching embed.pl about conditional symbols is a TODO.

p4raw-id: //depot/perl@27461

18 years agoTweak change 27457 - make PL_lasttoke MAD only, but teach makedef.pl
Nicholas Clark [Fri, 10 Mar 2006 12:09:39 +0000]
Tweak change 27457 - make PL_lasttoke MAD only, but teach makedef.pl
about MADness (and unMADness - maybe there is a special word for that)

p4raw-id: //depot/perl@27460

18 years agoSummon constman! S_checkcomma now has all 3 arguments const char.
Nicholas Clark [Fri, 10 Mar 2006 11:31:14 +0000]
Summon constman! S_checkcomma now has all 3 arguments const char.

p4raw-id: //depot/perl@27459

18 years agoTease apart the keyword/subroutine test in S_checkcomma.
Nicholas Clark [Fri, 10 Mar 2006 10:50:29 +0000]
Tease apart the keyword/subroutine test in S_checkcomma.

p4raw-id: //depot/perl@27458

18 years agoMove PL_lasttoke out of its PERL_MAD define in intrpvar.h.
Steve Peters [Fri, 10 Mar 2006 02:17:43 +0000]
Move PL_lasttoke out of its PERL_MAD define in intrpvar.h.
Otherwise, Win32 does not compile.

p4raw-id: //depot/perl@27457

18 years agoCorrect another gv_fetchpv() - it takes a bitmask of flags, not simply
Nicholas Clark [Thu, 9 Mar 2006 23:23:19 +0000]
Correct another gv_fetchpv() - it takes a bitmask of flags, not simply
TRUE/FALSE.

p4raw-id: //depot/perl@27456

18 years agoRemove two NEWSV()s in the non-ithread dump code that got missed.
Nicholas Clark [Thu, 9 Mar 2006 22:50:23 +0000]
Remove two NEWSV()s in the non-ithread dump code that got missed.

p4raw-id: //depot/perl@27455

18 years agoRe: [PATCH pod/perlfunc.pd] Re: truncate needs seek on filehandle
Ronald J. Kimball [Thu, 9 Mar 2006 16:29:12 +0000]
Re: [PATCH pod/perlfunc.pd] Re: truncate needs seek on filehandle
Message-ID: <20060309212912.GF22682@penkwe.pair.com>

p4raw-id: //depot/perl@27454

18 years agoAdd the Perl 5 to Perl 5 convertor scripts.
Nicholas Clark [Thu, 9 Mar 2006 22:03:55 +0000]
Add the Perl 5 to Perl 5 convertor scripts.

p4raw-id: //depot/perl@27453

18 years agoSuperior *printf-fu lets us avoid modifying source code.
Nicholas Clark [Thu, 9 Mar 2006 22:01:28 +0000]
Superior *printf-fu lets us avoid modifying source code.

p4raw-id: //depot/perl@27452

18 years agoAvoid scribbling on the passed in command line string for -i in
Nicholas Clark [Thu, 9 Mar 2006 21:49:20 +0000]
Avoid scribbling on the passed in command line string for -i in
moreswitches.

p4raw-id: //depot/perl@27451

18 years ago-Dmad works, albeit with some test failures:
Nicholas Clark [Thu, 9 Mar 2006 21:24:04 +0000]
-Dmad works, albeit with some test failures:
ext/B/t/f_sort.t                       20    5  25.00%  2 4-6 16
ext/B/t/optree_samples.t               20    1   5.00%  18
ext/List/Util/t/weak.t      1   256    22    1   4.55%  22
ext/Safe/t/safeops.t        0   134   354  288  81.36%  211-354
lib/warnings.t                        561    2   0.36%  3 10

p4raw-id: //depot/perl@27450

18 years agoOops. Failed to add madly.c
Nicholas Clark [Thu, 9 Mar 2006 20:53:11 +0000]
Oops. Failed to add madly.c

p4raw-id: //depot/perl@27449

18 years agoThe Coverity audit is upset that the scream_olds variable is not
Steve Peters [Thu, 9 Mar 2006 20:48:50 +0000]
The Coverity audit is upset that the scream_olds variable is not
directly initialized, although it is likely initialized after
another variable takes a reference to it.

p4raw-id: //depot/perl@27448

18 years agoAdd the madly parser. I believe that this is all the C code. Now to
Nicholas Clark [Thu, 9 Mar 2006 20:48:18 +0000]
Add the madly parser. I believe that this is all the C code. Now to
find the bugs that stowed away.

p4raw-id: //depot/perl@27447

18 years agoIt helps to run embed.pl in the correct directory. (not the symlink
Nicholas Clark [Thu, 9 Mar 2006 19:53:53 +0000]
It helps to run embed.pl in the correct directory. (not the symlink
tree)

p4raw-id: //depot/perl@27446

18 years agoThe remainder of the toke.c MAD changes. Now to investigate why MAD
Nicholas Clark [Thu, 9 Mar 2006 19:22:10 +0000]
The remainder of the toke.c MAD changes. Now to investigate why MAD
no longer builds.

p4raw-id: //depot/perl@27445

18 years agoAnother refactoring from Larry implicit in the MAD patch.
Nicholas Clark [Thu, 9 Mar 2006 19:00:26 +0000]
Another refactoring from Larry implicit in the MAD patch.

p4raw-id: //depot/perl@27444

18 years agoRefactoring from Larry implicit in the MAD patch.
Nicholas Clark [Thu, 9 Mar 2006 18:08:52 +0000]
Refactoring from Larry implicit in the MAD patch.

p4raw-id: //depot/perl@27443

18 years agoReplace PL_nextval[PL_nexttoke].opval with the macro NEXTVAL_NEXTTOKE
Nicholas Clark [Thu, 9 Mar 2006 17:06:46 +0000]
Replace PL_nextval[PL_nexttoke].opval with the macro NEXTVAL_NEXTTOKE
as MAD conditionally changes this consistently in 27 places.

p4raw-id: //depot/perl@27442

18 years agoNeed to update test count from change #27440.
Steve Peters [Thu, 9 Mar 2006 16:14:20 +0000]
Need to update test count from change #27440.
p4raw-link: @27440 on //depot/perl: 516d25e8e8c09c6c60bf2f46703fc4d5add0f5fb

p4raw-id: //depot/perl@27441

18 years agocroak in POSIX::sigaction() when passed a negative signal instead
Steve Peters [Thu, 9 Mar 2006 16:03:21 +0000]
croak in POSIX::sigaction() when passed a negative signal instead
dumping core.

p4raw-id: //depot/perl@27440

18 years agoMAD changes for bare skipspace()
Nicholas Clark [Thu, 9 Mar 2006 15:13:49 +0000]
MAD changes for bare skipspace()

p4raw-id: //depot/perl@27439

18 years agopatch@27385 - VMS thread fixes (was: threads and VMS)
John E. Malmberg [Wed, 8 Mar 2006 23:34:05 +0000]
patch@27385 - VMS thread fixes (was: threads and VMS)
From: "John E. Malmberg" <wb8tyw@qsl.net>
Message-id: <440FB03D.8010409@qsl.net>

p4raw-id: //depot/perl@27438

18 years agoAdd the MAD change to perly.y, which brings a new PEG token.
Nicholas Clark [Thu, 9 Mar 2006 13:45:30 +0000]
Add the MAD change to perly.y, which brings a new PEG token.

p4raw-id: //depot/perl@27437

18 years agoFix a memory leak in ck_grep(), spotted by coverity:
Marcus Holland-Moritz [Thu, 9 Mar 2006 12:54:19 +0000]
Fix a memory leak in ck_grep(), spotted by coverity:
perl -e'eval "grep" while 1'

p4raw-id: //depot/perl@27436

18 years agoTweaking the order of op creation/op free in Perl_ck_require to Larry's
Nicholas Clark [Thu, 9 Mar 2006 12:41:40 +0000]
Tweaking the order of op creation/op free in Perl_ck_require to Larry's
order doesn't make tests pass :-(. Nor does it make them fail.

p4raw-id: //depot/perl@27435

18 years agoFix C99ism in dump.c
Nicholas Clark [Thu, 9 Mar 2006 12:17:41 +0000]
Fix C99ism in dump.c

p4raw-id: //depot/perl@27434

18 years agoUnroll 27425 - keeping Larry's order in op.c means that exactly the
Nicholas Clark [Thu, 9 Mar 2006 12:01:36 +0000]
Unroll 27425 - keeping Larry's order in op.c means that exactly the
same sequence of pad usage is generated with and without MAD.

p4raw-id: //depot/perl@27433

18 years agoNote the weakref.t failure that commenting out this code caused.
Nicholas Clark [Thu, 9 Mar 2006 11:20:14 +0000]
Note the weakref.t failure that commenting out this code caused.
Pesky waterbeds.

p4raw-id: //depot/perl@27432

18 years agoNote the cause of another MAD test failure - an explicit abort trap
Nicholas Clark [Thu, 9 Mar 2006 11:04:23 +0000]
Note the cause of another MAD test failure - an explicit abort trap
for saving PL_formfeed is being hit.

p4raw-id: //depot/perl@27431

18 years agoBodge MAD code to get pack.t and ext/Encode/t/Encode.t passing.
Nicholas Clark [Thu, 9 Mar 2006 10:53:25 +0000]
Bodge MAD code to get pack.t and ext/Encode/t/Encode.t passing.

p4raw-id: //depot/perl@27430

18 years agoMake MAD at least compile and link once more. Although it's now failing
Nicholas Clark [Wed, 8 Mar 2006 23:33:05 +0000]
Make MAD at least compile and link once more. Although it's now failing
6 regression tests.

p4raw-id: //depot/perl@27429

18 years agoFurther MAD changes.
Nicholas Clark [Wed, 8 Mar 2006 22:36:30 +0000]
Further MAD changes.

p4raw-id: //depot/perl@27428

18 years agoOops. Didn't mean to commit 27426 to maint first. Integrate it:
Nicholas Clark [Wed, 8 Mar 2006 21:53:06 +0000]
Oops. Didn't mean to commit 27426 to maint first. Integrate it:
Avoid writing over the input string in the case 'F' in moreswitches.

p4raw-id: //depot/perl@27427
p4raw-integrated: from //depot/maint-5.8/perl@27425 'merge in' perl.c
(@27318..)

18 years agoThe last MAD change to op.c changes the pad offsets of the test that
Nicholas Clark [Wed, 8 Mar 2006 21:18:58 +0000]
The last MAD change to op.c changes the pad offsets of the test that
ext/B/t/f_map.t is using.

p4raw-id: //depot/perl@27425

18 years agoAlmost all the MAD changes for op.c
Nicholas Clark [Wed, 8 Mar 2006 20:34:16 +0000]
Almost all the MAD changes for op.c

p4raw-id: //depot/perl@27424

18 years agoChange from Larry in the MAD code which seems to be unrelated to MAD.
Nicholas Clark [Wed, 8 Mar 2006 18:17:09 +0000]
Change from Larry in the MAD code which seems to be unrelated to MAD.

p4raw-id: //depot/perl@27423

18 years agoMAD changes to dump.c
Nicholas Clark [Wed, 8 Mar 2006 16:29:26 +0000]
MAD changes to dump.c

p4raw-id: //depot/perl@27422

18 years agoGet rid of unnecessary "rmdir /s" calls in Win32 makefiles
Steve Hay [Wed, 8 Mar 2006 16:25:23 +0000]
Get rid of unnecessary "rmdir /s" calls in Win32 makefiles

They only existed for the benefit of old shells that didn't
understand "rmdir /s /q", but since change #27195 effectively
drops support for old shells that don't understand "xcopy /y"
it seems unnecessary now.

Windows NT4, 2000, XP and onwards all support "rmdir /s /q", so
it's probably only the Windows 9x shells that lose out.
p4raw-link: @27195 on //depot/perl: b6ed73147a6df30ce6f460cb8d6857100c9a83be

p4raw-id: //depot/perl@27421

18 years agoSort the ext/ and lib/ tests when running under t/harness
Steve Hay [Wed, 8 Mar 2006 15:20:28 +0000]
Sort the ext/ and lib/ tests when running under t/harness

Unless they are sorted then they are run in the same order in which
they are listed in MANIFEST, which is not always ideal. In particular,
the ext/Compress/IO/Zlib/t/*.t tests are not run in the correct order,
which causes some files to be left behind afterwards.

ExtUtils::Command::MM::test_harness() sorts test files, so it seems
sensible for t/harness to do likewise, rather than relying on the
ordering in MANIFEST.

p4raw-id: //depot/perl@27420