This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
perl5.git
14 years agoMerge op_pmreplstart and op_pmstash/op_pmstashpv into a union in
Nicholas Clark [Fri, 6 Apr 2007 23:31:14 +0000 (23:31 +0000)]
Merge op_pmreplstart and op_pmstash/op_pmstashpv into a union in
PMOP, as both pointers never need to be set at the same time.

p4raw-id: //depot/perl@30860

14 years agoassert() that OP_QR and OP_MATCH never set op_pmreplstart, which
Nicholas Clark [Fri, 6 Apr 2007 22:32:02 +0000 (22:32 +0000)]
assert() that OP_QR and OP_MATCH never set op_pmreplstart, which
simplifies the peep hole optimiser slightly.

p4raw-id: //depot/perl@30859

14 years agoAdd editor block.
Nicholas Clark [Fri, 6 Apr 2007 22:08:20 +0000 (22:08 +0000)]
Add editor block.

p4raw-id: //depot/perl@30858

14 years agoOnly append ?? match ops to the list used by reset. This saves memory
Nicholas Clark [Fri, 6 Apr 2007 21:53:46 +0000 (21:53 +0000)]
Only append ?? match ops to the list used by reset. This saves memory
and time.

p4raw-id: //depot/perl@30857

14 years agoAvoid accessing free()d memory when calling reset in one thread, after
Nicholas Clark [Fri, 6 Apr 2007 20:57:34 +0000 (20:57 +0000)]
Avoid accessing free()d memory when calling reset in one thread, after
deleting pattern match ops in another thread.

p4raw-id: //depot/perl@30856

14 years agoGive accurate failure diagnostics when unlike() fails.
Nicholas Clark [Fri, 6 Apr 2007 20:46:47 +0000 (20:46 +0000)]
Give accurate failure diagnostics when unlike() fails.

p4raw-id: //depot/perl@30855

14 years agoOoops. Forgot to update proto.h in change 30853.
Nicholas Clark [Fri, 6 Apr 2007 20:17:14 +0000 (20:17 +0000)]
Ooops. Forgot to update proto.h in change 30853.

p4raw-id: //depot/perl@30854

14 years agoRemove op_pmnext from PMOPs, and instead store the list for reset as
Nicholas Clark [Fri, 6 Apr 2007 19:50:12 +0000 (19:50 +0000)]
Remove op_pmnext from PMOPs, and instead store the list for reset as
an array hanging from the mg_ptr of the symbol table magic.
(Previously the linked list head was in the mg_obj member)

p4raw-id: //depot/perl@30853

14 years agoAllow any *DBM_File to work for DynaLoader testing
Jerry D. Hedden [Wed, 4 Apr 2007 09:57:16 +0000 (05:57 -0400)]
Allow any *DBM_File to work for DynaLoader testing
From: "Jerry D. Hedden" <jdhedden@1979.usna.com>
Message-ID: <1ff86f510704040657t732af552jbcbdb8804d134ff7@mail.gmail.com>

p4raw-id: //depot/perl@30852

14 years agoMAD do not ignore errors from 'use'
Gerard Goossen [Thu, 29 Mar 2007 18:02:29 +0000 (20:02 +0200)]
MAD do not ignore errors from 'use'
Message-ID: <20070329160229.GF20231@ostwald>

plus another similar correction

p4raw-id: //depot/perl@30851

14 years agomadprop about forced words
Gerard Goossen [Thu, 29 Mar 2007 18:10:42 +0000 (20:10 +0200)]
madprop about forced words
Message-ID: <20070329161042.GG20231@ostwald>

p4raw-id: //depot/perl@30850

14 years agoRe: Proposed changes and to regular expression interfaces in core
Ævar Arnfjörð Bjarmason [Thu, 29 Mar 2007 22:52:49 +0000 (22:52 +0000)]
Re: Proposed changes and to regular expression interfaces in core
From: "Ævar Arnfjörð Bjarmason" <avarab@gmail.com>
Message-ID: <51dd1af80703291552y1073bcb6r954b043eb68a4459@mail.gmail.com>

p4raw-id: //depot/perl@30849

14 years agog++ on NetBSD needs to have environ declared as well.
Steve Peters [Thu, 5 Apr 2007 16:32:19 +0000 (16:32 +0000)]
g++ on NetBSD needs to have environ declared as well.

p4raw-id: //depot/perl@30848

14 years agoC++ compilation fix by Steve Peters
Rafael Garcia-Suarez [Thu, 5 Apr 2007 16:26:18 +0000 (16:26 +0000)]
C++ compilation fix by Steve Peters

p4raw-id: //depot/perl@30847

14 years agoS_op_destroy() was not static. Also tidy all other STATIC/static
Nicholas Clark [Thu, 5 Apr 2007 08:56:10 +0000 (08:56 +0000)]
S_op_destroy() was not static. Also tidy all other STATIC/static
declarations onto the same line as the return type, to make checking
this easier.

p4raw-id: //depot/perl@30846

14 years agoAdd TODO tests for the (sometimes) crashing threads/op deletion/reset
Nicholas Clark [Wed, 4 Apr 2007 21:42:26 +0000 (21:42 +0000)]
Add TODO tests for the (sometimes) crashing threads/op deletion/reset
combination.

p4raw-id: //depot/perl@30845

14 years agoA test for reset.
Nicholas Clark [Wed, 4 Apr 2007 19:55:23 +0000 (19:55 +0000)]
A test for reset.

p4raw-id: //depot/perl@30844

14 years agoGet further through the build with read-only optrees. Free()ing cops
Nicholas Clark [Wed, 4 Apr 2007 17:22:02 +0000 (17:22 +0000)]
Get further through the build with read-only optrees. Free()ing cops
as currently implemented seems to want write activity to them. Also we
save a field in PL_curcop at one point, and sometimes this can be in
a read only COP. If so, the end of scope restoration is always to the
same value, but it's easier to cheat and skip the restoration than to
dig around and avoid the save.

p4raw-id: //depot/perl@30843

14 years agoSymbol export nit, found by Yves
Rafael Garcia-Suarez [Wed, 4 Apr 2007 08:40:52 +0000 (08:40 +0000)]
Symbol export nit, found by Yves

p4raw-id: //depot/perl@30842

14 years agoRe: pmdynflags and thread safety
Yves Orton [Wed, 4 Apr 2007 01:46:26 +0000 (03:46 +0200)]
Re: pmdynflags and thread safety
Message-ID: <9b18b3110704031646p7ac8dbearf9e41397a5f884d8@mail.gmail.com>

p4raw-id: //depot/perl@30841

14 years agoext\ExtUtils\t\Embed.t fails test when upgrading a perl with different core headers.
Yves Orton [Wed, 4 Apr 2007 03:17:21 +0000 (05:17 +0200)]
ext\ExtUtils\t\Embed.t fails test when upgrading a perl with different core headers.
Message-ID: <9b18b3110704031817l37c034d3gf3d2b1cd205ac517@mail.gmail.com>

p4raw-id: //depot/perl@30840

14 years agoRe: [PATCH] microperl plus missing config vars
Jarkko Hietaniemi [Tue, 3 Apr 2007 18:28:30 +0000 (14:28 -0400)]
Re: [PATCH] microperl plus missing config vars
Message-ID: <4612D50E.5010105@iki.fi>

p4raw-id: //depot/perl@30839

14 years agoRe: [PATCH] MAD prototype checking
Gerard Goossen [Thu, 29 Mar 2007 17:34:23 +0000 (19:34 +0200)]
Re: [PATCH] MAD prototype checking
Message-ID: <20070329153423.GE20231@ostwald>

p4raw-id: //depot/perl@30838

14 years agoRe: [PATCH] MAD disable constant inlining
Gerard Goossen [Thu, 29 Mar 2007 17:31:32 +0000 (19:31 +0200)]
Re: [PATCH] MAD disable constant inlining
Message-ID: <20070329153132.GD20231@ostwald>

p4raw-id: //depot/perl@30837

14 years agoC++ compilation patch by Jarkko
Rafael Garcia-Suarez [Tue, 3 Apr 2007 08:34:09 +0000 (08:34 +0000)]
C++ compilation patch by Jarkko

p4raw-id: //depot/perl@30836

14 years agoWe may not even have a list of slabs when Perl_Slab_Free is called.
Nicholas Clark [Mon, 2 Apr 2007 21:27:26 +0000 (21:27 +0000)]
We may not even have a list of slabs when Perl_Slab_Free is called.

p4raw-id: //depot/perl@30835

14 years agoCleanups for PERL_DEBUG_READONLY_OPS debugging code - remove a now-
Nicholas Clark [Mon, 2 Apr 2007 21:25:03 +0000 (21:25 +0000)]
Cleanups for PERL_DEBUG_READONLY_OPS debugging code - remove a now-
unneeded include of <sys/mman.h>, and make PERL_DEBUG_READONLY_OPS
appear in -V output.

p4raw-id: //depot/perl@30834

14 years agoAllow testing of DBM_Filter with any *DBM_File
Jerry D. Hedden [Mon, 2 Apr 2007 10:15:47 +0000 (03:15 -0700)]
Allow testing of DBM_Filter with any *DBM_File
From: "Jerry D. Hedden" <jdhedden@yahoo.com>
Message-ID: <642435.53086.qm@web30215.mail.mud.yahoo.com>

p4raw-id: //depot/perl@30833

14 years agoFor Perl_Slab_Alloc(), eliminate the unused parameter 'm'.
Nicholas Clark [Mon, 2 Apr 2007 19:53:06 +0000 (19:53 +0000)]
For Perl_Slab_Alloc(), eliminate the unused parameter 'm'.

p4raw-id: //depot/perl@30832

14 years agoDon't install dbm_filter_util.pl
Jerry D. Hedden [Mon, 2 Apr 2007 09:31:13 +0000 (02:31 -0700)]
Don't install dbm_filter_util.pl
From: "Jerry D. Hedden" <jdhedden@yahoo.com>
Message-ID: <721682.89546.qm@web30209.mail.mud.yahoo.com>

p4raw-id: //depot/perl@30831

14 years agoThe op slab allocator can call calloc() instead of malloc(), and save
Nicholas Clark [Mon, 2 Apr 2007 19:25:53 +0000 (19:25 +0000)]
The op slab allocator can call calloc() instead of malloc(), and save
a Zero(). Also, argument m is unused.

p4raw-id: //depot/perl@30830

14 years agoAdd a new compile option PERL_DEBUG_READONLY_OPS which marks the optree
Nicholas Clark [Mon, 2 Apr 2007 19:03:55 +0000 (19:03 +0000)]
Add a new compile option PERL_DEBUG_READONLY_OPS which marks the optree
as read only (or as much of it as it practical). This makes it trivial
to detect buggy code that is modifying the optree at runtime.

p4raw-id: //depot/perl@30829

14 years agoperlfunc.pod: Lexical visibility hint in require docu
Christian Winter [Sun, 1 Apr 2007 18:01:24 +0000 (20:01 +0200)]
perlfunc.pod: Lexical visibility hint in require docu
Message-ID: <460FD754.8050106@linux-config.de>

p4raw-id: //depot/perl@30828

14 years agoUpdate name of Perl executable on VMS following #30414.
Craig A. Berry [Sun, 1 Apr 2007 18:11:28 +0000 (18:11 +0000)]
Update name of Perl executable on VMS following #30414.

p4raw-id: //depot/perl@30827

14 years agoMark new regex glob test as a TODO on VMS.
Craig A. Berry [Sun, 1 Apr 2007 18:01:01 +0000 (18:01 +0000)]
Mark new regex glob test as a TODO on VMS.

p4raw-id: //depot/perl@30826

14 years agoREADME.$OS URLs
Steffen Müller [Sat, 31 Mar 2007 16:37:56 +0000 (18:37 +0200)]
README.$OS URLs
Message-ID: <20070331143757.26271.qmail@lists.develooper.com>

p4raw-id: //depot/perl@30825

14 years agoSymbian sync
Jarkko Hietaniemi [Sat, 31 Mar 2007 15:30:09 +0000 (11:30 -0400)]
Symbian sync
Message-ID: <460EB6C1.4020406@iki.fi>

p4raw-id: //depot/perl@30824

14 years agomicroperl plus missing config vars
Jarkko Hietaniemi [Sat, 31 Mar 2007 17:50:18 +0000 (13:50 -0400)]
microperl plus missing config vars
Message-ID: <460ED79A.5030809@iki.fi>

p4raw-id: //depot/perl@30823

14 years agoMake parser_free() be called slightly later,
Dave Mitchell [Sun, 1 Apr 2007 01:27:57 +0000 (01:27 +0000)]
Make parser_free() be called slightly later,
i.e. at the point where lexer vars are being restored in a LEAVE,
rather than at the end of yyparse()

p4raw-id: //depot/perl@30822

14 years agoEliminate the use of PL_curforce outside of toke.c
Dave Mitchell [Sun, 1 Apr 2007 01:22:20 +0000 (01:22 +0000)]
Eliminate the use of PL_curforce outside of toke.c

p4raw-id: //depot/perl@30821

14 years agoEliminate the use of PL_bufend outside of toke.c:
Dave Mitchell [Sun, 1 Apr 2007 01:20:02 +0000 (01:20 +0000)]
Eliminate the use of PL_bufend outside of toke.c:
give Perl_scan_vstring() an explicit 'end' arg rather than using
PL_bufend, and replace it with a local var in Perl_find_script()

p4raw-id: //depot/perl@30820

14 years agofree backtrack track when on unexpected exit, eg /(?{ die })/
Dave Mitchell [Sun, 1 Apr 2007 00:51:00 +0000 (00:51 +0000)]
free backtrack track when on unexpected exit, eg /(?{ die })/

p4raw-id: //depot/perl@30819

14 years agofix $^R scoping bug.
Dave Mitchell [Sun, 1 Apr 2007 00:27:02 +0000 (00:27 +0000)]
fix $^R scoping bug.
By setting the outer saved $^R to the current $^R just at the
end of a successful match, and ensuring that that the savestack
doesn't get popped beforehand, the code is simplified and fixes a
bug.

p4raw-id: //depot/perl@30818

14 years agoRearrange members of structures to reduce memory size on some
Nicholas Clark [Sat, 31 Mar 2007 23:15:15 +0000 (23:15 +0000)]
Rearrange members of structures to reduce memory size on some
platforms. On LP64 structs stackinfo, refcounted_he, and magic shrink
by 8 bytes, struct yy_parser by 16.

p4raw-id: //depot/perl@30817

14 years agoIntegrate:
Nicholas Clark [Sat, 31 Mar 2007 23:13:15 +0000 (23:13 +0000)]
Integrate:
[ 30652]
In safecalloc, use the system calloc() for the non-DEBUGGING case, as
it may be able to memset() to 0 for free.

[ 30661]
Remove the (sometimes) unused variable from change 30652.

[ 30670]
Subject: [PATCH] Change 30661 could result in undefined variable
From: "Jerry D. Hedden" <jdhedden@yahoo.com>
Message-ID: <915076.64236.qm@web30212.mail.mud.yahoo.com>
Date: Wed, 21 Mar 2007 12:47:52 -0700 (PDT)

[ 30771]
Subject: [PATCH] util.c [PATCH] perlhack.pod (Was: Re: threads crashes in Tru64)
From: Jarkko Hietaniemi <jhi@iki.fi>
Date: Mon, 26 Mar 2007 19:50:11 -0400
Message-ID: <46085C33.1030601@iki.fi>
p4raw-link: @30771 on //depot/perl: ad7244db27635ed088fc05a8a69e99bbb19c36d6
p4raw-link: @30670 on //depot/perl: 6460123ae692a9c25bc8c2253f12c8cf6f0ebdc0
p4raw-link: @30661 on //depot/perl: a93c8a9e4a51d34ca994257e736b80ca04ec6f9e
p4raw-link: @30652 on //depot/perl: e1a95402a0ef053bf8dd15ba3824743513f0725e

p4raw-id: //depot/perl@30816

14 years agoReorder the members of various regexp structs to reduce their size on
Nicholas Clark [Sat, 31 Mar 2007 20:52:53 +0000 (20:52 +0000)]
Reorder the members of various regexp structs to reduce their size on
LP64 platforms, by pairing up I32 and U32 members. Notably structs
_reg_trie_data, reg_ac_data, regexp and regmatch_state down by 8 bytes,
re_save_state by 16, and regmatch_slab up by 48 (ie one more state per
slab)

p4raw-id: //depot/perl@30815

14 years agoFix URL rot in README.vms
Craig A. Berry [Sat, 31 Mar 2007 19:42:35 +0000 (19:42 +0000)]
Fix URL rot in README.vms

p4raw-id: //depot/perl@30814

14 years agoMake Perl_vms_start_glob return the input pattern when there are
Craig A. Berry [Sat, 31 Mar 2007 19:40:43 +0000 (19:40 +0000)]
Make Perl_vms_start_glob return the input pattern when there are
no matches.

p4raw-id: //depot/perl@30813

14 years agoprecomp in struct regexp can be const char *.
Nicholas Clark [Sat, 31 Mar 2007 15:50:05 +0000 (15:50 +0000)]
precomp in struct regexp can be const char *.

p4raw-id: //depot/perl@30812

14 years agoFix bug #42224, perl -MO=Deparse -e 'print readpipe $x, $y' broken
Rafael Garcia-Suarez [Sat, 31 Mar 2007 14:11:40 +0000 (14:11 +0000)]
Fix bug #42224, perl -MO=Deparse -e 'print readpipe $x, $y' broken

p4raw-id: //depot/perl@30811

14 years agoFix the error message "Can't call method "DOES" on unblessed
Rafael Garcia-Suarez [Sat, 31 Mar 2007 06:23:12 +0000 (06:23 +0000)]
Fix the error message "Can't call method "DOES" on unblessed
reference".

p4raw-id: //depot/perl@30806

14 years agoDon't use utf8.pm in the test (avoiding embedding utf8 in the test
Rafael Garcia-Suarez [Sat, 31 Mar 2007 06:16:09 +0000 (06:16 +0000)]
Don't use utf8.pm in the test (avoiding embedding utf8 in the test
file), and use greek to test for something out of 8-bit latin, as
Jarkko suggested. Everything flows.

p4raw-id: //depot/perl@30805

14 years agoThis was released as 2.16.
Abhijit Menon-Sen [Sat, 31 Mar 2007 00:51:55 +0000 (00:51 +0000)]
This was released as 2.16.

p4raw-id: //depot/perl@30804

14 years agoAdd new hints files.
Abhijit Menon-Sen [Sat, 31 Mar 2007 00:34:13 +0000 (00:34 +0000)]
Add new hints files.

p4raw-id: //depot/perl@30803

14 years agoBump File::Glob's VERSION to go with change 30103.
Nicholas Clark [Fri, 30 Mar 2007 16:17:25 +0000 (16:17 +0000)]
Bump File::Glob's VERSION to go with change 30103.

p4raw-id: //depot/perl@30791

14 years agoRemove the never-used numeric and character switch ops.
Nicholas Clark [Fri, 30 Mar 2007 11:52:43 +0000 (11:52 +0000)]
Remove the never-used numeric and character switch ops.

p4raw-id: //depot/perl@30785

14 years agoA logical rearrangement of ops, to get the post 5.005 ops to their
Nicholas Clark [Fri, 30 Mar 2007 11:48:54 +0000 (11:48 +0000)]
A logical rearrangement of ops, to get the post 5.005 ops to their
logical groups.

p4raw-id: //depot/perl@30784

14 years agoOops. Failed to remove the now obsolete comment about custom ops.
Nicholas Clark [Fri, 30 Mar 2007 10:53:31 +0000 (10:53 +0000)]
Oops. Failed to remove the now obsolete comment about custom ops.

p4raw-id: //depot/perl@30783

14 years agoRemove the restriction that op_custom has to be the last op.
Nicholas Clark [Fri, 30 Mar 2007 10:50:10 +0000 (10:50 +0000)]
Remove the restriction that op_custom has to be the last op.
This allows more ops to be added during the life of a stable release.

p4raw-id: //depot/perl@30782

14 years agoWhen dup'ing a filehandle, retain the :utf8 layer
Rafael Garcia-Suarez [Fri, 30 Mar 2007 08:30:35 +0000 (08:30 +0000)]
When dup'ing a filehandle, retain the :utf8 layer

p4raw-id: //depot/perl@30781

14 years agoAdd two deprecation warnings:
Rafael Garcia-Suarez [Fri, 30 Mar 2007 07:48:59 +0000 (07:48 +0000)]
Add two deprecation warnings:
Opening dirhandle %s also as a file
Opening filehandle %s also as a directory

p4raw-id: //depot/perl@30780

14 years agocflags.SH: 30327 wasn't portable Bourne (avoid '!: not found')
Jarkko Hietaniemi [Fri, 30 Mar 2007 04:44:11 +0000 (07:44 +0300)]
cflags.SH: 30327 wasn't portable Bourne (avoid '!: not found')
Message-Id: <200703300144.l2U1iBSA490663@kosh.hut.fi>

p4raw-id: //depot/perl@30779

14 years agoUpgrade to CGI.pm-3.28
Steve Peters [Thu, 29 Mar 2007 23:11:29 +0000 (23:11 +0000)]
Upgrade to CGI.pm-3.28

p4raw-id: //depot/perl@30778

14 years agoShorten 'skipping...' test message
Jerry D. Hedden [Thu, 29 Mar 2007 05:56:51 +0000 (22:56 -0700)]
Shorten 'skipping...' test message
From: "Jerry D. Hedden" <jdhedden@yahoo.com>
Message-ID: <213529.35369.qm@web30214.mail.mud.yahoo.com>

p4raw-id: //depot/perl@30777

14 years agoBack out changes 30762 and 30763
Rafael Garcia-Suarez [Thu, 29 Mar 2007 07:56:11 +0000 (07:56 +0000)]
Back out changes 30762 and 30763

p4raw-id: //depot/perl@30776

14 years ago: perlguts.pod
John P. Linderman [Tue, 27 Mar 2007 12:33:39 +0000 (08:33 -0400)]
: perlguts.pod
From: "John P. Linderman" <jpl@research.att.com>
Message-Id: <200703271633.l2RGXd3u3443680@raptor.research.att.com>

p4raw-id: //depot/perl@30775

14 years agostop the cargo cult of (MEM_SIZE)~0
Jarkko Hietaniemi [Tue, 27 Mar 2007 15:07:52 +0000 (18:07 +0300)]
stop the cargo cult of (MEM_SIZE)~0
Message-Id: <200703271207.l2RC7qOC443040@kosh.hut.fi>

p4raw-id: //depot/perl@30774

14 years agoCompress::Zlib isn't a good example of extra modules,
Rafael Garcia-Suarez [Tue, 27 Mar 2007 11:15:51 +0000 (11:15 +0000)]
Compress::Zlib isn't a good example of extra modules,
since it's now core

p4raw-id: //depot/perl@30773

14 years agoFix return value of state assignment ($x=state $y=$z).
Rafael Garcia-Suarez [Tue, 27 Mar 2007 09:42:44 +0000 (09:42 +0000)]
Fix return value of state assignment ($x=state $y=$z).
Fix a form of state assignment (state($x)=$y).
Minor optimisation in pp_aassign for state variables.
New tests.

p4raw-id: //depot/perl@30772

14 years agoutil.c [PATCH] perlhack.pod (Was: Re: threads crashes in Tru64)
Jarkko Hietaniemi [Mon, 26 Mar 2007 19:50:11 +0000 (15:50 -0400)]
util.c [PATCH] perlhack.pod (Was: Re: threads crashes in Tru64)
Message-ID: <46085C33.1030601@iki.fi>

p4raw-id: //depot/perl@30771

14 years agoDocumented the PA-RISC End-Of_Life plan; more servers; some IPF
H.Merijn Brand [Tue, 27 Mar 2007 06:15:21 +0000 (06:15 +0000)]
Documented the PA-RISC End-Of_Life plan; more servers; some IPF
remarks. TODO: the various locations for downloading prebuilt
depots, and what their specifics are. The first is now in, but
no specifics yet.

p4raw-id: //depot/perl@30770

14 years agoIn struct regexp replace the two arrays of I32s accessed via startp
Nicholas Clark [Mon, 26 Mar 2007 22:52:18 +0000 (22:52 +0000)]
In struct regexp replace the two arrays of I32s accessed via startp
and endp with a single array of struct regexp_paren_pair, which has 2
I32 members. PL_regstartp and PL_regendp are replaced with a pointer
to regexp_paren_pair. The regexp swap structure now only has one
member, so abolish it and store the pointer to the swap array directly.
Hopefully keeping the corresponding start and end adjacent in memory
will help with cache coherency.

p4raw-id: //depot/perl@30769

14 years agoadd p55.t MAD tests, and convert renamed mad/nomad into a module
Gerard Goossen [Mon, 26 Mar 2007 21:48:48 +0000 (21:48 +0000)]
add p55.t MAD tests, and convert renamed mad/nomad into a module
Subject: Re: [PATCH] p55 tests
Message-ID: <20070322174056.GE24152@ostwald>

p4raw-id: //depot/perl@30768

14 years agoRename mad/nomad to mad/Nomad.pm
Gerard Goossen [Mon, 26 Mar 2007 21:42:48 +0000 (21:42 +0000)]
Rename mad/nomad to mad/Nomad.pm

Subject: Re: [PATCH] p55 tests
Message-ID: <20070322174056.GE24152@ostwald>

p4raw-id: //depot/perl@30767

14 years agoMove the declaration of two variables into the only block that uses
Nicholas Clark [Mon, 26 Mar 2007 21:40:12 +0000 (21:40 +0000)]
Move the declaration of two variables into the only block that uses
them.

p4raw-id: //depot/perl@30766

14 years agoNit to change 30762 by Gisle.
Rafael Garcia-Suarez [Mon, 26 Mar 2007 14:36:17 +0000 (14:36 +0000)]
Nit to change 30762 by Gisle.
Avoids storing the chars removed by 4-arg substr() in void context.

p4raw-id: //depot/perl@30763

14 years agoRe: substr($megabytes, 30) =
Yitzchak Scott-Thoennes [Mon, 12 Feb 2007 22:13:50 +0000 (14:13 -0800)]
Re: substr($megabytes, 30) =
From: "Yitzchak Scott-Thoennes" <sthoenna@efn.org>
Message-ID: <1119.63.226.247.186.1171347230.squirrel@63.226.247.186>

p4raw-id: //depot/perl@30762

14 years agoMake the warning "interpreted as function" a bit less annoying,
Rafael Garcia-Suarez [Mon, 26 Mar 2007 14:21:39 +0000 (14:21 +0000)]
Make the warning "interpreted as function" a bit less annoying,
and test it properly

p4raw-id: //depot/perl@30761

14 years agosay() should parse exactly like print()
Rafael Garcia-Suarez [Mon, 26 Mar 2007 13:51:30 +0000 (13:51 +0000)]
say() should parse exactly like print()

p4raw-id: //depot/perl@30760

14 years agoRemove the object_compatibility placeholder in the interpreter struct
Rafael Garcia-Suarez [Mon, 26 Mar 2007 11:52:21 +0000 (11:52 +0000)]
Remove the object_compatibility placeholder in the interpreter struct

p4raw-id: //depot/perl@30759

14 years agoOn Linux, don't compile with -DTHREADS_HAVE_PIDS if the
Rafael Garcia-Suarez [Mon, 26 Mar 2007 10:09:21 +0000 (10:09 +0000)]
On Linux, don't compile with -DTHREADS_HAVE_PIDS if the
kernel uses the NPTL threading model

p4raw-id: //depot/perl@30758

14 years agoSimplify the logic in Perl_re_dup() - copy the structure as 1 block,
Nicholas Clark [Mon, 26 Mar 2007 08:39:43 +0000 (08:39 +0000)]
Simplify the logic in Perl_re_dup() - copy the structure as 1 block,
then fix up what differs. Curiously -Os with gcc x86 produces the same
sized object code.

p4raw-id: //depot/perl@30757

14 years agoChange 30753 forgot to remove 2 now unused variables.
Nicholas Clark [Sun, 25 Mar 2007 23:40:00 +0000 (23:40 +0000)]
Change 30753 forgot to remove 2 now unused variables.

p4raw-id: //depot/perl@30756

14 years agoLooks like re_dup has been leaking references on 2 SVs for most regexps
Nicholas Clark [Sun, 25 Mar 2007 23:34:58 +0000 (23:34 +0000)]
Looks like re_dup has been leaking references on 2 SVs for most regexps
since dot (only to be recovered for certain at thread exit).

p4raw-id: //depot/perl@30755

14 years agoA more efficient copy in Perl_reg_temp_copy()
Nicholas Clark [Sun, 25 Mar 2007 22:32:11 +0000 (22:32 +0000)]
A more efficient copy in  Perl_reg_temp_copy()

p4raw-id: //depot/perl@30754

14 years agoRe: The performance problem of 30678
Marcus Holland-Moritz [Sat, 24 Mar 2007 18:55:11 +0000 (19:55 +0100)]
Re: The performance problem of 30678
Message-ID: <20070324185511.02f84668@r2d2>
Date: Sat, 24 Mar 2007 18:55:11 +0100

p4raw-id: //depot/perl@30753

14 years agoMake readline() default to *ARGV.
Rafael Garcia-Suarez [Sat, 24 Mar 2007 21:14:22 +0000 (21:14 +0000)]
Make readline() default to *ARGV.
Plus MAD fixes.

p4raw-id: //depot/perl@30750

14 years agoNow that readpipe defaults to $_, I should update the prototype table
Rafael Garcia-Suarez [Sat, 24 Mar 2007 16:58:28 +0000 (16:58 +0000)]
Now that readpipe defaults to $_, I should update the prototype table
in opcode.pl too.

p4raw-id: //depot/perl@30749

14 years agoMake sure readpipe without argument can be followed by //
Rafael Garcia-Suarez [Sat, 24 Mar 2007 16:51:13 +0000 (16:51 +0000)]
Make sure readpipe without argument can be followed by //

p4raw-id: //depot/perl@30748

14 years agoMake readpipe default to $_
Rafael Garcia-Suarez [Sat, 24 Mar 2007 16:46:02 +0000 (16:46 +0000)]
Make readpipe default to $_

p4raw-id: //depot/perl@30747

14 years agostartp and endp can allocate their array of I32s from the same block of
Nicholas Clark [Sat, 24 Mar 2007 14:44:28 +0000 (14:44 +0000)]
startp and endp can allocate their array of I32s from the same block of
memory, and Copy() and free it as one.
Probably these two pointers to array should merge into single pointer
to an array of structs.

p4raw-id: //depot/perl@30745

14 years agoIt's possible to write the dup of struct reg_substr_datum with a
Nicholas Clark [Fri, 23 Mar 2007 23:32:19 +0000 (23:32 +0000)]
It's possible to write the dup of struct reg_substr_datum with a
memcpy() replacing the member by member copy. Curiously gcc's -Os
produces the same sized code, but not all optimisers may manage this.
Also, by reading and re-assigning to the copied data for the sv_dup()s
we hope to avoid any cache misses on the copied from data.

p4raw-id: //depot/perl@30740

14 years agoDon't use PL_na.
Rafael Garcia-Suarez [Fri, 23 Mar 2007 18:25:43 +0000 (18:25 +0000)]
Don't use PL_na.

p4raw-id: //depot/perl@30730

14 years ago[perl #41775] Memoize bug (and patch): problem with LIST_CACHE => MERGE
Father Chrysostomos [Sat, 10 Mar 2007 12:18:35 +0000 (04:18 -0800)]
[perl #41775] Memoize bug (and patch): problem with LIST_CACHE => MERGE
From: Father Chrysostomos (via RT) <perlbug-followup@perl.org>
Message-ID: <rt-3.6.HEAD-30201-1173557915-1780.41775-75-0@perl.org>

p4raw-id: //depot/perl@30729

14 years agoUpgrade to Devel::PPPort 3.11_01
Marcus Holland-Moritz [Fri, 23 Mar 2007 17:21:15 +0000 (17:21 +0000)]
Upgrade to Devel::PPPort 3.11_01

p4raw-id: //depot/perl@30728

14 years agoMove CHECK_WORD in regcomp.c to a more generic memEQs() in handy.h
Nicholas Clark [Fri, 23 Mar 2007 16:55:13 +0000 (16:55 +0000)]
Move CHECK_WORD in regcomp.c to a more generic memEQs() in handy.h
and provide memNEs() too.

p4raw-id: //depot/perl@30726

14 years agoSilence compiler warning in util.c
Jerry D. Hedden [Fri, 23 Mar 2007 09:40:02 +0000 (02:40 -0700)]
Silence compiler warning in util.c
From: "Jerry D. Hedden" <jdhedden@yahoo.com>
Message-ID: <256688.25862.qm@web30215.mail.mud.yahoo.com>

p4raw-id: //depot/perl@30725

14 years agoChange 30080 was wrong to swap the BEGIN test to memEQ
Nicholas Clark [Fri, 23 Mar 2007 16:15:58 +0000 (16:15 +0000)]
Change 30080 was wrong to swap the BEGIN test to memEQ

p4raw-id: //depot/perl@30724

14 years agoRe: [perl #41683] [PATCH] v5.8.8 pod2html -- incorrect treatment of non-manual page...
Jari Aalto [Wed, 7 Mar 2007 11:08:24 +0000 (13:08 +0200)]
Re: [perl #41683] [PATCH] v5.8.8 pod2html -- incorrect treatment of non-manual page refs like "this(c)"
Message-ID: <87ejo1s9o7.fsf@w2kpicasso.cante.net>

with adjustments to the regexp

p4raw-id: //depot/perl@30722

14 years agoPerlFAQ sync.
Rafael Garcia-Suarez [Fri, 23 Mar 2007 14:33:22 +0000 (14:33 +0000)]
PerlFAQ sync.
Also, use a svn checkout done under an English locale, to
avoid having French dates in the headers. Hateful.

p4raw-id: //depot/perl@30721