SADAHIRO Tomoyuki [Thu, 4 Jan 2007 00:21:02 +0000]
Re: What if NV_PRESERVES_UV_BITS is undef? (was: Re: [PATCH] Configure test code fails to compile if SIGFPE is undefined)
Message-Id: <
20070104001943.6BC2.BQW10602@nifty.com>
p4raw-id: //depot/perl@29693
Steve Hay [Fri, 5 Jan 2007 09:45:08 +0000]
Silence various VC6 warnings
p4raw-id: //depot/perl@29692
Bo Lindbergh [Thu, 4 Jan 2007 18:22:09 +0000]
Re: [PATCH] perlio.c repair
Message-Id: <
E22DC961-6821-4AC2-BA2F-9B5B06927758@hagernas.com>
p4raw-id: //depot/perl@29691
Jan Dubois [Thu, 4 Jan 2007 12:37:48 +0000]
ANSIfy the PATH environment variable on Windows
Message-ID: <7coqp2pme3r30qaonticuv6c6licieg4cg@4ax.com>
p4raw-id: //depot/perl@29690
Jan Dubois [Thu, 4 Jan 2007 12:20:21 +0000]
Add error handling to win32_ansipath
Message-ID: <f2oqp2dbdglim8bda4q1kajt9k3cvpqqga@4ax.com>
p4raw-id: //depot/perl@29689
Rafael Garcia-Suarez [Thu, 4 Jan 2007 17:04:11 +0000]
Upgrade Module::Load::Conditional to 0.14
p4raw-id: //depot/perl@29688
Marcus Holland-Moritz [Mon, 1 Jan 2007 20:16:13 +0000]
4th patch from:
Subject: [PATCH] Cleanup SVf arguments (2nd try)
Message-ID: <
20070101201613.4120d9ef@r2d2>
Introduce an SVfARG() macro for %SVf (%-p here) arguments to
perl's printf
p4raw-id: //depot/perl@29687
Marcus Holland-Moritz [Mon, 1 Jan 2007 20:16:13 +0000]
The first three patches from:
Subject: [PATCH] Cleanup SVf arguments (2nd try)
Message-ID: <
20070101201613.4120d9ef@r2d2>
Adds (void*) casts to %-p and %p printf arguments
p4raw-id: //depot/perl@29686
Jarkko Hietaniemi [Thu, 4 Jan 2007 03:35:54 +0000]
Re: [PATCH] Symbian port: add Series 90 support
Message-ID: <
459CBC6A.4060806@iki.fi>
Fixes a VC7/VC7.1 linker error following #29650/#29651
p4raw-id: //depot/perl@29685
Rafael Garcia-Suarez [Thu, 4 Jan 2007 14:40:18 +0000]
Remove no longer used magic vtable
p4raw-id: //depot/perl@29684
Rafael Garcia-Suarez [Thu, 4 Jan 2007 14:39:08 +0000]
Silence a 'null argument' warning
p4raw-id: //depot/perl@29683
Yves Orton [Fri, 29 Dec 2006 21:45:51 +0000]
Re: [PATCH] Change implementation of %+ to use a proper tied hash interface and add support for %-
Message-ID: <
9b18b3110612291245q792fe91cu69422d2b81bb4f0b@mail.gmail.com>
p4raw-id: //depot/perl@29682
Jan Dubois [Thu, 28 Dec 2006 19:06:00 +0000]
Add note to perltodo.pod about Unicode and file globbing
Message-ID: <4e19p2lgme9qbqjkv78q7154vg6hlaar19@4ax.com>
p4raw-id: //depot/perl@29681
Dave Mitchell [Thu, 4 Jan 2007 01:03:22 +0000]
make parser_free actually free the parser!
also remove two SAVEs left over from #29655
p4raw-id: //depot/perl@29680
Nicholas Clark [Wed, 3 Jan 2007 23:13:43 +0000]
Rename OURSTASH to SvOURSTASH and OURSTASH_set to SvOURSTASH_set.
p4raw-id: //depot/perl@29679
Dave Mitchell [Wed, 3 Jan 2007 22:17:48 +0000]
handle cloning of parsers on the save stack
p4raw-id: //depot/perl@29678
Nicholas Clark [Wed, 3 Jan 2007 22:02:49 +0000]
Teach dump about OURSTASH and COP ranges stored in the NV union.
p4raw-id: //depot/perl@29677
Jan Dubois [Wed, 3 Jan 2007 08:48:22 +0000]
Update change 29607 to affect only Win32
Message-ID: <f9nnp2lf4hsb74jf9unlsj59ahsf4f1o1v@4ax.com>
p4raw-id: //depot/perl@29676
Jan Dubois [Thu, 28 Dec 2006 18:59:40 +0000]
[PATCH] Use short pathnames in $^X and @INC if the long form cannot be represented in the current codepage
Date: Thu, 28 Dec 2006 18:59:40 -0800
Message-ID: <vq09p2p09k6rcu6c9t0mab3vnc335ghg9m@4ax.com>
Subject: Re: [PATCH] Use short pathnames in $^X and @INC if the long form cannot be represented in the current codepage
From: Jan Dubois <jand@ActiveState.com>
Date: Wed, 03 Jan 2007 08:12:35 -0800
Message-ID: <orknp2pj17265modfosjkp2qtt4bdgtgjp@4ax.com>
p4raw-id: //depot/perl@29675
Jarkko Hietaniemi [Mon, 1 Jan 2007 14:35:46 +0000]
Re: g++ unhappy
Message-ID: <
45996292.5090000@iki.fi>
p4raw-id: //depot/perl@29674
Nicholas Clark [Wed, 3 Jan 2007 15:17:29 +0000]
Make the ptr_table functions available in an unthreaded perl, as other
code (such as Storable) can take advantage of them.
p4raw-id: //depot/perl@29673
Paul Marquess [Tue, 2 Jan 2007 13:14:26 +0000]
IO::Compress::* 2.003
From: "Paul Marquess" <Paul.Marquess@ntlworld.com>
Message-ID: <
02ca01c72e6f$eef88360$6401a8c0@myopwv.com>
p4raw-id: //depot/perl@29672
Steve Peters [Wed, 3 Jan 2007 01:46:41 +0000]
Upgrade to Time-HiRes-1.9704
p4raw-id: //depot/perl@29671
Nicholas Clark [Tue, 2 Jan 2007 21:45:08 +0000]
Update copyright years to include 2007. (Plus a couple of 2006s and
earlier we missed in av.h and hv.h)
p4raw-id: //depot/perl@29670
Nicholas Clark [Tue, 2 Jan 2007 21:10:13 +0000]
Assignment to a tainted variable was causing confusion if the source
value was an NV too large for an IV (bug #40708). Fix the confusion
by not promoting private flags to public flags in S_save_magic if
there are already public flags.
p4raw-id: //depot/perl@29669
Nicholas Clark [Tue, 2 Jan 2007 20:47:31 +0000]
PVMG can be isUV too.
p4raw-id: //depot/perl@29668
Dave Mitchell [Tue, 2 Jan 2007 19:44:23 +0000]
fix -Mad build breakage
p4raw-id: //depot/perl@29667
Nicholas Clark [Tue, 2 Jan 2007 18:14:26 +0000]
Copying to FORMATs should work (and not fail assertions if the value
is floating point). Copying to ARRAYs and HASHes can't work and should
croak.
p4raw-id: //depot/perl@29666
Nicholas Clark [Tue, 2 Jan 2007 17:41:06 +0000]
It's good to be able to dump FORMATs too :-)
p4raw-id: //depot/perl@29665
Nicholas Clark [Tue, 2 Jan 2007 16:28:48 +0000]
Remove the vestigal "#if 0"s from header files that defined same-sized
*allocated structs, as these are not going to be needed again.
p4raw-id: //depot/perl@29664
Nicholas Clark [Tue, 2 Jan 2007 16:05:02 +0000]
Probably better to make PAD_FAKELEX_ANON and PAD_FAKELEX_MULTI special
cases, instead of scanning pad.h to find them, and picking up 3 other
superfluous definitions as a side effect.
p4raw-id: //depot/perl@29663
Nicholas Clark [Tue, 2 Jan 2007 14:55:29 +0000]
Note that the examples for the structure copying bug example are no
longer relevant because of the XPVGV re-ordering.
p4raw-id: //depot/perl@29662
Steve Hay [Tue, 2 Jan 2007 14:50:56 +0000]
Fix a couple of linker errors on Win32
p4raw-id: //depot/perl@29661
Dave Mitchell [Tue, 2 Jan 2007 14:02:40 +0000]
remove conflict in pending_ident function and field name
p4raw-id: //depot/perl@29660
Steve Hay [Tue, 2 Jan 2007 11:18:26 +0000]
Check miniperl exists before using it to run uupacktool.pl
during distclean
p4raw-id: //depot/perl@29659
Marcus Holland-Moritz [Tue, 2 Jan 2007 00:08:27 +0000]
Configure test code fails to compile if SIGFPE is undefined
Message-ID: <
20070102000827.7acf11c8@r2d2>
p4raw-id: //depot/perl@29658
Jarkko Hietaniemi [Tue, 2 Jan 2007 03:15:42 +0000]
really should test compile in AIX or something (fixup after the Series 90 patch)
Message-Id: <
200701020115.l021FgtC495074@kosh.hut.fi>
p4raw-id: //depot/perl@29656
Dave Mitchell [Tue, 2 Jan 2007 01:51:05 +0000]
move easy lexer state variables into PL_parser
(where "easy" == "only appear in toke.c")
p4raw-id: //depot/perl@29655
Steve Peters [Tue, 2 Jan 2007 00:07:52 +0000]
Quiet a smoke warning in toke.c and bump the copyright year.
p4raw-id: //depot/perl@29654
Steve Peters [Mon, 1 Jan 2007 22:48:33 +0000]
Welcome to 2007! Time to update copyrights from changes earlier
today.
p4raw-id: //depot/perl@29653
Dave Mitchell [Mon, 1 Jan 2007 22:37:40 +0000]
split parser initialisation from parser execution
p4raw-id: //depot/perl@29652
Steve Peters [Mon, 1 Jan 2007 19:30:05 +0000]
Add file missed in change #29650.
p4raw-link: @29650 on //depot/perl:
53d44271720d88220a01b5620a93869665083b01
p4raw-id: //depot/perl@29651
Jarkko Hietaniemi [Mon, 1 Jan 2007 08:49:03 +0000]
Symbian port: add Series 90 support
Message-ID: <
4599114F.8020307@iki.fi>
p4raw-id: //depot/perl@29650
Dave Mitchell [Mon, 1 Jan 2007 11:24:07 +0000]
split parser stack from parser object
p4raw-id: //depot/perl@29649
Dave Mitchell [Sun, 31 Dec 2006 18:56:37 +0000]
in lex_start(), ensure that all lexer state vars are saved
and re-initialised. This is mostly overkill, but mimicks the
effect of having all the vars in a single parser object.
p4raw-id: //depot/perl@29648
Nicholas Clark [Sun, 31 Dec 2006 16:58:18 +0000]
Rationalise the logic in lex_start for creating a new temporary if the
passed in SV is read only or doesn't end in ';'. Also, allow a NULL
pointer to imply a zero length string, which saves creating a mortal
in pp_require, only for it to be ignored in favour of a new SV.
p4raw-id: //depot/perl@29647
Steve Peters [Sun, 31 Dec 2006 16:37:54 +0000]
Fix to test failures caused by change #29606 when PERLIO=stdio.
Based on suggestions from Ilya Zakharevich in
http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2006-12/msg00646.html
p4raw-link: @29606 on //depot/perl:
add1a1a3c3dc28dd49272f4754cfc04acae28e3b
p4raw-id: //depot/perl@29646
Jarkko Hietaniemi [Sun, 31 Dec 2006 08:51:52 +0000]
smoke signs suppression
Message-ID: <
4597C078.4000503@iki.fi>
p4raw-id: //depot/perl@29645
Nicholas Clark [Sun, 31 Dec 2006 12:31:09 +0000]
As freeing up PL_linestr is now done via the scope stack, there's no
need to make any temporary varariable a mortal, and waste effort on
the temp stack (not) clearing it up.
p4raw-id: //depot/perl@29644
Nicholas Clark [Sun, 31 Dec 2006 12:07:37 +0000]
PL_linestr needs to survive until the end of scope, not just the next
FREETMPS. Fixes the underlying cause of the thread cloning SEGV
reported in http://www.nntp.perl.org/group/perl.perl5.porters/63123
p4raw-id: //depot/perl@29643
Nicholas Clark [Sun, 31 Dec 2006 00:08:00 +0000]
Move SVt_BIND to be the lowest type after SVt_NULL. This will force all
code attempting to upgrade a BIND to anything into sv_upgrade(), which
for now will croak, but in future can DTRT, for whatever TRT is decided
to be. Make SvOK() check the flags of the referenant for a BIND, as I
envisage that the only flag bit that will get set on a BIND is SVf_UTF8
even if the referant has a defined value.
p4raw-id: //depot/perl@29642
Paul Marquess [Fri, 29 Dec 2006 20:50:26 +0000]
Update Compress Modules to version 2.002
From: "Paul Marquess" <Paul.Marquess@ntlworld.com>
Message-ID: <
000501c72b8a$f96b2ba0$6401a8c0@myopwv.com>
p4raw-id: //depot/perl@29641
Dave Mitchell [Sat, 30 Dec 2006 01:18:18 +0000]
fix a test failing under -Dmad
p4raw-id: //depot/perl@29640
Nicholas Clark [Fri, 29 Dec 2006 16:20:39 +0000]
With xlow and xhigh as U32, PAD_MAX shouldn't be IV_MAX, as on a 64 bit
system this can't be stored.
p4raw-id: //depot/perl@29639
Nicholas Clark [Fri, 29 Dec 2006 14:42:03 +0000]
I failed to run regen.pl to propagate my pad.c changes to perlintern.pod
p4raw-id: //depot/perl@29638
Nicholas Clark [Fri, 29 Dec 2006 00:17:18 +0000]
In Perl_fbm_compile, really rarest should be U32 not I32, as it is
set from a U32, and used as an array index.
p4raw-id: //depot/perl@29637
Dave Mitchell [Fri, 29 Dec 2006 00:08:35 +0000]
further fix for #29543: fix parser leaks caused by croaking
p4raw-id: //depot/perl@29636
Nicholas Clark [Thu, 28 Dec 2006 23:57:16 +0000]
Eliminate BmPREVIOUS_set - with the complexity gone from how the datum
is stored, there's no need for it.
p4raw-id: //depot/perl@29635
Nicholas Clark [Thu, 28 Dec 2006 23:35:54 +0000]
Move all the FBM data fields from the table into a struct xbm_s which
is part of the xnv union.
p4raw-id: //depot/perl@29634
Nicholas Clark [Thu, 28 Dec 2006 22:40:37 +0000]
With PAD_COMPNAME_GEN in SvUVX, SvCUR is trustworthy once more.
p4raw-id: //depot/perl@29633
Nicholas Clark [Thu, 28 Dec 2006 22:03:56 +0000]
Move PAD_COMPNAME_GEN from SvCUR to SvUVX.
p4raw-id: //depot/perl@29632
Nicholas Clark [Thu, 28 Dec 2006 21:41:52 +0000]
Move PAD_FAKELEX_ANON and PAD_FAKELEX_MULTI to pad.h, export them via
B.pm, so that B::Concise doesn't need to hard code magic numbers.
p4raw-id: //depot/perl@29631
Nicholas Clark [Thu, 28 Dec 2006 21:21:22 +0000]
Move the low/high cop sequences from NVX/IVX to a two U32 structure
in the xnv union. This frees up IVX for the PL_generation code, which
in turn will allow SvCUR to return to its real purpose.
p4raw-id: //depot/perl@29630
Nicholas Clark [Thu, 28 Dec 2006 20:02:03 +0000]
Abstract the pad code's overloaded use of SvNVX and SvIVX into
4 macros COP_SEQ_RANGE_LOW, COP_SEQ_RANGE_HIGH, PARENT_PAD_INDEX
and PARENT_FAKELEX_FLAGS
p4raw-id: //depot/perl@29629
Nicholas Clark [Thu, 28 Dec 2006 17:55:06 +0000]
Change PAD_MAX to something a bit less FORTRAN-ish than
999999999.
p4raw-id: //depot/perl@29628
Steve Peters [Wed, 27 Dec 2006 19:36:13 +0000]
Upgrade to Errno-1.10 (!). I have left the core tests as is for
now as well as for Makefile.PL, which need some looking into.
p4raw-id: //depot/perl@29627
Yves Orton [Wed, 27 Dec 2006 11:37:36 +0000]
Macroize char class tests in case to remove duplicated code.
Message-ID: <
9b18b3110612270237i40e5ef34ud80898459ee338eb@mail.gmail.com>
Date: Wed, 27 Dec 2006 11:37:36 +0100
p4raw-id: //depot/perl@29626
Steve Peters [Wed, 27 Dec 2006 14:13:27 +0000]
Upgrade to CPAN-1.83_66.
p4raw-id: //depot/perl@29625
Steve Peters [Tue, 26 Dec 2006 15:53:57 +0000]
Cannot use a L<> within a preformatted section of a Pod document.
p4raw-id: //depot/perl@29624
SADAHIRO Tomoyuki [Tue, 26 Dec 2006 23:11:38 +0000]
Re: Named-capture regex syntax
Message-Id: <
20061226231027.8320.BQW10602@nifty.com>
p4raw-id: //depot/perl@29623
Rafael Garcia-Suarez [Mon, 25 Dec 2006 17:09:07 +0000]
Fix a couple of warnings
p4raw-id: //depot/perl@29622
Yves Orton [Sun, 24 Dec 2006 14:38:15 +0000]
Re: Named-capture regex syntax
Message-ID: <
9b18b3110612240538m5c45654br7d27171835f6664@mail.gmail.com>
p4raw-id: //depot/perl@29621
Rafael Garcia-Suarez [Sun, 24 Dec 2006 09:28:22 +0000]
The PERLIO environment variable, when unset, doesn't default to :stdio.
p4raw-id: //depot/perl@29620
Rafael Garcia-Suarez [Sat, 23 Dec 2006 16:15:50 +0000]
Fix #6006 for taint mode too.
p4raw-id: //depot/perl@29619
Rafael Garcia-Suarez [Sat, 23 Dec 2006 15:52:27 +0000]
Remove -T from op/subst.t, and add a new test file, substT.t,
to run the s/// tests under taint mode; mark the test for
RT #6006 as TODO, but only with -T.
p4raw-id: //depot/perl@29618
Rafael Garcia-Suarez [Sat, 23 Dec 2006 15:38:50 +0000]
A better fix for RT #6006: revert change 29616, which was bogus,
and add OP_SCOPE to the list of "dangerous" ops that might appear
on the right side of s///.
However, this fix does not work under taint mode, and op/subst.t
has -T.
p4raw-id: //depot/perl@29617
Rafael Garcia-Suarez [Sat, 23 Dec 2006 13:52:02 +0000]
Fix RT #6006: Regexp replaces using large replacement variables fail
some of the time (i.e. when the substitution contains something
like ${10}). Patch derived from a patch by Hugo van der Sanden;
added also a second test (Yves Orton already added a TODO test
for this.)
p4raw-id: //depot/perl@29616
Rafael Garcia-Suarez [Sat, 23 Dec 2006 12:31:26 +0000]
Upgrade to Module::Build 0.2806
p4raw-id: //depot/perl@29615
Rafael Garcia-Suarez [Sat, 23 Dec 2006 11:45:28 +0000]
syswrite() wasn't reporting a warning when writing to a filehandle
opened only for input.
p4raw-id: //depot/perl@29614
Rafael Garcia-Suarez [Fri, 22 Dec 2006 15:44:51 +0000]
The Perl 6 specification is still evolving, so don't refer to it.
p4raw-id: //depot/perl@29613
Steve Peters [Fri, 22 Dec 2006 15:18:02 +0000]
miniperl may not have been successfully built before running a
"make distclean".
p4raw-id: //depot/perl@29612
Steve Peters [Fri, 22 Dec 2006 14:53:17 +0000]
Silence an unused variable warning in sv.c.
p4raw-id: //depot/perl@29611
Rafael Garcia-Suarez [Fri, 22 Dec 2006 11:30:47 +0000]
Unshare the shared string gp_file_hek only if it exists in a GP.
Not doing this will produce invalid frees in the global destruction
process, probably because the empty string will get freed several
times. This should be safe since gp_file_hek is only assigned a
new value with share_hek().
p4raw-id: //depot/perl@29610
Rafael Garcia-Suarez [Thu, 21 Dec 2006 10:24:05 +0000]
Upgrade to Text::Balanced 2.0.0
p4raw-id: //depot/perl@29609
Jerry D. Hedden [Wed, 20 Dec 2006 10:30:21 +0000]
threads 1.57
From: "Jerry D. Hedden" <jdhedden@yahoo.com>
Message-ID: <
20061220183021.79793.qmail@web30205.mail.mud.yahoo.com>
p4raw-id: //depot/perl@29608
Jan Dubois [Wed, 20 Dec 2006 18:53:51 +0000]
allow blib.pm to be used for testing Win32 module (again)
Message-ID: <vjtjo21l5ksgeab6k9l46d9pvlg5vf86rt@4ax.com>
p4raw-id: //depot/perl@29607
Ilya Zakharevich [Mon, 18 Dec 2006 01:13:59 +0000]
SelfLoader after fork()
Message-ID: <
20061218091359.GA14985@powdermilk.math.berkeley.edu>
p4raw-id: //depot/perl@29606
Jan Dubois [Wed, 20 Dec 2006 15:41:00 +0000]
Implement killpg() for MSWin32
Message-ID: <m8hjo2pasv3hdbhd1sj9sasen7tm38hrhp@4ax.com>
p4raw-id: //depot/perl@29605
Yves Orton [Wed, 20 Dec 2006 16:00:33 +0000]
Re: [PATCH] Making adding binary files possible
Message-ID: <
9b18b3110612200700k48157cd1n1f2305f306b7de32@mail.gmail.com>
plus matching updates to Makefile.SH and patching.pod
p4raw-id: //depot/perl@29604
H.Merijn Brand [Wed, 20 Dec 2006 12:58:41 +0000]
Open Configure to the world of 64bitonly environments
p4raw-id: //depot/perl@29603
Rafael Garcia-Suarez [Wed, 20 Dec 2006 11:15:29 +0000]
Fix leaks in label strings allocation in COPs
p4raw-id: //depot/perl@29601
Rafael Garcia-Suarez [Wed, 20 Dec 2006 09:30:06 +0000]
Don't free the original label string when assigning it to cop_label,
since it comes directly from the parser. (That was making op/goto.t
fail under threads and without debugging on some platforms.)
p4raw-id: //depot/perl@29600
Jerry D. Hedden [Tue, 19 Dec 2006 10:30:47 +0000]
threads::shared 1.06
From: "Jerry D. Hedden" <jdhedden@yahoo.com>
Message-ID: <525867.40748.qm@web30207.mail.mud.yahoo.com>
p4raw-id: //depot/perl@29599
Jan Dubois [Mon, 18 Dec 2006 21:37:25 +0000]
Update cwd() to return the "short" pathname if the long one doesn't fit the codepage
Message-ID: <3rteo219or8hqr511e4vg1fnsgvgemb4sh@4ax.com>
p4raw-id: //depot/perl@29598
Jan Dubois [Tue, 19 Dec 2006 13:49:40 +0000]
let readdir() return the alternate (short) filename if the long name is outside the current codepage
Message-ID: <8nmgo2dr5gtmv00gtfu07g7rdgitr859jk@4ax.com>
p4raw-id: //depot/perl@29597
Jerry D. Hedden [Mon, 18 Dec 2006 10:18:28 +0000]
Re: 'localtime' not thread-safe under HP-UX B.11.11/64
From: "Jerry D. Hedden" <jdhedden@yahoo.com>
Message-ID: <31291.28361.qm@web30205.mail.mud.yahoo.com>
p4raw-id: //depot/perl@29596
Rafael Garcia-Suarez [Tue, 19 Dec 2006 15:22:24 +0000]
Proper cleanup for Errno.pm
p4raw-id: //depot/perl@29595
Rafael Garcia-Suarez [Tue, 19 Dec 2006 15:14:47 +0000]
Fix paths of pack.pl in patching.pod
p4raw-id: //depot/perl@29594
Jos I. Boumans [Fri, 15 Dec 2006 20:51:45 +0000]
Making adding binary files possible
From: "Jos Boumans" <kane@dwim.org>
Message-ID: <19978.80.127.35.68.
1166212305.squirrel@webmail.xs4all.nl>
with a few tweaks
p4raw-id: //depot/perl@29593
Rafael Garcia-Suarez [Tue, 19 Dec 2006 10:50:13 +0000]
Add a regression test to verify that POSIX::AUTOLOAD works,
and silence a potential warning from it.
p4raw-id: //depot/perl@29592