SADAHIRO Tomoyuki [Mon, 9 Jan 2006 12:53:15 +0000]
L<PerlIO> and Pod::Html
Message-Id: <
20060109125220.3BB0.BQW10602@nifty.com>
p4raw-id: //depot/perl@27061
Steven Schubiger [Thu, 2 Feb 2006 16:02:41 +0000]
s/Nullcv/NULL/g
Message-ID: <
20060202150241.GF12591@accognoscere.homeunix.org>
Date: Thu, 2 Feb 2006 16:02:41 +0100
p4raw-id: //depot/perl@27060
Nicholas Clark [Fri, 3 Feb 2006 13:06:00 +0000]
Change PL_perlio_fd_refcnt from a fixed size static array to a pointer
to a dynamic array.
p4raw-id: //depot/perl@27059
Steve Peters [Thu, 2 Feb 2006 20:57:01 +0000]
It would be nice to have tests in B for the functions getting changed.
p4raw-id: //depot/perl@27057
Steve Peters [Thu, 2 Feb 2006 20:42:56 +0000]
Death to magic characters! No pun intended.
p4raw-id: //depot/perl@27056
Steven Schubiger [Thu, 2 Feb 2006 10:38:49 +0000]
Re: [PATCH] s/Null(av|ch)/NULL/g
Message-ID: <
20060202093849.GD12591@accognoscere.homeunix.org>
p4raw-id: //depot/perl@27054
Nicholas Clark [Thu, 2 Feb 2006 12:31:30 +0000]
gv_fetchpvn_flags ranks highly in the profile, and the ::/' scanning
loop is iterated over millions of times. Add a flag GV_NOTQUAL purely
as an optimisation, when the caller is passing in a string that is
known not to contain any package separators.
p4raw-id: //depot/perl@27053
Rafael Garcia-Suarez [Thu, 2 Feb 2006 11:22:43 +0000]
Update to Module::CoreList 3.04.
Cleanup corelist manpage.
Remove -w in corelist shebang to avoid spurious warnings.
p4raw-id: //depot/perl@27052
Gisle Aas [Thu, 2 Feb 2006 10:05:32 +0000]
perl-5.8.8 also contained change 27040.
p4raw-id: //depot/perl@27051
Nicholas Clark [Thu, 2 Feb 2006 00:36:38 +0000]
Yet more instances of gv_fetchpv... that should be GV_ADD rather than
TRUE.
Convert two gv_fetchpvn_flags to gv_fetchpvs.
p4raw-id: //depot/perl@27050
Nicholas Clark [Thu, 2 Feb 2006 00:24:06 +0000]
And another assumption.
p4raw-id: //depot/perl@27049
Nicholas Clark [Wed, 1 Feb 2006 23:38:53 +0000]
Since PERL_COPY_ON_WRITE became PERL_OLD_COPY_ON_WRITE it seems that
no-one has been testing it, as there was a const too far.
p4raw-id: //depot/perl@27048
Steve Peters [Wed, 1 Feb 2006 23:34:47 +0000]
Doc nit found by Stephen Gordon.
p4raw-id: //depot/perl@27047
Nicholas Clark [Wed, 1 Feb 2006 23:03:11 +0000]
doubleplusoops. Apart from the documented NUL termination assumption.
Now removed.
p4raw-id: //depot/perl@27046
Nicholas Clark [Wed, 1 Feb 2006 22:39:29 +0000]
References can't have SVp_IOK or SVp_NOK set, so no need for all the
conditional code to copy IVX and NVX if true. "You cannot get here".
p4raw-id: //depot/perl@27045
Nicholas Clark [Wed, 1 Feb 2006 22:03:37 +0000]
Oops. 1 NUL termination assumption remains in Perl_gv_fetchpvn_flags
p4raw-id: //depot/perl@27044
Nicholas Clark [Wed, 1 Feb 2006 20:26:46 +0000]
Correct my comment, so that it's actually useful. :-)
p4raw-id: //depot/perl@27043
Steve Purkis [Fri, 20 Jan 2006 12:35:06 +0000]
[[:...:]] is equivalent to \p{...}, not [:...:], tweaked from
Subject: Re: [:...:] and \p{...} character class equivalence in utf8 regexps
Message-Id: <
0DAE5956-3ECC-4692-A0C9-C62C8F790C97@multimap.com>
Date: Fri, 20 Jan 2006 12:35:06 -0500
p4raw-id: //depot/perl@27042
Rafael Garcia-Suarez [Wed, 1 Feb 2006 18:07:26 +0000]
Improve a bit the documentation for kill() with SIGNAL==0
p4raw-id: //depot/perl@27039
Nicholas Clark [Wed, 1 Feb 2006 18:05:20 +0000]
Mmm, it looks like it's a bad plan to rebuild the TOC from a tree that
has other modules in ext/
TODO - fix buildtoc to cross check with MANIFEST.
p4raw-id: //depot/perl@27038
Nicholas Clark [Wed, 1 Feb 2006 17:53:28 +0000]
Integrate perl588delta.pod from maint, and re-run pod/buildtoc.
p4raw-id: //depot/perl@27037
p4raw-branched: from //depot/maint-5.8/perl@27036 'branch in'
pod/perl588delta.pod (@27012..)
p4raw-edited: from //depot/maint-5.8/perl@24695 'edit in'
vms/descrip_mms.template (@24616..)
p4raw-integrated: from //depot/maint-5.8/perl@24695 'edit in'
Makefile.SH (@23936..) pod.lst pod/perltoc.pod win32/Makefile
win32/makefile.mk (@24616..) 'merge in' win32/pod.mak
(@23573..) MANIFEST (@24611..)
Rafael Garcia-Suarez [Wed, 1 Feb 2006 17:49:53 +0000]
Doc nit by Andy Dougherty
p4raw-id: //depot/perl@27036
Rafael Garcia-Suarez [Wed, 1 Feb 2006 16:14:07 +0000]
autouse actually requires perl 5.6 at least.
p4raw-id: //depot/perl@27035
Rafael Garcia-Suarez [Wed, 1 Feb 2006 15:51:35 +0000]
Avoid "Prototype mismatch" warnings with autouse.
p4raw-id: //depot/perl@27034
Rafael Garcia-Suarez [Wed, 1 Feb 2006 13:22:26 +0000]
Fix corelist data for 5.8.8
p4raw-id: //depot/perl@27033
Rafael Garcia-Suarez [Wed, 1 Feb 2006 12:41:37 +0000]
More history
p4raw-id: //depot/perl@27032
Rafael Garcia-Suarez [Wed, 1 Feb 2006 12:30:14 +0000]
Update to Module::CoreList 2.03 (includes 5.8.8 and 5.9.3)
p4raw-id: //depot/perl@27031
Jim Cromie [Tue, 31 Jan 2006 03:33:46 +0000]
[patch] arena rework - unify arenaroots
Message-ID: <
43DF3D0A.3040102@gmail.com>
Date: Tue, 31 Jan 2006 03:33:46 -0700
p4raw-id: //depot/perl@27030
Nicholas Clark [Tue, 31 Jan 2006 22:59:27 +0000]
Make Perl_gv_fetchpvn_flags actually heed the passed in length.
This means that \0 bytes in symbolic references now work.
p4raw-id: //depot/perl@27028
Steve Peters [Tue, 31 Jan 2006 18:42:31 +0000]
Intel C Compiler 9.0 hatefully defines __GNUC__. Unfortunately,
icc does not support gcc built-ins like __attribiute__.
p4raw-id: //depot/perl@27027
p.boven@sara.nl [Tue, 31 Jan 2006 05:22:28 +0000]
[perl #38380] Typo in perlfunc.pod
From: p.boven@sara.nl (via RT) <perlbug-followup@perl.org>
Message-ID: <rt-3.0.11-38380-129047.11.
4217468041333@perl.org>
Date: Tue, 31 Jan 2006 05:22:28 -0800
p4raw-id: //depot/perl@27022
Nicholas Clark [Tue, 31 Jan 2006 15:10:37 +0000]
Eliminate save_threadsv, as it is never linked to, and all it does
post 5005 threads is croak().
p4raw-id: //depot/perl@27021
Yves Orton [Tue, 31 Jan 2006 09:55:25 +0000]
Re: [PATCH] Re: replacing "inuse" Win files (was Re: Help with a Cwd.pm build error)
Message-ID: <
9b18b3110601310055h7aeb9aa1gdc2d63cc56d97768@mail.gmail.com>
p4raw-id: //depot/perl@27016
John E. Malmberg [Tue, 31 Jan 2006 04:37:27 +0000]
patch@26996 - patch around stat("_NLA0:") bug
From: "John E. Malmberg" <wb8tyw@qsl.net>Date: Mon, 30 Jan 2006 23:21:34 -0500
Message-id: <
43DEE5CE.7060108@qsl.net>
p4raw-id: //depot/perl@27015
Nicholas Clark [Tue, 31 Jan 2006 00:15:26 +0000]
Silence a warning from the MS compiler about signed/unsigned mismatch.
p4raw-id: //depot/perl@27014
Rafael Garcia-Suarez [Mon, 30 Jan 2006 17:47:56 +0000]
Add perl594delta
p4raw-id: //depot/perl@27010
Nicholas Clark [Mon, 30 Jan 2006 17:12:54 +0000]
Change the author aliasing to reflect updates to AUTHORS.
p4raw-id: //depot/perl@27009
Rafael Garcia-Suarez [Mon, 30 Jan 2006 16:58:47 +0000]
Bump version number to 5.9.4
Regenerate template config in Porting
p4raw-id: //depot/perl@27007
Nicholas Clark [Mon, 30 Jan 2006 16:50:38 +0000]
Silence another Irix compiler warning.
p4raw-id: //depot/perl@27006
Nicholas Clark [Mon, 30 Jan 2006 16:29:38 +0000]
Avoid a warning from the Irix C compiler.
p4raw-id: //depot/perl@27005
Steve Peters [Mon, 30 Jan 2006 15:04:50 +0000]
Teach the Makefile how to cleanup config.over.
p4raw-id: //depot/perl@27004
Nicholas Clark [Mon, 30 Jan 2006 13:42:56 +0000]
Code assumes that *FOO{SCALAR} will always return a scalar reference,
so make it so, creating $FOO if necessary.
(Effectively this is a policy decision that PERL_DONT_CREATE_GVSV is
visible to XS code, but isn't visible to Perl code)
p4raw-id: //depot/perl@27002
Steve Peters [Mon, 30 Jan 2006 13:25:04 +0000]
Add -DPTR_IS_LONG when building with -Duse64bitint on Solaris/i386.
Otherwise, the build fails in POSIX.
p4raw-id: //depot/perl@27001
Nicholas Clark [Mon, 30 Jan 2006 12:59:12 +0000]
Use a union for storing the shared hash key reference count, rather
than messy pointer hacks.
p4raw-id: //depot/perl@27000
Yitzchak Scott-Thoennes [Mon, 30 Jan 2006 01:06:24 +0000]
[DOC PATCH] $^X isn't necessarily argv[0]
Message-ID: <
20060130090624.GA3760@efn.org>
p4raw-id: //depot/perl@26999
Jarkko Hietaniemi [Sun, 29 Jan 2006 23:45:03 +0000]
maint and blead: unicos.sh hints
Message-Id: <
200601292145.k0TLj3mW320311@kosh.hut.fi>
p4raw-id: //depot/perl@26998
Nicholas Clark [Mon, 30 Jan 2006 00:18:52 +0000]
Assert that IVs and NVs can never be tainted.
p4raw-id: //depot/perl@26997
Nicholas Clark [Sun, 29 Jan 2006 22:00:11 +0000]
Solaris seems not to enjoy initialising constants from SIG_??? on 64
bit builds.
p4raw-id: //depot/perl@26996
Steve Peters [Sun, 29 Jan 2006 20:43:44 +0000]
Upgrade to podlators-2.0.3
p4raw-id: //depot/perl@26995
Rafael Garcia-Suarez [Sun, 29 Jan 2006 18:31:52 +0000]
Email change for Steven Schubiger
p4raw-id: //depot/perl@26994
Philippe M. Chiasson [Thu, 26 Jan 2006 13:53:07 +0000]
Re: [patch] s/ar ts/ar s/g
From: "Philippe M. Chiasson" <gozer@ActiveState.com>
Message-ID: <
43D944C3.60101@ActiveState.com>
p4raw-id: //depot/perl@26993
Nicholas Clark [Sun, 29 Jan 2006 00:25:04 +0000]
C compilers don't let you return void. Pay attention gcc, and
stop aping Shildt.
p4raw-id: //depot/perl@26991
Nicholas Clark [Sun, 29 Jan 2006 00:09:10 +0000]
Avoid using space to create 2 functions just to force mathoms to be
linked. Instead, force the link via the perl interpreter structure
via a function pointer that gets replaced at run time anyway.
p4raw-id: //depot/perl@26990
Nicholas Clark [Sat, 28 Jan 2006 17:46:56 +0000]
Call gv_fetchpvn_flags where we already know the length.
p4raw-id: //depot/perl@26989
Steve Peters [Sat, 28 Jan 2006 17:25:03 +0000]
Upgrade to podlators-2.0.2
p4raw-id: //depot/perl@26988
Nicholas Clark [Sat, 28 Jan 2006 17:15:21 +0000]
Squeeze more flag manipulations together in sv_setsv_flags.
p4raw-id: //depot/perl@26987
Steve Peters [Sat, 28 Jan 2006 16:50:32 +0000]
Upgrade to CPAN-1.83_59
p4raw-id: //depot/perl@26986
Steve Peters [Sat, 28 Jan 2006 16:43:43 +0000]
Upgrade to Archive-Tar-1.28
p4raw-id: //depot/perl@26985
Nicholas Clark [Sat, 28 Jan 2006 15:44:04 +0000]
Add gv_fetchpvs, which uses STR_WITH_LEN to call gv_fetchpvn_flags.
Many strlen()s saved.
p4raw-id: //depot/perl@26983
Nicholas Clark [Sat, 28 Jan 2006 14:46:35 +0000]
Merge the IOK and NOK clauses together in sv_setsv_flags.
p4raw-id: //depot/perl@26982
Nicholas Clark [Sat, 28 Jan 2006 14:23:49 +0000]
The flags manipulation in sv_setsv_flags can be more efficient.
p4raw-id: //depot/perl@26981
Nicholas Clark [Sat, 28 Jan 2006 14:02:10 +0000]
PERL_DONT_CREATE_GVSV wasn't as effective as it should have been.
p4raw-id: //depot/perl@26980
Nicholas Clark [Sat, 28 Jan 2006 13:38:28 +0000]
By changing SvVOK() from returning 0/1 to 0/pointer-to-magic we can
save duplicating mg_find calls, without changing the semantics in any
boolean context.
p4raw-id: //depot/perl@26979
Nicholas Clark [Sat, 28 Jan 2006 12:49:15 +0000]
Pull the other large chunk of PVGV assignment code into S_pvgv_assign.
There seems to be some duplication with S_glob_assign.
p4raw-id: //depot/perl@26978
Nicholas Clark [Sat, 28 Jan 2006 12:26:47 +0000]
Remove goto glob_assign;
p4raw-id: //depot/perl@26977
Nicholas Clark [Sat, 28 Jan 2006 12:08:01 +0000]
Pull glob_assign out from sv_setsv_flags into a static function.
p4raw-id: //depot/perl@26976
Rafael Garcia-Suarez [Sat, 28 Jan 2006 11:00:30 +0000]
Update Changes
p4raw-id: //depot/perl@26975
Rafael Garcia-Suarez [Sat, 28 Jan 2006 10:57:04 +0000]
Add perl 5.9.3 in release list
p4raw-id: //depot/perl@26974
Gisle Aas [Fri, 27 Jan 2006 07:21:16 +0000]
Re: Make Passive mode the default for Net::FTP
Message-ID: <lrslr91ytv.fsf@caliper.activestate.com>
p4raw-id: //depot/perl@26970
Rafael Garcia-Suarez [Fri, 27 Jan 2006 18:27:22 +0000]
Oops. Forgot to add this file.
p4raw-id: //depot/perl@26969
Rafael Garcia-Suarez [Fri, 27 Jan 2006 18:07:54 +0000]
Regenerate META.yml
p4raw-id: //depot/perl@26968
Rafael Garcia-Suarez [Fri, 27 Jan 2006 18:04:17 +0000]
Regenerate Changes; remove ^M chars in Changes
p4raw-id: //depot/perl@26967
Rafael Garcia-Suarez [Fri, 27 Jan 2006 17:47:32 +0000]
Remove patch number
p4raw-id: //depot/perl@26966
Rafael Garcia-Suarez [Fri, 27 Jan 2006 17:44:48 +0000]
Mention ptardiff in perldelta.
p4raw-id: //depot/perl@26965
Rafael Garcia-Suarez [Fri, 27 Jan 2006 17:37:27 +0000]
Don't install ptardiff in the module tree
p4raw-id: //depot/perl@26964
Rafael Garcia-Suarez [Fri, 27 Jan 2006 17:19:52 +0000]
Install the ptardiff utility that comes with Archive::Tar
p4raw-id: //depot/perl@26963
Rafael Garcia-Suarez [Fri, 27 Jan 2006 17:04:35 +0000]
Install nit : README.e2x should be installed, the other READMEs shouldn't
p4raw-id: //depot/perl@26962
Rafael Garcia-Suarez [Fri, 27 Jan 2006 16:31:33 +0000]
Regenerate PODs
p4raw-id: //depot/perl@26961
Rafael Garcia-Suarez [Fri, 27 Jan 2006 15:35:01 +0000]
Some spelling fixes by David Landgren.
Mention perltodo.
p4raw-id: //depot/perl@26960
Rafael Garcia-Suarez [Fri, 27 Jan 2006 15:26:05 +0000]
Revert change 24461, now that change 26757 allows TEST
to accept extra output from module tests
p4raw-id: //depot/perl@26959
Yitzchak Scott-Thoennes [Wed, 25 Jan 2006 19:16:09 +0000]
quiet gcc Info message in cygwin
Message-ID: <
20060126031608.GA2588@efn.org>
p4raw-id: //depot/perl@26958
Steve Hay [Fri, 27 Jan 2006 11:09:02 +0000]
Remove note about #elif, see :
Subject: Re: Change 26954: Simplify nested '#ifdef' by replacing it with '#elif defined()'
Message-ID: <
43D9FF4E.6010501@uk.radan.com>
p4raw-id: //depot/perl@26957
Nicholas Clark [Fri, 27 Jan 2006 10:36:05 +0000]
A description of the new constants.
p4raw-id: //depot/perl@26955
Gisle Aas [Fri, 27 Jan 2006 08:54:04 +0000]
Simplify nested '#ifdef' by replacing it with '#elif defined()'
p4raw-id: //depot/perl@26954
Gisle Aas [Fri, 27 Jan 2006 08:49:34 +0000]
Safer implementation of RTRIM().
Don't rely on SvCUR to be sane unless SvPOK.
p4raw-id: //depot/perl@26953
Philippe M. Chiasson [Thu, 26 Jan 2006 13:39:59 +0000]
Change 26781/26830 (SvRTRIM should also write a new '\0' at the end) can cause segfaults
From: "Philippe M. Chiasson" <gozer@ActiveState.com>
Message-ID: <
43D941AF.1060503@ActiveState.com>
Date: Thu, 26 Jan 2006 13:39:59 -0800
p4raw-id: //depot/perl@26952
Steve Peters [Thu, 26 Jan 2006 19:23:48 +0000]
Fix a few nits in perl593delta
p4raw-id: //depot/perl@26951
Rafael Garcia-Suarez [Thu, 26 Jan 2006 18:48:27 +0000]
miniperl isn't enough anymore to run pod2man.
p4raw-id: //depot/perl@26950
Nicholas Clark [Thu, 26 Jan 2006 18:35:56 +0000]
2 spelling corrections, and rephrase the entry on perlivp (better
grammer, more succinct)
p4raw-id: //depot/perl@26949
Rafael Garcia-Suarez [Thu, 26 Jan 2006 18:14:26 +0000]
More work on perldelta
p4raw-id: //depot/perl@26948
Yves Orton [Thu, 26 Jan 2006 15:40:42 +0000]
Make dump and -u croak on Win32 and Cygwin to avoid confusion
such as was seen here:
Subject: Weird Win32 error: "This application has requested the Runtime to terminate it in an unusual way."
Message-ID: <
9b18b3110601260640x34b2442cnf5379d2f83b74187@mail.gmail.com>
p4raw-id: //depot/perl@26947
Nicholas Clark [Thu, 26 Jan 2006 14:57:42 +0000]
SvREADONLY_off inside the magic routines is also dangerous on shared
hash key scalars, so decompose them to normal scalars if they ever
get here.
p4raw-id: //depot/perl@26946
Rafael Garcia-Suarez [Thu, 26 Jan 2006 11:49:22 +0000]
assertions.pm is dual-lived
p4raw-id: //depot/perl@26945
Stephen McCamant [Tue, 24 Jan 2006 07:23:00 +0000]
aelemfast optimization opclass problem (c.f. [perl #38279])
Message-ID: <17365.51166.604020.571992@conquest.OCF.Berkeley.EDU>
p4raw-id: //depot/perl@26943
Gisle Aas [Tue, 24 Jan 2006 05:01:28 +0000]
Simplify &IO::Handle::say
Message-ID: <lrzmll7pav.fsf@caliper.activestate.com>
p4raw-id: //depot/perl@26941
Rafael Garcia-Suarez [Tue, 24 Jan 2006 15:10:27 +0000]
Fix a regression on suidperl : the error message "No #! line" was
produced in erroneous cases, because we used to read values in a
string buffer that pointed to nothing.
p4raw-id: //depot/perl@26940
Steve Hay [Tue, 24 Jan 2006 12:13:15 +0000]
Remove d_nanosleep from win32 configs
Change 19899 removed them before, but they have accidentally crept back.
p4raw-id: //depot/perl@26936
Steve Hay [Tue, 24 Jan 2006 09:22:44 +0000]
SPRINTF_RETURNS_STRLEN isn't really a compile-time option
From: Gisle Aas <gisle@ActiveState.com>
Date: 23 Jan 2006 02:12:04 -0800
Message-ID: <lr3bjfb6dn.fsf@caliper.activestate.com>
Subject: RE: SPRINTF_RETURNS_STRLEN isn't really a compile-time option
From: "Jan Dubois" <jand@ActiveState.com>
Date: Mon, 23 Jan 2006 11:30:51 -0800
Message-ID: <
008001c62053$86adf680$2217a8c0@candy>
p4raw-id: //depot/perl@26935
Rafael Garcia-Suarez [Mon, 23 Jan 2006 19:24:06 +0000]
More work on perldelta
p4raw-id: //depot/perl@26933
Nicholas Clark [Mon, 23 Jan 2006 19:06:41 +0000]
constant_add_symbol needs to be static, else static builds can fail at
link time. (*How* many meanings is "static" overloaded with?)
p4raw-id: //depot/perl@26932
projects / p5sagit/p5-mst-13.2.git / log