Yves Orton [Fri, 20 Oct 2006 12:38:13 +0000]
only do \C hackery if \C is seen
Subject: Re: utf8 regexp performance problem
Message-ID: <
9b18b3110610200338k5df1d3afpf829fbc0f3663a75@mail.gmail.com>
p4raw-id: //depot/perl@29063
Alexander Gough [Fri, 20 Oct 2006 02:05:20 +0000]
B:: changes for UNITCHECK blocks
Message-ID: <
20061020010520.GC12290@the.earth.li>
p4raw-id: //depot/perl@29062
Rafael Garcia-Suarez [Fri, 20 Oct 2006 10:32:20 +0000]
Let's undocument -Dusesfio in INSTALL.
p4raw-id: //depot/perl@29061
Rafael Garcia-Suarez [Fri, 20 Oct 2006 10:20:18 +0000]
More leak fixes, by Jarkko
p4raw-id: //depot/perl@29060
Dave Mitchell [Fri, 20 Oct 2006 10:06:57 +0000]
zero length $1 took too long to detect utf8-ness
p4raw-id: //depot/perl@29059
Rafael Garcia-Suarez [Fri, 20 Oct 2006 09:04:49 +0000]
Silence some warnings
p4raw-id: //depot/perl@29058
Yves Orton [Wed, 18 Oct 2006 20:51:41 +0000]
Re: Off by one in the trie code?
Message-ID: <
9b18b3110610181151i3ca438cdied769ebaa4255079@mail.gmail.com>
1. code necessary to make patterns with interpolated vars behave
correctly under lexical re 'debug', including additional tests.
2. changes necessary to resolve the off by one error,
3. tweaks to re.pm to document that re 'debug' is lexical,
p4raw-id: //depot/perl@29057
Yves Orton [Wed, 18 Oct 2006 20:51:41 +0000]
Re: Off by one in the trie code?
Message-ID: <
9b18b3110610181151i3ca438cdied769ebaa4255079@mail.gmail.com>
change test files that do a require "./test.pl"; without a BEGIN block
to ensure prototypes are seen, plus fix any breakage this reveals.
p4raw-id: //depot/perl@29056
H.Merijn Brand [Thu, 19 Oct 2006 16:40:08 +0000]
<tr> isnb't the same as C<tr>, and <tr> looks so stupid in man
p4raw-id: //depot/perl@29055
Rafael Garcia-Suarez [Thu, 19 Oct 2006 16:20:41 +0000]
Trim perltodo.
p4raw-id: //depot/perl@29054
Alexander Gough [Thu, 19 Oct 2006 13:04:12 +0000]
stab at UNITCHECK blocks
Message-ID: <
20061019120412.GA12290@the.earth.li>
p4raw-id: //depot/perl@29053
Jos Boumans [Fri, 13 Oct 2006 17:12:57 +0000]
Add Log::Message and Log::Message::Simple to the core
From: "Jos Boumans" <kane@xs4all.net>
Message-ID: <13003.80.127.35.68.
1160752377.squirrel@webmail.xs4all.nl>
p4raw-id: //depot/perl@29052
Rafael Garcia-Suarez [Thu, 19 Oct 2006 12:17:18 +0000]
Make prototype("CORE::mkdir") work. That's a special case,
since from the OA_* flags we can't figure out which parameter
is going to default to $_.
p4raw-id: //depot/perl@29051
Rafael Garcia-Suarez [Thu, 19 Oct 2006 12:08:24 +0000]
Memory leak fix, by Jarkko
p4raw-id: //depot/perl@29050
Rafael Garcia-Suarez [Thu, 19 Oct 2006 10:27:20 +0000]
Adjust documentation for previous change
p4raw-id: //depot/perl@29049
Rafael Garcia-Suarez [Thu, 19 Oct 2006 10:26:54 +0000]
Allow the _ prototype character to be followed by optional arguments
p4raw-id: //depot/perl@29048
Rafael Garcia-Suarez [Thu, 19 Oct 2006 08:56:14 +0000]
Make all JAPH tests pass
p4raw-id: //depot/perl@29047
Rafael Garcia-Suarez [Thu, 19 Oct 2006 08:04:04 +0000]
Remove free() part of change #29045, which causes
panics from miniperl in FreeBSD/multithreaded
p4raw-link: @29045 on //depot/perl:
e47547a83f24974bde05453531e68d7dd8c5ec87
p4raw-id: //depot/perl@29046
Jarkko Hietaniemi [Wed, 18 Oct 2006 20:07:54 +0000]
blead valgrind finding
Message-ID: <
4536C1DA.4060600@iki.fi>
p4raw-id: //depot/perl@29045
Rafael Garcia-Suarez [Wed, 18 Oct 2006 20:26:37 +0000]
Make prototype("CORE::foo") return prototypes with _ when it should
(except for mkdir)
p4raw-id: //depot/perl@29044
Rafael Garcia-Suarez [Wed, 18 Oct 2006 19:37:01 +0000]
Add a regression test for change #29041
p4raw-link: @29041 on //depot/perl:
8c28b960db3547273cd8c89c0eaafc8e99cf70c1
p4raw-id: //depot/perl@29043
Nicholas Clark [Wed, 18 Oct 2006 19:11:55 +0000]
Using explicit printf-like APIs with a format of "%s" isn't the world's
most efficient idea.
p4raw-id: //depot/perl@29042
Rafael Garcia-Suarez [Wed, 18 Oct 2006 17:40:57 +0000]
Functions with a (_) prototype should behave as unary operators
(which is not the case of functions with a (;$) prototype)
p4raw-id: //depot/perl@29041
Nicholas Clark [Wed, 18 Oct 2006 17:35:06 +0000]
panic if we attempt to sv_setsv to or from an already freed SV.
p4raw-id: //depot/perl@29040
Rafael Garcia-Suarez [Wed, 18 Oct 2006 14:41:26 +0000]
Test for the interaction between ; and _ in prototypes
p4raw-id: //depot/perl@29039
Rafael Garcia-Suarez [Wed, 18 Oct 2006 14:34:26 +0000]
Document _ prototype in perlsub
p4raw-id: //depot/perl@29038
Rafael Garcia-Suarez [Wed, 18 Oct 2006 14:30:57 +0000]
Fix prototype to mimic built-in in perlsub examples
p4raw-id: //depot/perl@29037
Rafael Garcia-Suarez [Wed, 18 Oct 2006 14:01:59 +0000]
The _ character is only allowed to be at the end of prototypes
p4raw-id: //depot/perl@29036
Rafael Garcia-Suarez [Wed, 18 Oct 2006 12:54:34 +0000]
Improvements and fixes to the _ prototype
p4raw-id: //depot/perl@29035
Rafael Garcia-Suarez [Wed, 18 Oct 2006 08:49:01 +0000]
FAQ sync
p4raw-id: //depot/perl@29034
Dave Bailey [Tue, 17 Oct 2006 09:37:28 +0000]
[perl #40557] regexec.c saves context stack position improperly
From: Dave Bailey (via RT) <perlbug-followup@perl.org>
Message-ID: <rt-3.5.HEAD-4979-
1161103047-337.40557-75-0@perl.org>
p4raw-id: //depot/perl@29033
Rafael Garcia-Suarez [Tue, 17 Oct 2006 16:07:04 +0000]
First attempt at implementing the _ prototype
p4raw-id: //depot/perl@29032
Steve Peters [Tue, 17 Oct 2006 15:43:05 +0000]
Upgrade to Time-HiRes-1.94.
p4raw-id: //depot/perl@29031
SADAHIRO Tomoyuki [Tue, 17 Oct 2006 23:10:04 +0000]
Re: [PATCH] remove leaveit from toke.c:scan_const
Message-Id: <
20061017230958.41A2.BQW10602@nifty.com>
p4raw-id: //depot/perl@29030
Steve Peters [Tue, 17 Oct 2006 12:42:04 +0000]
Upgrade to CPAN-1.88_55.
p4raw-id: //depot/perl@29029
Rafael Garcia-Suarez [Tue, 17 Oct 2006 07:44:09 +0000]
Less brace nesting.
p4raw-id: //depot/perl@29028
Rafael Garcia-Suarez [Mon, 16 Oct 2006 12:59:07 +0000]
The "Unrecognized escape" warning can now be emitted in category
"regexp" after change 29026
p4raw-id: //depot/perl@29027
SADAHIRO Tomoyuki [Sun, 15 Oct 2006 22:23:00 +0000]
remove leaveit from toke.c:scan_const
Message-Id: <
20061015222223.BC38.BQW10602@nifty.com>
p4raw-id: //depot/perl@29026
SADAHIRO Tomoyuki [Sun, 15 Oct 2006 16:51:34 +0000]
Re: sprintf 64 test
Message-Id: <
20061015165052.77AD.BQW10602@nifty.com>
1. nullify 0 flag in integer conversions when precision is given
2. ignore space after a plus sign as a sign for a nonnegative number
3. make a negative precision through * working as if the precision
is omitted
p4raw-id: //depot/perl@29025
Nicholas Clark [Mon, 16 Oct 2006 04:41:30 +0000]
Fix change 28770 to cope with the strange 32 bit bool type on VMS.
(I should have seen this one coming - the two structures weren't
textually identical)
p4raw-id: //depot/perl@29024
Steve Peters [Sun, 15 Oct 2006 17:17:32 +0000]
Upgrade to Time-HiRes-1.93.
p4raw-id: //depot/perl@29023
Steve Peters [Sat, 14 Oct 2006 23:18:29 +0000]
Upgrade to CPAN-1.88_54.
p4raw-id: //depot/perl@29020
Steve Peters [Sat, 14 Oct 2006 23:11:23 +0000]
Upgrade to Digest-SHA-5.44.
p4raw-id: //depot/perl@29019
Nicholas Clark [Sat, 14 Oct 2006 23:06:52 +0000]
Avoid undefined behaviour for -DPERL_MEM_LOG by not using a
dereference of member of the structure being reallocated as part of
the size calculation. (This one was well hidden by macros)
p4raw-id: //depot/perl@29018
Steve Peters [Sat, 14 Oct 2006 23:00:35 +0000]
Upgrade to XSLoader-0.07.
p4raw-id: //depot/perl@29017
Nicholas Clark [Sat, 14 Oct 2006 22:40:00 +0000]
Avoid undefined behaviour for -DPERL_MEM_LOG by not using a direct
dereference of member of the structure being reallocated as part of
the size calculation. (There may be other similar bugs).
p4raw-id: //depot/perl@29016
Nicholas Clark [Sat, 14 Oct 2006 20:38:59 +0000]
Use Perl_croak_nocontext() rather than Perl_croak() for the snprintf()
and vsnprintf() macros.
p4raw-id: //depot/perl@29015
Nicholas Clark [Sat, 14 Oct 2006 14:32:28 +0000]
Fix typo.
p4raw-id: //depot/perl@29012
Jim Cromie [Fri, 13 Oct 2006 13:46:37 +0000]
Re: SVpvs vs SVpvn
Message-ID: <
452FED1D.1000504@gmail.com>
plus the last newSVpvn() in toke.c that his regexp missed.
p4raw-id: //depot/perl@29011
Steve Peters [Fri, 13 Oct 2006 14:11:04 +0000]
Upgrade to Time-HiRes-1.92.
p4raw-id: //depot/perl@29010
Nicholas Clark [Fri, 13 Oct 2006 13:53:16 +0000]
Correct one "parenthesis" to "parentheses".
p4raw-id: //depot/perl@29009
Rafael Garcia-Suarez [Fri, 13 Oct 2006 12:12:04 +0000]
Don't bother checking for bad characters in prototypes if we're not
going to warn about them.
p4raw-id: //depot/perl@29008
Robin Barker [Fri, 13 Oct 2006 11:49:11 +0000]
compiler warning noreturn
From: "Robin Barker" <Robin.Barker@npl.co.uk>
Message-ID: <
2C2E01334A940D4792B3E115F95B7226120A70@exchsvr1.npl.ad.local>
p4raw-id: //depot/perl@29007
Rafael Garcia-Suarez [Fri, 13 Oct 2006 09:15:43 +0000]
Fix nit in possessive quantifier descriptions.
Add some =head3 headings.
p4raw-id: //depot/perl@29006
Yves Orton [Thu, 12 Oct 2006 21:23:15 +0000]
Add possessive quantifiers to regex engine.
Message-ID: <
9b18b3110610121223m191e47ddtce3398cb0e8ba320@mail.gmail.com>
With doc tweaks
p4raw-id: //depot/perl@29005
Steve Peters [Thu, 12 Oct 2006 15:07:17 +0000]
Upgrade to PathTools-3.23.
p4raw-id: //depot/perl@29004
Jerry Hedden [Thu, 12 Oct 2006 07:47:37 +0000]
threads::shared 1.04
Message-ID: <
20061012144737.56305.qmail@web30207.mail.mud.yahoo.com>
p4raw-id: //depot/perl@29003
Jerry Hedden [Wed, 11 Oct 2006 05:57:43 +0000]
threads 1.44
Message-ID: <
20061011125743.68567.qmail@web30211.mail.mud.yahoo.com>
p4raw-id: //depot/perl@29002
Robin Barker [Wed, 11 Oct 2006 17:02:01 +0000]
fix format errors; was RE: g++ compile and make test 100%
From: "Robin Barker" <Robin.Barker@npl.co.uk>
Message-ID: <
2C2E01334A940D4792B3E115F95B7226120A69@exchsvr1.npl.ad.local>
p4raw-id: //depot/perl@29001
Robin Barker [Wed, 11 Oct 2006 16:58:10 +0000]
RE: g++ compile and make test 100%
From: "Robin Barker" <Robin.Barker@npl.co.uk>
Message-ID: <
2C2E01334A940D4792B3E115F95B7226120A68@exchsvr1.npl.ad.local>
p4raw-id: //depot/perl@29000
Yves Orton [Thu, 12 Oct 2006 14:45:25 +0000]
More regexp documentation
Message-ID: <
9b18b3110610120545m3002e17cqace30f908b0e2277@mail.gmail.com>
p4raw-id: //depot/perl@28999
Yves Orton [Thu, 12 Oct 2006 00:46:50 +0000]
Add Regex conditionals. Various bugfixes. More tests.
Message-ID: <
9b18b3110610111546j74ca490dg21bd9fd1e7e10d42@mail.gmail.com>
p4raw-id: //depot/perl@28998
Robin Barker [Wed, 11 Oct 2006 15:29:43 +0000]
RE: g++ compile and make test 100%
From: "Robin Barker" <Robin.Barker@npl.co.uk>
Message-ID: <
2C2E01334A940D4792B3E115F95B7226120A65@exchsvr1.npl.ad.local>
p4raw-id: //depot/perl@28997
Steve Peters [Wed, 11 Oct 2006 14:48:45 +0000]
Fix typo in NDBM_File.xs and bump up the version number.
p4raw-id: //depot/perl@28996
Rafael Garcia-Suarez [Wed, 11 Oct 2006 09:59:15 +0000]
Fix aliasing to an non-existent element of %+
(by Yves Orton)
p4raw-id: //depot/perl@28995
Rafael Garcia-Suarez [Wed, 11 Oct 2006 09:38:37 +0000]
Update comment (by Yves Orton), plus POD fixes
p4raw-id: //depot/perl@28994
Rafael Garcia-Suarez [Wed, 11 Oct 2006 08:03:03 +0000]
Missing % in format string
p4raw-id: //depot/perl@28993
Jarkko Hietaniemi [Wed, 11 Oct 2006 09:53:59 +0000]
ext re: C++: EXTERN_C needed, and where did sub install() go?
Message-Id: <
20061011065359.DFC4C8B3A1@seth.hut.fi>
p4raw-id: //depot/perl@28992
Steve Peters [Wed, 11 Oct 2006 00:52:58 +0000]
Grrr...moved the files, but forgot to update Soundex.pm
p4raw-id: //depot/perl@28991
Jarkko Hietaniemi [Tue, 10 Oct 2006 19:20:26 +0000]
C++: class is a keyword
Message-Id: <
20061010162026.190BE8B37C@seth.hut.fi>
p4raw-id: //depot/perl@28990
Rafael Garcia-Suarez [Tue, 10 Oct 2006 16:23:48 +0000]
Adjust test count for ExtUtils::Command
p4raw-id: //depot/perl@28989
Steve Peters [Tue, 10 Oct 2006 15:04:09 +0000]
Upgrade to ExtUtils-Manifest-1.49.
p4raw-id: //depot/perl@28988
Steve Peters [Tue, 10 Oct 2006 15:01:21 +0000]
Upgrade to ExtUtils-Command-1.12.
p4raw-id: //depot/perl@28987
Steve Peters [Tue, 10 Oct 2006 14:49:47 +0000]
Upgrade to CPAN-1.88_53.
p4raw-id: //depot/perl@28986
Anatoly Vorobey [Sun, 8 Oct 2006 17:58:16 +0000]
[perl #40473] sprintf width+precision fails on wide chars
From: Anatoly Vorobey (via RT) <perlbug-followup@perl.org>
Message-ID: <rt-3.5.HEAD-31259-
1160355496-1702.40473-75-0@perl.org>
p4raw-id: //depot/perl@28985
Steve Peters [Tue, 10 Oct 2006 14:33:53 +0000]
Upgrade to ExtUtils-Makemaker-6.31.
p4raw-id: //depot/perl@28984
Steve Peters [Tue, 10 Oct 2006 14:05:52 +0000]
Update to PathTools-3.22.
p4raw-id: //depot/perl@28983
H.Merijn Brand [Tue, 10 Oct 2006 14:02:16 +0000]
New shiny models
p4raw-id: //depot/perl@28982
Yves Orton [Mon, 9 Oct 2006 20:36:20 +0000]
Regexp Recurse by name.
Message-ID: <
9b18b3110610091136g48e5b154tf16d00d38e80a6dc@mail.gmail.com>
(with doc nits)
p4raw-id: //depot/perl@28981
Jarkko Hietaniemi [Tue, 10 Oct 2006 13:52:57 +0000]
enc2xs and C++: add extern "C" to data
Message-ID: <
452B7B89.201@iki.fi>
p4raw-id: //depot/perl@28980
Jarkko Hietaniemi [Tue, 10 Oct 2006 14:00:40 +0000]
Linux and Solaris hints: C++ vs dlerror
Message-ID: <
452B7D58.1090009@iki.fi>
p4raw-id: //depot/perl@28979
Steve Hay [Tue, 10 Oct 2006 10:25:06 +0000]
Updates for building on Win32 with Visual C++ 2005 Express Edition
Add some instructions in README.win32, clarify macro comments and add
missing macros to makefile.mk, and drop the GLOBEXEBUILD macro from
Makefile--the Express Edition of Visual C++ 2005 is free and includes
setargv.obj, so there is no need to support the .NET 2.0 SDK compiler
which didn't include it.
Unfortunately, the build still doesn't work yet--my attempt got as far
as building extensions, but miniperl.exe crashed. Hopefully this can
be sorted out soon.
p4raw-id: //depot/perl@28978
Craig A. Berry [Tue, 10 Oct 2006 03:29:50 +0000]
fix a doc typo and a warning typo in VMS::DCLsym
p4raw-id: //depot/perl@28977
Steve Peters [Mon, 9 Oct 2006 20:17:07 +0000]
Make g++ happy when compiling NDBM_File on a system that
is likely using the GDBM compatibility headers that
g++ (and other C++ implementations) cannot handle.
p4raw-id: //depot/perl@28976
Dave Mitchell [Mon, 9 Oct 2006 16:35:02 +0000]
fix bad free in do_exec3()
p4raw-id: //depot/perl@28975
Jarkko Hietaniemi [Mon, 9 Oct 2006 16:54:12 +0000]
Encode.xs: add an explicit cast to make g++ happier
Message-Id: <
20061009135412.47F7A8B3A0@seth.hut.fi>
p4raw-id: //depot/perl@28974
Rafael Garcia-Suarez [Mon, 9 Oct 2006 14:13:25 +0000]
Address update for Jerry D. Hedden
p4raw-id: //depot/perl@28973
Rafael Garcia-Suarez [Mon, 9 Oct 2006 12:53:40 +0000]
Update perldelta for recent regexp changes, based on a text by Yves Orton.
p4raw-id: //depot/perl@28972
Jarkko Hietaniemi [Mon, 9 Oct 2006 14:24:09 +0000]
Digest::SHA 5.43_01 - ANSIfy src/hmac.c
Message-ID: <
452A3159.3060003@iki.fi>
p4raw-id: //depot/perl@28971
Rafael Garcia-Suarez [Mon, 9 Oct 2006 10:32:06 +0000]
Add perl595delta
p4raw-id: //depot/perl@28970
Rafael Garcia-Suarez [Mon, 9 Oct 2006 10:11:09 +0000]
Bump version to 5.9.5
p4raw-id: //depot/perl@28969
Jarkko Hietaniemi [Mon, 9 Oct 2006 12:47:11 +0000]
Re: [PATCH] regexec.c: goto crossing initialization
Message-ID: <
452A1A9F.4080103@iki.fi>
p4raw-id: //depot/perl@28968
Yves Orton [Sun, 8 Oct 2006 20:14:24 +0000]
Re: [perl #40468] Not OK: perl 5.9.4 +patchaperlup: on i686-linux-64int 2.6.17-2-k7 (UNINSTALLED)
Message-ID: <
9b18b3110610081114g11dabafaw860181598ab54bd6@mail.gmail.com>
p4raw-id: //depot/perl@28967
Rafael Garcia-Suarez [Sun, 8 Oct 2006 19:39:00 +0000]
Revert change 28962 (at least until a better solution is found)
p4raw-id: //depot/perl@28966
Rafael Garcia-Suarez [Sun, 8 Oct 2006 17:07:42 +0000]
Clarifications on the usage of setlocale() and LC_NUMERIC to
affect the decimal point in output, by Jarkko.
p4raw-id: //depot/perl@28965
Nicholas Clark [Sun, 8 Oct 2006 00:25:07 +0000]
Document the growth policy, and Yves suggested a better arbitary
constant. (The original plan was add 80 if free space is less than 80,
which wasn't a noticable improvement. Hence 40 was 80/2 for plan B)
p4raw-id: //depot/perl@28964
Nicholas Clark [Sat, 7 Oct 2006 23:42:56 +0000]
Yves said that the XS version of Data::Dumper was inefficient because
it keeps triggering realloc() due to sv_cat(). Here's a rather brute
force approach to pre-stretching the buffer - if there are less than
40 bytes free, grow it by 50%.
Surprisingly effective for my test program
./perl -Ilib -MData::Dumper -MStorable=retrieve -we \
'print Dumper(retrieve(shift))' ~/.cpan/Metadata >/dev/null
Before
real 2m42.921s
user 1m43.321s
sys 0m55.611s
After
real 0m5.205s
user 0m4.885s
sys 0m0.255s
Same 25M of output, byte for byte. :-)
p4raw-id: //depot/perl@28963
Nicholas Clark [Sat, 7 Oct 2006 22:24:28 +0000]
No need to create a new magic vtable if it's all 0 pointers.
[And these are function pointers, so they're not NULL :-)]
p4raw-id: //depot/perl@28962
Nicholas Clark [Sat, 7 Oct 2006 22:06:19 +0000]
Make reference stringification (blessed and unblessed) about as fast as
is possible, because I'm told it's used quite frequently.
p4raw-id: //depot/perl@28961
Nicholas Clark [Sat, 7 Oct 2006 17:16:01 +0000]
Use packed addresses for the seen tracking hash, rather than
"stringified" references. These use less memory, and should be faster
as there is no call to sprintf().
p4raw-id: //depot/perl@28960
projects / p5sagit/p5-mst-13.2.git / log