| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
| |
p4raw-id: //depot/maint-5.005/perl@22270
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@22269
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@22214
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@22213
|
|
|
|
|
| |
and use -fPIC instead of -fpic under SPARC (under Linux)
p4raw-id: //depot/maint-5.005/perl@22212
|
|
|
|
|
| |
thanks to Jarkko Hietaniemi
p4raw-id: //depot/maint-5.005/perl@22211
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Here is a rather trivial patch that updates one of the files
in the "vos" subdir.
I keep an old copy of config_h.SH in there so I can see what
changes. (It has a funky name to prevent various clever bits
of Makefile trickery from recognizing it.) This lets me
update my own private copy of config.h easily. This patch
updates my copy of config_h.SH to be the same as the version
in the main directory; no changes are needed to any other
files.
I do not support the native (Configure-based) build method for
this release of perl. However, when perl5.005_03 is built
using the scripts found in the vos subdir, it builds just fine
on current releases of VOS.
p4raw-id: //depot/maint-5.005/perl@22210
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@22184
|
|
|
|
|
|
|
| |
thanks to Nicholas Clark
p4raw-id: //depot/maint-5.005/perl@22178
p4raw-integrated: from //depot/maint-5.6/perl-5.6.2@22158 'copy in'
hints/freebsd.sh
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@22158
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@22157
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@22156
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@22130
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@22066
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Patch by Paul Marquess <Paul.Marquess@btinternet.com>,
DB_File's maintainer. His description:
The patch is to work around a problem where some versions of
Linux have a C library with Berkeley DB version 2 embedded in
it. This makes life difficult if you want to build with a
version of Berkeley DB other than the one embedded in
libc. This problem is compounded by the way Perl used to
*always* include the Berkeley DB library when it was being
built. The DB_File patches solves the latter problem, by
retrofitting the fix to Configure et al that was included from
perl 5.6 on. I think your best course of action is to just
apply the 5.00503 patch to the 5.00504 source. I just tried
applying it and it seems ok.
p4raw-id: //depot/maint-5.005/perl@22065
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@22063
|
|
|
|
|
|
| |
From: "Marcus Holland-Moritz" <mhx-perl@gmx.net>
Subject: [PATCH perl-5.005xx] Locale-dependant test failures
p4raw-id: //depot/maint-5.005/perl@22062
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@22061
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@22060
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@21793
|
|
|
|
|
| |
Also, alphabetically sort MANIFEST
p4raw-id: //depot/maint-5.005/perl@21792
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@21791
|
|
|
|
|
|
|
|
|
|
| |
perl5.005_03 up to compiling on a modern linux system From:
Richard Clamp <richardc@unixbeard.net> To:
perl5-porters@perl.org, Chip Salzenberg <chip@pobox.com> Date:
Sun, 4 May 2003 20:53:33 +0100
First part: Teach makedepend about ggc3's extra output that
needs stripping (stolen from bleadperls makedepend.SH)
p4raw-id: //depot/maint-5.005/perl@21790
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@18681
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@18679
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@17648
|
|
|
|
|
| |
Subject: Re: [ID 20000421.002] Math::BigInt out of memory?
p4raw-id: //depot/maint-5.005/perl@17647
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@17646
|
|
|
|
|
| |
terminate -s switch processing only on C<-->, not on C<--foo>
p4raw-id: //depot/maint-5.005/perl@4788
|
|
|
|
|
| |
p4raw-link: @4785 on //depot/maint-5.005/perl: 3e8148feadabd0d0b1869fcc4d218a6475a5b0bc
p4raw-id: //depot/maint-5.005/perl@4787
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@4786
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
2821, 2916, 2930, 2931, 2937 2990, 3033, 3036, 3042, 3061, 3097,
3123, 3134, 3135, 3269, 3270, 3300, 3316, 3345 3350, 3353, 3362,
3363, 3396, 3419, 3423, 3445, 3446, 3447, 3449, 3450, 3451, 3487
3493, 3514, 3533, 3548, 3549, 3588, 3682, 3685, 3699, 3785,
3804, 3811, 3897, 4057 4102, 4103, 4230, 4401, 4420, 4504
--
enable better Win32::DomainName() by demand loading netapi32.dll
(from Jan Dubois)
--
win32_utime() on directories should use localtime() rather
than gmtime() (from Jan Dubois)
--
serious bug introduced by G_VOID changes in 5.003_96: scalar
eval"" did not pop stack correctly; C<$a = eval "(1,2)x1">
is one symptom of the problem
--
add $installarchlib/CORE to default linker search path on windows
--
fix memory leak in C<sub f { split ' ', "a b" } f() while 1>
--
fix memory leak in C<sub f { @_ = 1 } f() while 1>
--
avoid hiding child process window
--
optimizations could sometimes bypass bareword check
--
fix typo that caused INSTALLPRIVLIB to have doubled 'perl5'
--
Pod::Html tweak
From: jan.dubois@ibm.net (Jan Dubois)
To: perl5-porters@perl.org
Subject: [PATCH 5.005_58] pod2html: Missing chunk for VMS filenames
Date: Tue, 27 Jul 1999 22:14:12 +0200
Message-ID: <37a50af0.46171380@smtp1.ibm.net>
--
don't quit if =head* wasn't found (suggested by Roland Bauer
<roland.bauer@fff.at>)
--
avoid bug in win32_str_os_error() (from Jan Dubois)
--
applied suggested patch, along with later tweak
From: jan.dubois@ibm.net (Jan Dubois)
Date: Wed, 14 Jul 1999 23:53:43 +0200
Message-ID: <37a902e7.15977234@smtp1.ibm.net>
Subject: Merge ActivePerl Stylesheet support etc into Pod::Html.pm
--
use a better prefixify() heuristic than m/perl/ (prefix/lib/perl5
and prefix/lib/perl5/man are ass_u_med only if those directories
actually exist; else prefix/{lib,man} are used)
--
allow C<-foo> under C<use integer> (behavior of C<-$string>
is unchanged still)
--
avoid race condition in the CAPI extension bootstrap handler
--
sanity check to cover the case when perl is installed into the
X:\ (drive root)
--
truncate() has a peculiar exemption from strict barewords, even
though it has a non-filehandle prototype
--
change#3447 didn't do enough to exempt Foo->bar(qw/.../) from
strict 'subs'
--
change#3449 wasn't doing enough
--
make win32_spawnvp() inherit standard handles even when they
may be redirected
--
minor logic tweak for reserved word warning
--
oops, some files missing in change#3449
--
allow '*' prototype to autoquote even barewords that happen to be
function names; parens or ampersand continue to force the other
interpretation; makes C<sub Foo {'bar'} CORE::GLOBAL::require Foo;>
do the right thing, for example
--
redo change#2061 and parts of change#1169 with code in the
parser; PL_last_proto hackery gone, strict 'subs' in now
implemented in the optimizer where specifying the exceptional
cases is much more robust; '*' (bareword) prototype now works
reliably when used in second and subsequent arguments
--
remove redundant part of change#1169 superceded by change#2061;
avoid "future reserved word" warning on prototypical bearwords
--
s/isspace/isSPACE/g and make sure the CRT version is always
passed an unsigned char (fixes random occurrence of spaces in
arguments containing high-bit chars passed to spawned children,
on win32)
--
on win32, look for "site/5.XXX/lib" if "site/5.XXXYY/lib" isn't
found (brings sitelib intuition in line with privlib)
--
mortalize string allocations by win32_get_{priv,site}lib()
(fixes small memory leak in interpreter)
--
opendir(D,"x:") on win32 opens cwd() for drive rather than root;
stat() behaves similarly
--
documentation for Win32 builtins (somewhat modified)
From: jan.dubois@ibm.net (Jan Dubois)
Date: Tue, 30 Mar 1999 08:05:03 +0200
Message-ID: <37006783.1926460@smtp1.ibm.net>
Subject: Re: Issues with build 509
--
provide File::Copy::syscopy() via Win32::CopyFile() on win32
--
more bulletproof workaround for mangled paths;
provide Win32::GetLongPathName()
--
normalize $^X to full pathname on win32
--
work around mangled archname on win32 while finding privlib/sitelib;
normalize lib paths to forward slashes internally
--
avoid negative return value from Win32::GetTickCount()
From: jan.dubois@ibm.net (Jan Dubois)
Date: Sat, 03 Apr 1999 19:04:18 +0200
Message-ID: <37084742.22824479@smtp1.ibm.net>
Subject: Re: Win32::GetTickCount
--
adjust win32_stat() to cope with FindFirstFile() and stat() bugs
(makes opendir(D,"c:") work reliably)
--
fix buggy reference count on refs to SVs with autoviv magic
(resulted in C<my @a; $a[1] = 1; print \$_ for @a> and Data::Dumper
accessing free()d memory)
--
fix bug in change#3123 (off-by-one, caused C<qx[noargs]> to fail
on win32)
--
flip release & version in win32_uname()
--
support POSIX::uname() via win32_uname()
--
implement win32_spawnvp() internally, making it return true PIDs
for asynchronous spawns; fix win32_kill() to always deal with
PIDs
--
use yyerror() instead of croak() so that compile-time failures in
my(LIST) don't confuse globals with lexicals
--
allow custom comparison function in File::Compare::compare_text()
From: jan.dubois@ibm.net (Jan Dubois)
Date: Fri, 26 Feb 1999 21:56:09 +0100
Message-ID: <36db0838.8805651@smtp1.ibm.net>
Subject: Re: PodParser 1.07 (was: RE: C<stuff()> vs stuff())
--
slightly edited version of suggested patch
From: jan.dubois@ibm.net (Jan Dubois)
Date: Mon, 01 Mar 1999 00:32:05 +0100
Message-ID: <36dbcf2c.12325433@smtp1.ibm.net>
Subject: Re: [PATCH 5.005_55] Cleanup of File::Spec module
--
revert parts of change#2990 to preserve predictable usage of
Win32::Foo() as stacked list values
From: jan.dubois@ibm.net (Jan Dubois)
Date: Sat, 27 Feb 1999 18:24:17 +0100
Message-ID: <36e22849.36531259@smtp1.ibm.net>
Subject: Re: resend [PATCH 5.005_55] Various win32/win32.c cleanup
--
add File::Compare::compare_text()
From: jan.dubois@ibm.net (Jan Dubois)
Date: Fri, 26 Feb 1999 00:20:41 +0100
Message-ID: <36dcd8ab.20195659@smtp1.ibm.net>
Subject: Re: PodParser 1.07 (was: RE: C<stuff()> vs stuff())
--
From: jan.dubois@ibm.net (Jan Dubois)
Date: Thu, 18 Feb 1999 19:14:07 +0100
Message-ID: <36d15809.40853323@smtp1.ibm.net>
Subject: resend [PATCH 5.005_55] Various win32/win32.c cleanup
--
support Win32::GetFullPathName() and Win32::SetLastError()
From: jan.dubois@ibm.net (Jan Dubois)
Date: Tue, 09 Feb 1999 22:27:31 +0100
Message-ID: <36c1a2ed.8007554@smtp1.ibm.net>
Subject: [PATCH _54] Win32::GetFullPathName
--
backout change#2811 and add newer version based on File::Spec
From: Barrie Slaymaker <rbs@telerama.com>
Date: Thu, 11 Feb 1999 16:29:24 -0500
Message-ID: <36C34BB4.A62090E0@telerama.com>
Subject: (pod2html) Relative URLs using new File::Spec
--
From: Barrie Slaymaker <rbs@telerama.com>
Date: Thu, 11 Feb 1999 19:39:48 -0500
Message-ID: <36C37854.707D139@telerama.com>
Subject: Merging File::PathConvert in to File::Spec
--
back out change#2751, apply updated version
From: jan.dubois@ibm.net (Jan Dubois)
Date: Sat, 06 Feb 1999 01:06:29 +0100
Message-ID: <36bc844c.18763049@smtp1.ibm.net>
Subject: [PATCH] Cleanup of File::Spec module
--
pod2html misinterprets Foo::Bar as a URL
(fix suggested by Alexander Barilo
<Alexander.Barilo@aexp.com>)
--
devnull() support from Jan Dubois <jan.dubois@ibm.net> and others
p4raw-link: @3449 on //depot/perl: 18228614d1dac2db099ef3d8846e53f44accce72
p4raw-link: @3447 on //depot/perl: 7a52d87a7fbc7848e6b3e9e96db52d4070212cca
p4raw-link: @3123 on //depot/perl: 0aaad0ff610b01c0682abfc20594c83a6d49f148
p4raw-link: @2990 on //depot/perl: bb897dfcf82adc653513b0b92523fb44767a9837
p4raw-link: @2811 on //depot/cfgperl: 5a039dd3f529422cb070070772502cedaf09ae20
p4raw-link: @2751 on //depot/perl: 99804bbbf0b24ddc3b565419ea53f59e7410d1f4
p4raw-link: @2061 on //depot/perl: bf8481137c02593eb36f8d0e234a2ec41a1c92e4
p4raw-link: @1169 on //depot/perl: 2a841d1398ee9bbf30a942905192cc2591b3e92a
p4raw-id: //depot/maint-5.005/perl@4785
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@4784
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@4783
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@3219
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@3218
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@3199
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@3198
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@3197
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@3196
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@3195
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@3193
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@3192
|
|
|
|
|
|
|
|
|
|
|
| |
fix bogus OPf_REF context for the BLOCK in C<grep BLOCK @foo>
(sometimes caused bizarreness in the BLOCK)
p4raw-link: @3180 on //depot/perl: fd3835b3e90ede5576a7ad392beef86bf933ce15
p4raw-id: //depot/maint-5.005/perl@3191
p4raw-branched: from //depot/perl@3190 'branch in' t/op/grep.t
p4raw-integrated: from //depot/perl@3190 'merge in' op.c (@3124..)
MANIFEST (@3149..)
|
|
|
|
|
|
|
| |
warn about newfangled vfork() caveats
p4raw-link: @3147 on //depot/perl: 454f2888fcb3cc0830fef0020890cac2dabe9d66
p4raw-id: //depot/maint-5.005/perl@3190
|
|
|
| |
p4raw-id: //depot/maint-5.005/perl@3189
|
|
|
|
|
| |
p4raw-link: @3098 on //depot/perl: 0244c3a403af2426ac6678d042024bb183ebbfa9
p4raw-id: //depot/maint-5.005/perl@3188
|
|
|
|
|
| |
the Borland compiler)
p4raw-id: //depot/maint-5.005/perl@3187
|
|
|
|
|
| |
-DPERL_CAPI
p4raw-id: //depot/maint-5.005/perl@3186
|
|
|
|
|
|
|
|
|
|
|
| |
a modified version of suggested patch for pack template 'Z'; added docs
Message-ID: <%lOHpzIuGV@snark.ptc.spbu.ru>
Subject: lack of pack/unpack letter with useful symmetry for C null delimited strings
p4raw-link: @2846 on //depot/perl: 5a929a98cca1fca196d5fd6d9350568e529e8825
p4raw-id: //depot/maint-5.005/perl@3185
p4raw-integrated: from //depot/perl@3184 'merge in' t/op/pack.t
(@3023..)
|