This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
perl5.git
20 years agoPERL_IMPLICIT_SYS compiles but does not work.
Nick Ing-Simmons [Sun, 3 Dec 2000 22:57:46 +0000 (22:57 +0000)]
PERL_IMPLICIT_SYS compiles but does not work.

p4raw-id: //depot/perlio@7970

20 years agoIntegrate mainline (STDCHAR)
Nick Ing-Simmons [Sun, 3 Dec 2000 18:41:21 +0000 (18:41 +0000)]
Integrate mainline (STDCHAR)

p4raw-id: //depot/perlio@7966

20 years agoRaw zero bytes in text files confuse at least GNU patch 2.1.
Jarkko Hietaniemi [Sun, 3 Dec 2000 17:58:20 +0000 (17:58 +0000)]
Raw zero bytes in text files confuse at least GNU patch 2.1.

p4raw-id: //depot/perl@7965

20 years agoIntegrate perlio:
Jarkko Hietaniemi [Sun, 3 Dec 2000 17:35:56 +0000 (17:35 +0000)]
Integrate perlio:

[  7961]
Make iperlsys.h vector stdio not PerlIO.

p4raw-link: @7961 on //depot/perlio: eaf8b6989fc5883b4cde6853803087e4b807b2b7

p4raw-id: //depot/perl@7964

20 years agoUntangle the <stdio.h> #include nest for the stdchar test,
Jarkko Hietaniemi [Sun, 3 Dec 2000 17:33:27 +0000 (17:33 +0000)]
Untangle the <stdio.h> #include nest for the stdchar test,
from Andy Dougherty.

p4raw-id: //depot/perl@7962

20 years agoMake iperlsys.h vector stdio not PerlIO.
Nick Ing-Simmons [Sun, 3 Dec 2000 16:53:00 +0000 (16:53 +0000)]
Make iperlsys.h vector stdio not PerlIO.

p4raw-id: //depot/perlio@7961

20 years agoIntegrate mainline
Nick Ing-Simmons [Sun, 3 Dec 2000 14:43:01 +0000 (14:43 +0000)]
Integrate mainline

p4raw-id: //depot/perlio@7960

20 years agoUse DO_UTF8().
Jarkko Hietaniemi [Sun, 3 Dec 2000 08:16:36 +0000 (08:16 +0000)]
Use DO_UTF8().

p4raw-id: //depot/perl@7959

20 years agoRe: [ID 20001130.011] expression parsing bug ?
Mike Guy [Sat, 2 Dec 2000 17:27:13 +0000 (17:27 +0000)]
Re: [ID 20001130.011] expression parsing bug ?
Message-Id: <E142GRN-0003go-00@libra.cus.cam.ac.uk>

An extraneous argument.

p4raw-id: //depot/perl@7958

20 years agoA missing hunk.
Jarkko Hietaniemi [Sat, 2 Dec 2000 18:22:32 +0000 (18:22 +0000)]
A missing hunk.

p4raw-id: //depot/perl@7957

20 years agoRe: [ID 20001202.001] (Net::hostent)
Nicholas Clark [Sat, 2 Dec 2000 14:28:27 +0000 (14:28 +0000)]
Re: [ID 20001202.001] (Net::hostent)
Message-ID: <20001202142827.A12308@plum.flirble.org>

Test robustness.

p4raw-id: //depot/perl@7956

20 years agoAn inconvenient hang would happen if the stdio _ptr wasn't
Jarkko Hietaniemi [Sat, 2 Dec 2000 17:54:32 +0000 (17:54 +0000)]
An inconvenient hang would happen if the stdio _ptr wasn't
found in <stdio.h> and <stdio_impl.h> didn't exist.  (grep
would be grepping stdin for all eternity for the _ptr.)

p4raw-id: //depot/perl@7954

20 years agoUpdate Changes.
Jarkko Hietaniemi [Sat, 2 Dec 2000 00:07:15 +0000 (00:07 +0000)]
Update Changes.

p4raw-id: //depot/perl@7953

20 years agoRe: long shell lines
Andy Dougherty [Thu, 30 Nov 2000 12:45:37 +0000 (07:45 -0500)]
Re: long shell lines
Message-ID: <Pine.SOL.4.10.10011301044590.26388-100000@maxwell.phys.lafayette.edu>

Split overly long shell command lines.

p4raw-id: //depot/perl@7952

20 years agoTry to find stdchar also from <stdio_impl.h>.
Jarkko Hietaniemi [Fri, 1 Dec 2000 23:47:39 +0000 (23:47 +0000)]
Try to find stdchar also from <stdio_impl.h>.

p4raw-id: //depot/perl@7950

20 years agoRe: [ID 20001130.011] expression parsing bug ?
Ronald J. Kimball [Fri, 1 Dec 2000 16:26:08 +0000 (11:26 -0500)]
Re: [ID 20001130.011] expression parsing bug ?
Message-ID: <20001201162608.B1436@linguist.dartmouth.edu>

Reword the diagnostic message.

p4raw-id: //depot/perl@7949

20 years agoIntegrate mainline (for s/y/m as sub names stuff).
Nick Ing-Simmons [Fri, 1 Dec 2000 22:49:43 +0000 (22:49 +0000)]
Integrate mainline (for s/y/m as sub names stuff).

p4raw-id: //depot/perlio@7948

20 years agoIntegrate perlio:
Jarkko Hietaniemi [Fri, 1 Dec 2000 20:55:41 +0000 (20:55 +0000)]
Integrate perlio:

[  7942]
Fix 'mmap' lib/filehand.t (ungetc) test fail.
Make 'unix' pass most tests
- do unread by temporary push of layer ("pending") holding unread chars
- fast_gets is now based on per-handle flag
- relax one of io/pipe.t tests to allow print to fail and close to
succeed so that it passes on unbuffered "unix" layer.
Remaining fail is sporadic and a genuine race condition between
parent/child in fork test.

p4raw-link: @7942 on //depot/perlio: 5e2ab84baab5f2372dc1ffda47d5b89faa2613cd

p4raw-id: //depot/perl@7947

20 years agoUpgrade to CPAN 1.59_51, from Andreas König.
Jarkko Hietaniemi [Fri, 1 Dec 2000 20:36:51 +0000 (20:36 +0000)]
Upgrade to CPAN 1.59_51, from Andreas König.

p4raw-id: //depot/perl@7946

20 years agoIntegrate mainline
Nick Ing-Simmons [Fri, 1 Dec 2000 20:31:32 +0000 (20:31 +0000)]
Integrate mainline

p4raw-id: //depot/perlio@7945

20 years agoSome help for 20001130.011. Now one gets warnings like
Jarkko Hietaniemi [Fri, 1 Dec 2000 20:23:29 +0000 (20:23 +0000)]
Some help for 20001130.011.  Now one gets warnings like
"Ambiguous -f() resolved as a file test ..."

p4raw-id: //depot/perl@7944

20 years agoRetract #7941. Forbidding subs m/s/etc is too cruel because
Jarkko Hietaniemi [Fri, 1 Dec 2000 19:44:44 +0000 (19:44 +0000)]
Retract #7941.  Forbidding subs m/s/etc is too cruel because
that also breaks using them as methods.

p4raw-id: //depot/perl@7943

20 years agoFix 'mmap' lib/filehand.t (ungetc) test fail.
Nick Ing-Simmons [Fri, 1 Dec 2000 17:56:46 +0000 (17:56 +0000)]
Fix 'mmap' lib/filehand.t (ungetc) test fail.
Make 'unix' pass most tests
 - do unread by temporary push of layer ("pending") holding unread chars
 - fast_gets is now based on per-handle flag
 - relax one of io/pipe.t tests to allow print to fail and close to
   succeed so that it passes on unbuffered "unix" layer.
Remaining fail is sporadic and a genuine race condition between
parent/child in fork test.

p4raw-id: //depot/perlio@7942

20 years ago(Retracted by #7943.)
Jarkko Hietaniemi [Fri, 1 Dec 2000 15:28:58 +0000 (15:28 +0000)]
(Retracted by #7943.)

Reserve the short named string operator names.

p4raw-id: //depot/perl@7941

20 years agoGet the three different space character classes right under utf8.
Jarkko Hietaniemi [Fri, 1 Dec 2000 06:45:53 +0000 (06:45 +0000)]
Get the three different space character classes right under utf8.

p4raw-id: //depot/perl@7940

20 years agoTwo more tests that make no sense in UTF-8 since the test
Jarkko Hietaniemi [Fri, 1 Dec 2000 03:26:26 +0000 (03:26 +0000)]
Two more tests that make no sense in UTF-8 since the test
data is not in UTF-8.

p4raw-id: //depot/perl@7939

20 years ago\x{} doesn't any more require 'use utf8' outside regexen so why
Jarkko Hietaniemi [Fri, 1 Dec 2000 03:20:39 +0000 (03:20 +0000)]
\x{} doesn't any more require 'use utf8' outside regexen so why
should it be required inside regexen?

p4raw-id: //depot/perl@7938

20 years agoFix for 20001130.008 and 20001130.010, the PL_regnpar wasn't
Jarkko Hietaniemi [Fri, 1 Dec 2000 02:58:53 +0000 (02:58 +0000)]
Fix for 20001130.008 and 20001130.010, the PL_regnpar wasn't
stored and restored, and thusly was trounced by the utf8 swash
routines.

p4raw-id: //depot/perl@7937

20 years agoFix for 20001128.006, be more careful in Perl_sv_utf8_downgrade().
Jarkko Hietaniemi [Thu, 30 Nov 2000 20:41:39 +0000 (20:41 +0000)]
Fix for 20001128.006, be more careful in Perl_sv_utf8_downgrade().
Why the different platforms behave so differently (core dump vs
no core dump) on this bug is a but of a mystery, but if I had to
guess I would mumble something like 'alignment'.

p4raw-id: //depot/perl@7936

20 years agotoke.c perlio.c -Wformat nits
Robin Barker [Thu, 30 Nov 2000 14:27:12 +0000 (14:27 +0000)]
toke.c perlio.c -Wformat nits
Message-Id: <200011301427.OAA00030@tempest.npl.co.uk>

p4raw-id: //depot/perl@7935

20 years agoIn OS/390 'localhost' can be 'localhost.FOO.BAR'.
Jarkko Hietaniemi [Thu, 30 Nov 2000 16:04:49 +0000 (16:04 +0000)]
In OS/390 'localhost' can be 'localhost.FOO.BAR'.

p4raw-id: //depot/perl@7934

20 years ago[ID 20001129.007] Not OK: perl v5.7.0 +DEVEL7928 on VMS_AXP V7.1 (UNINSTALLED)
Peter Prymmer [Wed, 29 Nov 2000 15:16:45 +0000 (07:16 -0800)]
[ID 20001129.007] Not OK: perl v5.7.0 +DEVEL7928 on VMS_AXP V7.1 (UNINSTALLED)
Message-Id: <Pine.OSF.4.10.10011291515570.328738-100000@aspara.forte.com>

In VMS the localhost may be LOCALHOST.

p4raw-id: //depot/perl@7933

20 years agoRe: question about retlen in utf8.c:Perl_utf8_to_uv()
Peter Prymmer [Wed, 29 Nov 2000 12:36:40 +0000 (04:36 -0800)]
Re: question about retlen in utf8.c:Perl_utf8_to_uv()
Message-ID: <Pine.OSF.4.10.10011291233120.328738-100000@aspara.forte.com>

plus regen perlapi.pod.

p4raw-id: //depot/perl@7932

20 years agoget t/lib/filter-util.t to work on VMS
Charles Lane [Wed, 29 Nov 2000 17:15:11 +0000 (12:15 -0500)]
get t/lib/filter-util.t to work on VMS
Message-Id: <001129171439.6099b@DUPHY4.Physics.Drexel.Edu>

p4raw-id: //depot/perl@7931

20 years agoUpdate Changes.
Jarkko Hietaniemi [Wed, 29 Nov 2000 18:51:55 +0000 (18:51 +0000)]
Update Changes.

p4raw-id: //depot/perl@7930

20 years ago[ID 20001128.003] Not OK: perl v5.7.0 +DEVEL7891 on ppc-linux 2.2.17
Dominic Dunlop [Tue, 28 Nov 2000 16:41:57 +0000 (17:41 +0100)]
[ID 20001128.003] Not OK: perl v5.7.0 +DEVEL7891 on ppc-linux 2.2.17
Message-Id: <p04320400b649771b9797@[192.168.1.4]>

Patch to catch a core dump in the Configure va_copy test.

p4raw-id: //depot/perl@7928

20 years agoRetract #7921, the patch shouldn't be needed.
Jarkko Hietaniemi [Wed, 29 Nov 2000 17:31:59 +0000 (17:31 +0000)]
Retract #7921, the patch shouldn't be needed.

p4raw-id: //depot/perl@7927

20 years agoMake the Class::Struct import() wiser.
Jarkko Hietaniemi [Wed, 29 Nov 2000 17:01:41 +0000 (17:01 +0000)]
Make the Class::Struct import() wiser.

p4raw-id: //depot/perl@7926

20 years agoAll the core library users of Class::Struct seem to be
Jarkko Hietaniemi [Wed, 29 Nov 2000 16:36:33 +0000 (16:36 +0000)]
All the core library users of Class::Struct seem to be
using "use Class::Struct 'struct';" instead of the bare
"use Class::Struct;", which isn't documented in Class::Struct.
This can't be right.

p4raw-id: //depot/perl@7925

20 years agoA test works better if it has the right 1..$n output.
Jarkko Hietaniemi [Wed, 29 Nov 2000 16:32:15 +0000 (16:32 +0000)]
A test works better if it has the right 1..$n output.

p4raw-id: //depot/perl@7924

20 years agoForgot to MANIFEST the new Net::hostent test.
Jarkko Hietaniemi [Wed, 29 Nov 2000 16:29:11 +0000 (16:29 +0000)]
Forgot to MANIFEST the new Net::hostent test.

p4raw-id: //depot/perl@7923

20 years agoIntegrate perlio:
Jarkko Hietaniemi [Wed, 29 Nov 2000 16:27:49 +0000 (16:27 +0000)]
Integrate perlio:

[  7906]
Initial tidy of setmode() muddle (UNIX does not have, nor need it).

p4raw-link: @7906 on //depot/perlio: a4d3c1d3a59a079ee84191d2df8b5e232a8bee44

p4raw-id: //depot/perl@7922

20 years ago(Retracted by #7927.)
Simon Cozens [Wed, 29 Nov 2000 14:48:20 +0000 (14:48 +0000)]
(Retracted by #7927.)

Subject: [PATCH] Is infinity a number?
Message-ID: <20001129144820.A31339@pembro33.pmb.ox.ac.uk>

p4raw-id: //depot/perl@7921

20 years agonet_hostent.t (was Re: [ID 20001128.002] what's the point of example code if it is...
Robert Spier [Wed, 29 Nov 2000 02:17:49 +0000 (21:17 -0500)]
net_hostent.t (was Re: [ID 20001128.002] what's the point of example code if it is buggy?)
Message-ID: <14884.44445.760322.278647@localhost.localdomain>

Add a test for Net::hostent.

p4raw-id: //depot/perl@7920

20 years agoMake "use Class::Struct 'struct';" work again (broken by #7617);
Jarkko Hietaniemi [Wed, 29 Nov 2000 16:21:31 +0000 (16:21 +0000)]
Make "use Class::Struct 'struct';" work again (broken by #7617);
add a test for Class::Struct.

p4raw-id: //depot/perl@7919

20 years agoRe: Minor suggestion for Sys::Syslog [PATCH]
Robert Spier [Wed, 29 Nov 2000 02:25:39 +0000 (21:25 -0500)]
Re: Minor suggestion for Sys::Syslog  [PATCH]
Message-ID: <14884.44915.86028.422238@localhost.localdomain>

More checking in case someone has broken their services or
protocol databases.

p4raw-id: //depot/perl@7918

20 years ago$^O win32 -> MSWin32
Robert Spier [Wed, 29 Nov 2000 02:17:38 +0000 (21:17 -0500)]
$^O  win32 -> MSWin32
Message-ID: <14884.44434.340627.126145@localhost.localdomain>

plus similar nits for vms, err, VMS, and UNICOS.

p4raw-id: //depot/perl@7917

20 years agoTokeniser debugging
Simon Cozens [Wed, 29 Nov 2000 14:15:45 +0000 (14:15 +0000)]
Tokeniser debugging
Message-ID: <20001129141545.A30864@pembro33.pmb.ox.ac.uk>

p4raw-id: //depot/perl@7916

20 years agoRe: [PATCH 5.7.0] OUT keyword for xsubpp
Ilya Zakharevich [Wed, 29 Nov 2000 02:13:14 +0000 (21:13 -0500)]
Re: [PATCH 5.7.0] OUT keyword for xsubpp
Date: Wed, 29 Nov 2000 02:13:14 -0500
Message-ID: <20001129021314.A2532@monk.mps.ohio-state.edu>

OUT keyword nits.

Subject: Re: [PATCH 5.7.0] OUT keyword for xsubpp
From: Ilya Zakharevich <ilya@math.ohio-state.edu>
Date: Wed, 29 Nov 2000 03:09:04 -0500
Message-ID: <20001129030904.A2754@monk.mps.ohio-state.edu>

OUT and IN_OUT documentation.

p4raw-id: //depot/perl@7915

20 years agoRe: [ID 20001127.002] const subs hurt under debugger
John Tobey [Tue, 28 Nov 2000 01:13:41 +0000 (20:13 -0500)]
Re: [ID 20001127.002] const subs hurt under debugger
Message-Id: <m140e1N-000FOoC@feynman.localnet>

plus a performance speedup noticed by Sarathy.

p4raw-id: //depot/perl@7914

20 years agoUse "1 while unlink" so that VMS gets clean, too.
Jarkko Hietaniemi [Wed, 29 Nov 2000 03:16:59 +0000 (03:16 +0000)]
Use "1 while unlink" so that VMS gets clean, too.

p4raw-id: //depot/perl@7913

20 years agoOUT keyword for xsubpp
Ilya Zakharevich [Tue, 28 Nov 2000 03:27:09 +0000 (22:27 -0500)]
OUT keyword for xsubpp
Message-ID: <20001128032709.A23401@monk.mps.ohio-state.edu>

p4raw-id: //depot/perl@7912

20 years agoThis should have been part of #7872: no need to scan UTF-8
Jarkko Hietaniemi [Wed, 29 Nov 2000 03:06:30 +0000 (03:06 +0000)]
This should have been part of #7872: no need to scan UTF-8
until eternity.

p4raw-id: //depot/perl@7911

20 years agoOne more perltie.pod nit from Casey R. Tweten.
Jarkko Hietaniemi [Wed, 29 Nov 2000 02:50:04 +0000 (02:50 +0000)]
One more perltie.pod nit from Casey R. Tweten.

p4raw-id: //depot/perl@7910

20 years agoDocument the known sprintf test failures, exact standard
Jarkko Hietaniemi [Wed, 29 Nov 2000 02:44:23 +0000 (02:44 +0000)]
Document the known sprintf test failures, exact standard
reference from Dominic Dunlop, NonStop-UX testing from Tom Bates.

p4raw-id: //depot/perl@7909

20 years agoAdd the Encoding table format documentation.
Jarkko Hietaniemi [Wed, 29 Nov 2000 02:36:23 +0000 (02:36 +0000)]
Add the Encoding table format documentation.

p4raw-id: //depot/perl@7908

20 years agoRe: [PATCH] Updating perltie.pod for arrays
Casey R. Tweten [Tue, 28 Nov 2000 12:11:41 +0000 (07:11 -0500)]
Re: [PATCH] Updating perltie.pod for arrays
Date: Tue, 28 Nov 2000 12:11:41 -0500 (EST)
Message-ID: <Pine.OSF.4.21.0011281209050.25178-100000@home.kiski.net>

Subject: Re: [PATCH] Updating perltie.pod for arrays
From: "Casey R. Tweten" <crt@kiski.net>
Date: Tue, 28 Nov 2000 15:03:50 -0500 (EST)
Message-ID: <Pine.OSF.4.21.0011281458500.10331-100000@home.kiski.net>

p4raw-id: //depot/perl@7907

20 years agoInitial tidy of setmode() muddle (UNIX does not have, nor need it).
Nick Ing-Simmons [Tue, 28 Nov 2000 22:40:59 +0000 (22:40 +0000)]
Initial tidy of setmode() muddle (UNIX does not have, nor need it).

p4raw-id: //depot/perlio@7906

20 years agoIntegrate mainline
Nick Ing-Simmons [Tue, 28 Nov 2000 22:04:34 +0000 (22:04 +0000)]
Integrate mainline

p4raw-id: //depot/perlio@7905

20 years ago[ID 20001128.002] what's the point of example code if it is buggy?
Nicholas Clark [Tue, 28 Nov 2000 11:16:57 +0000 (11:16 +0000)]
[ID 20001128.002] what's the point of example code if it is buggy?
Message-Id: <E140ikr-00076N-00@fruitbat.babyhippo.co.uk>

p4raw-id: //depot/perl@7904

20 years agoUpdating perltie.pod for arrays
Casey R. Tweten [Mon, 27 Nov 2000 17:47:20 +0000 (12:47 -0500)]
Updating perltie.pod for arrays
Message-ID: <Pine.OSF.4.21.0011271746360.11659-100000@home.kiski.net>

p4raw-id: //depot/perl@7903

20 years agoRe: 5.6 bug: split /^/ implies /m modifier (from CLPM)
Robin Barker [Mon, 27 Nov 2000 17:56:44 +0000 (17:56 +0000)]
Re: 5.6 bug: split /^/ implies /m modifier (from CLPM)
Message-Id: <200011271756.RAA22706@tempest.npl.co.uk>

p4raw-id: //depot/perl@7902

20 years agosetmode() is a DOSish-only thing.
Jarkko Hietaniemi [Tue, 28 Nov 2000 16:28:04 +0000 (16:28 +0000)]
setmode() is a DOSish-only thing.

p4raw-id: //depot/perl@7901

20 years agoperlcc.PL cleanups
Simon Cozens [Tue, 28 Nov 2000 12:44:35 +0000 (12:44 +0000)]
perlcc.PL cleanups
Message-ID: <20001128124435.A5500@pembro33.pmb.ox.ac.uk>

p4raw-id: //depot/perl@7900

20 years agoUpdate Changes.
Jarkko Hietaniemi [Mon, 27 Nov 2000 15:02:32 +0000 (15:02 +0000)]
Update Changes.

p4raw-id: //depot/perl@7892

20 years agoRe: perlfaq style changes
Yitzchak Scott-Thoennes [Sat, 25 Nov 2000 23:32:26 +0000 (15:32 -0800)]
Re: perlfaq style changes
Message-ID: <KyLI6gzkgiRK092yn@efn.org>

plus a note from Ronald Kimball.

p4raw-id: //depot/perl@7891

20 years agoRe: [ID 20001122.006] weird behaviour of $|
Benjamin Holzman [Sun, 26 Nov 2000 16:27:33 +0000 (11:27 -0500)]
Re: [ID 20001122.006] weird behaviour of $|
Message-ID: <20001126162733.J25040@ecnvantage.com>

p4raw-id: //depot/perl@7890

20 years ago[ID 20001127.004] White space problem in perlamiga.pod
Larry W. Virden [Mon, 27 Nov 2000 09:32:06 +0000 (04:32 -0500)]
[ID 20001127.004] White space problem in perlamiga.pod
Message-Id: <200011271432.JAA09550@lwv26awu.cas.org>

p4raw-id: //depot/perl@7889

20 years agoRestore also the locale test to no-sprintf-taint state.
Jarkko Hietaniemi [Mon, 27 Nov 2000 14:32:34 +0000 (14:32 +0000)]
Restore also the locale test to no-sprintf-taint state.
Fixes the bug 20001127.003.

p4raw-id: //depot/perl@7888

20 years ago__FUNCTION__ isn't portable and trying to emulate it leads
Jarkko Hietaniemi [Mon, 27 Nov 2000 13:54:18 +0000 (13:54 +0000)]
__FUNCTION__ isn't portable and trying to emulate it leads
into practically spelling out the name of the function.
Takes care of the bug 20001127.001.

p4raw-id: //depot/perl@7886

20 years agoVMS test cleanup
Charles Lane [Mon, 27 Nov 2000 08:23:54 +0000 (03:23 -0500)]
VMS test cleanup
Message-Id: <001127082318.6381c@DUPHY4.Physics.Drexel.Edu>

p4raw-id: //depot/perl@7884

20 years agoIt seems that *both* the unused submatch loop cleanup
Jarkko Hietaniemi [Mon, 27 Nov 2000 05:48:41 +0000 (05:48 +0000)]
It seems that *both* the unused submatch loop cleanup
codes are needed.

p4raw-id: //depot/perl@7881

20 years agoClean .exists deeper.
Jarkko Hietaniemi [Mon, 27 Nov 2000 04:50:07 +0000 (04:50 +0000)]
Clean .exists deeper.

p4raw-id: //depot/perl@7880

20 years agoThe code in regcppop() (see #7878) contains the correct lower
Jarkko Hietaniemi [Mon, 27 Nov 2000 02:31:35 +0000 (02:31 +0000)]
The code in regcppop() (see #7878) contains the correct lower
limit for the unused submatch 'cleanup' loop so that under
"use utf8" the following code wouldn't dump core:
"," =~ /([^,]*,)*/  With the the wrong lower limit (>=1)
the cleanup loop in regtry() stomped beyond allocated area
in the startp[] array.  Therefore, copied the correct lower
loop limit (*PL_reglastparen) to regtry().  Note: something
may still not be quite right: why was the _higher_ loop limit
(prog->nparens) different in the utf8 case?

After this patch "./perl -Ilib -Mutf8 t/op/regexp.t" works
without core dumps, there were about 17 of them before
the patch (with us since Perl 5.7.0).  Two failures, still:
496 and 505 (though these may not be severe).

Patch #7881 is also needed since both the cleanup loops
seem to be needed.

Also, the t/op/pat#44 seems to core dump under utf8.
Plus a couple of failures.  UGH-8.

p4raw-id: //depot/perl@7879

20 years agoThe unused submatch cleanup code in regtry() seems to be more crucial,
Jarkko Hietaniemi [Mon, 27 Nov 2000 02:21:17 +0000 (02:21 +0000)]
The unused submatch cleanup code in regtry() seems to be more crucial,
the code in regcppop() seems to be redundant for the test suite --
but it contains a germ of truth, and it needed for the build
process itself: see #7879 and #7881.

p4raw-id: //depot/perl@7878

20 years agoComment on comment.
Jarkko Hietaniemi [Mon, 27 Nov 2000 01:53:09 +0000 (01:53 +0000)]
Comment on comment.

p4raw-id: //depot/perl@7877

20 years agoAdjust the docs to agree with #7875.
Jarkko Hietaniemi [Mon, 27 Nov 2000 00:49:59 +0000 (00:49 +0000)]
Adjust the docs to agree with #7875.

p4raw-id: //depot/perl@7876

20 years agoRetract #7863. It makes more sense not to taint formatted output,
Jarkko Hietaniemi [Mon, 27 Nov 2000 00:49:27 +0000 (00:49 +0000)]
Retract #7863.  It makes more sense not to taint formatted output,
not even by sprintf().

p4raw-id: //depot/perl@7875

20 years agoDebug dump of ANYOFUTF8 was garbage (data from ANYOF).
Jarkko Hietaniemi [Sun, 26 Nov 2000 23:36:02 +0000 (23:36 +0000)]
Debug dump of ANYOFUTF8 was garbage (data from ANYOF).
Not really fixed (should really dump the UTF-8 charclass),
but stopped displaying the garbage.

Also add a note on the (missing) Unicode PSXSPC and BLANK.

p4raw-id: //depot/perl@7874

20 years agoBOUND regex opcodes (\b, \B) could try to scan zero length UTF-8.
Jarkko Hietaniemi [Sun, 26 Nov 2000 21:31:13 +0000 (21:31 +0000)]
BOUND regex opcodes (\b, \B) could try to scan zero length UTF-8.

p4raw-id: //depot/perl@7873

20 years agoNo need to scan till infinity, 13 is enough.
Jarkko Hietaniemi [Sun, 26 Nov 2000 21:06:04 +0000 (21:06 +0000)]
No need to scan till infinity, 13 is enough.

p4raw-id: //depot/perl@7872

20 years agoTest line numbers are different with utf8.
Jarkko Hietaniemi [Sun, 26 Nov 2000 20:20:32 +0000 (20:20 +0000)]
Test line numbers are different with utf8.

p4raw-id: //depot/perl@7871

20 years agoMessage nit.
Jarkko Hietaniemi [Sun, 26 Nov 2000 20:10:12 +0000 (20:10 +0000)]
Message nit.

p4raw-id: //depot/perl@7870

20 years agoMake utf8_length() and utf8_distance() (the latter of which
Jarkko Hietaniemi [Sun, 26 Nov 2000 19:01:05 +0000 (19:01 +0000)]
Make utf8_length() and utf8_distance() (the latter of which
is unused at the moment) to be less forgiving about bad UTF-8.

p4raw-id: //depot/perl@7869

20 years agoRe: [ID 20001125.004] OK: perl v5.7.0 +DEVEL7825 on i686-linux 2.2.17 (UNINSTALLED)
Benjamin Holzman [Sat, 25 Nov 2000 20:40:20 +0000 (15:40 -0500)]
Re: [ID 20001125.004] OK: perl v5.7.0 +DEVEL7825 on i686-linux 2.2.17 (UNINSTALLED)
Message-ID: <20001125204020.A25040@ecnvantage.com>

Debian allows installing /usr/bin/locale without installing
any locales, an error message will ensue.

p4raw-id: //depot/perl@7868

20 years agoRe: [PATCH: perl@7825] SvTEMP-ness on rhs of aassign can wreak havoc
Benjamin Holzman [Sun, 26 Nov 2000 02:42:34 +0000 (21:42 -0500)]
Re: [PATCH: perl@7825] SvTEMP-ness on rhs of aassign can wreak havoc
Message-ID: <20001126024234.G25040@ecnvantage.com>

Patch for the bug 20000212.002.

p4raw-id: //depot/perl@7867

20 years agoPATCH: File::Temp fix on WindowsNT/VMS
Tim Jenness [Fri, 24 Nov 2000 18:38:34 +0000 (08:38 -1000)]
PATCH: File::Temp fix on WindowsNT/VMS
Message-ID: <Pine.LNX.4.21.0011241833230.18423-100000@lapaki.jach.hawaii.edu>

File::Temp 0.11.

p4raw-id: //depot/perl@7866

20 years agoMake the va_copy() test not to be so talkative.
Jarkko Hietaniemi [Sun, 26 Nov 2000 18:23:41 +0000 (18:23 +0000)]
Make the va_copy() test not to be so talkative.

p4raw-id: //depot/perl@7864

20 years agoFix locale inconsistencies unearthed by Hugo's work.
Jarkko Hietaniemi [Sun, 26 Nov 2000 18:11:02 +0000 (18:11 +0000)]
Fix locale inconsistencies unearthed by Hugo's work.
Now the floating point sprintf really does taint the result
string as perllocale promises (has promised for a long time)
if "use locale" is in the lexical scope.

p4raw-id: //depot/perl@7863

20 years agoRe: pp_add -> pp_i_add efficiency hack?
Nicholas Clark [Sun, 26 Nov 2000 00:07:50 +0000 (00:07 +0000)]
Re: pp_add -> pp_i_add efficiency hack?
Message-ID: <20001126000750.A22446@plum.flirble.org>

Infinit.

p4raw-id: //depot/perl@7862

20 years agoIntegrate perlio:
Jarkko Hietaniemi [Sun, 26 Nov 2000 00:01:26 +0000 (00:01 +0000)]
Integrate perlio:

[  7860]
Correct relative path from new ext\Filter\Util\Call location
back to miniperl.

[  7857]
All tests pass on Win32/gcc/USE_PERLIO.
- when crlf layer is pushed make unix-level binary.
- remove :crlf injection in pp_backtick in pp_sys.c
- (Change dependacy of extension .dll's to perldll.def rather than perl.exe)

[  7853]
Re-arrange crlf vs binary for platforms that care.
crlf layer is now "the" buffer layer and can turn its behaviour on/off.
Lip-service to making stdio layer work on such platforms (untested).
Now fails 3 tests rather than one, checkin to see if I can debug
the issue under linux. (Do not merge.)

p4raw-link: @7860 on //depot/perlio: 0727cc217546946a8c8db12875c3cacb5833e494
p4raw-link: @7857 on //depot/perlio: a77df51f3c9319336e0ec43ad003b02a0111d2af
p4raw-link: @7853 on //depot/perlio: f5b9d040829bb5deb93173e35ee6c82ce0a01c37

p4raw-id: //depot/perl@7861

20 years agoCorrect relative path from new ext\Filter\Util\Call location
Nick Ing-Simmons [Sat, 25 Nov 2000 22:04:17 +0000 (22:04 +0000)]
Correct relative path from new ext\Filter\Util\Call location
back to miniperl.

p4raw-id: //depot/perlio@7860

20 years agoIntegrate mainline.
Nick Ing-Simmons [Sat, 25 Nov 2000 21:26:54 +0000 (21:26 +0000)]
Integrate mainline.

p4raw-id: //depot/perlio@7859

20 years agoC<foreach my $x ...> in pseudo-fork()ed process may diddle
Gurusamy Sarathy [Sat, 25 Nov 2000 20:52:17 +0000 (20:52 +0000)]
C<foreach my $x ...> in pseudo-fork()ed process may diddle
parent's memory; fix it by keeping track of the actual pad
offset rather than a raw pointer (this change is probably also
relevant to non-ithreads case to avoid fallout from reallocs of
the pad array, but is currently only enabled for the ithreads
case in the interests of minimal disruption to existing "well
tested" code)

p4raw-id: //depot/perl@7858

20 years agoAll tests pass on Win32/gcc/USE_PERLIO.
Nick Ing-Simmons [Sat, 25 Nov 2000 20:01:28 +0000 (20:01 +0000)]
All tests pass on Win32/gcc/USE_PERLIO.
 - when crlf layer is pushed make unix-level binary.
 - remove :crlf injection in pp_backtick in pp_sys.c
        - (Change dependacy of extension .dll's to perldll.def rather than perl.exe)

p4raw-id: //depot/perlio@7857

20 years agoUndo the SOCKS workarounds, instead start using PerlIO
Jens Hamisch [Fri, 24 Nov 2000 18:31:30 +0000 (19:31 +0100)]
Undo the SOCKS workarounds, instead start using PerlIO
if SOCKS is selected.

Subject: perl@7847, [ID 20001030.005], close-patch, perlio - The big cleanup
Date: Fri, 24 Nov 2000 18:31:30 +0100
Message-ID: <20001124183130.E28337@Strawberry.COM>

Subject: Re: perl@7847, [ID 20001030.005], close-patch, perlio - Patch the patch ...
From: Jens Hamisch <jens@Strawberry.COM>
Date: Fri, 24 Nov 2000 19:11:51 +0100
Message-ID: <20001124191151.A28753@Strawberry.COM>

p4raw-id: //depot/perl@7855

20 years agoUndo #7848. Some of the code seems to use walkoptree(),
Jarkko Hietaniemi [Sat, 25 Nov 2000 17:39:08 +0000 (17:39 +0000)]
Undo #7848.  Some of the code seems to use walkoptree(),
some walkoptree_slow().  An unfinished renaming?
Now the sub is walkoptree() (which is @EXPORT_OK),
the walkoptree_slow() is a typeglob alias to walkoptree.
This makes the tests to pass, at least.

p4raw-id: //depot/perl@7854

20 years agoRe-arrange crlf vs binary for platforms that care.
Nick Ing-Simmons [Fri, 24 Nov 2000 22:24:33 +0000 (22:24 +0000)]
Re-arrange crlf vs binary for platforms that care.
crlf layer is now "the" buffer layer and can turn its behaviour on/off.
Lip-service to making stdio layer work on such platforms (untested).
Now fails 3 tests rather than one, checkin to see if I can debug
the issue under linux. (Do not merge.)

p4raw-id: //depot/perlio@7853

20 years agoAUTHORS edits.
Jarkko Hietaniemi [Fri, 24 Nov 2000 18:57:39 +0000 (18:57 +0000)]
AUTHORS edits.

p4raw-id: //depot/perl@7852

20 years agoAdd Jan-Erik Karlsson.
Jarkko Hietaniemi [Fri, 24 Nov 2000 17:34:56 +0000 (17:34 +0000)]
Add Jan-Erik Karlsson.

p4raw-id: //depot/perl@7851