Jan Dubois [Thu, 17 Dec 2009 19:15:38 +0000]
Implement win32_isatty()
Commit
827da6a38 added a custom isatty() implementation
in win32/perlhost.h, but that code will only be used when
perl is compiled with -DPERL_IMPLICIT_SYS. This change
makes sure that the custom implementation will be used
on Windows for all choices of build options.
Jan Dubois [Wed, 16 Dec 2009 23:42:19 +0000]
-t should only return TRUE for file handles connected to a TTY
The Microsoft C version of isatty() returns TRUE for all
character mode devices, including the /dev/null style "nul"
device and printers like "lpt1".
The included test has only been tested on Windows and Linux;
the device names for OS/2 and VMS are just best guesses...
Abigail [Wed, 16 Dec 2009 13:01:32 +0000]
Document issues when using named captures in combination with a branch reset pattern (see also #71136)
Rafael Garcia-Suarez [Wed, 16 Dec 2009 09:20:27 +0000]
Fix casting warnings
Gerard Goossen [Wed, 9 Dec 2009 16:23:00 +0000]
Add a test that "eval" does not create additional reference to ouside variables.
Gerard Goossen [Tue, 8 Dec 2009 19:41:28 +0000]
Store the PL_compcv instead of the the PL_comppad in parser stack, and make it reference counted. Properly solves [perl #66094]
Alex Davies [Wed, 16 Dec 2009 08:42:12 +0000]
tweak to pp_ctl.c gives smaller object code
Jesse Vincent [Wed, 16 Dec 2009 02:08:40 +0000]
Update File::Copy tests to skip on OpenBSD, as it mounts too many filesystems nosuid.
[perl #71334] is the TODO for a better fix for this
Tim Bunce [Tue, 15 Dec 2009 23:04:11 +0000]
Allow override of PERL5OPT etc in t/TEST
Rafael Garcia-Suarez [Tue, 15 Dec 2009 23:04:34 +0000]
Don't make C<use legacy> die on unknown legacy names
So we can use C<use legacy "qubits"> to avoid the new "qubit" behaviour,
without worrying about perls that didn't have qubit support at all. :)
Zefram [Tue, 15 Dec 2009 10:48:31 +0000]
[perl #22977] Bug in format/write
Zefram [Tue, 15 Dec 2009 10:17:35 +0000]
[perl #68640] Wrong error for undef constant name
Karl Williamson [Mon, 14 Dec 2009 16:18:29 +0000]
more regex folding tests
Father Chrysostomos [Mon, 14 Dec 2009 15:15:51 +0000]
[perl #70171] 5.10.0 -> 5.10.1 Regression in
fafafbaf70 (Big slowdown in 5.10 @_ parameter passing)
In this case my %x = %$x assigns a hash to itself. This causes the
hv_clear in pp_aassign to wipe away the hash before it can be copied.
The ‘panic: attempt to copy freed scalar’ error is triggered by this
line, which copies the value:
sv_setsv(tmpstr,*relem); /* value */
The solution is to make sure the OPpASSIGN_COMMON flag is on in such
cases, so that pp_aassign copies everything before doing the assignment.
Michael Breen [Fri, 11 Dec 2009 17:48:51 +0000]
fix bug 67156: overload: nomethod(..., '!') return value inverted
Gene Sullivan [Mon, 14 Dec 2009 13:07:27 +0000]
[perl #71204] diagnostics.pm suppresses 'Use of uninitialized value in range (or flip)' warning
(and bump version of diagnostics.pm)
Reini Urban [Mon, 14 Dec 2009 12:04:37 +0000]
cygwin IPv6 since 1.7
Use g++ as default linker
IPv6 support for 1.7
Father Chrysostomos [Mon, 14 Dec 2009 11:19:35 +0000]
[perl #70764] $' fails to initialized for pre-compiled regular expression matches
The match vars are associated with the regexp that last matched
successfully. In the case of $str =~ $qr or /$qr/, since the $qr could
be used in multiple scopes that need their own sets of match vars, the
$qr is cloned by Perl_reg_temp_copy as of change 30677/28d8d7f. This
happens in pp_regcomp before pp_match has stringified the LHS, hence the
bug. In short, /$gror/ is not equivalent to
($which = !$which) ? /$gror/ : /$gror/, which is weird.
Attached is a patch, which admittedly is a hack, but fixes this
particular side effect of what is probably a bad design, by stringifying
the LHS in pp_regcomp, and having pp_match skip get-magic in such cases.
A real fix far exceeds my capabalities, and would also be very intrusive
according to
<http://www.nntp.perl.org/group/perl.perl5.porters/2007/03/msg122415.html>.
Craig A. Berry [Mon, 14 Dec 2009 04:17:13 +0000]
Fix compile failure introduced in
37e2e78edfe0a224b8a615820f46db879584f523.
Solaris, VMS, and Win32 all failed to build after this change. In C99's
description of:
do statement while ( expression ) ;
the trailing semicolon does not appear to be optional. And at least
three compilers from three vendors agree.
Reini Urban [Sun, 13 Dec 2009 07:06:43 +0000]
Eliminate OP_SETSTATE from cop.h header
It had been added with change 3728 to track linenumbers in
optimized else, disabled by change 4309,
and removed with change 33072.
Bump copyright, latest change was "Fix MULTICALL in List-Util" 2009-03-07
with commit
1bbbfc50
Rafael Garcia-Suarez [Sun, 13 Dec 2009 14:47:45 +0000]
Make new is_utf8_X_* functions public
Rafael Garcia-Suarez [Sat, 12 Dec 2009 14:57:28 +0000]
Doc nits
Rafael Garcia-Suarez [Sat, 12 Dec 2009 08:33:51 +0000]
Merge commit 'khwilliamson/x' into blead
Jesse Vincent [Thu, 10 Dec 2009 18:20:27 +0000]
Notes on why PathTools is in Cwd/
Jesse Vincent [Thu, 10 Dec 2009 18:14:24 +0000]
Update Cwd / PathTools to 3.31 to get us a non-devel version number based on a chat with Steffen. No code changes.
Jesse Vincent [Thu, 10 Dec 2009 16:44:26 +0000]
Updated to Pod::Simple 3.11 from CPAN [perl #71004]
Chris Williams [Thu, 10 Dec 2009 12:24:16 +0000]
Update CPANPLUS to cpan version 0.89_12
Changes for 0.89_12 Mon Dec 7 13:33:16 2009
================================================
* Resolve RT #52348 Duplicate test output, reported by Apocalypse
* Fixed typo in Shell::Default, RT #52376, reported by Apocalypse
Changes for 0.89_11 Tue Dec 1 13:14:24 2009
================================================
* Fixed RT #52287 reported by Apocalypse regarding
Test::Reporter barfing on send()
* Change SQLite to AutoCommit, resolves RT #52308,
reported by Apocalypse
Changes for 0.89_10 Sat Nov 28 23:20:09 2009
================================================
* Resolve RT #51516 setting conf options which
include spaces.
* Explicitly use Cwd's chdir in _chdir()
Update to allow various perl smokers test before update to 0.90
Chris Williams [Thu, 10 Dec 2009 12:16:57 +0000]
Update Archive::Extract to cpan version 0.36
Changes for 0.36 Tue Nov 24 10:27:16 2009
============================================
* Explicitly use Cwd's chdir
Required for CPANPLUS update
Jan Dubois [Wed, 9 Dec 2009 23:55:09 +0000]
OS/2 doesn't support inplace editing without backups
Reini Urban [Wed, 9 Dec 2009 22:45:44 +0000]
Updated docpointer for illguts
Zefram [Wed, 9 Dec 2009 22:40:05 +0000]
Fix for [perl #70910] wrong line number in syntax error message
Andy Dougherty [Tue, 8 Dec 2009 16:53:14 +0000]
Merge branch 'blead' of ssh://perl5.git.perl.org/gitroot/perl into blead
Andy Dougherty [Tue, 8 Dec 2009 16:36:54 +0000]
Document config_args limitations reported in [perl #70912]
Gerard Goossen [Tue, 8 Dec 2009 11:42:58 +0000]
proper error on "grep $x (1,2,3)". Solves [perl #37314]
Tony Cook [Tue, 8 Dec 2009 11:48:59 +0000]
[perl #70802] -i'*' refuses to work
Add regression tests
(the bug was fixed by commit
c9930541bfa04399c3b648e83c9b750cee1154fb)
Bram [Sat, 1 Aug 2009 20:11:22 +0000]
undef $! before running the errno tests
Ricardo Signes [Mon, 7 Dec 2009 23:24:59 +0000]
legalize =begin foo bar
Ricardo Signes [Mon, 7 Dec 2009 23:21:44 +0000]
more aggressively deprecate L<section> and L<"section">
Ricardo Signes [Mon, 7 Dec 2009 23:19:28 +0000]
remove prohibition against L<text|href>
Rafael Garcia-Suarez [Tue, 8 Dec 2009 08:29:11 +0000]
Add latest test file to MANIFEST
Chip Salzenberg [Tue, 8 Dec 2009 08:17:05 +0000]
utf8cache test: Hugo asks, and I provide. Merry Christmas
Jan Dubois [Tue, 8 Dec 2009 01:19:18 +0000]
Throw away uncleanable scopes when exiting a pseudo-forked process.
Commit
adab9969 tried to clean up those additional scopes, but failed
because some of the memory was allocated from a different pool. To
avoid triggering the assert() in perl_destruct() this change instead
moves the one remaining scope back to the root of the stack, effectively
discarding the additional frames without any further processing.
Rafael Garcia-Suarez [Mon, 7 Dec 2009 16:51:16 +0000]
The deep arrays for testing ~~ were not deep enough
Without commit
6bc991bfb3c34a5d286a1202fcc0d740d72dcee7, this test
would now segfault.
Rafael Garcia-Suarez [Mon, 7 Dec 2009 14:03:48 +0000]
Fix a test that wasn’t testing was it purported to be testing
(noticed by Father Chrysostomos, further modified to avoid using
too much overloaded names like foo and Foo.)
Father Chrysostomos [Mon, 7 Dec 2009 13:58:40 +0000]
[perl #71076] sort with active sub (5.10 regression)
One of the tests in sort.t causes a bus error (or sometimes ‘Undefined
subroutine called’) if run multiple times. This is because sort
decreases the refcount of an active sub used as a comparison routine.
Ironically enough, this test was added by the very change that broke it
(25953/9850bf2).
Rafael Garcia-Suarez [Mon, 7 Dec 2009 12:55:49 +0000]
Allocate the right HV
This will solve an obscure bug in smart-match involving nested data
structures containing the same elements.
Rafael Garcia-Suarez [Mon, 7 Dec 2009 12:41:05 +0000]
Fix [perl #71078] Smart match against @_ gives false negatives
@_ can contain NULLs for undefined elements
Chip Salzenberg [Mon, 7 Dec 2009 08:57:10 +0000]
Merge branch 'blead' of ssh://perl5.git.perl.org/perl into blead
Tony Cook [Mon, 7 Dec 2009 03:43:41 +0000]
Fix for
7baa4690 breakage of threaded builds.
Jesse Vincent [Sun, 6 Dec 2009 20:13:48 +0000]
Import ExtUtils::MakeMaker 6.55_03 from CPAN
hv@crypt.org [Sun, 6 Dec 2009 21:24:39 +0000]
[perl #71000] Wrong variable name in warning
Add a new warning "Missing argument in %s"
Reini Urban [Sun, 6 Dec 2009 14:31:23 +0000]
Innocently looking union member swap
This helps statically initializing union members on gcc,
otherwise we get "initializer element is not computable at load time".
This speeds up initializing larger B::C/B::CC compiled programs with -O1/-O2 by 10%.
Karl Williamson [Sun, 6 Dec 2009 05:21:38 +0000]
qr/\X/ expansion
Chip Salzenberg [Thu, 29 Oct 2009 00:08:05 +0000]
avoid infinite chain of utf8 magic
Chip Salzenberg [Tue, 27 Oct 2009 02:39:20 +0000]
when removing magic, fix SvMAGICAL flags (unless save_magic is active)
Chip Salzenberg [Wed, 28 Oct 2009 23:41:02 +0000]
preserve readonly flag when saving and restoring magic flags
hv@crypt.org [Sat, 5 Dec 2009 00:50:02 +0000]
t/test.pl fresh_perl_like() bug
The existing code does the wrong thing if a string (rather than a compiled
regexp) is supplied for $expected. Happily perl will do the right thing
if you let it.
Hugo
Message-Id: <
200912041444.nB4EiqBj004436@zen.crypt.org>
Craig A. Berry [Fri, 4 Dec 2009 23:48:27 +0000]
Port lib/unicore/mktables to VMS.
This is mostly just making filename comparisons and lookups of filenames
in hashes case blind, which is necessary since filename case is not
(by default) preserved.
Nicholas Clark [Fri, 4 Dec 2009 15:16:04 +0000]
Silence Win32 compiler warning
Subject: Re: Smoke [5.11.2] v5.11.2-105-gc2123ae FAIL(F) MSWin32 Win2000 SP4 (x86/1 cpu)
Message-ID: <
20091204140536.GD2582@plum.flirble.org>
(amended to fix the obvious typo)
Reini Urban [Thu, 3 Dec 2009 19:47:55 +0000]
pp_sort.c typo: stabilty
I'm now working on the sort code in th perl compiler as you can see :)
--
Reini Urban
http://phpwiki.org/ http://murbreak.at/
From
b8c749be70f51499fe1ffd9e483ee3a0a8305d9b Mon Sep 17 00:00:00 2001
From: Reini Urban <rurban@cpan.org>
Date: Thu, 3 Dec 2009 19:44:57 +0000
Subject: [PATCH] fix typo: stabilty
Signed-off-by: H.Merijn Brand <h.m.brand@xs4all.nl>
Nicholas Clark [Thu, 3 Dec 2009 10:07:51 +0000]
Another C<return NORMAL> needed post
bb4c52e023e0, missed from 805bf316c58a.
Rafael Garcia-Suarez [Thu, 3 Dec 2009 10:34:08 +0000]
Revert makefile part of the Unicode 5.2 patch
lib/unicore/version was getting removed by "make distclean".
Rafael Garcia-Suarez [Thu, 3 Dec 2009 10:25:12 +0000]
Add new Unicode data file in MANIFEST
Karl Williamson [Thu, 3 Dec 2009 04:36:17 +0000]
Unicode 5.2
Tony Cook [Thu, 3 Dec 2009 09:20:09 +0000]
[perl #70802] -i'*' refuses to work
Nicholas Clark [Wed, 2 Dec 2009 16:01:10 +0000]
Ensure that pp_qr returns a new regexp SV each time. Resolves RT #69852.
Instead of returning a(nother) reference to the (pre-compiled) regexp in the
optree, use reg_temp_copy() to create a copy of it, and return a reference to
that. This resolves issues about Regexp::DESTROY not being called in a timely
fashion (the original bug tracked by RT #69852), as well as bugs related to
blessing regexps, and of assigning to regexps, as described in correspondence
added to the ticket.
It transpires that we also need to undo the SvPVX() sharing when ithreads
cloning a Regexp SV, because mother_re is set to NULL, instead of a cloned
copy of the mother_re. This change might fix bugs with regexps and threads in
certain other situations, but as yet neither tests nor bug reports have
indicated any problems, so it might not actually be an edge case that it's
possible to reach.
Karl Williamson [Tue, 1 Dec 2009 05:32:23 +0000]
work-around Carp/overloading miniperl problem
Jan Dubois [Wed, 2 Dec 2009 09:34:18 +0000]
Cleanup all scopes before exiting a pseudo-forked process.
perl_destruct() contains an assertion that the scope stack
is empty. The remaining scopes are due to fork() being
called from within a BEGIN block.
Jan Dubois [Wed, 2 Dec 2009 07:11:03 +0000]
Revert commit
cbdd5331.
Commit
a09dc31d1 says:
It's possible that someone has built a module with -DDEBUGGING,
but they're using it against a perl built non-DEBUGGING, [...]
which means that the PL_scopestack_name symbol must always be exported.
Jan Dubois [Wed, 2 Dec 2009 07:03:09 +0000]
Add another hint about setting WIN64=undef to build a 32-bit Perl.
Of course anyone who read README.win32 already knew this. :)
Jan Dubois [Wed, 2 Dec 2009 06:55:55 +0000]
Don't redefine isnan() macro if MinGW already defined one.
Leon Brocard [Tue, 1 Dec 2009 13:45:42 +0000]
Update to reflect that I release 5.11.2
Tony Cook [Tue, 1 Dec 2009 11:25:39 +0000]
-Dmad: double free or corruption
> If your perl has -Dmad, the following program crashes:
>
> $ bleadperl -we '$x="x" x 257; eval "for $x"'
> *** glibc detected *** bleadperl: double free or corruption (!prev): 0x
0000000001dca670 ***
Change
6136c704 changed S_scan_ident from:
e = d + destlen - 3;
to:
register char * const e = d + destlen + 3;
where e is used to mark the end of the buffer, this meant that the
various buffer end checks allowed the various buffers supplied
S_scan_ident to overflow.
Attached is a fix, various tests with fencepost checks on different
identifier lengths, and the specific case mentioned in the ticket.
Tony
Signed-off-by: H.Merijn Brand <h.m.brand@xs4all.nl>
Nicholas Clark [Tue, 1 Dec 2009 11:25:37 +0000]
make regen_perly following
5ca68648d92617d8.
Tony Cook [Tue, 1 Dec 2009 08:12:30 +0000]
-Dmad broken for c++
On Tue, Dec 01, 2009 at 08:22:38AM +0100, H.Merijn Brand wrote:
> perly.y: In function 'int Perl_madparse(PerlInterpreter*)':
> perly.y:335: error: cast from 'TOKEN*' to 'line_t' loses precision
> perly.y:348: error: cast from 'TOKEN*' to 'line_t' loses precision
> perly.y:430: error: cast from 'TOKEN*' to 'line_t' loses precision
Uses the same mechanism used in if, unless to retrieve an ival set in
toke.c
Signed-off-by: H.Merijn Brand <h.m.brand@xs4all.nl>
Rafael Garcia-Suarez [Mon, 30 Nov 2009 23:21:17 +0000]
Make the new test work in the core and bump version of Safe to 2.20
Tim Bunce [Mon, 30 Nov 2009 23:15:21 +0000]
[rt.cpan.org #51574] Safe.pm sort {} bug accessing $a and $b with -Dusethreads
Eric Brine [Sat, 28 Nov 2009 04:52:41 +0000]
Test modifiability of range elements
Zefram [Mon, 30 Nov 2009 10:42:14 +0000]
Document if.pm in perlfunc
Rafael Garcia-Suarez [Mon, 30 Nov 2009 09:30:41 +0000]
Make distclean work again
Eric Brine [Fri, 27 Nov 2009 00:16:16 +0000]
Fix -DPERL_NO_UTF16_FILTER
Rafael Garcia-Suarez [Sun, 29 Nov 2009 22:30:45 +0000]
Make split warn in void context
Nicholas Clark [Sun, 29 Nov 2009 19:02:05 +0000]
Handle $@ being assigned a read-only value (without error or busting the stack).
Discovered whilst investigating RT #70862.
Nicholas Clark [Sun, 29 Nov 2009 16:42:42 +0000]
Fix RT #70862 by converting ERRSV to GvSVn() to ensure a non-NULL GvSV().
Steve Peters [Sun, 29 Nov 2009 05:45:10 +0000]
Merge branch 'blead' of ssh://stevep@perl5.git.perl.org/gitroot/perl into blead
Steve Peters [Sun, 29 Nov 2009 05:43:11 +0000]
Add error codes for getaddrinfo() and getnameinfo()
Karl Williamson [Sat, 28 Nov 2009 19:04:34 +0000]
mktables performance improvement
The attached patch got the easiest performance improvements to mktables.
Hopefully this is good enough for now.
This involved:
1) Nicholas' patch
2) I stored complete_name instead of recomputing it each time.
3) Used $_[xxx] instead of shift in very heavily used subroutines
4) removed trace accidentally left in.
I also changed the misspelled subroutine name discovered by Craig Berry.
I searched for any other misspellings and didn't find any.
Also removed trailing white space that keeps creeping back in, and now
this doesn't generate pod entries if not outputting a pod file, and
clarified warning message if no mktables.lst is present.
I couldn't figure out a way to conditionally use 'no overloading', as
it is called at compile time. So I just commented out the old stuff
that will work for 5.8, with a note about using that if you want to
use 5.8
Moritz Lenz [Thu, 26 Nov 2009 23:33:09 +0000]
Document backreferences to groups that did not match
Also add a test for that, fill in test description, and sneak in a vim
modeline for re_tests
Vincent Pit [Sat, 28 Nov 2009 12:27:02 +0000]
Allow a closing brace after an "use VERSION"
This fixes [perl #70884] : use VERSION in BLOCK without semicolon -> syntax error
Zefram [Thu, 26 Nov 2009 16:41:22 +0000]
-Dmad minitest failure bisect
I wrote:
>In my tests with -Dmad, I'm still getting a test failure ("panic: input
>overflow") from t/op/incfilter.t. The underlying problem is the filter
>layer mishandling things when a filter function gives it a multiline
>string, so it generates an invalid SV state (strlen(SvPVX(PL_linestr))
>> SvCUR(PL_linestr)). This faulty state also occurs without -Dmad,
>and so doesn't appear to be Mad-related, it just doesn't in practice
>cause the test panic without -Dmad. I'm investigating this bug now.
It's fixed by the attached patch. Since the bug is an inconsistency
in the SV data structure, it can't be sensibly tested from Perl code,
so I'm at a loss for writing a test script. Hopefully that panic with
-Dmad is sufficient.
-zefram
Signed-off-by: H.Merijn Brand <h.m.brand@xs4all.nl>
Zefram [Thu, 26 Nov 2009 15:19:09 +0000]
-Dmad minitest failure bisect
Tony Cook wrote:
>Smokes with -Dmad have been failing during make minitest since the
>middle of last month.
Mostly fixed by the attached patch. The fault is a logic error on my
part, probably from the early phase of developing the lexer API patch,
when I didn't properly understand the various buffer pointer variables.
In my tests with -Dmad, I'm still getting a test failure ("panic: input
overflow") from t/op/incfilter.t. The underlying problem is the filter
layer mishandling things when a filter function gives it a multiline
string, so it generates an invalid SV state (strlen(SvPVX(PL_linestr))
> SvCUR(PL_linestr)). This faulty state also occurs without -Dmad,
and so doesn't appear to be Mad-related, it just doesn't in practice
cause the test panic without -Dmad. I'm investigating this bug now.
-zefram
Signed-off-by: H.Merijn Brand <h.m.brand@xs4all.nl>
Tim Bunce (via RT) [Wed, 25 Nov 2009 11:02:24 +0000]
perlbug reports "Message sent" even when it hasn't
perlbug doesn't check the return status of Mail::Send and so reports
"Message sent" even if it hasn't:
Zefram [Wed, 25 Nov 2009 22:17:52 +0000]
perl-5.11.2 breaks NYTProf savesrc option (Lexer API suspected)
Tim Bunce wrote:
>The primary issue is the off-by-one error in the array indexing.
There's a bit more to it than that. The indexing was off-by-one for
*some* places that process a new line, but correct for others, so the
saved source as a whole was mangled rather than simply offset. Also,
there were some redundant calls to update_debugger_info(), so some lines
got saved twice, in some cases off-by-one for one saving and not for
the other. The saved source is, therefore, hopelessly broken in 5.11.2.
Attached patch fixes the source saving. Includes a new test, which works
through all reachable places that source lines get saved. This should
close RT #70804.
-zefram
Jesse Vincent [Wed, 25 Nov 2009 20:47:41 +0000]
Jerry's last patch to add a TODO test didn't update the test count.
Thanks to Nicholas++ for pointing this out
Jesse Vincent [Wed, 25 Nov 2009 18:47:38 +0000]
Moved the original test file from the previous patch into t/op/lex.t,
Aded details on the bug it fixes to the tests.
Gerard Goossen [Tue, 24 Nov 2009 11:13:13 +0000]
add interpolation test for [perl #70091]
(Updated by Jesse Vincent to put the test in comp rather than a new
toplevel test directory)
Gerard Goossen [Tue, 24 Nov 2009 11:06:55 +0000]
Also skip spaces after variable if we are within lexical brackets. Fixes #70091: Segmentation fault in hash lookup in regex substitution
Gerard Goossen [Tue, 24 Nov 2009 10:27:04 +0000]
Add assertions that pp_padav and pp_padhv push scalars of the correct type.
Karl Williamson [Wed, 25 Nov 2009 15:30:57 +0000]
further windows makefiles fixes
Message-ID: <
4B0D4F71.3010606@khwilliamson.com>
Karl Williamson [Tue, 24 Nov 2009 20:47:37 +0000]
wrap uniprops.t; makefile changes for mktables
Message-ID: <
4B0C4744.7080401@khwilliamson.com>