This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
perl5.git
18 years agoassert() that we can't reach two unreachable code paths
Nicholas Clark [Sat, 11 Feb 2006 00:40:04 +0000 (00:40 +0000)]
assert() that we can't reach two unreachable code paths
(in Perl_magic_get and Perl_pp_shift).

p4raw-id: //depot/perl@27154

18 years agoThe return value from S_hv_auxinit can be ignored, as it's also stored
Nicholas Clark [Fri, 10 Feb 2006 22:37:23 +0000 (22:37 +0000)]
The return value from S_hv_auxinit can be ignored, as it's also stored
within the passed in hv parameter.

p4raw-id: //depot/perl@27153

18 years agoA better fix than 27148
Steve Hay [Fri, 10 Feb 2006 17:46:35 +0000 (17:46 +0000)]
A better fix than 27148

Subject: Win32 assert (was Re: Smoke [5.9.4] 27145 FAIL(M) MSWin32 WinXP/.Net SP2 (x86/2 cpu))
From: Nicholas Clark <nick@ccl4.org>
Date: Fri, 10 Feb 2006 14:58:47 +0000
Message-ID: <20060210145847.GA85020@plum.flirble.org>

Subject: RE: Win32 assert (was Re: Smoke [5.9.4] 27145 FAIL(M) MSWin32 WinXP/.Net SP2 (x86/2 cpu))
From: "Jan Dubois" <jand@ActiveState.com>
Date: Fri, 10 Feb 2006 09:03:41 -0800
Message-ID: <024401c62e63$f1e53c10$6062a8c0@candy>

p4raw-id: //depot/perl@27152

18 years agoWith PERL_POISON defined, ensure freshly malloc()ed memory isn't zeros,
Nicholas Clark [Fri, 10 Feb 2006 14:25:33 +0000 (14:25 +0000)]
With PERL_POISON defined, ensure freshly malloc()ed memory isn't zeros,
and when PERL_TRACK_MEMPOOL is also defined scribble on any extension
added by realloc().

p4raw-id: //depot/perl@27151

18 years agoFix the effects of 27136 on Win32 with DEBUGGING and PERL_IMPLICIT_SYS
Steve Hay [Fri, 10 Feb 2006 11:20:23 +0000 (11:20 +0000)]
Fix the effects of 27136 on Win32 with DEBUGGING and PERL_IMPLICIT_SYS

p4raw-id: //depot/perl@27148

18 years agoChange 27136 seems to have inadvertently upset some compilers,
Nicholas Clark [Thu, 9 Feb 2006 23:23:59 +0000 (23:23 +0000)]
Change 27136 seems to have inadvertently upset some compilers,
presumably where NULL was defined as 0, not (void *)0

p4raw-id: //depot/perl@27145

18 years agoFix breakage in S_set_csh() from change #27136.
Steve Peters [Thu, 9 Feb 2006 18:25:02 +0000 (18:25 +0000)]
Fix breakage in S_set_csh() from change #27136.
p4raw-link: @27136 on //depot/perl: 5f66b61cbe1f2ea985f00c0d9a66153e7a3b6ee1

p4raw-id: //depot/perl@27140

18 years agoDocument incompatible interpretation of chdir FOO.
Gisle Aas [Thu, 9 Feb 2006 18:09:01 +0000 (18:09 +0000)]
Document incompatible interpretation of chdir FOO.
Ref change 27125.

p4raw-id: //depot/perl@27139

18 years agoReally regenerate headers
Rafael Garcia-Suarez [Thu, 9 Feb 2006 17:49:18 +0000 (17:49 +0000)]
Really regenerate headers

p4raw-id: //depot/perl@27138

18 years agoRegenerated headers for change #27136.
Steve Peters [Thu, 9 Feb 2006 17:40:02 +0000 (17:40 +0000)]
Regenerated headers for change #27136.
p4raw-link: @27136 on //depot/perl: 5f66b61cbe1f2ea985f00c0d9a66153e7a3b6ee1

p4raw-id: //depot/perl@27137

18 years agoTrying my "remove the pTHXes" patch again
Andy Lester [Thu, 9 Feb 2006 09:40:18 +0000 (03:40 -0600)]
Trying my "remove the pTHXes" patch again
Message-ID: <20060209154018.GA14610@petdance.com>

p4raw-id: //depot/perl@27136

18 years agoSkip new chdir warning test if fchdir is not available
Rafael Garcia-Suarez [Thu, 9 Feb 2006 12:38:35 +0000 (12:38 +0000)]
Skip new chdir warning test if fchdir is not available

p4raw-id: //depot/perl@27135

18 years agoAvoid calling all the scanning code in gv_fetchpvn_flags when saving
Nicholas Clark [Thu, 9 Feb 2006 12:37:13 +0000 (12:37 +0000)]
Avoid calling all the scanning code in gv_fetchpvn_flags when saving
$1 etc, as we can perfom a hash lookup on %:: directly.

p4raw-id: //depot/perl@27134

18 years agolastparen in the regexp structure is never initialised, and so will
Nicholas Clark [Wed, 8 Feb 2006 20:57:24 +0000 (20:57 +0000)]
lastparen in the regexp structure is never initialised, and so will
still be uninitialised after a failed match, yet various points in
mg.c read it come what may. Should fix bug 38461.

p4raw-id: //depot/perl@27133

18 years agoBareword file handles not supported for chown/chmod/utime.
Gisle Aas [Wed, 8 Feb 2006 17:52:16 +0000 (17:52 +0000)]
Bareword file handles not supported for chown/chmod/utime.

p4raw-id: //depot/perl@27132

18 years agoWave hands about encoding::warnings
Rafael Garcia-Suarez [Wed, 8 Feb 2006 16:57:48 +0000 (16:57 +0000)]
Wave hands about encoding::warnings

p4raw-id: //depot/perl@27131

18 years ago[perl #38464] chdir() on closed handles
Peter Dintelmann [Wed, 8 Feb 2006 01:33:54 +0000 (17:33 -0800)]
[perl #38464] chdir() on closed handles
From: Peter Dintelmann (via RT) <perlbug-followup@perl.org>
Message-ID: <rt-3.0.11-38464-129771.3.46114092925276@perl.org>

Plus a regression test.

p4raw-id: //depot/perl@27130

18 years agoHandle unused args
Andy Lester [Tue, 7 Feb 2006 16:36:56 +0000 (10:36 -0600)]
Handle unused args
Message-ID: <20060207223656.GA5177@petdance.com>

p4raw-id: //depot/perl@27129

18 years ago[patch] several broken make *.valgrind targets
Jim Cromie [Tue, 7 Feb 2006 15:11:54 +0000 (08:11 -0700)]
[patch] several broken make *.valgrind targets
Message-ID: <43E91B2A.4060700@gmail.com>

p4raw-id: //depot/perl@27128

18 years agoBorland's C compiler warns that the & is unnecessary.
Nicholas Clark [Wed, 8 Feb 2006 00:05:44 +0000 (00:05 +0000)]
Borland's C compiler warns that the & is unnecessary.

p4raw-id: //depot/perl@27127

18 years agoMerging pp_bit_or and pp_bit_xor shrinks the object code by about .7K.
Nicholas Clark [Tue, 7 Feb 2006 18:01:26 +0000 (18:01 +0000)]
Merging pp_bit_or and pp_bit_xor shrinks the object code by about .7K.
The overloading tests are not free.

p4raw-id: //depot/perl@27126

18 years agoAllow bareword file handle as argument to chdir().
Gisle Aas [Tue, 7 Feb 2006 17:32:50 +0000 (17:32 +0000)]
Allow bareword file handle as argument to chdir().
This copies the mechanism used by truncate().
Fixes bug #38457.

p4raw-id: //depot/perl@27125

18 years agoAll the trancendental unary operators can be merged into PP_sin
Nicholas Clark [Tue, 7 Feb 2006 17:12:08 +0000 (17:12 +0000)]
All the trancendental unary operators can be merged into PP_sin
(cos, exp, log, sqrt)

p4raw-id: //depot/perl@27124

18 years agoAvoid gcc warning about possibly uninitialised variables.
Nicholas Clark [Tue, 7 Feb 2006 15:52:18 +0000 (15:52 +0000)]
Avoid gcc warning about possibly uninitialised variables.

p4raw-id: //depot/perl@27123

18 years agoCreate new macros AMG_CALLun_var, AMG_CALLun_var and tryAMAGICun_var
Nicholas Clark [Tue, 7 Feb 2006 15:28:41 +0000 (15:28 +0000)]
Create new macros AMG_CALLun_var, AMG_CALLun_var and tryAMAGICun_var
which don't do the pre-processor string manipulation internally.

p4raw-id: //depot/perl@27122

18 years agopp_pop can be implemented by pp_shift.
Nicholas Clark [Tue, 7 Feb 2006 14:57:36 +0000 (14:57 +0000)]
pp_pop can be implemented by pp_shift.

p4raw-id: //depot/perl@27121

18 years agoRe: [PATCH] Documentation patch for overload
Steffen Müller [Mon, 6 Feb 2006 10:58:44 +0000 (11:58 +0100)]
Re: [PATCH] Documentation patch for overload
Message-ID: <43E71DD4.5010108@sneakemail.com>

p4raw-id: //depot/perl@27120

18 years agopp_hex can be implemented trivially by pp_oct, making pp_hex a mathom.
Nicholas Clark [Tue, 7 Feb 2006 14:11:42 +0000 (14:11 +0000)]
pp_hex can be implemented trivially by pp_oct, making pp_hex a mathom.

p4raw-id: //depot/perl@27119

18 years agoMerge pp_index and pp_rindex - we have another mathom.
Nicholas Clark [Tue, 7 Feb 2006 13:13:44 +0000 (13:13 +0000)]
Merge pp_index and pp_rindex - we have another mathom.

p4raw-id: //depot/perl@27118

18 years agoSome refactoring to converge pp_index and pp_rindex
Nicholas Clark [Tue, 7 Feb 2006 12:41:27 +0000 (12:41 +0000)]
Some refactoring to converge pp_index and pp_rindex

p4raw-id: //depot/perl@27117

18 years agoFix bug 38454 (rindex corrects for $[ on bytes rather than UTF-8)
Nicholas Clark [Tue, 7 Feb 2006 12:19:17 +0000 (12:19 +0000)]
Fix bug 38454 (rindex corrects for $[ on bytes rather than UTF-8)

p4raw-id: //depot/perl@27116

18 years agoRe: Configure.com issue, more quoting needed.
Peter Prymmer [Mon, 6 Feb 2006 13:27:14 +0000 (08:27 -0500)]
Re: Configure.com issue, more quoting needed.
Message-id:  <OF3536E719.A8B8D167-ON8525710D.0065011E-8525710D.00655ED0@factset.com>

p4raw-id: //depot/perl@27115

18 years agopatch@27082 Allow fatal exceptions to bring up VMS debugger
John E. Malmberg [Sat, 4 Feb 2006 16:04:32 +0000 (11:04 -0500)]
patch@27082 Allow fatal exceptions to bring up VMS debugger
From: "John E. Malmberg" <wb8tyw@qsl.net>
Message-id: <43E516E0.1000508@qsl.net>

p4raw-id: //depot/perl@27114

18 years agoOptimise index so that if the big string is ISO-8859-1 but the little
Nicholas Clark [Mon, 6 Feb 2006 22:28:41 +0000 (22:28 +0000)]
Optimise index so that if the big string is ISO-8859-1 but the little
string is UTF-8, it tries to downgrade the little string, rather than
upgrade the big string. For half-meg big strings this is a fourfold
speed gain.

p4raw-id: //depot/perl@27113

18 years agoGiven that the memory allocated in Perl_bytes_from_utf8 and
Nicholas Clark [Mon, 6 Feb 2006 21:40:57 +0000 (21:40 +0000)]
Given that the memory allocated in Perl_bytes_from_utf8 and
Perl_bytes_to_utf8 will immediately be written to, I see no need to
allocate it zeroed.

p4raw-id: //depot/perl@27112

18 years agoRename S_glob_assign to S_glob_assign_glob
Nicholas Clark [Mon, 6 Feb 2006 20:52:41 +0000 (20:52 +0000)]
Rename S_glob_assign to S_glob_assign_glob
Rename S_pvgv_assign to S_glob_assign_ref

I don't think that they share enough common code to merge them.

p4raw-id: //depot/perl@27111

18 years agoI think that "merge Perl_sv_2[inpu]v" and "reduce duplication in
Nicholas Clark [Mon, 6 Feb 2006 20:12:58 +0000 (20:12 +0000)]
I think that "merge Perl_sv_2[inpu]v" and "reduce duplication in
sv_setsv_flags" are about as done as they can be.

p4raw-id: //depot/perl@27110

18 years agoIn Perl_sv_upgrade, the assignment to new_type_details only needs to
Nicholas Clark [Mon, 6 Feb 2006 20:04:35 +0000 (20:04 +0000)]
In Perl_sv_upgrade, the assignment to new_type_details only needs to
be done once.

p4raw-id: //depot/perl@27109

18 years agoDocument the limitations of calling exit() (instead of _exit()) from
Steve Peters [Mon, 6 Feb 2006 19:43:17 +0000 (19:43 +0000)]
Document the limitations of calling exit() (instead of _exit()) from
a child process on Solaris.

p4raw-id: //depot/perl@27108

18 years agoRemove a duplicate flag copy line from Perl_sv_setsv_flags.
Nicholas Clark [Mon, 6 Feb 2006 18:45:35 +0000 (18:45 +0000)]
Remove a duplicate flag copy line from Perl_sv_setsv_flags.

p4raw-id: //depot/perl@27107

18 years agoMove Perl_save_long, Perl_save_I16, Perl_save_I8, Perl_save_iv,
Nicholas Clark [Mon, 6 Feb 2006 18:07:28 +0000 (18:07 +0000)]
Move Perl_save_long, Perl_save_I16, Perl_save_I8, Perl_save_iv,
Perl_save_nogv, Perl_save_list, Perl_save_destructor to mathoms.c
Perl_save_svref can't actually move because it calls a static function.

p4raw-id: //depot/perl@27106

18 years agoExpunge USE_5005THREADS from perl.c
Nicholas Clark [Mon, 6 Feb 2006 16:50:22 +0000 (16:50 +0000)]
Expunge USE_5005THREADS from perl.c

p4raw-id: //depot/perl@27105

18 years agoFix bug #38439 - reference to typeglob assignment needs to be based
Nicholas Clark [Mon, 6 Feb 2006 13:08:34 +0000 (13:08 +0000)]
Fix bug #38439 - reference to typeglob assignment needs to be based
on SvROK(sstr) rather than SvTYPE(sstr) == SVt_RV.

p4raw-id: //depot/perl@27104

18 years ago5.8.8: Unicos test skip (and for the record: config.sh and harness results)
Jarkko Hietaniemi [Sun, 5 Feb 2006 12:16:37 +0000 (14:16 +0200)]
5.8.8: Unicos test skip (and for the record: config.sh and harness results)
Message-ID: <43E5D085.90400@iki.fi>

p4raw-id: //depot/perl@27103

18 years agoCleaning up shadowed variables
Andy Lester [Sun, 5 Feb 2006 21:56:43 +0000 (15:56 -0600)]
Cleaning up shadowed variables
Message-Id: <B375E848-FA07-479D-9571-FA3925106D16@petdance.com>

p4raw-id: //depot/perl@27102

18 years agoMerge the PVCV case into the common code in S_pvgv_assign.
Nicholas Clark [Sun, 5 Feb 2006 23:44:12 +0000 (23:44 +0000)]
Merge the PVCV case into the common code in S_pvgv_assign.

p4raw-id: //depot/perl@27101

18 years agoMerge the PVAV and PVHV clauses into the common code of S_pvgv_assign.
Nicholas Clark [Sun, 5 Feb 2006 23:20:21 +0000 (23:20 +0000)]
Merge the PVAV and PVHV clauses into the common code of S_pvgv_assign.

p4raw-id: //depot/perl@27100

18 years agoMerge the default case (GvSV) with the common code in S_pvgv_assign.
Nicholas Clark [Sun, 5 Feb 2006 23:02:58 +0000 (23:02 +0000)]
Merge the default case (GvSV) with the common code in S_pvgv_assign.

p4raw-id: //depot/perl@27099

18 years agoStart to merge the 6 arms of S_pvgv_assign into one.
Nicholas Clark [Sun, 5 Feb 2006 22:41:42 +0000 (22:41 +0000)]
Start to merge the 6 arms of S_pvgv_assign into one.

p4raw-id: //depot/perl@27098

18 years ago[patch] rework arenas - repair arenasets
Jim Cromie [Sun, 5 Feb 2006 05:31:14 +0000 (22:31 -0700)]
[patch] rework arenas - repair arenasets
Message-ID: <43E5F012.1080203@gmail.com>
Date: Sun, 05 Feb 2006 05:31:14 -0700

p4raw-id: //depot/perl@27097

18 years agoDocumentation patch for overload
Steffen Müller [Sun, 5 Feb 2006 12:26:00 +0000 (12:26 +0000)]
Documentation patch for overload
Message-ID: <43E5E0E9.1070707@sneakemail.com>

p4raw-id: //depot/perl@27096

18 years agoThe message "Can't upgrade that kind of scalar" can be far more
Nicholas Clark [Sun, 5 Feb 2006 20:51:43 +0000 (20:51 +0000)]
The message "Can't upgrade that kind of scalar" can be far more
informative.

p4raw-id: //depot/perl@27095

18 years agonew_body_type doesn't need to subtract the offset, that's what
Nicholas Clark [Sun, 5 Feb 2006 20:33:57 +0000 (20:33 +0000)]
new_body_type doesn't need to subtract the offset, that's what
new_body_allocated is for.

p4raw-id: //depot/perl@27094

18 years agoYes, I broke Win32 with change 27089 by having a { in the wrong place.
Nicholas Clark [Sun, 5 Feb 2006 20:26:40 +0000 (20:26 +0000)]
Yes, I broke Win32 with change 27089 by having a { in the wrong place.

p4raw-id: //depot/perl@27093

18 years agoRemove "Can't upgrade to undef" from perldiag, as the corresponding
Nicholas Clark [Sun, 5 Feb 2006 19:52:50 +0000 (19:52 +0000)]
Remove "Can't upgrade to undef" from perldiag, as the corresponding
code is now gone from sv_upgrade.

p4raw-id: //depot/perl@27092

18 years agoMerge and refactor the PVAV and PVHV code in Perl_sv_upgrade, which
Nicholas Clark [Sun, 5 Feb 2006 19:45:37 +0000 (19:45 +0000)]
Merge and refactor the PVAV and PVHV code in Perl_sv_upgrade, which
reduces the object size by about 80 bytes.

p4raw-id: //depot/perl@27091

18 years agoEliminate a croak we can't get to.
Nicholas Clark [Sun, 5 Feb 2006 18:51:17 +0000 (18:51 +0000)]
Eliminate a croak we can't get to.

p4raw-id: //depot/perl@27090

18 years agoWithout this Windows and Netware compilers should be complaining
Nicholas Clark [Sun, 5 Feb 2006 11:15:34 +0000 (11:15 +0000)]
Without this Windows and Netware compilers should be complaining
about perl.c having code before a declaration, *but only if compiling
with PERL_TRACK_MEMPOOL*. So, will it fix the smoke failures? Place
bets now.

p4raw-id: //depot/perl@27089

18 years agoDon't export PL_memory_debug_header to Win32 (and others) unless
Steve Peters [Sun, 5 Feb 2006 05:08:02 +0000 (05:08 +0000)]
Don't export PL_memory_debug_header to Win32 (and others) unless
PERL_TRACK_MEMPOOL is defined.

p4raw-id: //depot/perl@27088

18 years agoFixed embed.fnc entry for S_forbid_setid() and picked up a change in
Steve Peters [Sun, 5 Feb 2006 04:13:27 +0000 (04:13 +0000)]
Fixed embed.fnc entry for S_forbid_setid() and picked up a change in
pod/perlapi.pod for good measure when regenerating files.

p4raw-id: //depot/perl@27087

18 years agobcc was issuing a warning about this line in Perl_sv_2cv, so tweak it
Nicholas Clark [Sat, 4 Feb 2006 22:51:17 +0000 (22:51 +0000)]
bcc was issuing a warning about this line in Perl_sv_2cv, so tweak it
to keep it happy.

p4raw-id: //depot/perl@27086

18 years agoAdd regression tests to lib/charnames.t to ensure that unicore/Names.pl
Nicholas Clark [Sat, 4 Feb 2006 22:13:08 +0000 (22:13 +0000)]
Add regression tests to lib/charnames.t to ensure that unicore/Names.pl
is 100% 7 bit ASCII.

p4raw-id: //depot/perl@27085

18 years agoEnhance PERL_TRACK_MEMPOOL so that it also emulates the PerlHost
Nicholas Clark [Sat, 4 Feb 2006 19:09:17 +0000 (19:09 +0000)]
Enhance PERL_TRACK_MEMPOOL so that it also emulates the PerlHost
behaviour of freeing up all memory at thread exit. With this and
tools such as valgrind you will now get warnings as soon as you
read from the deallocated memory, rather than just a warning much
later about freeing to the wrong pool.

p4raw-id: //depot/perl@27084

18 years agoThe TODO 'Tidy up global variables' can go, replaced with a second
Nicholas Clark [Sat, 4 Feb 2006 18:48:09 +0000 (18:48 +0000)]
The TODO 'Tidy up global variables' can go, replaced with a second
paragraph on 'Ordering of "global" variables'.

p4raw-id: //depot/perl@27083

18 years agoI believe that mathoms have been completely TODOne.
Nicholas Clark [Sat, 4 Feb 2006 17:11:54 +0000 (17:11 +0000)]
I believe that mathoms have been completely TODOne.

p4raw-id: //depot/perl@27082

18 years agoRejig the definition of ARENAS_PER_SET to maximise the size of the
Nicholas Clark [Sat, 4 Feb 2006 13:27:00 +0000 (13:27 +0000)]
Rejig the definition of ARENAS_PER_SET to maximise the size of the
array whilst keeping the structure within PERL_ARENA_SIZE.

p4raw-id: //depot/perl@27081

18 years agoTweak arena sets to avoid assignment between different pointers.
Nicholas Clark [Sat, 4 Feb 2006 13:00:06 +0000 (13:00 +0000)]
Tweak arena sets to avoid assignment between different pointers.

p4raw-id: //depot/perl@27080

18 years ago[patch] arena rework - arena sets
Jim Cromie [Tue, 31 Jan 2006 04:52:06 +0000 (21:52 -0700)]
[patch] arena rework - arena sets
Message-ID: <43DF4F66.4080808@gmail.com>
Date: Tue, 31 Jan 2006 04:52:06 -0700

p4raw-id: //depot/perl@27079

18 years agowarnings.h does some evil(*) pointer arithmetic on (SV *)0, so a
Nicholas Clark [Sat, 4 Feb 2006 11:10:52 +0000 (11:10 +0000)]
warnings.h does some evil(*) pointer arithmetic on (SV *)0, so a
simple replace of Nullsv with NULL turns out to be a bad idea.
* Technically undefined behaviour, I believe.

p4raw-id: //depot/perl@27078

18 years agoRemove obsolete comment.
Nicholas Clark [Sat, 4 Feb 2006 10:37:08 +0000 (10:37 +0000)]
Remove obsolete comment.

p4raw-id: //depot/perl@27077

18 years agoS_find_beginning is only called from one place, so by hoisting the
Nicholas Clark [Fri, 3 Feb 2006 23:21:37 +0000 (23:21 +0000)]
S_find_beginning is only called from one place, so by hoisting the
forbid_setid check for -x into the caller, we can eliminate the
suidscript parameter from S_find_beginning.

p4raw-id: //depot/perl@27076

18 years agoAnd as we've now got to the point where all calls to Perl_moreswitches
Nicholas Clark [Fri, 3 Feb 2006 22:41:55 +0000 (22:41 +0000)]
And as we've now got to the point where all calls to Perl_moreswitches
have -1 as the second argument, we can remove the second argument,
which gets us back to where we started, only with the elimination of
a per-thread variable.

p4raw-id: //depot/perl@27075

18 years agoThe MSDOS-only call to moreswitches will never reach forbid_setid, so
Nicholas Clark [Fri, 3 Feb 2006 22:18:45 +0000 (22:18 +0000)]
The MSDOS-only call to moreswitches will never reach forbid_setid, so
it can have a parameter of -1 without changing any behaviour.
I see no reason to forbid "command line" switches when they are
actually coming from the #! line of the script itself (as read by perl)
as they must have been written by the owner of the script, rather than
being crafted by the user manipulating the command line, and therefore
cannot be subverted. (Or alternatively, can be subverted as easily as
the rest of the script, and we're permitting *that* to be run)

p4raw-id: //depot/perl@27074

18 years agoPull the variables fdscript and suidscript into a much smaller scope.
Nicholas Clark [Fri, 3 Feb 2006 21:47:04 +0000 (21:47 +0000)]
Pull the variables fdscript and suidscript into a much smaller scope.
Replace suidscript with literal -1 where it could only have been -1.
(Re-indent the other code inside the new block creating the scope.
 Rejig the indentation of the conditionals)

p4raw-id: //depot/perl@27073

18 years agoChange 27070 failed to initialise a variable. (But despite that still
Nicholas Clark [Fri, 3 Feb 2006 21:23:21 +0000 (21:23 +0000)]
Change 27070 failed to initialise a variable. (But despite that still
passed all tests locally.)

p4raw-id: //depot/perl@27072

18 years agoSilence a warning about "statement not reached" from the Sun C compiler.
Nicholas Clark [Fri, 3 Feb 2006 20:12:10 +0000 (20:12 +0000)]
Silence a warning about "statement not reached" from the Sun C compiler.
It was accurate - the conditional code used to give 2 return statements.

p4raw-id: //depot/perl@27071

18 years agoEliminating PL_suidscript is more tricky, and requires changing the
Nicholas Clark [Fri, 3 Feb 2006 20:10:07 +0000 (20:10 +0000)]
Eliminating PL_suidscript is more tricky, and requires changing the
prototype of Perl_moreswitches.

p4raw-id: //depot/perl@27070

18 years agoPassing the flag to S_forbit_setid as a single char rather than a
Nicholas Clark [Fri, 3 Feb 2006 18:27:52 +0000 (18:27 +0000)]
Passing the flag to S_forbit_setid as a single char rather than a
string is a slight space optimisation.

p4raw-id: //depot/perl@27069

18 years agoOops. I *thought* that I had checked that all changed files were open.
Nicholas Clark [Fri, 3 Feb 2006 18:10:45 +0000 (18:10 +0000)]
Oops. I *thought* that I had checked that all changed files were open.
Clearly not. (Fixes change 27066)

p4raw-id: //depot/perl@27068

18 years agoOops. When changing warnings.pl one should re-run it. Well spotted
Nicholas Clark [Fri, 3 Feb 2006 17:06:43 +0000 (17:06 +0000)]
Oops. When changing warnings.pl one should re-run it. Well spotted
Rafael.

p4raw-id: //depot/perl@27067

18 years agoIt's actually easier to get rid of PL_fdscript than we thought.
Nicholas Clark [Fri, 3 Feb 2006 17:06:04 +0000 (17:06 +0000)]
It's actually easier to get rid of PL_fdscript than we thought.

p4raw-id: //depot/perl@27066

18 years agoRe: [PATCH] s/Null(gv|hv|sv)/NULL/g
Steven Schubiger [Fri, 3 Feb 2006 16:24:49 +0000 (17:24 +0100)]
Re: [PATCH] s/Null(gv|hv|sv)/NULL/g
Message-ID: <20060203152449.GI12591@accognoscere.homeunix.org>
Date: Fri, 3 Feb 2006 16:24:49 +0100

p4raw-id: //depot/perl@27065

18 years agoFix file where internal XS functions are defined
Rafael Garcia-Suarez [Fri, 3 Feb 2006 15:18:18 +0000 (15:18 +0000)]
Fix file where internal XS functions are defined

p4raw-id: //depot/perl@27064

18 years agoHelp to compile on VMS with UNLINK_ALL_VERSIONS
John E. Malmberg [Tue, 13 Dec 2005 22:18:56 +0000 (17:18 -0500)]
Help to compile on VMS with UNLINK_ALL_VERSIONS
See :
Subject: bug@26347 embed.fnc/embed.h with unlnk macro
From: "John E. Malmberg" <wb8tyw@qsl.net>
Message-ID: <439F8F20.4050207@qsl.net>

p4raw-id: //depot/perl@27063

18 years ago[PAUSE] CPAN Upload: A/AN/ANDK/CPAN-1.83_64.tar.gz
Andreas König [Thu, 2 Feb 2006 12:35:41 +0000 (13:35 +0100)]
[PAUSE] CPAN Upload: A/AN/ANDK/CPAN-1.83_64.tar.gz
Message-ID: <87acda0z8y.fsf@k75.linux.bogus>

p4raw-id: //depot/perl@27062

18 years agoL<PerlIO> and Pod::Html
SADAHIRO Tomoyuki [Mon, 9 Jan 2006 12:53:15 +0000 (21:53 +0900)]
L<PerlIO> and Pod::Html
Message-Id: <20060109125220.3BB0.BQW10602@nifty.com>

p4raw-id: //depot/perl@27061

18 years agos/Nullcv/NULL/g
Steven Schubiger [Thu, 2 Feb 2006 16:02:41 +0000 (17:02 +0100)]
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

18 years agoChange PL_perlio_fd_refcnt from a fixed size static array to a pointer
Nicholas Clark [Fri, 3 Feb 2006 13:06:00 +0000 (13:06 +0000)]
Change PL_perlio_fd_refcnt from a fixed size static array to a pointer
to a dynamic array.

p4raw-id: //depot/perl@27059

18 years agoIt would be nice to have tests in B for the functions getting changed.
Steve Peters [Thu, 2 Feb 2006 20:57:01 +0000 (20:57 +0000)]
It would be nice to have tests in B for the functions getting changed.

p4raw-id: //depot/perl@27057

18 years agoDeath to magic characters! No pun intended.
Steve Peters [Thu, 2 Feb 2006 20:42:56 +0000 (20:42 +0000)]
Death to magic characters!  No pun intended.

p4raw-id: //depot/perl@27056

18 years agoRe: [PATCH] s/Null(av|ch)/NULL/g
Steven Schubiger [Thu, 2 Feb 2006 10:38:49 +0000 (11:38 +0100)]
Re: [PATCH] s/Null(av|ch)/NULL/g
Message-ID: <20060202093849.GD12591@accognoscere.homeunix.org>

p4raw-id: //depot/perl@27054

18 years agogv_fetchpvn_flags ranks highly in the profile, and the ::/' scanning
Nicholas Clark [Thu, 2 Feb 2006 12:31:30 +0000 (12:31 +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

18 years agoUpdate to Module::CoreList 3.04.
Rafael Garcia-Suarez [Thu, 2 Feb 2006 11:22:43 +0000 (11:22 +0000)]
Update to Module::CoreList 3.04.
Cleanup corelist manpage.
Remove -w in corelist shebang to avoid spurious warnings.

p4raw-id: //depot/perl@27052

18 years agoperl-5.8.8 also contained change 27040.
Gisle Aas [Thu, 2 Feb 2006 10:05:32 +0000 (10:05 +0000)]
perl-5.8.8 also contained change 27040.

p4raw-id: //depot/perl@27051

18 years agoYet more instances of gv_fetchpv... that should be GV_ADD rather than
Nicholas Clark [Thu, 2 Feb 2006 00:36:38 +0000 (00:36 +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

18 years agoAnd another assumption.
Nicholas Clark [Thu, 2 Feb 2006 00:24:06 +0000 (00:24 +0000)]
And another assumption.

p4raw-id: //depot/perl@27049

18 years agoSince PERL_COPY_ON_WRITE became PERL_OLD_COPY_ON_WRITE it seems that
Nicholas Clark [Wed, 1 Feb 2006 23:38:53 +0000 (23:38 +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

18 years agoDoc nit found by Stephen Gordon.
Steve Peters [Wed, 1 Feb 2006 23:34:47 +0000 (23:34 +0000)]
Doc nit found by Stephen Gordon.

p4raw-id: //depot/perl@27047

18 years agodoubleplusoops. Apart from the documented NUL termination assumption.
Nicholas Clark [Wed, 1 Feb 2006 23:03:11 +0000 (23:03 +0000)]
doubleplusoops. Apart from the documented NUL termination assumption.
Now removed.

p4raw-id: //depot/perl@27046

18 years agoReferences can't have SVp_IOK or SVp_NOK set, so no need for all the
Nicholas Clark [Wed, 1 Feb 2006 22:39:29 +0000 (22:39 +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