15 years agoRemove some empty lines in a code example
Rafael Garcia-Suarez [Mon, 20 Apr 2009 07:51:38 +0000]
Remove some empty lines in a code example

15 years agoDocument that ~~ overload will only work on the RHS
Rafael Garcia-Suarez [Mon, 20 Apr 2009 07:46:00 +0000]
Document that ~~ overload will only work on the RHS

15 years agoAdjustments to the ~~ dispatch table
Rafael Garcia-Suarez [Wed, 25 Mar 2009 22:55:34 +0000]
Adjustments to the ~~ dispatch table

Hash slice existence now uses 'exists' instead of the complex recursive
invocation to ~~.

Any ~~ Array matches now keep the "any" element on the LHS of the smart
match. More special cases have been added for complex values of "Any".

15 years agoInline macro tryAMAGICbinSET() in smart match routine
Rafael Garcia-Suarez [Tue, 24 Mar 2009 17:30:20 +0000]
Inline macro tryAMAGICbinSET() in smart match routine

(to build tricks on a sane base)

15 years agoMore tests. Simplify tests for ~~ overloading
Rafael Garcia-Suarez [Tue, 24 Mar 2009 17:21:59 +0000]
More tests. Simplify tests for ~~ overloading

15 years agoFix equivalent code in perlsyn for Array~~Hash
Rafael Garcia-Suarez [Tue, 24 Mar 2009 13:16:15 +0000]
Fix equivalent code in perlsyn for Array~~Hash

It's an "any", not an "all". The synopsis 3 for Perl 6 has this code :
when $_ is an Array, and X is a Hash, $_ ~~ X means X.{any @$_}:exists.
Note that the implementation is right already (and was right in 5.10).

15 years agoSpecify that the behaviour of $foo ~~ Range should only depend on the RHS
Rafael Garcia-Suarez [Tue, 24 Mar 2009 10:42:18 +0000]
Specify that the behaviour of $foo ~~ Range should only depend on the RHS

(and not from the value being checked, as Zefram remarks)

15 years agoMake []~~\&f and {}~~\&f match
Rafael Garcia-Suarez [Tue, 24 Mar 2009 10:37:35 +0000]
Make []~~\&f and {}~~\&f match

(Zefram remarks that all(empty set) is true)

15 years agoFix tests for ~~ overloading
Rafael Garcia-Suarez [Tue, 24 Mar 2009 10:25:54 +0000]
Fix tests for ~~ overloading

I had the overload function wrong. ~~ overloading actually works.

15 years agoImplement "~~ $scalar" for numbers and strings
Rafael Garcia-Suarez [Mon, 23 Mar 2009 23:27:32 +0000]
Implement "~~ $scalar" for numbers and strings

Make documentation more precise about the different cases,
and add tests.

15 years agoMake ~~ qr// non-commutative
Rafael Garcia-Suarez [Mon, 23 Mar 2009 16:17:03 +0000]
Make ~~ qr// non-commutative

Pass elements in correct order to the sub-smart-match when comparing
two arrays. And more tests

15 years agoRefactoring work on '~~ @array' implementation
Rafael Garcia-Suarez [Mon, 23 Mar 2009 15:57:57 +0000]
Refactoring work on '~~ @array' implementation

(remove commutativity)

15 years agoRevert order of ~~ in definition of smart matching against arrays
Rafael Garcia-Suarez [Mon, 23 Mar 2009 15:55:52 +0000]
Revert order of ~~ in definition of smart matching against arrays

This allows to remove a special case (Hash ~~ Array).
We're also going to reuse ~~ in the Array ~~ Hash smart match
definition.

More tests will be needed.

15 years agoAdd more tests with an empty hash on the right of ~~
Rafael Garcia-Suarez [Mon, 23 Mar 2009 14:45:55 +0000]
Add more tests with an empty hash on the right of ~~

15 years agoSmart match when a hash is on the RHS
Rafael Garcia-Suarez [Wed, 18 Mar 2009 19:20:17 +0000]
Smart match when a hash is on the RHS

This implement the new semantics of C<~~ HASH> smart matching.
This also reveals that overloading actually does not work at all.
Add also tests for C<undef ~~ %hash>.

15 years agoReverse the order of operands for several "~~ hash" tests
Rafael Garcia-Suarez [Wed, 18 Mar 2009 19:12:47 +0000]
Reverse the order of operands for several "~~ hash" tests

This way the hash is on the right.
Some of those tests will need to become symmetrical later.

15 years agoRemove obsolete or redundant tests
Rafael Garcia-Suarez [Wed, 18 Mar 2009 18:29:34 +0000]
Remove obsolete or redundant tests

15 years agoDocument that empty sets don't smart-match
Rafael Garcia-Suarez [Tue, 17 Mar 2009 14:34:24 +0000]
Document that empty sets don't smart-match

(and implement this for the C<~~ coderef> form)

15 years agoTest that in "~~ sub", the sub is not called on empty hashes and arrays
Rafael Garcia-Suarez [Tue, 17 Mar 2009 14:11:05 +0000]
Test that in "~~ sub", the sub is not called on empty hashes and arrays

15 years agoFix a warnings test that was failing due to smart match dispatch changes
Rafael Garcia-Suarez [Tue, 17 Mar 2009 13:34:46 +0000]
Fix a warnings test that was failing due to smart match dispatch changes

15 years agoIterative smart match over keys/elements when a coderef is on the RHS
Rafael Garcia-Suarez [Tue, 17 Mar 2009 09:31:45 +0000]
Iterative smart match over keys/elements when a coderef is on the RHS

15 years agoMake smart matching ~~ undef dispatch only on the RHS
Rafael Garcia-Suarez [Fri, 13 Mar 2009 10:47:58 +0000]
Make smart matching ~~ undef dispatch only on the RHS

15 years agoSome tests reorgarnisation for ~~ against undef and objects
Rafael Garcia-Suarez [Thu, 12 Mar 2009 09:55:44 +0000]
Some tests reorgarnisation for ~~ against undef and objects

15 years agoRemove special case for coderefs with empty prototypes in smart match
Rafael Garcia-Suarez [Fri, 6 Mar 2009 14:57:21 +0000]
Remove special case for coderefs with empty prototypes in smart match

(plus, fix a metasyntactic name clash in the switch test)

15 years agoMake the "..." flip-flop operator work in scalar context in when()
Rafael Garcia-Suarez [Fri, 6 Mar 2009 14:01:34 +0000]
Make the "..." flip-flop operator work in scalar context in when()

15 years agoThe new smart match dispatch table for 5.10.1 onwards
Rafael Garcia-Suarez [Wed, 4 Mar 2009 20:14:10 +0000]
The new smart match dispatch table for 5.10.1 onwards

This /does/ break backwards compatibility. Changes to come are:

- ~~ is no longer commutative. Dispatch is now done based on the
  (run-time) type of the right side argument.
- The right side can be a ".." range, which will be treated specially,
  and no longer as a boolean operator (as in scalar context).
- when() should treat "..." as a boolean operator.
- Coderefs with an empty prototype are no longer special.
- Any ~~ Array distributes the smart match test.
- Comparing two coderefs is no longer a special case and does not
  perform referential equality. (Use \&foo ~~ "".\&bar for that.)

This patch also documents that ~~ won't work on objects that don't
overload it, except in the expression $object ~~ undef.

15 years agoRemove TODO tests about && in when() clauses
Rafael Garcia-Suarez [Wed, 4 Mar 2009 19:40:59 +0000]
Remove TODO tests about && in when() clauses

when() actually behaves as documented in perlsyn. It treats "E1 && E2"
expressions as booleans only if E1 is boolean, and if not, it does the
equivalent of the smart match expression "$_ ~~ (E1 && E2)". As long as
the ~~ operator is not expected to distribute boolean expressions found
on its right side, the current behaviour of perl is correct.

So, should we expect ~~ to distribute boolean expressions ? Given that
this would be rather complex to implement and test correctly with
complex expressions, given that we don't expect =~ to distribute either,
and given that when() is already quite smart regarding boolean
operators, I see no motivation for that feature.

This allows to close bug #50538 as "won't fix".

15 years agoDocument that when() treats specially most of the filetest operators
Rafael Garcia-Suarez [Wed, 4 Mar 2009 18:40:59 +0000]
Document that when() treats specially most of the filetest operators

15 years agoMake // behave like || in when clauses
Rafael Garcia-Suarez [Wed, 4 Mar 2009 18:33:31 +0000]
Make // behave like || in when clauses

(plus minor documentation updates about "when")

15 years agoUpdate Archive::Extract to 0.31_02
Jos I. Boumans [Wed, 4 Mar 2009 12:04:19 +0000]
Update Archive::Extract to 0.31_02

15 years agoAdd a failing test for stack corruption with MULTICALL
Alex Vandiver [Wed, 4 Mar 2009 07:54:26 +0000]
Add a failing test for stack corruption with MULTICALL

15 years agoUpdate Parse::CPAN::Meta to 0.04_01
Steffen Mueller [Tue, 3 Mar 2009 20:18:30 +0000]
Update Parse::CPAN::Meta to 0.04_01

15 years agoAdd Parse::CPAN::Meta to Porting/Maintainers.pl
Steffen Mueller [Tue, 3 Mar 2009 20:25:27 +0000]
Add Parse::CPAN::Meta to Porting/Maintainers.pl

15 years agoAdd contact info for Simon Wistow in Porting/Maintainers.pl
Steffen Mueller [Tue, 3 Mar 2009 20:19:59 +0000]
Add contact info for Simon Wistow in Porting/Maintainers.pl

15 years agoRemove the MANIFEST from Compress-Raw-Bzip2
Rafael Garcia-Suarez [Tue, 3 Mar 2009 15:57:06 +0000]
Remove the MANIFEST from Compress-Raw-Bzip2

(it wasn't the same in the CPAN distribution anyway)

15 years agoBump versions to 0.12 to sync with CPAN version
Rafael Garcia-Suarez [Mon, 2 Mar 2009 14:05:40 +0000]
Bump versions to 0.12 to sync with CPAN version

15 years agoIntegrate Parse::CPAN::Meta 0.04 into core
Steffen Mueller [Tue, 6 Jan 2009 18:49:01 +0000]
Integrate Parse::CPAN::Meta 0.04 into core

15 years agoUpdate CPANPLUS to 0.85_06
Jos I. Boumans [Fri, 27 Feb 2009 15:07:53 +0000]
Update CPANPLUS to 0.85_06

15 years agoBump versions to 0.10 for CPAN sync
Rafael Garcia-Suarez [Mon, 2 Mar 2009 09:44:54 +0000]
Bump versions to 0.10 for CPAN sync

15 years agoMake util.c compile on systems without getcwd()
Rafael Garcia-Suarez [Sun, 1 Mar 2009 23:13:34 +0000]
Make util.c compile on systems without getcwd()

15 years agoBetter gitignore file for Compress-Raw-Bzip2
Rafael Garcia-Suarez [Sun, 1 Mar 2009 23:05:03 +0000]
Better gitignore file for Compress-Raw-Bzip2

15 years agoAdd Compress-Raw-Bzip2, by Paul Marquess
Rafael Garcia-Suarez [Sun, 1 Mar 2009 22:37:53 +0000]
Add Compress-Raw-Bzip2, by Paul Marquess

15 years agoMark modules upgraded today as "upstream from CPAN"
Rafael Garcia-Suarez [Sun, 1 Mar 2009 22:17:35 +0000]
Mark modules upgraded today as "upstream from CPAN"

15 years agoUpdate CPANPLUS::Dist::Build to 0.08
Chris 'BinGOs' Williams [Sun, 1 Mar 2009 21:27:17 +0000]
Update CPANPLUS::Dist::Build to 0.08

15 years agoUpdate to Term::ANSIColor 2.00
Rafael Garcia-Suarez [Sun, 1 Mar 2009 16:30:21 +0000]
Update to Term::ANSIColor 2.00

15 years agoMake deprecate.pm work on VMS.
Craig A. Berry [Sat, 28 Feb 2009 17:47:56 +0000]
Make deprecate.pm work on VMS.

15 years agoHash::Util::FieldHash 1.03 leaks SVs which are used as object IDs stored in mg->mg_obj
Goro Fuji [Sat, 28 Feb 2009 10:21:38 +0000]
Hash::Util::FieldHash 1.03 leaks SVs which are used as object IDs stored in mg->mg_obj

15 years agoNAME section must fit on one line
Rafael Garcia-Suarez [Fri, 27 Feb 2009 14:29:07 +0000]
NAME section must fit on one line

15 years agorewrote deprecate test using File::Spec
Robin Barker [Wed, 25 Feb 2009 18:05:09 +0000]
rewrote deprecate test using File::Spec

15 years agoadd standard documentation to deprecate
Robin Barker [Thu, 26 Feb 2009 17:08:35 +0000]
add standard documentation to deprecate

15 years ago6c0ec79a2e9c445ac9999df2 failed to build from distclean - also tweak make_ext.pl
Nicholas Clark [Thu, 26 Feb 2009 22:31:39 +0000]
6c0ec79a2e9c445ac9999df2 failed to build from distclean - also tweak make_ext.pl

15 years agoRemove spurious TODO indicators
Rafael Garcia-Suarez [Thu, 26 Feb 2009 10:00:40 +0000]
Remove spurious TODO indicators

By definition, the first test in a series for testing a
leak should always succeed

15 years agoCorrect the paths for the nonxs_ext targets, following the ext flattening.
Nicholas Clark [Thu, 26 Feb 2009 08:34:36 +0000]
Correct the paths for the nonxs_ext targets, following the ext flattening.
(Static and dynamic each build a target in lib/auto, which was unchanged.
nonxs_ext use a target inside the extension's directory, which was moved, but
Makefile.SH had not been updated to reflect this.)

15 years agoRemove needless sed line added with 4318d5a0158916acc059aa8e6af84037cb7296b9.
Nicholas Clark [Thu, 26 Feb 2009 08:16:39 +0000]
Remove needless sed line added with 4318d5a0158916acc059aa8e6af84037cb7296b9.

15 years agoMake non-blocking mode work on Windows in IO::Socket::INET
Yasuhiro Matsumoto [Thu, 26 Feb 2009 08:09:33 +0000]
Make non-blocking mode work on Windows in IO::Socket::INET

CPAN RT bug #43573

15 years agoUpdate Filter::Util::Call to 1.08
Paul Marquess [Wed, 25 Feb 2009 23:20:05 +0000]
Update Filter::Util::Call to 1.08

15 years agofix typo in pod/perlperf.pod
David Mitchell [Wed, 25 Feb 2009 22:32:37 +0000]
fix typo in pod/perlperf.pod

15 years agoMakeMaker is first-come in Maintainers.pl.
Craig A. Berry [Wed, 25 Feb 2009 13:43:23 +0000]
MakeMaker is first-come in Maintainers.pl.

Says Schwern in Message-ID: <49A5290F.1060400@pobox.com>.

15 years agotest script for deprecate.pm
Robin Barker [Wed, 25 Feb 2009 08:41:52 +0000]
test script for deprecate.pm

First pass at tests for deprecate.pm.

Had to modify the module to provide an interface for testing - need to
provide fake install directories, but deprecated reads from
%Config::Config, which is read only.

15 years agoadd-package.pl fixup
Jos I. Boumans [Wed, 25 Feb 2009 08:31:02 +0000]
add-package.pl fixup

Sometimes, using find, add-package.pl will construct filenames with 2
consecutive slashes, ie:

 Package-Name/bin//script.pl

As this literal string isn't in the MANIFEST yet, it'll be added
erroneously. The below patch fixes this.

15 years agoremove explicit caller depth in deprecate.pm by searching caller stack for use/requir...
Robin Barker [Tue, 24 Feb 2009 09:57:23 +0000]
remove explicit caller depth in deprecate.pm by searching caller stack for use/require Module.pm

15 years agoFix line number test for scalar attribute handlers
Rafael Garcia-Suarez [Tue, 24 Feb 2009 09:21:09 +0000]
Fix line number test for scalar attribute handlers

Un-TODO the tests for bleadperl and above

15 years agoFix to installing non-xs ext's in priv lib
Jerry D. Hedden [Mon, 23 Feb 2009 17:37:16 +0000]
Fix to installing non-xs ext's in priv lib

15 years agoUpgrade to threads::shared 1.28
Jerry D. Hedden [Mon, 23 Feb 2009 19:43:46 +0000]
Upgrade to threads::shared 1.28

15 years agoUpgrade to threads 1.72
Jerry D. Hedden [Mon, 23 Feb 2009 20:11:01 +0000]
Upgrade to threads 1.72

15 years agoIn S_parse_body(), sv hasn't been needed since -P bit the dust.
Nicholas Clark [Mon, 23 Feb 2009 23:37:35 +0000]
In S_parse_body(), sv hasn't been needed since -P bit the dust.
(And probably should have been removed with change 32958,
2f3529075cdaceeaf619b8bd308885878ec9828a)

15 years agoMerge branch 'blead' of ssh://stevep@perl5.git.perl.org/gitroot/perl into blead
Steve Peters [Mon, 23 Feb 2009 19:57:38 +0000]
Merge branch 'blead' of ssh://stevep@perl5.git.perl.org/gitroot/perl into blead

15 years agofileno() was failing when passed a NULL. I'm guessing that we don't really need...
Steve Peters [Mon, 23 Feb 2009 19:25:33 +0000]
fileno() was failing when passed a NULL.  I'm guessing that we don't really need to increase the refcount on a NULL fd.

15 years agoFix unused variable warning
Jerry D. Hedden [Mon, 23 Feb 2009 16:04:07 +0000]
Fix unused variable warning

Attached patch fixes the following build warnings:

perlmini.c: In function `S_incpush':
perlmini.c:4358: warning: unused variable `addoldvers'

perl.c: In function `S_incpush':
perl.c:4358: warning: unused variable `addoldvers'

From 87b6dbeb18691177a2b2e8dc971da242ac831c03 Mon Sep 17 00:00:00 2001
From: Jerry D. Hedden <jdhedden@cpan.org>
Date: Mon, 23 Feb 2009 10:58:14 -0500
Subject: [PATCH] Fix unused variable warning

15 years agoRe-apply cd5cc49dbc0e5ee748252c2da8b435855908e6d2 (Module::Build on VMS, #42724)
Nicholas Clark [Mon, 23 Feb 2009 14:43:09 +0000]
Re-apply cd5cc49dbc0e5ee748252c2da8b435855908e6d2 (Module::Build on VMS, #42724)

15 years agoUpdate to Module::Build 0.31_04 (with a tweak to MBTest.pm)
Nicholas Clark [Mon, 23 Feb 2009 14:15:49 +0000]
Update to Module::Build 0.31_04 (with a tweak to MBTest.pm)

15 years agoTemporarily reverse out change cd5cc49dbc0e5ee748252c2da8b435855908e6d2.
Nicholas Clark [Mon, 23 Feb 2009 13:23:30 +0000]
Temporarily reverse out change cd5cc49dbc0e5ee748252c2da8b435855908e6d2.

15 years agoModule::Build Changes from 0.31012, which hadn't been updated.
Nicholas Clark [Mon, 23 Feb 2009 12:29:31 +0000]
Module::Build Changes from 0.31012, which hadn't been updated.

15 years agoAdd deprecate.pm. Deprecate shipping Switch.pm in the core distribution.
Nicholas Clark [Mon, 23 Feb 2009 10:27:08 +0000]
Add deprecate.pm. Deprecate shipping Switch.pm in the core distribution.

15 years agovms.c - Remove .DIR; in UNIX mode.
John Malmberg [Sun, 22 Feb 2009 18:23:29 +0000]
vms.c - Remove .DIR; in UNIX mode.

When doing a readdir() or glob() with DECC$FILENAME_UNIX_REPORT active,
the ".DIR" suffix needs to be removed for directories.

15 years agoVMS-specific nit in S_init_perllib after 2cace6acdfad87b2d298bff18b91b339c18fd1e3.
Craig A. Berry [Mon, 23 Feb 2009 03:14:17 +0000]
VMS-specific nit in S_init_perllib after 2cace6acdfad87b2d298bff18b91b339c18fd1e3.

15 years agotime_t is unsigned 32-bit on VMS; pick sGMTIME_max and sLOCALTIME_max accordingly.
Craig A. Berry [Mon, 23 Feb 2009 00:28:13 +0000]
time_t is unsigned 32-bit on VMS; pick sGMTIME_max and sLOCALTIME_max accordingly.

15 years agoFor S_incpush(), dir is never NULL, and len is always > 0.
Nicholas Clark [Sun, 22 Feb 2009 22:17:47 +0000]
For S_incpush(), dir is never NULL, and len is always > 0.

15 years agoOptimise S_incpush() by avoiding repeatedly copying libdir to subdir.
Nicholas Clark [Sun, 22 Feb 2009 21:27:18 +0000]
Optimise S_incpush() by avoiding repeatedly copying libdir to subdir.
Specifically, copy it once with newSVsv(), then pass libdir to
S_incpush_if_exists(), and if that creates a new SV, use newSVsv() there to
re-do the copy. Otherwise reset the length of the passed-in SV (which is
subdir), back to the length of libdir, effectively truncating it back to be
equal to libdir. This avoids repeated copying of the same bytes over the same
memory that already holds those bytes.

15 years agoIn S_incpush, move the declaration of subdir inside the only block that uses it.
Nicholas Clark [Sun, 22 Feb 2009 21:11:19 +0000]
In S_incpush, move the declaration of subdir inside the only block that uses it.

15 years agoTidy up incpush. Re-order the macros, and re-indent some code.
Nicholas Clark [Sun, 22 Feb 2009 21:00:03 +0000]
Tidy up incpush. Re-order the macros, and re-indent some code.

15 years agoMove unversioned directories in PERL_OTHERLIBDIRS later in @INC.
Nicholas Clark [Sun, 22 Feb 2009 20:45:04 +0000]
Move unversioned directories in PERL_OTHERLIBDIRS later in @INC.
Split INCPUSH_ADD_SUB_DIRS into INCPUSH_ADD_VERSIONED_SUB_DIRS and
INCPUSH_ADD_ARCHONLY_SUB_DIRS, and tweak the flags used for the two halves of
PERL_OTHERLIBDIRS so that .../$archname and .../ come after .../$old_version

15 years ago7fc731078209e8c26ed8bba477edfd90724c7146 missed one (addsubdirs || addoldvers).
Nicholas Clark [Sun, 22 Feb 2009 20:29:32 +0000]
7fc731078209e8c26ed8bba477edfd90724c7146 missed one (addsubdirs || addoldvers).

15 years agoIn S_init_perllib(), push the bare VENDORLIB_STEM onto @INC after its subdirs.
Nicholas Clark [Sun, 22 Feb 2009 19:27:48 +0000]
In S_init_perllib(), push the bare VENDORLIB_STEM onto @INC after its subdirs.

15 years agoIn S_init_perllib(), push the bare SITELIB_STEM onto @INC after its subdirs.
Nicholas Clark [Sun, 22 Feb 2009 19:27:08 +0000]
In S_init_perllib(), push the bare SITELIB_STEM onto @INC after its subdirs.

15 years agoIn S_incpush(), move push @INC, ".../$archname" if -d ".../$archname"; later.
Nicholas Clark [Sun, 22 Feb 2009 19:24:49 +0000]
In S_incpush(), move push @INC, ".../$archname" if -d ".../$archname"; later.

This effects this change:

--- Start       2009-02-21 14:04:40.000000000 +0100
+++ /dev/fd/63  2009-02-22 16:41:18.000000000 +0100
@@ -1,14 +1,14 @@
 /Users/nick/Sandpit/511v/hashbangI/5.11.0/darwin-thread-multi-64int-2level
 /Users/nick/Sandpit/511v/hashbangI/5.11.0
-/Users/nick/Sandpit/511v/hashbangI/darwin-thread-multi-64int-2level
 /Users/nick/Sandpit/511v/hashbangI/5.11.-2/darwin-thread-multi-64int-2level
 /Users/nick/Sandpit/511v/hashbangI/5.11.-2
+/Users/nick/Sandpit/511v/hashbangI/darwin-thread-multi-64int-2level
 /Users/nick/Sandpit/511v/hashbangI
 /Users/nick/Sandpit/511v/cliI/5.11.0/darwin-thread-multi-64int-2level
 /Users/nick/Sandpit/511v/cliI/5.11.0
-/Users/nick/Sandpit/511v/cliI/darwin-thread-multi-64int-2level
 /Users/nick/Sandpit/511v/cliI/5.11.-1/darwin-thread-multi-64int-2level
 /Users/nick/Sandpit/511v/cliI/5.11.-1
+/Users/nick/Sandpit/511v/cliI/darwin-thread-multi-64int-2level
 /Users/nick/Sandpit/511v/cliI
 /Users/nick/Sandpit/511v/perl5lib0/5.11.0/darwin-thread-multi-64int-2level
 /Users/nick/Sandpit/511v/perl5lib0/5.11.0

15 years agoIn S_init_perllib(), initialise perl5lib to NULL, to keep trace flow happy.
Nicholas Clark [Sun, 22 Feb 2009 17:37:26 +0000]
In S_init_perllib(), initialise perl5lib to NULL, to keep trace flow happy.

15 years agoIn S_incpush(), replace (addsubdirs || addoldvers) with a constant variable.
Nicholas Clark [Sun, 22 Feb 2009 17:24:04 +0000]
In S_incpush(), replace (addsubdirs || addoldvers) with a constant variable.

15 years agoIn S_init_perllib(), s can be const char*
Nicholas Clark [Sun, 22 Feb 2009 13:05:38 +0000]
In S_init_perllib(), s can be const char*

15 years agoIn S_init_perllib(), only call PerlEnv_getenv("PERL5LIB") once.
Nicholas Clark [Sun, 22 Feb 2009 12:55:49 +0000]
In S_init_perllib(), only call PerlEnv_getenv("PERL5LIB") once.

15 years agoUnwind the implicit loop in S_init_perllib(), by writing the code out longhand.
Nicholas Clark [Sun, 22 Feb 2009 12:34:29 +0000]
Unwind the implicit loop in S_init_perllib(), by writing the code out longhand.
Call it only once, remove the old_vers parameter, and all the related
conditional code.

15 years agoChange the value passed into S_init_perllib() to a boolean flag.
Nicholas Clark [Sat, 21 Feb 2009 17:23:46 +0000]
Change the value passed into S_init_perllib() to a boolean flag.

15 years agoNetWare has PRIVLIB_EXP as something other than a string constant. Cope with it.
Nicholas Clark [Sat, 21 Feb 2009 07:06:06 +0000]
NetWare has PRIVLIB_EXP as something other than a string constant. Cope with it.

15 years agoUse "system" gmtime in time64 on VMS so C<vmsish 'time'> pragma works.
Craig A. Berry [Sat, 21 Feb 2009 20:46:27 +0000]
Use "system" gmtime in time64 on VMS so C<vmsish 'time'> pragma works.

It's not really the system gmtime; it's the homegrown one in vms/vms.c
that checks hints and does shifting between UTC and local time when
requested.

15 years agoIn S_init_perllib(), correct errors in the MACOS_TRADITIONAL conditional code.
Nicholas Clark [Sat, 21 Feb 2009 18:04:08 +0000]
In S_init_perllib(), correct errors in the MACOS_TRADITIONAL conditional code.
(Errors that originated in a26c0e281cb6068a8d148933281d8186f1eb4206 and
50d61629dc7fe34f077b9f66c50287d839e06378)

15 years agoRe-apply changes 4adc95e616bac7ee and 200cbd6aa595a074 (RT #61492)
Nicholas Clark [Sat, 21 Feb 2009 08:41:14 +0000]
Re-apply changes 4adc95e616bac7ee and 200cbd6aa595a074 (RT #61492)
(ExtUtils::MM_Win32 should not generate "mt" command when CRT is statically
linked)(We'd failed to send this one upstream. Bad us.)

15 years agoUpgrade to ExtUtils::MakeMaker 6.49_01
Nicholas Clark [Sat, 21 Feb 2009 08:11:13 +0000]
Upgrade to ExtUtils::MakeMaker 6.49_01

15 years agoSkip time64 when there is no 64-bit integer type available.
Craig A. Berry [Sat, 21 Feb 2009 04:58:55 +0000]
Skip time64 when there is no 64-bit integer type available.

The y2038 code currently requires that a 64-bit integer type be available,
so the build falls down hard if we try to use it where there isn't one.  A
better fix might be to try NV for time_t and a 32-bit integer for year.

N.B.  We test for the definedness of Quad_t rather than HAS_QUAD since the
latter is explicitly undef'd in the case where 64-bit types are available
but are not being used for Perl's IV and UV.

15 years agoAvoid VMS long filename support on older systems that don't have it.
Craig A. Berry [Fri, 20 Feb 2009 23:43:08 +0000]
Avoid VMS long filename support on older systems that don't have it.

15 years agoMove redefinition of lstat above its first use in vms/vms.c.
Craig A. Berry [Fri, 20 Feb 2009 23:37:47 +0000]
Move redefinition of lstat above its first use in vms/vms.c.