This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
perl5.git
10 years agoperldelta: 5.12.3, not .2
Father Chrysostomos [Mon, 21 Mar 2011 21:11:22 +0000 (14:11 -0700)]
perldelta: 5.12.3, not .2

10 years agoperldelta: More reminders for the Known Problems
Father Chrysostomos [Mon, 21 Mar 2011 21:05:00 +0000 (14:05 -0700)]
perldelta: More reminders for the Known Problems

10 years agoperldelta: Unicode db not installed with Perl
Karl Williamson [Mon, 21 Mar 2011 18:58:05 +0000 (12:58 -0600)]
perldelta: Unicode db not installed with Perl

10 years agoperldelta: nit in to/from uvuni() section
Karl Williamson [Mon, 21 Mar 2011 18:41:37 +0000 (12:41 -0600)]
perldelta: nit in to/from uvuni() section

10 years agoperldelta: clarify /\p{}/ means Unicode
Karl Williamson [Mon, 21 Mar 2011 18:38:15 +0000 (12:38 -0600)]
perldelta: clarify /\p{}/ means Unicode

10 years agoperldelta: clarify /\p{}/i
Karl Williamson [Mon, 21 Mar 2011 18:37:44 +0000 (12:37 -0600)]
perldelta: clarify /\p{}/i

10 years agoperldelta: Fill in missing module name
Karl Williamson [Mon, 21 Mar 2011 18:28:59 +0000 (12:28 -0600)]
perldelta: Fill in missing module name

10 years agoperldelta: Add \b{ and \B{ as deprecations
Karl Williamson [Mon, 21 Mar 2011 18:28:03 +0000 (12:28 -0600)]
perldelta: Add \b{ and \B{ as deprecations

10 years agoperldelta: \cX is deprecated not removed
Karl Williamson [Mon, 21 Mar 2011 18:26:47 +0000 (12:26 -0600)]
perldelta: \cX is deprecated not removed

10 years agoperldelta: Re-order 2 paragraphs
Karl Williamson [Mon, 21 Mar 2011 17:35:57 +0000 (11:35 -0600)]
perldelta: Re-order 2 paragraphs

Note that one can turn off sub-categories of utf8 warnings when
operating on them, and introduce the fact that there are now these
categories in the paragraph before it instead of after it

10 years agonit in perldelta
Karl Williamson [Mon, 21 Mar 2011 17:31:19 +0000 (11:31 -0600)]
nit in perldelta

10 years agoPerl no longer comes with Unicode db files
Karl Williamson [Mon, 21 Mar 2011 17:26:13 +0000 (11:26 -0600)]
Perl no longer comes with Unicode db files

10 years agoAdapt corelist-perldelta to the current perldelta style
Florian Ragwitz [Mon, 21 Mar 2011 16:39:33 +0000 (17:39 +0100)]
Adapt corelist-perldelta to the current perldelta style

Apparently we used to do

  =item C<Module>
  ...

but these days all the perldeltas seem to prefer

  =item *
  C<Module> ...

instead.

10 years agoSub::Name is working again for some time now
Florian Ragwitz [Mon, 21 Mar 2011 16:26:19 +0000 (17:26 +0100)]
Sub::Name is working again for some time now

10 years agoperldelta: Re-add the upgrade notice
Father Chrysostomos [Mon, 21 Mar 2011 16:20:27 +0000 (09:20 -0700)]
perldelta: Re-add the upgrade notice

removed by 5076a39

10 years agoperldelta: Delete the Testing section
Father Chrysostomos [Sun, 20 Mar 2011 01:54:37 +0000 (18:54 -0700)]
perldelta: Delete the Testing section

perldelta is still above 4000 lines, but this should help. :-)

10 years agoperldelta: ‘(5.12.3)’ markers
Father Chrysostomos [Sun, 20 Mar 2011 01:51:24 +0000 (18:51 -0700)]
perldelta: ‘(5.12.3)’ markers

10 years agoperldelta: 5.12.2 markers, omissions, and tweaks
Father Chrysostomos [Sun, 20 Mar 2011 01:46:01 +0000 (18:46 -0700)]
perldelta: 5.12.2 markers, omissions, and tweaks

10 years agoperldelta: ‘(5.12.1)’ markers
Father Chrysostomos [Sat, 19 Mar 2011 19:15:22 +0000 (12:15 -0700)]
perldelta: ‘(5.12.1)’ markers

and other omissions that were discovered in the process

10 years agoperldelta: Incorporate mention of VS 2010 from 912c63e
Father Chrysostomos [Sat, 19 Mar 2011 18:03:28 +0000 (11:03 -0700)]
perldelta: Incorporate mention of VS 2010 from 912c63e

10 years agoperldelta: Retitle a subsection
Father Chrysostomos [Sat, 19 Mar 2011 16:58:51 +0000 (09:58 -0700)]
perldelta: Retitle a subsection

10 years agoperldelta: Add a reminder about Known Problems
Father Chrysostomos [Sat, 19 Mar 2011 16:58:13 +0000 (09:58 -0700)]
perldelta: Add a reminder about Known Problems

10 years agoperldelta: Tentatively finish sorting out bug fixes
Father Chrysostomos [Sat, 19 Mar 2011 16:57:49 +0000 (09:57 -0700)]
perldelta: Tentatively finish sorting out bug fixes

10 years agoperldelta: Rearrange the debugger bug section
Father Chrysostomos [Sat, 19 Mar 2011 01:11:17 +0000 (18:11 -0700)]
perldelta: Rearrange the debugger bug section

so that things that non-internals people will understand are at the
top of the list

10 years agoperldelta: Clean up the magic bug section
Father Chrysostomos [Sat, 19 Mar 2011 01:09:10 +0000 (18:09 -0700)]
perldelta: Clean up the magic bug section

10 years agoperldelta: Clean up the Unicode bugs section
Father Chrysostomos [Fri, 18 Mar 2011 15:53:53 +0000 (08:53 -0700)]
perldelta: Clean up the Unicode bugs section

10 years agoperldelta: Clean up the glob/stash/mro bug section
Father Chrysostomos [Fri, 18 Mar 2011 05:49:33 +0000 (22:49 -0700)]
perldelta: Clean up the glob/stash/mro bug section

10 years agoperldelta: Clean up the parsing bugs section
Father Chrysostomos [Fri, 18 Mar 2011 03:26:08 +0000 (20:26 -0700)]
perldelta: Clean up the parsing bugs section

This rewords and reformats some entries and also rearranges them
to put similar entries nearer each other.

10 years agoperldelta: Move the C API bugs section down
Father Chrysostomos [Fri, 18 Mar 2011 00:01:14 +0000 (17:01 -0700)]
perldelta: Move the C API bugs section down

10 years agoperldelta: Some clean-up of I/O and regexp bugs
Father Chrysostomos [Fri, 18 Mar 2011 00:00:24 +0000 (17:00 -0700)]
perldelta: Some clean-up of I/O and regexp bugs

10 years agoperldelta: First pass at sorting out bug fixes
Father Chrysostomos [Thu, 17 Mar 2011 16:45:51 +0000 (09:45 -0700)]
perldelta: First pass at sorting out bug fixes

10 years agoperldelta: Cull bug fix entries that do not apply
Father Chrysostomos [Thu, 17 Mar 2011 13:32:43 +0000 (06:32 -0700)]
perldelta: Cull bug fix entries that do not apply

These describe bugs that were introduced after 5.12.

10 years agoperldelta: Add the [perl #86328] entry from c094a73d1
Father Chrysostomos [Thu, 17 Mar 2011 13:14:52 +0000 (06:14 -0700)]
perldelta: Add the [perl #86328] entry from c094a73d1

10 years agoperldelta: Put ‘Selected’ in the module updates title
Father Chrysostomos [Thu, 17 Mar 2011 13:06:31 +0000 (06:06 -0700)]
perldelta: Put ‘Selected’ in the module updates title

10 years agoperldelta: Add the entry about text-mode source from 270ca14
Father Chrysostomos [Thu, 17 Mar 2011 12:42:01 +0000 (05:42 -0700)]
perldelta: Add the entry about text-mode source from 270ca14

10 years agoperldelta: Clean up and rearrange teh Internal Changes section
Father Chrysostomos [Wed, 16 Mar 2011 19:52:33 +0000 (12:52 -0700)]
perldelta: Clean up and rearrange teh Internal Changes section

10 years agoperldelta: Sort out the platform section
Father Chrysostomos [Wed, 16 Mar 2011 05:18:12 +0000 (22:18 -0700)]
perldelta: Sort out the platform section

10 years agoperldelta: Add a disgruntled reminder to the Testing section
Father Chrysostomos [Wed, 16 Mar 2011 01:06:14 +0000 (18:06 -0700)]
perldelta: Add a disgruntled reminder to the Testing section

10 years agoperldelta: Clean up the Configuration section
Father Chrysostomos [Wed, 16 Mar 2011 01:05:55 +0000 (18:05 -0700)]
perldelta: Clean up the Configuration section

10 years agoperldelta: Clean up and rearrange the utilities section
Father Chrysostomos [Wed, 16 Mar 2011 01:00:14 +0000 (18:00 -0700)]
perldelta: Clean up and rearrange the utilities section

10 years agoperldelta: Reword and rearrange the Diagnostics section
Father Chrysostomos [Tue, 15 Mar 2011 23:50:23 +0000 (16:50 -0700)]
perldelta: Reword and rearrange the Diagnostics section

10 years agoperldelta: Add the Windows fork change from 3aa0ac5a
Father Chrysostomos [Tue, 15 Mar 2011 21:02:42 +0000 (14:02 -0700)]
perldelta: Add the Windows fork change from 3aa0ac5a

10 years agoperldelta: Clean up the Documentation section
Father Chrysostomos [Tue, 15 Mar 2011 20:59:46 +0000 (13:59 -0700)]
perldelta: Clean up the Documentation section

Mentioning little changes to documentation here and there, such as
adding an example, rewording a paragraph, removing mention of a mod-
ule, etc., does little to help someone who is reading perldelta to
find out what documentation elsewhere might be worth reading or
re-reading.

So this commit removes quite a few perldelta entries for doc changes.

10 years agoperldelta: Remove the Dual-Life Modules section
Father Chrysostomos [Tue, 15 Mar 2011 19:21:14 +0000 (12:21 -0700)]
perldelta: Remove the Dual-Life Modules section

Dual life is more a property of the CPAN releases than it is of perl.
And this delta file is too long anyway.

10 years agoperldelta: Updated Modules clean-up
Father Chrysostomos [Tue, 15 Mar 2011 18:21:39 +0000 (11:21 -0700)]
perldelta: Updated Modules clean-up

10 years agoperldelta: Cull uninteresting upgrades from the module list
Father Chrysostomos [Tue, 15 Mar 2011 13:30:51 +0000 (06:30 -0700)]
perldelta: Cull uninteresting upgrades from the module list

A huge list of version numbers is not helpful to most people.

10 years agoperldelta: Move the new C APIs back up to Core Enhancements
Father Chrysostomos [Tue, 15 Mar 2011 05:35:51 +0000 (22:35 -0700)]
perldelta: Move the new C APIs back up to Core Enhancements

On second thought, they seem better there. :-)

10 years agoperldelta: Consistency tweaks
Father Chrysostomos [Tue, 15 Mar 2011 05:32:35 +0000 (22:32 -0700)]
perldelta: Consistency tweaks

Consistent spelling of ‘optimi[sz]ation’ and more consistent use
of dashes.

10 years agoperldelta: Rearrange the Performance Enhancements section
Father Chrysostomos [Tue, 15 Mar 2011 05:29:29 +0000 (22:29 -0700)]
perldelta: Rearrange the Performance Enhancements section

This moves similar items closer to each other. It is somewhat subject-
ive, but basically speed optimisations are first, and then memory sav-
ings. Slight enhancements come last.

10 years agoperldelta: First pass in cleaning up Performance Enhancements
Father Chrysostomos [Tue, 15 Mar 2011 05:20:31 +0000 (22:20 -0700)]
perldelta: First pass in cleaning up Performance Enhancements

10 years agoperldelta: Some rewording of entries in Deprecations
Father Chrysostomos [Tue, 15 Mar 2011 04:33:08 +0000 (21:33 -0700)]
perldelta: Some rewording of entries in Deprecations

10 years agoperldelta: Rearrange the Deprecations section a bit
Father Chrysostomos [Tue, 15 Mar 2011 04:27:29 +0000 (21:27 -0700)]
perldelta: Rearrange the Deprecations section a bit

10 years agoperldelta: Rearrange and categorise Incompatible Changes
Father Chrysostomos [Tue, 15 Mar 2011 01:19:01 +0000 (18:19 -0700)]
perldelta: Rearrange and categorise Incompatible Changes

10 years agoperldelta: Add link from Core Enhancements to New APIs
Father Chrysostomos [Mon, 14 Mar 2011 23:50:04 +0000 (16:50 -0700)]
perldelta: Add link from Core Enhancements to New APIs

10 years agoperldelta: Some Incompatible Changes clean-up
Father Chrysostomos [Mon, 14 Mar 2011 23:36:22 +0000 (16:36 -0700)]
perldelta: Some Incompatible Changes clean-up

This rewords some sections and removes others that no longer apply
(e.g., that describe reverts of earlier 5.13 changes).

The section about blead API compatibility can go, as it does not apply
to stable releases. Anyone following blead releases should have seen
it already.

10 years agoperldelta: Add the CPAN version
Father Chrysostomos [Mon, 14 Mar 2011 21:08:44 +0000 (14:08 -0700)]
perldelta: Add the CPAN version

10 years agoperldelta: Clean up the Security section
Father Chrysostomos [Mon, 14 Mar 2011 21:06:03 +0000 (14:06 -0700)]
perldelta: Clean up the Security section

10 years agoperldelta: Reword Core Enhancements
Father Chrysostomos [Mon, 14 Mar 2011 21:02:24 +0000 (14:02 -0700)]
perldelta: Reword Core Enhancements

and cut down the size of some entries

10 years agoperldelta: Incorporate 994ae75
Father Chrysostomos [Mon, 14 Mar 2011 20:15:55 +0000 (13:15 -0700)]
perldelta: Incorporate 994ae75

10 years agoperldelta: Rearrange Core Enhancements into subsections
Father Chrysostomos [Mon, 14 Mar 2011 20:14:54 +0000 (13:14 -0700)]
perldelta: Rearrange Core Enhancements into subsections

Also delete the perl -h thing, which is not noteworthy.

And capitalise headers consistently

Some of the items now under Regular Expressions are not specific to
regexps, but that should not matter, as the point is to create a
coherent document that people can read without getting lost.

And ‘Return value of C<delete $+{...}>’ has been ousted.

10 years agoperldelta: Oust the XS-specific core enhancements
Father Chrysostomos [Mon, 14 Mar 2011 20:14:27 +0000 (13:14 -0700)]
perldelta: Oust the XS-specific core enhancements

They can go in their own section as Core Enhancements is far too big.

10 years agoperldelta: Merge some entries in Core Enhancements
Father Chrysostomos [Mon, 14 Mar 2011 20:13:05 +0000 (13:13 -0700)]
perldelta: Merge some entries in Core Enhancements

10 years agoperldelta: Oust some entries from Core Enhancements
Father Chrysostomos [Mon, 14 Mar 2011 20:12:36 +0000 (13:12 -0700)]
perldelta: Oust some entries from Core Enhancements

Some of those do not belong in there.

10 years agoperldelta: Add changes from 5.13.0
Father Chrysostomos [Mon, 14 Mar 2011 20:12:04 +0000 (13:12 -0700)]
perldelta: Add changes from 5.13.0

perl 5.13.0’s delta file was never completed. These are the miss-
ing entries.

10 years agoCombine all perl513*deltas (as of commit d24f2be2)
Father Chrysostomos [Mon, 14 Mar 2011 20:11:20 +0000 (13:11 -0700)]
Combine all perl513*deltas (as of commit d24f2be2)

(Wow! This thing is almost 5000 lines!)

10 years agoReplace perldelta.pod with the template
Father Chrysostomos [Mon, 21 Mar 2011 16:13:07 +0000 (09:13 -0700)]
Replace perldelta.pod with the template

so I can apply my commits from the perl5140delta branch (and so I
don’t have to bother learning git properly :-)

10 years agosilence "unused variable" compiler warning
Karl Williamson [Mon, 21 Mar 2011 15:07:50 +0000 (09:07 -0600)]
silence "unused variable" compiler warning

10 years agoregcomp.c: enums are tighter in C++
Karl Williamson [Mon, 21 Mar 2011 15:05:55 +0000 (09:05 -0600)]
regcomp.c: enums are tighter in C++

I was using 0 for a generic non-interesting character, which works in C
but not C++.

10 years agopack test failures with long doubles on x86/gcc
David Mitchell [Mon, 21 Mar 2011 14:14:52 +0000 (14:14 +0000)]
pack test failures with long doubles on x86/gcc

[perl #86534]

When using 80-bit extended precision floats, gcc-generated code
can sometimes copy 10 bytes and sometimes 12. This can lead to 2 random
bytes something appearing in the output of pack('F') and pack('D').

Work around this by using sv_2nv() rather than SvNV()
(which expands to (SvNOK(sv) ? SvNVX(sv) : sv_2nv(sv) )

The basic issue is that the NV return value of a function is returned
in a floating point register, which is then written to memory as 10 bytes,
whereas a direct assignment, e.g. NV nv1 = nv2, is done by a 12-byte
memory copy.

However, when the sv_2nv() function is called as part of the ?:
expression, its returned value is written as *10* bytes to a temp location
on the stack, which is then copied as a *12* byte value to its final
destination, picking up 2 bytes of random data from the stack, which then
appears in the output of pack(), and ultimately leads to a test failure.

10 years agoTry reducing the risk of blind copy-paste errors
Florian Ragwitz [Sun, 20 Mar 2011 20:40:32 +0000 (21:40 +0100)]
Try reducing the risk of blind copy-paste errors

10 years agoAt this point, there's no tag for the next release yet
Florian Ragwitz [Sun, 20 Mar 2011 20:40:08 +0000 (21:40 +0100)]
At this point, there's no tag for the next release yet

10 years agopod/buildtoc doesn't update pod.lst
Florian Ragwitz [Sun, 20 Mar 2011 20:29:10 +0000 (21:29 +0100)]
pod/buildtoc doesn't update pod.lst

10 years agoCreate a perldelta for 5.14.0
Florian Ragwitz [Sun, 20 Mar 2011 19:57:59 +0000 (20:57 +0100)]
Create a perldelta for 5.14.0

10 years agoAdd the 5.13.11 epigraph
Florian Ragwitz [Sun, 20 Mar 2011 18:47:51 +0000 (19:47 +0100)]
Add the 5.13.11 epigraph

10 years agoTrim some trailing whitespace from epigraphs.pod
Florian Ragwitz [Sun, 20 Mar 2011 18:47:34 +0000 (19:47 +0100)]
Trim some trailing whitespace from epigraphs.pod

10 years agoregex .t's: Remove now passing TODOs
Karl Williamson [Sun, 20 Mar 2011 17:57:20 +0000 (11:57 -0600)]
regex .t's: Remove now passing TODOs

10 years agoregexec.c: Rmv special code no longer needed
Karl Williamson [Sun, 20 Mar 2011 17:56:11 +0000 (11:56 -0600)]
regexec.c: Rmv special code no longer needed

The trickyness has been resolved elsewhere

10 years agoregcharclass: Add tricky fold characters.
Karl Williamson [Sun, 20 Mar 2011 01:29:17 +0000 (19:29 -0600)]
regcharclass: Add tricky fold characters.

The tricky fold characters need to be expanded to include the ones
that map to the same ones as the original set.  This isn't because the
new ones have a length issue, it's that they get left out of comparisons
because of the special regnodes generated for the tricky ones.

10 years agoregcomp.c: Remove FOLDCHAR generation
Karl Williamson [Sun, 20 Mar 2011 17:32:11 +0000 (11:32 -0600)]
regcomp.c: Remove FOLDCHAR generation

ANYOFV handles multi-char folds in ANYOF nodes, and it turns
out it is a superset of what FOLDCHAR does, which never got fully
implemented in regexec.c, whereas ANYOFV is.  FOLDCHAR may be the better
way to go in the long-term, as it takes less space and is faster, but
this gives us the functionality today, with no extra work.

FOLDCHAR had been generated only when the character in question is a
literal in the input stream, and wasn't touched for the probably more
common use of \N{} or \x, which were fixed from not doing anything
special to using ANYOFV earlier in the 5.13 series, and it turns out
that the code that does it all is in a part of the code that gets
executed anyway, so that simply removing the special FOLDCHAR code
causes execution to drop down to this code.

I'm thinking at the moment that for 5.16, ANYOV should be removed in
favor of branches, using the technique of recursion that has recently
been added to \N{}.  That would enable easier trie generation and
simplify things in regexec and the optimizer.

10 years agoregcomp.c: Use mnemonic instead of hex value
Karl Williamson [Sun, 20 Mar 2011 17:26:14 +0000 (11:26 -0600)]
regcomp.c: Use mnemonic instead of hex value

10 years agoregcomp.c: Handle inverse tricky folds
Karl Williamson [Sun, 20 Mar 2011 17:23:14 +0000 (11:23 -0600)]
regcomp.c: Handle inverse tricky folds

The tricky folds have only worked one direction.  This handles the
other, when it sees something the tricky fold folds to it converts that
to the tricky fold op.

10 years agoregcomp.c: Move opening block brace
Karl Williamson [Sun, 20 Mar 2011 17:18:17 +0000 (11:18 -0600)]
regcomp.c: Move opening block brace

This is in preparation for future commits.  The declarations don't
depend on the two code lines.

10 years agoregcomp.c: Add special case for a Unicode char
Karl Williamson [Sun, 20 Mar 2011 17:14:28 +0000 (11:14 -0600)]
regcomp.c: Add special case for a Unicode char

This failed under some circumstances

10 years agoreg_namedseq: Restructure so doesn't duplicate code
Karl Williamson [Sun, 20 Mar 2011 16:25:17 +0000 (10:25 -0600)]
reg_namedseq: Restructure so doesn't duplicate code

This routine now calls reg() recursively after converting the parse
to something the rest of the code understands.  This eliminates
duplicated code, and allows for uniform treatment of code points, as
things were getting out of sync.  It also eliminates the restrction on
how many characters a named sequence can expand to.

toke now converts its input (which is in Unicode terms) to native on
EBCDIC platforms, so the rest of the code can can continue to ignore
that.

The restriction on the length of the number of characters a named
sequence is hereby removed, because reg() handles that.

10 years agoAdd depth parameter to reg_namedseq
Karl Williamson [Sun, 20 Mar 2011 16:20:14 +0000 (10:20 -0600)]
Add depth parameter to reg_namedseq

10 years agoregcomp.c: Add element to structure
Karl Williamson [Sun, 20 Mar 2011 15:59:26 +0000 (09:59 -0600)]
regcomp.c: Add element to structure

10 years agoutf8.h: Add #define
Karl Williamson [Sun, 20 Mar 2011 15:58:44 +0000 (09:58 -0600)]
utf8.h: Add #define

10 years agoSimplify a tiny bit of the release manager guide v5.13.11
Florian Ragwitz [Sun, 20 Mar 2011 15:16:49 +0000 (16:16 +0100)]
Simplify a tiny bit of the release manager guide

10 years agoAdd 5.13.11 to perlhist
Florian Ragwitz [Sun, 20 Mar 2011 15:14:43 +0000 (16:14 +0100)]
Add 5.13.11 to perlhist

10 years agoUpdate corelist for 5.13.11
Florian Ragwitz [Sun, 20 Mar 2011 14:40:16 +0000 (15:40 +0100)]
Update corelist for 5.13.11

10 years agoAvoid warnings in pat_advanced.t in the logic for skipping a test.
Nicholas Clark [Sun, 20 Mar 2011 14:37:39 +0000 (14:37 +0000)]
Avoid warnings in pat_advanced.t in the logic for skipping a test.

Also, change the skip added in 2feceb76bc07c897 to a todo skip.

10 years agoFill the Acknowledgements section in perldelta
Florian Ragwitz [Sun, 20 Mar 2011 14:31:55 +0000 (15:31 +0100)]
Fill the Acknowledgements section in perldelta

10 years agoBump the perl version for 5.13.11
Florian Ragwitz [Wed, 9 Mar 2011 22:41:46 +0000 (23:41 +0100)]
Bump the perl version for 5.13.11

10 years agoFix a tiny pod nit in perldelta
Florian Ragwitz [Sun, 20 Mar 2011 13:58:19 +0000 (14:58 +0100)]
Fix a tiny pod nit in perldelta

10 years agoRemove empty perldelta sections
Florian Ragwitz [Sun, 20 Mar 2011 13:55:29 +0000 (14:55 +0100)]
Remove empty perldelta sections

Only the Acknowledgements section still needs filling in.

10 years agoComplete the perldelta for 5.13.11
Florian Ragwitz [Sun, 20 Mar 2011 13:49:47 +0000 (14:49 +0100)]
Complete the perldelta for 5.13.11

10 years agoSkip test in pat_advanced_thr.t as it triggers an unrelated thread bug.
Nicholas Clark [Sun, 20 Mar 2011 11:19:23 +0000 (11:19 +0000)]
Skip test in pat_advanced_thr.t as it triggers an unrelated thread bug.

10 years agoMinimise the actions of the BEGIN block in B.pm
Nicholas Clark [Sun, 20 Mar 2011 10:25:21 +0000 (10:25 +0000)]
Minimise the actions of the BEGIN block in B.pm

Retain the call to XSLoader::load() at BEGIN time, as we want the constants
loaded before the compiler meets OPf_KIDS below, as the combination of having
the constant stay a Proxy Constant Subroutine and its value being inlined
saves a little over .5K

10 years agoIn B, use typeglob aliasing instead of subref to typeglob assignment.
Nicholas Clark [Sun, 20 Mar 2011 10:03:15 +0000 (10:03 +0000)]
In B, use typeglob aliasing instead of subref to typeglob assignment.

Typeglob aliasing saves just about 1.25K, because fewer internal structures are
created. In the general case the behaviour of the two differs, but as the
only package variables of these names are subroutines, and we are within our
own namespace, there is no difference here.

10 years ago#84774: local $_ calls STORE when $_ is aliased to a tied hash element
Jan Dubois [Fri, 18 Mar 2011 22:37:20 +0000 (15:37 -0700)]
#84774: local $_ calls STORE when $_ is aliased to a tied hash element

local($_) will now strip all magic from $_, so that it is always safe
to localize $_, regardless what kind of special (or tied) variable it
may have been aliased to.