Karl Williamson [Tue, 22 Jan 2013 20:34:54 +0000 (13:34 -0700)]
Add deprecation warning for literal PATWS under /x
This is explained in the perldelta changes in this commit. We plan to
migrate to Unicode's definition of white-space to ignore under /x. That
means we should raise a deprecation warning in the meantime if anyone
currently uses these characters in such a way as to have the meaning
changed when the migration is completed.
Karl Williamson [Tue, 22 Jan 2013 20:35:15 +0000 (13:35 -0700)]
regcomp.c: Add macro for generating deprecated warnings
This will be used in a future commit
Karl Williamson [Tue, 22 Jan 2013 20:32:01 +0000 (13:32 -0700)]
regcomp.c: Add comments; no code changes
Karl Williamson [Tue, 22 Jan 2013 20:30:01 +0000 (13:30 -0700)]
regcharclass.h: Add macro for non-ASCII PATWS
This will be used to deprecate uses of non-ASCII Pattern White Space
Karl Williamson [Tue, 22 Jan 2013 20:29:10 +0000 (13:29 -0700)]
/regen/regcharclass.pl: white-space only; no code changes
Karl Williamson [Tue, 22 Jan 2013 20:27:44 +0000 (13:27 -0700)]
regen/regcharclass.pl: Add capability
This allows one to generate macros that exclude just the ASCII range
Karl Williamson [Tue, 22 Jan 2013 20:23:06 +0000 (13:23 -0700)]
reg_mesg.t: Remove repetitious boiler plate
All the messages have boiler plate that can be removed and inserted by
the subroutine that does insertion anyway.
Karl Williamson [Tue, 22 Jan 2013 18:40:54 +0000 (11:40 -0700)]
Move t/lib/warnings/regcomp to t/re/reg_mesg.t
reg_mesg.t has better infrastructure to more easily add and maintain
these warnings.
Karl Williamson [Tue, 22 Jan 2013 18:16:59 +0000 (11:16 -0700)]
regcomp.c: Change warning category to just deprecated
The warnings for \b{ and \B{ were added in the 5.17 series; they are a
deprecation warning which should be turned off by that category. One
should not have to turn off regular regexp warnings as well to get rid
of these.
Karl Williamson [Tue, 22 Jan 2013 18:08:27 +0000 (11:08 -0700)]
reg_mesg.t: Add cpabilities; improve output
This adds the capability to have tests that each generate multiple
warnings, and it improves the flow so that if a test fails that make
moot subsequent tests, those tests are skipped.
Karl Williamson [Tue, 22 Jan 2013 02:32:09 +0000 (19:32 -0700)]
re/reg_mesg.t: White-space only; no code changes
Karl Williamson [Tue, 22 Jan 2013 02:25:01 +0000 (19:25 -0700)]
re/reg_mesg.t: Add tests for suppressing warnings
This automatically adds a test for each warning to verify that turning
off the warning category works.
Karl Williamson [Wed, 23 Jan 2013 18:36:01 +0000 (11:36 -0700)]
perlapi: Fix SvIOK_UV, SvUOK descriptions
Commit
b630937b8bf49e835d8976fc1036e68c79585b04 changed the text
of these two macros to how they currently work, but we don't
want to be tied to this behavior in the future.
New wording suggested by Darin McBride
Chris 'BinGOs' Williams [Wed, 23 Jan 2013 14:59:03 +0000 (14:59 +0000)]
Update Object-Accessor to CPAN version 0.46
[DELTA]
Changes for 0.46 Wed Jan 23 14:09:11 GMT 2013
=================================================
* Add deprecate usage to warn if module is loaded
from corelib. Object::Accessor is leading core
with v5.20.0, but will remain available from
CPAN.
Chris 'BinGOs' Williams [Wed, 23 Jan 2013 13:29:20 +0000 (13:29 +0000)]
Update Log-Message to CPAN version 0.06
[DELTA]
Changes for 0.06 Wed Jan 23 12:21:48 GMT 2013
=================================================
* Add deprecate usage to warn if module is loaded
from corelib. Log::Message is leaving core with
v5.20.0, but will still be available from CPAN.
Chris 'BinGOs' Williams [Wed, 23 Jan 2013 11:25:01 +0000 (11:25 +0000)]
Update Log-Message-Simple to CPAN version 0.10
[DELTA]
Changes for 0.10 Wed Jan 23 08:10:42 GMT 2013
================================================
* Add deprecate usage to warn if module is loaded
from corelib. Log::Message::Simple is leaving
core with v5.20.0, but will still be available
on CPAN.
Brad Gilbert [Wed, 26 Dec 2012 21:00:06 +0000 (15:00 -0600)]
Clean up indenting on t/win32/fs.t
Previous commit removed the need for indents.
Brad Gilbert [Wed, 26 Dec 2012 20:46:49 +0000 (14:46 -0600)]
Remove dead code from t/win32/fs.t
tempfile() from t/test.pl already handles checking for existing files
and removing them at the end of the test.
This test also doesn't need Config loaded
James E Keenan [Wed, 23 Jan 2013 03:25:04 +0000 (22:25 -0500)]
Convert to use of t/test.pl
Patch supplied by Marcel Gruenauer.
For: RT #116511
James E Keenan [Wed, 23 Jan 2013 00:32:33 +0000 (19:32 -0500)]
Note why switchF1.t and switchp.t do not use test.pl.
Patches submitted by Marcel Gruenauer.
For RT #116507 and 116517.
Chris 'BinGOs' Williams [Tue, 22 Jan 2013 12:45:07 +0000 (12:45 +0000)]
Update Term-UI to CPAN version 0.34
[DELTA]
Changes for 0.34 Tue Jan 22 12:19:40 GMT 2013
=====================================================
* deprecate usage added if the library is found in
corelib location. Term::UI is leaving core with
v5.20.0
Tony Cook [Tue, 22 Jan 2013 12:27:13 +0000 (23:27 +1100)]
syntax check a directory: set the test process locale too
Ricardo Signes [Mon, 21 Jan 2013 22:16:03 +0000 (17:16 -0500)]
move B-Lint and File-CheckTree to ./cpan
Ricardo Signes [Mon, 21 Jan 2013 16:34:42 +0000 (11:34 -0500)]
add deprecate.pm usage to CPANPLUS
Chris 'BinGOs' Williams [Mon, 21 Jan 2013 20:43:45 +0000 (20:43 +0000)]
Update CPANPLUS to CPAN version 0.9134
[DELTA]
Changes for 0.9134 Mon Jan 21 20:00:03 2013
================================================
* $VERSION all library files
Chris 'BinGOs' Williams [Mon, 21 Jan 2013 12:09:18 +0000 (12:09 +0000)]
The MAPping for Test-Harness is not required since 3.26
Aaron Crane [Mon, 21 Jan 2013 11:20:54 +0000 (11:20 +0000)]
Mark 5.17.8 as done in release_schedule.pod
Chris 'BinGOs' Williams [Sun, 20 Jan 2013 23:17:31 +0000 (23:17 +0000)]
Module-CoreList-2.80 has been released to CPAN
Chris 'BinGOs' Williams [Sun, 20 Jan 2013 22:30:11 +0000 (22:30 +0000)]
Actually update Test-Harness to CPAN version 3.26
[DELTA]
Finishes off what was started with
83044e8d6
Aaron Crane [Sun, 20 Jan 2013 19:28:58 +0000 (19:28 +0000)]
Bump the Perl version in various places for 5.17.9
Aaron Crane [Sun, 20 Jan 2013 19:14:59 +0000 (19:14 +0000)]
Make a new perldelta for 5.17.9-to-be
Aaron Crane [Sun, 20 Jan 2013 19:11:26 +0000 (19:11 +0000)]
Add the 5.17.8 epigraph to epigraphs.pod
Aaron Crane [Sun, 20 Jan 2013 15:49:49 +0000 (15:49 +0000)]
Add 5.17.8 to perlhist
Aaron Crane [Sun, 20 Jan 2013 15:46:34 +0000 (15:46 +0000)]
Update Module::CoreList for 5.17.8
Aaron Crane [Sun, 20 Jan 2013 18:22:21 +0000 (18:22 +0000)]
Last-minute "known issue" in perldelta
Aaron Crane [Sat, 19 Jan 2013 20:32:57 +0000 (20:32 +0000)]
Update perldelta for 5.17.8
Including the generated Acknowledgements section.
Thanks to Smylers++ for reviewing an earlier draft.
Karl Williamson [Thu, 27 Dec 2012 22:21:47 +0000 (15:21 -0700)]
Add av_top() synonym for av_len()
av_len() is misleadingly named.
Karl Williamson [Sun, 20 Jan 2013 04:37:36 +0000 (21:37 -0700)]
Replace XXX by correct bug number
Commit
4d68ffa0f7f345bc1ae6751744518ba4bc3859bd failed to get the
correct bug number in a comment
Karl Williamson [Sun, 20 Jan 2013 03:37:37 +0000 (20:37 -0700)]
perlre: Fix typo
"<" isn't a metacharacter, therefore "\<" doesn't change its meaning.
"[" is a metacharacter, therefore "\[" does change its meaning.
Karl Williamson [Sun, 20 Jan 2013 03:29:42 +0000 (20:29 -0700)]
Deprecate certain rare uses of backslashes within regexes
There are three pairs of characters that Perl recognizes as
metacharacters in regular expression patterns: {}, [], and (). These
can be used as well to delimit patterns, as in:
m{foo}
s(foo)(bar)
Since they are metacharacters, they have special meaning to regular
expression patterns, and it turns out that you can't turn off that
special meaning by the normal means of preceding them with a backslash,
if you use them, paired, within a pattern delimitted by them. For
example, in
m{foo\{1,3\}}
the backslashes do not change the behavior, and this matches "f", "o"
followed by one to three more occurrences of "o".
Usages like this, where they are interpreted as metacharacters, are
exceedingly rare; we think there are none, for example, in all of CPAN.
Hence, this deprecation should affect very little code. It does give
notice, however, that any such code needs to change, which will in turn
allow us to change the behavior in future Perl versions so that the
backslashes do have an effect, and without fear that we are silently
breaking any existing code.
=head1 Performance Enhancements
Karl Williamson [Wed, 16 Jan 2013 20:02:53 +0000 (13:02 -0700)]
toke.c: White-space alignment only
Karl Williamson [Mon, 14 Jan 2013 17:56:06 +0000 (10:56 -0700)]
Revert "Deprecate literal unescaped "{" in regexes."
This reverts commit
2a53d3314d380af5ab5283758219417c6dfa36e9.
Not the entire commit was reverted, but the deprecation message is
gone. This caused too many problems. See thread
http://www.nntp.perl.org/group/perl.perl5.porters/2012/11/msg195425.html
(which lists previous threads).
Karl Williamson [Sat, 19 Jan 2013 04:58:10 +0000 (21:58 -0700)]
Extend strictness for qr/(?[ \N{} ])/
This recently added regex syntax imposes stricter rules on parsing than
normal. However, this did not include parsing \N{} constructs that
occur within it. This commit does that, making fatal the warnings that
come from \N{}
I will add to perldiag the newly added messages along with the others
for (?[ ]) before 5.18 ships
Chris 'BinGOs' Williams [Sat, 19 Jan 2013 11:35:27 +0000 (11:35 +0000)]
constant-1.25 has been released to CPAN
James E Keenan [Sat, 19 Jan 2013 02:15:04 +0000 (21:15 -0500)]
Remove checks for Perl 5.6; adjust tests; bump version number.
Patch contributed by Sébastien Aperghis-Tramoni
For: RT #114770.
# Please enter the commit message for your changes. Lines starting
# with '#' will be ignored, and an empty message aborts the commit.
# On branch blead
# Your branch is ahead of 'origin/blead' by 1 commit.
#
# Changes to be committed:
# (use "git reset HEAD^1 <file>..." to unstage)
#
# modified: dist/constant/lib/constant.pm
# modified: dist/constant/t/constant.t
#
James E Keenan [Tue, 8 Jan 2013 01:55:24 +0000 (20:55 -0500)]
Skip t/io/eintr.t on older Darwins.
Identify the OS version by capturing the first two parts of the M.m.p version
number.
For RT #116262
Chris 'BinGOs' Williams [Fri, 18 Jan 2013 16:54:50 +0000 (16:54 +0000)]
Update Socket to CPAN version 2.009
[DELTA]
2013/01/18
2.009 CHANGES:
* Fix building in core by skipping check_for() as it doesn't work
there (RT82760)
2012/12/27
2.008 CHANGES:
* Fix uninitialised memory read (RT82119)
2012/12/16
2.007 CHANGES:
* Test %Config keys for definedness, not mere existence (RT79854)
* Fix missing argument in sprintf in Socket.xs (from perl.git
5d6dfea82e1c4b6, RT82007)
Aaron Crane [Fri, 18 Jan 2013 15:13:34 +0000 (15:13 +0000)]
Update Pod-LaTeX to CPAN version 0.61
Upstream changes:
- Update site vs perl install for perl v5.12 and above
- Issue a deprecation warning if the module is being used in perl 5.17.0 and above.
This module will be removed from perl Core for perl 5.20. (Ricardo Signes)
It will still be available on CPAN.
Daniel Dragan [Thu, 17 Jan 2013 23:38:07 +0000 (18:38 -0500)]
rearrange initializations in S_intuit_more for better code gen
MS VC 2003 in my experiance does not reorder var initializations with
constants to reduce their liveness. This commit attempts to defer
initialization until right before the var is first used. I can't explain
exactly why less instructions or shorter addressing happened since I didnt
record what the asm looked like before. On VC 2003 -O1 -GL, S_intuit_more
was previously 0x4B5 bytes of 32 bit machine code long, after it this
change it is 0x4A3 bytes long. These changes should have no user visible
effect.
The scope of the vars was not reduced to avoid large indentation changes
which would be required under C89 and Perl code formatting policy.
Aaron Crane [Fri, 18 Jan 2013 14:24:13 +0000 (14:24 +0000)]
todo: revisit the super-linear cache
Suggested by Yves.
Ricardo Signes [Wed, 16 Jan 2013 23:40:44 +0000 (18:40 -0500)]
further refinement to EISDIR tests from `perl dir`
1. actually use the EISDIR string, rather than getting it and
not using it; this was a refactoring screw-up
2. don't hardcode the Win32 EACCES error, either, use the same
"$!" mechanism
Ricardo Signes [Wed, 16 Jan 2013 21:17:13 +0000 (16:17 -0500)]
use non-dev version for Safe
Ricardo Signes [Wed, 16 Jan 2013 21:16:39 +0000 (16:16 -0500)]
use non-dev version for IO
Ricardo Signes [Wed, 16 Jan 2013 21:14:51 +0000 (16:14 -0500)]
use non-dev version for I18N::LangTags::List
...and sync with I18N::LangTags
Ricardo Signes [Wed, 16 Jan 2013 21:10:55 +0000 (16:10 -0500)]
bump I18N::Langinfo to a non-dev version
Ricardo Signes [Fri, 18 Jan 2013 03:21:12 +0000 (22:21 -0500)]
tweak the way we mention if.pm
It should be clear even to readers of the documentation in text
format that "if" is a module and not the "if" flow control.
Andy Dougherty [Thu, 17 Jan 2013 15:34:33 +0000 (10:34 -0500)]
Added asserts() to check the arguments to S_copy_little_tm_to_big_TM.
The original version just zeroed dest if src == NULL, but that code path
was never used. (gcc -Os inlined the function and optimized the test
away anyway.)
Daniel Dragan [Thu, 17 Jan 2013 07:52:45 +0000 (02:52 -0500)]
remove an useless null check in S_copy_little_tm_to_big_TM
All callers of S_copy_little_tm_to_big_TM pass a true variable src.
Checking for null is pointless. On Win32 32 bit VC 2003 -O1 -GL,
perl517.dll's .text section went from 0xCO13F to 0xC012F bytes long. It can
be argued that the compiler should have figured this out on its own, but VC
for whatever reason didn't. Also pretty the indenting and align the
assignments. The null check blames to commit
806a119aef .
Karl Williamson [Wed, 16 Jan 2013 22:05:04 +0000 (15:05 -0700)]
utf8.c: Reword a warning message
This follows the suggestion by Aristotle Pagaltzis.
Karl Williamson [Wed, 16 Jan 2013 22:03:51 +0000 (15:03 -0700)]
perldiag: Move an out-of-alphabetical_order entry
Chris 'BinGOs' Williams [Wed, 16 Jan 2013 21:57:40 +0000 (21:57 +0000)]
Update File-Path to CPAN version 2.09
[DELTA]
2.09 2013-01-13 21:09:12 UTC
- merge two bugfixes from blead
see https://github.com/dland/File-Path/pull/1
Chris 'BinGOs' Williams [Wed, 16 Jan 2013 20:38:54 +0000 (20:38 +0000)]
Update Test to CPAN version 1.26
[DELTA]
1.26 2013-01-16 Jesse R Vincent <jesse@bestpractical.com>
* Bump version to a stable number. No functional changes
Karl Williamson [Wed, 16 Jan 2013 19:49:50 +0000 (12:49 -0700)]
utf8.h: Add comments
This also reorders one #define to be closer to a related one.
Andy Dougherty [Wed, 16 Jan 2013 17:30:43 +0000 (12:30 -0500)]
Avoid wraparound when casting unsigned size_t to signed ssize_t.
Practically, this only affects a perl compiled with 64-bit IVs on a 32-bit
system. In that instance a value of count >= 2**31 would turn negative
when cast to (ssize_t).
bulk88 (via RT) [Wed, 16 Jan 2013 05:54:30 +0000 (21:54 -0800)]
PATCH: [perl 116411]: code comment for commit
518a5310cc "Silence a MSVC++-specific warning"
There is no written investigation to google up for
the record. I don't want to forget that the #ifdef is benign and
accidentally reinvestigate it in the future. .text section of
perl517.dll was 0xC013F before and after the commit. No change.
-----------------------------------------------------------------
Ricardo Signes [Wed, 16 Jan 2013 13:56:57 +0000 (08:56 -0500)]
in testing EISDIR string, check known value first
It was not enough to ensure the English value, as some platforms
use a different string entirely. Rather than goof around with
figuring them out, just get the known value by making an EISDIR
and stringifying it, then compare to that.
Ricardo Signes [Wed, 16 Jan 2013 15:05:47 +0000 (10:05 -0500)]
Revert "Upgrade Socket to CPAN version 2.008"
This reverts commit
90ae46a10a094e68135385e525c78962d6572da3.
Socket 2.008 fails to build on Linux in core, and possibly other
platforms as well.
Ricardo Signes [Wed, 16 Jan 2013 14:36:47 +0000 (09:36 -0500)]
Upgrade Socket to CPAN version 2.008
2012/12/27
2.008 CHANGES:
* Fix uninitialised memory read (RT82119)
2012/12/16
2.007 CHANGES:
* Test %Config keys for definedness, not mere existence (RT79854)
* Fix missing argument in sprintf in Socket.xs (from perl.git
5d6dfea82e1c4b6, RT82007)
Aaron Crane [Wed, 16 Jan 2013 11:47:19 +0000 (11:47 +0000)]
perlfunc.pod: fix incorrect short description of `length`
Thanks to ranguard for pointing out the report at
https://github.com/perlorg/perlweb/issues/65
Chris 'BinGOs' Williams [Wed, 16 Jan 2013 08:21:49 +0000 (08:21 +0000)]
Update Test-Harness to CPAN version 3.26
[DELTA]
3.26 2013-01-16
- Renamed env.opts.t to env_opts.t (for VMS)
- Skipped some TAP::Formatter::HTML tests due to this bug: #82738
Nicholas Clark [Thu, 3 Jan 2013 15:32:26 +0000 (16:32 +0100)]
Remove the Rhapsody port.
Rhapsody was an Apple OS that later evolved into Darwin and Mac OS X. It was
initially only released to developers, but later became Mac OS X Server, with
releases in 1999 and 2000. It was obsoleted by Mac OS X 10.0, released in
March 2001.
Steffen Mueller [Wed, 16 Jan 2013 06:40:59 +0000 (07:40 +0100)]
New PathTools CPAN release 3.40
Steffen Mueller [Wed, 16 Jan 2013 06:40:23 +0000 (07:40 +0100)]
Cwd/File::Spec version bump to 3.40
Ricardo Signes [Tue, 15 Jan 2013 23:59:20 +0000 (18:59 -0500)]
avoid having to worry whether the test runs in a locale
This is a time-honored tradition from such places as t/op. Tony
Cook alerted me to failures caused by this test on machines smoking
in non-English locales.
Chris 'BinGOs' Williams [Tue, 15 Jan 2013 10:32:29 +0000 (10:32 +0000)]
Update Digest-SHA to CPAN version 5.81
[DELTA]
5.81 Mon Jan 14 05:17:08 MST 2013
- corrected load subroutine (SHA.pm) to prevent double-free
-- Bug #82655: Security issue - segfault
-- thanks to Victor Efimov and Nicholas Clark
for technical expertise and suggestions
Steve Hay [Tue, 15 Jan 2013 09:04:20 +0000 (09:04 +0000)]
Silence a MSVC++-specific warning
("function declared with __declspec(noreturn) has non-void return type" /
"function declared with __declspec(noreturn) has a return statement".)
Karl Williamson [Mon, 14 Jan 2013 19:12:50 +0000 (12:12 -0700)]
handy.h: Fix isIDCONT_utf8()
It was handling above-Latin1 code points as IDstarts instead of
continues.
Steve Hay [Mon, 14 Jan 2013 17:59:55 +0000 (17:59 +0000)]
Silence a couple of warnings
("'initializing' : conversion from 'I32' to 'U8', possible loss of data"
and "formal parameter n different from declaration".)
Karl Williamson [Sun, 13 Jan 2013 20:33:22 +0000 (13:33 -0700)]
Add warnings for "\08", /\017/
This was discussed in thread
http://perl.markmail.org/thread/avtzvtpzemvg2ki2
but I never got around to this portion of the consensus, until now.
I did a cpan grep
http://grep.cpan.me/?q=%28^|[^\\]%29\\[0-7]{1%2C2}[8-9]&page=1
and eyeballing the results, saw three cases where this warning might
show up; one of which was for EBCDIC. The others looked to be false
positives, such as in .css files.
Jerry D. Hedden [Fri, 11 Jan 2013 15:53:42 +0000 (10:53 -0500)]
bump threads::shared version 1.43
To sync with the forthcoming CPAN release.
Karl Williamson [Mon, 14 Jan 2013 00:22:34 +0000 (17:22 -0700)]
regcomp.c: Silence compiler warning
Some compilers say this can be used uninitialized. I don't believe them,
but silencing them will save effort in the long run.
Aaron Crane [Sun, 13 Jan 2013 10:08:02 +0000 (10:08 +0000)]
Make Data::Dumper XS ignore Freezer return value
Fixes RT #116364
Karl Williamson [Sun, 13 Jan 2013 20:31:38 +0000 (13:31 -0700)]
regcomp.c: Stop potential leaks
These two ref counts were not at the same level as their corresponding
dec.
Karl Williamson [Sun, 13 Jan 2013 20:28:09 +0000 (13:28 -0700)]
regcomp.c: Add comments; align ternary operator for clarity
Karl Williamson [Sun, 13 Jan 2013 18:46:27 +0000 (11:46 -0700)]
reg_mesg.t: Add, revise test names
Karl Williamson [Sun, 13 Jan 2013 18:11:20 +0000 (11:11 -0700)]
embed.fnc: Clarify that varargs suppresses embed.h
Macro don't have variable numbers of args, hence the entry in embed.h is
suppressed.
Daniel Dragan [Sun, 13 Jan 2013 02:02:33 +0000 (21:02 -0500)]
better POD for unpackstring
Someone asked me how to use unpackstring() since they couldn't figure it
out from the docs, so I added some things.
Karl Williamson [Sun, 30 Dec 2012 03:18:09 +0000 (20:18 -0700)]
Create deprecated fncs to replace to-be-removed macros
These macros should not be used as they are prone to misuse. There are
no occurrences of them in CPAN. The single use of either of them in
core has recently been removed (commit
8d40577bdbdfa85ed3293f84bf26a313b1b92f55), because it was a misuse.
Instead code should use isIDFIRST_lazy_if or isWORDCHAR_lazy_if
(isALNUM_lazy_if is also available, but can be confused with the Posix
alnum, which it doesn't mean).
Karl Williamson [Sun, 30 Dec 2012 05:02:49 +0000 (22:02 -0700)]
porting/args_assert.t: Needs to look in inline.h too
inline.h is a special header file that contains C functions, and hence
perhaps PERL_ARGS_ASSERTS.
Ricardo Signes [Sat, 12 Jan 2013 12:12:34 +0000 (07:12 -0500)]
expect a different error for `perl dir` on Win32
We've known that this is how Win32 behaves, as it was documented in
the ticket for which this is a fix. I don't think it's worth the
bother of ensuring we get EISDIR, as long as we don't just exit
silently!
Ricardo Signes [Mon, 7 Jan 2013 20:00:08 +0000 (15:00 -0500)]
perldelta for #61362, croak on `perl directory`
James E Keenan [Sun, 6 Jan 2013 03:25:29 +0000 (22:25 -0500)]
Add a test to detect error when attempting to syntax-check a directory.
For: RT #61362
Ricardo Signes [Sun, 6 Jan 2013 01:30:48 +0000 (20:30 -0500)]
croak on an attempt to run a directory as a script
How many times have I meant to run "perl -I lib myprog" but instead
run "perl lib myprog" only to exit 0 because that's what perl does
when you try to run a directory as a script (at least on unix)? Many.
perl should croak when instructed to execute a directory.
[perl #61362] suggests it already does so on Win32. Now it does it
everywhere. Tests not yet written.
Craig A. Berry [Sat, 12 Jan 2013 22:30:17 +0000 (16:30 -0600)]
Simplify ExtUtils::Install::_can_write_dir on VMS.
This reverts the code (but not test) portions of
3d55b451d9544fb.
The old solution of using catdir rather than a combination of
catdir and catpath works with and without extended filespecs
enabled, so there's no reason to maintain two versions, and thus
no reason to have all the boilerplate feature checking code at
the beginning of the module.
David Mitchell [Sat, 12 Jan 2013 10:26:15 +0000 (10:26 +0000)]
include SvREADONLY() in SvIMMORTAL() test
SvIMMORTAL() is currently defined as
((sv)==&PL_sv_undef || (sv)==&PL_sv_yes
|| (sv)==&PL_sv_no || (sv)==&PL_sv_placeholder)
Which is relatively slow. Some places do this:
if (SvREADONLY(sv) && SvIMMORTAL(sv)) ...
which quickly rejects most times.
This commit simply moves the SvREADONLY test into the SvIMMORTAL macro
so that *all* uses benefit from this speedup.
Karl Williamson [Sat, 12 Jan 2013 15:48:01 +0000 (08:48 -0700)]
perlapi: Clarify av_pop
This notes that the caller now has control of a reference count of the
returned SV.
Wording mostly suggested by Paul Evans
Karl Williamson [Fri, 11 Jan 2013 21:29:29 +0000 (14:29 -0700)]
Allow slop on a few locale tests
Four recently introduced tests in locale.t fail for two locales of all
the ones that get tested in our smoke farm. I investigated the failures
and it looks to me like the problem in each case is that the locale
definition is defective.
The tests were added because of finding and fixing a bug in Perl, so I
don't want to remove them. Instead these 4 tests will be marked as TODO
if at least 95% of locales pass on any given machine.
This works for our current smokers.
Aaron Crane [Sat, 12 Jan 2013 15:34:43 +0000 (15:34 +0000)]
perlre: fix typo
Nicholas Clark [Sat, 12 Jan 2013 10:45:26 +0000 (11:45 +0100)]
In S_regatom(), set flags to 0 at the start.
Without this, flags will be read uninitialised for the execution path that
ends up with "Internal urp". (Although it's not clear that that error message
is ever reachable.)
The bug was introduced in perl 5.000, which added (?#) comments and (?m)
style embedding flags. It's not present in alpha 9.
Karl Williamson [Fri, 11 Jan 2013 18:51:18 +0000 (11:51 -0700)]
Merge branch 'sets' into blead
This branch adds the experimental regular expression (?[ ]) feature.
Details are in the individual commit messages.