This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Karl Williamson [Sun, 3 May 2015 15:43:23 +0000 (09:43 -0600)]
locale.c: White-space, comment only
Add a comment, indent some nested #if's
Karl Williamson [Thu, 28 May 2015 02:35:43 +0000 (20:35 -0600)]
mktables: Under -annotate use dotted circle for combining marks
Previously the NBSP was used, but the dotted circle is what Unicode uses
in their books, and it looks better. Note that -annotate is not a
default option for mktables.
Karl Williamson [Sat, 30 May 2015 16:29:12 +0000 (10:29 -0600)]
perldelta_template: Add L<> around 2 links
Karl Williamson [Tue, 2 Jun 2015 03:44:46 +0000 (21:44 -0600)]
PATCH: [perl #125291] perlreref misstatement
This commit fixes this, and makes a few other clarifications to the
document. Now there is only one definition for white space, so no need
to mention the earlier differences.
Reini Urban [Fri, 24 Apr 2015 00:56:08 +0000 (20:56 -0400)]
actually load vars to test its new sideeffect
For: RT # 124369
Peter John Acklam [Tue, 28 Apr 2015 11:07:43 +0000 (13:07 +0200)]
Fix bug in Math::BigFloat's bceil() and bint() methods.
- When $x is a Math::BigFloat, $x -> bceil() and $x -> bint() for -1 < $x < 0
returns -0. Negative zero is never used by Math::Big(Int|Float), and care has
been taken to avoid it, so this bug is surely an oversight.
- This patch fixes CPAN RT #104012.
Peter John Acklam [Fri, 24 Apr 2015 13:13:33 +0000 (15:13 +0200)]
Speed up Math::BigFloat -> blog().
- Intermediate computations in blog() increases the number of digits
significantly in some cases. Reduce the number of digits by rounding.
However, keep some extra digits for remaining intermediate computations
before the final rounding.
- This patch fixes CPAN RT #43924.
Peter John Acklam [Fri, 17 Apr 2015 19:28:37 +0000 (21:28 +0200)]
Correct Math::BigInt -> new() for non-integer input.
The documentation in Math::BigInt says about input values that "Non-integer
values result in NaN." The actual behaviour doesn't comply with this.
- All input values in the range (-1,1) written as a decimal number, e.g.,
-0.75 and 0.5, now return NaN, not 0.
- Input values with a large (absolute value) negative exponent, e.g.,
1e-9999999, now return NaN. The former behaviour was to die with the
message "Quantifier in {,} bigger than 32766 in regex; marked by ..."
- This patch fixes CPAN RT #61887 and CPAN RT #63038.
Peter John Acklam [Tue, 14 Apr 2015 18:56:40 +0000 (20:56 +0200)]
Correct bdiv() and bmod() in Math::BigRat
- Added support for bdiv() in list context. Added corresponding tests.
- Changed bmod() so the result corresponds to Perl's % operator, as well as
the second output argument of bdiv() in list contest. Modified tests
correspondingly. Now the behaviour of bmod() is consistent between
Math::BigInt, Math::BigFloat, and Math::BigRat.
- Clearer POD for the bdiv() and bmod() methods.
Peter John Acklam [Tue, 14 Apr 2015 13:20:14 +0000 (15:20 +0200)]
Correct bdiv() and bmod() in Math::BigInt and Math::BigFloat
- The POD documentation, as well as the comments in the code, say that
$x->bdiv($y) in list context should return quotient $q and remainder $r
so that $x = $q * $y + $r, and that the remainder (modulo) $r should
correspond to Perl's % operator as well as the bmod() method. This has
not been the actual behaviour. This patch fixes this.
- Clearer POD documentation for the bdiv() and bmod() methods.
- Changed test files to match corrected behaviour.
- Removed some duplicated tests.
Daniel Dragan [Fri, 27 Mar 2015 07:02:58 +0000 (03:02 -0400)]
remove obsolete VOL in S_parse_body
commit
312caa8e97 removed dJMPENV from perl_parse_body (former perl_parse)
when it created perl_parse_body and perl_parse kept the dJMPENV, but this
commit forgot to remove the VOL tag on var dosearch. Since the VOL is
only required when the function has a setjmp in it, and this function
doesn't have a setjmp in it anymore, remove the VOL tag.
Dagfinn Ilmari Mannsåker [Fri, 20 Mar 2015 17:49:12 +0000 (17:49 +0000)]
Handle fcntl(F_GETPIPE_SZ) failure in t/io/eintr.t
Even if the F_GETPIPE_SZ constant is defined, the fcntl() call can
fail (e.g. with EINVAL), so fall back to the default in that case.
Zefram [Wed, 18 Mar 2015 23:49:28 +0000 (19:49 -0400)]
Impose deterministic order on cpp-definition options.
Heretofore, ExtUtils::CBuilder put cpp-definition options into the cc command line in
non-deterministic order. This produced noise when diffing build logs. Make
this order deterministic.
Add tests for ascii-betical order in t/04-base.t. Increment $VERSION in three
packages. Remove commented-out line.
For: RT #124106
David Golden [Tue, 17 Mar 2015 02:02:00 +0000 (22:02 -0400)]
Apply changes for CPAN-Meta-YAML v0.014.
Held to be applied to blead before perl-5.23.1.
One test failure was corrected: t/Porting/exec-bit.t.
set +x on this executable, as it is in the cpan distribution
Additional patch by Karen Etheridge.
Ricardo Signes [Mon, 1 Jun 2015 19:25:09 +0000 (15:25 -0400)]
epigraphs: the v5.22.0 epigraph
Ricardo Signes [Mon, 1 Jun 2015 18:58:42 +0000 (14:58 -0400)]
feature: add the 5.23 bundle (same as 5.22)
Ricardo Signes [Mon, 1 Jun 2015 18:43:46 +0000 (14:43 -0400)]
drop the perldeltas for 5.21.*
Chris 'BinGOs' Williams [Mon, 1 Jun 2015 18:33:29 +0000 (19:33 +0100)]
Update M-CL Changes to reflect v5.22.0 release
Karen Etheridge [Mon, 1 Jun 2015 18:32:32 +0000 (19:32 +0100)]
[PATCH] add changelog entries that were in perl52111delta for this release
Signed-off-by: Chris 'BinGOs' Williams <chris@bingosnet.co.uk>
Ricardo Signes [Mon, 1 Jun 2015 18:28:47 +0000 (14:28 -0400)]
corelist: add stubs for v5.23.0
Ricardo Signes [Mon, 1 Jun 2015 18:26:47 +0000 (14:26 -0400)]
version bump: make regen and make META.json
Ricardo Signes [Mon, 1 Jun 2015 18:23:04 +0000 (14:23 -0400)]
add perldelta for 5.23.0
Ricardo Signes [Mon, 1 Jun 2015 18:16:27 +0000 (14:16 -0400)]
bump version to v5.23.0
Ricardo Signes [Mon, 1 Jun 2015 16:33:26 +0000 (12:33 -0400)]
Coro is known broken on blead
see http://nntp.perl.org/group/perl.perl5.porters/228257
Ricardo Signes [Sat, 30 May 2015 15:50:01 +0000 (11:50 -0400)]
META.json: this is now a stable release
Ricardo Signes [Sat, 30 May 2015 15:45:31 +0000 (11:45 -0400)]
remove the RC marker from patchlevel.h for v5.22.0!
Jarkko Hietaniemi [Mon, 1 Jun 2015 11:51:42 +0000 (07:51 -0400)]
perldelta: Tru64: mention also the CPAN failures.
Jarkko Hietaniemi [Sun, 31 May 2015 19:38:40 +0000 (15:38 -0400)]
perldelta: Elaborate on the IRIX and Tru64 failures.
Karl Williamson [Sat, 30 May 2015 16:14:57 +0000 (10:14 -0600)]
perldiag: Clarify wording
Karl Williamson [Sat, 30 May 2015 16:13:57 +0000 (10:13 -0600)]
perldelta: Fix typos, make clarifications; nits
Karl Williamson [Sat, 30 May 2015 16:12:33 +0000 (10:12 -0600)]
perldelta: Capitalize perl when it means the language
There are plenty of other places in our documentationwhere the
captialization is iffy, but this one was unambiguous to me.
Karl Williamson [Sat, 30 May 2015 16:10:55 +0000 (10:10 -0600)]
perldelta: Add L<> for hyperlinks
Karl Williamson [Sat, 30 May 2015 16:08:52 +0000 (10:08 -0600)]
perldelta: Use American instead of UK spelling
Karl Williamson [Sat, 30 May 2015 16:06:25 +0000 (10:06 -0600)]
perldelta: Rmv unhelpful text
There no longer is any information about this module's changes, so don't
bother to mention it changed.
Karl Williamson [Sat, 30 May 2015 16:05:34 +0000 (10:05 -0600)]
perldelta: Rmv some intra 5.21 changes
Karl Williamson [Sat, 30 May 2015 15:38:41 +0000 (09:38 -0600)]
perldelta: Add/Rmv C<>
A few things inappropriately were wrapped with C<>, but mostly add them.
Karl Williamson [Fri, 29 May 2015 22:26:00 +0000 (16:26 -0600)]
perldelta: Italicize "e.g.", "i.e."
These are now consistent throughout this pod.
Ricardo Signes [Sat, 30 May 2015 02:08:14 +0000 (22:08 -0400)]
perldelta: correct the upgraded-to versions in modules section
Karl Williamson [Fri, 29 May 2015 21:29:37 +0000 (15:29 -0600)]
perldelta: Note that the 'locale' warnings category is new
The previous commit lost this information. This adds it back, and in
all places where it is applicable, consistent with the handling of other
new warnings categories.
Karl Williamson [Fri, 29 May 2015 21:27:01 +0000 (15:27 -0600)]
perldelta: Refactor grouped warnings
Several warning messages were in two groups at the end of the warnings
section. This commit removes them from their groups and places them
individually merged alphabetically with the rest of the messages.
Karl Williamson [Fri, 29 May 2015 21:20:16 +0000 (15:20 -0600)]
perldelta: Add missing warnings categories
Karl Williamson [Fri, 29 May 2015 20:58:19 +0000 (14:58 -0600)]
perldelta: Move 'no re' to bug fix section
This really isn't any more of an enhancement than any bug fix is
Karl Williamson [Fri, 29 May 2015 20:54:21 +0000 (14:54 -0600)]
perldelta: Move 2 fatal msgs into that section from warnings section
Ricardo Signes [Fri, 29 May 2015 22:47:57 +0000 (18:47 -0400)]
perldelta: close setting $! is a bugfix
Ricardo Signes [Fri, 29 May 2015 21:06:46 +0000 (17:06 -0400)]
perldelta: remove an obsolete notice
Ricardo Signes [Fri, 29 May 2015 14:13:08 +0000 (10:13 -0400)]
update epigraphs for v5.22.0 RCs
Ricardo Signes [Fri, 29 May 2015 12:10:23 +0000 (08:10 -0400)]
perldelta: add some known problem modules
based on http://analysis.cpantesters.org/beforemaintrelease?pair=5.20.2:5.22.0%20RC2
Ricardo Signes [Thu, 28 May 2015 23:02:43 +0000 (19:02 -0400)]
perldelta: alphabetize module upgrades
Karl Williamson [Thu, 28 May 2015 20:38:56 +0000 (14:38 -0600)]
perldelta: Add some details for Module changes
This needs to be scrubbed before release.
Karl Williamson [Thu, 28 May 2015 17:39:59 +0000 (11:39 -0600)]
perlfunc: Fix confusing entries for splice
=item splice ARRAY or EXPR,OFFSET,LENGTH,LIST
is what it used to say. Is that "or" to be taken literally as the 'or'
operator or what?
Based on irc feedback this was spit into two lines
=item splice ARRAY,OFFSET,LENGTH,LIST
=item splice EXPR,OFFSET,LENGTH,LIST
This is also the method used in the other like-entries in the pod
Dagfinn Ilmari Mannsåker [Thu, 21 May 2015 23:29:46 +0000 (00:29 +0100)]
perldelta: Revert erroneous C<> removal for Inf/NaN output
Also clarify the fact that these refer to the literal strings that get
output when stringifying infinite and not-a-numer values, not to the
values themselves.
Ricardo Signes [Thu, 21 May 2015 22:34:45 +0000 (18:34 -0400)]
add new address for kmx
KMX [Thu, 14 May 2015 07:20:08 +0000 (09:20 +0200)]
handle existing mkstemp() in mingw-w64-v4
Ricardo Signes [Thu, 21 May 2015 14:04:28 +0000 (10:04 -0400)]
update perlhist and patchlevel for RC2
Ricardo Signes [Thu, 21 May 2015 14:00:38 +0000 (10:00 -0400)]
use ~1 instead of ^ for parent commit
^ caused problems under CMD.EXE; fix suggested by Nicholas Clark
Craig A. Berry [Thu, 21 May 2015 11:57:35 +0000 (06:57 -0500)]
Correct default for usedevel on VMS.
Before
32cb1fe2e04a1dce it was hard-coded to undef, which wasn't
right but after that commit it was only set for the defined case,
and the omission didn't show up until PERL_VERSION became an even
number in the release candidate.
Tony Cook [Thu, 21 May 2015 04:32:30 +0000 (14:32 +1000)]
They're Unicode tables, not UTF-8 tables
Daniel Dragan [Tue, 19 May 2015 17:22:40 +0000 (13:22 -0400)]
pack('f', $NAN) must account for NAN_COMPARE_BROKEN platforms
VC6 was returning either packed float +inf or packed float -inf
(I dont remember) instead of packed float NAN in t/op/infnan.t .
This fixes #125203
Karl Williamson [Wed, 20 May 2015 03:59:37 +0000 (21:59 -0600)]
perldelta: Nits, clarifications, wordsmithing
This eliminates some redundancies and clarifies some wording.
Karl Williamson [Wed, 20 May 2015 03:57:13 +0000 (21:57 -0600)]
perldelta: Remove 5.21-only changes
As discussed on irc, the person reading this pod shouldn't be interested
in intra-5.21 changes.
Karl Williamson [Wed, 20 May 2015 03:55:55 +0000 (21:55 -0600)]
perldelta: Don't split =head across lines
My perldoc caused these split lines to be output wrongly.
Karl Williamson [Wed, 20 May 2015 03:53:46 +0000 (21:53 -0600)]
perldelta: List more changes in the pods
I looked at the differences of all pods in /pod, and found a bunch of
ones that I think are significant that weren't listed in perldelta.
Karl Williamson [Tue, 19 May 2015 20:20:20 +0000 (14:20 -0600)]
perldelta: Rmv reference to internal flag
SCF_DO_SUBSTR is a flag internal to the current implementation of the
regular expression optimizer. There is no need to proclaim its
existence to the outside world, and is just extraneous noise.
I myself do not understand this flag, and I've spent more time looking
at this code than all but a few people likely to be reading this
perldelta. If someone who does understand it could explain it to me, I
would add comments to the code (after the freeze) to aid future readers.
Karl Williamson [Mon, 18 May 2015 14:35:56 +0000 (08:35 -0600)]
perldelta: Add, subtract some C<> S<> F<>
Dagfinn Ilmari Mannsåker [Wed, 20 May 2015 01:00:59 +0000 (02:00 +0100)]
Use "UTF-8" consistently in perldelta
Except when referring to actual names of things.
Also update the diagnostic description in perldiag.
Tony Cook [Wed, 20 May 2015 01:40:24 +0000 (11:40 +1000)]
perlapi was updated to refer to Newx, not NewX
Tony Cook [Wed, 20 May 2015 00:22:39 +0000 (10:22 +1000)]
use the piconv tool to convert source files, not the encoding pragma
Ricardo Signes [Tue, 19 May 2015 19:52:44 +0000 (15:52 -0400)]
perldelta: re-balance a C<..>
Ricardo Signes [Tue, 19 May 2015 17:38:50 +0000 (13:38 -0400)]
remove the temporary perl5220delta from Porting
Ricardo Signes [Tue, 19 May 2015 12:13:19 +0000 (08:13 -0400)]
document infnan failure on VC6 as known problem
Ricardo Signes [Tue, 19 May 2015 12:09:49 +0000 (08:09 -0400)]
fill in tentative release date for v5.22.0
Ricardo Signes [Tue, 19 May 2015 12:08:30 +0000 (08:08 -0400)]
patchlevel: we are now releasing v5.22.0 RC1
Daniel Dragan [Tue, 19 May 2015 05:23:29 +0000 (01:23 -0400)]
dont use a 64 bit constant for a 32 bit value
Perl on MSVC6 doesnt support 64 bit ints (p5p choice not to support it)
so this macro isn't defined on MSVC6 builds, commit
e59642234e hid this
mistake from non-DEBUGGING builds. The mistake is a copy paste mistake
from commit
eacbb37937 . Miniperl fails at VC6 link time due to UINT64_C
symboil not being found.
..\pad.c(165) : warning C4013: 'UINT64_C' undefined; assuming extern
returning int
..\pad.c(165) : warning C4018: '!=' : signed/unsigned mismatch
Steve Hay [Mon, 18 May 2015 17:13:33 +0000 (18:13 +0100)]
Revert Windows test watchdog() to kill('KILL', ...)
There are suspicions that the process tree kill('-KILL', ...) might be
nuking too much. It was only done to kill the cmd.exe+perl.exe tree that
was unexpectedly launched by system(1, $cmd), but by switching to
system({$perl} 1, $perl, '-e', $prog) we can avoid the intermediate cmd.exe
and thus revert to normal process kill('KILL', ...) instead to kill the
perl.exe that is now launched directly.
See http://www.nntp.perl.org/group/perl.perl5.porters/2015/05/msg227859.html
Ricardo Signes [Tue, 19 May 2015 00:41:35 +0000 (20:41 -0400)]
corelist: regenerate the v5.22.0 corelist delta
Ricardo Signes [Tue, 19 May 2015 00:38:48 +0000 (20:38 -0400)]
perlhist: expected perl v5.22.0 release dates
Ricardo Signes [Tue, 19 May 2015 00:36:51 +0000 (20:36 -0400)]
perldelta: the v5.22.0 acknowledgements
Ricardo Signes [Tue, 19 May 2015 00:16:12 +0000 (20:16 -0400)]
switch perldelta module diff to "go ask corelist"
We may yet restore a summary of important changes, but for now,
this is it.
David Mitchell [Sun, 17 May 2015 18:38:44 +0000 (19:38 +0100)]
Revert "stop T_IN/OUT/INOUT/STDIO typemaps leaking"
This reverts commit
50e5165b9638b94be310f15477b42935c79e82d5.
That commit fixed the leak too well and instead introduced a potential
premature free.
This re-introduces the long-standing leak, which will be addressed post
5.22 release.
See RT #124181
David Mitchell [Mon, 18 May 2015 16:23:06 +0000 (17:23 +0100)]
davem's perldelta proofreading part 3
These are my changes from between
=head1 Selected Bug Fixes
inclusive up to EOF.
Some notes.
The C<$lexical = chr $foo> optimisations were added and then removed
during 5.21.x, so I've removed all references to them.
Ditto readline after <<>> was added.
Ditto "Only stringy classnames are now shared".
Ditto "two assertion failures introduced into C<-DPERL_OP_PARENT>"
The size overflow stuff of 123554 was mentioned in two parts; now
consolidated into one.
Moved the two flip-flop items to be next to each other.
David Mitchell [Mon, 18 May 2015 12:18:28 +0000 (13:18 +0100)]
davem's perldelta proofreading part 2
These are my changes from between
=head1 Modules and Pragmata
inclusive up to (but not including)
=head1 Selected Bug Fixes
Some specific comments:
I moved "Missing or undefined argument to require" from "changed diags"
to "new diags": the behaviour has changed, but its a completely new diag
message.
"<B> now compiles again on Windows" I assume was only broken within the
5.21.x series.
The headline description of Perl_sv_get_backrefs() made it sound like it
returns a boolean. The more detailed description I've deleted, since
people can get that from the api doc.
I've merged the two entries for cv_name.
David Mitchell [Mon, 18 May 2015 11:39:12 +0000 (12:39 +0100)]
clarify apidoc for Perl_sv_get_backrefs
Steve Hay [Mon, 18 May 2015 08:05:33 +0000 (09:05 +0100)]
PerlIO::via::QuotedPrint finally has a new release which synchronizes it with blead
David Mitchell [Sun, 17 May 2015 19:48:39 +0000 (20:48 +0100)]
perdelta tweaks based on bulk88 feedback.
See http://nntp.perl.org/group/perl.perl5.porters/228010
David Mitchell [Sun, 17 May 2015 18:48:25 +0000 (19:48 +0100)]
bump utf8 version after doc fix
Rafael Garcia-Suarez [Sun, 17 May 2015 08:03:11 +0000 (10:03 +0200)]
Fix a doc typo in the spelling of EBCDIC
James E Keenan [Sat, 16 May 2015 23:41:06 +0000 (19:41 -0400)]
Small grammatical correction in documentation of C<each>.
David Mitchell [Sat, 16 May 2015 20:22:55 +0000 (21:22 +0100)]
a few perldelta nits
Aaron Crane [Sat, 16 May 2015 16:05:42 +0000 (17:05 +0100)]
perldelta: typo fixes and wordsmithing
Herbert Breunung [Sat, 16 May 2015 12:05:03 +0000 (08:05 -0400)]
note that .. can also create string ranges
David Mitchell [Fri, 8 May 2015 13:46:01 +0000 (14:46 +0100)]
RT #124156: death during unwinding causes crash
v5.19.3-139-g2537512 changed POPSUB and POPFORMAT so that they also
unwind the relevant portion of the scope stack. This (sensible) change
means that during exception handling, contexts and savestack frames are
popped in lock-step, rather than all the contexts being popped followed by
all the savestack contents.
However, LEAVE_SCOPE() is now called by POPSUB/FORMAT, which can trigger
destructors, tied method calls etc, which themselves may croak. The new
unwinding will see the old sub context still on the context stack and call
POPSUB on it again, leading to double frees etc.
At this late stage in code freeze, the least invasive change is to
use an unused bit in cx->blk_u16 to indicate that POPSUB has already
been called on this context frame.
Sometime later, this whole area of code really needs a thorough overhaul.
The main issue is that if cxstack_ix-- is done too early, then calling
destructors etc can overwrite the current context frame while we're still
using using it; if cxstack_ix-- is done too late, then that stack frame
can end up getting unwound twice.
Tony Cook [Fri, 15 May 2015 04:14:05 +0000 (14:14 +1000)]
perldelta for
5cd35aacb089
Tony Cook [Fri, 15 May 2015 04:06:45 +0000 (14:06 +1000)]
[perl #123998] prevent a missing chcp causing failures and noise on Win32
Tony Cook [Fri, 15 May 2015 01:47:11 +0000 (11:47 +1000)]
bump all of the EUMM module versions to avoid confusion
and the customized housekeeping to match
Tony Cook [Thu, 14 May 2015 03:09:36 +0000 (13:09 +1000)]
backport
37c49881 from the EUMM repository
Tony Cook [Wed, 13 May 2015 00:28:22 +0000 (10:28 +1000)]
backport
1e2f0871 from the EUMM repository
Aaron Crane [Thu, 14 May 2015 16:52:02 +0000 (17:52 +0100)]
perldelta: add obituary entry for Brian McCauley
This borrows heavily from Barbie's longer obituary:
http://blogs.perl.org/users/barbie/2015/05/brian-mccauley-1967-2015.html
David Mitchell [Thu, 14 May 2015 15:21:35 +0000 (16:21 +0100)]
davem's perldelta proofreading part 1
These are my changes up to (but not including)
=head1 Modules and Pragmata
I've expanded on things mainly where I personally was confused by what the
entry was trying to say, and where I had to refer back to the src or RT
ticket.
Some specific stuff:
"qr/foo/x> now ignores any Unicode pattern white space" appeared under
both "Core Enhancements" and "Incompatible Changes"; I've consolidated
both entries into one, in the latter section.
\C now in fact warns; it was already deprecated in an earlier release.
Performance efactoring of pp_tied and pp_ref is really a minor internal
detail not significant enough for a perldelta.
"Pathtools doesn't try to load XS on miniperl" - I've moved this from
"performance" to "build", since it doesn't affect the performance of perl
binary, but only how long it takes to build that binary.
"A typo fix reduces the size of the C<< OP >> structure." - this both
broke and was fixed within 5.21.x, so I've removed the entry.
Tony Cook [Thu, 14 May 2015 02:01:55 +0000 (12:01 +1000)]
make minitest on win32
Tony Cook [Thu, 14 May 2015 01:41:09 +0000 (11:41 +1000)]
test-prep-gcc was broken and fixed within 5.21