This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Thibault DUPONCHELLE [Thu, 17 Jun 2021 06:26:33 +0000 (08:26 +0200)]
Update UCD version. Remove changes to cpan Encode. Regen
Thibault DUPONCHELLE [Thu, 17 Jun 2021 05:20:07 +0000 (07:20 +0200)]
blogs.perl.org is not https
Thibault DUPONCHELLE [Wed, 16 Jun 2021 17:11:49 +0000 (19:11 +0200)]
Fix several unicode.org links
Thibault DUPONCHELLE [Wed, 16 Jun 2021 17:04:59 +0000 (19:04 +0200)]
Fix link for unicode faq
Thibault DUPONCHELLE [Wed, 16 Jun 2021 17:00:45 +0000 (19:00 +0200)]
pmd/cpd moved to GitHub and GitHub pages
Thibault DUPONCHELLE [Wed, 16 Jun 2021 16:49:32 +0000 (18:49 +0200)]
Fix perlbuzz link
H.Merijn Brand [Wed, 16 Jun 2021 06:30:15 +0000 (08:30 +0200)]
Disambiguate what user sees in interactive Configure
User should expect to see configuration questions spelled 'UTF-8', not
'UTF8' (even if we store information internally or in config.sh without
the hyphen).
Correct user prompts accordingly.
For: #18844
https://github.com/Perl/perl5/pull/18870 by @jkeenan
Karl Williamson [Wed, 19 Aug 2020 03:13:42 +0000 (21:13 -0600)]
perl.h: Pay attention to if can have non-int bitfields
Karl Williamson [Sun, 30 May 2021 17:55:33 +0000 (11:55 -0600)]
Fix ASSUME definition
Commit
5d5b9c460e2a06563d2b5e35a1a79991460696eb fixed the definition of
ASSUME for some purposes, but broke it on Windows. This commit should
fix that.
Felipe Gasper [Tue, 15 Jun 2021 04:25:38 +0000 (00:25 -0400)]
Fix spelling: precede
Matthew Horsfall [Mon, 14 Jun 2021 12:56:53 +0000 (08:56 -0400)]
Add perldelta entry try/catch deparsing fix -
846e32ebef
Matthew Horsfall [Thu, 10 Jun 2021 19:58:47 +0000 (15:58 -0400)]
B::Deparse: Handle try/catch when catch has an ENTER/LEAVE pair
Hugo van der Sanden [Tue, 1 Jun 2021 14:06:04 +0000 (15:06 +0100)]
regcomp.c: comments
Comment change suggestions from @hvds in PR #18835.
Karl Williamson [Mon, 31 May 2021 21:38:21 +0000 (15:38 -0600)]
regcomp.c: White-space only
My attempt to insulate from the leading tab removal the year-old commits
finally pushed as
77a6d54c0deb1165b37dcf11c21cd334ae2579bb and
403d7eb3e4320188571cf61b9dab62ff10799f49 failed miserably.
I spent a bunch of time sorting it all out, and this is the result.
Ricardo Signes [Sun, 13 Jun 2021 15:22:27 +0000 (11:22 -0400)]
perlgov: re-alphabetize the name list
Ricardo Signes [Sat, 12 Jun 2021 16:22:11 +0000 (12:22 -0400)]
use cpan.org, not ftp.funet.fi, to check the CPAN
Ricardo Signes [Sat, 12 Jun 2021 16:21:53 +0000 (12:21 -0400)]
CPAN mirror list: stop building one
James E Keenan [Sun, 13 Jun 2021 13:11:56 +0000 (13:11 +0000)]
perldelta entry for 52f0602 (IO upgrade)
James E Keenan [Sun, 13 Jun 2021 12:51:25 +0000 (12:51 +0000)]
perldelta entry for 1c1bafd (Unicode-Collate upgrade)
Max Maischein [Sun, 13 Jun 2021 08:27:27 +0000 (10:27 +0200)]
Update Socket from CPAN to 2.032
Karl Williamson [Sat, 5 Jun 2021 14:33:25 +0000 (08:33 -0600)]
Configure: apply recent metaconfig changes
Several new probes have been added.
Karl Williamson [Sat, 5 Jun 2021 13:33:13 +0000 (07:33 -0600)]
Configure changes for HAS_NON_INT_BITFIELDS
This probe has been available without my realizing it had been merged.
Some compilers, chiefly IBM, use only integer-size bitfields, warning
when presented with other-sized ones. Allowing other-sized fields is an
extension to the C standard.
These warnings don't affect the correctness of the code generated, but
very many are generated per run, potentially overwhelming the reader
into not noticing warnings that are important.
Karl Williamson [Fri, 4 Jun 2021 03:47:24 +0000 (21:47 -0600)]
regcomp.c: Fix typo in comment
Karl Williamson [Tue, 1 Jun 2021 11:59:20 +0000 (05:59 -0600)]
regcomp.h: Clarify comment
Nicholas Clark [Fri, 11 Jun 2021 12:17:44 +0000 (12:17 +0000)]
Add -Werror=vla by default. C99 variable-length arrays are dangerous.
C11 makes them optional, and MSVC doesn't support them. We can't use them
(but even if we could, we should not).
Dave Lambley [Thu, 10 Jun 2021 20:22:41 +0000 (21:22 +0100)]
Add missing quotes to IO::Socket::INET example
Also clean trailing white space.
Committer: Dave Lambley is now a Perl author.
James E Keenan [Wed, 2 Jun 2021 22:10:57 +0000 (22:10 +0000)]
Skip tests of CORE::Dump on FreeBSD
Because on this OS this test writes to /var/log/messages
For: https://github.com/Perl/perl5/issues/18847
Move FreeBSD skip condition near to other 'skip_all' conditions, per
review by TonyC.
Tomasz Konojacki [Fri, 4 Jun 2021 20:05:04 +0000 (22:05 +0200)]
replace all instances of PERL_IMPLICIT_CONTEXT with MULTIPLICITY
Since the removal of PERL_OBJECT
(
acfe0abcedaf592fb4b9cb69ce3468308ae99d91) PERL_IMPLICIT_CONTEXT and
MULTIPLICITY have been synonymous and they're being used interchangeably.
To simplify the code, this commit replaces all instances of
PERL_IMPLICIT_CONTEXT with MULTIPLICITY.
PERL_IMPLICIT_CONTEXT will stay defined for compatibility with XS
modules.
Tony Cook [Tue, 4 May 2021 05:04:25 +0000 (15:04 +1000)]
avoid a prototype warning assigning \&getcwd to *cwd
This would produce a warning if we fallback to using getcwd() where
getcwd() has a prototype.
Tony Cook [Wed, 5 May 2021 00:12:31 +0000 (10:12 +1000)]
don't fallback to simple pwd
When _backtick_pwd invokes $pwd_cmd it first clears the PATH, and since
the command has no shell metacharacters, it perl won't invoke the
shell, so it will always fail.
An alternative here might be to use "/bin/sh -c pwd" but there's no
guarantee that pwd is available as a shell builtin.
Tony Cook [Tue, 4 May 2021 04:55:50 +0000 (14:55 +1000)]
remove code that assuming finding pwd on the path is reasonable
We deliberately clear PATH when invoking pwd, so this search is
useless.
Tony Cook [Thu, 3 Jun 2021 05:51:01 +0000 (15:51 +1000)]
prevent undefined behaviour at a language level while probing getenv
This test in Configure tries to probe for undefined behaviour in
getenv(), but provokes undefined behaviour in C/C++ by falling off
the end of a function with a non-void return type.
Without optimization clang++ generated a ud2 instruction here on
amd64 producing an illegal instruction exception. With optimization
the test case fell off the end and started re-executing main(),
eventually producing a SIGBUS.
Simply dropping the value of getenv() here and returning NULL wasn't
useful, under -O2 the compiler optimized away the getenv() call,
voiding the whole point of the test.
Scott Baker [Wed, 2 Jun 2021 20:31:28 +0000 (13:31 -0700)]
Simply the uc() and lc() documentation for novices
Tomoyuki Sadahiro [Sun, 6 Jun 2021 22:23:17 +0000 (22:23 +0000)]
Synch with CPAN version 1.30
Per https://rt.cpan.org/Ticket/Display.html?id=133952
James E Keenan [Sun, 6 Jun 2021 15:40:51 +0000 (11:40 -0400)]
Obituary for Spider Boardman
Per posting on perl.module-authors by Neil Bowers and
https://www.nashuatelegraph.com/obituaries-memorials/obituaries/2021/05/29/spider-raun-boardman/
Scott Baker [Wed, 26 May 2021 18:12:09 +0000 (11:12 -0700)]
Add some examples to the glob() documentation
Leon Timmermans [Tue, 12 Jan 2021 22:47:55 +0000 (23:47 +0100)]
Call magic on all elements on %SIG delocalization
Leon Timmermans [Fri, 14 May 2021 12:18:03 +0000 (14:18 +0200)]
Correctly call delete magic on all hash magic
Previously it would only call it correctly if the hash magic was
RMAGICAL, which is only set if a magic either has clear magic or has
neither get or set magic. This means any magic with get or set would
break.
Paul "LeoNerd" Evans [Mon, 31 May 2021 21:35:40 +0000 (22:35 +0100)]
Rename G_ARRAY to G_LIST; provide back-compat when not(PERL_CORE)
Hugo van der Sanden [Tue, 1 Jun 2021 13:33:57 +0000 (14:33 +0100)]
gh18770: stop scanning for substrs after *COMMIT
*ACCEPT already avoids this (because it is "ENDLIKE"), but gets a
related fix to stop scanning for start class.
Karl Williamson [Sat, 2 May 2020 12:58:57 +0000 (06:58 -0600)]
regcomp.c: white-space; comments
Michael G Schwern [Wed, 5 May 2021 13:18:01 +0000 (07:18 -0600)]
Base *.[ch] files: Replace leading tabs with blanks
This is a rebasing by @khw of part of GH #18792, which I needed to get
in now to proceed with other commits.
It also strips trailing white space from the affected files.
Karl Williamson [Sat, 2 May 2020 12:45:01 +0000 (06:45 -0600)]
regcomp.c: Extract code from a too-large-function
S_regclass() is unwieldy. This commit splits it into two nearly equal
size parts. More could be done.
Karl Williamson [Mon, 31 May 2021 15:10:21 +0000 (09:10 -0600)]
perl5340delta: Clarify /x{,3}/
This was suggested by @LanX
This fixes GH #18812
Karl Williamson [Mon, 31 May 2021 15:01:58 +0000 (09:01 -0600)]
perluniprops: Remove references to Unicode::Unihan
This CPAN module doesn't work on recent Unicode versions
This fixes GH #18787
Ricardo Signes [Sat, 8 May 2021 14:12:18 +0000 (10:12 -0400)]
Porting/docs-team-charter.pod: the docs team charter
Hugo van der Sanden [Thu, 8 Aug 2019 16:27:14 +0000 (17:27 +0100)]
[gh 17847] data->pos_delta should stick at infinity
The expression we're about to add to data->pos_delta in this part of
study_chunk() can be both positive or negative; however while we apply
an overflow check to avoid exceeding OPTIMIZE_INFTY, we were happily
subtracting from it when the expression was negative, making it no longer
infinite.
Hugo van der Sanden [Sun, 4 Aug 2019 13:07:22 +0000 (14:07 +0100)]
[gh 17847] avoid overflow on delta in study_chunk
delta and pos_delta may hold OPTIMIZE_INFTY to represent infinity.
Hugo van der Sanden [Sun, 4 Aug 2019 13:31:29 +0000 (14:31 +0100)]
[gh 17847] Include data->pos_delta in #if'd-out diagnostic
Hugo van der Sanden [Mon, 31 May 2021 13:08:54 +0000 (14:08 +0100)]
gh18842: fix a pair of sprintf format warnings
The format is '%s', so it is not appropriate to wrap the char * arguments
in SVfARG().
James E Keenan [Fri, 21 May 2021 03:35:02 +0000 (23:35 -0400)]
Improve grammar
Per suggestion by oodler577 in https://github.com/Perl/perl5/pull/18736.
With correction reported as needed by grinnz in
https://github.com/Perl/perl5/pull/18806.
積丹尼 Dan Jacobson [Sat, 22 May 2021 04:24:43 +0000 (12:24 +0800)]
Add difference between <=> and cmp example
For https://github.com/Perl/perl5/issues/18742
Matthew Horsfall [Tue, 25 May 2021 00:06:51 +0000 (20:06 -0400)]
perlhacktips: Add callgrind profiling documentation
One of the benefits of callgrind is you can profile perls that
haven't been built with debugging symbols, which can be real useful
on live-ish datasets
Karl Williamson [Mon, 31 May 2021 13:39:10 +0000 (07:39 -0600)]
Bump feature.pm Version to 1.66
H.Merijn Brand [Mon, 31 May 2021 13:25:55 +0000 (15:25 +0200)]
Sync Configure with META/DIST : >&4 changes and a typo
Felipe Gasper [Thu, 20 May 2021 14:22:07 +0000 (10:22 -0400)]
Clarify descriptions of unicode_eval and evalbytes.
Issue #18801
Hugo van der Sanden [Sat, 22 May 2021 18:39:11 +0000 (19:39 +0100)]
gh17824: zero curlocales[]
Static analysis tools such as Coverity and clang report that we can
otherwise end up reading uninitialized data, and inspection agrees.
Karl Williamson [Sun, 16 May 2021 19:11:17 +0000 (13:11 -0600)]
Refactor UTF_START_MARK()
This allows the removal of a conditional in a very low level (called a
lot) macro
Tomasz Konojacki [Sat, 29 May 2021 22:02:24 +0000 (00:02 +0200)]
The previous commit wasn't meant to be pushed
This reverts commit
e9f285aca6c125ef50e64ed9b85265c9896b7b39.
Tomasz Konojacki [Sat, 29 May 2021 22:01:23 +0000 (00:01 +0200)]
ssh
Karl Williamson [Sun, 16 May 2021 19:15:56 +0000 (13:15 -0600)]
UTF8_IS_NEXT_CHAR_DOWNGRADEABLE() check before deref
Reorder the clauses to check first before dereferencing
Chris 'BinGOs' Williams [Sat, 29 May 2021 08:53:03 +0000 (09:53 +0100)]
Happy World Digestive Health Day
Dan Book [Sat, 29 May 2021 04:29:55 +0000 (00:29 -0400)]
perlcommunity - add a webchat link for Libera.Chat #perl
Karl Williamson [Tue, 30 Jun 2020 20:04:24 +0000 (14:04 -0600)]
utf8.h: Simplify UNICODE_IS_SURROGATE()
This uses inRANGE() with mnemonics to make it clearer with no increase
in the number of conditionals
Karl Williamson [Sun, 16 May 2021 22:31:35 +0000 (16:31 -0600)]
utf8.h: Use inRANGE for UNICODE_IS_32_CONTIGUOUS_NONCHARS
This leads to a single conditional instead of two.
Karl Williamson [Sun, 16 May 2021 22:07:11 +0000 (16:07 -0600)]
utf8.h: Refactor UNICODE_IS_NONCHAR()
This adds branch prediction and re-orders so that an unlikely to succeed
test is done before the likely to succeed one, so that the latter
usually doesn't need to be executed. Since both conditions must succeed
for the entire expression to succeed, this doesn't change what the whole
expresson matches.
s# Please enter the commit message for your changes. Lines starting
Karl Williamson [Tue, 5 May 2020 23:48:55 +0000 (17:48 -0600)]
regexec.c: Move parameter cast into macro
Instead of calling the macro with a cast parameter, do the cast inside
the macro so the caller doesn't have to be bothered with it.
Karl Williamson [Wed, 26 May 2021 15:37:37 +0000 (09:37 -0600)]
inline.h: Clarify comment
Karl Williamson [Tue, 25 May 2021 17:16:07 +0000 (11:16 -0600)]
regcomp.h: Fix typos in comments
Karl Williamson [Mon, 10 May 2021 13:30:24 +0000 (07:30 -0600)]
regexec.c: Replace code with equivlent inline fcn
Don't repeat a paradigm
Karl Williamson [Tue, 18 May 2021 02:31:57 +0000 (20:31 -0600)]
perl.h: Fix typo in comment
Karen Etheridge [Mon, 1 Mar 2021 22:17:13 +0000 (14:17 -0800)]
Vulnerabilities in cpan/ modules still affect perl; we should handle them
Karl Williamson [Thu, 13 May 2021 17:50:44 +0000 (11:50 -0600)]
isUTF8_CHAR() Check ptr before dereferencing
It is legal to call this function, though not so done in core, with
empty input. By swapping two conditions in the same 'if', we check if
empty before trying to access it.
Tony Cook [Wed, 26 May 2021 00:45:42 +0000 (10:45 +1000)]
remove MSVC100/vcexpress2010 from CI, at least for now
Due to failing installation of VC express 2010, that version isn't
being tested anyway, and is just producing errors.
If we work out a solution for the missing download from Microsoft
we can reinstate it, but at this point it's just noise
reducing the utility CI gives us.
This does not make VC2010 unsupported, it's purely a CI change.
Richard Leach [Thu, 13 May 2021 09:17:52 +0000 (10:17 +0100)]
Perl_clear_defarray - array does not actually need Zeroing
AvREIFY_only() is about to be applied
Richard Leach [Mon, 19 Apr 2021 20:28:01 +0000 (21:28 +0100)]
Perl_clear_defarray: faster array creation via new macro+function
Richard Leach [Mon, 19 Apr 2021 20:27:03 +0000 (21:27 +0100)]
Add Perl_av_new_alloc() function and newAV_alloc_x/z() macros
Nicholas Clark [Tue, 25 May 2021 19:25:12 +0000 (19:25 +0000)]
Data::Dumper 2.180 was released on 2021-05-26.
Richard Leach [Tue, 25 May 2021 17:32:48 +0000 (18:32 +0100)]
perldelta.pod: entry for 2583ef6 (undef = )
Richard Leach [Tue, 11 May 2021 20:04:45 +0000 (21:04 +0100)]
Make scalar assignment to undef a compile time error
Nicholas Clark [Mon, 24 May 2021 15:21:47 +0000 (15:21 +0000)]
Bump Data::Dumper's $VERSION and update Changes, for a dev release.
Nicholas Clark [Mon, 24 May 2021 15:10:12 +0000 (15:10 +0000)]
Restore the explicit Data::Dumper Makefile.PL
Data::Dumper's upstream repository is (effectively) blead, and whilst we can
auto-generate Makefile.PL suitable for blead, this doesn't provide some
other functionality needed for CPAN.
This also fixes CPAN #12282, which has been missed forever.
Really we should decide *which* bugtracker is canonical for Data::Dumper
(and will be checked and acted upon) and then record that metadata in this
Makefile.PL so that metacapn.org etc link to it.
Tony Cook [Tue, 25 May 2021 01:07:28 +0000 (11:07 +1000)]
Tony Cook [Tue, 25 May 2021 00:51:05 +0000 (10:51 +1000)]
File::Copy: compare inode numbers as strings
as recommended in perldoc -f stat
Todd Rinaldo [Fri, 21 May 2021 15:01:32 +0000 (10:01 -0500)]
Update version from CPAN to 0.9929
[DELTA]
0.9929
* Upgrade to ppport 3.62
* Import fix for core build warning
Leon Timmermans [Tue, 19 Jan 2021 23:37:54 +0000 (00:37 +0100)]
Make use open :std push layers without any munging
Previously it would only apply the first layer that was either :utf8 or
any other layer with and argument (e.g. :encoding(utf8)).
Leon Timmermans [Mon, 19 Apr 2021 16:49:52 +0000 (18:49 +0200)]
Do not cache utf8 offsets for non-canonical lengths
In particular, if the length is beyond the end, it should not be stored as the end.
Nicholas Clark [Mon, 24 May 2021 07:18:33 +0000 (07:18 +0000)]
Bump Data::Dumper's $VERSION and update Changes, for a dev release.
Nicholas Clark [Sun, 23 May 2021 20:11:52 +0000 (20:11 +0000)]
5.8.1 provides all of isASCII, isPRINT, isALPHA and isIDFIRST.
So no need to emulate them.
Nicholas Clark [Sun, 23 May 2021 19:23:26 +0000 (19:23 +0000)]
Dumper.xs can use more emulation functions from ppport.h
ppport.h provides HvNAME_get and isWORDCHAR, so we can drop our duplicate
definitions.
We're already using the ppport.h emulation of utf8_to_uvchr_buf, so remove
ours. "my_sprintf is not supported by ppport.h" is no longer true, so "NEED"
ppport.h's my_sprintf, and remove our conditional compilation.
Max Maischein [Mon, 24 May 2021 08:49:41 +0000 (10:49 +0200)]
Update Maintainers.PL with experimental version info from 0.022 to 0.24
This has no actual changes, since the changes were imported
with
acd998d76362b480eb8e517c712de155d2659d03 , but didn't update
Maintainers.PL
Tony Cook [Tue, 30 Mar 2021 04:36:32 +0000 (15:36 +1100)]
mark test modules (found in ext/*/t/) as no_index
fixes #18632
Tony Cook [Mon, 24 May 2021 04:55:39 +0000 (14:55 +1000)]
update PathTools changes for the inode comparison change
Tony Cook [Wed, 12 May 2021 02:24:59 +0000 (12:24 +1000)]
bump PathTools to 3.81
Tony Cook [Mon, 17 May 2021 05:58:55 +0000 (15:58 +1000)]
PathTools: compare inode numbers as strings
as recommended in perldoc -f stat
Tony Cook [Tue, 27 Apr 2021 06:22:35 +0000 (16:22 +1000)]
fix utime() to set $!/errno when called on a closed handle
My
8334cae65 was intended to fix this, but the test was faulty, and
didn't correctly fail.
This started showing as a failure on cygwin, with the fixed test it
also fails on Linux, so fix doio.c as well.
Tony Cook [Wed, 12 May 2021 01:36:39 +0000 (11:36 +1000)]
update dist/PathTools/Changes
Nicholas Clark [Sun, 23 May 2021 13:32:39 +0000 (13:32 +0000)]
Bump Data::Dumper's $VERSION and update Changes, for a dev release.
Nicholas Clark [Sun, 23 May 2021 13:29:51 +0000 (13:29 +0000)]
Eliminate $WANT, inline &SKIP_TEST into &TEST_BOTH, and pass $want to &TEST.
This eliminates the action-at-a-distance file scoped lexical.
Nicholas Clark [Sun, 23 May 2021 12:36:04 +0000 (12:36 +0000)]
Convert all the Data::Dumper "numbers" and "strings" tests to TEST_BOTH.