10 months agoadd ~~x and x~~ overloads, deprecating ~~ overloads tonyc/smartmatch
Tony Cook [Tue, 24 Nov 2015 00:29:34 +0000]
add ~~x and x~~ overloads, deprecating ~~ overloads

10 months agoreturn a generic true/false for ~~ overloads
Tony Cook [Mon, 23 Nov 2015 23:09:25 +0000]
return a generic true/false for ~~ overloads

rather than the return value of the overload method or function

10 months agomake ~~ return a generic true/false with a subref on the right
Tony Cook [Thu, 17 Sep 2015 04:56:59 +0000]
make ~~ return a generic true/false with a subref on the right

Previously it would return the value returned by the sub

10 months agomake when(/foo/) act like when(qr/foo/)
Tony Cook [Thu, 10 Sep 2015 03:56:20 +0000]
make when(/foo/) act like when(qr/foo/)

10 months agoadd postfix when BLOCK
Tony Cook [Mon, 13 Jul 2015 07:07:53 +0000]
add postfix when BLOCK

10 months agothe behaviour of objects in overloading has changed
Tony Cook [Mon, 13 Jul 2015 06:20:30 +0000]
the behaviour of objects in overloading has changed

re-worked separately because we might re-instate the older behaviour

10 months agorip out most of the smartmatch documentation, it's much simpler now
Tony Cook [Mon, 13 Jul 2015 06:07:24 +0000]
rip out most of the smartmatch documentation, it's much simpler now

10 months agoWarn at compile time where possible.
Tony Cook [Tue, 7 Jul 2015 04:13:51 +0000]
Warn at compile time where possible.

This commit probably needs work, and the very idea needs discussion.

10 months agoreintroduce conversion of $foo ~~ /bar/ to $foo ~~ qr/bar/
Tony Cook [Thu, 2 Jul 2015 06:22:15 +0000]
reintroduce conversion of $foo ~~ /bar/ to $foo ~~ qr/bar/

This does not convert an implicit /bar/ as in when (/bar/)

10 months agore-enable overload ~~ any
Tony Cook [Thu, 2 Jul 2015 04:14:54 +0000]
re-enable overload ~~ any

10 months agoupdate non-autodie tests to work under the new smartmatch limits
Tony Cook [Thu, 2 Jul 2015 00:32:16 +0000]
update non-autodie tests to work under the new smartmatch limits

10 months agodisallow smartmatches not explicitly permitted (runtime)
Tony Cook [Mon, 22 Jun 2015 04:06:29 +0000]
disallow smartmatches not explicitly permitted (runtime)

10 months agoNew whirled order for givwhen/last/next
Father Chrysostomos [Thu, 13 Sep 2012 07:18:28 +0000]
New whirled order for givwhen/last/next

given now responds to last and next.

break only breaks out of given, and does not complain about an
inner foreach.

when and default do an implicit regular next, not a ‘special’ next
that only breaks out of foreach() and given.

10 months agotoke.c: use goto to reduce repetitition
Father Chrysostomos [Wed, 12 Sep 2012 23:33:49 +0000]
toke.c: use goto to reduce repetitition

I couldn’t think of a name for the label that didn’t just repeat what
the code was doing (increment_lex_allbrackets_and_set_PL_expect), so I
gave it an arbitrary short name.

10 months agoswitch to done_testing()
Tony Cook [Tue, 23 Jun 2015 05:15:18 +0000]
switch to done_testing()

10 months agoSmart Match 3™
Father Chrysostomos [Wed, 12 Sep 2012 07:21:12 +0000]
Smart Match 3™

I was persuaded by Smylers’ argument concerning string equivalence:
<20120905143040.GZ1742@stripey.com>

But it is not too late for me to be overruled since this is not in
blead yet.

The list is:

      RHS type
   1. undef
   2. ~~ overloading
   3. reftype eq CODE
   4. reftype eq REGEXP
   5. fall back to eq

One question still remains:  Should undef ~~ "string" give a warning?
Currently it doesn’t, but neither does it call eq overloading on the
RHS if we actually have undef ~~ $overloaded_object_without_smartmatch.

This cannot be merged until we solve the fact that smart match
is used in:

cpan/autodie/lib/Fatal.pm
cpan/autodie/t/exceptions.t

This does not deal with loop controls yet, just smartmatch and the
smartmatch aspects of of givwhen.

10 months agoTest.pm: avoid noise on STDERR.
David Mitchell [Mon, 23 Nov 2015 17:32:01 +0000]
Test.pm: avoid noise on STDERR.

If a TODO test fails, the diagnostic output should go to STDOUT rather
than STDERR.

This was originally

    https://rt.cpan.org/Public/Bug/Display.html?id=105423

but since Test has been moved from cpan/ to dist/, it can be applied
locally.

10 months agoPorting/Maintainers.pl - Fix ExtUtils::Constant
Steve Hay [Mon, 23 Nov 2015 17:41:20 +0000]
Porting/Maintainers.pl - Fix ExtUtils::Constant

Now the CUSTOMIZED file appears in customized.dat

10 months agoSilence warnings in encoding::warnings
David Mitchell [Mon, 23 Nov 2015 17:11:15 +0000]
Silence warnings in encoding::warnings

This is Karl's patch from

    https://rt.cpan.org/Public/Bug/Display.html?id=100538

which can be applied locally now that encoding::warnings has moved from
cpan/ to dist/.

10 months agomove encoding-warnings from cpan/ to dist/
David Mitchell [Mon, 23 Nov 2015 17:07:44 +0000]
move encoding-warnings from cpan/ to dist/

A private communication with Audrey indicated that to allow us to make
mods, this was the preferred solution.

10 months agoMaintainers.pl: update ExtUtils::Constant status
David Mitchell [Mon, 23 Nov 2015 16:38:55 +0000]
Maintainers.pl: update ExtUtils::Constant status

It used to indicate that the blead version had diverged dramatically from
the 0.16 CPAN version. Since then a 0.23 has been released on CPAN which
mostly matches blead - blead has one test file locally modified.

NB: I ran "t/porting/customized.t --regen" and for some reason it
didn't think that customized.dat needed updating.

10 months agoperlfunc: specify valid inputs precisely [perl #126437]
Aristotle Pagaltzis [Mon, 23 Nov 2015 14:31:32 +0000]
perlfunc: specify valid inputs precisely [perl #126437]

10 months agoMerge branch 'clang-thread-safety-analysis' into blead
Jarkko Hietaniemi [Mon, 23 Nov 2015 13:10:44 +0000]
Merge branch 'clang-thread-safety-analysis' into blead

clang thread safety analysis

10 months agobump threads version
Jarkko Hietaniemi [Fri, 20 Nov 2015 17:27:32 +0000]
bump threads version

10 months agoanalysis cannot tell that the mutex was locked
Jarkko Hietaniemi [Thu, 19 Nov 2015 02:25:04 +0000]
analysis cannot tell that the mutex was locked

It was locked in S_ithread_create() - in case it returned a non-NULL thread.
Because of this conditionality the S_ithread_create() cannot be annotated.

threads.xs:1161:9: warning: releasing mutex 'thread->mutex' that was not held [-Wthread-safety-analysis]
        MUTEX_UNLOCK(&thread->mutex);

10 months agoanalysis does not like mutex being either held or released
Jarkko Hietaniemi [Thu, 19 Nov 2015 02:18:37 +0000]
analysis does not like mutex being either held or released

threads.xs:1020:1: warning: mutex 'thread->mutex' is not held on every path through here [-Wthread-safety-analysis]
}
^

10 months agoclang specific diagnostics ignore/restore
Jarkko Hietaniemi [Sat, 21 Nov 2015 02:40:00 +0000]
clang specific diagnostics ignore/restore

10 months agoannotate S_ithread_create create_destruct_mutex use
Jarkko Hietaniemi [Wed, 18 Nov 2015 03:15:09 +0000]
annotate S_ithread_create create_destruct_mutex use

If the pool is passed in, and the create_destruct_mutex unlocked also
on success (so that all the unlocks happen in S_ithread_create()),
the create_destruct_mutex use can be annotated.

threads.xs:757:9: warning: releasing mutex 'my_poolp->create_destruct_mutex' th
at was not held [-Wthread-safety-analysis]
        MUTEX_UNLOCK(&MY_POOL.create_destruct_mutex);

threads.xs:1000:9: warning: releasing mutex 'my_poolp->create_destruct_mutex' t
hat was not held [-Wthread-safety-analysis]
        MUTEX_UNLOCK(&MY_POOL.create_destruct_mutex);

threads.c:1319:1: warning: mutex 'my_poolp->create_destruct_mutex' is not held on every path through here [-Wthread-safety-analysis]
}
^

10 months agoannotate S_ithread_count_inc mutex use
Jarkko Hietaniemi [Sat, 14 Nov 2015 14:09:38 +0000]
annotate S_ithread_count_inc mutex use

No warnings issued, but a clear case for annotations.

10 months agoannotate S_ithread_free mutex use
Jarkko Hietaniemi [Fri, 6 Nov 2015 00:13:22 +0000]
annotate S_ithread_free mutex use

threads.xs:276:13: warning: releasing mutex 'thread->mutex' that was not held [-Wthread-safety-analysis]
                  MUTEX_UNLOCK(&thread->mutex);

threads.xs:282:5: warning: releasing mutex 'thread->mutex' that was not held [-Wthread-safety-analysis]
          MUTEX_UNLOCK(&thread->mutex);

threads.xs:394:1: warning: mutex 'thread->mutex' is still held at the end of
      function [-Wthread-safety-analysis]
}

threads.xs:677:1: warning: mutex 'thread->mutex' is still held at the end of
      function [-Wthread-safety-analysis]
}

10 months agosplit off the $0 setting so mutex use can be annotated
Jarkko Hietaniemi [Thu, 5 Nov 2015 23:15:35 +0000]
split off the $0 setting so mutex use can be annotated

No warnings were emitted since the use of the PL_dollarzero_mutex
was correctly bracketed by mutex lock and unlock, but by splitting
off the code and annotating it is more likely to stay correct.

10 months agocleanup the mutex use of PerlIOStdio_close
Jarkko Hietaniemi [Fri, 6 Nov 2015 00:00:01 +0000]
cleanup the mutex use of PerlIOStdio_close

Before: clang -Wthread-safety found the flow quite suspect:
one conditional mutex lock, with two conditional unlocks.
The code *looked* like the being okay logic-wise now, but
rather fragile, so let's make it clearer and more solid
by hoisting the locking earlier.

This is a fd close, this should not be a performance sensitive spot.

And it's in PerlIOStdio, so it should be pretty dead anyway.

perlio.c:3283:18: warning: mutex 'PL_perlio_mutex' is not held on every path through here [-Wthread-safety-analysis]
        result = PerlSIO_fclose(stdio);

perlio.c:3299:6: warning: releasing mutex 'PL_perlio_mutex' that was not held [-Wthread-safety-analysis]
            MUTEX_UNLOCK(&PL_perlio_mutex);

10 months agoannotate the mutex use of S_more_refcounted_fds
Jarkko Hietaniemi [Thu, 5 Nov 2015 23:58:56 +0000]
annotate the mutex use of S_more_refcounted_fds

perlio.c:2232:2: warning: releasing mutex 'PL_perlio_mutex' that was not held [-Wthread-safety-analysis]
        MUTEX_UNLOCK(&PL_perlio_mutex);

10 months agoannotate the mutex use of atfork_lock and atfork_unlock
Jarkko Hietaniemi [Thu, 5 Nov 2015 23:57:58 +0000]
annotate the mutex use of atfork_lock and atfork_unlock

util.c:2729:1: warning: mutex 'PL_perlio_mutex' is still held at the end of function [-Wthread-safety-analysis]
}

util.c:2729:1: warning: mutex 'PL_op_mutex' is still held at the end of function [-Wthread-safety-analysis]
}

util.c:2739:5: warning: releasing mutex 'PL_perlio_mutex' that was not held [-Wthread-safety-analysis]
    MUTEX_UNLOCK(&PL_perlio_mutex);

util.c:2744:5: warning: releasing mutex 'PL_op_mutex' that was not held [-Wthread-safety-analysis]
    OP_REFCNT_UNLOCK;

10 months agoannotate the acquire/release of PL_op_mutex
Jarkko Hietaniemi [Thu, 5 Nov 2015 23:57:17 +0000]
annotate the acquire/release of PL_op_mutex

op.c:1190:1: warning: mutex 'PL_op_mutex' is still held at the end of function
}

op.c:1199:5: warning: releasing mutex 'PL_op_mutex' that was not held
    OP_REFCNT_UNLOCK;

10 months agoadd -Wthread-safety for clang 3.5 or later
Jarkko Hietaniemi [Sun, 22 Nov 2015 16:40:36 +0000]
add -Wthread-safety for clang 3.5 or later

With this change there is no need to explicitly e.g.
Configure ... -Accflags='-Wthread-safety'
since doing just -Dusethreads with recent enough clang
will automatically add the -Wthread-safety

10 months agoclang thread safety annotations
Jarkko Hietaniemi [Thu, 5 Nov 2015 23:52:20 +0000]
clang thread safety annotations

http://clang.llvm.org/docs/ThreadSafetyAnalysis.html

Static (compile-time) annotations for declaring the multithreaded
behavior of functions, variables, and capabilities (like mutexes).
Available since about clang 3.5.

./Configure -des -Dusedevel -Dusethreads -Dcc=clang -Accflags='-Wthread-safety'

clang -Wthread-safety then checks the validity of the annotations.

10 months agoassertion failure on foo(my $x : bar)
David Mitchell [Mon, 23 Nov 2015 08:15:40 +0000]
assertion failure on foo(my $x : bar)

RT #126257

'my var : attr' injects a void-context sub call just after the
pad op. However, Perl_ck_entersub_args_list() tries to impose lvalue
context on each of its args, which causes an assertion failure.

This commit makes Perl_ck_entersub_args_list() skip calling op_lvalue()
on any args which are OP_ENTERSUB/OPf_WANT_VOID. Strictly
speaking it should check that the sub call is actually an attribute
method call (e.g. first child is a const("attibutes") etc), but this
was far too much like hard work.

10 months agoTODO newish re/speed.t tests on VMS.
Craig A. Berry [Sun, 22 Nov 2015 23:02:55 +0000]
TODO newish re/speed.t tests on VMS.

The tests that dump regex debug info get their output mangled by the
record-oriented pipes but otherwise look fine.  So mark them to-do.

10 months agomake_ext.pl: fix operator precedence error from b4c079ca5
Tony Cook [Sun, 22 Nov 2015 23:37:47 +0000]
make_ext.pl: fix operator precedence error from b4c079ca5

10 months agobase: fix a mis-statement in Changes file
Ricardo Signes [Sat, 21 Nov 2015 21:31:31 +0000]
base: fix a mis-statement in Changes file

10 months agorelease schedule: use a verbatim block for a list-y paragraph
Ricardo Signes [Sat, 21 Nov 2015 20:01:09 +0000]
release schedule: use a verbatim block for a list-y paragraph

10 months agoGet re/uniprops.t to pass on minitest
Karl Williamson [Sat, 21 Nov 2015 19:20:32 +0000]
Get re/uniprops.t to pass on minitest

locale handling doesn't work without POSIX module being able to load, so
doesn't work on minitest.  Prior to this patch, the code checked for
only one case of locale handling to skip when there was no POSIX, but
there was a 2nd case if failed to detect.

10 months agobase: new CPAN release
Ricardo Signes [Sat, 21 Nov 2015 18:16:47 +0000]
base: new CPAN release

10 months agoTick!
Abigail [Sat, 21 Nov 2015 16:55:07 +0000]
Tick!

10 months agoHash-Util-FieldHash version bump.
Jarkko Hietaniemi [Fri, 20 Nov 2015 21:53:26 +0000]
Hash-Util-FieldHash version bump.

10 months agoHUF_get_trigger0 unused since at least d74d639b
Jarkko Hietaniemi [Fri, 20 Nov 2015 21:43:19 +0000]
HUF_get_trigger0 unused since at least d74d639b

The 0b057af7 exposed this by static-ing it, so that compilers
started warning about the unused static function.

10 months ago[perl #116677] always deparse <> as either glob or readline
Lukas Mai [Sat, 21 Nov 2015 00:15:24 +0000]
[perl #116677] always deparse <> as either glob or readline

10 months ago[perl #124280] don't warn for 'my $foo, *bar'
Lukas Mai [Fri, 20 Nov 2015 23:16:31 +0000]
[perl #124280] don't warn for 'my $foo, *bar'

It doesn't make sense to warn 'Parentheses missing around "my" list' if
adding the parens causes a syntax error (you can't declare typeglobs).

But it does make sense to warn for 'local $foo, *bar' because typeglobs
can be localized.

Thus modify the heuristic to only warn for '*' if we're not lexically
declaring something.

10 months agoBump the perl version in various places for 5.23.6
Abigail [Fri, 20 Nov 2015 19:15:14 +0000]
Bump the perl version in various places for 5.23.6

10 months agoNew perldelta for 5.23.6
Abigail [Fri, 20 Nov 2015 18:00:09 +0000]
New perldelta for 5.23.6

10 months agoVarious tests: use centralized locale detection
Karl Williamson [Wed, 21 Oct 2015 03:40:36 +0000]
Various tests: use centralized locale detection

These tests were using individually defined heuristics to decide whether
to do locale testing or not.  However t/loc_tools.pl provides functions
that are more reliable and complete for determining this than the
hand-rolled ones in these tests.

10 months agolib/locale.pm: Fix so works on platforms without LC_CTYPE
Karl Williamson [Fri, 20 Nov 2015 02:57:11 +0000]
lib/locale.pm: Fix so works on platforms without LC_CTYPE

These may not actually exist in the wild, but it is better to be
general.

This also adds an XXX comment about future possibilities.

10 months agot/loc_tools.pl: Improve locale detection
Karl Williamson [Fri, 20 Nov 2015 02:53:27 +0000]
t/loc_tools.pl: Improve locale detection

This would return FALSE if the platform didn't have LC_CTYPE; it really
should test if the desired locale category is available, and return TRUE
if so, regardless of LC_CTYPE's presence.  Now it may be that all
platforms that have locales have LC_CTYPE, but it is better to not make
that assumption.

It also centralizes the tests for basic locale handling into one common
function called by the others.

10 months agot/loc_tools.pl: Fix incomplete description in comment
Karl Williamson [Fri, 20 Nov 2015 02:51:53 +0000]
t/loc_tools.pl: Fix incomplete description in comment

10 months agot/loc_tools.pl: White-space only
Karl Williamson [Fri, 30 Oct 2015 16:14:08 +0000]
t/loc_tools.pl: White-space only

10 months agoop.c: Fix typo in comment
Karl Williamson [Fri, 20 Nov 2015 17:10:31 +0000]
op.c: Fix typo in comment

This was very recently introduced, by
09681a13f07600082e5a7efbc9c978f62b24c502

10 months agoAdded epigraph for 5.23.5
Abigail [Fri, 20 Nov 2015 17:27:31 +0000]
Added epigraph for 5.23.5

10 months agoMerge branch 'release-5.23.5' into blead
Abigail [Fri, 20 Nov 2015 17:15:03 +0000]
Merge branch 'release-5.23.5' into blead

10 months agoperldelta: acknowledgements v5.23.5
Abigail [Fri, 20 Nov 2015 13:40:19 +0000]
perldelta: acknowledgements

10 months agoperlhist: added entry for 5.23.5
Abigail [Fri, 20 Nov 2015 13:35:01 +0000]
perlhist: added entry for 5.23.5

10 months agoperldelta: Modules and Pragmata
Abigail [Fri, 20 Nov 2015 13:29:33 +0000]
perldelta: Modules and Pragmata

10 months agohexfp: if HEXFP_NV the shift2 is unused
Jarkko Hietaniemi [Fri, 20 Nov 2015 01:47:49 +0000]
hexfp: if HEXFP_NV the shift2 is unused

HEXFP_NV happens with e.g. -Duselongdouble.  The unusedness is
warned about by e.g. gcc/g++ at least since 4.6, but seemingly
not by clang, even at 3.7.

10 months agoUpdate Module::CoreList for 5.23.5
Abigail [Fri, 20 Nov 2015 10:34:11 +0000]
Update Module::CoreList for 5.23.5

10 months agoUpdate Porting/Maintainers.pl for previous commit
Steve Hay [Fri, 20 Nov 2015 08:16:28 +0000]
Update Porting/Maintainers.pl for previous commit

(Customization is incorporated into the new CPAN release.)

10 months agoUpdate Module-Metadata to version 1.000030
Karen Etheridge [Fri, 20 Nov 2015 03:15:18 +0000]
Update Module-Metadata to version 1.000030

1.000030  2015-11-20 03:03:24Z (TRIAL RELEASE)
  - temp dirs cleaned up during tests (Steve Hay)
  - more accurately mark tests as TODO, so as to have a quieter and less
    confusing test run without passing TODO tests. This release is primarily
    intended for the perl 5.23.5 release.

10 months agopod/perldelta: Add punctuation
Abigail [Thu, 19 Nov 2015 18:14:35 +0000]
pod/perldelta: Add punctuation

10 months agoop_free(): don't assert op_private ok when erred
David Mitchell [Thu, 19 Nov 2015 15:49:03 +0000]
op_free(): don't assert op_private ok when erred

[perl #126258]

op_free includes an assert to make sure each op's op_private field
only has the flags set that are expected for that op. It's a thing I added
at the same time I added the regen/op_private mechanism, and is more a
general "make sure people are only setting the flags we know about" test.

However, if the op tree is being freed after a compilation error, some
op's flags may be in an inconsistent state; so skip the assert in that case.

e.g.

    perl -e 'grep$0,0}'

10 months agoTurn new diag message into a link.
Abigail [Thu, 19 Nov 2015 15:46:45 +0000]
Turn new diag message into a link.

This turned out to be a bit of work. When running the porting tests,
the pod checker checks whether all links are valid. To do so, it
creates a database of all valid targets, and then for all links,
checks whether the target exist. However, the targets have been POD
processed, while the links have not -- except for E<sol> and
E<verbar>. So, if the target contains S<>, the link *must not*.

10 months agopodcheck.t: Deindent by 1 space to left margin.
Abigail [Thu, 19 Nov 2015 12:32:00 +0000]
podcheck.t: Deindent by 1 space to left margin.

10 months agopodcheck.t: Line up comments.
Abigail [Thu, 19 Nov 2015 12:28:57 +0000]
podcheck.t: Line up comments.

A block with all comments lined up, save for one? That hurts the eyes.

10 months agopodcheck.t: Fix typos in comments
Abigail [Thu, 19 Nov 2015 12:28:35 +0000]
podcheck.t: Fix typos in comments

10 months agoQuell "used only once" warning in test suite
Aaron Crane [Thu, 19 Nov 2015 11:29:24 +0000]
Quell "used only once" warning in test suite

10 months agomaint policy: allow fixes for new features
Aaron Crane [Sun, 15 Nov 2015 15:10:54 +0000]
maint policy: allow fixes for new features

As (Steve Hay)++ said on-list:

    I think it's a shame if users get a 5.X.0 release with some shiny new
    features which I hope they're excited about and start making use of,
    but then find bugs in those features and have to wait until 5.X+2.0,
    rather than 5.X.1, for fixes.

rjbs++ clarified that, although such changes constitute a break in backwards
compatibility, there are two reasons why bugs in new features should be
fixed in maint:

    * they delay people from using a feature for a year, because it does
      something stupid

    * they risk enshrining bad behavior under the usual program of bugward
      compatibility

10 months agosilence compiler warnings using INCMARK/POPMARK
David Mitchell [Thu, 19 Nov 2015 10:03:04 +0000]
silence compiler warnings using INCMARK/POPMARK

v5.23.3-305-g6cae08a introduced debugging variants of INCMARK/POPMARK, and
replaced a number of "PL_markstack_ptr--;" with "POPMARK;" etc.

This spews a bunch of "value computed is not used" warnings; so
add some "(void)"s.

Also indent the new definitions of INCMARK/POPMARK correctly.

10 months agoInitial work for pod/erldelta for 5.23.5
Abigail [Fri, 13 Nov 2015 22:46:02 +0000]
Initial work for pod/erldelta for 5.23.5

Processed commits up to cb4547118517bc34243bbb938eeccfbd8016c64a.

Work that still needs to be done on perldelta before 5.23.5:
  -  New/Changed/Delete modules
  -  Acknowledgements
  -  And commits after cb4547118517bc34243bbb938eeccfbd8016c64a.
  -  Get the L<> link to work referring to a new diagnostic message.

10 months agosv_inc/dec_nomg(): croak on GVs etc
David Mitchell [Wed, 18 Nov 2015 15:30:37 +0000]
sv_inc/dec_nomg(): croak on GVs etc

RT #126637

Prior to v5.23.4-100-g20e9643, pp_postinc() etc checked whether the SV was
an AV, non-fake GV, or any other such non-modifiable thinngy; and if so,
called Perl_croak_no_modify(). That commit of mine removed the check, and
instead relied on the underlying sv_inc_nomg() function (called by
pp_postinc()) to do the checking instead. It turns out that while
sv_inc_nomg() etc does some checks, it wasn't as thorough as pp_postinc().
So something like

    my $y = $_++ for *FOO;

now crashes with an assertion failure in sv_inc_nomg() rather than
croaking.

This commit adds such checks to sv_inc_nomg() and sv_dec_nomg() - which
need them anyway, since they may be called from places other than
pp_postinc() etc.

10 months agoUpdate Config::Perl::V to 0.25
H.Merijn Brand [Wed, 18 Nov 2015 08:19:08 +0000]
Update Config::Perl::V to 0.25

10 months agotru64: Freshen the section on required process data size limits.
Jarkko Hietaniemi [Tue, 17 Nov 2015 02:00:27 +0000]
tru64: Freshen the section on required process data size limits.

10 months agoUpgrade Time::HiRes from version 1.9727_02 to 1.9728
Steve Hay [Tue, 17 Nov 2015 13:24:36 +0000]
Upgrade Time::HiRes from version 1.9727_02 to 1.9728

10 months agoUpgrade Math::BigInt::FastCalc from version 0.35 to 0.37
Steve Hay [Tue, 17 Nov 2015 13:21:11 +0000]
Upgrade Math::BigInt::FastCalc from version 0.35 to 0.37

10 months agoUpgrade Math::BigInt from version 1.999707 to 1.999710
Steve Hay [Tue, 17 Nov 2015 13:18:20 +0000]
Upgrade Math::BigInt from version 1.999707 to 1.999710

10 months ago[perl #126193] don't use a trash SV if no indexes are provided to pp_lslice
Tony Cook [Wed, 11 Nov 2015 04:57:09 +0000]
[perl #126193] don't use a trash SV if no indexes are provided to pp_lslice

10 months ago[perl #126602] make sure targ isn't freed from under us
Tony Cook [Wed, 11 Nov 2015 00:29:03 +0000]
[perl #126602] make sure targ isn't freed from under us

10 months agohandle empty directory lists in File::Find
Mohammed El-Afifi [Mon, 9 Nov 2015 21:49:08 +0000]
handle empty directory lists in File::Find

10 months agoConfigure: gcc 5 exists.
Jarkko Hietaniemi [Mon, 16 Nov 2015 00:45:06 +0000]
Configure: gcc 5 exists.

10 months agoUpdated Maintainers.pl for recent CPAN releases
Chris 'BinGOs' Williams [Sun, 15 Nov 2015 18:25:07 +0000]
Updated Maintainers.pl for recent CPAN releases

10 months agoAdd epigraph for 5.22.1-RC2
Steve Hay [Sun, 15 Nov 2015 15:26:53 +0000]
Add epigraph for 5.22.1-RC2

10 months agoPerl 5.22.1-RC2 today
Steve Hay [Sun, 15 Nov 2015 14:14:10 +0000]
Perl 5.22.1-RC2 today

10 months agoRevert "Module::CoreList updates for 5.22.1"
Steve Hay [Sun, 15 Nov 2015 13:45:01 +0000]
Revert "Module::CoreList updates for 5.22.1"

This reverts commit 89a7d40e89d2f2e64fe5421cb28d5a1f29bb25b4.

(The need for 5.22.1-RC2 means that 5.22.1 will not be released before
5.23.5 now, so remove the 5.22.1 data from here for now. I will
reinstate it after 5.23.5 has been released.)

10 months agoPATCH: [perl #126615] Segfault/assert fail with regex_sets
Karl Williamson [Sat, 14 Nov 2015 18:41:30 +0000]
PATCH: [perl #126615] Segfault/assert fail with regex_sets

This is a the result of where I added an assert to make sure that
where I didn't think a situation could happen, in fact couldn't.  But it
turns out it can happen when there is a syntax error, so raise an error,
instead of doing the assert.

10 months agoAppease the gods of stupid tests.
Jarkko Hietaniemi [Sat, 14 Nov 2015 16:25:38 +0000]
Appease the gods of stupid tests.

10 months agoSome more threads.xs comments, courtesy jhedden
Jarkko Hietaniemi [Sat, 7 Nov 2015 23:24:16 +0000]
Some more threads.xs comments, courtesy jhedden

10 months agoDocument S_ithread_run lock+unlock in more detail, courtesy davem
Jarkko Hietaniemi [Fri, 6 Nov 2015 21:59:54 +0000]
Document S_ithread_run lock+unlock in more detail, courtesy davem

10 months agoPathTools: new CPAN release
Ricardo Signes [Fri, 13 Nov 2015 23:44:16 +0000]
PathTools: new CPAN release

This release gets the latest blead code onto CPAN, but blead gets
the updated Makefile.PL, which fixes INSTALLDIRS to use site
on 5.12-ish and later.

10 months agobase: no longer works on v5.6, require v5.8
Ricardo Signes [Fri, 13 Nov 2015 16:27:56 +0000]
base: no longer works on v5.6, require v5.8

10 months agoInclude the error message on IPC::Open3 exec() failure
Dagfinn Ilmari Mannsåker [Tue, 10 Nov 2015 10:59:24 +0000]
Include the error message on IPC::Open3 exec() failure

10 months ago[perl #126325] don't read past the end of the source for pack [Hh]
Tony Cook [Thu, 22 Oct 2015 01:03:05 +0000]
[perl #126325] don't read past the end of the source for pack [Hh]

With a utf8 target but a non-utf8 source, pack Hh would read past the
end of the source when given a length, due to an incorrect condition.

10 months agofix symbian XS too, missed this when checking over the patch
Tony Cook [Tue, 10 Nov 2015 22:20:56 +0000]
fix symbian XS too, missed this when checking over the patch

Unfortunately the symbian SDK doesn't appear to be available anymore.

10 months agoXSLoader 0.21: use dl_find_symbol 3rd optional argument
Reini Urban [Fri, 13 Mar 2015 23:31:41 +0000]
XSLoader 0.21: use dl_find_symbol 3rd optional argument

to skip the worthless dl_last_error message