This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
perl5.git
11 months agot/porting/readme.t: TODO tests now passing on EBCDIC
Karl Williamson [Sun, 21 Jun 2020 17:57:55 +0000 (11:57 -0600)]
t/porting/readme.t: TODO tests now passing on EBCDIC

11 months agoop/infnan.t: Todo tests are now passing on EBCDIC
Karl Williamson [Sun, 21 Jun 2020 17:50:34 +0000 (11:50 -0600)]
op/infnan.t: Todo tests are now passing on EBCDIC

11 months agoFix uninitialized fold table for EBCDIC with no locales
Karl Williamson [Sun, 21 Jun 2020 17:54:04 +0000 (11:54 -0600)]
Fix uninitialized fold table for EBCDIC with no locales

The C locale table for EBCDIC platforms was not getting initialized at
compile time, but only not for EBCDIC platforms.  It normally gets
changed when a new locale is switched to, but if that doesn't happen, it
doesn't get initialized at all.  IBM on z/OS has not implemented locales
with threads, so on threaded perls, the table was never getting
initialized.

11 months agoregen/ebcdic.pl: Declare tables with correct size
Karl Williamson [Mon, 22 Jun 2020 14:09:14 +0000 (08:09 -0600)]
regen/ebcdic.pl: Declare tables with correct size

We know the size, might as well include it.

11 months agoregen/ebcdic.pl: Add locale-related ability
Karl Williamson [Mon, 22 Jun 2020 14:04:35 +0000 (08:04 -0600)]
regen/ebcdic.pl: Add locale-related ability

A locale-related table won't be const; that is the only difference from
other ones.

11 months agoregexec.c: Don't use sizeof()
Karl Williamson [Mon, 22 Jun 2020 14:15:51 +0000 (08:15 -0600)]
regexec.c: Don't use sizeof()

A future commit will change this array so that its size isn't known at
compilation time.

11 months agoRemove PL_freq
Karl Williamson [Sat, 20 Jun 2020 22:22:13 +0000 (16:22 -0600)]
Remove PL_freq

This hasn't actually been used since commit
8922e4382e9c1488fdbe46a0f52493860dc897a6.

11 months agoRemove EBCDIC PL_freq
Karl Williamson [Sat, 20 Jun 2020 22:15:49 +0000 (16:15 -0600)]
Remove EBCDIC PL_freq

Instead, do a translation of the ASCII one at runtime.  This is because
the table doesn't account for the various EBCDIC flavors, so isn't
actually accurate.

11 months agoperl.h: White-space only
Karl Williamson [Sat, 20 Jun 2020 17:02:20 +0000 (11:02 -0600)]
perl.h: White-space only

Properly indent some nested pre-processor lines

11 months agoext/File-Glob/t/basic.t: Check error returns
Karl Williamson [Fri, 29 May 2020 15:55:45 +0000 (09:55 -0600)]
ext/File-Glob/t/basic.t: Check error returns

This test was blindly assuming various system calls succeed.
Permissions problems, for one, could cause failures.

11 months agot/test.pl: Display an ESC control char as \e
Karl Williamson [Sat, 14 Dec 2019 20:46:51 +0000 (13:46 -0700)]
t/test.pl: Display an ESC control char as \e

11 months agot/loc_tools.pl: Don't test unsafe locales
Karl Williamson [Sun, 14 Jun 2020 14:10:35 +0000 (08:10 -0600)]
t/loc_tools.pl: Don't test unsafe locales

Under threads, locales may or may not be safe to use.  Skip the testing
of them when unsafe.

11 months agoAdd z/OS locale categories
Karl Williamson [Sat, 13 Jun 2020 20:28:50 +0000 (14:28 -0600)]
Add z/OS locale categories

z/OS has two locale categories, LC_SYNTAX and LC_TOD,  not found outside
IBM products.  This makes Perl know about them, so that a program can
refer to them, but like other similar categories found on other OS's,
nothing more is done with them.

11 months agoembed.fnc: Mark some internal functions as core
Karl Williamson [Fri, 17 Jul 2020 01:32:15 +0000 (19:32 -0600)]
embed.fnc: Mark some internal functions as core

These are documented in perlguts as for internal use, but they were
listed here as public API

11 months agoembed.fnc: utf8 foo_msgs fcns no longer experimental
Karl Williamson [Fri, 17 Jul 2020 01:17:49 +0000 (19:17 -0600)]
embed.fnc: utf8 foo_msgs fcns no longer experimental

I'm now comfortable that these functions that I wrote are stable, and so
don't need to be marked experimental

11 months agoembed.fnc: Add info to a comment
Karl Williamson [Fri, 17 Jul 2020 00:16:29 +0000 (18:16 -0600)]
embed.fnc: Add info to a comment

11 months agoembed.fnc: Fix typo in comment
Karl Williamson [Fri, 17 Jul 2020 00:15:41 +0000 (18:15 -0600)]
embed.fnc: Fix typo in comment

11 months agoembed.fnc: Reorder some comments
Karl Williamson [Fri, 17 Jul 2020 00:07:54 +0000 (18:07 -0600)]
embed.fnc: Reorder some comments

A block of text is moved to after something that is helpful to have read
first.

11 months agoMake SvTRUEx identical to SvTRUE
Karl Williamson [Sun, 14 Jun 2020 18:30:56 +0000 (12:30 -0600)]
Make SvTRUEx identical to SvTRUE

There is no longer any need for the 'x' version, as SvTRUE evaluates its
argument just once since 5.32.

11 months agoUpdate pod for SvTRUE, to indicate single param evaluation
Karl Williamson [Sun, 14 Jun 2020 18:26:02 +0000 (12:26 -0600)]
Update pod for SvTRUE, to indicate single param evaluation

5.32 changed this macro into an inline function so that 'sv' only gets
evaluated once, but didn't update the documentation to reflect that.

11 months agosv.h: Wanted UOK, but said IOK
Karl Williamson [Sun, 28 Jun 2020 18:03:54 +0000 (12:03 -0600)]
sv.h: Wanted UOK, but said IOK

I don't know the consequences of this bug

11 months agoMove restricted-use api items to perlintern
Karl Williamson [Thu, 18 Jun 2020 01:06:52 +0000 (19:06 -0600)]
Move restricted-use api items to perlintern

A few currently API items say they are not for general use.  So move
them to perlintern.

11 months agoMove api items to perlintern where macro should instead be used
Karl Williamson [Thu, 18 Jun 2020 00:54:06 +0000 (18:54 -0600)]
Move api items to perlintern where macro should instead be used

There are some items in perlapi whose entries say "don't use this;
instead use the macro equivalent".  Such items really shouldn't be in
perlapi, but instead be in perlintern.

11 months agosv.h: Change formal parameter name to reflect its usage
Karl Williamson [Sun, 14 Jun 2020 00:44:32 +0000 (18:44 -0600)]
sv.h: Change formal parameter name to reflect its usage

The formal parameter in these macros is 'lp', which elsewhere is used to
signifiy a pointer to length.  But instead here it isn't a pointer, but
the plain length

11 months agoMUTABLE_PTR() Rmv non-standard syntax
Karl Williamson [Mon, 29 Jun 2020 15:21:24 +0000 (09:21 -0600)]
MUTABLE_PTR() Rmv non-standard syntax

Variables in C are beginning with an underscore are reserved for use by
the C implementation.  Change this non-conformant usage.

11 months agoautodoc.pl: varargs fcns require full 'Perl_foo'
Karl Williamson [Fri, 17 Jul 2020 00:04:24 +0000 (18:04 -0600)]
autodoc.pl: varargs fcns require full 'Perl_foo'

Prior to this commit, perlapi indicated that you could say foo(...),
when in fact that doesn't work for these functions, which have to be
called as Perl_foo(aTHX_ ...)

11 months agoXS-APItest: Silence a compiler warning
Karl Williamson [Sat, 18 Jul 2020 03:18:37 +0000 (21:18 -0600)]
XS-APItest: Silence a compiler warning

This gives an unused variable warning on 32 bit platforms

11 months agotime64: declare `tm_zone` as `const char*` unconditionally
Dagfinn Ilmari Mannsåker [Tue, 16 Jun 2020 12:47:35 +0000 (13:47 +0100)]
time64: declare `tm_zone` as `const char*` unconditionally

We only ever assign from `struct tm` to `struct TM`, not the other way
around, so making this const regardless of what `struct tm` has should
be safe.

11 months agofix 5.32.0 release date; remove 5.31.x releases already complete, duplicate section
Karen Etheridge [Fri, 17 Jul 2020 22:01:03 +0000 (15:01 -0700)]
fix 5.32.0 release date; remove 5.31.x releases already complete, duplicate section

11 months agoRe-run Porting/makemeta
James E Keenan [Fri, 17 Jul 2020 22:02:16 +0000 (22:02 +0000)]
Re-run Porting/makemeta

When I was running 'make test_porting', I got a failure in porting/regen.t.

    $ cd t;./perl harness -v porting/regen.t; cd -

    ok 1 - generated perly.act is up to date
    ...
    ok - regen/feature.pl feature.h
    not ok - Porting/makemeta META.json
    'META.json-new' and 'META.json' differ
    ok - Porting/makemeta META.yml
    Failed 1/45 subtests

    Test Summary Report
    -------------------
    porting/regen.t (Wstat: 0 Tests: 45 Failed: 1)
      Failed test:  44

This was alleviated by re-running: ./perl -Ilib Porting/makemeta, which
brought the content of META.json up-to-date.

11 months agoMerge branch 'blead' of github.com:Perl/perl5 into blead
Sawyer X [Fri, 17 Jul 2020 21:54:57 +0000 (00:54 +0300)]
Merge branch 'blead' of github.com:Perl/perl5 into blead

11 months agodist/IO no indirect, strict & warnings
Nicolas R [Tue, 23 Jun 2020 22:57:14 +0000 (16:57 -0600)]
dist/IO no indirect, strict & warnings

Make sure dist/IO is not using indirect
calls.

11 months agoUpdate Socket from v2.029 to v2.030
Nicolas R [Tue, 7 Jul 2020 20:15:16 +0000 (14:15 -0600)]
Update Socket from v2.029 to v2.030

Match recent release on CPAN

11 months agoUpdate JSON-PP from 4.04 to 4.05
Nicolas R [Wed, 8 Jul 2020 21:57:58 +0000 (15:57 -0600)]
Update JSON-PP from 4.04 to 4.05

11 months agoUpdate IPC-SysV from 2.07 to 2.08
Nicolas R [Tue, 7 Jul 2020 20:24:45 +0000 (14:24 -0600)]
Update IPC-SysV from 2.07 to 2.08

11 months agoBump the perl version in various places for 5.33.1
Sawyer X [Fri, 17 Jul 2020 21:32:05 +0000 (00:32 +0300)]
Bump the perl version in various places for 5.33.1

11 months agonew perldelta for 5.33.1
Karen Etheridge [Fri, 17 Jul 2020 20:41:18 +0000 (13:41 -0700)]
new perldelta for 5.33.1

11 months agocorrupt.t: Correct spelling error
James E Keenan [Fri, 17 Jul 2020 20:36:14 +0000 (20:36 +0000)]
corrupt.t: Correct spelling error

Discovered when file was run with warnings as part of Perl 7 research.

For: https://github.com/Perl/perl5/issues/17970

11 months agoTime-HiRes: Use core version compare
Karl Williamson [Fri, 17 Jul 2020 20:29:31 +0000 (14:29 -0600)]
Time-HiRes: Use core version compare

This had its own copy of some macros that are now in core.

11 months agoBump the perl version in various paces for 5.33.1
Sawyer X [Fri, 17 Jul 2020 20:25:16 +0000 (23:25 +0300)]
Bump the perl version in various paces for 5.33.1

11 months agoHTTP::Tiny needs two modules for https support
James E Keenan [Thu, 9 Apr 2020 20:26:07 +0000 (20:26 +0000)]
HTTP::Tiny needs two modules for https support

When running sync-with-cpan on a machine for the first time, internal
subroutine wget() was silently failing.  After adding debugging code and
running through the debugger, the HTTP::Tiny mirror() method call
displayed this error message:

  IO::Socket::SSL 1.42 must be installed for https support
  Net::SSLeay 1.49 must be installed for https support

11 months agoDumpvalue.t: Tighten scope of package variables.
James E Keenan [Sun, 5 Jul 2020 21:51:29 +0000 (17:51 -0400)]
Dumpvalue.t:  Tighten scope of package variables.

Rename some variables for clarity.

In one place, get() is returning a list, so we should test for contents of
list (via Test::More::is_deeply) rather than for string equality.

Signed-off-by: James E Keenan <jkeenan@cpan.org>
11 months agogroups.t: Correct one syntax error which was warning
James E Keenan [Fri, 3 Jul 2020 14:32:25 +0000 (10:32 -0400)]
groups.t:  Correct one syntax error which was warning

When run with warnings, this file would get:

        Scalar value @sup_group[2] better written as $sup_group[2]

Using '@' rather than '$' was probably unintentional.  Let's fix it.

11 months agoAdd PERL_VERSION_GE macro, and kin
Karl Williamson [Wed, 15 Jul 2020 21:11:47 +0000 (15:11 -0600)]
Add PERL_VERSION_GE macro, and kin

This moves things from vutil.h, and into handy.h; this is required for
v7

11 months agonew perldelta for 5.33.1
Sawyer X [Fri, 17 Jul 2020 19:57:34 +0000 (22:57 +0300)]
new perldelta for 5.33.1

11 months agoUpdate release schedules, as muchas I can
Sawyer X [Fri, 17 Jul 2020 19:52:45 +0000 (22:52 +0300)]
Update release schedules, as muchas I can

11 months agoUpdate index() documentation with some code examples
Scott Baker [Tue, 30 Jun 2020 18:08:03 +0000 (11:08 -0700)]
Update index() documentation with some code examples

11 months agoAdd Scott Baker to AUTHORS
Karl Williamson [Fri, 17 Jul 2020 19:42:09 +0000 (13:42 -0600)]
Add Scott Baker to AUTHORS

11 months agoupdate epigraph
Sawyer X [Fri, 17 Jul 2020 19:44:08 +0000 (22:44 +0300)]
update epigraph

11 months agosmall version number orrection v5.33.0
Sawyer X [Fri, 17 Jul 2020 19:27:02 +0000 (22:27 +0300)]
small version number orrection

11 months agoadd new release to perlhist
Sawyer X [Fri, 17 Jul 2020 10:35:25 +0000 (13:35 +0300)]
add new release to perlhist

11 months agoUpdate perldelta for 5.33.0
Sawyer X [Fri, 17 Jul 2020 10:34:37 +0000 (13:34 +0300)]
Update perldelta for 5.33.0

11 months agoUpdate Module::CoreList for 5.33.0
Sawyer X [Fri, 17 Jul 2020 10:12:05 +0000 (13:12 +0300)]
Update Module::CoreList for 5.33.0

11 months agohv.c: Remove obsolete/confusing constant 17291/head
Eric Herman [Fri, 10 Jul 2020 07:07:42 +0000 (09:07 +0200)]
hv.c: Remove obsolete/confusing constant

The HV_FILL_THRESHOLD is no longer used and is misleading.

See also:
commit 8bf4c4010cc474d4000c2a8c78f6890fa5f1e577
Date:   Mon Jun 20 22:51:38 2016 +0200
Change scalar(%hash) to be the same as 0+keys(%hash)

11 months agoMake perl compile under minix again
Leon Timmermans [Mon, 29 Jun 2020 21:52:31 +0000 (23:52 +0200)]
Make perl compile under minix again

Minix has a #define for si_signo to _info._signo, despite not having
a struct siginfo; this caused compilation errors.

11 months agofix bad version number in INSTALL
Karen Etheridge [Sat, 27 Jun 2020 23:59:06 +0000 (16:59 -0700)]
fix bad version number in INSTALL

This appears to have started with a mistakenly-unapplied s/// in the 5.31.10
release that carried forward from there.

11 months agoperlpolicy: Adjust the list of moderators.
Andy Dougherty [Mon, 6 Jul 2020 16:20:10 +0000 (12:20 -0400)]
perlpolicy: Adjust the list of moderators.

11 months agoRemove redundant ternary in safesysmalloc
Dagfinn Ilmari Mannsåker [Wed, 1 Jul 2020 11:26:02 +0000 (12:26 +0100)]
Remove redundant ternary in safesysmalloc

Commit b001a0d149ed99df18916796f3a72b2c888b94d8 changed it to bumping
the size from 0 to 1 earlier in the function, but didn't remove the
then-redundant ternary that tried to avoid passing zero to
PerlMem_malloc().

11 months agoUpdate epigraphs
Sawyer X [Tue, 30 Jun 2020 18:18:58 +0000 (21:18 +0300)]
Update epigraphs

11 months agoBumps, perldeltas, pods
Sawyer X [Sat, 27 Jun 2020 23:26:45 +0000 (02:26 +0300)]
Bumps, perldeltas, pods

11 months agoBump to 5.33.0
Sawyer X [Sat, 27 Jun 2020 22:12:28 +0000 (01:12 +0300)]
Bump to 5.33.0

12 months agoTick the release
Sawyer X [Sat, 20 Jun 2020 21:15:46 +0000 (00:15 +0300)]
Tick the release

12 months agoadd new release to perlhist release-7.0.0 v5.32.0
Sawyer X [Sat, 20 Jun 2020 13:07:37 +0000 (16:07 +0300)]
add new release to perlhist

12 months agoUpdate perldelta
Sawyer X [Sat, 20 Jun 2020 12:44:51 +0000 (15:44 +0300)]
Update perldelta

12 months agoskip the failing hexfloat test if it fails on NetBSD
Tony Cook [Wed, 17 Jun 2020 02:07:19 +0000 (02:07 +0000)]
skip the failing hexfloat test if it fails on NetBSD

perl (in theory) supports C89, where expl() doesn't exist (though
long doubles do), so our tests shouldn't fail if expl() isn't
available.

Of course, the NetBSD libm does supply an implementation of expl(),
it's just not a long double implementation.

12 months agowarn the user that NetBSD's long double support isn't very good
Tony Cook [Wed, 17 Jun 2020 02:06:51 +0000 (02:06 +0000)]
warn the user that NetBSD's long double support isn't very good

12 months agoop.h: Add additional padding to struct opslab to ensure proper alignment
John Paul Adrian Glaubitz [Fri, 19 Jun 2020 14:40:38 +0000 (16:40 +0200)]
op.h: Add additional padding to struct opslab to ensure proper alignment

On m68k, the natural alignment is 16 bits which causes the opslab_opslot
member of struct opslab to be aligned at a 16-bit offset. Other 32-bit
and 64-bit architectures have a natural alignment of at least 32 bits, so
the offset is always guaranteed to be at least 32-bit-aligned.

Fix this by adding additional padding bytes before the opslab_opslot
member, both for cases when PERL_DEBUG_READONLY_OPS defined and not
defined to ensure the offset of oplab_slots is always 32-bit-aligned.
On architectures which have a natural alignment of at least 32 bits,
the padding does not affect the alignment, offsets or struct size.

12 months ago[Doc] GH #17868: Fix a few things on perlcommunity:
Sawyer X [Sat, 20 Jun 2020 08:14:32 +0000 (11:14 +0300)]
[Doc] GH #17868: Fix a few things on perlcommunity:

* YAPC was rebranded.
* OSCON and OSDC are discontinued.
* The Perl Review calendar link doesn't work.

12 months agoSmall fixup to explain conference names
Sawyer X [Sat, 20 Jun 2020 08:01:44 +0000 (11:01 +0300)]
Small fixup to explain conference names

12 months agoperldelta: Update advice for z/OS
Karl Williamson [Fri, 19 Jun 2020 00:29:32 +0000 (18:29 -0600)]
perldelta: Update advice for z/OS

12 months agoperldelta: minor pod fix
SHIRAKATA Kentaro [Thu, 18 Jun 2020 18:25:06 +0000 (03:25 +0900)]
perldelta: minor pod fix

12 months agoFix cygwin Perl_utf16_to_utf8 parameter types
Dagfinn Ilmari Mannsåker [Wed, 17 Jun 2020 14:08:06 +0000 (15:08 +0100)]
Fix cygwin Perl_utf16_to_utf8 parameter types

Commit f46dcac2d383af532d3e8c014488fba83a514309 changed the I32
arguments of Perl_utf16_to_utf8() to Size_t and updated all callers
that called it via the utf16_to_utf8 macro, but missed the
fully-qualified call in cygwin.c.

12 months agoperlhist: Fix typo in maint-5.32 description
Dagfinn Ilmari Mannsåker [Mon, 15 Jun 2020 10:37:07 +0000 (11:37 +0100)]
perlhist: Fix typo in maint-5.32 description

12 months agoperldelta: add shorter heading for POSIX::mblen() etc.
SHIRAKATA Kentaro [Sat, 13 Jun 2020 16:52:38 +0000 (01:52 +0900)]
perldelta: add shorter heading for POSIX::mblen() etc.

12 months agoAdd Thibault Duponchelle to AUTHORS file
Thibault DUPONCHELLE [Tue, 9 Jun 2020 08:48:38 +0000 (10:48 +0200)]
Add Thibault Duponchelle to AUTHORS file

12 months agoMinor perldelta fix to link to the correct issue tracker numbers
Thibault DUPONCHELLE [Tue, 9 Jun 2020 07:49:18 +0000 (09:49 +0200)]
Minor perldelta fix to link to the correct issue tracker numbers

12 months agoRemove bogus 5.031012 release from Module::CoreList
Chris 'BinGOs' Williams [Tue, 9 Jun 2020 06:34:48 +0000 (07:34 +0100)]
Remove bogus 5.031012 release from Module::CoreList

12 months agoUpdate meta
Sawyer X [Sun, 7 Jun 2020 22:30:49 +0000 (01:30 +0300)]
Update meta

12 months agoDocument epigraph
Sawyer X [Sun, 7 Jun 2020 21:22:34 +0000 (00:22 +0300)]
Document epigraph

12 months agodisarm RC1 bump
Sawyer X [Sun, 7 Jun 2020 21:14:12 +0000 (00:14 +0300)]
disarm RC1 bump

12 months agoUpdate META files v5.32.0-RC1
Sawyer X [Sun, 7 Jun 2020 18:38:29 +0000 (21:38 +0300)]
Update META files

12 months agoBump version to RC1
Sawyer X [Sun, 7 Jun 2020 18:38:07 +0000 (21:38 +0300)]
Bump version to RC1

12 months agoadd new release to perlhist
Sawyer X [Sun, 7 Jun 2020 18:37:04 +0000 (21:37 +0300)]
add new release to perlhist

12 months agoUpdate perldelta
Sawyer X [Sun, 7 Jun 2020 18:36:56 +0000 (21:36 +0300)]
Update perldelta

12 months agoUpdate Module::CoreList for 5.32.0-RC2
Sawyer X [Sun, 7 Jun 2020 18:14:44 +0000 (21:14 +0300)]
Update Module::CoreList for 5.32.0-RC2

12 months agoAdd z/OS platform notes to 5.32 perldelta
Karl Williamson [Sun, 7 Jun 2020 14:25:14 +0000 (08:25 -0600)]
Add z/OS platform notes to 5.32 perldelta

12 months agoUpdate perldelta on security fixes:
Sawyer X [Sun, 7 Jun 2020 06:11:24 +0000 (09:11 +0300)]
Update perldelta on security fixes:

* CVE-2020-10543
* CVE-2020-10878
* CVE-2020-12723

They were fixed on v5.28.3 and 5.30.3 and the fixes were backported.

12 months agoAPItest/t/utf16_to_utf8.t: Skip on EBCDIC
Karl Williamson [Fri, 29 May 2020 15:54:37 +0000 (09:54 -0600)]
APItest/t/utf16_to_utf8.t: Skip on EBCDIC

How this should behave is not well defined.

12 months agoAPItest/t/svpv.t: Generalize for EBCDIC
Karl Williamson [Fri, 29 May 2020 14:51:46 +0000 (08:51 -0600)]
APItest/t/svpv.t: Generalize for EBCDIC

This had ASCII-centric tests, which have now been converted to work on
EBCDIC as well.

12 months agoGH #17826: Remove entire section, thanks @richardleach
Sawyer X [Mon, 1 Jun 2020 17:46:12 +0000 (20:46 +0300)]
GH #17826: Remove entire section, thanks @richardleach

12 months agoperlhacktips.pod - don't explicitly list supported ASan combinations
Richard Leach [Wed, 3 Jun 2020 21:28:14 +0000 (22:28 +0100)]
perlhacktips.pod - don't explicitly list supported ASan combinations

As suggested in #16910

12 months agoCorrect warning categories
Smylers [Wed, 3 Jun 2020 04:54:08 +0000 (05:54 +0100)]
Correct warning categories

The 2 no-longer-needed warning categories were mentioned with each other's changes.

12 months agoperlhacktips.pod - update ASan section
Richard Leach [Tue, 2 Jun 2020 20:33:47 +0000 (21:33 +0100)]
perlhacktips.pod - update ASan section

Addresses #16910 plus other minor updates.

12 months agoTweak check_extended_testing job
Nicolas R [Tue, 2 Jun 2020 06:00:01 +0000 (00:00 -0600)]
Tweak check_extended_testing job

12 months agoAjust authors logic
Nicolas R [Tue, 2 Jun 2020 06:05:13 +0000 (00:05 -0600)]
Ajust authors logic

12 months agoSet CONFIGURE_ARGS env correctly for sanity check
Nicolas R [Fri, 22 May 2020 21:23:03 +0000 (15:23 -0600)]
Set CONFIGURE_ARGS env correctly for sanity check

also add a debug statement for viewing GitHub context

12 months agoMerge smoke informations to testsuite workflow
Nicolas R [Fri, 22 May 2020 20:24:17 +0000 (14:24 -0600)]
Merge smoke informations to testsuite workflow

Smoke informations should belong to the testsuite
and only display when run on the main repo.

Note: moved the 'if' condition in the run as a 'if:'
check, avoiding to start a VM when not needed.

12 months agoOnly smoke extra flavors when EXTENDED_TESTING=1
Nicolas R [Fri, 22 May 2020 20:23:51 +0000 (14:23 -0600)]
Only smoke extra flavors when EXTENDED_TESTING=1

This is avoiding using too much resources on fork repositories.

fixup

12 months agoRun a sanity check before running all jobs
Nicolas R [Fri, 22 May 2020 20:16:15 +0000 (14:16 -0600)]
Run a sanity check before running all jobs

Abort earlier if we cannot compile a simple
configuration on linux.

This is avoiding wasting resources and will come
back earlier on errors. Note that this is adding
the time of an extra runtime job as we are
waiting for the sanity_check workflow before starting
any other jobs.

12 months agoMerge all workflows as a single testsuite workflow
Nicolas R [Fri, 22 May 2020 20:02:41 +0000 (14:02 -0600)]
Merge all workflows as a single testsuite workflow

The existing smoke-* workflows are part of the testsuite.
By merging them together we can link them and add some
conditions to reduce resources usages.