Hey everyone,

Following is the p5p (Perl 5 Porters) mailing list summary for the past week. Enjoy!

May 2st-9th

News and updates

Ricardo Signes released Perl 5.24.0! Perl 5.25.0 shortly followed.

Dave Mitchell provided 2nd grant report #125, #126, and #127. Dave also provided the monthly grant report for April.

Another highlight related to Coro, which given a simple patch provided below, can compile and work with Perl 5.22 and above.

Issues

New issues

• Perl #128052: threads 2.07 t/test.pl uses re is_regexp which required 5.10.
• Perl #128083: Silent encoding of filenames with UTF8 flag set.
• Perl #128086: SIGSEGV in hek_eq_pvn_flags.
• Perl #128093: Broken build of perl5.24.0-RC5 on Mac OS X.
• Perl #128095: Failed test dist/IO/t/cachepropagate-unix on OS X with Perl 5.22.
• Perl #128085: SIGSEGV in S_regmatch with S_study_chunk.
• Perl #128096: use open doesn't affect magic diamond operator.
• Perl #128100: EXECUTABLE_NAME and hardlinks.

Resolved issues

• Perl #127877: Concat with uninitialized vars.

Proposed patches

Aaron Crane provided a patch to Perl #127887 to include a warning in the documentation of sprintf regarding operations that produce numbers which cannot be represented by the underlying C layer.

Following an analysis from Hugo van der Sanden, Yves Orton provided a patch to Perl #128085.

Discussion

Following a recent blog post regarding the state of Coro and Perl 5.22, Dave Mitchell suggested reverting a commit relating to the compilation problem mentioned in the blog post. This was an unusual request considering 5.24.0 was already about to be released. A conversation continued trying to understand whether Coro could work without changing core and what original problem Coro worked around, since no test is available in Coro for the issue. All too quick, Nicholas Clark provided a patch to build and test Coro that did not require a single change to core, included no need to revert any commits.

In Perl #127855, Dave Mitchell provides a more in-depth analysis of SvGROW. A conversation ensued with Yves Orton shedding light on the interaction of growing SVs (whether SvGROW or sv_grow) and COW.

James E. Keenan had researched mandoc and the patch provided in Perl #128012, and recommends merging the patch. Another +1 from Karl Williamson.