It was relatively quiet in the aftermath of the release of Perl 5.8.1.
Unfortunately, the first bugs in 5.8.1 were found also. Some annoying, such as a memory leak when initializing a shared array:
http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2003-09/msg01615.html
and, more seriously, a bug with random numbers after fork() which basically cause each child to get the same random numbers as the parent:
http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2003-10/msg00037.html
Fortunately, the upcoming mod_perl release 1.29, will take this bug into account when running with 5.8.1. More bugs will no doubt be found and more fixes will be made in the coming weeks.
Johan Vromans found some warnings when installing 5.8.1 RPM's that indicated dependencies of core modules on non-core modules.
http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2003-09/msg01644.html
This thread turned into a discussion whether Perl should / could be responsible for packaging on different platforms.
http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2003-10/msg00021.html
Some consensus was reached in that it was felt to be a good idea to supply a META.yml file with each Perl distribution, which would allow packagers to more easily package.
And then Ken Williams realized there suddenly was a META.yml in bleadperl. With incorrect information. But that all got sorted out:
http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2003-10/msg00088.html
Hugo released a snapshot and fires up bleadperl development again!
http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2003-10/msg00086.html
A flurry of documentation patches started when Shlomi Fish returned to p5p:
http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2003-10/msg00055.html
As a side note, Shlomi also started some action to refurbish learn.perl.org on the Perl Advocacy list:
http://www.mail-archive.com/advocacy%40perl.org/msg01816.html
Michael Schwern found an(other) difference between 5.8.0 and 5.8.1. This time it involves configuration on 64bit platforms. Previously, some configurations would silently fall back. In 5.8.1, they exit with an error.
http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2003-10/msg00059.html
Jos Boumans reported a bug in the handling of attributes with our and my. Richard Clamp explained why my attributes are handled at execution time and our at compile time. But there may still be something amiss:
http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2003-10/msg00105.html
Abigail added TODO functionality to one of the core-test internal dialects of Test::xxx.
Robin Barker found an oddity when using perl -V:'?flags'. The ? makes it possible to search for any string, rather than the ones anchored to the beginning (because it makes the /^ optional).
This week's summary was produced by Elizabeth Mattijsen. Summaries are published weekly on http://use.perl.org/ and on a mailing list, which subscription address is perl5-summary-subscribe@perl.org. Corrections and comments are welcome. Next week Rafael will be taking the honours again.