This Week on perl5-porters (7-13 June 2004)

dev.perl.org
 

This Week on perl5-porters (7-13 June 2004)

This week, a small summary is better than no summary at all.

h2ph

Rafael Garcia-Suarez finds that h2ph can't process the newest C headers from the glibc, because they include inline functions (instead of good old macros). He forced a bit of heuristics in it to decode the less convoluted C inline functions that might be there. Your mileage may vary.

IO::File on windows

Roderich Schupp finds that IO::File fails on open a file that is found on another volume on Windows (bug #30132). The problem comes from the Cwd and File::Spec modules (and their handling of volumes in pathnames), on which Ken Williams is working. He released thereafter a new beta of Cwd.pm, on which Mark Mielke provided some performance considerations.

    http://groups.google.com/groups?selm=1CDC3096-BCF9-11D8-9A4F-000A95BD9874%40mathforum.org

DB_File fooled by substr

Ton Hospel finds (bug #30237) that the DB_File methods don't behave correctly when they're passed results of substr() in argument. Paul Marquess investigates.

    http://groups.google.com/groups?selm=rt-3.0.9-30237-90394.14.8512570678496%40perl.org

Obscure bug of the week

Ton Hospel (again) finds that a small snippet of code involving the locale pragma, regexp manipulation and loading the POSIX module at run-time dies with an insecure dependency error when run under -T. (bug #30068). Nobody commented on the possible causes of this regression.

    http://groups.google.com/groups?selm=rt-3.0.9-30068-89825.7.20442262108094%40perl.org

Data::Dumper deparsing indentation

Mathieu Arnold proposed a patch to improve the indentation of the output of Data::Dumper when it is configured to deparse code references (bug #30197).

Safe Panic

Perl panics when one tries to use split() on an utf8-string from your average Safe compartment, as Rusty Conover reported (bug #30258). This is probably due to perl trying to load some data behind the scenes, which is forbidden by Safe, so the operation is aborted, and this abortion doesn't end well for some unclear reason.

    http://groups.google.com/groups?selm=rt-3.0.9-30258-90458.7.13228983473336%40perl.org

.. optimized

The range .. operator is usually optimized in a foreach loop, so that the list is not constructed in place. However, for alphabetic ranges, and when the range is not the only thing in the foreach() argument, this optimization does not occur (bug #30123, reported by Michael G Schwern).

About this summary

This summary was written by Rafael Garcia-Suarez. Weekly summaries are published on http://use.perl.org/ and posted on a mailing list, which subscription address is perl5-summary-subscribe@perl.org. Comments and corrections welcome.