Follows the fourth grants updates report.
- Port PyYAML to Perl -- Ingy döt Net
- Fixing Bugs in the Archive::Zip Perl Module -- Alan Haggai Alavi
Alan has finished adding support for archives with unicode filenames on Windows, and is now tackling the Perl side, making it possible to create archives with unicode filenames in them.
- Extending BSDPAN -- Colin Smith
- Test::Builder 2 -- Michael G Schwern
- Perl cross-compilation for linux and wince -- Vadim Konovalov
- Tcl/Tk access for Rakudo -- Vadim Konovalov
- Perl Survey -- Kieren Diment
KD has received the dumps from Skud, but has just begun a new job and is trying to get the last touches on a book, so those are taking priority right now.
- Mojo Documentation Project -- Sebastian Riedel
Sebastian is taking things easy on doctors orders after back surgery. His grant manager (Adrian Howard) agreed a hiatus with him and he'll be starting up again on March 14th.
- Embedding perl into C++ applications -- Leon Timmermans
Due to personal circumstances I haven't been able to work on my project for some time, but now I'm back. I've been trying to tackle the regular expression bindings. It turns out that is much harder than I had initially anticipated.
My initial plan was to wrap the Perl 5.10 regexp api, as described in 'perlreapi'. This turns out to be more difficult than initially hoped for. Mostly because the API that is exposed is very low level. This effectively means I would have to reimplement much of the m// and s/// operators (qr// is turns out to be easy). It should be feasible by re-using some code from perl itself, but it's not an easy route.
There are two alternatives: The first is to call the implementations (pp ops) of the operators, directly. This turns out to be more complicated than I had hoped for, and seems to require building a small OP code tree. This is possibly even harder and definitely messier.
The other option is to compile perl subs an execute those to do the matching. This is the easier way to do (and I already have a partial implementation), but also the slowest by far. This would really be the least preferred option.
So now I'm back at my initial plan. This is turning out to be more complicated than I hoped for :-|
In more positive news, I'll probably be giving a presentation about my project at the Dutch Perl Workshop (on March 6th).
- Moose docs -- Dave Rolsky
The first two milestones, Moose::Manual, and revising all the existing
recipes, are done and released to CPAN. Dave is working on the next milestone, which is writing several additional
recipes. He is about halfway through that right now. That leaves comprehensive API documentation as the last milestone to do, which he will be getting to some time in March.
More detailed news can be read in his blog
- Integrating Padre with Parrot and Rakudo -- Gabor Szabo