Parrot 0.4.17 Released

by chromatic

Will Coleda, Parrot project manager, has just announced the release of Parrot 0.4.17. Parrot's seen a lot of progress in the past month, with several features improved and a few reclaimed. In particular, the design documentations for several important subsystems have left draft status and are awaiting implementation.

We'll release Parrot 0.5.0 on 20 November 2007, with a full implementation of Parrot's object system. (It narrowly missed this release.) This should also give us much better speed, as well as the ability to make much further progress on implementing advanced languages including Perl 6 and Cardinal (Ruby on Parrot). I'm also preparing to check GC-debugging code that should let us track down memory errors very quickly and effectively.)

Next month's bug day/new contributor day is 17 November 2007, so if you're curious about the project and want to get involved, feel free to join #parrot on between now and then. There's plenty of work to do even if you don't know Perl or C.


Daniel Berger
2007-10-18 11:58:38
Will 0.5.0 build on Solaris? There's a (known) issue with the 0.4.x branch.

I'd like to tinker with it. :)

2007-10-18 12:13:30
@Daniel, that depends on which version of Solaris and which compiler you use. GCC should be pretty solid. We've done some work to get Sun Studio to work better too. If you know someone with a Solaris box and can get some warnings and errors from it, we can fix those up for 0.5.0.

Paul T. Cochrane will likely add some coding standards tests so that we'll catch the kinds of constructs that are likely to break the build there too.

2007-10-22 17:07:59
Is there any hope for perl 6 alpha for Christmas?
2007-10-22 18:11:18
@dextius, it really depends on two things. First, Patrick has one update to make to the Parrot Compiler Tools (to add namespace support), at which point we can port over the existing Perl 6 implementation to use NQP instead of PIR. (This will simplify writing Perl 6, in that we can use a cut-down version of Perl 6 to write the compiler instead of Parrot's high-level assembly language.)

Second, there can't be any really big surprises in what Parrot can do, or any weird bugs that are difficult to fix. I've been trying to find and remove all of the weird bugs that might get in the way.