Re^2: Announce: Rakudo Star - a useful, usable, "early adopter" distribution of Perl 6

[moritz]

I expect most of the Perl 5 code reuse to happen by loading Perl 5 modules in Perl 6, not by using (semi)automatic translation.

A prototype for that already exists (search for blizkost), and it works for some frequently-used modules already (CGI.pm, Tk).

Parsing Perl 5 is fragile, and only works if you neglect certain things (like function prototypes). If you do that, you'll arrive add PPI (p5 code) or STD_P5 (p6 code).

Another thing to try is the MAD emitter that Larry hacked into perl5 core before the 5.10.0 release, which uses perl to parse Perl 5 code, and emits an XML-like description of the parse tree.

[Anonymous Monk]

Could you please provide some more info on STD_P5? Is it's purpose to allow you to write a minimal Perl 5 parser in Perl 6? Is STD_P5 still being actively developed? What's the Perl 6 developers' opinion on its value?

If STD_P5 can be used to parse a fairly large subset of Perl 5, then could that "subset-of-Perl-5" code be run from inside my Perl 6 program?

[moritz]

Could you please provide some more info on STD_P5? Is it's purpose to allow you to write a minimal Perl 5 parser in Perl 6?

It is meant to be a (mostly complete, as far as that works) Perl 5 parser in Perl 6.

Is STD_P5 still being actively developed?

Not currently, blocking on TimToady being on vacation first, then YAPC::EU and finally needing most of his Perl 6 tuits for answering questions from the other developers.

But it's not abandoned, in the sense that he plans to hack on it again, once he gets around to it.

What's the Perl 6 developers' opinion on its value?

I can't speak for the masses of the p6 developers, but I for one welcome our new p5 embracing overlords.