Stories
Slash Boxes
Comments
NOTE: use Perl; is on undef hiatus. You can read content, but you can't post it. More info will be forthcoming forthcomingly.

All the Perl that's Practical to Extract and Report

use Perl Log In

Log In

[ Create a new account ]

autrijus (1505)

autrijus
  (email not shown publicly)
http://pugs.blogs.com/

Journal of autrijus (1505)

Tuesday July 12, 2005
09:23 AM

Day 159: Night before 6.2.8. A plan!

[ #25653 ]
The pre-release bughunting fest went all day long, amassing over 80 commits and still counting upward. PIR now passes all sanity tests again; the main test suite is down to 2 failed test files.

Aankhen, Stevan, bsmith, dudley, fglock, gaal, geoffb, iblech, nothingmuch, putter, rafl all chimed in with bugfixes and more test cases.

putter interviewed me about the imminent, PIL-based "Pugs as the Compiler" change, and put up a writeup in docs/notes/plan which received lots of editing help from geoffb. If you were wondering what's the Next Big Thing for Pugs development, please check it out -- I think I'll merge it with my (rejected) hw2005 paper to produce the much delayed Pugs Apocryphon 2...

The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
 Full
 Abbreviated
 Hidden
More | Login | Reply
Loading... please wait.
  • This is very cool! Is there any plan to do a PIL-to-Cpp (or C) emitter? Native compilation would be the killer feature in Perl 6 (at least for me). Excuse if the question is dumb! No idea about how XS is being dealt with in Perl 6...
    • Parrot can generate a native executate from PIR, so once Pugs compiling to Parrot is in place, generating native executables is pretty much there.

      As for XS in Perl 6, as far as running Perl 6 on Parrot goes, Parrot has a native calling interface to easily call into C libraries. So maybe we'll not have XS style things, but simply write a PIR module to directly interface with the required C library.
    • It's not a dumb question; it's entirely valid. However, we probably will not have the resource to directly compile to runtime-less C code, since that will mean maintaining a set of primitive functions.

      However if indeed we can target Haskell with PIL, then it is conceivable to generate runtime-less ANSI C via Jhc [repetae.net], an experimental Haskell compiler. This is in its early speculative stage, though.