pugs.blogs.com | pugscode.org | pugs.kwiki.org | paste: sial.org/pbot/perl6 | <stevan> Moose... it's the new Camel ":P | .pmc == PPI source filters!
Set by Alias_ on 16 March 2006.
stevan audreyt: I just uploaded Moose::Autobox to CPAN 02:06
audreyt: it makes heavy use of roles to create the 4 basic Perl 5 "types" (Scalar, Array, Hash & Code)
it's very rough still, but it's fun to play with :) 02:07
obra seen audreyt 02:28
QtPlatypus hasn't seen audreyt today. 02:55
audreyt ooooh 03:27
stevan++ stevan++ stevan++ 03:28
Automoose
Alias_ audreyt: ping? 08:37
gaal nothingmuch: ping 09:23
nothingmuch gaal: pong 10:59
wolverian good morning 11:10
spinclad good $rollingearth 11:35
svnbot6 r10590 | audreyt++ | * array_cap.t: adjust test count. 14:34
jedi06 hello i need help with par 16:14
xerox Anybody using audrey's GHC bin dist for macintels? 17:28
The link the the GHC track is b0rked ): 17:29
jedi06 i try to install version .94 par from repository bribe.org and it looks like it upgrades when i type upgrade -install par 17:42
<jedi06> then i got to properties par and it says version .90
putter any parsey around? i'm puzzling over blocks... 17:59
err, s/any/anyone/
issue is...
putter oh, fribble. I get it. 18:04
ok, thinking aloud... 18:13
blocks... a block gets parsed, and if in retrospect it can be interpreted as hash literal, it is. a bit bizarre, but nicely local. 18:14
putter the trailing }... to get map:{3;}.foo to be (map:{3;}).foo, while f {3;},42; still works normally, the {block} term is unusual. 18:16
Alias_ rindolf: I'm here, but can't private message 18:17
putter one can deal with it in several ways. the '}' lexeme can establish a mutant post-term parser context which only accepts a few specially labeled ops ( , : => ). so hitting .foo creates a parse-fail/end-of-stream for the argument list expr. 18:20
putter or, if all the special ops nest, err, left, so the argument list can be wrapped up in a single parser reduce step, one can use a collaborative precedence. i think. maybe. but it seems an unfortunate restriction. 18:23
the gotcha with adding a mutant post-term state, is it doesnt deal with this: 18:25
f:{3;}.foo is (f:{3;}).foo, but simple {3;}.foo is just fine. I don't immediately know which way 2,'a',{3;}.foo,13 is supposed to go, but eh. the problem is, it looks like {3;} has to be treated differently in the two cases. ie, the parser has _non-local_ state, "I am/am-not parsing a (argument?) list". which seems to 18:30
reach down into core shift/reduce decisions.
so... what am I missing? is there a clean way to do this? 18:31
putter I (very) fuzzily _think_ you can do it as a after-the-parse rewrite step. maybe. but blech. 18:40
I bit of state (literally;) in the parser would be fine. I just didn't expect it. Hopefully I'm confused. 18:46
bjoern_ pugs.blogs.com/pugs/2006/06/possible_job_ta.html 21:10
sounds like great stuff! 21:11
fglock this looks like the perl6 idea of a run-time optimizer: psyco.sourceforge.net/introduction.html 22:46
webmind are there any javascript to perl6 encoders/compilers ? 22:55
fglock webmind: I don't think so - there are currently only perl6->x translators, and a few perl5->perl6 attempts 23:00
webmind ok, awell.. too bad 23:03
I had some horrid js code I wanted to read.. :)
fglock webmind: try a code formatter, like www.prettyprinter.de/ 23:12

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!