Perl 6 - the future is here, just unevenly distributed

IRC log for #perl11, 2013-09-26

| Channels | #perl11 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
04:03 tokuhirom cool
06:24 Will_the_Chill what do y'all think about STD_P5?
06:44 tokuhirom is it works well?
06:58 Will_the_Chill I've been talking to fglock about STD_P5
06:58 Will_the_Chill he says it works pretty good
06:58 Will_the_Chill still nothing but Pumpkin (P5P) /usr/bin/perl parses Perl
06:59 Will_the_Chill fglock says Perlito's parser and STD_P5 are both very close to Pumpkin Perl
06:59 Will_the_Chill like >90%
07:00 Will_the_Chill rurban says he is at P2 parsing 50% of Perl
07:00 Will_the_Chill personally I want low magic anyway so I don't need to compile 100% of Perl, I just need to compile the low magic bits and keep them compatible with the high magic parts running through the VM/interpreter
13:23 Will_the_Chill joined #perl11
13:55 bluescreen joined #perl11
15:03 stevan joined #perl11
15:54 rurban like gperl
16:14 Will_the_Chill rurban: how fast will gperl be?
16:24 rurban Will_the_Chill: Alrady said: gperl is 3x faster than p2 in fibonacci (simple function calls)
16:25 Will_the_Chill oh good.
16:25 Will_the_Chill but why do we need yet another new Perl project?
16:26 rurban It already runs fine a testsuite similar to mine
16:26 rurban because it's ofun, and different
16:27 rurban The gperl parser does more than mine, ~80%
16:27 rurban and it compiles to LLVM
16:32 Will_the_Chill LLVM is interesting!
16:32 rurban It's big but fast
16:32 Will_the_Chill right
16:33 Will_the_Chill where does gperl's speed come from?
16:33 Will_the_Chill is the speed due to LLVM?
16:33 Will_the_Chill does gperl require the use of a type system?
16:33 rurban statically typed only and optimizing LLVM compiler
16:33 Will_the_Chill cool
16:33 Will_the_Chill what about XS compatibility for gperl?
16:33 rurban so very simple
16:33 rurban no
16:34 rurban p2 in opposite is dynamically typed only, I will add static types later
16:34 rurban and a thread scheduler
16:35 rurban currently I'm adding iterators to simplify foreach and such
16:35 Will_the_Chill rperl is statically typed only
16:36 Will_the_Chill and rperl should maintain full XS compatibility
16:42 rurban I believe XS is fundamentally wrong. I designed a proper external API first, and then put a glue XS layer on top of that
16:42 rurban p5 iterators also
16:43 rurban for my $i (list) {block} compiles the block to a closure taking one arg, the itervar $i
16:44 rurban depending on the list type (decl, listvar, range, lazy list via iterfunction) it calls the closure in sequence
16:45 rurban tuple/hash each(block($i))
16:46 rurban range(from,to) => to(block($i))
16:46 rurban lazy => lazy(block($i))
17:56 stevan joined #perl11
18:10 dalek p2/for: effb03e | rurban++ | / (5 files):
18:10 dalek p2/for: for/foreach prototypes (not yet compiled)
18:10 dalek p2/for:
18:10 dalek p2/for: parse foreach(){} and for(;;;){} with (from..to) ranges (ranges no in lists yet)
18:10 dalek p2/for: review: https://github.com/perl11/p2/commit/effb03e924
18:23 stevan_ joined #perl11
18:31 travis-ci joined #perl11
18:31 travis-ci [travis-ci] perl11/p2#114 (for - effb03e : Reini Urban): The build passed.
18:31 travis-ci [travis-ci] Change view : https://github.com/perl11/p2/commit/effb03e92466
18:31 travis-ci [travis-ci] Build details : http://travis-ci.org/perl11/p2/builds/11833755
18:31 travis-ci left #perl11
18:44 dalek p2/for: e9685f9 | rurban++ | core/compile.c:
18:44 dalek p2/for: compile for: list expr...
18:44 dalek p2/for: review: https://github.com/perl11/p2/commit/e9685f933d
19:04 travis-ci joined #perl11
19:04 travis-ci [travis-ci] perl11/p2#115 (for - e9685f9 : Reini Urban): The build was broken.
19:04 travis-ci [travis-ci] Change view : https://github.com/perl11/p2/compare/effb03e92466...e9685f933d9e
19:04 travis-ci [travis-ci] Build details : http://travis-ci.org/perl11/p2/builds/11834890
19:04 travis-ci left #perl11
19:25 stevan joined #perl11
20:08 rurban1 joined #perl11
21:10 rurban1 joined #perl11
21:19 Will_the_Chill joined #perl11
21:49 stevan joined #perl11
22:24 Will_the_Chill XS may be fundamentally wrong, but we must maintain backward compatibility with XS and CPAN one way or another
23:00 rurban1 joined #perl11
23:21 rurban1 joined #perl11
23:44 Will_the_Chill joined #perl11
23:53 rurban1 joined #perl11
23:53 Will_the_Chill what do y'all think about Perlito?

| Channels | #perl11 index | Today | | Search | Google Search | Plain-Text | summary