pugscode.org/ | nopaste: sial.org/pbot/perl6 | pugs: [~] <m oo se> (or rakudo:, kp6:, smop: etc.) || We do Haskell, too | > reverse . show $ foldl1 (*) [1..4] | irclog: irc.pugscode.org/
Set by TimToady on 25 January 2008.
Auzon perl6: my $a = { 3 }; $a.WHAT; 02:30
exp_evalbot kp6: No output (you need to produce output to STDOUT)
..pugs: RESULT[::Block]
..rakudo: RESULT[Method 'perl' not found for invocant of class ''␤current instr.: '_block10' pc 35 (EVAL_11:16)␤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 785 (src/PCT/HLLCompiler.pir:458)␤called from Sub 'parrot;PCT::HLLCompiler;evalfiles' pc 1059 (src/PCT/HLLCompiler.pir:585)␤called from
..Sub 'parrot;PCT::HLLCompiler;command_line' pc 1238 (src/PCT...
Auzon perl6: my $a = { 3 }; say $a.WHAT; 02:31
exp_evalbot kp6: OUTPUT[Int␤]
..pugs: OUTPUT[Block␤]
..rakudo: OUTPUT[Code␤]
Auzon ^_^'
perl6: my $a = { 3 }; my $b = 3; say ($a eqv $b); 02:33
exp_evalbot kp6: OUTPUT[no method 'APPLY' in Class 'Undef'␤ at compiled/perl5-kp6-mp6/lib/KindaPerl6/Runtime/Perl5/MOP.pm line 345␤ KindaPerl6::Runtime::Perl5::MOP::__ANON__('HASH(0x824c004)', 'APPLY', 'HASH(0x8e52504)') called at compiled/perl5-kp6-mp6/lib/KindaPerl6/Runtime/Perl5/MOP.pm line 169␤
..main::DISPATCH('HASH(0x824c004)', 'APPLY', 'HASH(0x8e52504)') ca...
..pugs: OUTPUT[␤]
..rakudo: OUTPUT[Null PMC access in invoke()␤current instr.: '_block10' pc 79 (EVAL_11:36)␤called from Sub 'parrot;PCT::HLLCompiler;eval' pc 785 (src/PCT/HLLCompiler.pir:458)␤called from Sub 'parrot;PCT::HLLCompiler;evalfiles' pc 1059 (src/PCT/HLLCompiler.pir:585)␤called from Sub
..'parrot;PCT::HLLCompiler;command_line' pc 1238 (src/PCT/HLLCompiler.pir:674)...
Auzon Should fudged tests be marked as passing, or perhaps passing under some condition? 03:59
TimToady at the moment fudge throws in an exit(1) at the end to make sure the file is considered "dubious" even if all the tests officially pass 04:03
Auzon I'm applying for the Summer of Code, and I was wondering if, as some bonus work, you'd like fudged cases to be differentiated from strictly failing tests somehow. 04:04
TimToady the test harness does that already by emiting # TODO as a comment 04:05
which is why the smoke matrix puts dark green instead of red on those
Auzon It all makes sense now. Thanks TimToady.
Are there any other test-related tasks that might be good for bonus work? 04:06
TimToady well, there's still a whole lot of tests to convert to fudge format
and maybe another 20,000 or so tests to write :) 04:07
Auzon I don't think I'll be able to do more than 1000 in the summer, even in a best case scenario.
TimToady you only have to do it by "christmas" :) 04:08
Auzon When you say converting tests to Fudge format, does that mean modifying fudge or the tests? Or both? 04:09
TimToady the tests
fudge is pretty much done
most pugs tests have :todo as a test argument 04:10
that has to move out to comments for fudge
so that each implementation can have its own todo list
Auzon Makes sense
TimToady and ignore all the todos of other impls
some of that could perhaps be automated 04:11
Auzon I guess my "bonus work" could be looking through existing tests.
TimToady but you'd need to find a good ad hoc text processing language :)
Auzon Hm. I think we have one or two around here somewhere ;)
TimToady also, they're slowing getting renamed into t/spec/*
so that they match up with the synopses numbers 04:12
Auzon Is there any procedure for adding tests that I should be aware of before applying?
TimToady just need a pugs commit bit 04:13
Auzon I didn't think there would be any administrative stuff, but I thought I'd ask ;) 04:14
TimToady the pugs repo has a very low bar
parrot's is a little higher
everything in the pugs repo works on a "easier to get forgiveness than permission" principle 04:15
though, of course, if you're not sure about something it never hurts to ask first
Auzon Right. Sounds good to me :) 04:17
How hard is Rakudo to get working for passing tests? 04:18
TimToady that's more of a #parrot question, I expect
(on irc.perl.org) 04:19
Auzon Alright.
TimToady but, tends to compile easily, if that's what you're asking
Auzon That helps. I still haven't gotten Pugs to compile :-/ 04:21
TimToady me either, someone needs to ask forgiveness :)
I'm just now downloading ghc681 rpm to see if it works better
Auzon From my experience earlier today on my Fedora 8 install, no. 04:22
TimToady well, I'll try to duplicate that result, then :)
Auzon with ghc661, I couldn't figure out how to get runhaskell for HsSyck 04:23
681 went further but still stumbled
TimToady 661 always complains: Setup: HsSyck.cabal:14: 'Executable' stanza starting with field 'license'
did you delete 661? 04:24
Auzon You got further with 661 than I did then
I think I have it installed in parallel with 681 at present
but I tried them both alone
TimToady up to 35/109 04:29
had to remake my symlinks in /usr/bin
Auzon further than I got with either time.
TimToady did you make clean and perl Makefile.PL? 04:30
Auzon No, I didn't run perl on that makefile.
I'll try that again when I get a chance.
TimToady seems to compile pugs faster
crumb, failed recompiling the prelude... 04:32
TimToady 睡 & 04:44
lichtkind can anybody help me to understand what smop exactly is? 05:32
lichtkind whats the real advantage to something like SMOP ? 05:40
Auzon if I understand it correctly, SMOP is an implementation like perl5 for running Perl 6. 05:40
lichtkind yeah but what was ruasos goal? 05:41
Auzon I am uncertain.
lichtkind why smop if kp6 can let run perl 6 on perl
Auzon KP6 was slow, if I recall.
lichtkind but with smop faster? 05:42
Auzon SMOP is written in C in the style of Perl 5's interpreter .
if I'm correct.
lichtkind yes so far i see it too
Auzon That approach worked once before, so it might be worth pursuing again. :) 05:43
lichtkind am im correct if i say that you can tie with smop different interpreter together to get best of many worlds?
Auzon I don't know.
lichtkind why was kp6 slow if perl is so fast? 05:44
Auzon KP6 is Perl 6 compiled to Perl 5.
so we have generated Perl 5 interpreting Perl 6. 05:45
lichtkind so it generated slow code
Auzon that may have been part of the problem
but also consider how many layers of interpretation there are in that case
lichtkind you mean with smop there would be less layer? 05:46
Auzon so we're interpreting code on code translated to run on another interpreter
right. SMOP would be in C, which is faster to begin with than Perl 5.
but we also don't have the translation layer
Have you looked at the wiki page for SMOP?
lichtkind sure 05:47
Auzon I'm not entirely sure if SMOP was meant to be a stepping stone or a maintained implementation of Perl 6 05:48
I know that KP6/MP6, redsix, and elf are more of stepping stones.
lichtkind what is elf?
Auzon from what I understand, it was a recent build of compiling Perl 6 to Perl 5, similar to KP6, but with less of a scope, I think 05:50
mainly a stepping stone tool
It could also emit (for instance) Lisp if someone wrote the backend
lichtkind were getting lot of stones lately :) 05:51
stone rich like we say in german
Auzon: anyway night is over and i had no sleep thank you for the answers
Auzon Good night :)
lichtkind :)
Auzon [particle]: I sent you another GSoC draft 06:49
it should be good.
sorry about waiting so long though :-/ 06:53
Auzon sleep & 07:21
lichtkind has still anybody hints whats the motivation and purpose behind SMOP ? 12:52
pmurias lichtkind: the purpose of smop is too build a runtime for Perl 6 14:02
pmurias while it's possible to run a subset of p6 on perl5 with (near) perl5 speed, doing some things efficiently is very hard 14:08
lichtkind pmurias: thanks i read te word runtime , i am wandering what special advantages should smop have? 14:10
just stepping stone for bootstraping or real alternative to run 14:11
pmurias a real runtime alternative 14:20
it's simpler then for example parrot 14:21
the whole vm is object oriented 14:25
lichtkind pmurias: and is it true that you can include other interpreter into smop to get best of several worlds? 14:59
pmurias lichtkind: it's intented as a smop feature, but it's possible with other interpreters also 17:31
lichtkind pmurias: and do you know what gave the pressure that started smop ? 17:47
pmurias lichtkind: the need for a fast runtime for kp6 and problems with parrot 17:49
lichtkind thanks
pmurias: and what exactly is it called what emmits kp6 and what executes smop? 17:50
pmurias kp6 can't use smop yet 17:52
lichtkind i know
i just want to know how is that structure called that executes smop
pmurias the stuff smop executes? 17:54
the only stuff smop can execute now is sm0p 17:58
lichtkind pmurias: wb 18:39
pmurias rehi 18:41
;) 18:42
pmurias trying to get by printer to work after a foolish attempt to install an ipod-- on my gentoo box, so had to reboot to change the kernel 18:43
lichtkind pmurias: is smop/kp6 a serious kombatant in the race who first runs the std.pm? 18:44
pmurias lichtkind: no 18:45
lichtkind: kp6 is stalled
lichtkind but they still try to make it run on smop? 18:46
pmurias smop still needs work and nobody is working on kp6 nowdays 18:47
lichtkind but smop is under development? 18:49
lichtkind ruaso is even one of the busiest in the wiki :) 18:52
pmurias lichtkind: re under development, you can look at the commit log 18:55
lichtkind where?
pmurias: but im currently finishing that article anyhow :) 18:56
Auzon Well, my SoC application is in :) 18:59
lichtkind congrats 19:00
Auzon Now... I wait. :-/ 19:01
Skaman_Sam Hello, all! 19:03
I have a really stupid question - but here it goes-
Why would I want to use Perl6 over Perl5 ?
Auzon Right now? You probably wouldn't 19:04
When it's done, however, there are a lot of great changes.
Skaman_Sam cool - now I feel more secure about myself ;) 19:05
Auzon for instance, here's one neat piece of syntax added:
pugs: 1,3,5,7 Z 2,4,6,8
exp_evalbot RESULT[((1, 2), (3, 4), (5, 6), (7, 8))]
Skaman_Sam I have been hesitant to embrace the Parrot Paradigm
ohh...
Auzon The idea is that we'll eventually have many compatible implementations of Perl 6. 19:06
Unlike Perl 5, Perl 6 has a specification other than "the perl executable's behavior"
lichtkind Skaman_Sam: i would use perl 6 over perl 5 because it let me write shorter programs :) 19:08
Skaman_Sam on the parrot site, it says it lets perl hackers hack better 19:22
lichtkind Skaman_Sam: yes perl 5 is more hackish :) 19:43

Powered by Cro and the Raku® Programming Language.

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

Thank you!