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.
pugs_svn r21322 | ruoso++ | [smop] use sm0p labels in test/14... still with some memory leaks... 00:11
pugs_svn r21323 | ruoso++ | [smop] Some more changes trying to catch the memory leak... but this kind of debug is incredibly hard... 01:08
meppl good night 01:15
pugs_svn r21324 | pmurias++ | [sm0p] move_capturize is no longer a special case 09:48
r21325 | pmurias++ | [sm0p] moved misc/sm0p to v6/smop/sm0p 09:57
pmurias how should i get hold of the stack backtrace in c (i need to to get it to print where the reference counters get --/++) what's the best way to do it? 10:50
using gdb, inline assembler, or gprof output are things which come to my mind 10:51
pmurias ruoso: hi 11:16
ruoso pmurias, hi
pmurias re copy, are you sure you haven't fixed something, it didn't work when i tried 11:17
ruoso I've just changed the test count... you had 1..2 where you had only one tes
test
and I added a SMOP_RELEASE call for the value you initialized outside the sm0p block 11:18
pmurias well i find things work strangely sometimes, evil spirits living inside my laptop most likely ;) 11:20
pmurias C 11:21
SORRY
(caps lock)--
ruoso: regarding .call, implementing it would require lots of code duplication if it was possible to specifiy the responder interface would the move_* methods be required? 11:24
ruoso pmurias, doesn't move_* accept 0 as target parameter? 11:26
pmurias ruoso: you mean making .call all the move_stuff? 11:29
ruoso calling the move_* from the .call method
moving to the current node...
pmurias would it be possible to use a q:sm0p block for that? 11:31
ruoso hmmm... I don't think so... 11:32
pmurias what would be the main problem? 11:36
(besides it being slow)
ruoso the fact that you would need to recurse, and the CurrentFrame wouldn't be the frame you want to move the values from 11:38
pmurias ruoso: i would use a diffrent frame 11:41
ruoso yes... how would you move_responder from the original frame?
pmurias the code duplication starts to look more inviting now ;) 11:43
ruoso heh...
but you can SMOP_DISPATCH manually
and then you stay in the same frame
pmurias doing SMOP_DISPATCH manually is horrible 11:44
ruoso heh...
pmurias so i think it's either having the move methods or call 11:46
if nodes was an array we could have the move methods in sm0p/perl6 11:49
ruoso pmurias, the nodes are an array... you just need to expose the api 11:50
pmurias i'm looking at libjit now, have you thought how are we going to do NCI? 12:19
pugs_svn r21326 | pmurias++ | sm0p was moved to v6/smop 15:29
Auzon t/spec/S02-builtin_data_types/multi_dimensional_array.t seems to assume that $foo[0][0] means the same as $foo[0;0]. Is that correct? It doesn't seem like it to me. 18:56
Also, is there any difference between using an array or an array stored in a scalar when it comes to multidimensional arrays? 18:57
moritz_ tries to check 18:58
Auzon The spec was pretty silent about the [0][0] syntax 18:59
moritz_ the difference between an array and an array in a scalar is purely syntactical, because the array container forwards method calls to its content
so my Array of Int @x declares an array of Arrays of Int
whereas my Array of Int $x; declares an Array of Int (afaict) 19:00
Auzon OK. As far as I can tell, @foo[0][0] or $foo[0][0] is like the Perl 5 style, but @foo[0;0] is true multidimensional (in the sense that it has a defined boundary) and isn't the same thing. 19:01
moritz_ those @array[0][0] tests are pretty old, if you look at svn blame 19:02
r1653
Auzon wow
moritz_ maybe send a mail to p6l asking for clarification
autovivification.t has a lot of redundant smartlinks 19:03
5 consecutive smartlinks that are displayed in a row in the spec doc 19:04
Auzon re mail, sounds good
re smartlinks, I'll clean that up unless you've started on it already 19:05
moritz_ I haven't 19:06
Auzon ok, done 19:07
moritz_ Auzon++
pugs_svn r21327 | Auzon++ | [gsoc_spectest] smartlink cleanups
Auzon hm. I was descriptive there ;) 19:10
moritz_ sufficient ;) 19:10
Auzon What is the terminology for $foo[0][0] in Perl 6? $foo is a capture of an array, and $foo[0] is also a capture of an array, right? 19:11
moritz_ I think so, yes 19:12
or you can just say that both $foo and $foo[0] are Array objects
Auzon and $foo[0;0] refers to a multidimensional array object, right?
moritz_ pugs: my @x = <a b c>; my $y = @x; my $z = \@x; say $y.perl, ' ', $z.perl 19:13
p6eval pugs: OUTPUT[\\("a", "b", "c") \\("a", "b", "c")␤]
moritz_ $foo in $foo[0;0] referst to a multidim array object
Auzon ok, create
*great
moritz_ $foo[0;0] referst to a stored item
s:first /st/t/ 19:14
Auzon doesn't know how 'great' became 'create'
moritz_ my $str = 'create' ; # ;-) 19:15
moritz_ Auzon: still around? 19:57
moritz_ rakudo: say ("a" ~~ m/a/).WHAT 19:59
p6eval rakudo 29398: OUTPUT[Statement not terminated properly at line 1, near "/).WHAT"␤current instr.: 'parrot;PGE::Util;die' pc 120 (runtime/parrot/library/PGE/Util.pir:82)␤]
Auzon moritz_: I am now. 20:04
moritz_ Auzon: I'm looking at rakudo's ROADMAP, and collecting stuff that's missing there
nopaste.snit.ch/13574 this is what I've got so far 20:05
Auzon: do you know other major stuff (from top of your head) that's missing?
Auzon Something that's a big deal for the tests is that eval's scope is wrong (e.g. my $foo = 5; eval 'say $foo'; doesn't see $foo in the eval) 20:06
moritz_ that's a big annoyance, but nothing that has to appear in the roadmap; pmichaud is workiing on that (mostly parrot issues blocking) 20:07
Auzon Right.
I can't think of anything big missing.
moritz_ ok, thanks for looking
pmurias ruoso: if libjit is GPL, can we use it? 20:16
moritz_ rakuo: say "abc".bytes 20:32
rakudo: say "abc".bytes
p6eval rakudo 29398: OUTPUT[Method 'bytes' not found for invocant of class 'Perl6Str'␤current instr.: '_block11' pc 21 (EVAL_13:13)␤]
moritz_ rakudo: say "abc".codes
p6eval rakudo 29398: OUTPUT[Method 'codes' not found for invocant of class 'Perl6Str'␤current instr.: '_block11' pc 21 (EVAL_13:13)␤]
moritz_ rakudo: say "abc".graphs
p6eval rakudo 29398: OUTPUT[Method 'graphs' not found for invocant of class 'Perl6Str'␤current instr.: '_block11' pc 21 (EVAL_13:13)␤]
Auzon rakudo: say "abc".chars 20:33
p6eval rakudo 29398: OUTPUT[3␤]
Auzon At least it has one of them ;)
moritz_ rakudo: say "äü".chars 20:34
p6eval rakudo 29398: OUTPUT[2␤]
moritz_ rakudo: say "A҄".chars 20:35
p6eval rakudo 29398: OUTPUT[2␤]
moritz_ ok, current .chars corresponds to .codes actually
(the A҄ was entered as two codepoints) 20:36
Auzon ah 20:37
Is there a one-char version of that?
moritz_ don't think so
it's a cyrillic combining accent
;-)
rakudo: say "a𝉄".chars 20:38
p6eval rakudo 29398: OUTPUT[2␤]
moritz_ I have no font that would display it correctly ;)
U+1D244 COMBINING GREEK MUSICAL PENTASEME 20:39
pugs_svn r21328 | moritz++ | [irclog] dalek in #parrot is a bot 20:40
Auzon ^ a statement of fact. 20:42
pmurias ruoso: how should pure sm0p file (without c) be distinguished a 'use sm0p;' at front or a heuristic? 21:11

Powered by Cro and the Raku® Programming Language.

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

Thank you!