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_svnbot r21090 | lwall++ | [STD5] now supports :i including in LTMish spots 00:44
r21090 | lwall++ | [STD] now (more) correctly parses character classes
r21091 | lwall++ | [Cursor5] be slightly more unicodey on :i 00:50
TimToady decommuting & 01:01
pugs_svnbot r21092 | lwall++ | [Cursor5] more unicodification
pugs_svnbot r21093 | lwall++ | [STD] now gets 15% of the way through parsing itself 04:59
s11 Auzon: is there vim syntax highlighter somewhere for the test files? 05:46
also, what's the difference between the .rakudo and the .t test files? 05:48
TimToady s11: the .t files are the standard test files. the .rakudo is a preprocessed version of the .t that takes into account the "fudging" needed to run under rakudo 06:56
pugs_svnbot r21094 | lwall++ | [STD] 23%
TimToady the perl6.vim file can be found in util/
TimToady diff the .t and the .rakudo and you'll see what fudge does 07:01
zzz & # thunk 07:02
pugs_svnbot r21095 | moritz++ | [spec] misc: unfudged a test, modified other, marked one as unspecced 14:03
dolmen perl6: my @a = 0..9; say @a.delete(5..9).perl; say @a.perl; 14:16
p6eval elf 21095: OUTPUT[Can't call method "delete" without a package or object reference at (eval 123) line 4.␤ at ./elf_f line 3861␤]
..kp6 21095: OUTPUT[syntax error at position 10, line 1 column 10:␤my @a = 0..9; say @a.delete(5..9).perl; say @a.perl␤ ^ HERE␤]
..pugs: OUTPUT[(5, 6, 7, 8, 9)␤\(0, 1, 2, 3, 4, undef, undef, undef, undef)␤]
..rakudo 28780: OUTPUT[get_integer() not implemented in class 'Range'␤current instr.: 'parrot;Perl6Array;delete' pc 4215 (src/gen_builtins.pir:2796)␤]
dolmen I would like to fix this in rakudo, but I don't know how to handle it 14:19
moritz_ dolmen: try to grep for get_integer in the rakudo part of the repo, and look how it's done for other stuff 14:20
dolmen Do we have to explicitelty handle Range objects (or more generally objects which "can" 'iterator()') or it is a list expansion issue? 14:21
moritz_ don't know, but you'd better ask on #parrot where rakudo implementation questions are normally handled 14:22
dolmen I tried to implement get_integer in Range, but it will be called only once, so it will only return the first element of the range
moritz_ I'd guess get_integer returns the number of items, not the first item (which doesn't have to be an integer) 14:25
dolmen this answers my question: I think that we should handle Range objects explicitely in delete 14:27
if not only the first element of the range will be deleted and this is not what the user expects 14:28
pmurias TimToady: is it ok if i overload operators on Cursor5 objects? (stringification specificly) 14:38
TimToady yes 14:39
I keep forgetting to say .text as it is...
just try to keep your changes localized, because I'll be working offline for half a day or so and will want to merge 14:41
TimToady afk & 14:43
pugs_svnbot r21096 | dolmen++ | Check Array.delete with a Range argument. 14:45
rakudo_svn r28785 | pmichaud++ | [rakudo]: 15:52
r28785 | pmichaud++ | * spectest-progress update for 6-28: 75 files, 1072 passing, 17 failing
jferrero perlbot, say "Hola" 16:16
pmichaud perl6: say "Hola" 16:17
p6eval elf 21096: OUTPUT[Hola␤]
..kp6 21096: OUTPUT[Hola␤]
..pugs: OUTPUT[Hola␤]
..rakudo 28785: OUTPUT[Hola␤]
jferrero perl6: print "Thanks, pmichaud" 16:18
p6eval elf 21096: OUTPUT[Thanks, pmichaud]
..kp6 21096: OUTPUT[Thanks, pmichaud]
..pugs: OUTPUT[Thanks, pmichaud]
..rakudo 28785: OUTPUT[Thanks, pmichaud]
rakudo_svn r28786 | pmichaud++ | [rakudo]: 16:20
r28786 | pmichaud++ | * Add temporary implementation of Order::Same, Order::Increase,
r28786 | pmichaud++ | Order::Decrease
r28786 | pmichaud++ | * t/spec/S03-operators/comparison.t passes again
Juerd Who maintains p6eval? 16:33
moritz_ 16:43
pugs_svnbot r21097 | moritz++ | [SLAVES] exp_evalbot is now called p6eval 16:46
rhr does the Order enum conflict with the Order role? how do you declare an enum and a role (other than the role made by the enum) of the same name? 16:59
our enum Order ...; role Order is also {...} ?
moritz_ I think that an enum is a way to define a role (not sure, though) 17:04
S12:1650 17:05
or S12/"Enums" if you like that better ;)
rhr right, but Order needs more than just an enum
moritz_ "An enum is a low-level class that can function as a role or property."
pmichaud I'm thinking that eventually we'll be able to say something like (in Prelude): enum Order << :Increase(-1), Same, Decrease >>; 17:18
pugs_svnbot r21098 | moritz++ | [spec] delete.t: fix plan, improve fudging for rakudo 17:19
rakudo_svn r28787 | moritz++ | [rakudo] delete.t in spectest_regression.data is no longer "pure", 17:21
r28787 | moritz++ | dolmen++ added new tests.
pugs_svnbot r21099 | moritz++ | [spec] delete.t: exclusive ranges are 2^..4, not ^2..4 17:22
pmichaud rakudo: say 1^..10; 17:34
p6eval rakudo 28787: OUTPUT[101010101010101010␤]
pmichaud oops.
moritz_ rakudo: 1^..10 17:36
p6eval rakudo 28787: RESULT[1^..10]
moritz_ rakudo: (1^..10).perl.say
p6eval rakudo 28787: OUTPUT[1^..10␤]
rakudo_svn r28788 | pmichaud++ | [rakudo]: 18:00
r28788 | pmichaud++ | * Fix exclusive from ranges (1^..5)
r28789 | pmichaud++ | [rakudo]: 18:11
r28789 | pmichaud++ | * Make sure Array.delete flattens its arguments in list context.
pugs_svnbot r21100 | pmichaud++ | S29-array/delete.t: 18:14
r21100 | pmichaud++ | * Uncomment #?rakudo skip on ranges
r21100 | pmichaud++ | * Fix one incorrect test result.
rakudo_svn r28791 | pmichaud++ | [rakudo]: 18:24
r28791 | pmichaud++ | * Adjust <=>, cmp, leg to return Order::* values.
pugs_svnbot r21101 | pmurias++ | [pixie] 01-sanity/01-tap.t passes 19:02
pmurias is Common Lisp fast enough to get resonable performance using smop style responder interfaces? 19:25
vixey yes 19:27
pmurias If you mean CLOS ?
SamB if CLOS can do it, I imagine you can too (in theory) 19:28
CLOS is, as I understand it, implemented (almost?) 100% in standard CL 19:29
pmurias vixey: i meant $obj->{_dispatcher}(...)
if CLOS is not specialy hard coded in the compiler, i guess the performance wouldn't suffer much 19:30
SamB hmm, apparantly there is an unofficial MOP... 19:32
pmurias SamB: in Common Lisp? 19:33
SamB www.lisp.org/mop/index.html
lambdabot Title: MetaObject Protocol
pmurias by smop style responder interfaces i meant adding a layer of indirection where the object itself desides how a method call should be done 19:37
s/call/dispatch/
SamB well, it shouldn't be any more unreasonable than in any other environment 19:38
pmurias it's was extremely unreasonable in kp6 perl5 backend 19:39
because the difference between $obj->{_DISPATCH}($obj,"method",...args...) and $obj->method(...args...) is significant in perl5 19:41
SamB well, this is CL we are talking about
pmurias how fast is CL compared to C? 19:43
vixey very fast 19:44
pmurias vixey: compared to C? 19:46
;)
vixey pmurias: Yes
pmurias ;) 19:48
pugs_svnbot r21102 | Auzon++ | [gsoc_spectest] minimal whitespace test. (added 8 tests) 20:50
cathya anyone awake 21:44
i've got a doubt about somthing not related to perl6
vixey *not* related to perl6 21:45
cathya yep not 21:48
TimToady I doubt there's anything not related to Perl 6. 21:53
pugs_svnbot r21103 | lwall++ | [STD] 47%
vixey hehe TimToady 21:54
rakudo_svn r28796 | chromatic++ | [Rakudo] Fixed SVN metadata. 22:25

Powered by Cro and the Raku® Programming Language.

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

Thank you!