Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2008-01-14

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:26 Southen joined #perl6
00:44 Limbic_Region joined #perl6
00:48 ruoso joined #perl6
00:58 meppuru good night
01:00 _sprocket_ left #perl6
01:00 cnhackTNT joined #perl6
01:11 jferrero joined #perl6
01:47 ruz joined #perl6
01:53 Jedai joined #perl6
01:55 skids perl5 embedded in pugs -- any way to get functions that modify parameters to work?
01:56 skids e.g. IO::Socket::INET::recv($dat,1,0) modifies $dat, but it is not modified from perl6
01:57 allbery_b you have to do something special to get variables propagated, IIRC, but I don't' recall what.  (and it may not be implemented yet, even; I foret)
01:57 BinGOs_ joined #perl6
02:03 BinGOs joined #perl6
02:10 xdg joined #perl6
02:13 BinGOs joined #perl6
02:20 BinGOs joined #perl6
02:22 agentzh joined #perl6
02:22 agentzh morning!
02:22 lambdabot agentzh: You have 1 new message. '/msg lambdabot @messages' to read it.
02:22 agentzh @message
02:22 lambdabot Maybe you meant: messages messages?
02:22 agentzh ....
02:22 agentzh @messages
02:22 lambdabot fglock said 1m 3d 11h 5m 17s ago: there is a v6.pm cpan-testers report with "t/01-sanity/02-counter...............greediness control not implemented: [ '\»' | . ]" - is this PCR related?
02:23 agentzh fglock: not sure...
02:23 agentzh 1m 3d...
02:24 Auzon Hi
02:29 BinGOs joined #perl6
02:39 BinGOs joined #perl6
02:46 BinGOs joined #perl6
03:02 ntgrl joined #perl6
03:09 BinGOs joined #perl6
03:13 penk joined #perl6
03:41 kanru joined #perl6
04:22 Auzon pugs: 1,2 X 3,4
04:22 exp_evalbot RESULT[((1, 3), (1, 4), (2, 3), (2, 4))]
04:22 Auzon pugs: (1,2) X (1,2)
04:22 exp_evalbot RESULT[((1, 1), (1, 2), (2, 1), (2, 2))]
04:22 Auzon pugs: 1,2 X 1,2 X 1,2
04:22 exp_evalbot RESULT[((1, 1, 1), (1, 1, 2), (1, 2, 1), (1, 2, 2), (2, 1, 1), (2, 1, 2), (2, 2, 1), (2, 2, 2))]
04:22 Auzon excellent.
04:23 Auzon pugs: 0,1,2 X 0,1,2 X 7,8,9
04:23 exp_evalbot RESULT[((0, 0, 7), (0, 0, 8), (0, 0, 9), (0, 1, 7), (0, 1, 8), (0, 1, 9), (0, 2, 7), (0, 2, 8), (0, 2, 9), (1, 0, 7), (1, 0, 8), (1, 0, 9), (1, 1, 7), (1, 1, 8), (1, 1, 9), (1, 2, 7), (1, 2, 8), (1, 2, 9), (2, 0, 7), (2, 0, 8), (2, 0, 9), (2, 1, 7), (2, 1, 8), (2, 1, 9), (2, 2, 7), (2, 2, 8), (2,
04:23 exp_evalbot ..2, 9))]
04:34 Auzon pugs: 0,1,2 X 0,1,2 X 7,8,9,0
04:34 exp_evalbot RESULT[((0, 0, 7), (0, 0, 8), (0, 0, 9), (0, 0, 0), (0, 1, 7), (0, 1, 8), (0, 1, 9), (0, 1, 0), (0, 2, 7), (0, 2, 8), (0, 2, 9), (0, 2, 0), (1, 0, 7), (1, 0, 8), (1, 0, 9), (1, 0, 0), (1, 1, 7), (1, 1, 8), (1, 1, 9), (1, 1, 0), (1, 2, 7), (1, 2, 8), (1, 2, 9), (1, 2, 0), (2, 0, 7), (2, 0, 8), (2,
04:34 exp_evalbot ..0, 9), (2, 0, 0), (2, 1, 7), (2, 1, 8), (2, 1, 9)...
04:35 Auzon pugs: 0 X 0,1,2 X 7,8,9,0
04:35 exp_evalbot RESULT[((0, 0, 7), (0, 0, 8), (0, 0, 9), (0, 0, 0), (0, 1, 7), (0, 1, 8), (0, 1, 9), (0, 1, 0), (0, 2, 7), (0, 2, 8), (0, 2, 9), (0, 2, 0))]
04:35 Auzon I love this feature.
04:40 kanru joined #perl6
05:14 pmccann joined #perl6
05:18 coumbes joined #perl6
05:42 cnhackTNT joined #perl6
05:55 pmccann left #perl6
06:00 awwaiid joined #perl6
06:02 alc joined #perl6
06:35 devogon joined #perl6
07:02 pen1 joined #perl6
07:03 lyokato_ joined #perl6
07:14 pen2 joined #perl6
07:14 overlast_ joined #perl6
07:16 pen3 joined #perl6
07:19 pen4 joined #perl6
07:40 Aankhen`` joined #perl6
07:42 iblechbot joined #perl6
07:43 IllvilJa joined #perl6
07:47 Tene So, question about method delegation.
07:47 Tene S12 says "has $the_real_me handles *;"
07:47 mattz joined #perl6
07:48 Tene Is there any corresponding way to specify a function that gets called when no other function defined on the object matches?
07:49 Tene I think ruby calls it method_missing or some such.
07:49 Tene That seems like the sort of thing that should be possible in Perl 6's meta-model, but I can't find anything in the spec about it.
07:54 japhb_ Tene: I vaguely recall something like "AUTOMETHOD", though that's probably ancient spec-wise
07:54 Aankhen`` I coulda sworn there was an "AUTOMETHDEF" or something like that, but I can't find it in the spec.
07:56 Aankhen`` Okay, from <http://perlcabal.org/syn/S13.html#Fallbacks>: "If an attempt is made to dispatch to a declared but undefined routine, Perl will redispatch to an AUTODEF submethod [conjectural] as appropriate to define the routine."
07:56 Aankhen`` I guess it's not fleshed out yet.
08:28 pen4 left #perl6
08:29 xinming joined #perl6
08:31 penk joined #perl6
09:00 riffraff joined #perl6
09:07 riffraff joined #perl6
09:08 riffraff hi
09:15 Khisanth joined #perl6
09:17 moritz_ hi riffraff
10:10 drbean joined #perl6
10:15 njbartlett_ joined #perl6
10:18 ebassi joined #perl6
10:25 bwisti joined #perl6
10:44 pen1 joined #perl6
10:56 njbartlett_ joined #perl6
11:38 pugs_svn r19474 | cosimo++ | Changed the todo mark with new compiler directives
11:42 meppuru joined #perl6
11:43 masak joined #perl6
11:53 penk joined #perl6
12:14 polettix joined #perl6
12:27 Aankhen`` joined #perl6
12:29 njbartlett_ joined #perl6
12:54 ebassi joined #perl6
12:56 farmerjoe joined #perl6
13:50 cmarcelo joined #perl6
13:55 falesca joined #perl6
14:05 chris2 joined #perl6
14:13 xinming_ joined #perl6
14:20 xinming_ joined #perl6
14:20 falesca joined #perl6
14:20 iblechbot joined #perl6
14:20 kanru joined #perl6
14:20 alester joined #perl6
14:20 charsbar joined #perl6
14:20 skids joined #perl6
14:20 mncharity joined #perl6
14:20 silug joined #perl6
14:20 qmole_ joined #perl6
14:20 pbuetow joined #perl6
14:20 djm joined #perl6
14:20 avar joined #perl6
14:20 Grrrr joined #perl6
14:20 broquaint joined #perl6
14:20 bloonix joined #perl6
14:20 allbery_b joined #perl6
14:20 nnunley joined #perl6
14:20 GeJ joined #perl6
14:23 jhorwitz joined #perl6
14:25 Aankh|Clone joined #perl6
14:33 renormalist joined #perl6
14:33 Auzon left #perl6
14:34 renormalist joined #perl6
14:36 jferrero joined #perl6
14:39 rdice joined #perl6
15:10 TJCRI joined #perl6
15:13 bphillip1 joined #perl6
15:39 Khisanth joined #perl6
15:41 pbuetow joined #perl6
15:57 pugs_svn r19475 | particle++ | add todo markers to a few spec tests
16:13 pmurias joined #perl6
16:13 pmurias Aankh|Clone: CANDO
16:16 Aankh|Clone Ah, thanks.
16:19 ispy_ joined #perl6
16:40 kanru joined #perl6
16:44 barney joined #perl6
16:45 pasteling "ruoso" at 195.23.61.1 pasted "I'm with limited internet access right now, but here is a piece of code where we program SMOP in SMOP... This is the code for CPS call to DESTROYALL done by gc." (38 lines, 1.9K) at http://sial.org/pbot/29849
17:00 DarkWolf84 joined #perl6
17:05 moritz_ TimToady: any thoughts on the "For 5.12: named parameters + prototypes + methods + aliases" discussion on p5p?
17:14 njbartlett_ joined #perl6
17:20 bwisti joined #perl6
17:21 manfred joined #perl6
17:22 IllvilJa joined #perl6
17:31 aindilis joined #perl6
17:33 jjore joined #perl6
17:35 BinGOs joined #perl6
17:42 TimToady moritz_: I'm sure if I go and read the discussion it will give me high blood pressure  :)
17:44 moritz_ ;)
18:04 Grrrr joined #perl6
18:07 braceta joined #perl6
18:25 polettix joined #perl6
18:25 TimToady joined #perl6
18:29 ruoso joined #perl6
18:33 cognominal_ joined #perl6
18:37 rhr TimToady: what's the default normalization form for Str.print, .normalize, .codes, and .ord?  should any of them fail by default?
18:42 cognominal_ joined #perl6
18:50 Limbic_Region joined #perl6
18:54 pmurias joined #perl6
18:54 lambdabot joined #perl6
19:01 rindolf joined #perl6
19:05 TimToady print should know what its handle wants.
19:06 TimToady I'm not sure .normalize is useful under graphemes
19:06 TimToady .ord/.chr should probably be based on grapheme id normally, so NFC
19:08 tobeya joined #perl6
19:08 TimToady if "use codepoints" declares whether the default is nfc or nfd, maybe .normalize defaults to that
19:08 TimToady (in that lexical scope)
19:10 TimToady and maybe under use bytes you also get a default encoding, so .normalize.encode can take any Str and turn it into, say, :nf<d> :enc<utf-8>
19:11 TimToady so maybe the defaults in Functions.pod are wrong, and should default to something in the lexical scope
19:11 TimToady like $?NF and $?ENC
19:12 TimToady I suppose open could also default to $?NF and $?ENC
19:15 TimToady or maybe we don't tie it to "use bytes" and "use codepoints", but have "use nf 'c'" and "use enc 'utf-16le'" instead, since you also might want $?NF and $?ENC to default for other things like converting Str to Buf
19:15 TimToady even under the grapheme view
19:16 TimToady hmm, maybe it's just "constant $?NF = 'C'" in that case, and no use is required
19:18 TimToady but if the other arguments are :nf<> and :enc<>, then maybe "use uni :nf<kc> :enc<utf-32>" is what people would prefer to write to set their defaults
19:18 TimToady for consistency
19:18 stevan_ joined #perl6
19:20 ruoso TimToady, is it ok for a specific environment to declare new $?FOO variables?
19:20 ruoso I mean... smop will probably have a $?STACK
19:21 TimToady why not?
19:21 TimToady except that would imply that $?STACK is a compile-time constant
19:21 ruoso I mean, I was just wondering if that was not something to be specced
19:21 TimToady which I kinda doubt
19:22 ruoso $?STACK is kinda a compile-time constant that means, the current stack on that block
19:22 ruoso the value is not constant, but the meaning is
19:22 TimToady as soon as you say "current" it's not a constant
19:22 TimToady no, that's why we got rid of $?SELF
19:22 alester afternoon all
19:23 TimToady if it varies, you can use $*STACK
19:23 ruoso but then it means it can be local
19:23 ruoso and it's not the case
19:23 ruoso better yet,
19:23 TimToady maybe we need a dynamic sigil...
19:23 TimToady *twigil
19:23 ruoso the "current" stack only means something inside a block
19:24 ruoso like http://sial.org/pbot/29849
19:24 lambdabot Title: Paste #29849 from "ruoso" at 195.23.61.1
19:24 ruoso it should probably be a macro, actually
19:24 TimToady like "self" really
19:24 ruoso but, should I use a plain bareword?
19:25 TimToady __STACK__ if we reserve those for implementation details
19:25 rhr doesn't $+ kind of do that?
19:25 ruoso __STACK__ seems fine to me...
19:25 ruoso weird enough to not cause any confusion
19:25 TimToady $+ scans for a contextual variable
19:26 ruoso I think the point is to get a pattern for this implementation-dependant macros
19:26 TimToady rhr: did you see my unicode microrant above?
19:26 ruoso I mean... HOW is available in all implementations, but __STACK__ isn't
19:26 rhr yes, thanks
19:26 ruoso I should probably follow the same pattern to RI
19:26 ruoso __RI__(obj)
19:26 ruoso or obj.__RI__
19:26 rhr I was thinking .normalize would be used to tell a graph Str how to become a codepoint Str when asked, is there another way to do that?
19:27 ruoso because it's only available in some Perl 6 implementations
19:27 TimToady shades of python...
19:28 TimToady I think it is probably not useful for a Str to know how to become a codepoint without knowing how it's going to be used
19:28 TimToady it's the context that uses it that knows how it wants it
19:28 TimToady and two contexts could want it differently
19:29 TimToady you pass a Str into a module, and it's written in "use codepoint; use :nf<c>"; and then it sends it to a random handle which is set to :nf<d>
19:29 * ruoso will consider the __FOO__ pattern for implementation-exclusive-macros from now on..
19:29 TimToady at each of those context switches it should do the right thing
19:30 rhr OK, that sounds good
19:30 TimToady note that you might have some trouble if you link in a python program and call $obj.__STACK__
19:31 TimToady if you think it's a macro
19:31 TimToady but python thinks it's an ordinary method
19:31 TimToady maybe we should go with something newish
19:31 TimToady maybe ___STACK___
19:32 ruoso hmm... should I stick with __STACK__() only?
19:32 ruoso I mean
19:32 ruoso __FOO__() only
19:32 ruoso __STACK__ doesn't get any parameter
19:32 ruoso or __RI__(__STACK__)
19:32 ruoso to get the responder interface of the current stack
19:32 TimToady that would always dispatch mmd-wise so would stay in the Perl universe
19:32 TimToady it's only the $obj.foo form that can interpret "foo" according to the language of $obj
19:33 TimToady but I'd say just for avoiding cultural confusion to use triple underscore
19:33 TimToady and since it's going to generally be hidden from the user
19:33 TimToady and implementation vars are going to be fairly rare
19:33 TimToady so longer is okay
19:34 ruoso yeah... that is for very deep things, like implementing CPS in core
19:34 ruoso probably some module will hide it away
19:34 TimToady maybe for continuations we can use 52 underscores :)
19:34 ruoso heh
19:34 ruoso I mean, did you see the code snippet in the paste?
19:34 ruoso that's how to implement CPS in SMOP... :)
19:36 TimToady cool!
19:37 pugs_svn r19476 | ruoso++ | [smop] starting to implement the first test. Small Perl 6 code snippet implementing Continuation-Passing-Style during object destruction
19:39 pugs_svn r19477 | ruoso++ | [smop] implementation-specific-and-exclusive-macros follow the ___FOO___ name pattern, and can only be used as in ___FOO___(arg) and never as arg.___FOO___.
19:40 * ruoso going to some bar &
19:42 * ruoso will backlog, as usual...
19:44 jisom joined #perl6
20:06 meppuru joined #perl6
20:07 pugs_svn r19478 | lwall++ | Some updates to reflect distinctions between lexical graphemes/codepoints/bytes scopes
20:15 TimToady I think I may also do away with the Uni type, since we now always have everything based on Int values, including grapheme ids
20:16 TimToady and chr/ord simply to Int/Str as people would expect from p5
20:16 TimToady s/to/do
20:17 TimToady (least surprise, and all that...)
20:29 thoughtpolice joined #perl6
20:30 rhr TimToady: I find Uni to be somewhat confusing, so that would be cool with me
20:44 alester joined #perl6
20:44 pugs_svn r19479 | lwall++ | [Functions] did away with the Uni type
20:47 polettix joined #perl6
21:14 mnemonic joined #perl6
21:15 mnemonic hi
21:15 moritz_ hi
21:15 pbuetow hi
21:16 pmurias hello
21:16 TimToady lo
21:20 pbuetow jo
21:27 meppuru joined #perl6
21:49 shachaf joined #perl6
22:11 dduncan joined #perl6
22:11 dduncan left #perl6
22:13 cosimo joined #perl6
22:21 meppl joined #perl6
22:25 thoughtpolice joined #perl6
22:26 Torment joined #perl6
22:48 meppl joined #perl6
22:51 braceta joined #perl6
22:54 pugs_svn r19480 | cosimo++ | moved quotemeta.t to t/spec/S29-str/
22:54 moritz_ there is quotemeta in p6?
22:55 moritz_ for backward compatibility with p5 regexes?
22:55 pugs_svn r19481 | cosimo++ | moved index.t to t/spec/S29-str/
22:56 moritz_ cosimo: that quotemeta commit just delete a file, it didn't add it in another place
22:56 cmarcelo joined #perl6
22:56 moritz_ cosimo: same with index.t
22:56 moritz_ ALL UR TESTS ARE BEL0NG TO US!
22:57 mnemonic left #perl6
23:08 pugs_svn r19482 | cosimo++ | [t] Refactored reverse (Str & List) tests into t/spec/S29-list/reverse.t
23:08 pugs_svn r19482 | cosimo++ | Changed todo marker to new compiler directive
23:08 pugs_svn r19483 | cosimo++ | Test suite refactoring. Moved index.t to t/spec/S29-str/
23:11 cosimo Should be everything in place now.
23:11 cosimo moritz_: thanks for the notice!
23:14 jferrero joined #perl6
23:18 TimToady still don't see quotemeta
23:19 TimToady did you check in lower down than your "svn mv"?
23:19 cosimo TimToady: yes, I see it here: http://svn.pugscode.org/pugs​/t/spec/S29-str/quotemeta.t
23:19 TimToady oh, nevermind, I see it
23:19 TimToady it does A before D, which makes sense
23:20 cosimo Seems to me that there's no missing file now
23:22 moritz_ confirmed
23:23 ruoso joined #perl6
23:26 pugs_svn r19484 | cosimo++ | Test suite refactoring.
23:26 pugs_svn r19484 | cosimo++ | Moved ord & chr tests in t/spec/S29-conversions/
23:33 pugs_svn r19485 | cosimo++ | Test suite refactoring.
23:33 pugs_svn r19485 | cosimo++ | Removed chomp.t from builtins.
23:33 pugs_svn r19485 | cosimo++ | Was already merged in t/spec/S29-str/chomp.t
23:34 meppl joined #perl6
23:42 BillR joined #perl6
23:50 braceta joined #perl6
23:50 pugs_svn r19486 | cosimo++ | Changed todo markers with new compiler directives
23:54 pugs_svn r19487 | cosimo++ | Changed todo markers to new compiler directives (S29-str/chop.t)
23:56 pugs_svn r19488 | cosimo++ | Changed todo markers with new compiler directives (S29/comb.t)
23:57 meppl joined #perl6
23:58 cosimo To anyone on these frequencies. Is it correct to mark todos as `#?pugs:' rather than `#?perl6:' ?
23:58 [particle] yes, but pugs doesn't understand it yet
23:58 TimToady sure, if it's an old todo
23:58 lambdabot joined #perl6
23:58 TimToady pugs will be taught
23:58 TimToady with our fantastic new preprocessor
23:59 cosimo yes, I mean old ones that were already in the tests.
23:59 TimToady go ahead and change them
23:59 [particle] pugs is old & busted. <the implementation yet to be named> is the new hotness.
23:59 japhb_ TimToady: to bad we can't just pipe it through Tom anymore ....
23:59 TimToady it's a shell game

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo