Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2007-12-26

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:03 Limbic_Region joined #perl6
00:05 theorb joined #perl6
00:13 fglock joined #perl6
00:30 agjal joined #perl6
01:05 penk joined #perl6
01:10 araujo joined #perl6
01:14 cnhackTNT joined #perl6
01:33 lyokato joined #perl6
01:59 oneuponzero joined #perl6
02:14 oneuponzero left #perl6
02:35 agjal good night
02:54 jatt` joined #perl6
02:57 jatt`` joined #perl6
03:04 kanru joined #perl6
03:48 neuromu joined #perl6
04:19 ruz joined #perl6
04:44 chacha_chaudhry joined #perl6
04:44 amnesiac joined #perl6
04:55 drbean_ joined #perl6
05:41 poppper joined #perl6
06:11 Ashizawa joined #perl6
06:29 kanru joined #perl6
06:41 Aankhen`` joined #perl6
07:16 Schwern joined #perl6
07:42 jedai joined #perl6
08:04 cnhackTNT joined #perl6
08:36 franck___ joined #perl6
08:40 franck____ joined #perl6
08:42 pjcj joined #perl6
08:46 franck_____ joined #perl6
08:53 Ashizawa joined #perl6
09:11 franck___ joined #perl6
09:22 Caelum joined #perl6
09:24 pugs_svn r19245 | Darren_Duncan++ | ext/Muldis-DB/ : added new Example Engine file Value.pm, replaces the previous PhysType.pm
09:41 Schwern How does one declare a class method?
09:42 franck____ joined #perl6
10:04 Juerd_ Schwern: method foo { ... }
10:06 Juerd_ Oh, method ^foo { ... }
10:08 Schwern Seriously?
10:09 Schwern That seems like huffman encoding gone wild
10:14 Tene Schwern: howso?
10:17 franck___ joined #perl6
10:17 Schwern How often does one declare a class method that they can't write "class"?
10:17 Schwern And what does ^ have to do with a class?
10:17 Tene ^ is used in a few other places to refer to class methods.
10:18 Schwern That doesn't answer either question.
10:18 Tene $obj.^foo is the same as $obj.WHAT.foo, as I recall
10:18 Schwern HOW
10:18 Tene Ahh, right.
10:19 Tene Oh, you're right, that isn't what you asked.
10:19 Schwern Didn't mean for that to come our rude, but I just watched the Matrix Reloaded again. :)
10:19 Schwern Nevermind
10:20 Tene No, no rudeness.
10:20 Schwern What a second
10:20 Schwern $obj.^foo... but ^foo is a class method...
10:20 Schwern And you're calling it like an object method
10:21 Tene Yes, used when you want to find out stuff about the class.
10:21 Tene "Do something with the class that this object is an instance of."
10:21 Schwern I see.
10:22 Schwern At least its internally consistent.
10:23 Schwern I'm sure there's some clever mnemonic like "^ is an up arrow indicating you should move up the aristotlean conceptual hierarchy from object to class" :)
10:24 Tene Heh, sounds as good as any other mneumonic.
10:48 franck____ joined #perl6
10:53 pmurias joined #perl6
10:54 pmurias Schwern: $foo.^method is the shortcut for $foo.HOW.method
10:55 pmurias they are more like metaclass instance methods than class methods
11:04 franck_____ joined #perl6
11:09 agjal joined #perl6
11:31 fglock joined #perl6
11:33 franck___ joined #perl6
11:41 lichtkind joined #perl6
11:50 wolverian http://comments.gmane.org/gmane.comp.lang.perl.perl5.porters/52692 interesting issue with regexes and copy semantics
11:50 lambdabot Title: gmane.comp.lang.perl.perl5.porters, http://tinyurl.com/26m4de
11:57 fglock [particle]: ping
12:00 ludan joined #perl6
12:14 franck___ joined #perl6
12:14 turrepurre joined #perl6
12:20 kanru joined #perl6
12:26 lichtkind sorry for that question but how to text an print output of a script?
12:29 fglock joined #perl6
12:34 pmurias joined #perl6
12:38 pmurias joined #perl6
13:19 amnesiac joined #perl6
13:20 devogon joined #perl6
13:23 pugs_svn r19246 | ruoso++ | [yap6] YAP6_NEW actually doesnt make sense...
13:26 ruoso fglock, the Scalar, List and Hash need ways to coerce between themselves, doesn't it?
13:27 ruoso I mean, Scalar should have a List and Hash methods and vice-versa for the there
13:27 ruoso s/the there/the three/
13:28 fglock__ joined #perl6
13:31 fglock__ ruoso: re coercion, there are several methods and functions, such as: @(), %(), +(), ~(), ?()
13:34 fglock__ and: postcircumfix:<{ }>, postcircumfix:<[ ]>, .Str, .true, ...
13:37 pmurias fglock__: have you seen my question about the containers?
13:37 dwave_ joined #perl6
13:38 fglock__ pmurias: I think v6.pm does have enough support containers, but I need to check it again to be sure
13:38 fglock__ support for
13:41 pmurias lunch&
13:41 fglock__ I think it's reasonable to backport kp6 to v6.pm,
13:41 fglock__ like I'm trying to do with kp6->perl6
13:48 franck___ joined #perl6
14:03 pugs_svn r19247 | ruoso++ | [yap6] all dispatchers have all the coercion methods...
14:14 fglock__ pmurias: re container methods, some things - binding, a "readonly" trait, a "type"
14:16 lorn joined #perl6
14:28 pbuetow joined #perl6
14:37 Psyche^ joined #perl6
14:42 Psyche^_ joined #perl6
14:48 alc joined #perl6
14:59 fglock__ [particle]: ping
15:14 njbartlett joined #perl6
15:15 chris2 joined #perl6
15:20 fglock__ [particle]: ../../parrot nqp.pbc t/10-sub.t  # Segmentation fault
15:21 [particle] fglock: that happens occasionally. lemme diagnose that today, after i rebuild parrot
15:26 fglock__ thanks
15:26 fglock__ [particle]: I'm playing with parrot scratchpads, I'm trying out PIR+nqp
15:26 pmurias joined #perl6
15:26 pmurias b
15:26 [particle] excellent
15:27 fglock__ no progress yet - the hardcoded ":outer" is hard to play with
15:29 pmurias is there an official example of a custom container?
15:29 fglock__ I wonder if I could generate parrot bytecode "by hand", or use low-level pasm
15:29 fglock__ pmurias: I don't think so
15:31 fglock___ joined #perl6
15:31 fglock__ pmurias: there is "Lvalue subroutines" in S06
15:32 pmurias fglock__: i think i know how Scalar's should work
15:32 pmurias i'm not sure about VAR(%hash) and VAR(@array)
15:33 fglock__ VAR() exists for Scalar only
15:33 fglock__ (S12)
15:34 pmurias thanks that cleared a lot
15:34 fglock__ spec-searchplugin ++
15:40 jhorwitz joined #perl6
15:40 fglock__ compiling a PIR program to PASM, it seems to just throw away the :outer information
15:41 fglock__ looking for a decompiler
15:42 [particle] there isn't one
15:42 fglock__ hmm - writing a decompiler, then :P
15:44 fglock__ [particle] is there a difference between compilers/ and languages/ ?
15:44 drrho joined #perl6
15:45 [particle] yes, compilers/ are guaranteed to ship with parrot
15:46 [particle] languages/ is a place where hll devs can work in the parrot tree, but many of these will be ripped out later
15:46 fglock__ thanks
15:46 fglock__ reading compilers/imcc
15:48 fglock__ "imcc.l" has this message: This file is generated automatically by the Parrot build process from the file compilers/imcc/imcc.l
15:48 fglock__ :)
15:49 [particle] yes, it does, inside a %top{ } block
15:49 [particle] so the output file that flex generates will have that at the top of it
15:50 fglock__ got it
15:50 [particle] maybe we can put a comment in there before %top{} to mention that the following block isn't to be taken literally
15:53 pmurias [particle]: aren't compilers/ the compiler tools?
15:53 [particle] compilers/ contains imcc, which isn't a compiler tool, it's the pir/pasm compiler
15:54 [particle] we haven't been including that in our definition of pct
15:55 [particle] it's possible that we could create a pir->pbc emitter, then we could ship a headless parrot (no compilers) and run bytecode on it
15:55 [particle] right now, only imcc can do that (emit pbc)
16:02 Psyche^ joined #perl6
16:03 fglock__ looking at newclosure() - pdd20 seems to imply that LexEnv is settable
16:03 fglock__ but we may need a dummy :outer anyway
16:03 [particle] there may be tests to look at
16:03 fglock__ looking
16:07 Psyche^_ joined #perl6
16:08 fglock__ t/op/lexicals.t
16:08 fglock__ I could add probably add some set_* methods for kp6
16:13 jferrero joined #perl6
16:17 fglock__ hmm - get_lexenv() is implemented in PIR
16:20 fglock__ this doesn't help - I think we need a Sub.set_outer()
16:27 [particle] correct. we do need that, and don't have it yet
16:27 fglock__ [particle]: if I subclassed 'Sub', I don't think there is a way to actually instantiate mySubs, right?
16:28 fglock__ so I need to add a method to Sub itself
16:28 ludanolo joined #perl6
16:28 [particle] i haven't tried it, but it's possible if you map your subclassed Sub to Sub using the .HLL directive, it'll Just Work
16:29 fglock__ the same way perl6 does it, right?
16:30 fglock__ I'll try it out
16:30 [particle] i'm not sure it'll work, but it's worth trying
16:34 fglock__ is there a way to mix C and PIR in the same file?
16:37 [particle] no
16:38 [particle] what do you want to do? if you need to write a pmc, i can help
16:43 pugs_svn r19248 | fglock++ | [kp6-nqp] added Sub.pmc stub
16:43 fglock__ sure - I need to get the parameter and store it to sub->outer_sub
16:44 fglock__ and then make a test program
16:44 fglock__ v6/v6-KindaPerl6/misc/nqp/outer.pmc
16:46 fglock__ the usage is likely to be: $inner.set_outer($outer)
16:47 [particle] i expect you'll take a PMC *outer param, and set sub->outer_sub
16:49 fglock__ yes, I just found an example
16:52 fglock__ what is the command line to compile a .pmc?
16:52 [particle] you'll have to change the const nature of sub, though
16:52 [particle] look it perl6's makefile
16:53 idiotmax joined #perl6
16:53 fglock__ ok
16:53 |Jedai| joined #perl6
16:59 fglock__ $ perl tools/build/pmc2c.pl --dump ~/pugs-svn-2/trunk/v6/v6-KindaPerl6/misc/nqp/outer.pmc  # no pmclass found
17:01 [particle] perhaps you should name it perl6sub.pmc instead of outer.pmc or modify pmclass to say 'Outer' instead of 'Perl6Sub'
17:03 fglock__ fixed
17:06 pugs_svn r19249 | fglock++ | [kp6-nqp] added preprocessing instructions
17:12 pugs_svn r19250 | fglock++ | [kp6-nqp] fixed module name, compilation instructions don't work yet
17:21 pugs_svn r19251 | fglock++ | [kp6-nqp] perl6sub.pmc compiles ok
17:22 fglock__ hmm - now what
17:29 fglock__ what is the PIR/nqp command to load a pmc
17:32 [particle] if the library is compiled in, new 'Other' should do it
17:32 [particle] (in pir)
17:32 [particle] $P0 = new 'Other' # sorry
17:33 fglock__ is there a command for "dynamic loading"?
17:33 rindolf joined #perl6
17:33 fglock__ compiled in == copied to parrot lib/ ?
17:34 [particle] oh, no. there are makefile commands for that
17:34 [particle] eg. PERL6_GROUP
17:35 fglock__ you mean Parrot makefile? I'm currently in v6/v6-KindaPerl6/misc/nqp/
17:35 [particle] look in perl6's makefile
17:35 fglock__ ok
17:37 fglock__ I need to create a config/makefiles/root.in, right?
17:37 [particle] do you have a languages dir under parrot?
17:37 fglock__ yes
17:38 fglock__ you mean, languages/kp6? no
17:38 [particle] okay, yes, config/... is better then
17:38 [particle] yes
17:38 fglock__ I wonder if it would be easier to patch perl6 instead
17:39 fglock__ or nqp
17:39 [particle] if all you need is that one method, i'd try that first
17:39 [particle] not nqp, no runtime there
17:40 fglock__ I'll try to organize kp6-parrot as a perl6 patch
17:45 lichtkind_ joined #perl6
17:47 pmurias fglock__: you mean the runtime?
17:48 pugs_svn r19252 | fglock++ | [kp6-parrot] created v6/v6-KindaPerl6/parrot
17:48 pugs_svn r19252 | fglock++ | - this directory replicates the Parrot directory structure;
17:48 pugs_svn r19252 | fglock++ | it will contain patches to "parrot" and "perl6"
17:48 fglock__ pmurias: yes, "perl6" and "kp6" will share the runtime
17:49 pmurias fglock__: that's good
17:52 fglock__ it seems the BEGIN and .perl code was lost - I'll patch that in later
17:54 fglock__ what's the opposite of "fork"?
17:54 [particle] join
17:55 fglock__ thanks
18:00 pugs_svn r19253 | fglock++ | [kp6-parrot] added perl6sub.pmc from misc/nqp;
18:00 pugs_svn r19253 | fglock++ | - added config file from "perl6" (unmodified)
18:01 fglock__ now I need a merge script
18:06 pugs_svn r19254 | fglock++ | fixed last commit
18:08 pugs_svn r19255 | fglock++ | [kp6] fixed last(2) commit
18:13 fglock__ looking for a "merge" in cpan
18:15 fglock__ clkao: ping
18:16 fglock__ clkao: I wonder if svk could handle merging files in different repositories
18:21 jjore-m joined #perl6
18:26 pmurias fglock__: editors such as vim and emacs have a merge function
18:26 pmurias see vimdiff
18:27 fglock__ pmurias: I'm writing an automated merge
18:27 ajs joined #perl6
18:27 ajs left #perl6
18:28 pmurias fglock__: there is a merge command on my system
18:29 fglock__ sure, but I need one that knows which files to merge, etc
18:29 fglock__ I forget command line switches quickly
18:29 turrepurre_ joined #perl6
18:30 pmurias from app-text/rcs
18:30 pmurias (it's a gentoo box)
18:38 lichtkind_ TimToady: can i make a suggestion due the titles of the synopses?
18:42 pugs_svn r19256 | fglock++ | [kp6-parrot] added merge.pl script
18:42 pugs_svn r19256 | fglock++ | - shows how to diff and patch "kp6-parrot" x "perl6"
18:45 gbacon joined #perl6
18:48 ludanolo joined #perl6
18:58 pugs_svn r19257 | fglock++ | [kp6-parrot] added t/kp6/01-outer.t (fails)
19:00 iblechbot joined #perl6
19:01 pugs_svn r19258 | fglock++ | [kp6-parrot] fixed test; does nothing useful yet
19:01 fglock__ [particle]: I think I've got everything in place - now testing...
19:02 TimToady lichtkind: what's that?
19:03 lichtkind TimToady: howdy, im currently resurecting the perl6::bible from the grave, its now called perl6::Doc but one of the problems i ran into that the titles of the S, E and A are not CPAN friendly
19:04 fglock__ "Method 'set_outer' not found for invocant of class 'Closure'"
19:04 lichtkind TimToady: a perfekt title is : =head1 NAME
19:04 lichtkind Synopsis_01 - Overview
19:05 lichtkind TimToady: i wanted the rename by myself but there were no longer in Pugs SVN
19:06 fglock__ lichtkind: how about fetching the most up-to-date text at Makefile.PL time?
19:06 fglock__ (maybe it already does this)
19:06 lichtkind fglock__: what you mean?
19:07 fglock__ instead of bundling the documentation with the module
19:07 lichtkind fglock__: yes i had this idea too
19:07 fglock__ cool
19:07 lichtkind i already have an script that does that :9
19:08 lichtkind iit has to be cleaned and styled  up a bit but it works and isnt that dirty
19:11 pugs_svn r19259 | fglock++ | [kp6-parrot] added Closure class
19:11 pugs_svn r19259 | fglock++ | - trying to fix this error message in "perl6":
19:11 pugs_svn r19259 | fglock++ | "Method 'set_outer' not found for invocant of class 'Closure'"
19:13 fglock__ I'm still getting the "Method not found" error
19:14 fglock__ [particle]: do I need to modify the "perl6_group" files somwhow?
19:15 fglock__ some
19:15 fglock__ home &   # will backlog
19:37 Auzon Or not ;)
19:40 [particle] TimToady: no call today, correct?
19:50 lichtkind why synopses no longer in pugs\docs\Perl6\Spec ?
19:50 Lorn_ joined #perl6
19:58 ispy_ joined #perl6
19:58 Tene [particle]: that was the decision last I heard.
20:00 Tene Oh, nevermind, I'm thinking of something else.
20:07 jedai joined #perl6
20:29 __Ace__ joined #perl6
21:05 pmurias lichtkind: to restrict the number of people able to edit them
21:05 lichtkind pmurias: as i could think myself i just want to get shure there was no accident :)
21:09 Limbic_Region joined #perl6
21:15 lichtkind pmurias: thanks
21:20 pmurias lichtkind: i only guessed the reason, but i'm sure it isn't an accident
21:46 REPLeffect joined #perl6
21:46 jjore-m left #perl6
21:52 jjore-w left #perl6
22:06 justatheory joined #perl6
23:01 chris2 joined #perl6
23:18 fglock___ joined #perl6
23:23 pugs_svn r19260 | fglock++ | [kp6-parrot] fixed makefile
23:37 fglock___ joined #perl6
23:38 fglock___ left #perl6
23:38 pugs_svn r19261 | ruoso++ | [yap6] declared the YAP6__CORE__num type, which is the return of the YAP6_NUMBR macro. I think all the coercions are declared now.
23:39 fglock_ joined #perl6
23:40 r0bby joined #perl6
23:41 meppel joined #perl6
23:49 pugs_svn r19262 | ruoso++ | [yap6] some cleaning and reordering of the yap6.h file
23:49 pugs_svn r19263 | ruoso++ | [yap6] some cleaning and reordering of the yap6.h file
23:57 fglock__ joined #perl6

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

Perl 6 | Reference Documentation | Rakudo