Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2007-09-21

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:05 theorbtwo joined #perl6
00:06 d4l3k_ joined #perl6
00:07 dalek joined #perl6
00:17 dduncan pugscode is back now
00:19 davidfetter yay!
00:21 Juerd joined #perl6
00:32 thoughtpolice joined #perl6
00:45 ferreira joined #perl6
00:55 takanori joined #perl6
01:06 c9s joined #perl6
01:15 hoelzro joined #perl6
01:23 avar @tell fglock If you tell me what the hell you're trying to do with global.lisp maybe I could implement it >:)
01:23 lambdabot Consider it noted.
01:25 davidfetter left #perl6
01:27 ahmadz hi, is feather right now offline?
01:27 ahmadz @seen feather ? ;-)
01:27 lambdabot I haven't seen feather.
01:28 ahmadz @karma feather
01:28 lambdabot feather has a karma of 0
01:28 ahmadz @localtime feather
01:30 avar @localtime avar
01:30 lambdabot Local time for avar is Fri Sep 21 01:30:08 2007
01:30 ahmadz @localtime ahmadz
01:30 lambdabot Local time for ahmadz is Fri Sep 21 04:30:35 2007
01:30 ahmadz @karma ahmadz
01:30 lambdabot You have a karma of 2
01:30 ahmadz not bad ;-)
01:30 daxim_ joined #perl6
01:31 hoelzro @localtime hoelzro
01:31 avar ahmadz++
01:31 avar YOU GROW STRONGAH
01:31 ahmadz @karma ahmadz
01:31 lambdabot You have a karma of 3
01:31 hoelzro what, have I no local time?
01:31 hoelzro @karma hoelzro
01:31 lambdabot You have a karma of 3
01:31 avar @remotetime hoelzro
01:31 lambdabot Unknown command, try @list
01:31 avar @list
01:31 lambdabot http://www.cse.unsw.edu.au​/~dons/lambdabot/COMMANDS
01:32 ahmadz i slept to wake and work on runpugs in the morning but pugscode/feather seem to be offline :-(
01:33 hoelzro =(
01:34 ferreira when feather is online again, the next micro-article will there at http://feather.perl6.nl/~ferrei​ra/perl6-operators/repeat.html and repeat.pod6
01:34 [particle1 joined #perl6
01:34 hoelzro you plan on working on pugs, ahmadz?
01:34 ahmadz runpugs
01:35 ahmadz hoelzro: finishing the tutorial part of "run perl6 on your browser"
01:35 hoelzro ah
01:35 ferreira Juerd warned about interruptions from 0AM to 4AM. I forgot that and don't have a single copy of the article in another machine.
01:35 ferreira see you tomorrow.
01:35 ferreira left #perl6
01:52 Jamtech joined #perl6
01:56 ahmadz @localtime juerd
01:56 lambdabot Local time for Juerd is Fri Sep 21 03:56:48 2007
02:05 Schwern joined #perl6
02:12 d4l3k_ joined #perl6
02:12 pasteling "pugsbot_r17185" at 193.200.132.135 pasted "Pugs build failure" (6 lines, 278B) at http://sial.org/pbot/27608
02:12 pugs_svnbot joined #perl6
02:12 pugsbot_r17185 joined #perl6
02:23 ajelenak joined #perl6
02:27 Juerd joined #perl6
02:27 Juerd Flipped the wrong switch, sorry
02:29 vdrab joined #perl6
03:02 ryporter joined #perl6
03:02 takanori_ joined #perl6
03:05 elmex_ joined #perl6
03:06 amnesiac joined #perl6
03:20 justatheory joined #perl6
03:20 salzh joined #perl6
03:25 lyokato joined #perl6
03:44 Chris joined #perl6
03:52 mj41__ joined #perl6
04:05 IllvilJa joined #perl6
04:34 mj41__ joined #perl6
04:34 amnesiac joined #perl6
04:34 rgs joined #perl6
04:34 platypus joined #perl6
04:34 broquaint joined #perl6
04:34 beppu joined #perl6
04:34 shmem joined #perl6
04:34 ingy joined #perl6
04:34 perlbot joined #perl6
04:34 meppl joined #perl6
04:35 dduncan left #perl6
04:42 vsoni joined #perl6
04:51 PerlJam joined #perl6
04:54 drupek1245 joined #perl6
05:41 Aankhen`` joined #perl6
05:42 hoelzro left #perl6
05:44 Ashizawa joined #perl6
05:49 lyokato joined #perl6
05:53 Aankhen`` avar: Yeesh, why on Earth did you change everything to camelCase?
05:53 Aankhen`` Rather, even worse, hyphenated-CamelCase.
05:53 platypus The worse of all worlds.
05:54 * Aankhen`` grins.
05:54 iblechbot joined #perl6
05:59 avar Aankhen``: to mirror the p6 names
06:01 Aankhen`` "Bah", I say.  "Bah".
06:02 Aankhen`` avar: A few things… first, what is the value of a container?  Second, why do you specify a VALUE slot for kp6-Hash when it already inherits one from kp6-Container?
06:03 avar Aankhen``: hlagh
06:03 Aankhen`` Third, why did you remove kp6-object from kp6-Container's superclass list?
06:03 avar clean it up then:)
06:03 Aankhen`` Answer the first and third questions first. :-P
06:13 pugs_svnbot r18034 | Aankhen++ | [kp6-lisp]
06:13 pugs_svnbot r18034 | Aankhen++ | * kp6-Container inherits from kp6-Object.
06:13 pugs_svnbot r18034 | Aankhen++ | * added a BIT slot to kp6-Object.
06:13 pugs_svnbot r18034 | Aankhen++ | * changed STR method to be defined with DEFGENERIC.
06:13 pugs_svnbot diff: http://dev.pugscode.org/changeset/18034
06:13 lambdabot Title: Changeset 18034 - Pugs - Trac
06:16 pugs_svnbot r18035 | Aankhen++ | * [kp6-lisp] removed references to KP6-GLOBAL package.
06:16 pugs_svnbot diff: http://dev.pugscode.org/changeset/18035
06:16 lambdabot Title: Changeset 18035 - Pugs - Trac
06:17 avar fix say() kthx
06:17 pugs_svnbot r18036 | Aankhen++ | * [kp6-lisp] indentation in GLOBAL.lisp
06:17 pugs_svnbot diff: http://dev.pugscode.org/changeset/18036
06:17 lambdabot Title: Changeset 18036 - Pugs - Trac
06:20 pugs_svnbot r18037 | Aankhen++ | * [kp6-lisp] initarg for kp6-Bit's BIT slot.
06:20 pugs_svnbot diff: http://dev.pugscode.org/changeset/18037
06:20 lambdabot Title: Changeset 18037 - Pugs - Trac
06:21 Aankhen`` o_O
06:21 Aankhen`` Interesting hash stuff.
06:22 avar light it up
06:23 Aankhen`` Sure.
06:24 Aankhen`` Seeing as how you wrote KP6-LOOKUP to operate on an instance of KP6-HASH and then pass it a quoted symbol instead, I think I may have found the problem.
06:29 * Aankhen`` goes to shower.
06:33 devogon joined #perl6
06:55 drrho joined #perl6
07:02 BinGOs joined #perl6
07:02 pdcawley joined #perl6
07:13 Aankhen`` avar: Okay, got the packages things done in a simple form.
07:13 pugs_svnbot r18038 | Aankhen++ | [kp6-lisp]
07:13 pugs_svnbot r18038 | Aankhen++ | * added kp6-Package class.
07:13 pugs_svnbot r18038 | Aankhen++ | * removed initarg for VALUE slot of kp6-Hash.
07:13 pugs_svnbot r18038 | Aankhen++ | * added Object.lisp to load list.
07:13 pugs_svnbot r18038 | Aankhen++ | * fixed syntax for kp6-Object class definition.
07:13 pugs_svnbot r18038 | Aankhen++ | * added package manipulation functions in GLOBAL.lisp
07:13 pugs_svnbot r18038 | Aankhen++ | * changed Code_say to use TERPRI instead of FORMAT for the newline.
07:13 pugs_svnbot diff: http://dev.pugscode.org/changeset/18038
07:13 lambdabot Title: Changeset 18038 - Pugs - Trac
07:16 Aankhen`` I'm out, later.
07:16 mr_ank joined #perl6
07:18 ahmadz joined #perl6
07:26 pdcawley_ joined #perl6
07:34 wolverian joined #perl6
07:39 Azure-BOT joined #perl6
07:41 Schwern joined #perl6
07:52 franck___ joined #perl6
07:58 ahmadz joined #perl6
08:21 jisom anyone around?
08:25 orafu joined #perl6
08:29 jisom the bot might answer me, http://74.134.230.123/
08:55 chris2 joined #perl6
09:15 cognominal_ joined #perl6
09:16 cognominal_ joined #perl6
09:17 jisom_ joined #perl6
09:34 pugs_svnbot r18039 | ruoso++ | [kp6] prefix:<$> calls scalar on the object. Match.scalar returns result or self.str
09:34 pugs_svnbot diff: http://dev.pugscode.org/changeset/18039
09:34 lambdabot Title: Changeset 18039 - Pugs - Trac
09:40 pugs_svnbot r18040 | ruoso++ | [kp6] using attr directly. Accessors are bad right now...
09:40 pugs_svnbot diff: http://dev.pugscode.org/changeset/18040
09:40 lambdabot Title: Changeset 18040 - Pugs - Trac
09:43 pugs_svnbot r18041 | ruoso++ | [kp6] new failing test. Object acessors are not working...
09:43 pugs_svnbot diff: http://dev.pugscode.org/changeset/18041
09:43 lambdabot Title: Changeset 18041 - Pugs - Trac
09:49 pugs_svnbot r18042 | ruoso++ | [kp6] object accessors now work...
09:49 pugs_svnbot diff: http://dev.pugscode.org/changeset/18042
09:49 lambdabot Title: Changeset 18042 - Pugs - Trac
10:07 rindolf joined #perl6
10:19 ahmadz joined #perl6
10:22 pugs_svnbot r18043 | ruoso++ | [kp6] params to constructor inside rule compile fine but do nothing...
10:22 pugs_svnbot diff: http://dev.pugscode.org/changeset/18043
10:22 lambdabot Title: Changeset 18043 - Pugs - Trac
10:30 gaal joined #perl6
10:30 snearch joined #perl6
10:30 gaal Haskell is less painful than LISP: http://scienceblogs.com/loom/2007​/09/21/mathematical_markings.php
10:30 lambdabot Title: The Loom : Mathematical Markings
10:36 flokuehn joined #perl6
10:40 devogon joined #perl6
10:40 ferreira joined #perl6
10:41 ferreira comments and suggestions on the draft micro-article on repeat operators are very welcome. Read it at http://feather.perl6.nl/~ferrei​ra/perl6-operators/repeat.html
10:41 lambdabot Title: repeat.pod6
10:44 Aankhen`` joined #perl6
10:48 cognominal_ joined #perl6
10:57 sunnavy joined #perl6
11:09 ruoso ferreira, shouldn't Marker->new be Marker.new ?
11:10 pugs_svnbot r18044 | fglock++ | [kp6] fix ::Obj() emitter
11:10 pugs_svnbot diff: http://dev.pugscode.org/changeset/18044
11:10 lambdabot Title: Changeset 18044 - Pugs - Trac
11:12 Blwood joined #perl6
11:17 ferreira ruoso: oops. indeed
11:18 ferreira ferreira: fixed
11:19 ferreira hey, I am talking to myself ;-)
11:20 ferreira by the way, yesterday the article on stitching (the ~ operator) hitted the 12th position on The Hot 25. Maybe this series get popular :)
11:20 ferreira The Hot 25 is seen at the right of this O'Reilly page: http://www.oreillynet.com/blogs/
11:20 lambdabot Title: O'Reilly Network Blogs
11:26 fglock joined #perl6
11:26 szbalint cool
11:28 pugs_svnbot r18045 | fglock++ | [kp6] set svk eol, mime-type properties
11:28 pugs_svnbot diff: http://dev.pugscode.org/changeset/18045
11:28 lambdabot Title: Changeset 18045 - Pugs - Trac
11:33 c9s joined #perl6
11:40 fglock Aankhen``: ping
11:40 lambdabot fglock: You have 4 new messages. '/msg lambdabot @messages' to read them.
11:40 Aankhen`` fglock: What's up?
11:40 fglock OSX doesn´t know the difference between Package.lisp and package.lisp, and it messed up
11:41 Aankhen`` Whoops.
11:41 Aankhen`` Sorry, I'll fix that.
11:41 fglock thanks
11:41 Aankhen`` Give me a minute, waiting for the VM to be restored.
11:44 Aankhen`` Oh man.
11:47 Aankhen`` Done.
11:47 pugs_svnbot r18046 | Aankhen++ | * [kp6-lisp] renamed package.lisp to defpackage.lisp to prevent name conflicts.
11:47 pugs_svnbot diff: http://dev.pugscode.org/changeset/18046
11:47 lambdabot Title: Changeset 18046 - Pugs - Trac
11:47 Aankhen`` Sorry about the delay, my VM lost its network connection.
11:50 pugs_svnbot r18047 | ryporter++ | change double to single quotes, to fix a unit test
11:50 pugs_svnbot diff: http://dev.pugscode.org/changeset/18047
11:50 lambdabot Title: Changeset 18047 - Pugs - Trac
11:55 wolv joined #perl6
12:19 ahmadz joined #perl6
12:19 ahmadz kp6: my @a = <foo 1 2 3>;
12:19 exp_evalbot r18047: OUTPUT[syntax error at position 6, line 1 column 6:␤my @a = <foo 1 2 3>␤      ^ HERE␤]
12:20 ahmadz perl6: my @a = <foo 1 2 3>;
12:20 exp_evalbot kp6: OUTPUT[syntax error at position 6, line 1 column 6:␤my @a = <foo 1 2 3>␤      ^ HERE␤]
12:20 exp_evalbot ..pugs: OUTPUT[["foo", "1", "2", "3"]␤]
12:20 exp_evalbot ..p6: OUTPUT[Null PMC access in invoke()␤current instr.: 'parrot;Perl6::Compiler;main' pc 138 (perl6.pir:98)␤]
12:20 exp_evalbot ..nqp: OUTPUT[Statement not terminated properly at line 1, near "@a = <foo "␤current instr.: 'parrot;PGE::Util;die' pc 123 (runtime/parrot/library/PGE/Util.pir:83)␤called from Sub 'parrot;NQP::Grammar;statement_list' pc 3574 (src/Grammar_gen.pir:1275)␤called from Sub 'parrot;NQP::Grammar;TOP' pc
12:20 exp_evalbot ..537 (src/Grammar_gen.pir:103)␤called from Sub 'parrot;PCT::HLLCo...]
12:21 ahmadz pugs: my @a = <foo 1 2 3>;
12:21 exp_evalbot OUTPUT[["foo", "1", "2", "3"]␤]
12:23 Aankhen`` What the…
12:24 ruoso fglock, wooohooo
12:24 ruoso I made it
12:24 Aankhen`` Something blew away my defpackage.lisp.
12:24 ruoso it's weird and dirty
12:24 ruoso but works
12:25 pasteling "ruoso" at 82.154.252.25 pasted "named argument support in new" (27 lines, 1K) at http://sial.org/pbot/27635
12:25 pugs_svnbot r18048 | azawawi++ | [runpugs] minified version of runpugs.js
12:25 pugs_svnbot diff: http://dev.pugscode.org/changeset/18048
12:25 lambdabot Title: Changeset 18048 - Pugs - Trac
12:25 ruoso fglock, ^
12:25 fglock looking
12:27 fglock looks good :)
12:27 pugs_svnbot r18049 | Aankhen++ | * [kp6-lisp] fixed defpackage.lisp, updated export list.
12:27 pugs_svnbot diff: http://dev.pugscode.org/changeset/18049
12:27 lambdabot Title: Changeset 18049 - Pugs - Trac
12:27 * ruoso recompiling kp6-kp6 with that
12:29 cmarcelo joined #perl6
12:29 pugs_svnbot r18050 | Aankhen++ | * [kp6-lisp] fixed Package.lisp
12:29 pugs_svnbot diff: http://dev.pugscode.org/changeset/18050
12:29 lambdabot Title: Changeset 18050 - Pugs - Trac
12:32 pugs_svnbot r18051 | Aankhen++ | * [kp6-lisp] kp6-Package should inherit from kp6-Hash (temporary!).
12:32 pugs_svnbot diff: http://dev.pugscode.org/changeset/18051
12:32 lambdabot Title: Changeset 18051 - Pugs - Trac
12:44 ruoso fglock, echo -n 'say "Hello!"' | perl -Ilib-kp6-kp6-p5 kp6-kp6.pl | perltidy
12:44 pugs_svnbot r18052 | ruoso++ | [kp6] implemented arguments to new. WOOOOHOOOO... kp6 now emits *some* ast.. still wrong.. but emits...
12:44 pugs_svnbot diff: http://dev.pugscode.org/changeset/18052
12:44 lambdabot Title: Changeset 18052 - Pugs - Trac
13:06 pugs_svnbot r18053 | Aankhen++ | [kp6-lisp]
13:06 pugs_svnbot r18053 | Aankhen++ | * added foreign.lisp containing CL->PERL and PERL->CL methods.
13:06 pugs_svnbot r18053 | Aankhen++ | * removed the unused code in GLOBAL.lisp and changed the "say" function to use CL->PERL.
13:06 pugs_svnbot diff: http://dev.pugscode.org/changeset/18053
13:06 lambdabot Title: Changeset 18053 - Pugs - Trac
13:32 blindfish joined #perl6
13:42 jhorwitz joined #perl6
13:43 Boes joined #perl6
13:49 TJCRI joined #perl6
13:50 iblechbot joined #perl6
14:23 drupek1245 joined #perl6
14:27 junp joined #perl6
14:31 junp left #perl6
14:31 Limbic_Region joined #perl6
14:47 DarkWolf84 joined #perl6
14:47 calin joined #perl6
14:49 TJCRI joined #perl6
14:49 barney joined #perl6
14:50 rhizo joined #perl6
14:53 turrepurre joined #perl6
14:53 pugs_svnbot r18054 | fglock++ | [kp6] added Runtime::Perl6::Scope.pm
14:53 pugs_svnbot diff: http://dev.pugscode.org/changeset/18054
14:53 lambdabot Title: Changeset 18054 - Pugs - Trac
14:54 adelfino joined #perl6
14:55 Boes joined #perl6
14:55 amnesiac joined #perl6
14:57 LimbicRegion joined #perl6
14:58 rhr ferreira: $title.elems should be $title.chars.  also, sub prefix:<~> should probably be multi method *prefix:<~>
14:58 lambdabot rhr: You have 1 new message. '/msg lambdabot @messages' to read it.
14:59 rhr @messages
14:59 lambdabot ferreira said 3d 1h 35m 23s ago: Thanks. Fixed.
15:10 pugs_svnbot r18055 | fglock++ | [kp6] added a test for Scope.pm
15:10 pugs_svnbot r18056 | ruoso++ | recompile
15:10 pugs_svnbot diff: http://dev.pugscode.org/changeset/18056
15:10 lambdabot Title: Changeset 18056 - Pugs - Trac
15:24 franck___ joined #perl6
15:29 drupek1245 joined #perl6
15:35 ferreira rhr: I fixed the *prefix:<~> signature and used .chars as it should be. thanks
15:46 cognominal_ joined #perl6
15:52 barney joined #perl6
15:55 pugs_svnbot r18057 | fglock++ | [kp6] *exists()
15:55 pugs_svnbot diff: http://dev.pugscode.org/changeset/18057
15:55 lambdabot Title: Changeset 18057 - Pugs - Trac
15:58 pugs_svnbot r18058 | ruoso++ | [kp6] uses NamedArgument in Match at p5 token. Grab string from $_ if not defined
15:59 pugs_svnbot diff: http://dev.pugscode.org/changeset/18058
15:59 lambdabot Title: Changeset 18058 - Pugs - Trac
16:09 TimToady ferreira: actually, multi method *prefix:<~> is wrong several ways.
16:09 TimToady there's no point in putting "multi" on a method with no arguments
16:10 pugs_svnbot r18059 | fglock++ | [kp6] lazy hash lookup
16:10 pugs_svnbot diff: http://dev.pugscode.org/changeset/18059
16:10 lambdabot Title: Changeset 18059 - Pugs - Trac
16:10 TimToady nor is there any reason to put the *, since we're looking for a deep coercion on the object, and not using a global operator
16:11 TimToady and prefix:<~> by itself isn't going to specify the deep coercion anyway
16:12 ferreira well, it seems I do not understand operators like ~, +, and ?
16:12 ferreira How do I specify an object Marker to stringify as its attribute $.text ?
16:13 TimToady currently it's specced in S13 with "is deep" but I'm not terribly happy with that
16:13 justatheory joined #perl6
16:14 moritz_ TimToady: what about an additional coercion interface via roles?
16:15 moritz_ so a ~ first looks for a method prefix:<~>, and if it doesn't find one, a role "stringification" or whatever you will call it
16:15 ferreira multi sub *prefix<~> ( Marker $m ) is deep { $m.text }    ??
16:16 TimToady prefix:<~> is the name of a syntax, not the name of a deep operation
16:16 PerlJam ferreira: btw, you suck.  :-) When I applied for a microgrant to write about perl 6 regex, they said they were full up on documentation grants.  So it looks like you got the last one.
16:16 PerlJam though, it's my own fault for not applying sooner.
16:17 TimToady ferreira: that's almost according to current spec, but you don't need the * on a multi
16:17 TimToady but I still don't really like "is deep" when we should be using type names
16:18 TimToady we can already specify from the receiving type a function like Str()
16:18 pdcawley joined #perl6
16:18 TimToady so it seems to me that the sending class should have a way to override that
16:18 TimToady method Str or method :to(Str) or some such
16:19 pugs_svnbot r18060 | fglock++ | [kp6] fixed exists()
16:19 pugs_svnbot diff: http://dev.pugscode.org/changeset/18060
16:19 lambdabot Title: Changeset 18060 - Pugs - Trac
16:19 TimToady also the parens in (1) xx $x are unnecessary
16:19 PerlJam perl5 habits die hard :)
16:19 Limbic_Region joined #perl6
16:20 mj41__ joined #perl6
16:20 TimToady and certainly we want $x.Str to work, so for now maybe just make it "method Str"
16:21 TimToady maybe "method Str is export"
16:21 TimToady if you want to get the Str() form as well
16:22 [particle] are submethods still around?
16:22 TimToady yes
16:23 pugs_svnbot r18061 | fglock++ | [kp6]?\194?\160t/kp6/internals/01-scope.t pass
16:23 pugs_svnbot diff: http://dev.pugscode.org/changeset/18061
16:23 lambdabot Title: Changeset 18061 - Pugs - Trac
16:27 pdcawley_ joined #perl6
16:30 pugs_svnbot r18062 | ruoso++ | [kp6] p5 tokens are now part of the ast, all grammars test pass.
16:30 pugs_svnbot r18063 | ruoso++ | [kp6] fixing test plan...
16:30 pugs_svnbot diff: http://dev.pugscode.org/changeset/18063
16:30 lambdabot Title: Changeset 18063 - Pugs - Trac
16:30 ruoso wheeeeeeeee... only three tests failing
16:36 pugs_svnbot r18064 | ruoso++ | [kp6] GLOBAL::exit
16:36 pugs_svnbot diff: http://dev.pugscode.org/changeset/18064
16:36 lambdabot Title: Changeset 18064 - Pugs - Trac
16:37 fglock ruoso: coffee?
16:37 ruoso yep ;)
16:40 rhizo joined #perl6
16:40 pdcawley joined #perl6
16:42 justatheory joined #perl6
16:46 silug joined #perl6
16:58 hoelzro joined #perl6
17:03 hoelzro is svn down?  I can't svn update
17:05 ferreira hey, I lost the connection with feather: "Host does not exist"
17:07 ferreira TimToady: I will replace the scary "multi method ..." thing with "method Str { $.text }" which looks nicer.
17:09 ferreira So there is no point in saying "multi method no_arg { ... }" I thought it be a kind of multi declaration that took precedence over "multi sub one_arg( ThisClass obj )"
17:11 lumi joined #perl6
17:15 ferreira PerlJam: sorry to have let you down. It looks like they did some simple mathematics on the 10 microgrants.
17:16 ferreira Having used two for porting stuff (Steve Peters and Steve Pritchard), one for hardware (Juerd), one for travel (fglock, for evangelizing purposes), one for tools/application (Phil Crow), they closed the ones for documentation after the first.
17:17 Aankh|Clone joined #perl6
17:17 ferreira Probably the remaining are preferred for coding and testing, and other topics that have not been contemplated.
17:22 [particle] looks like perljam isn't around
17:22 [particle] but maybe he could get a microgrant for interpretive dance
17:24 ferreira yep, he could try it. particle: do you know why feather is gone and svn.pugscode.org as well?
17:24 [particle] no idea
17:24 kyrbe joined #perl6
17:25 moritz_ even dns for feather.perl6.nl is b0rked
17:25 kyrbe left #perl6
17:29 Aankh|Clone joined #perl6
17:30 fglock moritz_: i've been reading irc.pugscode.org while committing, but after a few hours it gets too big for my phone - i wonder if it would be possible to have a paginate option
17:30 fglock commuting
17:32 moritz_ fglock: would a plain text only version help as well?
17:32 fglock sure
17:32 ahmadz joined #perl6
17:32 fglock and it should be easier to implement
17:33 ahmadz hi
17:33 ahmadz is feather down?
17:35 ahmadz @seen moritz
17:35 lambdabot moritz has changed nick to moritz_.
17:35 lambdabot moritz_ is in #perl6. I last heard moritz_ speak 2m 40s ago.
17:35 hoelzro moritz is on vacation
17:35 hakobe_ joined #perl6
17:35 Aankhen`` joined #perl6
17:36 moritz_ not yet, but I will be from tomorow, 20H
17:36 hoelzro oh hi
17:36 moritz_ hi ;)
17:36 ahmadz moritz: hi ;-)
17:37 moritz_ I thought my notebook was b0rked, but occasionally it still works... that's why I thought I couldn't IRC to you guys before vacation ;)
17:38 ahmadz moritz: i used the camel moose pic as a console background on runpugs. it looked cool but then feather went down and so no commit :-(
17:39 moritz_ ahmadz: ;)
17:40 ahmadz moritz: and a new click-click sound for the runpugs keyboard is coming also ;-)
17:42 moritz_ ahmadz: please don't... I always hate sound on websites ;)
17:45 ahmadz moritz: consider it noted ;-)
17:45 moritz_ ;)
17:46 ahmadz moritz: i guess having a command !kp6 or kp6: is not useful on runpugs as it compiles far less than pugs atm
17:47 justatheory joined #perl6
17:48 * ahmadz thinks about !tutorial and !load example/*.pl in runpugs
17:49 moritz_ ahmadz: probably, yes
17:49 vel joined #perl6
17:56 snearch joined #perl6
17:59 Aankhen`` joined #perl6
18:00 moritz_ fglock: http://moritz.faui2k3.org/irclog/text.pl
18:00 moritz_ fglock: supports the same parameters as the normal 'out.pl'
18:00 Aankh|Clone joined #perl6
18:01 fglock i'm working with ruoso on a high level implementation of scopes for bootstrapping
18:01 fglock moritz_: thanks!!!
18:03 moritz_ fglock: you're welcome... more elaborate version to follow after my vacations ;)
18:05 moritz_ fglock: now the HTML versions have a "Plain Text" link
18:06 fglock moritz_: the phone wants to download it instead of displaying - but i think it's a problem on my side
18:07 moritz_ fglock: it's Content-Type: text/html
18:07 moritz_ fglock: perhaps you can configure it...
18:07 fglock yes, looking
18:07 dalek joined #perl6
18:09 hoelzro left #perl6
18:12 fglock hmm - it just doesn't support "text"
18:13 ahmadz fglock: what kinda of phone?
18:13 fglock ahmadz: nokia 6070
18:14 ahmadz feather is back ;-)
18:15 fglock yay - committing...
18:15 ahmadz 11% packet loss when you ping it for a while
18:18 moritz_ pugs_svnbot is missing ( hilight diakopter here...)
18:18 ahmadz fglock: most phones dont support text/plain, your need wml or xhtml (http://www.forum.nokia.com/devices/6070)
18:18 lambdabot Title: Device Details -- Nokia 6070
18:18 justatheory joined #perl6
18:19 ahmadz s/your/you/
18:19 fglock how about a simple-html version like, just replace \n with <br/>
18:19 ahmadz fglock: xhtml or wml will do the trick
18:20 ahmadz fglock: however on wml, you need to escape things a lot
18:21 Aankhen`` Why not just HTML?
18:21 ahmadz moritz: seems to me than feather's dns was only down
18:23 ahmadz if it is well formed, then most phones should have no problem reading it.
18:23 ruoso left #perl6
18:23 Aankhen`` Indeed.
18:23 Aankhen`` The link you gave mentions HTML.
18:23 Aankhen`` They might leave out HTML in order to avoid the processing burden of error correction, but IIRC most phones do support it anyway.
18:23 fglock it does read html, but gives out of memory with big pages
18:24 Aankhen`` And not with XHTML?
18:24 ahmadz 4MB memory or less
18:26 ahmadz fglock: did u try google's mobilizer: http://www.google.com/gwt/n
18:26 lambdabot Title: Enter a URL
18:26 fglock ahmadz: will try
18:27 amnesiac_ joined #perl6
18:27 ahmadz fglock: we could actually provide a link to on the irc log for simplicity and it should transcode it easily.
18:27 fglock ahmadz: sure - it looks great
18:28 fglock http://www.google.com/gwt/n?u=http%3A%​2F%2Firc.pugscode.org&amp;_gwt_noimg=1
18:28 lambdabot Title: IRC log for #perl6, 2007-09-21, http://tinyurl.com/2g3p79
18:28 moritz_ it's not very readable though
18:28 moritz_ I can create a more readable, tiny version when I get back
18:29 fglock moritz_: the google version is good enough
18:29 moritz_ fglock: ok
18:29 ahmadz moritz: most phones = small screen+memory, so more formatting and the browser breaks.
18:32 ahmadz moritz, fglock:  i guess a version without the "left/joined" events would make sense and let google mobilizer do the rest.
18:33 fglock cool - it works now
18:35 moritz_ fglock: please try http://moritz.faui2k3.org/irclog/​text.pl?channel=perl6;2007-09-21 again - it now emits a html header
18:36 moritz_ and it omits all join/leave messages
18:36 ahmadz moritz_++
18:37 moritz_ from a technical point of view it's horrible ;)
18:37 moritz_ and it's not safe yet
18:37 fglock looking (where is lambdabot with the tinyurl?)
18:37 fglock http://moritz.faui2k3.org/irclog/​text.pl?channel=perl6;2007-09-21
18:37 PerlJam joined #perl6
18:38 moritz_ or just http://moritz.faui2k3.org/irclog/text.pl for today
18:40 fglock moritz_: is it marked as html? the phone wants to save it
18:41 moritz_ fglock: yes, it's send as text/html
18:42 moritz_ but it's in quirks mode (no doctype)
18:44 moritz_ fglock: perhaps a caching issue?
18:44 fglock maybe - cleaning
18:48 fglock nope
18:49 fglock 4 minutes to clean the cache and retry :P
18:49 moritz_ omg
18:49 moritz_ fglock: then you'll have to use google for the moment
18:49 fglock moritz_: np, that's good enough
18:50 moritz_ fglock: my fault, I'm too dump...
18:50 moritz_ fglock: I didn't update the server...
18:51 sunnavy joined #perl6
18:52 moritz_ fglock: try http://moritz.faui2k3.org/irclog/text.pl again please - last time now ;)
18:52 lambdabot Title: IRC Logs
18:52 fglock k
18:54 IllvilJa joined #perl6
18:55 fglock moritz_: out of memory
18:56 moritz_ fglock: ok, I give up
18:57 ahmadz phone development can be very difficult to test; i do this everyday ;-)
18:57 moritz_ decommute &
18:58 * ahmadz ponders about a wml version of runpugs ;-)
18:58 ahmadz check out this http://feather.perl6.nl:8096/runpugs/ ;-)
18:58 lambdabot Title: Run Perl 6 Now -- in your browser!
18:58 fglock ahmadz: that would be great
18:59 jisom joined #perl6
19:11 fglock svnbot is down
19:13 fglock home &
19:27 buubot joined #perl6
19:31 hoelzro joined #perl6
19:35 hoelzro hooray!  I can svn update again!
19:43 ferreira left #perl6
19:45 justatheory joined #perl6
19:50 Blwood joined #perl6
19:58 hoelzro left #perl6
20:38 Aankhen`` avar: Make sure to `svn up`. :-)
20:40 avar Aankhen``: what stuff works now?
20:41 avar say() still fails..
20:41 Aankhen`` I dunno about the emitter, I've been working on the Lisp side.
20:42 Aankhen`` Load up Runtime.lisp, switch to the KP6-CL package, and try this:L
20:42 Aankhen`` s/L$//
20:43 Aankhen`` (kp6-apply-function "Code_substr" (mapcar #'cl->perl (list "abcdef" 2 2)))
20:43 Aankhen`` (kp6-apply-function "Code_say" (mapcar #'cl->perl (list "ab" #\c #\d "ef")))
20:43 avar yeah i see
20:43 avar sweet
20:43 rlb3_work left #perl6
20:44 Aankhen`` ^_^
20:44 avar Are you using slime?
20:44 Aankhen`` But of course.
20:44 avar I've had some issues with loading the lisp files, seems I'm in the wrong dir..
20:44 Aankhen`` Yeah, Runtime.lisp uses absolute paths.
20:45 Aankhen`` You'll probably need to start emacs from the v6-KindaPerl6/ dir, then (load "lib/KindaPerl6/Runtime/Lisp/Runtime.lisp").
20:46 Aankhen`` I should really just create an ASDF system for it.
20:48 avar M-x cd v6-KindaPerl6 M-x slime C-x C-f lib/.../Runtime.lisp C-c C-k worked
20:48 Aankhen`` Ugl-eh.
20:48 Aankhen`` But it's better than what I was using, I suppose.
20:49 avar I like how package.lisp uses the Hash.lisp stuff
20:49 Aankhen`` packages.lisp, you mean
20:49 Aankhen`` ?
20:49 avar yeah
20:49 Aankhen`` I'm all about layers. <G>
20:50 avar what lisp are you using?
20:50 rlb3_work joined #perl6
20:50 Aankhen`` SBCL.
20:50 Aankhen`` BRB.
20:51 avar clisp here, my box crashes on sbcl;/
20:53 Aankhen`` What OS?
20:53 avar ew tabs for indenting
20:53 avar it's running on xen, known bug
20:54 Aankhen`` Oh, sucks.
20:54 Aankhen`` Tabs for indenting?  Yuck!
20:54 Aankhen`` (So says the person who used them exclusively for nine years.)
20:55 avar you put some in global.lisp :)
20:55 Aankhen`` Hmm, how?
20:55 avar (define-kp6-function ("say" :returns 'true) (&rest strs)
20:55 avar (format t "~{~A~}~%" (mapcar #'kp6-value strs)))
20:55 Aankhen`` I only edited it in emacs, which is set to use spaces rather than Tabs.
20:55 Aankhen`` Where's the Tab?
20:55 avar you fuuckeeed uuup?:)
20:56 Aankhen`` I see two spaces before the (format.
20:56 avar :value #'(lambda ,params
20:56 avar ,@body
20:56 avar ,@(when (not (null returns))
20:56 avar there
20:56 avar no "say" is all right
20:56 Aankhen`` Still only seeing spaces.
20:57 Aankhen`` A lot of them, but spaces nonetheless.
20:57 avar try C-x h M-x untabify on the file and diff it
20:58 Aankhen`` I already shut down my VM, unfortunately.
21:01 avar how do you call the "say" function?
21:02 Aankhen`` (kp6-apply-function "Code_say" (mapcar #'cl->list (list "foo" "bar" "baz")))
21:02 Aankhen`` General template, for now, is: (kp6-apply-function "Code_foo" (mapcar #'cl->list (list ...args...)))
21:03 Aankhen`` I'll add a wrapper function tomorrow.
21:05 Aankhen`` I'm a little confused.  Is there an `apply` and a `call` in Perl 6?
21:05 avar that's a kp6 thing
21:06 Aankhen`` Not in Perl 6 proper, then?
21:06 avar KP6-CL> (kp6-apply-function "Code_say" (mapcar #'cl->perl (list "a" "b" "c")))
21:06 avar abc
21:06 avar #<KP6-BIT #x000333E6E488>
21:06 avar It's just something kp6 uses for dispatch I think
21:07 Aankhen`` I wrote the KP6-APPLY-FUNCTION, I was just wondering about whether I was stomping a Perl 6 builtin. :-)
21:07 avar Aankhen``: If you can tell me how that would work where "a" "b" and "c" are (make-instance 'kp6-Str :value "a") I can get it working..
21:07 Aankhen`` Or rather, confusing names with a builtin.
21:07 Aankhen`` The idea is that the MAPCAR runs CL->PERL on each of them, which basically turns them into instances of kp6-Str.
21:07 Aankhen`` So if you already have instances of kp6-Str, just pass a list of them instead.
21:08 avar hrm
21:08 avar yeah, I see
21:08 Aankhen`` KP6-APPLY-FUNCTION takes the function name and a list of arguments, which need to be Perl 6 objects.
21:09 Aankhen`` It's just one step above the bare minimum, which is to look up the function yourself.
21:09 Aankhen`` Oh, it can also take an optional :PACKAGE argument to indicate which package you want to search for the function in.
21:09 avar I'll add a generic function that takes kp6-str and just passes it through
21:09 Aankhen`` Not sure I understand.
21:10 Aankhen`` Oh, a method for CL->PERL specialized on kp6-Str?
21:10 avar KP6-CL> (mapcar #'cl->perl (list "foo" (kp6-value (make-instance 'kp6-Str :value "bar"))))
21:10 avar (#<KP6-STR #x000333E64398> #<KP6-STR #x000333E643B8>)
21:10 avar thi works
21:11 avar KP6-CL> (mapcar #'cl->perl (list "foo" (make-instance 'kp6-Str :value "bar")))
21:11 avar this doesn't
21:11 Aankhen`` You could do that, but you'd be better off doing this: (defmethod cl->perl ((object kp6-Object)) object)
21:11 avar so there needs to be a cl->perl version that passes it through
21:11 avar yeah, that's what I was doing:)
21:11 Aankhen`` Yes, you're right.
21:11 Aankhen`` I'm not sure CL->PERL is the right place to do that though.
21:12 Aankhen`` Oh well, it'll do for now.
21:12 avar your stuff doesn't work with kp6-Object, works with kp6-Str
21:12 avar maybe it needs a type-of call instead?
21:13 Aankhen`` Wha?
21:13 Aankhen`` How so?
21:13 avar (defmethod cl->perl ((object kp6-Str) &key)
21:13 avar object)
21:13 Aankhen`` That works but not kp6-Object?
21:13 avar this works, the same thing with kp6-Object doesn't work
21:14 Aankhen`` Strange.
21:14 amnesiac omg it's avar...
21:14 Aankhen`` I'll look into it tomorrow.
21:14 hoelzro joined #perl6
21:16 Aankhen`` avar: What exactly happens if you substitute kp6-Object for kp6-Str?
21:17 avar no generic function is found
21:17 Aankhen`` I see.
21:17 Aankhen`` Something wrong with the inheritance hierarchy, maybe.
21:17 avar It just needs something equivalent to ->isa("kp6-Object") as opposed to ref $obj eq 'kp6-Str'
21:17 avar yeah
21:17 Aankhen`` Dangit, I hate having to boot up a VM to do any programming.
21:17 Aankhen`` It's just another barrier!
21:20 avar urgh, subtypep is odd
21:21 Tene Aankhen``: why do you need a VM for programming?
21:21 Aankhen`` SUBTYPEP is for checking whether one class is a subclass of another.
21:21 Aankhen`` Tene: My primary OS is Windows.
21:21 Aankhen`` I've done the whole Win32 programming bit for many, many years, and I finally got tired of it.
21:21 Tene Oh.
21:21 hoelzro ugh, Win32 programming...
21:21 Aankhen`` Er, s/Win32 programming/programming on Windows/.
21:21 Aankhen`` I didn't mean I've been writing Windows apps.
21:22 hoelzro be thankful
21:22 * Aankhen`` is.
21:22 Tene So... why are you running on Windows?
21:22 Aankhen`` Because for everything except development, it works. :-)
21:22 Aankhen`` And I'm not particularly fond of Linux.
21:22 avar you know slime supports connecting to a remote image
21:22 Tene Okay.
21:22 Aankhen`` It's better than Windows in many ways, but it's still just the lesser of two evils.
21:22 Aankhen`` And with that I'll shut up on the subject.
21:23 hoelzro to each his own
21:23 Aankhen`` avar: Yessir.
21:24 avar what do you use windows for?
21:24 avar might be less of a pita to run it in vmware:)
21:24 Aankhen`` Everything except development.  Surfing the 'Net, playing games, wasting time on IRC…
21:25 Aankhen`` Oh, and watching lots of animated series & movies.
21:25 Aankhen`` And listening to music.
21:25 Aankhen`` Until foobar2000 is released for Linux, I'm sticking with Windows.
21:25 hoelzro what's foobar2000?
21:25 Aankhen`` Audio player.
21:26 hoelzro never heard of it
21:26 hoelzro interesting
21:26 avar use emacs for playing music, dewd:)
21:26 avar the interface is similar to foobar
21:26 hoelzro emacs?!
21:27 hoelzro blasphemy!
21:27 hoelzro I'm a vimmer myself
21:27 hoelzro =P
21:28 avar Aankhen``: ATTACK
21:28 Aankhen`` EEK
21:29 Tene If you don't use linux, why do you care about developing software for it?
21:30 avar Aankhen``: Why do you hate america?
21:30 Aankhen`` I don't hate America, and I don't hate Linux. :-)
21:30 Tene I didn't say hate, just "don't use"
21:31 Tene I'm really just curious.
21:31 Aankhen`` Tene: As I said, Linux is the lesser of two evils… more importantly, though, I develop software because I love programming.
21:31 Tene Ahh.
21:31 Tene Okay. :)
21:31 avar Aankhen``: Your CL works, but the emitter really likes stuff in a certain order
21:32 Aankhen`` avar: How so?
21:32 avar and I (kp6-apply-function "Code_say" (mapcar #'cl->perl (list (make-instance 'kp6-Str :value "a")
21:32 avar (make-instance 'kp6-Str :value "b")
21:32 avar this works now
21:32 avar but..
21:32 avar (make-instance 'kp6-Str :value "c"))))
21:32 avar http://sial.org/pbot/27644
21:32 lambdabot Title: Paste #27644 from Someone at 208.78.101.240
21:32 avar this is what the emitter spits out
21:32 Aankhen`` You're going to make me start up the VM again, at 3:00 AM, aren't you? :-\
21:33 avar http://sial.org/pbot/27645 # with these patches
21:33 lambdabot Title: Paste #27645 from Someone at 208.78.101.240
21:33 avar nah, this looks more like a task for fglock
21:33 Aankhen`` Why did you remove the &allow-other-keys on CL->PERL?
21:33 * Kattana also cant move to linux until foobar2000 is ported
21:33 Aankhen`` And PERL->CL.
21:34 avar oh that was me screwing around earlier, nevermind that
21:34 Aankhen`` avar: Oh, I didn't realize what you were trying to use SUBTYPEP for.  What you want is TYPEP.  (typep object 'kp6-Object)
21:34 avar the relevant stuff is EmitLisp.pm
21:34 Aankhen`` The invocant EmitLisp.pm spits out is way harsh.
21:35 Aankhen`` I'm not sure what it's doing, exactly.
21:35 hoelzro none of us are.
21:35 hoelzro =P
21:36 * Aankhen`` giggles.
21:36 avar it's spitting out AST nodes
21:36 Aankhen`` avar: Yes, I mean, I'm not sure why it has so many colons, for one.
21:36 avar Either the AST needs to be mutated or the lisp stuff needs to work in that order too
21:36 avar I added that to see what was doing what
21:36 Aankhen`` Hrm, hrm.
21:37 Aankhen`` This stuff is whack, man!
21:37 Aankhen`` </dude>
21:38 Aankhen`` avar: What is the invocant it's printing?  Can it be separated into a package and a name?
21:39 * avar fails to parse that
21:39 Aankhen`` (kp6-apply-function (kp6-lookup obj:'(kp6-lookup obj:'kp6-Hash_KP6 index:(make-instance 'kp6-Str :value "GLOBAL") index:(make-instance 'kp6-Str :value "Code_say") ; AIUI, that huge second argument is the function being called
21:40 Aankhen`` Can it be separated into just the package and the name?
21:40 avar ugh, I don't know
21:41 avar hrm
21:41 Aankhen`` I'm sorry, I'm very ignorant of KP6's architecture; perhaps I should have held back before jumping into the CL side. :-\
21:42 avar nah it's fiine
21:42 * avar tries something
21:44 Aankhen`` Okay.
21:46 Aankhen`` The KP6-OBJECT thing seems okay here.
21:46 Aankhen`` Actually, I think I'd left the &key off the one I wrote in-channel.
21:47 Aankhen`` I've checked in the working version.
21:48 avar yes, you did:)
21:48 Aankhen`` Hold on, hold on.
21:48 Aankhen`` How on Earth did KP6-VALUE stop inheriting from KP6-OBJECT? O_O
21:49 Aankhen`` Fixed.
21:49 avar ah, was that the problem?
21:49 Aankhen`` As I said, KP6-VALUE wasn't inheriting from KP6-OBJECT for some reason.
21:51 avar sh-3.1$ echo 'say "a", "b", "c"' | perl kp6-mp6-perl5.pl --lisp | clisp | grep ^Break.*abc
21:51 avar Break 1 [2]> abc
21:51 avar awesomeness
21:51 Aankhen`` ?
21:51 Aankhen`` What does the Break indicate?
21:52 Aankhen`` I only used CLISP very briefly, when I was very new to CL.
21:52 avar nothing, it's just spewing out tons of shit
21:52 Aankhen`` Ah.
21:53 avar #my $op := $.code.emit_lisp;
21:53 avar my $op := '"Code_say"';
21:53 avar naughty
21:54 hoelzro which spec contains the information on the relationships between frontend,backend,emitter/etc?
21:54 avar of kp6?
21:54 hoelzro of Perl 6 in general
21:54 hoelzro I want to understand the roles that Pugs, kp6, etc all play
21:55 avar they each work in their own way
21:55 avar and it's all light on docs;/
21:55 hoelzro kp6 is just a Perl 6 -> CL translator?
21:56 Aankhen`` avar: LMAO.
21:56 hoelzro I think I'm confused
21:56 avar kp6 emits perl5, html, cl and probably something else
21:56 Aankhen`` hoelzro: No, Common Lisp is just one language for which KP6 has an emitter.
21:56 hoelzro I see
21:57 Aankhen`` hoelzro: There's a part of KP6 that parses Perl 6, and a separate part that emits that code in a different language.
21:57 hoelzro ah
21:57 hoelzro understood
21:57 avar ls lib/KindaPerl6/Visitor/Emit*
21:58 hoelzro are there plans for Perl 6 -> PIL with kp6?
21:58 avar $ echo 'say "a", "b", "c"' | perl kp6-mp6-perl5.pl --lisp | head -n3
21:58 avar global say
21:58 avar (in-package #:cl-user)
21:58 avar (load (compile-file "lib/KindaPerl6/Runtime/Lisp/Runtime.lisp"))
21:58 avar urgh, where the hell does that "global say" come from
21:58 Aankhen`` What's with the "global say" at the beginning?
21:58 Aankhen`` Heh.
21:59 avar hoelzro: If someone does it, suure:)
21:59 hoelzro is PIL a big deal, or some idea we're just throwing around?
21:59 avar The current plan for it is to bootstrap it with the STD parser, me and Aankhen`` are just goofing around
21:59 avar what's PIL?:)
22:00 Aankhen`` Pugs Intermediate Language?
22:00 hoelzro I think it's Parrot Intermediate Language, or more likely Perl Intermediate Language
22:00 Aankhen`` Parrot Intermediate Language?
22:00 Aankhen`` One of the two, I think.
22:00 avar Oh that thing
22:00 avar pugs or parrot?
22:00 hoelzro I think it's parrot
22:01 TimToady that's PIR, not PIL
22:01 * Aankhen`` checks the IRC logs to see if they're smarter.
22:01 Aankhen`` Pugs, yay.
22:01 Aankhen`` The logs were smarter than me after all. :-D
22:01 hoelzro ah
22:01 avar http://search.cpan.org/~particle/p​arrot-0.4.16/docs/glossary.pod#PIL
22:01 lambdabot Title: docs/glossary.pod - Parrot Glossary - search.cpan.org, http://tinyurl.com/yw6bue
22:02 avar If someone steps up sure, but pugs doesn't seem like an especially good /runtime/ at the moment
22:02 hoelzro I'm trying to help with Pugs, but I need guidance
22:03 hoelzro and the trac at dev.pugscode isn't very helpful =/
22:03 avar well "Pugs" is a lot of things, are you mainly interested in the pugs haskell compiler ?
22:03 avar i.e. the pugs(1) binary
22:03 TimToady I'm using pugs to breadboard STD currently, but that's because pugs's set of things it does right is closer to what STD needs than the set of things that everyone else's platform does right.
22:04 hoelzro the haskell compiler
22:04 avar TimToady: Are you runnning the desugared version?
22:05 TimToady I've run the metholated version of STD as far as parsing 42+1 via the operator precedence parser. mostly I'm fiddling with small examples to figure out how to optimize DFAable longest tokens out to the current language
22:06 TimToady so when you call "expect_term" from a particular language, it knows right there what all the possible tokens are
22:06 TimToady basically, I'm working on self-lexing grammars
22:06 TimToady which interacts oddly with extensible grammars
22:07 TimToady because when two different languages call expect_term, they have two different sets of longest tokens
22:07 TimToady so that info has to be stored in the right place
22:07 avar ah, so implementations aren't holding you back at the moment
22:08 TimToady well, I've spent a fair share of my time working around pugs bugs...
22:09 TimToady for instance, I can't really create a Match object from Perl 6
22:10 TimToady a lot of the bugs come back to the incomplete object system
22:10 TimToady but some of them are just Parsec being wonky
22:13 TimToady a lot of what I want to do is things like mixins of roles into method objects, but that doesn't work either.
22:13 avar ;/
22:13 avar everything sucks:)
22:14 avar apperently clisp's idea of --quiet or --silent is "please spew out the same irrelivant 50 lines of debug info"
22:14 Aankhen`` LMAO.
22:15 wolverian joined #perl6
22:16 TimToady I think a good error message would be "foobar at baz line 42; see http://[127.0.0.1]/aos24jakj9 for more"
22:16 TimToady where that contains the traceback and long explanations
22:16 TimToady and maybe the cockpit recorder
22:17 szbalint where is the terminal colour code for orange when you need it
22:17 szbalint :)
22:17 TimToady you do that with the TURN_TERMINAL_ORANGE_ON_ERROR env var
22:18 avar <Starship Troopers Announcer>Error on line 42, would you like to know more?
22:19 TimToady and the http page could have a link that says "Send me in as a bug report"
22:28 avar kp6: class Fleh {} Fleh.new
22:28 exp_evalbot r18087: OUTPUT[syntax error at position 0, line 1 column 0:␤( ( do { class Fleh {} Fleh.new } ).perl ).sa␤^ HERE␤]
22:30 TimToady that should be a syntax error according to spec
22:30 TimToady requires ; after the {}
22:30 TimToady but that's probably not what it's complaining about there...
22:31 TimToady biab &
22:41 avar avar@oe:/tmp/v6/v6-KindaPerl6$ time ./kp6-hellow |grep abc
22:41 avar abc
22:41 avar real    0m0.086s
22:41 avar the first Perl 6 to machine code compiler via CL and sbcl :)
22:56 TimToady what is CL's attitude toward's Unicode?
22:57 avar It's implementation defined, most do native unicode but if you want it portable across implementations you'll need your own (or some pre-written) library
22:57 avar (length "ævar")
22:57 avar 4
22:58 avar that's clisp, same with sbcl, might not be the case with some others..
23:00 weinig joined #perl6
23:03 avar TimToady: So about as well as C:)
23:03 avar w 5
23:05 polettix joined #perl6
23:06 avar How do I see (in kp6) what class something belongs to?
23:06 avar my $op := $.code.emit_lisp;
23:06 avar I want to find out what class that's being dispatched to
23:06 avar not $.code.WHAT
23:06 avar (ref $foo).to(Perl6) :)
23:10 TimToady .WHAT gives you the protoobject, while .HOW gives you the metaclass object
23:10 TimToady (in theory)
23:11 TimToady there is no such thing as a class object in Perl 6 :)
23:13 dduncan joined #perl6
23:14 avar I just want to see what class it belongs to
23:16 TimToady define "belongs"
23:16 BinGOs_ joined #perl6
23:17 TimToady there is no object that "owns" your object...
23:18 amnesiac avar, introspection?
23:21 TimToady introspection would be in the bailywick of the .HOW metaclass object
23:22 amnesiac yes, about the behaviors of the object
23:22 amnesiac TimToady, but what abou attributes?
23:22 amnesiac TimToady, I guess avar wants to know the class it belongs to, not as a relationship, just about with which class was it created
23:24 avar yeah, what C<class {}> the method will be dispatched to
23:25 TimToady that would be controlled by the metaobject instance, which could support either class-based or prototype-based method dispatch
23:26 TimToady but each class {} would typically get its own instance
23:26 * amnesiac ponders
23:30 TimToady the standard MO is class based, but there's no guarantee.  a MOP could easily set up a responder interface without classes at all, as in prototype based languages.  if there's a concept of class, your .HOW should be able to tell you
23:30 TimToady but for the standard MO, it'll probably just tell you the protoobject, whose main function is to carry the .HOW without being a real object.
23:31 TimToady so when you say "class Foo {}" you're setting up a Foo protoobject that isn't a real object
23:31 TimToady nonetheless Foo.HOW returns the metaclass instance for it.
23:32 TimToady otherwise you'd have to say Foo.new.HOW to get at it.
23:33 TimToady Foo is just a typed undef in Perl 6
23:33 TimToady Foo.HOW is the closest thing to a "class"
23:34 TimToady the "class" keyword is a good indication that you're using the standard MO though
23:34 TimToady a different MOP would likely use a different keyword
23:34 TimToady much like roles and subsets are different ways to view types
23:35 TimToady (though roles are a part of the standard MOP)
23:37 wolverian random_mop_class Foo { } # macro
23:39 TimToady currently package_declarator:<random_mop_class> in STD
23:41 wolverian ach, my ideas are pre-empted
23:42 TimToady we're serious about extensibility in Perl 6, and have been so for 'bout seven years now...
23:42 TimToady so there's been a lot of pre in which to empt...
23:46 BinGOs joined #perl6
23:54 Eidolos possible microgrant idea: implement a few popular modules in idiomatic p6

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

Perl 6 | Reference Documentation | Rakudo