Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2008-02-06

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:01 cj :)
00:02 pbuetow (:
00:04 wolverian holy crap, is v6/v6-KindaPerl/compiled supposed to be in the repo and take 36MB?
00:05 ruoso wolverian, considering it takes more time to compile than to download...
00:06 wolverian oh, sure. :)
00:06 wolverian just my hd that's full.. I know, my new laptop is coming in next week. phew.
00:07 meppl good night
00:13 cj anybody want to fund me to work on p6 on the CLI VM for a year?
00:13 cj sounds like more fun than generating invalid web pages...
00:14 ruoso pugs: class Foo { method a { return "a" } }; role Bar { method a { return "b" } }; my $a = Foo.new(); $a does Bar; role Bar is also { method a { return "c" } }; say $a.a;
00:14 exp_evalbot OUTPUT[*** Unknown syntactic construct: Syn "does" [Var "$a",Val (VType (mkType "Bar"))]␤    at /tmp/Gyy5lKVIoe line 1, column 97-108␤]
00:15 ruoso pugs: class Foo { method a { return "a" } }; role Bar { method a { return "b" } }; my $a = Foo.new(); class Foo is also { does Bar }; say $a.a;
00:15 exp_evalbot OUTPUT[a␤]
00:16 ruoso pugs: class Foo { method a { return "a" } }; role Bar { method a { return "b" } }; my $a = Foo.new(); class Foo is also { does Bar }; my $b = Foo.new(); say join ',', $a.a, $b.a
00:16 exp_evalbot OUTPUT[a,a␤]
00:17 cj what is "P�"?
00:17 cj \n ?
00:17 ruoso cj, you mean, in the OUTPUT line?
00:18 ruoso you probably want to configure your irc client to use utf8 (if that's the case)
00:18 cj yeah, that
00:18 ruoso pugs: class Foo { method a { return "a" } }; role Bar { method b { return "b" } }; my $a = Foo.new(); class Foo is also { does Bar }; my $b = Foo.new(); say join ',', $a.a, $b.a, $a.b, $b.b
00:18 exp_evalbot OUTPUT[a,a,b,b␤]
00:18 cj I just configured it to use utf-8
00:19 * cj sets his LOCALE
00:19 ruoso it seems that pugs object composition is not quite right...
00:19 TimToady yes, MO was going to be the next big release (6.28) but it never made it
00:20 wolverian cj, ␤ U+2424 "SYMBOL FOR NEWLINE", category: "Control Pictures", utf8 bytes: E2 90 A4
00:20 ruoso but what was it supposed to do in the following code?
00:20 ruoso pugs: class Foo { method a { return "a" } }; role Bar { method b { return "b" } method a { return "c" }; my $a = Foo.new(); class Foo is also { does Bar }; my $b = Foo.new(); say join ',', $a.a, $b.a, $a.b, $b.b
00:20 exp_evalbot OUTPUT[*** ␤    Unexpected end of input␤    expecting operator, ":", "," or "}"␤    at /tmp/3LYfJo4XG3 line 2, column 1␤]
00:21 ruoso pugs: class Foo { method a { return "a" } }; role Bar { method b { return "b" } method a { return "c" }}; my $a = Foo.new(); class Foo is also { does Bar }; my $b = Foo.new(); say join ',', $a.a, $b.a, $a.b, $b.b
00:21 exp_evalbot OUTPUT[a,a,b,b␤]
00:21 ruoso should any of these be "c"?
00:21 ruoso night &
00:22 cj wolverian: strange... what all do I need to set to get irssi to play nice with utf-8 chars?  do you know?  Setting gnome-terminal's encoding to utf-8 is obviously not enough...
00:23 wolverian cj, it's easiest if your locale is utf8, then everything (including screen and irssi) work automatically. the manual method is doing /set term_charset utf8 and setting screen to utf8 with ^A :utf8
00:23 wolverian cj, (but setting the locale requires restarting irssi and screen)
00:27 Juerd I've found it necessary to set the locale and restart screen.
00:27 Juerd ^A:utf8, -U, etc, never worked well for me.
00:27 wolverian I haven't actually tried the manual method for some years, since I haven't needed to.
00:28 wolverian so you might be right in that it doesn't even work.
00:28 wolverian of course since you can set the locale per-program, it doesn't make much sense even to do that :)
00:28 wolverian $ LANG=foo.utf8 screen irssi
00:29 wolverian but really.. are there any modern distros that ship with non-utf8 locales by default?
00:29 wolverian s/ship with/use/
00:29 wolverian food ->
00:31 wolverian s/->/→ /
00:32 __Ace__ joined #perl6
00:32 pbuetow joined #perl6
00:32 pcbuetow joined #perl6
00:32 cj wolverian: do you know an easy way to change a debian testing box to a utf8 locale?
00:34 pugs_svn r19823 | lwall++ | Remove some fossil () noticed by putter++
00:34 pugs_svn r19823 | lwall++ | Fix whitespace assertion on statement_control not to accidentally allow fatarrow after newline
00:38 TimToady ruoso: no, a class's methods always override a role's methods
00:40 rakudo_svn r25543 | jonathan++ | [rakudo] Generalize 'is' on classes to call trait_auxiliary:is, and implement a multi-sub that accepts a trait that is a class, which applies the inheritance. Now it's all MMD, as it should be.
00:42 wolverian cj, to make your user use utf8 by default, put export LANG=en_US.utf8 (or so) in your appropriate startup file, e.g. .bashrc
00:42 wolverian (I'm not sure which one you're supposed to really use. .bash_profile, perhaps?)
00:42 pugs_svn r19824 | lwall++ | [STD_extract] answer some of the questions
00:43 wolverian cj, to change the system default locale, see dpkg-configure locales, or so. Juerd can help here, probably.
00:43 wolverian (I'm used to ubuntu and its gui tools :)
00:44 [particle] well, looks like rakudo got the new object model before pugs. who would have predicted that a year ago?
00:45 jnthn [particle]: Erm, the object model in Rakudo has a LONG way to go.
01:00 kst joined #perl6
01:05 cj wolverian, Juerd: this look right? [*] en_US.UTF-8 UTF-8
01:06 cj wolverian: yeah, there are comments at the top of ~/.bash_foo that say which one is the 'login' version and which gets eval'd whenever a new bash starts
01:06 cj wolverian: I can never remember which is which
01:07 cj okay.  going home now.
01:10 cj wolverian: are you working on any of the various p6 grammars/parsers/emitters?
01:11 cj Juerd: you, too... I'm building a .deb of the library bits required to do late binding, etc with the CLI
01:11 cj s/required/used/
01:12 TimToady I think .bash_profile is for login, and .bashrc is every time
01:12 Juerd cj: I'm not working on anything perl6 related :(
01:15 namelessuser left #perl6
01:18 wolverian cj, no, I'm not. I only have time to procrastinate..
01:34 monomorph joined #perl6
01:47 Juerd CHS addressing is HARD.
01:51 TimToady it's easier than YMD  :)
01:52 TimToady I shudder thinking about leapbytes...
02:02 kst joined #perl6
02:16 allbery_b modern drives have worse than that, actually.  but the on-drive controller hides the ugliness.  (all modern CHS is fake)
02:16 nothingmuch why are modern drives worse?
02:17 allbery_b spare sectors, potentially different sectors/track in different zones
02:18 nothingmuch oh right, i knew about spare sectors at the very least... what's this about zones though? never heard of that
02:19 allbery_b how close the head is to the outer edge vs. center of the platter
02:19 nothingmuch real CHS required that sectors are really radiant?
02:20 allbery_b real CHS was based on the assumption that S was constant for all C, yes
02:20 nothingmuch ouch
02:20 allbery_b it may still be, or might not
02:21 allbery_b the drive manufacturers hide the real geometry so they can play whatever games they want to maximize storage (or speed, etc.)
02:21 nothingmuch *nod*
02:24 allbery_b (plug:  you should be able to find more details than you ever want to know at http://www.dssc.ece.cmu.edu/ )
02:26 Auzon joined #perl6
02:46 alester joined #perl6
03:01 mncharity joined #perl6
03:01 mncharity re r19822 commit message, lol :)
03:01 lambdabot mncharity: You have 1 new message. '/msg lambdabot @messages' to read it.
03:03 mncharity spinclad: re debian ruby 1.9, my fuzzy recollection is debian is serving a pre-release copy of 1.9.
03:05 mncharity so might check that version >= 1.9.0 2007-12-25
03:10 dmq joined #perl6
03:16 justatheory joined #perl6
03:16 kst joined #perl6
03:18 mncharity TimToady: re Why am I reinventing cheat/metholate?, hmm... my impression is c/m basically transforms STD into code which can be run on a pugs-level p6 implementation.
03:20 mncharity STD_extract has the (still work in progress) objective of providing enough information to "hand" implement a p6 parser when you don't have p6, and only a partial regex engine.  So it tries / will try to answer questions like
03:22 mncharity what do all the code blocks (which need to be translated to p5/ruby/whatever) look like.  what are all the rx pattern features used.  what are all the subrules used.
03:23 mncharity what are all the instances of mumble, so I can metaprogram draft code (ast nodes, or whatever) for them.
03:23 cathya joined #perl6
03:23 cathya left #perl6
03:24 mncharity That all seemed sufficiently far from the intent of c/m, that it seemed worth spending a day rolling something new, optimized for that exercise specifically.
03:26 mncharity what do you think?
03:26 cj mncharity is chatty tonight!
03:27 mncharity responding to notes in svn. :)
03:27 mncharity hmm, and sounds like I should backlog...
03:27 cj TimToady: that sounds right to me, but I won't COMMIT until I read the docs :)
03:28 Winkie joined #perl6
03:52 mncharity spinclad: oh, nm.  found the discussion in the backlog.
03:57 mncharity ruoso: congrats on progress.  bummer about grant.  need to print out the grant rejections for later posting on wall to mock the reviews - it's traditional.
04:14 mncharity TimToady: I'd like to see some version of STD running against pugs t/.  any feel for how near at hand that is?  thanks.
04:14 mncharity good night all. &
04:24 cathya joined #perl6
04:39 cathya anyone worked on security
04:39 cathya and integration
04:39 cathya ok nevermind offtopic
04:39 kst joined #perl6
05:54 justatheory joined #perl6
06:04 charsbar joined #perl6
06:23 kst joined #perl6
07:07 Daveman joined #perl6
07:37 moritz_ moin
07:38 jisom joined #perl6
07:39 barney joined #perl6
07:54 Aankhen`` joined #perl6
08:28 cosimo joined #perl6
08:32 meppl joined #perl6
08:41 pyfrag joined #perl6
08:43 pyfrag left #perl6
08:50 iblechbot joined #perl6
08:52 pugs_svn r19825 | bloonix++ | kicked use lib /home/... from Linux.pm :-)
08:52 bloonix joined #perl6
09:19 spinclad joined #perl6
09:20 kst joined #perl6
09:21 [particle] joined #perl6
09:26 zamolxes joined #perl6
09:29 devogon joined #perl6
09:58 mattz_ joined #perl6
10:01 Psyche^ joined #perl6
10:02 IllvilJa joined #perl6
10:06 Aankh|Clone joined #perl6
10:10 viklund joined #perl6
10:21 Aankhen`` joined #perl6
10:43 blindfish joined #perl6
10:55 Aankh|Clone joined #perl6
11:05 chris2_ joined #perl6
11:24 ebassi joined #perl6
11:39 djm left #perl6
11:40 wknight-away joined #perl6
11:58 ruoso TimToady, re: class method overrides role method. Isn't by the role overriding the class' method that "0 but true" works?
11:58 cognominal_ joined #perl6
12:01 ruoso TimToady, nevermind... I've found out... it's my example that is a bad one...
12:02 ruoso pugs: class Foo { method a { "a" } }; class Bar is Foo; role Baz { method a {"b" } }; my $a = Bar.new(); class Bar is also { does Baz }; my $b = Bar.new(); say join ',', $a.a, $b.a
12:02 exp_evalbot OUTPUT[b,b␤]
12:09 ruoso pugs: class Foo { method a { "a" } }; class Bar is Foo; role Baz { method a {"b" } }; class Buz { method a {"c"} }; my $a = Bar.new(); class Bar is also { does Baz };class Bar is also { is Buz }; my $b = Bar.new(); say join ',', $a.a, $b.a
12:09 exp_evalbot OUTPUT[b,b␤]
12:10 moritz_ uhm, pugs' OO model isn't all that good ;)
12:12 ruoso moritz_, but I think the last one is right
12:12 ruoso becaus I'm adding MI to Bar
12:12 ruoso and the MRO would call Foo's a
12:13 ruoso http://www.perlfoundation.org/perl6/index.cgi?smop_p6opaque_implementation
12:13 lambdabot Title: SMOP p6opaque Implementation / Perl 6, http://tinyurl.com/2z5spt
12:13 ruoso moritz_, take a look at the "composition" section in this sketch I'm writing
12:15 moritz_ ruoso: I don't know enough about Perl 6 OO (and MRO in general) to really assess that
12:22 mattz joined #perl6
12:34 franck____ joined #perl6
12:47 pbuetow joined #perl6
13:00 renormalist joined #perl6
13:07 blindfish joined #perl6
13:45 integral joined #perl6
13:50 kanru joined #perl6
13:57 monomorph joined #perl6
14:02 alester joined #perl6
14:04 Lorn joined #perl6
14:04 cognominal_ joined #perl6
14:06 integral joined #perl6
14:28 icwiener joined #perl6
14:34 acmoore joined #perl6
14:45 riffraff joined #perl6
14:51 Aankhen`` joined #perl6
15:14 TJCRI joined #perl6
15:23 peeps[work] joined #perl6
15:41 jferrero joined #perl6
15:44 paq joined #perl6
15:46 otronick joined #perl6
15:49 manfred joined #perl6
15:50 kyrbe joined #perl6
15:51 kyrbe left #perl6
15:54 Aankh|Clone joined #perl6
16:11 Alias__ joined #perl6
16:11 kst joined #perl6
16:24 Psyche^ joined #perl6
16:35 njbartlett joined #perl6
16:58 TimToady ruoso: those look correct to me.  the does inserts the 'b' method into Bar directly, and all the other .a's are in base classes, so the role wins
16:58 TimToady that's because "is also" is modifying the class in place, not doing a mixin
16:58 pugs_svn r19826 | moritz++ | [irclog] fixed stupid bug in the logging bot that caused
16:58 pugs_svn r19826 | moritz++ | weird and wrong nick names
16:59 TimToady if it were a mixin (as in "0 but true") then it would be different
16:59 TimToady with a mixin you get a new anonymous class, and then a does in the original class is considered part of the original class, not part of the new anonymous class
17:18 justatheory joined #perl6
17:19 kst joined #perl6
17:28 rindolf joined #perl6
17:51 meppl joined #perl6
17:54 paq joined #perl6
17:59 cognominal_ joined #perl6
18:01 pugs_svn r19827 | moritz++ | [irclog] added a few #parrot specific bots
18:04 * moritz_ wonders why the dynamic nick coloring doesn't work for the #parrot logs
18:12 pugs_svn r19828 | moritz++ | [irclog] determine nick colors local to current channel ;)
18:14 moritz_ the nick coloring predates the "multiple channels" feature ;)
18:23 tobeya joined #perl6
18:26 Lorn moritz_: hey, can you help me? i forget my password for the commit bit :P
18:26 moritz_ Lorn: just a sec
18:27 moritz_ damned, commitbit not respondiing
18:29 moritz_ Lorn: sent. If it doesn't work, or doesn't arrive, ping me again
18:35 paq joined #perl6
18:38 njbartlett_ joined #perl6
18:42 TechJournalist joined #perl6
18:44 cj morning homies
18:44 Tene morning
18:55 ruoso_ joined #perl6
18:58 rindolf joined #perl6
19:18 jisom joined #perl6
19:19 manfred joined #perl6
19:30 paq joined #perl6
19:32 pugs_svn r19829 | lorn++ | commit bit test
19:32 Lorn moritz_: work :) thanks
19:33 moritz_ Lorn: you're welcome
19:35 moritz_ ruoso: Vienna.pm is giving out grants in the "winter of code" project
19:35 moritz_ ruoso: maybe you could apply for a grant there
19:37 Aankhen`` joined #perl6
19:37 paq joined #perl6
19:45 psapo joined #perl6
19:51 __Ace__ joined #perl6
19:59 smtms joined #perl6
20:03 moritz_ @tell unobe I didn't apply your last irclog patch because the problem was cause by caching, not by wrong DB entries. Mail to you hangs at some smtp server :/
20:03 lambdabot Consider it noted.
20:13 buchet joined #perl6
20:15 buchet left #perl6
20:23 ebassi joined #perl6
20:23 peepsalot joined #perl6
21:00 DaGo joined #perl6
21:53 alester left #perl6
21:54 smtms joined #perl6
22:14 Auzon joined #perl6
22:15 smtms joined #perl6
22:28 alester joined #perl6
22:47 DarkWolf84 joined #perl6
22:56 ludan joined #perl6
23:08 wknight-away joined #perl6
23:40 alester joined #perl6
23:54 cathyal joined #perl6
23:54 cathyal hi
23:54 cathyal just wondering
23:54 Tene Yes?
23:54 cathyal anyone worked on ircdaemons
23:55 wknight-away left #perl6
23:55 cathyal ..i know ot
23:55 spinclad we have an irclogger
23:56 spinclad http://svn.pugscode.org/pugs/misc/irclog
23:56 lambdabot Title: Revision 19829: /misc/irclog

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

Perl 6 | Reference Documentation | Rakudo