Time  Nick          Message
23:59 masak         bug brought to you courtesy of a November being brought back from the Mostly Dead.
23:59 colomon       masak++
23:59 * masak       submits rakudobug
23:59 p6eval        rakudo : OUTPUT«Method 'WHAT' not found for invocant of class 'Hash'␤  in 'foo' at line 22:/tmp/qK5WsYzzEI␤  in main program body at line 22:/tmp/qK5WsYzzEI␤»
23:59 masak         rakudo: sub foo($a, %h?) { say %h.WHAT }; foo(42)
23:58 p6eval        rakudo : OUTPUT«Method 'HOW' not found for invocant of class 'Hash'␤  in 'bar' at line 1␤  in 'bar' at line 22:/tmp/dkqeNBk8Hm␤  in 'foo' at line 22:/tmp/dkqeNBk8Hm␤  in main program body at line 22:/tmp/dkqeNBk8Hm␤»
23:58 masak         rakudo: sub foo($a, %h?) { bar($a, %h) }; sub bar($a, %h?) { say "alive!" }; foo(42)
23:34 dalek         niecza: review: https://github.com/sorear/niecza/commit/f958e9c529cf49c4f310fe8020d33fc1cdef6506
23:34 dalek         niecza: Implement :i "str" / <sym>
23:34 dalek         niecza: f958e9c | sorear++ | / (3 files):
23:34 dalek         dalek-poller: review: https://github.com/sorear/dalek-poller/commit/395a68fd905a191a57628144c40f1a85eb2db205
23:34 dalek         dalek-poller: Add ad-hoc github link shortening
23:34 dalek         dalek-poller: 395a68f | sorear++ | modules/local/karmalog.pm:
23:27 lue           I'll look at it later, maybe it'll make sense then. afk
23:16 lue           in the gist I put up earlier, I think method addline($stuffz) is to blame.
23:14 masak         just kidding. that kind of trick is so 2003.
23:13 lue           noooo!
23:12 masak         in other news, if you're using Linux you have to pay me license money.
23:11 * masak       would prefer not to talk about that
23:11 Tene          sorear: It's been around for forever; it just used to go by its acronym, SCO.  It was involved with a lawsuit recently, which prompted the name adjustment.
23:09 masak         I imported ~200 posts from use.perl when starting it.
23:09 masak         it isn't.
23:09 sorear        jul 09?  didn't know sc.org was that old
23:06 masak         sorear: this might clarify some aspects of it: http://strangelyconsistent.org/blog/str-and-buf-i-think-i-get-it-now
23:05 masak         sorear: yes, I did GSoC work on the latter type, Buf.
23:05 masak         where 'char' has some more formal definition, maybe 'grapheme'.
23:05 sorear        I understand you've done some work on this
23:04 masak         sorear: by keeping a clear separation between sequences-of-chars and their encodings as sequences-of-ints.
23:04 sorear        masak: I'd be interested to hear /how/
22:57 masak         sorear: I hear they're thinking of doing it in some other way in Perl 6. :)
22:57 masak         oops. maybe let's not go there, on second thought. again.
22:57 masak         sorear: I hear they're thinking of fixing that in the subsequent major version.
22:47 sorear        perl5-- # for making $bytestring eq $charstring silently DTWT
22:47 sorear        so karmalog is comparing your name as a byte string (in CREDITS) to your name as a character string (in the commit packet)
22:46 sorear        the CREDITS parser doesnt'
22:46 sorear        JSON::XS automatically decodes UTF-8 input
22:46 sorear        tadzik: I think I see the problem
22:24 sorear        type constraints and @ don't mix yet
22:24 sorear        first off, has Pea @.foo is BAD
22:23 lue           can anyone check it out for me? From what I did in p6eval above, it seems to be a problem with the object rather than the loop.
22:21 dalek         dalek-poller: review: https://github.com/sorear/dalek-poller/commit/847cbe70a00e3744ecbb11aa2f369a580e1eb4f3
22:21 dalek         dalek-poller: Save copies of GitHub announce packets
22:21 dalek         dalek-poller: 847cbe7 | sorear++ | push.psgi:
22:20 lue           https://gist.github.com/710830 this fits in the module I described yesterday. The problem is that the @.parsetree array ends up containing an object for each line in the file test.pod, instead of two objects for each block.
22:14 dalek         niecza: review: https://github.com/sorear/niecza/commit/05d77c1974487f1a32230493ce9cdb0558c15d0e
22:14 dalek         niecza: [v6] fix labels, Z+, X+
22:14 dalek         niecza: 05d77c1 | sorear++ | v6/tryfile:
22:14 dalek         niecza: review: https://github.com/sorear/niecza/commit/36db696213656bd34e74a5418c7cbf32dca19358
22:14 dalek         niecza: [v6] fix canonicalize_name for GLOBAL::<$foo>
22:14 dalek         niecza: 36db696 | sorear++ | v6/tryfile:
22:10 p6eval        rakudo : OUTPUT«(timeout)»
22:09 lue           rakudo: my class C { has $.type; has $.stuff; method HAI($a) { $.stuff ~= $a };}; my @a = <. 1 2 . 3 4>; my C @b; for @a { if $_ == "." { @b.push(C.new(:type('.'))); } elsif $_ ~~ /\d/ { @b[*-1].HAI($_); };}; say @b.perl;
22:09 lue           my class C { has $.type; has $.stuff; method HAI($a) { $.stuff ~= $a };}; my @a = <. 1 2 . 3 4>; my C @b; for @a { if $_ == "." { @b.push(C.new(:type('.'))); } elsif $_ ~~ /\d/ { @b[*-1].HAI($_); };}; say @b.perl;
22:06 p6eval        rakudo : OUTPUT«Cannot modify readonly value␤  in '&infix:<=>' at line 1␤  in <anon> at line 22:/tmp/e9ai7QPP9B␤  in main program body at line 1␤»
22:06 diakopter     rakudo: my @a = <. 1 2 . 3 4>; my @b; for @a { if $_ ~~ /\d/ { @b[*-1] ~= $_ }}; say @b.perl
22:04 sorear        TimToady: What are the other elements of $*FILE?
22:04 p6eval        rakudo : OUTPUT«["12", "34"]␤»
22:04 lue           rakudo: my @a = <. 1 2 . 3 4>; my @b; for @a { if $_ == "." { @b.push(""); } elsif $_ ~~ /\d/ { @b[*-1] ~= $_; };}; say @b.perl;
22:01 diakopter     "Cellar Door"
22:00 lue           Tim und Struppi! [as they're called in the German version of the comics]
22:00 lue           ***we now return to our regularly scheduled 0x7F and lower codepoints. Hopefully.***
21:59 masak         Tintin: \δ/
21:56 diakopter     \Ö/
21:55 Tene          is an option too
21:55 phenny        U+00D6 LATIN CAPITAL LETTER O WITH DIAERESIS (Ö)
21:55 Tene          .u Ö
21:55 diakopter     \oo/
21:55 diakopter     oo
21:55 Tene          is how I type it, at least
21:55 phenny        U+0308 COMBINING DIAERESIS (◌̈)
21:55 phenny        U+004F LATIN CAPITAL LETTER O (O)
21:55 Tene          .u Ö
21:55 Tene          diakopter: Ö
21:54 diakopter     I mean the capital O
21:54 TimToady      diakopter: *you* last set the topic; you tell us
21:54 lue           a butterfly :)
21:54 jnthn         Head inflamed due to mosquito bites
21:54 diakopter     what's the O with the two dots over it
21:53 diakopter     \O/
21:53 masak         now that's adaptability.
21:53 lue           \./ # Fine. Have it your way, jnthn :)
21:52 jnthn         No need to be big-headed...
21:52 lue           \0/
21:52 jnthn         :)
21:52 jnthn         /o\
21:51 masak         \o/
21:30 dalek         6model: review: https://github.com/jnthn/6model/commit/f531f67cf33322c2035033200c3bed236dfcc735
21:30 dalek         6model: [dotnet] Until we're bootstrapped and can introspect, need to keep list of things declared in the setting so they're known as lexical; add NQPCapture to this list.
21:30 dalek         6model: f531f67 | jonathan++ | dotnet/compiler/Actions.pm:
21:29 colomon       \o
21:28 p6eval        std a194beb: OUTPUT«Potential difficulties:␤  $x is declared but not used at /tmp/_MpplE2J2a line 1:␤------> { my $x⏏ = 1 }␤ok 00:01 121m␤»
21:28 sorear        std: { my $x = 1 }
21:27 lue           ohayo o/
21:21 jnthn         diakopter: oh, fails for me locally too.
21:21 jnthn         diakopter: Does that work for you locally?
21:21 jnthn         hmm
21:20 p6eval        nqpnet: OUTPUT«␤Unhandled Exception: System.InvalidOperationException: No such method new␤  at Rakudo.Metamodel.KnowHOW.KnowHOWBootstrapper.<Bootstrap>m__7 (Rakudo.Runtime.ThreadContext TC, Rakudo.Metamodel.RakudoObject Ignored, Rakudo.Metamodel.RakudoObject Cap) [0x00000] in <filename unknown>:0…
21:20 jnthn         nqpnet: my $x := NQPCapture.new; $x[0] := 42; say($x[0]);
21:15 TimToady      diakopter: eh? you mean the new self, not the invocant of the new method
21:04 jnthn         moritz_: or BUILDALL...
21:02 moritz_       .bless NYI?
21:00 diakopter     ran into this other places too
21:00 jnthn         Yes
21:00 diakopter     new() *must* return self
21:00 diakopter     oh I see the problem
20:59 jnthn         Which should probably explode.
20:59 diakopter     oh, I see; build needs called
20:59 jnthn         So this is setting $!cap in the type object.
20:59 jnthn         The issue is that in new you don't have an object yet
20:58 jnthn         oh...sorry, I mis-read
20:58 diakopter     I want an NQPCapture
20:57 diakopter     but I don't want that
20:57 jnthn         s/NQPCapture/self/
20:57 jnthn         $!cap := nqp::instance_of(NQPCapture);
20:57 jnthn         diakopter:
20:55 dalek         6model: review: https://github.com/jnthn/6model/commit/3400d13eecb80f362b371fc6949f37711cc81175
20:55 dalek         6model: [dotnet] Fix vivification/default values on named parameters.
20:55 dalek         6model: 3400d13 | jonathan++ | dotnet/runtime/Runtime/Signatures/SignatureBinder.cs:
20:55 p6eval        nqpnet: OUTPUT«␤Unhandled Exception: System.InvalidOperationException: No such method bind_pos␤  at Rakudo.Metamodel.KnowHOW.KnowHOWBootstrapper.<Bootstrap>m__7 (Rakudo.Runtime.ThreadContext TC, Rakudo.Metamodel.RakudoObject Ignored, Rakudo.Metamodel.RakudoObject Cap) [0x00000] in <filename unkno…
20:55 diakopter     nqpnet: Capture.new.bind_pos(0, 4);
20:51 diakopter     lc()
20:51 * jnthn       needs to make those failures less noisy
20:51 p6eval        nqpnet: OUTPUT«␤Unhandled Exception: System.NullReferenceException: Object reference not set to an instance of an object␤  at NQPOutput_1290459063.block_6 (Rakudo.Runtime.ThreadContext TC, Rakudo.Metamodel.RakudoObject Block, Rakudo.Metamodel.RakudoObject Capture) [0x00000] in <filename unknown>:…
20:51 diakopter     nqpnet: sub foo(:$x) { say($x // 'ok'); }; foo();
20:50 p6eval        nqp: OUTPUT«ok␤»
20:50 jnthn         nqp: sub foo(:$x) { say($x // 'ok'); }; foo();
20:02 dalek         tpfwiki: (Herbert Breunung)++ | http://www.perlfoundation.org/perl6/index.cgi?perl_6_variable_tablet
20:00 moritz_       sorear: uhm, don't we have plenty of free ban slots?
19:46 dalek         tpfwiki: (Herbert Breunung)++ | http://www.perlfoundation.org/perl6/index.cgi?perl_6_variable_tablet
19:44 colomon       ;)
19:43 TimToady      now that sorear++ is drawing within 9% of STD, I'd  better start hacking on the std STD some more...  :)
19:37 sorear        moritz_: please don't waste perfectly good ban slots on lambdabot, not while I have an admin bit on it anyhow
19:36 diakopter     heh
19:36 sorear        @part #perl6
19:35 lichtkind     lue: hi
19:35 lichtkind     Tene: thnaks but moritz done this already :)
19:34 lue           hello o/
19:32 sorear        good * #perl6
19:32 sorear        81.95%, 54m 45s...
19:18 dalek         rakudo: review: https://github.com/rakudo/rakudo/commit/90cb8447157443ee18f53b3d8de03c025a9976b1
19:18 dalek         rakudo: [conf] rename revision comparison module
19:18 dalek         rakudo: 90cb844 | moritz++ | / (4 files):
19:00 phenny        Tene: I'll pass that on when cygx is around.
19:00 Tene          phenny: tell cygx please add http://perl6.org/documentation/ in the link section below on http://try.rakudo.org/
18:19 dalek         tpfwiki: (Herbert Breunung)++ | http://www.perlfoundation.org/perl6/index.cgi?perl_6_variable_tablet
17:25 dalek         tpfwiki: (Herbert Breunung)++ | http://www.perlfoundation.org/perl6/index.cgi?perl_6_delta_tablet
17:22 lichtkind     recently
17:21 lichtkind     perl 6 gets much more funny revently
17:14 jnthn         :)
17:14 jnthn         Maybe now it's a Lol
17:14 jnthn         I think it used to be a Slicel, but... :)
17:13 * colomon     suspects the spec has changed since we implemented Xop....
17:12 colomon       what sort of object is (<a b>; 1,2)?  lol?
17:09 dalek         tpfwiki: (Herbert Breunung)++ | http://www.perlfoundation.org/perl6/index.cgi?perl_6_delta_tablet
17:08 jnthn         colomon: I didn't think tehre was a .cross method either.
17:07 colomon       huh.  I don't think we implemented it that way....
17:06 lichtkind     colomon: http://perlcabal.org/syn/S03.html#Cross_operators
17:00 colomon       .cross?
16:58 dalek         tpfwiki: (Herbert Breunung)++ | http://www.perlfoundation.org/perl6/index.cgi?perl_6_variable_tablet
16:41 p6eval        rakudo : OUTPUT«Method 'cross' not found for invocant of class 'Array'␤  in main program body at line 22:/tmp/Ly4m3WwZjn␤»
16:41 jnthn         rakudo: my (@foo, @bar); @foo.cross(@bar)
16:41 lichtkind     dont worry i dont boost your ego just extract some vital infos for the docs :)
16:41 jnthn         lichtkind: interview...uh-oh. :)
16:38 lichtkind     jnthn: i want you to interview again for the perl 6 history tablet
16:37 lichtkind     ciau
16:36 jnthn         ahoj, lichtkind
16:34 lichtkind     jnthn: greets
16:19 jnthn         evening, #perl6
15:42 lichtkind     thank you
15:42 moritz_       lichtkind: I've added the link to the template. The next update from source should get it
15:39 moritz_       lambdabot? didn't we kick and ban that spammy bot for good?
15:38 lichtkind     dalek tell cygx please add http://perl6.org/documentation/ in the link section below on http://try.rakudo.org/
15:38 lambdabot     Unknown command, try @list
15:38 lichtkind     @dalek tell cygx please add http://perl6.org/documentation/ in the link section below on http://try.rakudo.org/
15:37 lichtkind     dalek: tell cygx please add http://perl6.org/documentation/ in the link section below on http://try.rakudo.org/
15:19 lichtkind     please add http://perl6.org/documentation/ in the link section below
15:19 lichtkind     who is responsible for http://try.rakudo.org/ ?
15:18 lichtkind     moritz_: but i have friends in ffm so have several reasons
15:18 lichtkind     moritz_: i understand that one :)
15:16 moritz_       negative; too much to do
15:15 lichtkind     yes
15:14 moritz_       so, upcoming weekend
15:10 lichtkind     moritz_: http://wiki.perl-community.de/Community/PlanungCommunityWorkshop2010#Wann
15:09 * MindosCheng living in Tropical and subtropical for most of his life.
15:08 dalek         tpfwiki: (Herbert Breunung)++ | http://www.perlfoundation.org/perl6/index.cgi?perl_6_variable_tablet
15:07 moritz_       (though the answer is most likely "no" anyway)
15:07 moritz_       lichtkind: when?
15:06 lichtkind     moritz_: will you come to frankfurt?
15:06 lichtkind     moritz_: true
15:04 moritz_       when you go to work by bike, snow is less annoying than rain
15:04 lichtkind     have to protect my herbs in the garden a bid
15:04 lichtkind     moritz_: its actually snow here
15:03 tadzik        :)
15:02 moritz_       (rainy day here, need my arms for protection)
15:02 moritz_       /o\
15:01 MindosCheng   \o/
13:27 takadonet     o/
13:26 colomon       \o
13:12 takadonet     morning all
11:16 dalek         niecza: review: https://github.com/sorear/niecza/commit/4b8ffc9f1a2c737743164afea864fc5924a6e6a6
11:16 dalek         niecza: Fix <.ws> LTM
11:16 dalek         niecza: 4b8ffc9 | sorear++ | lib/Cursor.cs:
11:16 sorear        actually at this point it looks like a runtime bug
11:15 sorear        niecza is miscompiling the STD parser
11:15 moritz_       sorear: thought so
11:15 sorear        moritz_: EWRONGMETALEVEL
11:01 p6eval        niecza 28f112a: OUTPUT«foo␤»
11:00 moritz_       niecza: say q"foo"
10:58 sorear        more breakage - q"foo" -> Undeclared routine
10:20 dalek         niecza: review: https://github.com/sorear/niecza/commit/28f112a757ef2d6f553d144dd8f8b9a1de17c71b
10:20 dalek         niecza: Fix stupid bug in alpha
10:20 dalek         niecza: 28f112a | sorear++ | / (3 files):
10:20 dalek         niecza: review: https://github.com/sorear/niecza/commit/4dc0dbe35836937590ee64cffddeff741fe290dd
10:20 dalek         niecza: Subrule entry/exit for RX trace
10:20 dalek         niecza: 4dc0dbe | sorear++ | lib/Cursor.cs:
09:53 masak         wasn't meaning to introduce an actual bug, just to make a bad pun.
09:53 jnthn         Let me guess. That was not the bug you were looking for. ;)
09:53 moritz_       that's what you get for not testing the code you show us :-)
09:53 masak         oops.
09:53 jnthn         } not )
09:50 masak         class Bug { has $.name }; my Bug $bug .= new( :name<Spot> }; # Spot the bug! :)
09:21 moritz_       not too hard :-)
09:21 sorear        token alpha { <+INTERNAL::alpha+[_]> \w* } # Spot the bug
09:20 moritz_       (for the innocent bystanders: transfiguring things into liquids and gases is strictly forbidden, because transfigurations are not permanent, and it will harm life if some part of the body suddenly turns into something else)
09:17 masak         hah :)
09:17 moritz_       masak: in the "order of the phoenix" movie Dumbledore turns some attacking magic by Voldemort into water :-)
09:16 masak         moritz_: yes, last chapter.
09:15 moritz_       masak: have you read about the safety rules for transfiguration already?
09:15 moritz_       there were other nits too, but most are really minor
09:15 moritz_       masak: I'm currently supervising a lab experiment about lenses :-)
09:14 masak         moritz_: after reading part of the fanfic, it amuses me that your only nit seems to be about some refraction index or other. :P
09:13 moritz_       there's even a scene that's blurred before he puts on the glasses
09:12 masak         he's just pretending to be near-blind without them. simplest explanation.
09:12 HarryS        off-camera
09:12 HarryS        he got laser correction
09:11 moritz_       I found it funny how Harry is supposed to see nearly nothing without his glasses, but you can look straight through the glasses without any distortions :-)
09:10 moritz_       not bad, IMHO
09:09 * moritz_     saw the movie yesterday
09:08 masak         I've had that as a tradition, to pre-read the books lazily.
09:08 masak         luckily, I finished the last book right before I went to see the first last movie :)
09:07 moritz_       it contains a lot of spoilers if you don't know all 7 books
09:05 masak         oh, ok.
09:05 moritz_       it's 60 chapters now, and maybe 2 more to go
09:05 moritz_       but it's close to the end, I think
09:03 masak         it's still in progress? I'll read slowly, then :)
09:03 moritz_       masak: I'm through all the chapters now (finished on Saturday), and waiting for the next chapter :-)
09:02 masak         moritz_: by the way, I'm .25 through http://www.fanfiction.net/s/5782108/1/Harry_Potter_and_the_Methods_of_Rationality -- it's way above my (admittedly unfounded) expectations on fanfic.
08:54 sorear        also broken: also is Foo; $Foo (but $! fine).  All signs point to longname
08:44 sorear        module { } # still works.  I guess I broke <longname>
08:44 sorear        actually, it's trying to parse Foo as the block
08:40 sorear        my module Foo { } # Unable to parse module definition
08:40 frettled      lue: Z tight
08:39 sorear        and I've just utterly broken STD with that last push
08:37 lue           I think I've burned my eyeballs and relevant portions of my brain for tonight. Goodnight o/
08:36 lue           The Front for the Liberation of Palestinian People of course!
08:34 frettled      ZzzZ?
08:33 diakopter     z
08:33 TimToady      zzz &
08:29 frettled      lue: But are we talking about the People's Front for the Liberation Palestine, or the Popular Palestinian Liberation Front?
08:29 moritz_       not the first time for me :(
08:29 sorear        I think this is the first time I've seen spam on perlmonks
08:27 p6eval        rakudo : OUTPUT«No applicable candidates found to dispatch to for 'infix:<Z>'. Available candidates are:␤:()␤:(Any $lhs, Any $rhs)␤␤  in main program body at line 22:/tmp/C9tNScxicR␤»
08:27 masak         rakudo: 1,2,3 Z 4,5,6 Z 7,8,9; say "alive"
08:27 p6eval        alpha : OUTPUT«alive␤»
08:27 masak         alpha: 1,2,3 Z 4,5,6 Z 7,8,9; say "alive"
08:27 moritz_       zip() doesn't work either
08:26 masak         except it doesn't work as a list-assoc infix in Rakudo :(
08:26 moritz_       I was wondering if I should :-)
08:26 masak         I was thinking the same.
08:26 sorear        you can still introduce infix:<Z>
08:26 masak         "(Wow, I got the Perl 6 reply in before moritz!)" # :D
08:25 lue           moritz_:  I suddenly get an image of Catholicism and Protestantism, only s/Catholicism/Perl 5/ and s/Protestantism/Perl 6/ :)
08:25 masak         sorear: but sometimes, I can't not use them...
08:24 sorear        masak: there is dialectal variation on the interpretation of double negatives in English, so it's best to avoid them in prose intended to be portable.
08:24 masak         lue: I have no memory of that. but you may be right.
08:23 lue           masak: release 0 was one way it was called initially [zero-indexing you know]
08:23 moritz_       \o/ http://www.perlmonks.org/?node_id=872821 other monks now start with Perl 6 replies :-)
08:21 lue           I need to change that, too lazy right now though :) [I will create a special one for 23 Nov. that'll be created to last around 1 year]
08:21 moritz_       :-)
08:20 masak         moritz_: if one "can't hardly wait", does that mean that one has almost no problem waiting? :P
08:20 masak         moritz_: the Update part of the top post is very cute, though. :)
08:19 dalek         niecza: review: https://github.com/sorear/niecza/commit/b85fe1c64b707161d8d7ae94c279251d1c10a287
08:19 dalek         niecza: Revert 4819d25 - STD depends on opposite behavior
08:19 dalek         niecza: b85fe1c | sorear++ | / (2 files):
08:19 sorear        ok. blasts work.
08:18 masak         lue: in your blog header, what does "Rakudo Star release 0" mean? as far as I know, there never was a release 0, and besides we're already on the fourth-soon-fifth release.
08:18 frettled      They wouldn't be Perl 5 users, by any chance? :)
08:17 moritz_       http://www.perlmonks.org/?node_id=872809 # it seems that for some developers it's REALLY hard to distinguis between a language and a compiler
08:16 TimToady      s/specced/specced (poorly)/  # :)
08:14 TimToady      note that STD's behavior is in fact specced at S05:2607
08:14 masak         this is why p6eval is perhaps the most important bot around, too. its technological innovation is in some sense secondary to its social contribution: it brings Actual Code and its evaluation into the IRC conversation.
08:11 * sorear      has been known to use "bionic memory" to refer to irclogs
08:10 masak         sorear: the Internet is not just some truck that I can dump stuff on. but backlog, RT, source repository, test suite and other similar digital containers are. I use them as only slightly detached extensions of my nervous system.
08:06 TimToady      yes, well, STD very much assumes that the LTM of A||B is the LTM of A
08:05 sorear        so I git blame the test suite and it tells me exactly when I made that decision
08:04 sorear        I know I made the decision for || to kill LTM on purpose
08:04 sorear        also, this is why a testsuite is awesome
08:01 sorear        TimToady: http://irclog.perlgeek.de/perl6/2010-09-07#i_2793271
08:00 lue           .oO(As the clock struck midnight, lue wondered if sleep is something he should fight. He thought, .oO(Oh, plenty fine I'll be, cos I've got plenty o' caffeine!))
07:58 * sorear      checks the irclogs
07:58 sorear        looks like I added the test on Sep 7 00:23
07:57 masak         scp1: cool. let me know how it goes.
07:56 TimToady      whereas the default case is requiring a block by calling <block> directly
07:56 TimToady      that's dropping into <statement> instead of <block>
07:56 p6eval        std a194beb: OUTPUT«===SORRY!===␤Bogus statement at /tmp/0jnDc4vQ4u line 1 (EOF):␤------> lazy⏏<EOL>␤Parse failed␤FAILED 00:01 119m␤»
07:56 TimToady      std: lazy
07:55 sorear        +s
07:55 scp1          masak, I've started to implement it with inotify. Nice idea btw.
07:54 * sorear      makes a one-line change to RxOp::SeqAlt::lad and runs the test
07:54 p6eval        std a194beb: OUTPUT«===SORRY!===␤Missing block at /tmp/WuUqaJdQ3d line 1:␤------> default ⏏42␤Parse failed␤FAILED 00:01 117m␤»
07:54 TimToady      std: default 42
07:53 TimToady      in the block case, we can get to it by multiple paths
07:53 TimToady      correct, that's how STD does LTM
07:52 sorear        so, concretely, "b" !~~ / [ [ a || b ] | c ] / ?
07:52 TimToady      in that particular case, we could warn that B is unreachable
07:52 masak         sorear: DIRECTIONS -- very nice. just what I was looking for.
07:52 TimToady      that's correct behavior
07:51 sorear        because LTM sees that neither [A || B] nor C has B as a prefix, and so tries neither
07:51 sorear        which will never match B
07:51 sorear        I had it that way earlier - it ran into issues with [ [A || B] | C ]
07:50 TimToady      in that case it might fail the ||
07:50 TimToady      it's only there if you call <block> outside of an LTM
07:50 TimToady      in this case, it can never call B via LTM
07:50 TimToady      yes
07:50 sorear        Does [A || B] in general make exactly the same LTM as A?
07:49 masak         sorear: looking.
07:48 TimToady      only for the right side of the ||
07:47 sorear        in <block>
07:46 sorear        Shouldn't the || stop LTM
07:45 TimToady      sorear: blast has an alternation of <block> | <statement>, and should never LTM into block unless it starts with the '{' from the <?before '{'>
07:43 lue           .oO(those ++'s look like bangheadworthy objects...)
07:43 sorear        masak: I'd like you to look at blob/master/docs/DIRECTIONS and tell me if that's what you wanted
07:42 dalek         niecza: review: https://github.com/sorear/niecza/commit/72c4e124d9ad60c6f14515e6a2c2c3823a914f21
07:42 dalek         niecza: [docs] prose status, more LHF stuff
07:42 dalek         niecza: 72c4e12 | sorear++ | docs/ (2 files):
07:42 masak         lue++ # blagging
07:42 masak         told you that more eyeballs would help :)
07:41 lue           *bangs head on nearby bangheadworthy object*
07:40 masak         HTH
07:40 masak         (and boy would I sometimes like to have debug output that told me such debug information about my grammars)
07:39 masak         lue: here's why the line '=begin Text' doesn't match <directive><delimited>: the string 'Text' doesn't march the rule <blocktype>
07:38 lue           well, at least its current incarnation as a file in the top directory. (maybe expand & mv to t/ ?)
07:37 masak         lue: your file test.pod sounds like a genuinely good idea. I don't think you should remove it as soon as things work.
07:36 masak         lue: you have a variable called $file in the sub 'importformat'. it seems to contain the slurped contents of the file. this qualifies as confusing naming to me. YMMV.
07:36 masak         | does LTM, || does strict left-to-right.
07:35 masak         no, that's not the difference.
07:35 lue           from quick skimming of S05, | would require terms to match the same beginning and end point, while || doesn't, right?
07:30 masak         lue: secondly, as a purely stylistic micro-optimization, prefer using comma (',') to tilde ('~') when printing several strings in a 'say' call. it creates fewer strings internally, is therefore marginally faster, and is easier on the eye.
07:29 lue           I'm referring to the way the encode them in URLs (the funny % stuff). I don't know, see that gives me the (false?) impression it hasn't caught up.
07:28 sorear        hex
07:28 sorear        lue: they have.  it's quite specifically UTF-8 then encode bytes in here
07:28 masak         lue: otherwise, when it finally is, all your code might break.
07:27 masak         lue: first off, I would recommend always using || until the specific behavior of | is implemented in Rakudo.
07:27 lue           .oO(Hm. You'd think whatever controls URL encoding would've caught up to the 21st century by now...)
07:26 lue           ololblagtime! https://rdstar.wordpress.com/2010/11/22/why-grammar-i-learn-cant%E2%80%BD/
07:22 * lue         almost done writing blag toast
07:18 * sorear      notices a lack of Encode:: calls in karmalog.pm
07:17 moritz_       mornin'
07:16 masak         morning, moritz_.
07:16 * sorear      debugs some more
07:16 moritz_       well, I might get a few downvotes, but I can cope with them :-)
07:15 sorear        the problem may be your name.  I hate encoding soup
07:15 tadzik        oh, hi moritz_ :)
07:15 tadzik        oh, someone courageous posted that on proggit
07:14 moritz_       http://www.reddit.com/r/programming/comments/e9to1/perl_6_grammars_not_only_for_parsing/
07:13 tadzik        oh wow, I'm mentioned on twitter again
07:11 tadzik        sorear: https://github.com/parrot/parrot/commit/8e5d0aca8 here I am
07:11 tadzik        edenc++ and ruoso++ for coming up with the idea
07:10 masak         tadzik++ # http://ttjjss.wordpress.com/2010/11/21/perl-6-grammars-%E2%80%93-not-only-for-parsing/
07:08 sorear        tadzik: most likely it doesn't exist
07:05 tadzik        sorear: what's wrong about the entry?
06:56 * masak       cheers sorear on
06:56 sorear        I've never done this "releasing" thing before
06:56 sorear        I'm not sure what that comprises though
06:55 sorear        I'd like to get niecza into a state where I can sanely ANNOUNCE: it before Tues
06:55 masak         sorear++
06:55 sorear        masak: I'm writing up a prose status document now
06:55 masak         scp1: yes. I've seen others use it for this purpose.
06:55 masak         also, when $loud is set, it uses the Mac OS X-specific command-line program 'say'.
06:55 scp1          Linux::Inotify2 is great (and inotify rocks)
06:54 masak         seems I just poll modification times once every $SLEEP_TIME seconds.
06:54 masak         I wish.
06:53 scp1          using inotify I guess?
06:53 masak         scp1: https://gist.github.com/709620
06:53 masak         doesn't need to do that at all. but it does.
06:52 masak         oh yes. it hardcodes the files it's monitoring.
06:52 masak         it's still too tied to individual repositories, if I recall correctly.
06:52 masak         I can make a gist.
06:50 scp1          masak, you should push it, I want to see the source
06:48 masak         uppsala here.
06:48 masak         ooh.
06:48 scp1          masak, norrköping - I saw your github page
06:47 masak         scp1: are you from .se?
06:47 masak         I had tote set up to verbally reward me when I passed more tests. it says "checkpoint! yay!", which seems to trigger endorphin releases.
06:47 scp1          masak, ooh... the world is small mate. god morgon ;)
06:46 masak         it's actually a little bit dangerous, from the viewpoint of forgetting to eat etc.
06:46 scp1          I have yet to manage to _install_ rakudo though :C Still stuck with perl5
06:46 scp1          heh, nice
06:46 masak         scp1: (it rocks)
06:45 masak         scp1: oh, I've used it privately for over a year now...
06:45 masak         lue: and what sorear said. it's more of an engineering thing.
06:45 scp1          masak, Cool, have you started hacking on it yet? :)
06:45 masak         lue: I don't have a formal CS background either.
06:45 sorear        TDD isn't taught in real CS classes anyway
06:44 phenny        U+203D INTERROBANG (‽)
06:44 lue           .u interrobang
06:44 lue           I have yet to take a formal CS class (the best offered at my high school is the usual "And this is how you open WORD! Word is used to WRITE PAPERS AND STUFF!!1!")
06:43 masak         sorear: "Completed ROADMAP items" in https://github.com/rakudo/rakudo/blob/master/docs/ROADMAP
06:42 masak         scp1: http://strangelyconsistent.org/blog/helpfully-addictive-tdd-on-crack
06:42 sorear        masak: hrm.  Is there a good example of what you're looking for in Rakudo?
06:42 scp1          What's tote?
06:42 masak         oops. should have released tote by now... :/
06:42 masak         it's even more awesome with something like tote.
06:42 scp1          It's awesome
06:41 masak         lue: I see you have some TDD to learn and look forward to. when you do, I bet you'll like it.
06:41 sorear        TimToady: How is <blast> supposed to work?  niecza is entering <blast> and immediately hitting <?before '{'> || <.panic: "Missing block">, and I can't see any reason why it *shouldn't*
06:41 masak         lue: what scp1 said.
06:41 scp1          tests first! then more tests! the documenation, and then code!
06:41 masak         sorear: unless you're referring to a big comment header, that's not the level of accessibility I was referring to.
06:40 lue           Let me say this: I haven't written a single test yet because there's nothing to test! :(
06:40 sorear        in the past I'd been relying on test.pl for that
06:39 masak         sorear: how would you describe the current state of the niecza project? is there a file somewhere that describes the current state of the niecza project?
06:37 dalek         niecza: review: https://github.com/sorear/niecza/commit/62645be2fd5a4ddae949f7f3b684d2bc7e9e8a7b
06:37 dalek         niecza: [v6] Add optional Niecza::Grammar use
06:37 dalek         niecza: 62645be | sorear++ | / (2 files):
06:37 masak         lue: as soon as the first test passes -- which is, like, five minutes into coding, right? -- you have something that no longer doesn't work. :)
06:36 * lue         writes blog post
06:36 lue           I will publish the problem code in my blog post (Although putting something on git that *doesn't work* seems silly to me, so not that kind of publishing.)
06:36 sorear        this stuff is breaking a lot
06:36 sorear        hee.
06:35 masak         it's official. sorear has now transformed into a Perl 6 compiler.
06:34 sorear        ===SORRY!=== Regex()<instance> at (eval) line 1
06:34 masak         every time I do that, good comes out of it in one way or other.
06:34 masak         whole or in parts.
06:33 masak         also, you should publish your code.
06:33 lue           OK-ish. I think there's a flaw in my understanding of grammars .oO(I should make a blag toast)
06:33 masak         you're welcome :)
06:33 lue           thanks for turning me to theoatmeal.com, by the way
06:32 masak         lue: how's the Pod hacking coming along?
06:31 masak         greetings, lue.
06:30 lue           hai masak o/
06:23 TimToady      sorear: I can't answer that until I know how they diverge
06:12 colomon-droid o/
05:51 sorear        TimToady: which is more correct - the speccish rules about bracketed postfix constructs, or the STDish rules about <EXPR(item %methodcall)> <?after <[ \] ) } > ]>> ?
05:50 lue           ohayo o/
05:06 masak         nom &
05:06 p6eval        std a194beb: OUTPUT«ok 00:01 121m␤»
05:06 masak         std: sub syntax {}; sub error {}; my $x; "$x[syntax error]"
04:59 masak         \o/
04:58 TimToady      that's better
04:58 p6eval        std a194beb: OUTPUT«ok 00:01 120m␤»
04:58 TimToady      std: my $x; "$x\[syntax error]"
04:58 p6eval        std a194beb: OUTPUT«===SORRY!===␤Undeclared routines:␤        'error' used at line 1␤ 'syntax' used at line 1␤Check failed␤FAILED 00:01 121m␤»
04:58 TimToady      std: my $x; "$x[syntax error]"
04:57 masak         nope.
04:57 TimToady      not a bug
04:57 TimToady      oh, of course
04:57 TimToady      hmm
04:57 masak         :)
04:57 TimToady      oops
04:57 p6eval        std a194beb: OUTPUT«===SORRY!===␤Variable $y is not predeclared at /tmp/sMj1q7mt8A line 1:␤------> my $x; "$x\[$y⏏]"␤Check failed␤FAILED 00:01 120m␤»
04:57 TimToady      std: my $x; "$x\[$y]"
04:57 p6eval        std a194beb: OUTPUT«===SORRY!===␤Variable $y is not predeclared at /tmp/IFbJELh2v3 line 1:␤------> my $x; "$x[$y⏏]"␤Check failed␤FAILED 00:01 121m␤»
04:56 TimToady      std: my $x; "$x[$y]"
04:56 p6eval        rakudo : OUTPUT«<␤»
04:56 masak         rakudo: say "\<"
04:56 * masak       submits rakudobug
04:55 p6eval        rakudo : OUTPUT«postcircumfix:<{ }> not defined for type Str()␤  in main program body at line 1␤»
04:55 masak         rakudo: my $foo = "work"; say "This should $foo\</a>"
04:55 masak         er.
04:55 p6eval        rakudo : OUTPUT«This should work␤»
04:55 masak         rakudo: my $foo = "work"; say "This should $foo\[]"
04:55 masak         oh, I remember that now.
04:55 TimToady      metoo, which is why we outlawed unspace in quotes (at top level only, of course)
04:54 masak         oh, and by the way: that's good news. I like that a backslash solves the problem.
04:53 * masak       updates the post
04:53 masak         TimToady++
04:53 TimToady      that's why I said it.  :)
04:53 masak         that is news to me.
04:53 TimToady      but it's specced to work
04:53 masak         I see.
04:52 TimToady      mind, rakudo doesn't implement it right yet
04:52 masak         oh!
04:52 TimToady      that's in a qq, so it's not an unspace
04:52 TimToady      Won't work, due to a wonderful little invention called unspace,
04:52 masak         TimToady: where?
04:51 masak         so I must have written something unclearly somewhere.
04:51 masak         er. but I know that part.
04:51 * masak       does so
04:50 TimToady      see S02:3802
04:50 TimToady      masak: your blog is incorrect
04:48 masak         ah. I'd never understood that usage of 'interrupt vector'. thanks.
04:47 sorear        code indirections get called vectors sometimes.  especially stuff like interrupt vectors and opcode vectos
04:47 masak         I know that the set of multis that can be seen can vary between every lexical scope. what I meant with 'lexical caching' was simply storing that information on the lexical block somewhere after calculating it once.
04:47 sorear        revectoring is just making a sub point somewhere else
04:46 masak         :)
04:46 sorear        I'm not familiar with the term 'lexical caching'.
04:46 masak         is it related to the thing with lexical caching?
04:45 masak         I'm not familiar with the term 'revectoring'.
04:45 sorear        but it's not
04:45 sorear        look at jnthn's problems.  sub revectoring can be orthogonal to dispatch
04:45 masak         I especially like the fact that a wrapping can be hotswapped without the programmer having to worry about the other wrappings above and/or below.
04:44 masak         can you back that up with reasons? because to the extent that I grok it, I quite like it.
04:43 sorear        the wrapping process feels brokenly overengineered to me
04:38 masak         more generally, can the wrapping process be written out as Perl 6?
04:38 masak         but I wanted to ask: is there a way to query a wrap handle for the underlying wrapped function?
04:37 masak         this is not spec-conformant, and there's a rakudobug about it in RT already.
04:37 p6eval        rakudo : OUTPUT«2␤»
04:37 masak         rakudo: sub foo {}; say &foo.wrap({ nextsame })
03:34 masak         that works, though.
03:34 p6eval        rakudo : OUTPUT«42␤»
03:34 masak         rakudo: class A { my $x = 42; method foo { say $x } }; A.new.foo
03:32 * masak       submits rakudobug
03:28 masak         ah. that explains a lot.
03:28 p6eval        rakudo : OUTPUT«Method 'x' not found for invocant of class 'A'␤  in 'A::foo' at line 22:/tmp/Vf8EKOC9DJ␤  in main program body at line 22:/tmp/Vf8EKOC9DJ␤»
03:28 masak         rakudo: class A { my $.x = 42; method foo { say $.x } }; A.new.foo
03:28 p6eval        alpha : OUTPUT«42␤»
03:28 masak         alpha: class A { my $.x = 42; method foo { say $.x } }; A.new.foo
03:27 masak         oh. thought you meant between lots of Perl 6 and little socializing with friends. ;)
03:26 diakopter     (in case someone didn't give me the benefit of the doubt in that statement :))
03:26 diakopter     between Perl 6 and socializing with friends, I mean, of course
03:26 masak         heh :)
03:25 diakopter     masak: wait. what's the distinction? ;)
03:04 masak         sorry about the late. I deviated from my usual strict regimen of lots of Perl 6 and no socializing with friends. ;)
03:03 masak         blog post! http://strangelyconsistent.org/blog/november-21-2010-just-a-few-wrinkles-to-iron-out
02:58 dalek         6model: review: https://github.com/jnthn/6model/commit/5f2da698d8841e2cf2925e6dc5aa1f70cbaf7772
02:58 dalek         6model: [Regex] start on PAST::Regex emission. Pending moar help from jnthn++
02:58 dalek         6model: 5f2da69 | diakopter++ | / (3 files):
02:54 dalek         niecza: review: https://github.com/sorear/niecza/commit/907d2774986ba96801bbc99713f1eb99b1ca4bd8
02:54 dalek         niecza: Box constants exactly once
02:54 dalek         niecza: 907d277 | sorear++ | / (2 files):
02:54 dalek         niecza: review: https://github.com/sorear/niecza/commit/148689cfb052d6c7a4df65f536eebd346a103091
02:54 dalek         niecza: Use a more efficient method for boxing constants
02:54 dalek         niecza: 148689c | sorear++ | / (4 files):
02:33 masak         \o
02:33 colomon       o/
02:09 masak         seems so. tadzik++
02:08 masak         ttjjss === tadzik?
02:08 masak         seems we need to do more of this: http://twitter.com/NegativeK/status/6463457343832065
02:07 Khisanth      indeed needs more colors ...
02:06 diakopter     who knows if "C" is in programming context, though.
02:06 diakopter     SQL & C are way up there at 5% (of *all* job descriptions)
02:05 diakopter     pretty soon, all programming job descriptions will include all of these keywords ;)
02:05 diakopter     :P
02:05 masak         that sounds scary on many levels.
02:04 diakopter     python rises quickly too
02:04 masak         worse is better.
02:04 diakopter     scarily, the growth of php continues unabated
02:03 diakopter     Perl seems to be converging with C++ though, maybe overtaking it eventually :)
02:03 masak         yes.
02:03 diakopter     oh wait.. /me misread too. C# & javascript overtook Perl in mid '06\
02:00 masak         then I have no further questions, your honor.
02:00 masak         right, that's C#.
02:00 diakopter     C# & javascript overtaking perl recently
02:00 masak         oh.
02:00 diakopter     cobol's at the bottom
02:00 diakopter     where do you see that
01:59 masak         diakopter: COBOL overtaking Perl? apart from that, I don't see anything in that graph that surprises me.
01:58 * masak       is late with his November blogging today
01:58 diakopter     http://www.indeed.com/jobtrends?q=cobol,+ada,+ruby,+python,+flex,+php,+perl,+C%2B%2B,+C%23,+javascript,+java&l=
01:58 masak         :)
01:58 ch3ck         gugu masak
01:57 masak         'sup, #perl6.
00:52 dalek         tpfwiki: (Herbert Breunung)++ | http://www.perlfoundation.org/perl6/index.cgi?perl_6_index_tablet
00:47 dalek         tpfwiki: (Herbert Breunung)++ | http://www.perlfoundation.org/perl6/index.cgi?perl_6_lookup_tablet
00:31 dalek         tpfwiki: (Herbert Breunung)++ | http://www.perlfoundation.org/perl6/index.cgi?perl_6_index_tablet
00:29 ch3ck         bye
00:29 lichtkind     kthxgby :)
00:28 ch3ck         lichtkind: at the moment i don't need help, thx
00:28 lichtkind     ch3ck: almost leaving can i help you?
00:27 ch3ck         hi lichtkind
00:27 lichtkind     hai ch3ck
00:26 ch3ck         rehi
00:20 dalek         tpfwiki: (Herbert Breunung)++ | http://www.perlfoundation.org/perl6/index.cgi?perl_6_index_tablet
00:14 lichtkind     TimToady: and should chars for consistancy not return a list with all character which would in num context do the same as now?
00:09 lichtkind     mean
00:09 lichtkind     i mena the lines command
00:09 lichtkind     when does lines know when to read a file, when its a string quoted with the adverb :path ?