perl6-projects.org/ | nopaste: sial.org/pbot/perl6 | evalbot: 'perl6: say 3;' | irclog: irc.pugscode.org/
Set by mncharity on 25 March 2009.
pasteling "ZuLu" at 193.140.225.206 pasted "use v6; my @aral" (5 lines, 53B) at sial.org/pbot/35734 01:05
ZuLuuuuuu Hello this code makes CPU 100%: sial.org/pbot/35734
and does not display any result
what might be the problem?
I'm using Rakudo on Parrot 1.0 01:06
bacek ZuLuuuuuu: lazy lists are not implemented yet 01:09
meppl good night
ZuLuuuuuu bacek: oh ok thanks
meppl: good night
meppl ;)
s1n frioux: pong 01:39
ZuLuuuuuu bacek: smart matching is also not fully imlemented I guess? 01:45
bacek ZuLuuuuuu: which part?
ZuLuuuuuu like 01:46
I try to smart match
a rule substitution with a string
or an undefined value with a scaler variable
bacek rakudo: say "Matched" if undef ~~ 1; 01:48
p6eval rakudo 7af829: OUTPUT«Use of uninitialized value␤»
bacek rakudo: say "Matched" if 'a' ~~ /<a>/;
p6eval rakudo 7af829: OUTPUT«Unable to find regex 'a'␤Null PMC access in invoke()␤current instr.: 'parrot;PGE;Grammar;' pc 287 (EVAL_17:124)␤» 01:49
bacek rakudo: say "Matched" if 'bar' ~~ /a'/; 01:49
rakudo: say "Matched" if 'bar' ~~ /a/;
p6eval rakudo 7af829: OUTPUT«perl6regex parse error: No closing ' in quoted literal at offset 31, found ''␤current instr.: 'parrot;PGE;Perl6Regex;parse_error' pc 10179 (compilers/pge/PGE/Perl6Regex.pir:1219)␤»
rakudo 7af829: OUTPUT«Matched␤»
ZuLuuuuuu hmmm it worked
rakudo: say "Matched" if $isim ~~ undef; 01:50
p6eval rakudo 7af829: OUTPUT«Scope not found for PAST::Var '$isim' in ␤current instr.: 'parrot;PCT;HLLCompiler;panic' pc 146 (src/PCT/HLLCompiler.pir:102)␤»
ZuLuuuuuu shouldn't this say "Matched"
bacek rakudo: say "Matched" if my $isim ~~ undef; 01:52
p6eval rakudo 7af829: OUTPUT«Matched␤» 01:53
ZuLuuuuuu oh ok 01:54
rakudo: say "Matched" if 'bar' ~~ s/a/b/; 01:56
p6eval rakudo 7af829: OUTPUT«Statement not terminated properly at line 1, near "/;"␤␤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:83)␤»
bacek std: say "Matched" if 'bar' ~~ s/a/b/; 01:58
p6eval std 26019: OUTPUT«ok 00:04 36m␤»
bacek looks like a bug
s1n frioux: i have some time blocked off for tomorrow, i'll be on after dinner 03:36
frioux s1n: eh, I lose freedom after 3pm till ca. 10 tomorrow 04:31
sorry
frioux anyone know where pir.vim is? 05:11
bacek frioux: pugs/editor 05:13
frioux: oops. sorry. 05:14
parrot/editor.
You have to "make" in this directory
frioux one sec
ah 05:15
great
thanks
rakudo: 5 % 3 == 0 05:16
p6eval rakudo 7af829: RESULT«Bool::False»
frioux rakudo: 6 % 3 == 0
p6eval rakudo 7af829: RESULT«Bool::True»
frioux Does the init method get called when a class is instantiated with new? 05:20
no. 05:21
frioux rakudo: 2 % 4 == 0 05:22
p6eval rakudo 7af829: RESULT«Bool::False»
frioux rakudo: 2 % 4
p6eval rakudo 7af829: RESULT«2»
frioux rakudo: 4 % 2 == 0
p6eval rakudo 7af829: RESULT«Bool::True» 05:23
frioux is sad that you can't put method calls into junctions with smartmatches yet 05:25
bacek rakudo: say "Matched" if .lc eq "foo" ~~ "Bar"|"FOO" 05:27
p6eval rakudo 7af829: OUTPUT«Use of uninitialized value␤»
bacek rakudo: say "Matched" if .lc eq "foo" ~~ ("Bar"|"FOO")
p6eval rakudo 7af829: OUTPUT«Use of uninitialized value␤»
frioux I meand
bacek hmm...
frioux er
meant
$f ~~ .bar & .baz
bacek std: my $f; $f ~~ .bar & .baz 05:28
p6eval std 26019: OUTPUT«ok 00:02 35m␤»
frioux I tried it with a minutes old build of rakudo though, it doesn't work atm 05:29
bacek rakudo: say "hi" if "foo" ~~ .lc 05:30
p6eval rakudo 7af829: OUTPUT«hi␤»
bacek rakudo: say "hi" if "foo" ~~ (.lc & .uc)
p6eval rakudo 7af829: OUTPUT«Use of uninitialized value␤Use of uninitialized value␤»
frioux rakudo: "foo" ~~ .uc 05:31
p6eval rakudo 7af829: RESULT«Bool::True»
frioux rakudo: "foo" ~~ .uc & .lc
p6eval rakudo 7af829: OUTPUT«Use of uninitialized value␤Use of uninitialized value␤»
frioux rakudo: "foo" ~~ (.uc & .lc)
p6eval rakudo 7af829: OUTPUT«Use of uninitialized value␤Use of uninitialized value␤»
bacek rakudo: class Foo { method bar { say "bar"; 1 } }; my $f = Foo.new; say "hi" if $f ~~ .bar; 05:32
p6eval rakudo 7af829: OUTPUT«bar␤hi␤»
bacek rakudo: class Foo { method bar { say "bar"; 1 } }; my $f = Foo.new; say "hi" if $f ~~ (.bar & .bar); 05:33
p6eval rakudo 7af829: OUTPUT«Method 'bar' not found for invocant of class 'Failure'␤current instr.: 'parrot;P6metaclass;dispatch' pc 637 (src/classes/ClassHOW.pir:161)␤»
frioux yep
that's what I get here
bacek frioux: parser error.
frioux well
I didn't use say ... if
bacek rakudo: class Foo { method bar { say "bar"; 1 } }; my $f = Foo.new; say "hi" if $f ~~ all(.bar, .bar);
frioux I did if ... { }
p6eval rakudo 7af829: OUTPUT«Method 'bar' not found for invocant of class 'Failure'␤current instr.: 'parrot;P6metaclass;dispatch' pc 637 (src/classes/ClassHOW.pir:161)␤»
bacek rakudo: class Foo { method bar { say "bar"; 1 } }; my $f = Foo.new; say "hi" if $f ~~ all(&Foo.bar, &Foo.bar); 05:34
p6eval rakudo 7af829: OUTPUT«bar␤bar␤get_number() not implemented in class 'Foo'␤current instr.: 'parrot;Int;ACCEPTS' pc 4751 (src/classes/Int.pir:50)␤»
bacek rakudo: class Foo { method bar { say "bar"; True } }; my $f = Foo.new; say "hi" if $f ~~ all(&Foo.bar, &Foo.bar);
p6eval rakudo 7af829: OUTPUT«bar␤bar␤Method 'ACCEPTS' not found for invocant of class ''␤current instr.: 'parrot;Junction;ACCEPTS' pc 9140 (src/classes/Junction.pir:154)␤»
frioux see? it's sad. 05:36
bacek rakudo: class Foo { method bar { say "bar"; True } }; my $f = all(&Foo.bar, &Foo.bar); say $f.perl 05:43
p6eval rakudo 7af829: OUTPUT«bar␤bar␤all(.new())␤» 05:44
bacek rakudo: class Foo { method bar { say "bar"; True } }; my $f = Foo.new; say "hi" if $f ~~ all( { .bar }, { .bar } );
p6eval rakudo 7af829: OUTPUT«bar␤bar␤hi␤»
bacek ok. This is bug.
And this is workaround :) 05:45
frioux why did it work in a block?
I like the workaround thought :-)
bacek rakudo evaluates methods too early.
frioux ah 05:46
got it
in other news: should this fail: { .foo 'bar' }
because it does
bacek std: { .foo 'bar' } 05:47
p6eval std 26019: OUTPUT«##### PARSE FAILED #####␤Malformed block␤Syntax error (two terms in a row?) at /tmp/D3te9cpxND line 1:␤------> { .foo 'bar' }␤ expecting any of:␤ infix or meta-infix␤ infix stopper␤ standard stopper␤ statement modifier loop␤ terminator␤FAILED 00:02
..35m␤»
bacek it should :)
frioux huh
why?
bacek "two terms in row" 05:49
std: { .foo.assuming('bar') } 05:50
p6eval std 26019: OUTPUT«ok 00:02 35m␤»
frioux eh
I'd just use regular parens then
what if I wanted to do this:
my @codes = @bars.map: { .foo($_) }
but not yet?
like...
to be evaluated later 05:51
my @codes = @bars.map: { { .foo($_) } } ?
bacek if you have class Bar { method foo(Bar $b) { ... } };
rakudo: my @codes = <a b c>.map: { .uc }; say ~@codes;
p6eval rakudo 7af829: OUTPUT«A B C␤» 05:52
bacek there is block already
frioux rakudo: <a b c>.map: { .uc }
p6eval rakudo 7af829: RESULT«["A", "B", "C"]»
frioux doesn't look like a block to me.. 05:53
bacek { .uc } is block
frioux well yeah
but I am saying I want to do this:
bacek Or you want create a lot of closures?
frioux yeah
that
I tried map: { sub { .. } }
but it wasn't happy
bacek rakudo: my @c=<a b c>.map { sub { .uc } }; say "hi"; say ~@c 05:54
p6eval rakudo 7af829: OUTPUT«Statement not terminated properly at line 1, near "{ sub { .u"␤␤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:83)␤»
bacek rakudo: my @c=<a b c>.map { -> { .uc } }; say "hi"; say ~@c.map: { .() }
p6eval rakudo 7af829: OUTPUT«Statement not terminated properly at line 1, near "{ -> { .uc"␤␤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:83)␤»
bacek rakudo: my @c=<a b c>.map: { -> { .uc } }; say "hi"; say ~@c.map: { .() } 05:55
p6eval rakudo 7af829: OUTPUT«hi␤C C C␤»
frioux hoho!
bacek oh shi...
frioux that's a weird bug
bacek rakudo: my @c=<a b c>.map: { -> { .uc }.assuming($_) }; say "hi"; say ~@c.map: { .() } 05:56
p6eval rakudo 7af829: OUTPUT«hi␤C C C␤»
bacek summon masak
afk # f1 Australian GP
frioux bacek: also there is something wrong it seems with that first thing 06:00
it gets the invocant wrong 06:01
and things $_ is the invocant
I want the invocant to be set at ~~ time
but hey, I got my workaround for the day :-)
literal @tell masak your comments are requested -> tinyurl.com/c8j8vw 07:29
lambdabot Consider it noted.
lucs literal: Is it just me, or are the lines annoyingly wrapped in that proposal? 07:33
literal yeah, the google groups thing apparently cuts them off at 70-something characters 07:34
lucs :( Painful to read
literal what I pasted didn't exceed 80 chars per-line so I thought it was safe
lucs: you can read it in a nice format by following the url I pointed to 07:35
lucs Better :)
Nice project. 07:38
masak rakudo: say ?("5[aaaaa]" ~~ /(\d+) '[' a**$0 ']' /) 12:03
lambdabot masak: You have 1 new message. '/msg lambdabot @messages' to read it.
masak @massage
lambdabot literal said 4h 33m 24s ago: your comments are requested -> tinyurl.com/c8j8vw
p6eval rakudo 7af829: OUTPUT«perl6regex parse error: Error in closure quantifier at offset 34, found '$'␤current instr.: 'parrot;PGE;Perl6Regex;parse_error' pc 10179 (compilers/pge/PGE/Perl6Regex.pir:1219)␤»
masak literal: excellent. 12:05
masak literal: I'm ready to discuss design ideas for proto, anytime. 12:05
literal: typo: s/but I where the/but where the/ 12:06
literal: also count on me reviewing your commits once you start working on "psi". have you decided on an implementation language? :) 12:07
@tell literal irclog.perlgeek.de/perl6/2009-03-29#i_1023806 12:08
lambdabot Consider it noted.
masak pmichaud: ping 13:07
mberends masak: good $summer-time! 13:18
masak :)
and to you, sir.
shortest weekend of the year.
mberends: I've reviewed your changes in a local branch; will push them now. 13:19
(re proto)
there, pushed. 13:20
mberends great. just backlogged over literal's plans for psi: looks good too.
masak aye.
even if Perl 6 Christmas doesn't come this year, we'll be ready next year.
or at least, I should say, relatively ready. 13:21
mberends we'll be ready. I see you've translated my Fortran into Perl 6 ;) 13:23
masak :) I just like not repeating myself.
most of the commits were actually on string contents.
mberends laziness. a very good virtue. yes, I see that. 13:24
masak two further questions: can we move Configure.pm into a lib/ ? can we rename s/hello.bash/create-new-project/ ?
mberends yes, and yes. 13:25
masak great, I'll do that now.
I think the create-new-project rocks...
...but as soon as I finished reading it, I wanted more. :)
mberends is in a good mood, he'll agree to almost anything today 13:26
masak we should create a README stub for the developer.
and help her pick a LICENSE.
and stub an AUTHORS file.
mberends anything you say, boss 13:27
masak :P just drawing guidelines, but they're just pencilled in. 13:28
mberends is the Artistic License compatible with BSD ( I know it's OK with GNU ) ?
masak dunno.
licenses confuse me to some degree.
maybe we should provide the Death and Retribution (D&R) license as an alternative? 13:29
it's almost always dual-licensed, though.
s/Retribution/Repudiation/
cpansearch.perl.org/src/AUDREYT/YAM...05/COPYING 13:30
mberends now that's a license. to kill. 13:31
masak :D
mberends or to die for. (hey, that's [almost] Perl poetry) 13:32
masak .die for lines("LICENSE"); 13:32
masak really likes lines()
mberends noticed 13:33
masak can't wait to start to use it in the next developer release!
good thing they come once a month.
mberends the BSD camp has some strong anti-GNU arguments.
masak URL? 13:34
mberends and Linus won't move from gnu 2 to gnu3, afaik 13:34
masak so I heard.
I understand, and even approve of, his arguments.
mberends scratches head
masak ok, enough tinkering. I'll symbolically unlock proto again now. 13:38
mberends masak: www.freebsd.org/doc/en_US.ISO8859-1...ticle.html 13:39
masak mberends: looks like a good read. will get back to you about it in two weeks. 13:40
today I plan to work some on the Lobster.
been wanting to do that all week, but $WORK has been getting in the way. :/ 13:41
mberends cool, if lobster is good when cool.
masak I should think so.
mberends packs stuff for a week in Groningen 13:42
BinGOs aaaaaaaaaaaaaaaaaaa/win 17 14:33
oops
masak oh, and for anyone who's interested: my work with the Druid tests is going forward at a steady pace: github.com/masak/druid/blob/master/...me-rules.t 14:39
the Test::Ix framework has drawbacks, but I like it for this small purpose.
masak Ovid writes a good article about roles. use.perl.org/~Ovid/journal/38649 15:02
masak I hadn't quite realized how much I appreciate roles; but I realize I use them regularly, and that there is surprisingly little pain involved. 15:03
ihrd hi 15:26
ihrd rakudo: my @a = 1,3,2; say @a.=sort: {1}; 15:26
p6eval rakudo 7af829: OUTPUT«Statement not terminated properly at line 1, near ": {1};"␤␤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:83)␤»
ihrd my @a = 1,3,2; say @a.sort: {1}; 15:27
rakudo: my @a = 1,3,2; say @a.sort: {1};
p6eval rakudo 7af829: OUTPUT«132␤»
ihrd rakudo: my @a = 1,3,2; say @a.=sort;
masak ihrd: that's correct, if you ask me.
p6eval rakudo 7af829: OUTPUT«123␤»
ihrd masak, why? 15:28
masak ihrd: the {1} returns 1 all the time, meaning 'the right operand is larger'.
ihrd masak, see first example 15:29
wrong topic
ihrd why .=sort: { ... } do not work? 15:30
masak I do not know.
ihrd this is my question :)
masak ihrd: submit a rakudobug :)
ihrd I will, if nobody answer :) 15:31
masak rakudo: my @a = 1,2,3; @a .= sort: { $^a cmp $^b }
p6eval rakudo 7af829: OUTPUT«Statement not terminated properly at line 1, near ": { $^a cm"␤␤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:83)␤»
masak that is indeed strange. 15:32
ihrd std: my @a = 1,2,3; @a.=sort: {1}
p6eval std 26019: OUTPUT«ok 00:02 37m␤»
ihrd bug 15:33
masak yes. STD and Rakudo disagree. I say submit now.
ihrd uhu 15:34
mistake in your example, spaces around .= 15:36
masak ihrd: are you saying that's not allowed?
ihrd I think so
std: akudo: my @a = 1,2,3; @a .= sort: { $^a cmp $^b } 15:37
p6eval std 26019: OUTPUT«ok 00:02 37m␤»
masak :)
ihrd std: my @a = 1,2,3; @a .= sort: { $^a cmp $^b }
p6eval std 26019: OUTPUT«ok 00:02 37m␤»
masak likes spaces around operators
ihrd hm
sorry
masak no problem at all. 15:38
masak makes a sound like a lobster 15:41
masak std: my $a = ~~ 5; 16:06
p6eval std 26019: OUTPUT«ok 00:02 37m␤»
masak rakudo: class A { has &!a = { say "OH HAI" }; method foo { &!a() } }; A.new.foo 16:33
p6eval rakudo 7af829: OUTPUT«No such attribute '!a'␤current instr.: 'parrot;A;foo' pc 267 (EVAL_20:112)␤»
masak submits rakudobug
literal @tell masak Implementation language? I thought Perl 6 was assumed :) 17:11
lambdabot Consider it noted.
moritz_ re 17:13
pugs_svn r26020 | moritz++ | [t/spec] 17:20
r26020 | moritz++ | * caps.t: update to current spec; tests for .chunks
r26020 | moritz++ | * S05-mass/rx.t: change infix:<also> to && in some places
r26020 | moritz++ | * a complete pass through the rakudo skip's, improving tests
r26020 | moritz++ | and finer graded fudging
r26020 | moritz++ | Overall +221 passing tests
dalek kudo: 370dd76 | (Moritz Lenz)++ | docs/ChangeLog:
[docs] update to ChangeLog
 17:22
kudo: dde9758 | (Moritz Lenz)++ | docs/guide_to_setting.pod:
[docs] more guidelines to adding methods to the setting
kudo: ab367f6 | (Moritz Lenz)++ | README:
[README] information on how to obtain git SHA-1
kudo: 0d52048 | (Moritz Lenz)++ | README:
[README] spell "Perl 6" with non-breaking space
kudo: bc32a56 | (Moritz Lenz)++ | src/setting/Match.pm:
update Match.caps and Match.chunks to meet current spec
frioux does anyone here know how I can make a closure (like this: { .foo } ) with the first arg curried, but not the invocant? 18:19
moritz_ frioux: do you want the closure to be a method? 18:24
frioux I guess?
I am not really sure I understand 18:25
moritz_ it could also be a sub
frioux well, I think this is ultimately a work around
but I'll explain
if you do $foo ~~ .bar it works
but if you use this: $foo ~~ .bar & .baz it fails 18:26
frioux because rakudo is too eager 18:26
frioux so you have to do $f ~~ { .bar } & { .baz } 18:26
moritz_ rakudo: my $s = 0; my $c = sub($i, $l) { $i.substr($c, $s, $l) }; $s = 1; $c('abcdefg', 2);
p6eval rakudo 370dd7: OUTPUT«get_integer() not implemented in class 'Sub'␤current instr.: 'parrot;Any;substr' pc 12519 (src/builtins/any-str.pir:295)␤»
moritz_ rakudo: my $s = 0; my $c = sub($i, $l) { $i.substr($i, $s, $l) }; $s = 1; $c('abcdefg', 2); 18:27
p6eval rakudo 370dd7: OUTPUT«too many arguments passed (5) - at most 3 params expected␤current instr.: 'parrot;Any;substr' pc 12519 (src/builtins/any-str.pir:295)␤»
moritz_ I don't understand that output
frioux: I don't know if that's a rakudobug that you've found, or if it's a limitiation by the spec 18:28
frioux bacek said it was a bug I think
and that using { } was a workaround
00:44 < bacek> ok. This is bug.
00:45 < bacek> And this is workaround :)
szabgab rakudo: sub f($x){}; sub f($y, $z) {} 18:40
p6eval rakudo 370dd7: RESULT«{ ... }»
szabgab shouldn't rakuod complain about function redefinition ? 18:41
rakudo: sub f($x){}; sub f($y, $z) {}; f(2, 3)
moritz_ it should
p6eval rakudo 370dd7: OUTPUT«too many arguments passed (2) - 1 params expected␤current instr.: 'f' pc 89 (EVAL_16:52)␤»
moritz_ an iirc we have tests for that...
but a ticket couldn't hurt ;-)
szabgab oh you are worried I start to write a test :-)
moritz_ (although I wouldn't complain about making subs multis by default, and require an 'only sub f' to enforce just that) 18:42
no
I just wanted to say that it's not completely unknown to me
szabgab sure :-)
frioux rakudo: say #[ foo ] "test"; 18:55
p6eval rakudo 370dd7: OUTPUT«say requires an argument at line 1, near " #[ foo ] "␤␤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:83)␤»
moritz_ NYI
frioux but that's how it should work though, right/
moritz_ right.
frioux k 18:56
moritz_ added that to the status page as known todo 19:02
pugs_svn r26021 | moritz++ | [t/spec] test for RT #62730 19:13
pugs_svn r26022 | moritz++ | [t/spec] test for RT #62728 19:19
moritz_ rakudo: my Int $x = +'2' 19:22
p6eval rakudo 370dd7: OUTPUT«Type mismatch in assignment.␤current instr.: 'die' pc 15621 (src/builtins/control.pir:204)␤»
moritz_ rakudo: undef === undef 19:24
p6eval rakudo 370dd7: RESULT«Bool::False»
moritz_ don't know if that's correct.... undef doesn't seem to be immutable, so it's not a value type, so two instances of undef might not be === after all 19:29
rakudo: say ++undef
p6eval rakudo 370dd7: OUTPUT«1␤»
moritz_ std: class a { has $.attrib is build { ... } }; 19:34
p6eval std 26022: OUTPUT«##### PARSE FAILED #####␤Malformed class␤Unable to parse class definition␤Unexpected block in infix position (two terms in a row, or previous statement missing semicolon?) at /tmp/7DeXHpjlG9 line 1:␤------> class a { has $.attrib is build { ... } };␤ expecting
..an...
pugs_svn r26023 | moritz++ | [t/spec] remove tests that used 'is build' or 'will build' from A12 19:37
moritz_ rakudo: sub foo($.bar) { ... }; say "alive"; 19:38
p6eval rakudo 370dd7: OUTPUT«alive␤»
moritz_ rakudo: ~Bool::True 19:40
p6eval rakudo 370dd7: RESULT«"1"»
moritz_ rakudo: ~Bool::True.name
p6eval rakudo 370dd7: OUTPUT«Method 'name' not found for invocant of class 'Bool'␤current instr.: 'parrot;P6metaclass;dispatch' pc 637 (src/classes/ClassHOW.pir:161)␤»
moritz_ RT really needs a button that expires a ticket if there's no activity in a certain time frame 19:50
rakudo: token text { \w+ }; say "ab" ~~ /<text>/; 19:51
p6eval rakudo 370dd7: OUTPUT«get_attr_str() not implemented in class 'String'␤current instr.: 'parrot;PGE;Grammar;' pc 382 (EVAL_18:164)␤»
moritz_ rakudo: sub foo(:$w=4){say $w};foo('w'=>3);foo(w=>3); 19:53
p6eval rakudo 370dd7: OUTPUT«w 3␤3␤»
moritz_ for all German readers: in the current I'x magazine (April 2009, but already available) there are two Perl 6 articles 20:02
skids wonders what's up with commitbit.pugscode.org 22:20
moritz_ it needs slapping... 22:21
... and my password safe hangs while trying to open the file. great 22:22
moritz_ could somebody with root on feather1 please send a SIGHUP to the jifty process? 22:28
(or start if there's none running)
mikehh rakudo (370dd76) builds on parrot r37815 - make test PASS, make spectest FAIL - 1/339 test programs. 0/10054 subtests failed. 22:30
cspencer S32 indicates that the log of 0 should be an error, whereas the spec test suite expects -Inf - should the test suite be updated to reflect this?
moritz_ mikehh: oh, which one failed? 22:31
cspencer: -Inf looks fine to me
mikehh t/spec/S29-conversions/ord_and_chr.rakudo - Parse errors: Bad plan. You planned 444 tests but ran 300.
moritz_ OH NOEZ
mikehh I ran make spectest earlier on rakudo (7af829f) and all was ok 22:34
moritz_ that's a failure that comes and goes 22:35
I bet when you execute it three times, it passes at least once
skids: commitbit.pugscode.org:6666/ now works
mikehh let me try a make on it
moritz_ ah, and it alsow works without the explicit port
mikehh it starts off - 1..444 - goes through to 300 then exits with - Failed 144/444 subtests (less 96 skipped subtests: 204 okay) 22:40
moritz_ mikehh: could you please run the .rakudo file directly, without the harness? 22:41
diakopter ok 22:43
oh
mikehh morritz_: it gets to ok 204 - chr() works for \3 22:44
No applicable methods. 22:45
current instr.: '_block308' pc 2745 (EVAL_20:835)
called from Sub '_block14' pc 1711 (EVAL_20:508)
called from Sub '!UNIT_START' pc 17383 (src/builtins/guts.pir:358) 22:46
mikehh called from Sub 'parrot;PCT;HLLCompiler;eval' pc 950 (src/PCT/HLLCompiler.pir:527) 22:46
moritz_ bah
mikehh called from Sub 'parrot;PCT;HLLCompiler;evalfiles' pc 1275 (src/PCT/HLLCompiler.pir:688)
moritz_ I got some other occasional errors during 'make spectest', but they all went away when I ran the individual file 22:47
mikehh called from Sub 'parrot;PCT;HLLCompiler;command_line' pc 1470 (src/PCT/HLLCompiler.pir:789)
moritz_ it's easy to blame the crappy GC in that case, but I don't know what's wrong with your rakudo
mikehh called from Sub 'parrot;Perl6;Compiler;main' pc 21813 (perl6.pir:164)
I am running Ubuntu Intrepid Amd86 22:48
cspencer is MMD taking into account the type of the invocant when choosing the method to execute? 22:49
moritz_ cspencer: yes
cspencer hmmm...i'm attempting to implement the 'log' method for the Any class 22:50
but i can't get the Complex log method to work
moritz_ rakudo: say log(1+2i)
p6eval rakudo 370dd7: OUTPUT«0.804719+1.10715i␤»
cspencer moritz: i was doing it in the setting
moritz_ cspencer: what's the problem? is the Complex method never called? 22:51
test the method form first; I think there were some problems that exported method currently didn't declare the invocant type directly 22:52
cspencer the method form with work if i put in an invocant type of Complex into Complex.pm
cspencer without it, i get an Ambiguous candidates error 22:52
however, the sub version (ie. log(2i)) still calls the Any version 22:53
moritz_ I guess you can't get it to work with 'is export' atm 22:54
try to define a separate sub manually
cspencer ah, ok, that works :) 22:56
mikehh moritz_: the test passes on my previous build - make t/spec/S29-conversions/ord_and_chr.t, but fails withe the same result
cspencer if i submit a patch with the sub and a "TODO: remove this sub eventually" note, that'll be okay? 22:57
moritz_ cspencer: yes
mikehh with ./perl6 t/spec/S29-conversions/ord_and_chr.t
moritz_ mikehh: thank you
mikehh that is at the first skip 22:58
it skips 205 through 408 passes 409 through 422 and skips the last 2 23:03
in my latest build it exits after 300
moritz_ sounds like a job for pmichaud :/ 23:05
mikehh the previous build was 6-7 hours ago (that passed) - let me try a rebuild 23:06
mikehh make t/spec/S29-conversions/ord_and_chr.t - Failed 144/444 subtests (less 96 skipped subtests: 204 okay) 23:13
mikehh the earlier build was at 7af829f on parrot r37813 at Sun Mar 29 17:19:32 BST 2009 23:16
this build (first time) was at 370dd76 on parroi r37815 at Sun Mar 29 21:35:26 BST 2009 23:19
moritz_ literal: nice gsoc proposal. I'd also be available as a mentor or backup mentor 23:25
literal great
cspencer rakudo: my $pat = "b"; say "abcefg" ~~ /$pat/
p6eval rakudo 370dd7: OUTPUT«Null PMC access in get_string()␤current instr.: 'parrot;PGE;Grammar;' pc 249 (EVAL_17:106)␤»
cspencer do regex interpolate variables into the patterns yet?
moritz_ no :(
cspencer doh
is there a workaround? 23:26
moritz_ eval
cspencer eval "rx/$pat/"?
rakudo: my $pat = "b"; my $rx = eval("rx/$pat/"); say "abcefg" ~~ $rx
p6eval rakudo 370dd7: OUTPUT«b␤»
cspencer ah
moritz_ rakudo: my $x = '\w+'; my $rx = eval "regex { $x }"; say "foo" ~~ $rx
p6eval rakudo 370dd7: OUTPUT«0␤» 23:27
moritz_ you were faster, and yours worked ;-)
rakudo: my $x = '\w+'; my $rx = eval "rx { $x }"; say "foo" ~~ $rx
p6eval rakudo 370dd7: OUTPUT«0␤» 23:27
moritz_ rakudo: my $x = '\w+'; my $rx = eval "rx { $x }"; say $!; say "foo" ~~ $rx
p6eval rakudo 370dd7: OUTPUT«Syntax error at line 1, near "rx \\w+"␤␤0␤»
moritz_ rakudo: my $x = '\w+'; my $rx = eval "rx/ $x /"; say $!; say "foo" ~~ $rx
p6eval rakudo 370dd7: OUTPUT«Use of uninitialized value␤␤foo␤»
mikehh moritz_: the only possibility I can see is in src/setting/Match.pm | 23 +++++++++++++++-------- 23:28
mikehh all the other changes were in docs 23:29
moritz_ mikehh: that commit only modified .caps and .chunks, both of which aren't used in that test (nor anywhere else, for that matter)
if that really caused trouble, it's somewhere much deeper
mikehh as far as I can see changes in parrot r37814 and 27815 were pod or doc changes 23:32
37815 23:33
skids thanks moritz++ for smacking commitbit back into shape. 23:36
moritz_ skids: you're welcome
moritz_ goes to bed now
mikehh moritz_: night 23:37
mikehh if I run ./perl6 t/spec/S29-conversions/ord_and_chr.rakudo I get a Segmentation fault 23:47
after ok 300 - # SKIP named args 23:48

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!