»ö« | perl6-projects.org/ | nopaste: paste.lisp.org/new/perl6 | evalbot usage: 'perl6: say 3;' | irclog: irc.pugscode.org/ | UTF-8 is our friend!
Set by moritz_ on 30 July 2009.
00:05 SmokeMachine left 00:07 kubrat joined 00:12 kubrat left 00:17 ruz_mac left 00:23 justatheory left 00:39 jferrero left 00:49 kane_ left, kane_ joined
meppl good night 00:54
00:55 xomas_ left 00:56 meppl left 01:10 jferrero joined 01:31 kane_ left 01:33 M_o_C left 02:17 jferrero left 02:33 sri_kraih_ left 02:40 PZt left 02:47 Whiteknight left 02:48 lumi_ joined 02:53 lumi left 03:00 jauaor joined 03:16 vigith joined 03:23 tann joined 03:25 lumi joined 03:29 justatheory joined 03:39 lumi_ left 03:48 jauaor left 04:00 JimmyZ joined 04:04 JimmyZ_ joined 04:05 JimmyZ_ left 04:08 flexibeast joined 04:11 justatheory left 04:12 lumi_ joined 04:13 lumi left 04:15 alester joined 05:13 Guest32604 left 05:35 lumi joined 05:37 molaf joined, lumi_ left 05:40 orafu left, orafu joined 05:41 JimmyZ left 05:45 JimmyZ joined 05:46 PZt joined 06:01 missingthepoint joined
JimmyZ rakudo: $_ = Num; $_('h22h').say; 06:02
p6eval rakudo e02bc0: OUTPUT«invoke() not implemented in class 'Float'␤in Main (/tmp/nPcB1YsEgT:2)␤» 06:03
JimmyZ rakudo: $_ = Int; $_('h22h').say;
p6eval rakudo e02bc0: OUTPUT«invoke() not implemented in class 'Integer'␤in Main (/tmp/sj7wvqoC5t:2)␤»
JimmyZ rakudo: nt('h22h').say;
rakudo: Int('h22h').say;
p6eval rakudo e02bc0: OUTPUT«Could not find non-existent sub nt␤»
rakudo e02bc0: OUTPUT«invoke() not implemented in class 'Integer'␤in Main (/tmp/KT3SBPtHKV:2)␤»
JimmyZ rakudo: ('2,',d','/3','sdf').elems.say 06:13
p6eval rakudo e02bc0: OUTPUT«Statement not terminated properly at line 2, near "','/3','sd"␤in Main (src/gen_setting.pm:3363)␤»
JimmyZ rakudo: ('2,',',d','/3','sdf').elems.say 06:14
p6eval rakudo e02bc0: OUTPUT«4␤»
missingthepoint JimmyZ: you can talk to the evalbot directly ;) 06:21
06:37 JimmyZ left 06:39 alester left 06:45 iblechbot left 06:46 ruz_mac joined 06:48 masak joined
masak rakudo: { "Morning, $^someone!" }.("perl6ers").say 06:49
p6eval rakudo e02bc0: OUTPUT«Morning, perl6ers!␤»
06:51 zamolxes left
missingthepoint masak: care to exposit that line for me? :) 06:52
masak: good morning!
masak missingthepoint: morning. yes, sure.
missingthepoint: $^someone is an implicit parameter. explained in S06, I think. 06:53
it gets bound to the argument "perl6ers" I send in.
so the whole block returns the string you see printed.
missingthepoint i'm confused by the method call syntax 06:54
ah, the block's an object 06:55
and it returns and object on which you call say
masak yes. a string.
missingthepoint because everything's an object! :)
masak sure, whatever :) 06:56
rakudo: { &^c("$^a, $^b") }.("morning", "perl6ers", &say)
p6eval rakudo e02bc0: OUTPUT«Parameter type check failed; expected Callable, but got Code for c in call to _block50␤in Main (/tmp/lu21Y4RHO7:2)␤»
masak submits rakudobug
missingthepoint masak: new bug need for this? 06:57
rakudo: say [**] 4,3,2
p6eval rakudo e02bc0: OUTPUT«4096␤»
masak rakudo: { &^foo("OH HAI") }.(&say) 06:58
p6eval rakudo e02bc0: OUTPUT«Parameter type check failed; expected Callable, but got Code for foo in call to _block50␤in Main (/tmp/DSpLRRPsT2:2)␤»
masak rakudo: say 4**(3**2)
p6eval rakudo e02bc0: OUTPUT«262144␤»
masak rakudo: say (4**3)**2
p6eval rakudo e02bc0: OUTPUT«4096␤»
masak missingthepoint: when in doubt, submit. :) go right ahead. 07:00
missingthepoint JimmyZ found it yesterday 07:01
masak missingthepoint: care to report it?
missingthepoint yes
masak rakudo: say [-]() 07:03
p6eval rakudo e02bc0: OUTPUT«0␤»
07:07 Su-Shee joined
Su-Shee good morning. :) 07:07
lambdabot Su-Shee: You have 1 new message. '/msg lambdabot @messages' to read it.
masak I was going to add .warn as a method of Any in the spec, but it looks like it's already a method of Object, at least according to S32/Basics.
Su-Shee: morning!
missingthepoint morning Su-Shee :) 07:08
masak rakudo: { &^foo }.(&say) 07:09
p6eval rakudo e02bc0: OUTPUT«Parameter type check failed; expected Callable, but got Code for foo in call to _block50␤in Main (/tmp/ItwenHGieA:2)␤»
masak rakudo: -> &foo {}.(&say) 07:10
p6eval rakudo e02bc0: OUTPUT«Parameter type check failed; expected Callable, but got Code for foo in call to _block50␤in Main (/tmp/tbgIXOFn8F:2)␤»
masak rakudo: sub (&foo) {}.(&say)
p6eval rakudo e02bc0: OUTPUT«Statement not terminated properly at line 2, near ".(&say)"␤in Main (src/gen_setting.pm:3363)␤»
masak rakudo: (sub (&foo) {}).(&say)
p6eval rakudo e02bc0: OUTPUT«Parameter type check failed; expected Callable, but got Code for foo in call to _block50␤in sub _block50 (/tmp/5TxfI8OcpU:2)␤called from Main (/tmp/5TxfI8OcpU:2)␤»
missingthepoint gah typos... RT #68114 07:12
JimmyZ++
Su-Shee: your last blog post was excellent. :) 07:13
Su-Shee the pretty perl stuff?
missingthepoint yah
japhb link? 07:14
Su-Shee it's unbelievable what one can do in a few hours if you're not have to poke around with xlib and c.
japhb: sushee.schreibsturm.org
missingthepoint Su-Shee: that is why i switched to perl! :)
japhb Su-Shee: thx 07:15
Su-Shee missingthepoint: besides that "gui" is my favorite subject technically, that's what I have in mind for perl6/parrot: a really nice suite of "gui stuff" to prototype windowmanagers and sceenlets and whateverlets and gui apps.
masak for people from the future, here's the absolute URL: sushee.schreibsturm.org/2009/07/pretty-perl.html
missingthepoint masak++ for remembering this is more than just a chatroom 07:16
Su-Shee I really can't stress OLPC's "sugar" as a great example.
masak Su-Shee++ # blog post
Su-Shee (and let's not forget that we have inotify, dbus, usb, foo and bar and whatever perl bindings to work with..) 07:17
masak rakudo: class A { method poscircumfix:<( )>() { say "OH HAI" } }; A.new().() 07:18
p6eval rakudo e02bc0: OUTPUT«invoke() not implemented in class 'A'␤in Main (/tmp/tqdQGJ3fbR:2)␤»
masak how do I make a custom Callable class? 07:19
missingthepoint Su-Shee: i would love that... I've been dreaming of an "emacs-but-in-perl6"... 07:23
Xemacs, even
Su-Shee missingthepoint: wouldn't that be an editor engine with perl6 scripting and free form keyboard shortcuts to be customized as either vi, emacs or win-style? 07:24
missingthepoint: editorcode-wise it's probably totally meaningless wether I call dd or something else to delete a line. 07:25
missingthepoint Su-Shee: well, my specific dreams are: 1) perl6 macros eqv to elisp, 2) everything gui done in perl6 to allow custom *graphical* extensions to the editor
Su-Shee (make it a parrot editor engine, why stop at perl6...)
missingthepoint: let's face it, we all got perl6 projects up until ... 2021 ;) 07:26
07:26 JimmyZ joined
missingthepoint Su-Shee: true, true 07:26
masak (which is kinda ironic, considering that the world is coming to an end in 2012.)
JimmyZ afternoon, #perl6 07:27
missingthepoint hi JimmyZ! :)
masak JimmyZ: hi
Su-Shee masak: it does? I thought the next coming to an end is 2038? ;)
JimmyZ hello missingthepoint and masak
masak Su-Shee: oh sure, that one too.
Su-Shee: we have like a schedule of apocalypses laid out. 07:28
Su-Shee ah, and btw on the subject of perl6 bindings and all that: www.programmableweb.com/ is the most important source for all "web api" things.
JimmyZ hello masak, do you have a rakudo commit?
masak JimmyZ: a commitbit? yes. 07:29
missingthepoint JimmyZ: you have a bugfix? :)
missingthepoint takes a look at assign.pir 07:31
JimmyZ no, just move grammer from STD.pm to rakudo, that makes some new spec partial working which still needs to be extended 07:32
masak JimmyZ: you could nopaste it, and I could have a look. 07:33
JimmyZ: have you run the spectests on it?
JimmyZ yes, and I added new spectests to pugs repo. 07:34
masak wow, cool.
so, what's the patch?
JimmyZ wait... 07:36
the problem is that, I don't know how to create a patch on windows. 07:37
missingthepoint masak: an apocalypse is just a revealing... :) 07:39
masak missingthepoint: yes, but the thing being revealed might be the end of the world... as we know it. 07:42
JimmyZ: ok. neither do I, unfortunately. 07:43
lisppaste3 jimmy pasted "patch" at paste.lisp.org/display/84610 07:47
missingthepoint masak: i was going to do Text::CSV first, but after seeing the contortions it goes through to deal with a broken format, Scruffy might be an easier first port
JimmyZ masak: that's the patch
masak JimmyZ: thanks. 07:48
missingthepoint: as mentioned, I'm willing to support you in any way I can with the Scruffy port.
JimmyZ I think <postfix_prefix_meta_operator>? is no need.
masak I believe it would be a very good thing to have.
JimmyZ but I can't have a try.
missingthepoint masak: thank you, i appreciate it a lot. :) 07:49
masak missingthepoint: let me know when you have a github repo (or equiv). 07:50
JimmyZ: ok, I see changes to three files.
the first change looks like a non-destructive simplification. 07:51
the second change removes an unused local variable.
both of these changes are fine, I guess.
the third change is the interesting one, I guess. 07:52
lisppaste3 jimmy pasted "another patch" at paste.lisp.org/display/84611
masak ok, I'll need to look this up in STD.pm. 07:53
JimmyZ thanks.
parrot still can't 'make' with chinse TEMP PATH. 07:54
should say chinese character 07:55
masak so a postfix_prefix_meta_operator would be the '>>' of '@obj>>.meth()', for example?
JimmyZ yes
add now support @obj.>>.meth()
masak I'm running spectests now.
I'll apply your patch, and then run them again. 07:56
JimmyZ masak++
masak if it looks good to me, I'll commit it.
JimmyZ masak: this patch just makes S03-operators/hyper.t more happy. 08:01
masak JimmyZ: so there's an opportunity for unfudging after applying your patch?
masak reads S03-operators/hyper.t 08:02
hm, the tests on line 137 and the rest in that block look wrong to me. 08:05
they say »+«, but I think they should say »+»
um, and some of them should say «+« 08:06
masak fixes
JimmyZ maska: perlcabal.org/svn/pugs/revision/?rev=27830 08:08
masak : that's the tests. perlcabal.org/svn/pugs/revision/?rev=27830
masak JimmyZ: ok. those look good to me. 08:10
running spectests on your changes now.
rakudo: sub sub($foo) { say "OH HAI" }; sub(1) 08:13
p6eval rakudo e02bc0: OUTPUT«OH HAI␤»
masak closes #65804 08:14
see? I don't just open tickets. :)
JimmyZ ;) 08:15
masak rakudo: role A {}; say A ~~ Role 08:18
p6eval rakudo e02bc0: OUTPUT«1␤»
masak rakudo: my %h = B => 2; my $x = "ABC"; $x .= subst(/(<[BD]>)/, {%h{$0}},:global); say $x; 08:22
p6eval rakudo e02bc0: OUTPUT«A2C␤»
masak rakudo: my %h = B => 2; my $x = "ABC"; $x.=subst(/(<[BD]>)/, {%h{$0}},:global); say $x;
p6eval rakudo e02bc0: OUTPUT«Use of uninitialized value␤Use of uninitialized value␤AC␤»
masak rakudo: say (42 if 0).perl 08:25
p6eval rakudo e02bc0: OUTPUT«[]␤»
masak rakudo: say (42 if 0).WHAT
p6eval rakudo e02bc0: OUTPUT«List()␤»
08:39 mberends joined
tann rakudo: class A { }; say A.WHAT; my $o = A.new; say $o ~~ A; 08:44
p6eval rakudo e02bc0: OUTPUT«A()␤1␤»
pugs_svn r27832 | masak++ | [t/spec] fixed some hyper-op typos 08:48
08:49 unitxt left
dalek kudo: f76b3c7 | masak++ | src/builtins/ (2 files):
[src/builtins] some PIR simplifications
 08:49
kudo: e89cc1e | masak++ | src/parser/grammar.pg:
[src/parser/grammar.pg] aligned more with STD.pm

  JimmyZ++ for the patch.
masak JimmyZ: applied. so I can close RT #68098 now, right?
general question: let's say I have a base class called 'rw', from which I want other classes to derive. how do I write a class that derives from 'rw'? 08:53
rakudo: undef ~~ /foo/ 08:55
p6eval rakudo e89cc1: OUTPUT«Object must be created by a class.␤in regex PGE::Grammar::_block51 (/tmp/xwkCY66Yz0:1)␤called from Main (/tmp/xwkCY66Yz0:2)␤»
masak rakudo: class class { method foo { say "OH HAI" } } 08:59
p6eval rakudo e89cc1: ( no output )
masak rakudo: class class { method foo { say "OH HAI" } }; my class $a; $a.foo 09:00
p6eval rakudo e89cc1: OUTPUT«OH HAI␤»
masak \o/
09:05 M_o_C joined
missingthepoint rakudo: class rw { method foo { say 'rw foo'; } }; class rw_and_then_some is rw { method bar { say 'rw+bar' } }; my $hmm = rw_and_then_some.new; $hmm.bar 09:13
p6eval rakudo e89cc1: OUTPUT«rw+bar␤»
missingthepoint rakudo: class rw { method foo { say 'rw foo'; } }; class rw_and_then_some is rw { method bar { say 'rw+bar' } }; my $hmm = rw_and_then_some.new; $hmm.bar; $hmm.foo;
p6eval rakudo e89cc1: OUTPUT«rw+bar␤Method 'foo' not found for invocant of class 'rw_and_then_some'␤»
09:13 tann__ joined, tann__ left
missingthepoint ah 09:14
masak see? :)
missingthepoint :)
ambiguity fail? 09:15
masak missingthepoint: I have mberends sitting in a sofa next to me. he aptly invoked the 'doctor, it hurts when I do this' rule.
but still. :)
missingthepoint i am inclined to agree... but still :)
is :rw ? 09:16
(as a newbie, that would seem a nice fit)
masak a constructive idea to fix the problem! I like that.
yes, to me that feels more ortho.
09:23 jedai joined 09:24 jedai left 09:30 tann left
missingthepoint "This application has requested the Runtime to terminate it in an unusual way." 09:45
rakudo: class A is KitchenSink { method b { say ~ .^parents }}; my $a = A.new; $a.b() 09:49
p6eval rakudo e89cc1: OUTPUT«No applicable candidates found to dispatch to for 'trait_mod:is'␤in Main (/tmp/mmJ7tdD8s1:2)␤» 09:50
masak std: class KitchenSink {}
p6eval std 27832: OUTPUT«ok 00:02 36m␤»
masak oh, is KitchenSink gone from the set of core types? 09:51
nope, still in CORE.setting
missingthepoint masak: wanna see something disgusting? 09:52
(codewise)
masak missingthepoint: is it Perl 6?
then, sure! 09:53
if it's COBOL or PHP, then perhaps not... :)
lisppaste3 missingthepoint pasted "untitled" at paste.lisp.org/display/84612
masak interesting, but what does it do? 09:54
it creates a new class at runtime...
and then instantiates it at runtime.
missingthepoint mainly, kills rakudo with the error message above
masak oh! you subclass everything in the whole language. 09:55
I see now.
rakudo: KitchenSink 09:56
p6eval rakudo e89cc1: OUTPUT«Could not find non-existent sub KitchenSink␤»
masak missingthepoint: well, KitchenSink isn't implemented in Rakudo.
the 'trait_mod:is' error is a less-than-awesome way to express that the base class doesn't exist. 09:57
missingthepoint true 09:58
M_o_C does "isn't implemented" implies that the spec includes a class named "KitchenSink"? o.O 09:59
s/implies/imply
jnthn masak: I already showed here on channel how to improve that error. :-P
dobre rano, by the way.
masak jnthn: boa dia!
M_o_C: spec no (AFAIK), STD.pm yes. 10:00
missingthepoint S06 mentions "KitchenSink role..."
masak missingthepoint: oh! 10:01
and here I thought all along it was just a private TimToady joke.
jnthn So Peerl 6 really doens't have everything but the kitchen sink. 10:02
masak it's in good company with Emacs and Mozilla Firefox. 10:03
jnthn If we embed Perl 6 in Emacs and Firefox, does the universe implode? 10:04
masak no, but the kitchen sink might overflow.
jnthn OH NOES the potato peals will go everywhere! 10:05
masak: Are you at the Perl 6 tut? 10:06
masak jnthn: no, I'm doing presentation procrastination in a sofa at our hostel, next to mberends. 10:07
jnthn: pmichaud said he'd cover the tut today.
jnthn Nice
OK
I may wander along to it later.
jnthn procrastinated his presentation a bit more last night
I found a hard bug in Rakudo along the way. :-( 10:08
masak which one? :)
jnthn .wrap doesn't work in a trait_mod
masak oh :/ 10:09
jnthn Yeah, which I am using in an example.
So need to work out why and fix it.
:/
masak jnthn++ 10:10
jnthn Suspect horrible lexicals issue. 10:11
10:11 greyfox joined
masak ouch. 10:13
greyfox hi 10:15
masak greyfox: oh hai
greyfox How're you ? 10:16
masak greyfox: very good. you?
greyfox fine.
Am learning file manipulation in perl, now 10:17
masak greyfox: aha. Perl 5?
greyfox v5.8.8
yes.
How do I upgrade it ?
masak to 5.10?
greyfox to perl 6
masak usually, you want to keep both perl 5 and perl 6 on your system. 10:18
greyfox Yes.
masak because they are good at different things.
greyfox brb
masak as for installing Perl 6, you can go here rakudo.org/how-to-get-rakudo and follow the instructions. 10:19
or you can go here github.com/masak/proto and follow the instructions. 10:20
here's a nice Ruby blog post about code reading: blog.rubybestpractices.com/posts/gr...tdlib.html -- I hope we can inspire people in the Perl 6 community to do similar code reads and then blog about them. 10:22
JimmyZ masak: Is there any problem with my patch? 10:25
jnthn pmichaud: Planning to walk a little bit first today, and then try and come and join the 6tut. 10:26
JimmyZ masak: RT #68098 should be closed, that'a bad patch.
jnthn ...if I can find the info on where it is. 10:27
masak JimmyZ: no problem with your patch AFAIK.
JimmyZ: ok, closing RT #68098.
JimmyZ masak++
masak jnthn: hold on, I have instructions.
jnthn: yapceurope2009.org/ye2009/wiki?node...ueLocation yapceurope2009.org/ye2009/wiki?node...ToTheVenue 10:29
argh, I should put them in the reverse order. 10:30
jnthn: anyway, the second URL contains instructions to get you to C3. the second one shows where C8 (the place for the class) is.
s:2nd/second/first/ 10:31
JimmyZ rakudo: (1, 4, 9).>>.++.perl.say; 10:41
p6eval rakudo e89cc1: OUTPUT«Statement not terminated properly at line 2, near ".>>.++.per"␤in Main (src/gen_setting.pm:3363)␤»
JimmyZ rakudo: (1, 4, 9).>>.++.succ.say;
p6eval rakudo e89cc1: OUTPUT«Statement not terminated properly at line 2, near ".>>.++.suc"␤in Main (src/gen_setting.pm:3363)␤»
JimmyZ rakudo: (1, 4, 9).>>.succ.say;
p6eval rakudo e89cc1: OUTPUT«2510␤»
10:41 meppl joined
JimmyZ rakudo: (1, 4, 9).>>.++.say; 10:41
p6eval rakudo e89cc1: OUTPUT«Statement not terminated properly at line 2, near ".>>.++.say"␤in Main (src/gen_setting.pm:3363)␤»
JimmyZ rakudo: (1, 4, 9).>>.pred.perl.say; 10:42
p6eval rakudo e89cc1: OUTPUT«[0, 3, 8]␤»
JimmyZ rakudo: (1, 4, 9)>>pred.perl.say;
p6eval rakudo e89cc1: OUTPUT«Statement not terminated properly at line 2, near ">>pred.per"␤in Main (src/gen_setting.pm:3363)␤»
JimmyZ rakudo: (1, 4, 9)>>++.perl.say;
p6eval rakudo e89cc1: OUTPUT«Statement not terminated properly at line 2, near ">>++.perl."␤in Main (src/gen_setting.pm:3363)␤»
JimmyZ rakudo: (1, 4, 9)>>.pred.perl.say; 10:44
p6eval rakudo e89cc1: OUTPUT«[0, 3, 8]␤»
10:51 Su-Shee left 10:56 Front_slash joined 11:06 clintongormley joined
JimmyZ perl6, another more writting only language than perl5 ;) 11:10
masak bah, consistency and common sense go a long way. I don't believe it is the duty of a language to restrict thought until only acceptable code can be produced. 11:12
mberends line noise has never been more beautiful 11:14
masak :)
jnthn "If you can't think about bad code, you can't write bad code." ;-) 11:34
masak Sapir-Whorf strikes again! 11:36
jnthn "...to the extent which language limits thought"
masak jnthn: you're beginning to sound like my insurance company. :) 11:37
jnthn :-P
masak: Coming to the pre-reg this evening? 11:38
masak jnthn: dunno. I might.
jnthn plans to
masak haven't really made plans for the day. this feels like the first unscheduled day of my vacation. 11:39
jnthn :-) 11:41
OK, see you there if you're there 11:42
jnthn goes for a wander and lunch
masak o/
11:47 mtp2 joined 11:49 missingthepoint left 11:50 mtp2 left
JimmyZ rakudo: my $a[] = 1; $a[] = 2; $a.perl.say; 11:56
p6eval rakudo e89cc1: OUTPUT«Statement not terminated properly at line 2, near "[] = 1; $a"␤in Main (src/gen_setting.pm:3363)␤»
JimmyZ rakudo: my $a[] = 1,2; $a.perl.say;
p6eval rakudo e89cc1: OUTPUT«Statement not terminated properly at line 2, near "[] = 1,2; "␤in Main (src/gen_setting.pm:3363)␤» 11:57
masak JimmyZ: looks like you're trying to do two things in the first statement.
(1) a 'my' declaration, (2) a (zen) slice operation
rakudo: my $a = []; $a[] = 1; say $a.perl 11:58
p6eval rakudo e89cc1: OUTPUT«[]␤»
masak JimmyZ: there you go :)
(next question is perhaps, why would you want to assign values to an empty slice?)
JimmyZ nothing ouputs
masak JimmyZ: correct. because you assign to an empty slice. 11:59
also, consider using an array variable @a instead.
more informative.
JimmyZ I just wonder if support '$foo.bar.baz.bletch.whatever.attr[] = 1,2,3;'
11:59 Whiteknight joined
masak I think so, but you shouldn't need the []. 12:00
but you make a good point: sometimes .[] means 'the whole array'.
maybe it should mean that in the case of assignment also.
JimmyZ and I wonder if perl6 supports assign to array likes php 12:01
;)
masak JimmyZ: where it means .push? I hope not. 12:03
JimmyZ yes
masak: I prefer $a[] = 'something' 12:04
masak JimmyZ: you can always write a module that changes Perl 6 to behave like that.
PHP::ArrayPush
JimmyZ masak: yep, but that needs to do more work 12:05
masak more work than what? convincing people in here to make $a[] = 'something' mean $a.push('something')? I doubt that.
JimmyZ masak: no, just my personal like, and I give it a try to perl6 12:07
12:08 jauaor joined
masak JimmyZ: fair enough. 12:09
JimmyZ: I don't care much for that syntax/semantics, and I don't see why it should mean .push rather than something else.
Juerd $a[] for push is nice in general, but it simply clashes with the notion of $a[] already being the whole array :)
masak Juerd: why is it nice in general? 12:10
Juerd masak: For no academic reason, but it's easy to get used to, and not ugly.
masak fair enough. 12:11
Juerd Also, shorter :)
But that's the golfer part of my brain that says that
masak I'm not condemning the syntax, just not very excited by it.
Juerd I've never been excited by it :)
But then, I'm not so excited about .[] in p6 either. 12:12
masak it makes sense in interpolated strings.
JimmyZ $a[] = ( 1, 2, 3); $a[]= 4; # now $a = (1,2,3,4)
Juerd It works, but it doesn't really makes sense to me.
masak JimmyZ: yah, I get it.
Juerd JimmyZ: That'll be $a.push: 4;
JimmyZ Juerd: as a phper, I prefer the former 12:14
Juerd Noted 12:15
masak de gustibus non disputandum est.
Juerd masak: Awwh, but we can try? :)
masak Juerd: :)
Juerd: by the way, it might please you to know that I'm currently doing slides for a talk for the grant project that is the direct result of the emails you wrote in 2006 to p6-u. 12:17
Juerd masak: It does :)
masak the project itself has taken me on strange paths, and affected me as a developer. :) 12:18
Juerd I'm really glad that I did write those. Due to personal circumstances I haven't been able to be involved with Perl 6 anymore (it'll take a while before I'll get back at the old level, if ever) and it is very comforting to see this being built 12:19
masak Juerd: are you coming to YAPC::EU?
Juerd Unfortunately not
masak ok.
anyway, it would be fun to sit down with you sometime and explain the current architecture of the project. 12:20
Juerd I'm involved with organizing HAR2009, for which buildup starts in a few days. A single congress can exhaust me at the moment, so I couldn't go to both.
masak understandable.
Juerd Also it'd feel weird to go to a Perl congress without speaking there ;) 12:22
masak judging from the list of participants, not everyone thinks like that. 12:23
jauaor morning
masak ...either that, or there are a lot of people who will feel weird at the conference. :)
jauaor: morning!
jauaor hi masak ! 12:24
:D
masak jauaor: you sound like you're having a fine morning? is it sunny where you are? 12:26
jauaor masak: haha yeah, sunny 12:27
birds singing and all
haha
:P
Juerd I've given a full or lightning talk at every Perl congress/workshop I've been to, except one German Perl Workshop.
masak Juerd++ 12:28
Juerd But I wouldn't know what to speak about now. Haven't used Perl since last YAPC I think
Well, except for oneliners and throwaway scripts
masak that's still more than 99% of humanity. 12:30
12:32 mberends left
Juerd Heh, sure, but it doesn't produce inspiration for talks :) 12:35
masak time for lunch. :) 12:38
12:38 masak left 13:21 sri_kraih joined, arthur-_ left 13:27 M_o_C left 13:32 Front_slash left 14:27 greyfox is now known as greyfox` 14:32 greyfox` is now known as greyfox 14:34 kimtaro_ joined 14:44 fridim_ joined 14:46 kimtaro__ joined, kimtaro_ left, arthur-_ joined, vigith left 14:47 zamolxes joined 14:51 kimtaro left 14:52 skids_ joined 14:54 Psyche^ joined 14:58 greyfox left 15:01 zamolxes left 15:03 unitxt joined 15:09 lumi_ joined, lumi left 15:10 Patterner left, Psyche^ is now known as Patterner 15:33 kimtaro joined 15:35 Front_slash joined 15:44 JimmyZ_ joined 15:46 zamolxes joined 15:50 kimtaro left, kimtaro__ left 15:59 JimmyZ left 16:00 clintongormley left, icwiener joined 16:02 mberends joined 16:03 masak joined 16:13 justatheory joined 16:16 Whiteknight left 16:21 JimmyZ_ left 16:41 Whiteknight joined, mberends left 16:52 zamolxes left
moritz_ re 17:08
masak moritz_: o/
moritz_ masak: how's YAPC, how's life?
masak moritz_: YAPC: starts tomorrow, life: good. :)
I was to szabgab's course yesterday. 17:09
today jnthn++ and pmichaud++ are covering it, AFAIK.
it's a bit hard to tell, because they are in 'Net shadow over there.
moritz_: I just ran smartlinks.pl on Web.pm! \o/ 17:10
moritz_ git-svnsvn rebase's
pugs_svn r27833 | moritz++ | [t/spec] avoid warnings in unspecified.t 17:11
r27834 | moritz++ | [t/spec] bring unspecified.t closer to my understanding of the spec
r27835 | moritz++ | [t/spec] remove want() from assign.t (and TODO)
r27836 | moritz++ | [t/spec] remove outdated reference to want.t 17:12
r27837 | moritz++ | [t/spec] unbreak two smartlinks, and break one differently than before
r27838 | moritz++ | [t/spec] more tests and a smartlink for pointy-rw.t
r27839 | moritz++ | [t/spec] merge mis-placed mutating_listops.t into S04-statements/for.t
r27840 | moritz++ | [t/spec] smartlinks and a simplification
masak après nous la deluge.
pugs_svn r27841 | moritz++ | [t/spec] smartlink and done_testing for exhaustive.t (the plan was incorrect anyway)
r27842 | moritz++ | [t/spec] more smartlinks; simplified subrule.t a bit
r27843 | moritz++ | [t/spec] interpolation of hashes into regexes is gone; smartlinks 17:13
r27844 | moritz++ | [t/spec] more test simplifications, more smartlinks
masak done_testing++
moritz_ there's one thing I don't like about done_testing
pugs_svn r27845 | moritz++ | [t/spec] move smartlinks from POD to ordinary comments
r27845 | moritz++ |
r27845 | moritz++ | our current implementation does not find them in POD sections.
r27845 | moritz++ | Also remove some redundant POD blocks.
r27846 | moritz++ | [t/spec] more smartlinks; less than 100 remain without smartlinks
r27847 | moritz++ | [t/spec] when our closure parameter requires and Int, we should give it one
moritz_ it's hard to find out how many tests are in there, which we need to generate our pretty charts
pugs_svn r27848 | moritz++ | [t/spec] update and simplify S02-names_and_variables/syntax.t; don't assume 17:14
r27848 | moritz++ | that infix:<:=> is a method.
r27848 | moritz++ | [t] remove syntax/signature.t, it relies heavily on unspecced (and unlikely)
r27848 | moritz++ | behaviour
r27849 | moritz++ | [t/spec] test multi dispatch over the signature of closure parameters
r27850 | moritz++ | [t/spec] Tests for RT 68074: class composition time multi dispatch
r27851 | moritz++ | [t/spec] basic tests for the "defines" declarator
r27852 | moritz++ | [t/spec] basic tests for "need" declarator
masak moritz_: one can always run the test file. :)
moritz_ masak: ...if one has an implementation to run it on, yes 17:15
masak oh, right.
we should write one, so that we can count our tests. 17:16
moritz_ good idea 17:22
I've set up a repo on github already
oh wait, it was pmichaud++ ;-)
masak :P 17:23
17:34 KyleHa joined
moritz_ it seems S03-operators/hyper.t was unfudged over-eagerly 17:34
masak KyleHa++: oh hai
moritz_: oh, so that's what it was. I saw the failures when spectesting. 17:35
33-37, right?
KyleHa Hello #perl6!
Some tests were added to hyper.t that should have been fudged but weren't.
moritz_ oh, ok
KyleHa Here's the change: dev.pugscode.org/changeset/27830
When I looked at them, they seemed wrong--like they shouldn't pass at all--but I wasn't sure. 17:36
moritz_ I'm not sure about the relative precdence of . and = 17:37
but I guess . is tighter
so it tries to call »++ on a List, which is immutable
KyleHa Yeah, that's what I thought.
Not only that, the test results show the last element not being changed.
moritz_ writing it as (@r = 1, 4, 9)»++ should do
ouch
and there's really no reason for the eval 17:38
I'll fix those after my spectest run finishes
KyleHa BTW, we both wrote tests for this ticket: rt.perl.org/rt3/Ticket/Display.html?id=68074
moritz_ you're right; I didn't look through RT before I pushed those 17:39
(and was offline during the weekend)
KyleHa I think your tests are better. You check a couple of things I didn't. 17:40
moritz_ s/better/more extensive/
let's revert your commit then 17:41
KyleHa I haven't done that before. 17:42
moritz_ git revert $hash
KyleHa Excellent, thanks.
masak git++
moritz_ maybe remove the bug number from the commit message
otherwise we'll get a third mail sent to the same ticket ;-) 17:43
KyleHa OK, that reverts it here, but I'm not sure it got back to svn. 17:44
moritz_ git svn dcommit
KyleHa Oh, there it goes.
pugs_svn r27853 | kyle++ | Revert RT #68074 tests (more extensive tests appear elsewhere).
r27853 | kyle++ |
r27853 | kyle++ | This reverts commit d2783ead01f44813594163f02abd8d904315391e.
moritz_ it's just a normal commit, that happens to have the inverse effect of another commit 17:45
KyleHa Press toe.
Cool. Thanks, moritz++
17:53 jauaor left 17:55 hudnix joined
dalek kudo: c57fbd7 | moritz++ | t/spectest.data:
[t/spectest.data] remove a test that was merged; add a few newly passing tests
 17:55
kudo: ed636b8 | moritz++ | t/spectest.data:
one more passing test file, KyleHa++
moritz_ rakudo: my @a = 1, 4; @a»++; 17:56
p6eval rakudo e89cc1: OUTPUT«Statement not terminated properly at line 2, near "\x{bb}++;"␤in Main (src/gen_setting.pm:3363)␤»
Juerd latin1 fail :)
moritz_ rakudo: my @a = 1, 4; @a>>++;
p6eval rakudo e89cc1: OUTPUT«Statement not terminated properly at line 2, near ">>++;"␤in Main (src/gen_setting.pm:3363)␤»
pugs_svn r27854 | kyle++ | [t/spec] Test for RT #65804 17:57
moritz_ Juerd: I think it escapes codepoints, not bytes
Juerd: so \x{bb} is fine
Juerd Oh, okay :)
You're right
Especially I should know. :(
17:58 lucs left
moritz_ :/ 17:58
pugs_svn r27855 | moritz++ | [t/spec] correct and fudge @array>>++ tests in hyper.t 17:59
moritz_ we all mix up Perl 5 and Perl 6 from time to time ;-)
masak speak for yourself. :P 18:00
(actually, yes, it does happen. but I don't find it a big deal -- it's no worse than the momentary confusion of switching keyboards or keyboard layouts.) 18:01
moritz_ daniel.ruoso.com/categoria/perl/too-much-perl-6 ruoso++ knows it, too ;-)
masak :) 18:02
moritz_ actually I think it's a very interesting post 18:04
the built-in ST is really a *very* handy feature
and it was trivial to add to Perl 6
so I thought about how to back-port it to Perl 5
and found that it wasn't all that easy
because sort can't easily introspect the block to find out if its arity is only 1 18:05
ENOTENOUGHMAGIC so to say 18:06
pmichaud greetings 18:09
masak pmichaud: evning.
pmichaud: how was class today?
18:09 nihiliad joined, Whiteknight left
pmichaud class went well 18:10
I didn't get much otherwork done (and no wireless), but it was very interesting 18:11
I have some notes of more things to "fix" in rakudo :-)
moritz_ PerlJam: ping
masak pmichaud: cool.
moritz_ PerlJam: I have some commits to perl6-docs locally; one of them changes the regex tutorial to using the spaced form by default 18:12
masak I think it's a great asset to both implementors and students to be in the same room during such a class.
moritz_ PerlJam: would agree with this change? I don't want to commit it without your approval.
18:17 rfordinal left 18:35 zamolxes joined, Chillance joined
masak does anyone know how the smoke results for the HTML spec files are generated? 18:38
moritz_ "by a cron job" 18:40
masak yes, sure. but is there a CPAN module that generates a YAML file, or something like that?
moritz_ I think so, yes
the script is in docs/feather/script/auto-smoke
masak obrigado. 18:41
moritz_ which in turn runs util/run-smoke.pl at some point
masak right.
so it's util/make-smoke.pl that creates the smoke.yml somehow. 18:42
s/make/run/
moritz_ which in turn calls util/yaml_harness.pl 18:43
masak just saw that :)
moritz_ which in turn does scary magic.
masak "you're in a maze of twisty little scripts..."
yes, but at least it imports the modules I've been looking at for the past half hour. 18:44
ugh, but it's long :/
moritz_ rakudo is now passing 12169
it subclasses one of those...
masak moritz_++
moritz_ turns away in pain 18:45
rakudo: say 14913 / (37*60+24) # average tests/second 18:49
p6eval rakudo c57fbd: OUTPUT«6.64572192513369␤»
18:55 tann joined 18:56 zamolxes left 19:21 molaf left 19:22 Front_slash left 19:27 icwiener left 19:30 desertm4x joined
masak rakudo: say get $*IN 19:35
p6eval rakudo c57fbd: OUTPUT«Land der Berge, Land am Strome,␤»
masak I didn't know there was a sub form of IO.get...
where is that specced?
oh, S07. 19:37
cbk masak, has anything changed in S05 regarding named regex's? 19:39
19:39 M_o_C joined
moritz_ lots of things change ;-) 19:39
masak cbk: don't think so. is there a problem?
cbk masak, well if you don't mind checking my regex's and a sub of mine out.... will post in a sec. 19:40
masak I might have posted this before: elliotlovesperl.com/2009/06/16/moos...mentation/ -- it strikes me that we don't have a 'Maybe' type yet. :) 19:41
cbk: gladly.
lisppaste3 cbk pasted "named regex's" at paste.lisp.org/display/84632 19:42
cbk ok there it is 19:43
masak looks
(btw its "regexes" lol)
cbk oh yah
moritz_ it seems whitespaces are really precious to you ;-)
cbk what do you mean? 19:44
masak yes, is that intentional?
cbk: he means that you never match on \s
moritz_ ... and that you don't use whitespaces in all those regex defintions, even though they don't have an effect
cbk well there is a space in between each var I want to match
masak yes, but it isn't significant, that's the point. 19:45
per default, whitespace in regexes don't mean anything.
cbk this is what I did have before I tried using named regexe's
moritz_ try this: /:s <teamNick> <teamName> <teamCity> <teamWin> <teamLoss> <teamLogoFile>/
cbk # if $line ~~ rule { (<[A..Z]>+) (<[a..zA..Z]>+) (<[a..zA..Z]>+) (\d+) (\d+) ([\w|\-]+\.gif) } { # PerlJam's way
moritz_, ok
masak cbk: observe 'rule'. 19:46
moritz_ cbk: do you know the /x modifier in Perl 5?
masak cbk: whitespace _is_ significant in rules.
moritz_ that's enabled in Perl 6 by default
cbk yah where do I put 'rule' in a named regexe
moritz_ and it seems those regexes could be tokens insteads
masak can never tell 19:47
cbk well maybe so, I'm just learning going off of S06
and then building up from there
masak cbk++
moritz_ cbk: have you read PerlJam's regex tutorial yet?
masak (S05, you mean)
moritz_ if not, that might be a good start 19:48
cbk don't really even need to do it this way, I had it allraedy working
masak, yes S05
moritz_, I'm not sure on PerlJam'S regexe tutorial where is it? well maybe I have it. Too much loose paper to look through.... 19:49
moritz_ cbk: in github.com/perlpilot/perl6-docs somewhere 19:50
masak cbk: 'my $currentObj = @teamsObj.end;' this looks non-idiomatic to me. you might want to create the object first, assign it all its properties, and _then_ push it to the array.
moritz_ in the intro dir
cbk moritz_, thanks
masak cbk: also, you can give it all its properties at creation if you like.
cbk masak, well I was just trying to find a good use for '.end ' and put it to use in my program so I have an example of it for latter. :-) 19:52
masak cbk: :P
cbk: since I implemented it (wrongly) almost a year ago, I haven't used it once, I think.
19:53 teslos joined
moritz_ I hope somebody fixed it in the mean time ;-) 19:53
masak yes.
don't remember who, but I hope it was me.
moritz_ (fixing own stuff)++
masak I had it implemented as meaning self[self.end]
moritz_ ah, I remeber that 19:54
cbk I kinda of like it . Can be usefull to easly get the last element in an array.
moritz_ @array[*-1] is shorter ;-)
lambdabot Unknown command, try @list
moritz_ @shutup
lambdabot Unknown command, try @list
masak cbk: yah, but in your case you simply shouldn't put it there in the first place.
moritz_ masak: don't be so picky; timtowtdi and all ;-)
cbk by the way, what does nick++ mean? like cbk++ 19:55
masak allright, allright, I'll try to be more understanding. :P 19:56
cbk: it means that we like you.
cbk sometimes you guys do that, I have no clue on it's meaning
masak @karma cbk
lambdabot cbk has a karma of 2
masak @karma moritz_
lambdabot moritz_ has a karma of 109
masak @karma TimToady
lambdabot TimToady has a karma of 93
moritz_ @karma masak
lambdabot masak has a karma of 284
moritz_ @karma lwall
lambdabot lwall has a karma of 546
cbk oh ic
masak lwall++
moritz_ @karma moritz_ 19:57
lambdabot You have a karma of 109
moritz_ @karma moritz
lambdabot moritz has a karma of 917
moritz_ IRC karma vs. committing karma ;-)
masak committing++
cbk can I trade my Karma in for prizes or something, maybe a perl 6 book 19:58
moritz_ (giving karma to random facts)++
cbk: sure, you just have to find a sponsor first ;-)
tann @karma tann 20:00
lambdabot You have a karma of 3
tann @karma tann_
lambdabot tann_ has a karma of 0
cbk well I rather spend my time working on perl 6 first, then I'll work on being a karma hound
tann :)
20:00 pmurias joined
tann tann++ 20:00
@karma tann
lambdabot You have a karma of 3
tann so no self-promoting here
moritz_ a bot I know decreses your karma if you try that (over in #perlde) ;-) 20:01
masak cbk: when you reach karma of 1000, you win your own metaoperator. that's why moritz is working so hard on the commits. :) 20:03
moritz_ actually I should have mine already, but lambdabot lost its karma records some time ago 20:04
I made 1429 commits to the test suite, and I don't think anybody gave me 500 -- ;-) 20:05
Tene @karma tene
lambdabot You have a karma of 45
moritz_ such is the dire, thankless world ;-)
masak moritz_: you sure you weren't given a metaoperator already, but you simply haven't gone and picked it up? 20:06
tann why isn't it ++nick instead of the inefficient nick++? :P
20:06 desertm4x left
Tene tann: since it's not being assigned to anything, the optimizer catches it 20:07
masak tann: because we don't believe in premature optimisation.
moritz_ masak: ok, I take the reduce ;-)
I always liked that one ;-)
masak :)
yeah, reduce is my favourite metaoperator too. but assignment is pretty sweet also. 20:08
tann moritz: sounds like a republican selling point - reduce reduce and reduce
moritz_ tann: the interesting qustion is always what to reduce ;-) 20:10
masak likes to reduce strings
moritz_ for example I'd love to reduce survaillance
and lists
tann ah, i see..not spending and gov size :) 20:11
masak rakudo: my @surveillance = <BIG BROTHER IS WATCHING YOU!>; say [~] map { $_, '-' }, @surveillance
p6eval rakudo c57fbd: OUTPUT«BIG-BROTHER-IS-WATCHING-YOU!-␤»
tann so, the roadmap suggests that rakudo 1.0 won't have threads?
masak there, reduced @surveillance.
M_o_C ~ concatenates? 20:12
masak M_o_C: right.
moritz_ tann: I'd be surprised if it did. Then again Tene++ pleasantly surprised me a few times already ;-)
masak tann: the roadmap doesn't mention 1.0... but no, it probably won't.
tann tene++
masak tEnE++ 20:13
20:13 kane_ joined
moritz_ the roadmap mostly mentions things that are needed to run STD.pm, or for other boostrapping tasks 20:13
doing massive parallel IO on an encrypted disc partition is not a good idea... 20:18
20:24 asciiville joined, pmurias left
masak rakudo: get 42 20:32
p6eval rakudo c57fbd: OUTPUT«No such attribute '$!PIO'␤in method IO::get (src/gen_setting.pm:797)␤called from Main (/tmp/CItChk1TK1:2)␤»
masak this could be seen as a bug.
maybe &get should only accept Iterator objects? 20:33
eternaleye Does anyone know if parameterized classes are possible? I'm aware of parameterized roles, but what if I wanted to implement a class which does Positional? Does the class inherit the parameterization from Positional? Or would something like 'class Foo does Positional[::T] {}' be needed? 20:34
masak rakudo: class Foo does Positional[::T] {}; say Foo ~~ Positional 20:35
p6eval rakudo c57fbd: OUTPUT«Null PMC access in isa()␤in Main (/tmp/WOxUdI8r1s:2)␤»
masak submits rakudobug
KyleHa rakudo: my $foo = 'foo'; say $foo R~= 'foo'; 20:36
p6eval rakudo c57fbd: OUTPUT«No applicable methods.␤in Main (/tmp/QD1ldJ8zKv:2)␤»
moritz_ eternaleye: you have to supply the parameters at class composition time... 20:37
masak rakudo: role A {}; class Foo does A[::T] {}
moritz_ eternaleye: or write a role yourself, instead of a class
p6eval rakudo c57fbd: OUTPUT«Null PMC access in isa()␤in Main (/tmp/8yKVKHWRBk:2)␤»
eternaleye Also, the spec on Set (S32/Containers) is unclear: it says "Like an ordinary list pick, but returns keys of the bag weighted by values, as if the keys were replicated the number of times indicated by the corresponding value and then list pick used." - but if it's a Hash of True, there is no integer value to use as described
A Bag has replication of elements, but a set is defined as being composed of unique members 20:38
pugs_svn r27856 | kyle++ | [t/spec] Test for RT #65962
r27857 | kyle++ | [t/spec] Test for RT #64818
moritz_ eternaleye: True numifies to 1, False to 0
eternaleye moritz_: Then it's redundant - there is no 'number of times', only presence or absence - which renders the 'replicated' bit nonsensical 20:39
Also, S14 says "a role may not be directly instantiated. 20:40
moritz_ that is a bit misleading
eternaleye Whick leaves one with a dilemma: how can one make generic collections?
moritz_ I stumbled over that today too 20:41
eternaleye Oh, are they punned to classes?
moritz_ if you try to instantiate a role, it will automatically as a class of the same name
eternaleye Ah
moritz_ I'll have to ask jnthn/TimToady what that means 20:42
eternaleye moritz_: It means what you just said
moritz_ then it should says what I just said
eternaleye Cool
It seems an odd restriction that roles can be parameterized and not classes, though 20:43
moritz_ it seems that roles are in general much more powerful
and it's amazing how much of S12++ and S14 jnthn implemented so far 20:44
masak it is. 20:45
tann i can't tell why there's need for both package and module? 20:46
moritz_ tann: I think packages are there for Perl 5 compatibility
tann am i right tis that package > module > class ? ie...one contains another
masak contains in what sense?
certainly not in the most literal sense, at least. 20:47
tann package P { module M { class C } }
masak they are different abstractions.
time for food &
20:47 masak left
tann likewise, dimsum is callin' :) 20:49
eternaleye rakudo: role TreeNode[::T] does Positional[T] { has TreeNode[T] @!children[2] handles 'postcircumfix:<[ ]>'; has T $.data is rw; }; my $tree = TreeNode[Int].new; $tree.data = 3; $tree[0] = TreeNode[Int].new; $tree[1] = TreeNode[Int].new; $tree[0].data = 1; $tree[1].data = 4; .say for $tree.data, $tree[0,1]>>.data; 20:50
p6eval rakudo c57fbd: OUTPUT«Statement not terminated properly at line 2, near "[2] handle"␤in Main (src/gen_setting.pm:3363)␤»
moritz_ rakudo doesn't know about sized arrays
eternaleye Ah
moritz_ has ... @!children[2] doesn't work yet 20:51
eternaleye rakudo: role TreeNode[::T] does Positional[T] { has TreeNode[T] @!children handles 'postcircumfix:<[ ]>'; has T $.data is rw; }; my $tree = TreeNode[Int].new; $tree.data = 3; $tree[0] = TreeNode[Int].new; $tree[1] = TreeNode[Int].new; $tree[0].data = 1; $tree[1].data = 4; .say for $tree.data, $tree[0,1]>>.data;
p6eval rakudo c57fbd: OUTPUT«Could not find non-existent sub T␤»
eternaleye Should that have a line number? 20:52
moritz_ yes
and it shouldn't complain that way, too
my guess is that the T is not yet bound where it says 'does Positional[T]'
eternaleye rakudo: role TreeNode[::T] does Positional { has TreeNode[T] @!children handles 'postcircumfix:<[ ]>'; has T $.data is rw; }; my $tree = TreeNode[Int].new; $tree.data = 3; $tree[0] = TreeNode[Int].new; $tree[1] = TreeNode[Int].new; $tree[0].data = 1; $tree[1].data = 4; .say for $tree.data, $tree[0,1]>>.data;
p6eval rakudo c57fbd: OUTPUT«src/call/pcc.c:613: failed assertion 'PObj_is_PMC_TEST(sig_pmc)'␤Backtrace - Obtained 25 stack frames (max trace depth is 32).␤/home/evalenv/rakudo/parrot/blib/lib/libparrot.so.1.4.0 [0xb7c85742]␤/home/evalenv/rakudo/parrot/blib/lib/libparrot.so.1.4.0(Parrot_confess+0x9a)
..[0xb7c858…
eternaleye Ooh, ouch 20:53
moritz_ please submit both of them
eternaleye What's the rakudobug email address? 20:54
moritz_ [email@hidden.address]
eternaleye Thanks
moritz_ thank you for submitting
tann moritz: some of the rt tkts seem to have gotten fixed ... should there be a bug scrub to rid those? 20:57
moritz_ if you find some, just paste their numbers here, sombody will close them 20:59
(KyleHa, masak or me, usually)
eternaleye But that assertion one will be interesting to find a fix for, I bet 21:02
rakudo: role TreeNode[::T] { has TreeNode[T] @!children handles 'postcircumfix:<[ ]>'; has T $.data is rw; }; my $tree = TreeNode[Int].new; $tree.data = 3; $tree[0] = TreeNode[Int].new; $tree[1] = TreeNode[Int].new; $tree[0].data = 1; $tree[1].data = 4; .say for $tree.data, $tree[0,1]>>.data; 21:03
p6eval rakudo c57fbd: OUTPUT«maximum recursion depth exceeded␤in Main (/tmp/MOCBPubqbz:2)␤»
eternaleye ...
It seems I'm on a roll today
moritz_ you shall not define trees, or the tree will defy you 21:09
21:09 teslos left
eternaleye I'm thinking it may be @!children that's doing it - any ideas? 21:10
Tene So, select() is NYI in rakudo, yes?
moritz_ eternaleye: it could be that postcircumfix:<[ ]> is not yet overridable or subject to 'handles' 21:12
eternaleye Maybe. 21:13
moritz_ Tene: seems so. I think that t/spec/S32-io/IO-Socket-INET.t tests all the socket things that Rakudo implements right now
(which is not all that much)
Tene so looks like there's no real way to deal with multiple sockets sanely... :( 21:14
pmichaud good evening, #perl6 21:25
eternaleye phenny: tell masak The Astaire docs have an error: the useragent example has :agent => /regex/ - the colons don't belong IIUC
phenny eternaleye: I'll pass that on when masak is around.
eternaleye pmichaud: Hello
21:25 iblechbot joined
KyleHa If Rakudo gives a segmentation fault, that's really a Parrot bug, right? 21:27
moritz_ KyleHa: not always; rakudo also contains some C code that might sgefault 21:28
KyleHa: so we submit them as rakudobugs
KyleHa Oh, OK. Good to know, thanks.
Well, I got the segfault while testing an existing bug. 21:29
And it only happens when I stick my new tests at the end of the test file with other stuff.
My new tests by themselves run.
moritz_ bah 21:30
KyleHa I guess I should try to isolate it.
21:39 maja left 21:43 maja joined 21:55 iblechbot left 22:00 M_o_C left 22:25 ihrd joined 22:28 KyleHa left 22:33 ab5tract joined
Tene With some rakudo patches and parrot patches, I finally have an echo server working in Rakudo with multiple clients.. :) 22:39
Tene eagerly waits for async IO on parrot. 22:42
japhb Tene: Do you have a copy of Network Programming with Perl?
Tene japhb: no. 22:43
japhb Tene: Ah. I was going to say, "When we can get everything in that book working, we'll be up to Perl 5 level ... then we get to work on APUE2 ...." 22:44
Tene :)
japhb And FWIW, NPwP is an excellent book, though much will be review if you already know APUE2. 22:45
Tene I don't recognize the acronym APUE2. 22:46
japhb Advanced Programming in the Unix Environment, 2nd Ed.
(AKA APUE in the Linux and FreeBSD world) 22:47
Another in the list of "Addison-Wesley: Why You Still Buy Dead Trees." 22:48
Tene Huh. I'll have to buy it.
anyway, AFK 22:49
cognominal APUE2? 22:59
ha ok
22:59 jauaor joined 23:00 fridim_ left
japhb
.oO( How many copies of NPwP have I convinced people to buy over the years ...?)
 23:03
23:08 jensjenny joined, jensjenny left 23:15 ruz_mac left 23:17 mikehh joined 23:23 buubot left 23:24 buubot joined 23:29 DemoFreak joined 23:31 ruz_mac joined
mikehh All tests PASS (pre/post-config, smolder, nqp_test, fulltest) at r40390 - Ubuntu 9.04 amd64 23:44

Powered by Cro and the Raku® Programming Language.

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

Thank you!