perl6-projects.org/ | nopaste: sial.org/pbot/perl6 | evalbot: 'perl6: say 3;' | irclog: irc.pugscode.org/
Set by mncharity on 25 March 2009.
skids A where {...} would seem the appropriate perl6 syntax for Contextual::Return 00:18
meppuru good night 03:42
s1n how does submethod differ from using 'my method'? 04:13
dukeleto can someone point me in the right direction regarding NCI with Rakudo/Parrot? Is there a document describing how to add bindings to a third-party library via NCI ? 04:15
bacek_ dukeleto: docs/pdds/draft/pdd16_native_call.pod in parrot 04:23
dukeleto: also examples/nci in parrot can give some... erm.. examples :)
dukeleto bacek: thanks, I didn't look in examples for some dumb reason ... 04:26
bacek_ dukeleto: what you are trying to bind? 04:29
dukeleto bacek: Oh, I have lots of things in mind, but GSL is the first. It is a fairly large library, but I already have a large test suite in Perl 5 for Math::GSL that could be hijacked 04:32
bacek_ dukeleto: looks great. 04:34
dukeleto i have made a new trac ticket at trac.parrot.org/parrot/ticket/531 to decide on a pilot/co-pilot for Parrot-related affairs this summer
bacek_ will never apply for this :) 04:36
dukeleto bacek: thanks. if parrot had GSL bindings, than there is the possibility for any language using Parrot to get GSL bindings "for free.
bacek_ I've got enough manager's responsibility at $work...
dukeleto bacek: yeah, just a general notice for the channel, wasn't directed at you or anything :)
that is pretty cool, since every dynamic language I know of that has GSL bindings put a lot of sweat, blood and tears into making it happen. 04:37
bacek_ ...one ring to bind them all... :) 04:38
dukeleto exactly
s1n frioux|home: ping 04:46
frew|work: ping 04:47
s1n if i wanted to lock a method, i.e. make it closed, is there any such way for allowing subclasses from inheriting but not augmenting? 05:11
that is, a method in a class
s1n i see the closed/nonfinal pragmas, but i'd really like perl method control 05:12
s/perl/per/
moritz_ rakudo: say {$^a + $^b}.arity 06:41
p6eval rakudo 78cb4c: OUTPUT«2␤»
moritz_ rakudo: say {$^a + $^b}.(2, 3, 4) 06:49
p6eval rakudo 78cb4c: OUTPUT«too many arguments passed (3) - 2 params expected␤current instr.: '_block20' pc 104 (EVAL_19:55)␤»
Matt-W Morning 07:30
moritz_ good morning
mikehh Matt-W: good morning
bacek good evening 09:31
bacek just used last century technology. 09:32
I've faxed CLA :)
Matt-W wow 09:39
fax
I remember thinking fax machines were really cool
bacek I'm just was too bored to understand how to scan-to-pdf... 09:43
ihrd hi there 10:54
ihrd rakudo: say /./ ~~ Regex 10:54
p6eval rakudo 78cb4c: OUTPUT«0␤»
ihrd rakudo say /./ ~~ Code
rakudo: say /./ ~~ Code 10:55
p6eval rakudo 78cb4c: OUTPUT«1␤»
ihrd why /./ ~~ Regex fail? this is bug or I miss something? 10:56
jnthn ihrd: Waiting on some (probably coming soon) PGE changes to be able to fix that. 10:57
ihrd do you need ticket? 10:58
jnthn I *think* we already have one. 10:59
bacek moritz_: why you marked rt.perl.org/rt3/Ticket/Display.html?id=62730 as "stalled", not "closed"? 11:03
moritz_ bacek: my mistake 11:08
bacek moritz_: np. I closed it :) 11:10
bacek sometimes hates masak... 11:13
It was less than 250 tickets couple of weeks ago. Now it 289!
jnthn masak++ 11:16
I'll try and cut that down by a few tomorrow. 11:18
moritz_ bacek++ 11:37
ruoso Hello! 11:58
moritz_ moans because he can't write closures in C++ 12:01
sbp anybody work on rakudo.org here? 12:05
"documentation" on the home page body text links to how-to-get-rakudo
rather than the documentation
ruoso moritz_, there is a library that provides closures in C
moritz_ sbp: thanks, fixed 12:07
sbp "pointy blocks ( -> $foo { ... } )" (implemented) 13:47
— rakudo.org/status
well, with -> $a or $^a, certainly, but:
rakudo: for ('abc', 'def', 'ghi') -> { say $_ } 13:48
p6eval rakudo 78cb4c: OUTPUT«Use of uninitialized value␤␤Use of uninitialized value␤␤Use of uninitialized value␤␤»
sbp is that a known bug? I tried here:
rt.perl.org/rt3//Public/Search/Simp...24_+rakudo
but no apparent report of it
(or is my code simply wrong?)
moritz_ what's wrong with it?
you declare explicitly that your block doesn't want a parameter
sbp well, as I understand it -> { ... } means that $_ can be used much like $^a would be 13:49
moritz_ so it gets none
sbp hmm
moritz_ no, that would be { ... } without the arrow in front
sbp oh, hmm. I could've sworn I read about -> { $_ } for this somewhere...
thanks
rakudo: for ('abc', 'def', 'ghi') { say $_ } 13:50
p6eval rakudo 78cb4c: OUTPUT«abc␤def␤ghi␤»
sbp moritz_: if '-> { }' explicitly decares that the block doesn't want a parameter, how come $^a can be used therein? does $^a override the explicit declaration that you don't want parameters? 13:52
moritz_ sbp: it can't
rakudo: for <abc def ghi> -> { say $^a } 13:53
sbp oh, you're right. sorry; I'd used for (...) { ... } in my test without even realising
p6eval rakudo 78cb4c: OUTPUT«Cannot use placeholder var in block with signature. at line 1, near " }"␤␤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:83)␤»
sbp moritz_: so is this wrong? 'while something() -> { ... $^thing ... }' - S04 13:56
moritz_ sbp: I'd say so
@ask TimToady S04: while something() -> { ... $^thing ... } # should the arrow be removed here? 13:57
lambdabot Consider it noted.
moritz_ but I'm never quite sure
sbp heh, thanks
moritz_ because the synopsis are a large and complex beast
ruoso std: for 1,2,3 -> { say $^a } 13:58
sbp aye. I can only find one other example for -> { ... }:
lives_ok -> { $foo.x[0] = 3 }, "Array initialized in auto-constructor is not unwritable...";
p6eval std 26041: OUTPUT«##### PARSE FAILED #####␤Malformed block␤Placeholder variable $^a cannot override existing signature ( ) at /tmp/YXtWE3F5tP line 1:␤------> for 1,2,3 -> { say $^a }␤ expecting desigilname␤FAILED 00:02 36m␤»
sbp — t/spec/S12-construction/construction.t lines 7–67
ruoso moritz_, I think that answers your question ;) 13:59
sbp not entirely sure what's going on there. what is the use case in general for declaring that you *don't* want parameters?
moritz_ ok
ruoso sbp, to fail if anyone tries to send one? 14:00
rakudo: -> { say 'live' }.(123);
p6eval rakudo 78cb4c: OUTPUT«live␤»
ruoso oops
hmm
rakudo: -> { say $_ }.(123);
p6eval rakudo 78cb4c: OUTPUT«Use of uninitialized value␤␤»
ruoso hmm... for sure a rakudobug 14:01
moritz_ ruoso: care to submit?
ruoso sure...
rakudo: sub foo() {}; foo(123); 14:02
p6eval rakudo 78cb4c: ( no output )
ruoso hmm 14:03
the bug is more generic than that
pugs_svn r26042 | moritz++ | [S04] fixed illegal construct, sbp++
ruoso rakudo: sub foo() { say @_[0] }; foo(123) 14:04
p6eval rakudo 78cb4c: OUTPUT«123␤»
ruoso rakudo: -> { say @_[0] }.(123);
p6eval rakudo 78cb4c: OUTPUT«123␤»
ruoso am I right to think that both cases above should mean an empty signature instead of the default signature? 14:06
ruoso just submitted rakudobug 14:07
ruoso moritz_, #64344 if you want to add something 14:11
moritz_ ruoso: thanks 14:12
ruoso rakudo: sub foo() {}; say &foo.signature.perl 14:20
p6eval rakudo 78cb4c: OUTPUT«:()␤»
pmichaud Parrot isn't able to throw an exception for zero-parameter functions.
I.e., if a sub is declared as having zero arguments, Parrot silently accepts multiple arguments.
ruoso but where is @_ coming from?
pmichaud ah, that's likely a bug, yes. 14:21
I'd have to check.
ruoso PhatEddy has pointed me in pvt that this looks similar to 56366
but I thought it would be different because it's not just failing to check empty signatures 14:22
it's implying the default signature in empty signatures...
pmichaud right -- in a couple of places we've added code to explicitly check for arguments passed to zero-arg subs.
pmichaud as far as default signature in empty signatures... I'll have to check. 14:22
ruoso I mean... considering :(*@_, *%_) is the default signature 14:23
pmichaud right now it appears that Rakudo treats any use of @_ as meaning "add a slurpy param"
ruoso hmm... is that spec?
pmichaud Yes. 14:24
S06:154
oh wait, that's for placeholders. 14:25
okay, we'll have to add some code for fixups. 14:27
that's likely to be a bit challenging 14:28
{ @_ } # uses outer @_
{ $^a; @_ } # has @_ as a parameter
{ @_; $^a } # has @_ as a parameter.... but we can't determine that until we see the $^a placeholder. 14:29
ruoso { $^a; @_ } must use outer @_ as well 14:30
pmichaud no, that's not what S06 says.
ruoso ouch 14:31
pmichaud (S06:154)
ruoso hmm
rakudo: sub foo { say $^a; say @_ }; say &foo.signature.perl
p6eval rakudo 78cb4c: OUTPUT«:(Any $a)␤»
ruoso rakudo: sub foo { say $^a; say @_ }; say &foo.signature.arity # most important, actually 14:32
p6eval rakudo 78cb4c: OUTPUT«Could not locate a method 'arity' to invoke on class 'Signature'.␤current instr.: 'die' pc 15621 (src/builtins/control.pir:204)␤»
pmichaud rakudo: sub foo { say $^a; say @_ }; say &foo.arity; 14:33
p6eval rakudo 78cb4c: OUTPUT«1␤» 14:33
ruoso rakudo: sub foo { say $^a; say @_ }; foo(1,2,3); 14:34
p6eval rakudo 78cb4c: OUTPUT«1␤23␤»
ruoso hmmm
that sounds weird
sbp rakudo: sub foo { say @_; say $^a }; foo(1,2,3); 14:35
p6eval rakudo 78cb4c: OUTPUT«23␤1␤»
ruoso rakudo: sub foo { say $^a; }; foo(1,2,3); 14:36
p6eval rakudo 78cb4c: OUTPUT«too many arguments passed (3) - 1 params expected␤current instr.: 'foo' pc 91 (EVAL_17:52)␤»
ruoso ok...
rakudo: sub foo { say $^a; say @_ }; say &foo.signature.perl;
rakudo: sub foo { say $^a; }; say &foo.signature.perl;
p6eval rakudo 78cb4c: OUTPUT«:(Any $a)␤»
ruoso these two were supposed to have different signatures... 14:37
as per S06:154
skids hrm should .arity return a range? 14:38
moritz_ no, .arity is only for required
.count is for optional
PhatEddy So if 64344 stands as a bug it would seem to be distinguished from 56366 by its use of @_ which might be worth a comment and perhaps even a subject change for the former? 14:40
ruoso well... the bug 64344 points is the generation of *@_ even without any other placeholder parameters... 14:43
56366 is about the innability to check empty signatures... 14:44
PhatEddy rakudo: say rand(999)
p6eval rakudo 78cb4c: OUTPUT«too many arguments passed - 0 params expected␤current instr.: 'die' pc 15621 (src/builtins/control.pir:204)␤»
PhatEddy isn't the @_ the cause of the innability to check empty signatures? 14:45
pmichaud No, @_ isn't the cause. It's a Parrot bug. 14:51
(for those functions that have been written in PIR) 14:52
actually, it's a Parrot bug in general.
PhatEddy the subject of 64344 is "empty signatures"
ruoso PhatEddy, well... the bug is related to empty signatures ;) 14:53
maybe that can be narrowed down a bit more
dalek kudo: e5f4a34 | pmichaud++ | docs/spectest-progress.csv:
spectest-progress.csv update: 340 files, 8049 passing, 0 failing
PhatEddy rakudo: sub foo() { }; foo("abc") 15:02
p6eval rakudo e5f4a3: ( no output )
PhatEddy rakudo: class A { method foo() { } }; A.foo("abc") 15:02
p6eval rakudo e5f4a3: OUTPUT«too many arguments passed (3) - 1 params expected␤current instr.: 'parrot;A;foo' pc 117 (EVAL_21:61)␤»
diakopter pugs: say "TECO reborn" 15:28
p6eval pugs: OUTPUT«TECO reborn␤»
ruoso later & 15:34
PhatEddy rakudo: class A {method foo() {say "hello"}}; A.new.foo() 16:41
p6eval rakudo e5f4a3: OUTPUT«hello␤»
PhatEddy rakudo: class A {method foo() {say "hello"}}; A.new.foo("abc")
p6eval rakudo e5f4a3: OUTPUT«too many arguments passed (3) - 1 params expected␤current instr.: 'parrot;A;foo' pc 129 (EVAL_21:62)␤»
zev rakudo: sqrt(9) + 1 19:23
p6eval rakudo e5f4a3: ( no output )
zev rakudo: say sqrt(9) + 1
p6eval rakudo e5f4a3: OUTPUT«4␤»
zev rakudo: say sqrt (9) + 1
p6eval rakudo e5f4a3: OUTPUT«3.16227766016838␤»
moritz_ why am I sick of junction discussions on p6l? 19:31
literal heh 19:34
maybe because there's no clear resolution to the questions yet? 19:35
jnthn It feels to me more like, there's a clear design, but people don't like it. 19:37
erm
*some* people don't like it
diakopter I think this squeaky wheel might need some grease. 19:38
zev if anyone out there wants to watch the talk that Larry is giving at MIT this afternoon, you can follow a link from sipb.mit.edu/projects/talks/2009/larry_wall/ in about 10 minutes 20:09
I believe there will also be a recording afterwards 20:10
literal cool 20:11
jnthn nice
brunoV zev++ 20:12
pmichaud alas, can't see the MIT video :-| 20:30
oh well, I'm probably too busy anyway.
dukeleto can anybody add some meat to which parts of the Perl 6 test suite need to be worked on at www.perlfoundation.org/perl5/index....9_projects ? 20:31
it would be much appreciated
or a few links to mailing list discussions which talk about it 20:32
zev it just got better
direct link: rtsp://ten-streaming.mit.edu:554/2009/04/wall.sdp
moritz_ dukeleto: I can do that 20:33
dukeleto moriz: that would be awesome, i will then tweet it 20:34
moritz_ zev: I get the video, but no audio
zev interesting 20:35
moritz_ Audio: no sound
PerlJam I got sound
Uri's talking now.
jnthn has sound
pmichaud oh. my Firefox doesn't understand rtsp: 20:36
moritz_ oh, I get audio with vlc, and video with mplayer
PerlJam yeah, I'm using mplayer. FF is buggy for me.
pmichaud yes, I get it now.
thanks.
(mplayer++ ftw)
zev cool
pmichaud I wonder if #perl is on a screen somewhere. Hi, Larry! 20:37
nice hat.
PerlJam heh, I was just thinking that
moritz_ well, now I have two applications open
PerlJam makes him look swash-buckling.
I expect indiana jones type music to be playing 20:38
jnthn hat win!
dukeleto where are the slides? 20:39
pmichaud on his laptop?
:-)
dukeleto and who is humping the speaker?
PerlJam ah, someone heard us
dukeleto yay, slides! 20:40
moritz_ someone++
PerlJam but now where's the Larry? :)
This presentation is very takahshi it seems. 20:41
moritz_ dukeleto: I've added a few bullet points - is that a good start? 20:47
dukeleto will check it out 20:50
moritz_ do you have specific candidates who are interested? 20:51
if you have, they should contact me ;-)
ruoso Heya 20:54
PerlJam heh! Camelia makes an appearance! 20:58
pmichaud note to people watching larry's talk (in real time): to help perl 6 development, write perl 6 applications! :-) 22:13
jnthn :-) 22:14
pmichaud: Great to see you're digging back into PGE! 22:15
pmichaud yes, these are all things that have been on my back burner for a while as I dealt with other "more immediate" issues.
I'm glad it's back on the front burner.
jnthn Same.
moritz_ write perl 6 applications - I fully agree :-) 22:18
jnthn will be Rakudo-daying tom^Wtoday :-) 22:19
pmichaud excellent. Although I can't believe it's Wednesday already. :-)
Tene where's larry talking? 22:20
jnthn I can't believe it's *April* already.
pmichaud Tene: MIT
jnthn needs to start prepping for NPW.
pmichaud Tene: just finished (at least my stream of it did)
dukeleto yep, i enjoyed it very much 22:22
Tene Aw. 22:23
jnthn Yes, was good to watch. 22:25
Tene: I think video will be posted later also.
moritz_ it would be nice to have always the latest release announcement somewhere on rakudo.org 22:27
dukeleto indeed
Tene so, jnthn and pmichaud, any consensus on leave? 22:29
jnthn Tene: Did you read the discussion I think yesterday on that?
Tene For the past few days I've been offline. 22:30
So, no.
I've read it now. 22:31
jnthn irclog.perlgeek.de/perl6/2009-03-31#i_1028123
ah
:-)
It felt consensus-ish. 22:34
Tene Yes, I asked Larry whether you could catch leave exceptions, and he waffled for a bit, and then trailed off.
Tene If Parrot is going to have block-exit stuff that we can use, then yes, that would be better. 22:37
I was expecting to implement all of that myself. 22:38
And if we're doing it ourselves, exceptions looks like the way to do it (when you have a hammer...)
pmichaud (latest release announcement) I've asked how we can get news items to appear on the front page, but no response yet. 22:40
moritz_ :(
pmichaud I'll ask again and we'll give it a week and if no response we'll look into something more drastic. 22:41
jnthn pmichaud: I asked too about that.
pmichaud afk # kid soccer practice (I got called to coach today's practice at last minute) 22:44
moritz_ I've sorta finished my next tidings post, but I'm so tired i'd better re-read it tomorrow morning before publishing it ;-) 22:46
Tene argh $realjob calls
'night moritz
jnthn Aye, I'll probably call it a night for now too... See you all tomorrow :-) 22:47
cspencer if i'm writing a Perl 6 class and I wanted to be able to numify it (with the + prefix op), am i supposed to just defined a multi method called Num? 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!