»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend!
Set by sorear on 4 February 2011.
00:00 lestrrat left 00:01 lestrrat joined 00:03 derrazo joined 00:05 Araq left
moritz sorear: no, p6eval is on feather3 00:49
sorear: but the compilers are built on feather1 and rsync'ed to feather3, because feather3 doesn't have enough memory for building stuff
01:18 benabik joined
sorear ah. 01:18
01:22 wolfman2000 joined 01:27 tokuhiro_ left, mucker joined, tokuhiro_ joined 01:32 tokuhiro_ left 01:38 fgomez joined, fgomez left, fgomez joined 01:43 mucker left, xinming left 01:44 fgomez left 01:49 fgomez joined
timotimo is creating instances in rakudo especially slow? how about math with floating points i'd get from sin/cos? 01:50
01:58 whiteknight left 02:22 leprevost joined 02:25 orafu left, orafu joined 02:29 derrazo left 02:50 sporous left, sporous joined 03:14 sftp left 03:39 libertyprime left 03:41 libertyprime joined, envi_ left 04:04 thou joined 04:09 drbean joined, havenn joined 04:29 kaleem joined 04:46 fgomez left
moritz \o/ new HPMoR chapter 05:03
05:03 rsimoes left 05:04 thou left 05:09 birdwindupbird joined 05:15 liuchong joined
tadzik good morning :) 05:16
moritz good morning tadzik 05:17
dalek ast: 58512ac | moritz++ | S32-exceptions/misc.t:
test index out of range exception
 05:21
05:22 kaare_ joined 05:26 havenn left, leprevost left 05:30 mdupont left
moritz p6 05:31
p6: say "foo"[2..*]
p6eval niecza v15-6-gefda208: OUTPUT«(timeout)»
..pugs, rakudo 4373f0: OUTPUT«␤»
05:46 rsimoes joined 05:52 wtw joined 05:53 JimmyZ joined
Teratogen p6 say 1..10; 05:54
p6: say 1..10; 05:55
p6eval pugs: OUTPUT«12345678910␤»
..rakudo 4373f0, niecza v15-6-gefda208: OUTPUT«1..10␤»
Teratogen WHOOP!
wait, what?
dalek ast: 87dea8d | moritz++ | S02-types/lists.t:
RT #112216
 05:57
06:01 lestrrat left 06:02 lestrrat joined 06:09 vlixes joined 06:15 NamelessTee joined 06:21 tarch joined, SHODAN left 06:25 lestrrat left 06:26 SHODAN joined 06:27 lestrrat joined 06:54 JimmyZ left 06:56 fhelmberger joined 07:00 lestrrat left 07:01 lestrrat joined 07:08 brrt joined 07:20 fglock joined 07:32 ggoebel left 07:35 NamelessTee left 07:44 hillu_ is now known as hillu 07:50 bbkr1 left 08:01 vlixes left 08:07 ggoebel joined 08:21 kaleem left 08:24 kaleem joined 08:28 lestrrat left 08:30 lestrrat joined
masak good now, #perl6. 08:35
Teratogen: pugs listifies (and doesn't put in spaces in between). Niecza and Rakudo are more up to current spec, and print the .gist of the Range, a succinct description of it.
Teratogen cool 08:37
p6: say 1..; 08:38
p6eval rakudo 4373f0: OUTPUT«===SORRY!===␤Confused␤at /tmp/rV7yfvhrZe:1␤»
..pugs: OUTPUT«*** unknown parse error␤ at /tmp/cTRVS0Pv7S line 1, column 9␤»
..niecza v15-6-gefda208: OUTPUT«===SORRY!===␤␤Bogus term at /tmp/TWlJRHkesD line 1:␤------> say 1..⏏;␤␤Parse failed␤␤»
Teratogen p6: say 1..inf;
p6eval niecza v15-6-gefda208: OUTPUT«===SORRY!===␤␤Undeclared routine:␤ 'inf' used at line 1␤␤Unhandled exception: Check failed␤␤ at /home/p6eval/niecza/boot/lib/CORE.setting line 1362 (die @ 3) ␤ at /home/p6eval/niecza/src/STD.pm6 line 1147 (P6.comp_unit @ 33) ␤ at /home/…
..pugs: OUTPUT«*** No such subroutine: "&inf"␤ at /tmp/t_7JAk0dvc line 1, column 5-11␤»
..rakudo 4373f0: OUTPUT«===SORRY!===␤CHECK FAILED:␤Undefined routine '&inf' called (line 1)␤»
Teratogen baw
masak Teratogen: it's "Inf".
Teratogen p6: say 1..Inf;
ooooooooh 08:39
p6eval pugs: OUTPUT«Stack space overflow: current size 8388608 bytes.␤Use `+RTS -Ksize -RTS' to increase it.␤»
..rakudo 4373f0, niecza v15-6-gefda208: OUTPUT«1..Inf␤»
masak Teratogen: and once upon a time the syntax was actually '1...' with three dots. but we found a better use for three dots :)
08:47 fhelmberger left 08:48 fhelmberger joined
masak pugs dies there for the same reason it outputs the finite list :) 08:48
slurping and infinity -- often a bad combo.
08:49 mishin__ joined
masak phenny: tell lichtkind ask on p6l about $=POD vs $=pod. TheDamian hangs out there, and he's likely to reply. 08:51
phenny masak: I'll pass that on when lichtkind is around.
08:53 dakkar joined 08:56 envi_ joined
masak nice: areweslimyet.com/ 09:09
we should do something like that.
moritz you mean, other than simply showing a "no"? :-)
masak graphing things. it's an efficient way of (1) paying attention to leanness (2) showing you pay attention to leanness. 09:10
s/efficient/effective/
09:11 netstar_ left
masak on HN one of the first comments is 'Our engineering team's motto is "if you measure it, it will improve".' 09:11
09:13 netstar joined 09:14 netstar left 09:16 max___ joined 09:18 max___ left 09:19 netstar joined
netstar is there a perl6 dev book published yet? 09:19
moritz netstar: there are PDF snapshots of "Using Perl 6" available for download 09:21
netstar who is the author?
moritz masak, jnthn, pmichaud, me, PerlJam
netstar cool
moritz and there are several other contributors
like Util 09:22
netstar thanks
moritz if you have any feedback on the PDF, please let us know 09:23
netstar ok
moritz (it's not yet ready for printing, and several pieces are still missing)
netstar I remember talking to the barefoot professor in 2007 about perl6 how much closer is it to deployment now 5 years on?
moritz we have two usable compilers that have monthly releases. 09:24
netstar Okay...that's cool
moritz they aren't feature complete, and the module infrastructure still is very sparse, but it's fun to use for early adopters
netstar so u want modules written i guess 09:25
moritz and occasionally we get reports that someone or another used Perl 6 in production
yes
and more docs
netstar I fell for the parrot joke 09:26
moritz and better module installers, faster compilers, more robust compilers and IDE support, and... :-)
netstar IDE support as in syntax highlighting or as in delphi
moritz I didn't do any programming back then, so I only heard about it afterwards
netstar: well, vim and emacs already have Perl 6 syntax hilighting 09:27
netstar that's cool
moritz netstar: but that's not the same as context sensitive autocompletion, jumping to error locations and so on
I usually don't use those features, but others do
netstar sorry i have been in hospital for 1 year 09:28
i'm trying to catch up
moritz oh, I hope you're better now
and no need to apologize
netstar I'm kinda wondering if i am stuck in 2005
moritz perl6.org/compilers/features should give you a basic overview over what's implemented in which compiler 09:29
netstar no peers to verify my skills
sounds cool
09:36 renormalist left, renormalist joined 09:37 gianlucadv joined, renormalist left, renormalist joined 09:38 renormalist left, renormalist joined 09:40 renormalist left, renormalist joined 09:43 kaleem left 09:45 kaleem joined 10:03 pernatiy left, mcero joined 10:11 mucker joined 10:18 y3llow_ joined 10:21 xenu__ joined, sporous left 10:22 sporous joined, wooden_ joined 10:27 y3llow left, PZt left, xenu left, Vlavv_ left, wooden left, silug left, y3llow_ is now known as y3llow, SHODAN left, pomysl left 10:33 Vlavv_ joined, silug joined 10:35 pomysl joined, pomysl left, pomysl joined 10:36 zipf joined 10:43 iHerb joined 10:45 iHerb left
colomon r: say 4 < 5 < 6 < 7 10:46
p6eval rakudo 4373f0: OUTPUT«True␤»
colomon r: say 4 < 5 < 6 < 4
p6eval rakudo 4373f0: OUTPUT«False␤»
colomon std: say 4 < 5 < 6 < 4 10:50
p6eval std 1ad3292: OUTPUT«ok 00:00 41m␤»
10:50 zipf left
tadzik std: say "foo" << "bar" 10:52
p6eval std 1ad3292: OUTPUT«===SORRY!===␤Unsupported use of << to do left shift; in Perl 6 please use +< or ~< at /tmp/3a8BnvPYSv line 1:␤------> say "foo" <<⏏ "bar"␤Check failed␤FAILED 00:00 41m␤»
tadzik heh, almost like C++ ;)
masak r: enum A <e1 e2 e3>; multi infix:<< < >>(A, A) { False }; multi infix:<< < >>(e1, e2) { True }; multi infix:<< < >>(e2, e3) { True }; multi infix:<< < >>(e3, e1) { True }; say e1 < e2 < e3 < e1
p6eval rakudo 4373f0: OUTPUT«False␤»
masak aww :)
why doesn't the above work? 10:53
tadzik masak: doesn't it?
masak no. read carefully what it does.
tadzik oh, right, it doesn't
masak r: enum A <e1 e2 e3>; multi infix:<< < >>(e1, e2) { True }; say e1 < e2
p6eval rakudo 4373f0: OUTPUT«True␤»
masak r: enum A <e1 e2 e3>; multi infix:<< < >>(e3, e1) { True }; say e3 < e1 10:54
p6eval rakudo 4373f0: OUTPUT«False␤»
masak there we go.
that's the bit that's failing.
tadzik r: enum A <e1 e2 e3>; multi infix:<< < >>(A, A) { False }; multi infix:<< < >>(e1, e2) { True }; multi infix:<< < >>(e2, e3) { True }; multi infix:<< < >>(e3, e1) { die "LOL"; True }; say e1 < e2 < e3 < e1
p6eval rakudo 4373f0: OUTPUT«False␤»
tadzik yep
masak right.
it's never run at all.
10:54 NamelessTee joined
masak r: enum A <e1 e2>; sub foo(e1) { die "MATCHED" }; foo e1 10:55
p6eval rakudo 4373f0: OUTPUT«MATCHED␤ in sub foo at /tmp/E03yKbC5Z_:1␤ in block <anon> at /tmp/E03yKbC5Z_:1␤␤»
masak seems to be related to operators, then.
tadzik r: enum A <e1>; multi prefix:<->(e1) { die "AWW" }; say -e1
p6eval rakudo 4373f0: OUTPUT«AWW␤ in sub prefix:<-> at /tmp/LdnFrauMSw:1␤ in block <anon> at /tmp/LdnFrauMSw:1␤␤»
colomon r: enum A <e1 e2 e3>; multi infix:<< < >>(e1, e2) { say "hello!" }; say e1 < e2
p6eval rakudo 4373f0: OUTPUT«True␤»
tadzik r: enum A <e1, e2>; multi infix:<->(e1, e2) { die "AWW" }; say e1 - e2
p6eval rakudo 4373f0: OUTPUT«===SORRY!===␤Invalid typename in parameter declaration at line 1, near ", e2) { di"␤»
tadzik r: enum A <e1 e2>; multi infix:<->(e1, e2) { die "AWW" }; say e1 - e2
p6eval rakudo 4373f0: OUTPUT«AWW␤ in sub infix:<-> at /tmp/tvYdwSv0aA:1␤ in block <anon> at /tmp/tvYdwSv0aA:1␤␤»
tadzik hmm
colomon n: enum A <e1 e2 e3>; multi infix:<< < >>(e1, e2) { say "hello!" }; say e1 < e2 10:56
p6eval niecza v15-6-gefda208: OUTPUT«===SORRY!===␤␤Lexical e1 is not a package at /tmp/yiM14bThI6 line 1:␤------> num A <e1 e2 e3>; multi infix:<< < >>(e1⏏, e2) { say "hello!" }; say e1 < e2␤␤A type must be provided at /tmp/yiM14bThI6 line 1:␤------> n…
colomon n: enum A <e1 e2 e3>; multi infix:<< < >>(A::e1, A::e2) { say "hello!" }; say e1 < e2
p6eval niecza v15-6-gefda208: OUTPUT«===SORRY!===␤␤GLOBAL::A::e1 names a non-package at /tmp/qS5DPEXHih line 1:␤------> A <e1 e2 e3>; multi infix:<< < >>(A::e1⏏, A::e2) { say "hello!" }; say e1 < e2␤␤A type must be provided at /tmp/qS5DPEXHih line 1:␤-----…
colomon r: enum A <e1 e2 e3>; multi infix:<< < >>(A, A) { say "hello!" }; say e1 < e2 10:57
p6eval rakudo 4373f0: OUTPUT«True␤»
colomon r: enum A <e1 e2>; multi infix:<< - >>(e1, e2) { die "AWW" }; say e1 - e2 10:58
p6eval rakudo 4373f0: OUTPUT«-1␤»
colomon think that's your problem 10:59
std: enum A <e1 e2>; multi infix:<< - >>(e1, e2) { die "AWW" }; say e1 - e2
p6eval std 1ad3292: OUTPUT«ok 00:00 45m␤»
colomon r: enum A <e1 e2 e3>; multi infix:«<»(A, A) { say "hello!" }; say e1 < e2 11:00
p6eval rakudo 4373f0: OUTPUT«hello!␤True␤»
colomon yup
masak submits rakudobug 11:02
tadzik++ colomon++ # golfing
colomon n: enum A <e1 e2 e3>; multi infix:«<»(A, A) { say "hello!" }; say e1 < e2
p6eval niecza v15-6-gefda208: OUTPUT«hello!␤True␤» 11:03
colomon n: enum A <e1 e2 e3>; multi infix:«<»(e1, e2) { say "hello!" }; say e1 < e2
p6eval niecza v15-6-gefda208: OUTPUT«===SORRY!===␤␤Lexical e1 is not a package at /tmp/vQoBtcaOdU line 1:␤------> enum A <e1 e2 e3>; multi infix:«<»(e1⏏, e2) { say "hello!" }; say e1 < e2␤␤A type must be provided at /tmp/vQoBtcaOdU line 1:␤------> enum…
colomon n: enum A <e1 e2 e3>; multi infix:<< < >>(A, A) { say "hello!" }; say e1 < e2
masak n: enum A <e1 e2 e3>; multi infix:«<»(A where { $_ === e1 }, A where { $_ === e2}) { say "hello!" }; say e1 < e2
p6eval niecza v15-6-gefda208: OUTPUT«hello!␤True␤»
niecza v15-6-gefda208: OUTPUT«===SORRY!===␤␤Multiple prefix constraints not yet supported at /tmp/sdI3ZwnXCU line 1:␤------> >; multi infix:«<»(A where { $_ === e1 }⏏, A where { $_ === e2}) { say "hello!" }␤␤Unhandled exception: Unable to resolve met…
colomon n: enum A <e1 e2 e3>; multi infix:<< < >>(A, A) { say "hello!" }; say e1 < e2 11:04
p6eval niecza v15-6-gefda208: OUTPUT«hello!␤True␤»
masak n: enum A <e1 e2 e3>; multi infix:«<»(A $ where { $_ === e1 }, A $ where { $_ === e2}) { say "hello!" }; say e1 < e2
p6eval niecza v15-6-gefda208: OUTPUT«hello!␤True␤»
11:04 lestrrat left 11:05 lestrrat joined 11:12 xenu__ is now known as xenu
masak r: multi infix:<< < >>(2, 7) { die "whoops!" }; say 2 < 7 11:13
p6eval rakudo 4373f0: OUTPUT«True␤»
masak r: multi infix:«<»(2, 7) { die "whoops!" }; say 2 < 7
p6eval rakudo 4373f0: OUTPUT«True␤»
masak hrm.
tadzik r: enum A <e1 e2 e3>; multi infix:«<»(A, A) { False }; multi infix:«<»(e1, e2) { True }; multi infix:«<»(e2, e3) { True }; multi infix:«<»(e3, e1) { True }; say e1 < e2 < e3 < e1 11:15
p6eval rakudo 4373f0: OUTPUT«True␤»
11:19 jerome_ left 11:23 jerome_ joined
masak \o/ 11:25
for great justice! 11:26
r: enum A <e1 e2 e3>; multi infix:«<»(A, A) { False }; multi infix:«<»(e1, e2) { True }; multi infix:«<»(e2, e3) { True }; multi infix:«<»(e3, e1) { True }; say e1 < e2 < e3 < e2
p6eval rakudo 4373f0: OUTPUT«False␤»
masak \o/
less-than comparison: now also on circular domains :D 11:27
11:29 pernatiy joined 11:30 agentzh joined
moritz comes back from lunch, and masak has already submitted a rakudobug 11:30
masak couldn't resist... 11:31
11:31 snearch joined
moritz did you even try? :-) 11:31
masak r: try { resist }
p6eval rakudo 4373f0: OUTPUT«===SORRY!===␤CHECK FAILED:␤Undefined routine '&resist' called (line 1)␤»
masak r: try resist; sub resist { die "I... I can't!" }; say $! 11:32
p6eval rakudo 4373f0: OUTPUT«I... I can't!␤ in sub resist at /tmp/k9Lz31F6B_:1␤ in block <anon> at /tmp/k9Lz31F6B_:1␤␤»
masak "I generally avoid temptation unless I can't resist it." -- Mae West
that's not really an autopun, but it still feels like something fairly classifiable. 11:33
it's like when we say "...except when they don't" about rules and exceptions here on the channel.
brrt r: sub foo { if (True) { say "hoi"; } else { say "doei"; } } 11:34
p6eval rakudo 4373f0: ( no output )
masak brrt: don't need parens around True.
also, consider calling foo if you want to see it execute ;)
tadzik it's like this "an exception proves the rule" proverb
I never understood it
masak yeah, but it's more like, it's funny because the addendum undermines the whole thing. 11:35
tadzik maybe it means "the fact that there is an exception to it makes it a rule", but I still find it an autopun of some sort :)
masak it's like "It's always nice weather around here. Except the days when it's bad weather."
it's not an autopun, because there's no use/mention mixing.
tadzik I don't always eat junk food, but when I do, it's every single day
I see 11:36
masak this would be both such a joke and an autopun: "I'm always inconsistent, in fact, I'm inconsistent right now. No, wait."
tadzik hehe 11:37
lunch & 11:38
masak (probably stolen from some Hofstadter opus or other. any sufficiently self-referent joke is indistinguishable from something Hofstadter's already written.) 11:39
moritz the famous "don't parse HTML with regexes" SO post is a nice autopun 11:41
stackoverflow.com/questions/1732348...54#1732454 # this one
11:41 NamelessTee left
arnsholt Yeah, that's a classic 11:42
Oh, wait. Different one from what I thought you meant
moritz there's another classic one? 11:43
arnsholt There's a pretty good one by tchrist where he first says "Don't do that", and then goes on to do it anyways
au yes. stackoverflow.com/questions/4231382...91#4234491 11:44
arnsholt Yeah, what au said
moritz I never liked it
brrt masak: thanks, in rakudo using parens usually doesn't work in fact :-) 11:45
moritz because he uses a loop for tokenizing and progressing, so it's not really "parsing with a regex", but rather "parsing with the help of regexes" 11:46
brrt: note that if you use parens, you need a whitespace between the if and the (
brrt: if you find a case where the parens cause it to not work, and you've correctly added that space, please submit a bug report 11:47
brrt i will check, but it probably my error
masak heh, "Phenomenally Exquisite Regex Language"
moritz though rakudo's error message could really be better
rakudo: if(True) { } 11:48
p6eval rakudo 4373f0: OUTPUT«===SORRY!===␤Confused␤at /tmp/TZ9FdRfgZk:1␤»
moritz std: if(True) { }
p6eval std 1ad3292: OUTPUT«===SORRY!===␤Word 'if' interpreted as 'if()' function call; please use whitespace instead of parens at /tmp/WKBOrOuOgl line 1:␤------> if⏏(True) { }␤Unexpected block in infix position (two terms in a row) at /tmp/WKBOrOuOgl line 1…
brrt my error indeed :-) did not know that
11:51 plutoid joined
masak brrt: generally, in Perl 6, things have different meanings if you write them together or if you write them apart. 11:59
brrt: like in %h<foo>, the '<' introduces a hash lookup. but in %h < 5, it's just the regular comparison op. 12:00
12:00 xinming joined
tadzik r: my $h = 3; $h < 5 > 6 12:02
p6eval rakudo 4373f0: ( no output )
tadzik r: my $h = { 5 => sub { say $^a + 1 } }; $h < 5 > 6 12:03
p6eval rakudo 4373f0: ( no output )
tadzik r: my $h = { 5 => sub { say $^a + 1 } }; $h<5> 6
p6eval rakudo 4373f0: OUTPUT«===SORRY!===␤Confused␤at /tmp/sJj5bRUmgm:1␤»
tadzik r: my $h = { 5 => sub { say $^a + 1 } }; $h <5> 6
p6eval rakudo 4373f0: ( no output )
cognominal_ read Pheromonally instead of Phenomenally in masak statement
pmichaud good morning, #perl6 12:04
moritz good am, pm 12:05
Juerd I'll keep it in mind for tomorrow :)
(afternoon here)
Good localtime
tadzik heh, in au example, are those regexes even Regular Expressions? :) 12:06
pmichaud moritz: (rt #112216) What my last comment was trying to say (however badly) is that I don't have an opinion what the (consistent/inconsistent) behavior should be... only that I wouldn't make any further changes until we had agreement and spectests. :) 12:08
moritz pmichaud: ok :-) 12:09
pmichaud: my opinion is that we should be consistent, one way or another. And currently we aren't.
pmichaud the current behavior is consistent with what I've been told in the past, even if it's not self-consistent with other parts of the Perl 6 spec :)
au! o/ 12:12
12:14 PacoAir joined
brrt hmm i have a case of sigil confusion... if I have declared an array attribute on an object, what sigil do i use to access it? 12:19
$ works, btw, but it seems... wrong
moritz brrt: use @ 12:20
pmichaud brrt: if you declared the attribute as @xyz, it should be accessed via @xyz
moritz brrt: thing is, attributes are named @!foo (or also @foo, though I find it confusing). If you write 'has @.foo', the attribute is still called @!foo
brrt: but the form with the dot also gives you an accessor, which is a very normal method 12:21
and $.foo is short for $( self.foo ), just as @.foo is short for @( self.foo )
tadzik TIL
moritz though rakudo ignores the sigils for the method calls
niecza doesn't, I think
tadzik r: class A { method foo { return 5 }; method bar { @.foo } }; say A.new.foo.perl 12:22
p6eval rakudo 4373f0: OUTPUT«5␤»
tadzik r: class A { method foo { return 5 }; method bar { @.foo } }; say A.new.bar.perl
p6eval rakudo 4373f0: OUTPUT«5␤»
moritz p6: class A { has @.a; method sayit { .say for $.a } }; A.new(a => (2, 3, 4)).sayit
p6eval pugs, rakudo 4373f0: OUTPUT«2␤3␤4␤»
..niecza v15-6-gefda208: OUTPUT«2 3 4␤»
moritz niecza++ is correct here
tadzik indeed 12:23
masak tadzik: if you think about it, it makes perfect sense? what's a public attribute? it's a private attribute with an (auto-generated) accessor method. so we just conflate the syntax of methods and public attributes.
pmichaud does that have a rakudobug ?
12:23 tarch_ joined, tarch left
masak pmichaud: hm, I think so. 12:23
I think I submat it last time we tulk about it.
tadzik masak: Sure it is, but I'm amused nonetheless :) 12:24
pmichaud that should be really easy to fix
since we already handle $( ) and @( )
12:24 tarch_ left
moritz is @( foo ) short for foo.list or for list(foo) ? 12:24
pmichaud foo.list
masak pmichaud: rt.perl.org/rt3//Public/Bug/Displa...?id=111914
pmichaud masak: that bug seems different than this one 12:25
masak yeah, I'm thinking the same. 12:26
masak submits this one
12:26 brrt left 12:27 jaldhar left, brrt joined
brrt thanks :-) from outside the object, obviously there is no way to access it w/o the method 12:27
moritz ... unless you go meta
or cheat
brrt www.youtube.com/watch?v=AGBpIqROW9c 12:28
masak rt.perl.org/rt3/Ticket/Display.html?id=112236
brrt daskfj
pmichaud masak++
brrt oh, it was not my intention to post that, i'm sorry 12:29
i thought my client was still disconnected
moritz any students here who would like to get paid for hacking on rakudo this summer? 12:31
brrt oh yes, me me :-)
pmichaud paid? students? how do I become a student then? ;-)
moritz pmichaud: enroll at your university of choice
brrt although i probably do not have nearly the required skills
masak aw, don't say that. 12:32
moritz brrt: there's the Google Summer of Code project
masak I don't speak for anyone else here, but personally I get lots done out of the depths of my incompetence :P
brrt i'm aware :-) i was actually hoping to implement bioperl6
masak oh, you're that guy!
brrt the same 12:33
masak brrt++
moritz brrt: the Parrot folks are offering to do some Rakudo projects through the parrot foundation, so if you're interested, talk to to the #parrot folks on irc.perl.org
specifically whiteknight
pmichaud in my case, a lot of what I've accomplished was based on suspending any idea of what could(n't) be achieved :)
moritz brrt: and do it quick, because the student application deadline is Friday
brrt :-) thanks, i will
masak pmichaud: "in order to clear this level, you have to disregard all obstacles" :D 12:34
moritz what does &.foo mean, precisely? 12:35
masak hm. 12:36
by extension, it means &( self.foo ) :P
moritz and what does &( TERM ) mean?
masak is such a big help
I have no idea. 12:37
pmichaud perhaps not the correct answer, but S13 has:
The expected semantics of C<&.()> is that of a type coercion which may
or may not create a new object.
I think that's something different though.
masak it is.
moritz pmichaud: in that case, &.() is a short name for postcircumfix:<( )>
masak it's a short form of postcircumfix:<( )>
tadzik to it's a long, not a short ;) 12:38
pmichaud S12 has:
&.foo(1,2,3); # calls self.foo under & context
&.foo.(1,2,3); # &.foo().(1,2,3), call .() on closure returned by .foo
which is what masak++ said :)
and equally helpful, at that. :)
moritz and what the heck does "under & context" mean?
masak it means under the context of something callable. 12:39
I mean, "I don't know"... :)
moritz masak.helpfulness == p6l.helpfulness # at least right now
pmichaud S03-operators mentions &() without any description of what it does :) 12:40
moritz tries to patch $.foo to DWIS 12:41
12:43 liuchong left 12:46 netstar left
pmichaud afk for a while 12:46
moritz patch compiles. Good.
[Coke] sees pmichaud and au almost have a conversation on irc. 12:47
moritz and seems to do what it should. Great
[Coke] *boggle*
PerlJam Coke: Is that like a "crossing the streams" moment? :) 12:49
masak [Coke]: that can't have been the first time, though. pm was doing PGE at the time Pugs broke into the physical universe.
PerlJam (gutenTag btw) 12:50
masak moritz: in a slightly more helpful vein, wouldn't it be enough to go as-strict-as-possible and say that only callable thingies (trivially) cast to callable, and everything else is an error?
moritz more like recrosssing the streams
masak at least until we know what the heck &(42) means. 12:51
PerlJam masak: +1 12:52
12:52 jferrero left
moritz ok, what's the method call for that? 12:53
$ => item, % => hash, @ => list, & => code? 12:54
masak wfm
they're all 4lw, which is cute.
PerlJam indeed 12:55
masak wants a t-shirt with those four pairs
12:55 mucker left
moritz "and I heard you said that code is just a four-letter word" -- Joan Baez. Very nearly. 12:55
au rehi pmichaud! \o </timeDilatedConversation>
12:56 Araq joined
PerlJam YAPC 2008 (?) in Chicago had t-shirts with stars from the Chicago flag replaced with $ @ % & IIRC 12:56
Adding the method names seems like fodder for a future YAPC :) 12:57
12:57 jferrero joined
moritz eeks. Lots of fallout from the patch. 12:58
n: class A { has %.h; method x { %.h<a> } }; say A.new(h => {a => 1 }) 13:00
p6eval niecza v15-6-gefda208: OUTPUT«A.new(...)␤»
moritz n: class A { has %.h; method x { %.h<a> } }; say A.new(h => {a => 1 }).x
p6eval niecza v15-6-gefda208: OUTPUT«1␤»
13:04 benabik left
masak r: say Date.new("2012-02-25") 13:05
p6eval rakudo 4373f0: OUTPUT«2012-02-25␤»
masak r: say Date("2012-02-25")
p6eval rakudo 4373f0: OUTPUT«Method 'Date' not found for invocant of class 'Str'␤ in <anon> at src/gen/BOOTSTRAP.pm:800␤ in any <anon> at src/gen/BOOTSTRAP.pm:797␤ in block <anon> at /tmp/q7LAlDjKnF:1␤␤»
masak r: .say for Date.new("2012-02-25") .. Date.new("2012-03-02")
p6eval rakudo 4373f0: OUTPUT«2012-02-25␤2012-02-26␤2012-02-27␤2012-02-28␤2012-02-29␤2012-03-01␤2012-03-02␤»
masak r: .say for Date.new("2011-02-25") .. Date.new("2011-03-02") 13:06
p6eval rakudo 4373f0: OUTPUT«2011-02-25␤2011-02-26␤2011-02-27␤2011-02-28␤2011-03-01␤2011-03-02␤»
masak \o/
moritz oh, I see why my patch has this much fallout 13:10
$.foo desugaring to self.foo.item
masak yes?
moritz think about rw accessors
masak oh!
:/
moritz oh.
masak wouldn't the other forms, like @.foo, suffer from this as well? 13:11
moritz aye
masak because sometimes you want to @.foo = 1, 2, 3
except no-one ever does that.
moritz n: class A { has $.x; method set { $.x = 5 } }; given A.new { .set; say .x } 13:13
p6eval niecza v15-6-gefda208: OUTPUT«5␤»
masak waitaminnit. 13:14
moritz n: class A { has $.x; method set { $.x.item = 5 } }; given A.new { .set; say .x }
p6eval niecza v15-6-gefda208: OUTPUT«5␤»
masak that's a non-rw attribute. 13:15
moritz masak: niecza simply doesn't do ro/decontainerization
13:15 agentzh left
masak ah. 13:15
moritz iirc it's even in the feature matrix
so, design problem.
I can see two solutions 13:16
1) $.foo doesn't desugar to $.foo.item, but rather to some itemize primitive that doesn't affect rw-ness. Same for % and @ 13:17
2) methods .item on Any and .list on Positional / .hash on Associative declare their invocant as rw, and their return value too
for efficiency reasons I'd prefer 1 13:18
maybe we can extend that to @(...) in general
and if you use @(...) on something that isn't a Positional container, it could still call .list 13:19
whereas $(...) could always itemize without requiring a method call
13:20 thou joined
PerlJam fwiw, #1 is how I imagined such things would work in general. 13:21
moritz hm, it might be that the $( ... ) -> ....item translation is indeed a rakudoism 13:23
dalek kudo/nom: f607966 | moritz++ | tools/build/check-versions.pl:
[build] recommend --with-parrot with absolute paths, because the build system does not like relative paths
 13:24
kudo/accessor-sigils: 0082372 | moritz++ | src/Perl6/Actions.pm:
@.foo, $.foo and %.foo now enforce context

Note that this breaks rw attributes
kudo/accessor-sigils: 915e0da | moritz++ | src/core/ (2 files):
be more careful with sigils in the setting
moritz when in doubt, branch
13:27 sftp joined
moritz branch, fork and spoon :-) 13:29
masak :) 13:33
moritz++ 13:34
13:37 thou left
gianlucadv moritz: I would like to ask you about irclog.perlgeek.de/ 13:39
moritz gianlucadv: sure. Feel free to /msg me if it's not related to this channel in particular
13:42 thou joined 13:47 _ilbot left, _ilbot joined 13:51 agentzh joined 13:52 thou left, agentzh left 14:04 wolfman2000 left 14:14 benabik joined 14:31 snearch left 14:40 JimmyZ joined, mcero left 14:44 plutoid left 14:46 benabik left, benabik joined 14:48 wtw left
[Coke] p: say (1,2,3).pick(); 14:50
p6eval pugs: OUTPUT«2␤»
14:53 derrazo joined
masak r: say (1,2,3).roll(); 14:54
p6eval rakudo 4373f0: OUTPUT«2␤»
masak p: say (1,2,3).roll();
p6eval pugs: OUTPUT«*** No such method in class Array: "&roll"␤ at /tmp/_N8HlqN5AA line 1, column 5-19␤»
masak gets an itch in his programming fingers
moritz p: say (1,2,3).pick(5, :repl);
p6eval pugs: OUTPUT«*** No compatible multi variant found: "&pick"␤ at /tmp/9I63TFPJHH line 1, column 5-27␤»
moritz p: say (1,2,3).pick(5, :rep);
p6eval pugs: OUTPUT«*** No compatible multi variant found: "&pick"␤ at /tmp/QFO4ASpe45 line 1, column 5-26␤»
masak I mean, how hard could it be, to just add a .roll method? :)
colomon p: say (1,2,3).pick(5, :replace)
p6eval pugs: OUTPUT«*** No compatible multi variant found: "&pick"␤ at /tmp/UhrIPeY7P4 line 1, column 5 - line 2, column 1␤»
masak could be that Pugs predates that named parameter. 14:55
colomon either that or it had a different name then. 15:00
[Coke] masak: I just fudged that spec test for you. have fun. ;) 15:04
You'll want to implement "so" first, though.
dalek ast: 30d3be9 | coke++ | S32- (2 files):
pugs fudge
 15:05
moritz [Coke]: I had a branch for 'so'; did you try that?
dalek gs.hs: e5911bf | coke++ | t/spectest.data:
run (fudged) tests
[Coke] moritz: ISTR it failed to work.
moritz (or am I misremembering stuff?)
:(
[Coke] I can try it again. momentito.
perl6: say so 3 15:06
p6eval pugs: OUTPUT«*** No such subroutine: "&so"␤ at /tmp/jnJDVKvLCj line 1, column 5 - line 2, column 1␤»
..rakudo 4373f0, niecza v15-6-gefda208: OUTPUT«True␤»
[Coke] urk. I cannot build your so branch on feather. 15:07
Preprocessing executable 'pugs' for Pugs-6.2.13.20120203...
<command line>: cannot satisfy -package-id FindBin-0.0.5-c1291d79cbf3e04ef6f8304a7717f332:
doing a clean, a cabal update, and a rebuild, now I'm getting pages of things like: 15:09
package terminfo-0.3.2.3 is broken due to missing package
15:09 birdwindupbird left
masak :/ 15:09
still work to do on getting a working dev environment for Pugs, in other words. 15:10
[Coke] haven't done a build in weeks. I assume, perhaps, that ghc was updated.
moritz I get the same error on master
15:10 fsergot joined, spider-mario joined
fsergot hi o/ 15:10
:)
[Coke] yah. I reinstalled, e.g. coke@feather:~/sandbox/Pugs.hs/Pugs$ cabal install text-0.11.1.13
.. and now that error is gone from the build. 15:11
jlaire '
[Coke] wishes for a "cabal rebuild everything" 15:18
moritz [Coke]: you're right, the 'so' branch doesn't work :/ 15:20
masak fsergocie! \o/ 15:21
[Coke] moritz: sorry, man.
I can't build master or so on feather for pugs. 15:25
No instance for (MonadIO (InputT IO))
moritz I did a git clean -xdf 15:26
rm -rf ~/.ghc/
and then cabal install 'mtl < 2'
then typed 'make', and then cabal install <all the packages it complained about>
masak I recognize that error. I think that's what I got in my attempts to get a working dev env on feather. 15:27
jlaire \ 15:29
[Coke] so, how did our cabals get screwed up?
15:32 fglock left 15:39 mucker joined
[Coke] blowing away ~/.ghc helped. 15:40
build succeeds now.
masak \o/ 15:42
15:47 Psyche^ joined 15:51 Patterner left, Psyche^ is now known as Patterner 15:53 kaleem left 15:55 jferrero left 15:57 derrazo left 15:58 jferrero joined 16:00 JimmyZ left 16:07 brrt left 16:14 wolfman2000 joined 16:20 DAA2WSX joined, VAAD1KD joined 16:24 fglock joined 16:27 vlixes joined 16:40 pernatiy left 16:47 tokuhirom joined 16:54 dakkar left 17:02 PZt joined 17:11 tokuhirom left, tokuhirom joined
timotimo are there any tools for profiling in rakudo? i've got some code that runs much faster than i feel it should 17:12
runs much slower*
benabik There's a profiling run core in Parrot, but I'm not familiar with how to use it and don't know if it works well with Rakudo. 17:13
17:14 lichtkind joined 17:15 tokuhirom left
timotimo should i be expecting slowness if i create many short-lived instances of a class? the class in question is my XYZ class that does a little bit of vector arithmetic. i use it as key into hashes a lot. 17:16
it's an immutable class so any arithmetic on it would create new instances, sometimes instances that will be thrown away instantly 17:17
17:19 Chillance joined
lichtkind any clue why thies doesnt work locally? 17:20
phenny lichtkind: 08:51Z <masak> tell lichtkind ask on p6l about $=POD vs $=pod. TheDamian hangs out there, and he's likely to reply.
lichtkind r: say '\\'
p6eval rakudo 4373f0: OUTPUT«\␤»
lichtkind k
geekosaur works for me locally... but if you;re using perl6 -e, be aware of shell quoting shenanigans 17:21
specifically, the naive: perl6 -e "say '\\'" # will fail because the shell interprets \\ inside the doublequotes, so perl6 receives: say '\' 17:22
lichtkind ah
thanks
\\\\ worked :)
17:24 gianlucadv left
colomon So, is someone going to implement Notch's DCPU-16 in Perl 6? 17:28
0x10c.com/doc/dcpu-16.txt
;) 17:29
17:29 derrazo joined
benabik Wait, wait... Mars Effect is a real game? 17:32
17:32 wolfman2000 left
[Coke] benabik: as opposed to Mass Effect ? 17:33
benabik [Coke]: @notch posted that information about a game called "Mars Effect" on 4/1, with the note that the name was carefully chosen to avoid any lawsuits like over Scrolls.
I hadn't heard that it's actually a game. Although, obviously, not called Mars Effect 17:34
derrazo doo hoo hoo 17:53
it's like what 17:54
0x10c now?
benabik derrazo: Yup
17:55 mucker left 17:57 simcop2387 left 17:59 skids joined 18:00 simcop2387 joined 18:01 havenn joined 18:04 spaceships joined 18:06 fhelmberger left
spider-mario r: constant $x = m/42/ 18:10
p6eval rakudo 4373f0: OUTPUT«===SORRY!===␤Method 'match' not found for invocant of class 'Mu'␤»
spider-mario r: constant $x = /42/
p6eval rakudo 4373f0: ( no output )
spider-mario why does this happen?
benabik r: $_ = 'test'; say m/es/ 18:11
p6eval rakudo 4373f0: OUTPUT«=> <es>␤␤»
benabik r: $_ = 'test'; say /es/
p6eval rakudo 4373f0: OUTPUT«_block1011␤»
spider-mario p6: constant $x = m/42/
PerlJam the first one tries to match against $_ and store the result in $x. The second one just stores the pattern in $x
p6eval niecza v15-6-gefda208: OUTPUT«Unhandled exception: Unable to resolve method match in class Any␤ at /tmp/VGKyKPUmD7 line 1 ($x init @ 2) ␤ at <unknown> line 0 (ExitRunloop @ 0) ␤ at /home/p6eval/niecza/src/NieczaBackendDotnet.pm6 line 0 (downcall @ 1) ␤ at /home/p6eval/niecza/src/Nie…
..pugs: OUTPUT«Error eval perl5: "if (!$INC{'Pugs/Runtime/Match/HsBridge.pm'}) {␤ unshift @INC, '/home/p6eval/.cabal/share/Pugs-6.2.13.20120203/blib6/pugs/perl5/lib';␤ eval q[require 'Pugs/Runtime/Match/HsBridge.pm'] or die $@;␤}␤'Pugs::Runtime::Match::HsBridge'␤"␤*** Can't locate P…
..rakudo 4373f0: OUTPUT«===SORRY!===␤Method 'match' not found for invocant of class 'Mu'␤»
spider-mario oh, ok
thanks
18:11 derrazo left
PerlJam r: m/42/ 18:11
p6eval rakudo 4373f0: OUTPUT«Method 'match' not found for invocant of class 'Any'␤ in block <anon> at /tmp/oMhkcncqcx:1␤␤»
PerlJam that's interesting. 18:12
PerlJam wonders where the divide is between Any and Mu there
spider-mario r: given 'there is a 42 in here' {m/42/; $/.perl.say}
p6eval rakudo 4373f0: OUTPUT«Match.new(orig => "there is a 42 in here", from => 11, to => 13, ast => Any, list => ().list, hash => EnumMap.new())␤»
spider-mario nice
PerlJam (How did $_ become either an Any or a Mu?)
spider-mario r: given 'there is a 42 in here' {m/42/.perl.say} 18:13
p6eval rakudo 4373f0: OUTPUT«Match.new(orig => "there is a 42 in here", from => 11, to => 13, ast => Any, list => ().list, hash => EnumMap.new())␤»
masak good eve, #perl6. 18:16
lichtkind hai hai
spider-mario o/
masak はいはい 18:17
spider-mario .ui coi
masak wow, Lojban. 18:20
18:20 mishin__ left
spider-mario I’d like to actually be able to speak in it 18:21
I barely know a few constructs and words 18:22
masak I hear it's possible to become fluent in it, though not that common.
masak settles down for the evening, with a beer and Ch. 82 of HPMoR
spider-mario what is your native language, btw? 18:23
(if I may ask)
masak phenny: "Svenska"?
phenny masak: "Swedish" (sv to en, translate.google.com)
spider-mario ok :)
I didn’t know about phenny, what is it? 18:24
masak bot that keeps messages for us, and knows Unicode, and translates using GT.
spider-mario oh :)
phenny: "merci pour l’explication"? 18:25
phenny spider-mario: "thank you for the explanation" (fr to en, translate.google.com)
masak phenny: tell spider-mario if you ever need to leave a message, this is how to do it. it's triggered with your next utterance on the channel.
phenny masak: I'll pass that on when spider-mario is around.
spider-mario great :D
phenny spider-mario: 18:25Z <masak> tell spider-mario if you ever need to leave a message, this is how to do it. it's triggered with your next utterance on the channel.
masak .u smiling
phenny U+263A WHITE SMILING FACE (☺)
spider-mario .u pile
phenny spider-mario: Sorry, no results for 'pile'.
spider-mario and yet there is www.fileformat.info/info/unicode/ch.../index.htm 18:26
masak yeah, phenny doesn't have the latest silliness from the Unicode Consortium.
also, I don't know why that character is everyone's favorite when there's WOMAN WITH BUNNY EARS.
arnsholt spider-mario: BTW, did I see you mentioning that you're in Rouen? 18:27
spider-mario yes
arnsholt Cool. I lived there, some years back
spider-mario oh :)
arnsholt Almost ten years actually, now that I think about it
18:27 pernatiy joined
lichtkind sir what is your opinion about my latest peace - U+1F4A9 18:28
nuff said :)
masak: any pointer to things i aybe not adressed ? 18:29
18:31 thou joined
[Coke] # 04/04/2012 - rakudo++ ; niecza (95.44%); pugs (38.82%) 18:32
niecza slips a bit, pugs gains. 18:33
benabik Wasn't pugs at like 5% a few weeks ago?
[Coke] gist.github.com/1476841#file_perl6_pass_rates 18:34
# 02/21/2012 - niecza++ ; pugs (14.83%); rakudo (99.04%)
thou o/ everyone 18:35
[Coke] pugs has gone from 3009 passes on 2/21 to 8229 today.
masak lichtkind: I don't understand the context of your question? _all_ the things?
benabik Ah, %15. Still awesome to see its meteoric rise.
lichtkind hi thou
[Coke] benabik: eh. mainly just fudging tests. hop on. ;)
thou hallo, lichtkind. long time no see :-)
lichtkind masak: some dirty corners of perl 6 i might not know 18:36
[Coke] though au++, moritz++, and masak++ have all poked at things to make it do a little more.
lichtkind yes
[Coke] er, whoops. 8276 today
lichtkind masak: we should get camelia into unicode
thou lichtkind: i've been out of perl6 land for a long time, still finding other things are higher on the list right now in my life
[Coke] (pushed)
benabik [Coke]: Brushing off the bit rot from tests is still progress. :-)
18:36 brrt joined
thou but i still love it :-) 18:36
lichtkind thou: yes but i try to cataloge the lang 18:37
[Coke] I imagine I can hit 10K without writing more haskell.
lichtkind thou: im maybe longer there thatn you and when i come back so much has changed 18:40
masak lichtkind: I'll have a look.
lichtkind: by the way, providing a URL helps whenever you want others to look at something on the web ;) 18:41
thou i'm still there (or not fully here, that is). i'm occasionally peeking in to see where things are at.
lichtkind thou: may wanna help (its easy) :) 18:42
masak things are at: awesome, and improving. 18:45
r: $_ = "The answer seems to be 2 + 40"; s[\d+ \s* '+' \s* \d+] = eval ~$/; .say
p6eval rakudo 4373f0: OUTPUT«The answer seems to be 42␤»
masak \o/
flussence how much of pugs is written in plain perl6? 18:46
masak relatively little, IIRC.
most of the parsing stuff and the runtime are in Haskell.
18:47 tyatpi joined, brrt left, brrt joined
flussence wonders if various core bits written as perl6 could be separated out into a standard library that all impls can share 18:48
lichtkind allright my task for today sanitising all the contextualizers
masak lichtkind: you still haven't provided a URL. 18:50
lichtkind wick lacks like an old toilet
masak: for what?
masak lichtkind: you asked me if anything was obviously missing.
lichtkind: I'm trying to access www.perlfoundation.org/perl6/index....ics_tablet but I can't get through.
getting "could not connect to www.perlfoundation.org" 18:52
[Coke] flussence: I think that's a stretch goal. so far whenever we've tried to do that, different niecza & rakudo have implemented different other things, so it's never a straight copy.
writing the perl6 parts that you can in perl6 cross-implementationally would require a lot of effort to figure out which stuff is core. 18:53
lichtkind masak right i thought ti was obvious mom
[Coke] (or CORE::, I suppose)
masak is not lichtkind's mom :)
PerlJam maybe lichtkind has become british? but then it would be "mum", wouldn't it? 18:54
arnsholt flussence: Common Lisp has parts of that I think. Apparently there's one implementation of the LOOP macro, and noone but the author really understands how it works =)
18:55 voraratis joined
masak lichtkind: my point isn't that it's not obvious where the information you want me to look at can be found. my point is that providing a URL makes it unambiguous where it can be found, and saves me work guessing. 18:55
lichtkind: besides, it may make others click through as well.
thirdly, it provides context in the IRC logs.
18:55 voraratis left
lichtkind masak: i understood it from staart i just wrestling for wiki access myself 18:55
[Coke] is this work still happening under auspices of a grant, also? Having a URL would help in that case, too. ;)
18:56 voraratis joined
lichtkind that would be even better :) 18:56
masak lichtkind: why is providing a URL contingent on getting wiki access? don't you know where the stuff you're working with is located?
18:57 voraratis_ joined, voraratis left
lichtkind masak: becasue i was nervously fiddling to save my changes and was not paying full attention to the channel let alone copy the url 18:57
masak ok. 18:59
let me know when you are ready.
lichtkind www.perlfoundation.org/perl6/index....dex_tablet 19:02
plus my precious clipboars was filled with data i had to store first
benabik TPF's site seems to be non-responsive. :-/ 19:03
masak envisions a wild hog in the forest, keeping sheets of paper together
benabik masak++
19:06 tyatpi left
lichtkind the call frame method "line" tells me count of lines or first line of a sub? 19:07
brrt p6eval: say [max] <39 29 10 48> 19:09
masak what method is that? where did you read about it?
brrt: 'r' or 'n' or 'p6'
Araq are there any efforts to provide a luajit backend for perl6?
brrt p6: say [max] <39 28 40 11>
p6eval rakudo 4373f0, niecza v15-6-gefda208: OUTPUT«40␤»
..pugs: OUTPUT«*** ␤ Unexpected "28"␤ expecting operator, ":" or ","␤ at /tmp/esyiVW2dvF line 1, column 15␤»
lichtkind masak: in the synopsis for routines :)
masak r: say [max] <2 11>
p6eval rakudo 4373f0: OUTPUT«2␤» 19:10
masak n: say [max] <2 11>
p6eval niecza v15-6-gefda208: OUTPUT«11␤»
masak inneresting.
masak submits rakudobug
brrt max compares them as strings, seems to be the explanation
masak lichtkind: ok, I'll just find that synopsis myself then, and try to find the reference in question.
brrt: they are strings. 19:11
[Coke] buubot_backup: say 3
buubot_backup [Coke]: Couldn't match input.
masak brrt: actually, they are weird Str/Int hybrids, and the Int aspect should probably win when sorting.
19:12 voraratis_ left
brrt i would hope so 19:12
19:14 wolfman2000 joined
lichtkind masak: feather.perl6.nl/syn/S06.html#The_c..._functions 19:15
masak thank you ;)
brrt p6: say <9 10 200>.sort 19:16
p6eval rakudo 4373f0: OUTPUT«10 200 9␤»
..pugs: OUTPUT«102009␤»
..niecza v15-6-gefda208: OUTPUT«9 10 200␤»
masak lichtkind: I take it as meaning "the current line", where we're running right now.
lichtkind masak: but its underspacced
masak brrt: aye, same problem. 19:17
lichtkind: granted. but you asked what it meant, and that's my interpretation.
lichtkind: I find neither "count of lines" or "first line of a sub" makes much sense.
lichtkind masak: but on the searchlight, this is a case for toady-man
masak: true 19:18
spider-mario the behavior varies between (rakudo, pugs) and niecza 19:19
r and p sort as strings 19:20
(+ pugs does’n insert spaces when it stringifies)
niecza sort them as integers
sorts*
brrt masak: is there a way to convert them to ints
19:20 envi_ left
spider-mario ».Int 19:20
masak p6: say <9 10 200>».Int.sort 19:21
p6eval pugs: OUTPUT«910200␤»
..rakudo 4373f0, niecza v15-6-gefda208: OUTPUT«9 10 200␤»
masak spider-mario++
spider-mario :)
19:21 birdwindupbird joined, wolfman2000 left
masak I still can't determine whether (1) we're in the middle of things, and stuff like this is shining through and making Str/Int duality seem like a really bad idea, or (2) the Str/Int duality is a really bad idea. 19:22
Araq (2)
brrt i'm not sure either 19:23
i usually like it
geekosaur I had the notion we were trying to get away from it, as one of the more annoying p5-isms
having it suddenly come back still seems like something is wrong somewhere 19:24
flussence reminds me of Scalar
masak geekosaur: no, Perl 6 has never been trying to get away from it. the Cool type sort of manifests the fact that we still like auto-casting between various scalar types. 19:25
benabik n: <5 6>[0].WHAT.say
p6eval niecza v15-6-gefda208: OUTPUT«Unhandled exception: Attempted to access slot $!value of type object for IntStr␤ at /home/p6eval/niecza/lib/CORE.setting line 0 (Str.say @ 1) ␤ at /tmp/1t0SRx5NAI line 1 (mainline @ 2) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3842 (ANON @ 3) ␤ at …
[Coke] I would love to see more explanations of things like this next to the spec. "here's why this was added in"
benabik ... That's not what I was expecting.
n: <5 6>».WHAT.say
p6eval niecza v15-6-gefda208: OUTPUT«IntStr() IntStr()␤»
masak geekosaur: but the <1 2 3> syntax, along with command-line arguments, is special in that it tries to "guess" what type things "really" are, besides being ordinary strings. 19:26
flussence: I never understood what Scalar is doing in the spec. I think I've had it explained to me a few times. it's something like a scalar container, analogously with Array or Hash. 19:27
spider-mario I didn’t know there was actually such a thing as IntStr
masak flussence: but it just seems to be that it means a hundred things and nothing whenever it's used in the spec.
brrt p6: <5 6>>>.WHAT.say
masak spider-mario: there isn't. that's just Niecza's name for it.
p6eval rakudo 4373f0: OUTPUT«Parcel()␤»
..niecza v15-6-gefda208: OUTPUT«IntStr() IntStr()␤»
..pugs: OUTPUT«StrStr␤»
spider-mario ok 19:28
masak r: say <5 6>[0].^name
p6eval rakudo 4373f0: OUTPUT«Str␤»
masak r: say <5 6>[0].WHAT
p6eval rakudo 4373f0: OUTPUT«Str()␤»
masak hm.
19:29 NamelessTee joined
masak r: say 5 ~~ Int 19:29
p6eval rakudo 4373f0: OUTPUT«True␤»
masak r: say <5 6>[0] ~~ Int
p6eval rakudo 4373f0: OUTPUT«False␤»
masak huh :)
brrt anyway... you could probably solve by modifying max to convert to int
but what happens then, if you use [max] <9 10 200 foobar> 19:30
spider-mario 'foobar'.Int = 0 19:32
r: <9 10 200 foobar>».Int.perl
p6eval rakudo 4373f0: ( no output )
spider-mario r: <9 10 200 foobar>».Int.perl.say
p6eval rakudo 4373f0: OUTPUT«(9, 10, 200, 0)␤»
masak spider-mario: no, it's a conversion error.
brrt: right. we remind ourselves of this problem once in a while. same with &sort, for example. 19:33
spider-mario not in rakudo, appearently
masak spider-mario: no, rakudo is lagging.
spider-mario n: <9 10 200 foobar>».Int.perl.say
p6eval niecza v15-6-gefda208: OUTPUT«Unhandled exception: Cannot parse number: foobar␤ at /home/p6eval/niecza/lib/CORE.setting line 1366 (die @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3405 (ANON @ 10) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3407 (NumSyntax.str2num @ 4) ␤ a…
spider-mario indeed
brrt p6: { my $input = "9 10 200"; my @data = $input.split(' '); say @data.sort }
spider-mario that’s a good thing
p6eval rakudo 4373f0, niecza v15-6-gefda208: OUTPUT«10 200 9␤»
..pugs: OUTPUT«102009␤»
spider-mario I’d have disliked it being the expected result
masak brrt: converting to a number won't fly, because &sort and [min] and [max] aren't numeric by default. they adapt to the type they're comparing. 19:34
brrt: and if two objects in the list can't be compared, they fail.
spider-mario p6: <9 10 200 foobar>.sort.perl.say
p6eval niecza v15-6-gefda208: OUTPUT«(val("9"), val("10"), val("200"), "foobar").list␤»
..pugs, rakudo 4373f0: OUTPUT«("10", "200", "9", "foobar")␤»
spider-mario p6: <9 10 200 foobar>.max(*.Int).perl.say 19:35
p6eval rakudo 4373f0: OUTPUT«"200"␤»
..pugs: OUTPUT«1797693134862315907729305190789024733617976978942306572734300811577326758055009631327084773224075360211201138798713933576587897688144166224928474306394741243777678934248654852763022196012460941194530829520850057688381506823424628814739131105408272371633505106845862982399472…
..niecza v15-6-gefda208: OUTPUT«Unhandled exception: Cannot parse number: foobar␤ at /home/p6eval/niecza/lib/CORE.setting line 1366 (die @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3405 (ANON @ 10) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3407 (NumSyntax.str2num @ 4) ␤ a…
spider-mario whence pugs’ output?
19:35 havenn left
spider-mario p: 'foobar'.Int 19:35
p6eval pugs: ( no output )
spider-mario p: 'foobar'.Int.say
p6eval pugs: OUTPUT«0␤»
spider-mario I can’t see where it could possibly come from 19:36
p: <9 10 200 foobar>.map(*.Int).perl.say
p6eval pugs: OUTPUT«*** Cannot cast from VList [VInt 1797693134862315907729305190789024733617976978942306572734300811577326758055009631327084773224075360211201138798713933576587897688144166224928474306394741243777678934248654852763022196012460941194530829520850057688381506823424628814739131105…
spider-mario hm
p: <9 10 200 foobar>».Int.perl.say 19:37
p6eval pugs: OUTPUT«(9, 10, 200, 0)␤»
19:37 pernatiy left
spider-mario p: <9 10 200 foobar>.map({.Int}).perl.say 19:37
p6eval pugs: OUTPUT«(9, 10, 200, 0)␤»
benabik p: say *.Int
p6eval pugs: OUTPUT«1797693134862315907729305190789024733617976978942306572734300811577326758055009631327084773224075360211201138798713933576587897688144166224928474306394741243777678934248654852763022196012460941194530829520850057688381506823424628814739131105408272371633505106845862982399472…
spider-mario Whatever causes stranges issues
oh
appearently, it’s as simple as that
.
masak yeah. 19:38
benabik I think pugs doesn't do WhateverCode
masak probably some sort of MAXINT, albeit with arbitrary precision.
Pugs does indeed not do WhateverCode. it predates it.
brrt predates, as in eats? 19:39
spider-mario :D
benabik pre-date is a palm tree
TimToady Araq: diakopter was doing something with a lua backend, but I think he gave up on it 19:42
Araq ok thanks, TimToady 19:43
19:44 pernatiy joined 19:46 tt joined
tt hello 19:46
19:46 tt is now known as Guest2724
Guest2724 hello 19:47
brrt hello
sjohnson 8hello
spider-mario hello
sjohnson oops
i blew it
spider-mario after a non-prime number of occurrences 19:48
:(
(unless mine has come too late)
lichtkind hurray wiki works 19:52
masak: please check just that table www.perlfoundation.org/perl6/index....xtualizers
moritz \o 19:53
19:53 Guest2724 left
lichtkind hai moritz 19:54
19:57 nebuchadnezzar left 19:58 nebuchadnezzar joined
masak lichtkind: it's too bad things don't line up well since it's not a monospaced font. 20:03
lichtkind: &() => code() 20:04
lichtkind im working on formating
that will come s soon
thatnk i missed code()
thanks
thatswhy i asked
masak apart from that, it looks fine. 20:05
lichtkind thank you 20:06
cognominal_ is there a bug filed about the nqp/nqpq shell being broken? 20:11
brrt whats the best mailing list to subscribe to
for dev discussions 20:12
moritz [email@hidden.address]
cognominal_ I answer my own quesiton yes, I filed it code.activestate.com/lists/perl6-compiler/7620/
20:17 xenu left
lichtkind masak: formating is fixed and code to main index added too 20:20
tadzik good evening! 20:23
flussence ooh, I just found www.mono-project.com/Mono_LLVM - anyone tried it on niecza? 20:24
20:24 fglock left 20:25 tyatpi joined
flussence (and I just realised gentoo's mono doesn't seem to have llvm support at all, bah) 20:25
lichtkind tadzik: hai 20:29
20:31 localhost left 20:33 localhost joined
lichtkind hope to scratch the 800 this weak 20:36
tadzik 800?
moritz didn't we already have 1000 this week? 20:37
whatever those numbers mean :-)
20:38 derrazo joined
lichtkind moritz: i mean entires in index A and no we just past 700 5 days ago 20:38
i just wondered if there are 1000 possible 20:39
20:40 spaceships left 20:41 derrazo left, brrt left 20:43 birdwindupbird left 20:44 fsergot left 20:52 tadzik is now known as iDzik
jnthn home 21:04
iDzik hello jnthn
[Coke] jnthn: yay, safe travels. 21:05
jnthn: how was the 80 hour flight? ;)
jnthn [Coke]: It departed 8 hours late.
iDzik ohh
jnthn [Coke]: Then the bus (this is crazy) from the plane to the terminal at Frankfurt that I was on got in an accident. :/
jnthn only got a few cuts
huf that's like a 2minute bus ride, what :/ 21:06
jnthn Right.
How the hell did they screw that up. :/
huf what did the bus hit? another bus? one of those flat towy things?
[Coke] jnthn: YIKES 21:07
jnthn The latter, and it actually managed not to hit it, but braked sharply enough to send everyone who was standing up flying. :/
huf oh :(
jnthn Anyway, I'm basically fine. Just tired, and home quite a bit later than planned. 21:08
21:08 bbkr left, bbkr joined
jnthn $dayjob were kind enough to cancel tomorrow morning's meeting though. :) 21:08
[Coke] \o 21:09
er, \o/
colomon o/
iDzik gosh, glad you're ok 21:10
jnthn Anyways, the rest of the vacation was all very nice. :) 21:12
Once I'm caught up on sleep, I look forward to digging back in to Rakudo. 21:13
21:14 kaare_ left
jnthn ooh, pmichaud++ dropped by with another patch :) 21:16
cognominal_ I can't wait to see you both on track to see rakudo dev back to cruise speed 21:26
masak jnthn! \o/ 21:27
jnthn o/ masak
masak jnthn: I guess you're about to go to sleep. 21:28
jnthn: glad to hear you're safe, and not mortally wounded in the bus accident.
jnthn masak: Will sleep soonish, just having a little snack first. My body thinks it's dinner time... 21:29
masak: You managed to post from the future again on Planetsix :P
cognominal_ :)
masak yeah :/
cognominal_ wish he could send patch from the future 21:30
masak moritz++ suggests I put in a check for this in my blogging tool. I will do this.
cognominal_: all of Perl 6 is a patch from the future ;)
cognominal_ except the future is slow to percolate back into the present. 21:31
jnthn At least we'll be leaving the PAST behind soon... 21:32
21:32 jaldhar joined
masak \o/ 21:33
maybe even tomorrow.
cognominal_ jnthn++
jnthn masak: I think the transition will take more than a day :P
iDzik will we be using FUTURE now? ;)
cognominal_ we will be QAST in the future?
jnthn
.oO( QAST::Lots )
 21:34
masak .oO( the PAST wasn't abandoned in a single day ) 21:36
jnthn
.oO( the PAST was slowly QAST away )
lichtkind r: %h = 1=>2, 3=>4; say %h.end 21:41
p6eval rakudo 4373f0: OUTPUT«===SORRY!===␤Variable %h is not declared␤at /tmp/IBjgGEYRrO:1␤»
lichtkind r: %h = 1=>2, 3=>4; say @(%h).end
p6eval rakudo 4373f0: OUTPUT«===SORRY!===␤Variable %h is not declared␤at /tmp/26cSF_71x6:1␤»
lichtkind r: my %h = 1=>2, 3=>4; say @(%h).end
p6eval rakudo 4373f0: OUTPUT«1␤»
lichtkind r: my %h = 1=>2, 3=>4; say %h.end
p6eval rakudo 4373f0: OUTPUT«1␤»
lichtkind thats strange 21:42
i expected a pair
jnthn r: my %h = 1=>2, 3=>4; say @(%h).perl 21:43
p6eval rakudo 4373f0: OUTPUT«("1" => 2, "3" => 4).list␤»
jnthn r: my @a = <foo bar baz>; say @a.end
p6eval rakudo 4373f0: OUTPUT«2␤»
jnthn lichtkind: .end doesn't give you the item at the end, it gives you the end index
masak I also once thought .end would give you the last item. it doesn't. 21:44
.end is the Perl 6 equiv of $#array
so often it simply corresponds to .elems - 1 21:45
lichtkind yes thathow i wrote it but i was curious how reacts on hashes
was not thinking that clear 21:46
Araq masak: what do you mean "often .elems-1" ?
lichtkind elems is amount of elements of a list
21:46 IngisKahn joined
masak Araq: well, you can declare arrays to have other indexing schemes. 21:47
lichtkind do +@a or @a.elems
TimToady only by using .{} instead of .[]
.[] is always 0-based
masak right.
lichtkind $[ finally died hurayyyyy :) 21:48
i mean in perl 5 too
masak I also don't quite remember how multi-dim arrays declare their .end
TimToady eval: say $[
buubot_backup TimToady: ERROR: Can't locate arybase.pm in @INC (@INC contains:) at (eval 20) line 2. BEGIN failed--compilation aborted at (eval 20) line 2.
masak hopefully in a sensible way :)
21:48 IngisKahn left
TimToady the sensible way might be to refuse to honor .end :) 21:50
masak well, even a multidimensional array has a first dimension. 21:54
21:58 benabik left
TimToady the problem is that a multi-dim array has more than one end 22:03
and multi-dim array subscripts are rather like surreal numbers 22:04
masak I don't immediately see the connection. 22:05
masak .oO( sorear numbers ) 22:06
Araq nothing in computing is like surreal numbers; computing is all about *discrete* math 22:09
TimToady Perl 6 has many indiscretions :) 22:10
TimToady wonders if there will be a better way to write rosettacode.org/wiki/Identity_matrix after S09 is implemented 22:12
spider-mario S09 looks really promising 22:13
TimToady well, other than turning @id[$i][$j] into @id[$i;$j] 22:14
spider-mario I wish I were able to contribute to one of its implementations
(I could say that for a lots of things)
TimToady realistically, it's probably faster to clear a batch of memory to 0's and then set @id[$i;$i] = 1 22:15
Araq TimToady: just make it another built-in operator \\??&56 # produce identity(65)
spider-mario :D
Araq or perhaps |\| 56 22:17
22:17 Trashlord left
Araq that even looks like ascii art for an identity matrix 22:17
masak TimToady: en.wikipedia.org/wiki/Kronecker_delta 22:18
Tene @id .= δ; 22:19
TimToady perl6: .say for ([1, 0 xx 5-1], *.rotate ... *)[^5]
p6eval niecza v15-6-gefda208: OUTPUT«1 0 0 0 0␤0 0 0 0 1␤0 0 0 1 0␤0 0 1 0 0␤0 1 0 0 0␤»
..rakudo 4373f0: OUTPUT«1 0 0 0 0␤0␤0␤0␤0␤»
..pugs: OUTPUT«*** unknown parse error␤ at /tmp/GaIyoJuBrC line 1, column 41␤»
TimToady whoops
perl6: .say for ([1, 0 xx 5-1], *.rotate(-1) ... *)[^5]
masak off-by-some :)
p6eval niecza v15-6-gefda208: OUTPUT«1 0 0 0 0␤0 1 0 0 0␤0 0 1 0 0␤0 0 0 1 0␤0 0 0 0 1␤»
..pugs: OUTPUT«*** unknown parse error␤ at /tmp/_y5HmFAdUO line 1, column 45␤»
..rakudo 4373f0: OUTPUT«1 0 0 0 0␤0␤1␤0␤0␤»
TimToady um... 22:20
rakudobug?
masak seemingly.
masak submits it
TimToady perl6: .say for ([1, 0 xx 5-1], *.rotate(-1).Array ... *)[^5] 22:21
p6eval niecza v15-6-gefda208: OUTPUT«Unhandled exception: Unable to resolve method Array in class Parcel␤ at /tmp/fjGCQhWYb3 line 1 (ANON @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3115 (ANON @ 2) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3122 (ANON @ 17) ␤ at <unknown> line …
..pugs: OUTPUT«*** unknown parse error␤ at /tmp/vYNECUwYYn line 1, column 51␤»
..rakudo 4373f0: OUTPUT«1 0 0 0 0␤0␤1␤0␤0␤»
TimToady perl6: .say for ([1, 0 xx 5-1], { [.rotate(-1)] } ... *)[^5]
p6eval niecza v15-6-gefda208: OUTPUT«===SORRY!===␤␤Method call found where infix expected (change whitespace?) at /tmp/jJEchhwcQ9 line 1:␤------> .say for ([1, 0 xx 5-1], { [⏏.rotate(-1)] } ... *)[^5]␤␤Parse failed␤␤»
..rakudo 4373f0: OUTPUT«1 0 0 0 0␤0 1 0 0 0␤0 0 1 0 0␤0 0 0 1 0␤0 0 0 0 1␤»
..pugs: OUTPUT«*** unknown parse error␤ at /tmp/v7DYQRMOSF line 1, column 50␤»
TimToady huh, niecza is trying to parse it as a reduction 22:22
perl6: .say for ([1, 0 xx 5-1], { [ .rotate(-1) ] } ... *)[^5]
p6eval rakudo 4373f0, niecza v15-6-gefda208: OUTPUT«1 0 0 0 0␤0 1 0 0 0␤0 0 1 0 0␤0 0 0 1 0␤0 0 0 0 1␤»
..pugs: OUTPUT«*** unknown parse error␤ at /tmp/3jJwelWhrj line 1, column 52␤»
masak submits nieczaissue
spider-mario -> $n {.say for ([1, 0 xx $n - 1], {[.rotate: -1]} ... *)[^$n]}.(5)
oops
r: -> $n {.say for ([1, 0 xx $n - 1], {[.rotate: -1]} ... *)[^$n]}.(5)
p6eval rakudo 4373f0: OUTPUT«1 0 0 0 0␤0 1 0 0 0␤0 0 1 0 0␤0 0 0 1 0␤0 0 0 0 1␤»
spider-mario r: -> $n {([1, 0 xx $n - 1], {[.rotate: -1]} ... *)[^$n]}.(5).perl.say 22:23
p6eval rakudo 4373f0: OUTPUT«([1, 0, 0, 0, 0], [0, 1, 0, 0, 0], [0, 0, 1, 0, 0], [0, 0, 0, 1, 0], [0, 0, 0, 0, 1])␤»
TimToady yes, Array.rotate is specced to return Array, not list, so rakudobug 22:24
masak it's now in RT.
rt.perl.org/rt3/Ticket/Display.html?id=112256
TimToady RT is short for rotate, right?
masak :P 22:25
colomon std: .say for ([1, 0 xx 5-1], { [.rotate(-1)] } ... *)[^5]
p6eval std 1ad3292: OUTPUT«===SORRY!===␤Method call found where infix expected (change whitespace?) at /tmp/GMpwlVpWB1 line 1:␤------> .say for ([1, 0 xx 5-1], { [⏏.rotate(-1)] } ... *)[^5]␤ expecting infix or meta-infix␤Parse failed␤FAILED 00:00 43m␤»…
spider-mario std: -> $n {([1, 0 xx $n - 1], {[.rotate: -1]} ... *)[^$n]}.(5).perl.say 22:27
p6eval std 1ad3292: OUTPUT«ok 00:00 45m␤»
spider-mario std: -> $n {([1, 0 xx $n - 1], {[.rotate(-1)]} ... *)[^$n]}.(5).perl.say
p6eval std 1ad3292: OUTPUT«===SORRY!===␤Method call found where infix expected (change whitespace?) at /tmp/wwhCTNvJK_ line 1:␤------> -> $n {([1, 0 xx $n - 1], {[⏏.rotate(-1)]} ... *)[^$n]}.(5).perl.say␤ expecting infix or meta-infix␤Parse failed␤FAILED…
spider-mario oh
masak TimToady: anyway, my ideal way of spec'ing I($n) would be mapping δ($i, $j) with some variant of &map that respected the shape of arrays :)
sub I($n) { shape-map -> $i, $j { δ($i, $j) }, (^$n X ^$n) } 22:29
one can dream.
TimToady that's a bit too APLish for my Poohbrane 22:32
Araq make 'shape-map' a type inspecting macro and it should work
er, sorry, my brain was thinking in another programming language 22:33
masak well, I'm wondering whether the ordinary &map shouldn't be made to respect shape.
22:34 tyatpi_ joined
jnthn -> sleep 22:34
catch y'all tomorrow
TimToady respect in the sense of ignore, maybe 22:35
22:35 wolfman2000 joined
masak respect in the sense of shape(out) === shape(in) 22:35
TimToady that's what I meant :)
masak ok. in maths, we call "ignore" "respect" :P
TimToady but also as in applying the mapping to each leaf regardless of its location 22:36
masak right.
much like »op, I guess. 22:37
timotimo FWIW, perf helped me see what's wrong with my program. it's 10% pool_allocate from Parrot_perl6_box_int_p_i 22:38
TimToady except that it might be more tightly tracked to the actual declared shape
masak anyway. I'll sleep now, and dream of maps over shaped data structures. 'night
TimToady o/
lichtkind buy buy :) 22:39
timotimo 5% from parrot_hash_get_bucket_string and 50% of that is from Parrot_sub_find_pad
22:39 spider-mario left
timotimo not sure what exactly that is, but i believe it has to do with lots of hash accesses? 22:40
dalek d: 3d13d53 | larry++ | STD.pm6:
fix [.rotate] parse problem
 22:43
22:50 NamelessTee left 22:51 vlixes left 22:59 Chillance left 23:08 Araq left 23:30 tokuhirom joined 23:34 envi_ joined
timotimo how come if 0 <= ($x + 1) <= 2 and 0 <= ($y + 1) <= 2 and 0 <= ($z + 1) <= 2 { 23:38
is a thousand times faster than
if $x & $y & $z == -1 | 0 | 1
-1 <= $x & $y & $z <= 1 is equally slow
the difference is really, really, really huge
23:39 fgomez joined 23:43 tyatpi_ left 23:46 tyatpi left
lichtkind i think only jnthn can answer that 23:51
23:55 thou left

Powered by Cro and the Raku® Programming Language.

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

Thank you!