»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! | tinyurl.com/p6contest
Set by moritz_ on 28 December 2010.
diakopter sorear: cool :) (unspace) 00:04
00:05 bacek left
sorear just checking if it works before I use it... 00:06
00:07 shi left
dalek rixel: 9e487ae | diakopter++ | sprixel/ (7 files):
new class/attributes/methods pass prepended/working.

admittedly not kosher Perl 6, but for this extremely tiny strict subset, it'll make writing/emitting efficient code much simpler.		 00:10
00:16 Chillance joined 00:17 jhuni joined, Kovensky left 00:19 masonkramer left 00:20 bacek joined 00:23 rindolf joined
diakopter cool, strongly typed Hash-es "just work" 00:24
(translated to Dictionary<string, T>)
that'll be good enough for now
00:26 gbacon left, jaldhar joined 00:29 pumphaus1 is now known as pumphaus
sorear push to github has stopped working. 00:33
00:36 bacek_ joined
dalek ecza: c9bdf1a | sorear++ | v6/ (2 files):
[v6] Translate final third of Metamodel
 00:36
00:39 bacek left
colomon t/spec/integration/advent2009-day18....get_attr_str() not implemented in class 'Sub' 00:41
sorear: github has been dodgy all days. give it a couple more tries.
(unless you already have, of course.)
Are other people getting the integration/advent2009-day18 error? That's with jnthn's broken patch reversed. 00:42
00:42 gbacon joined
colomon gads, "all day". 00:43
I had about five failed pushes or pulls today.
Repeated efforts eventually worked, but it's still annoying.
00:46 felliott left 00:50 envi joined
sorear niecza: sub foo() { say $*quux }; sub bar($*quux) { foo }; bar(15) 00:50
p6eval niecza v1-35-gc9d17a7: OUTPUT«15␤»
flussence colomon: I've got a clone of roast that fixes the day18 error
it's just a small typo 00:51
colomon where is it?
flussence github.com/flussence/roast
colomon can you send me a patch? ;)
contrawise, do you mind if I just duplicate the patch and give you credit in the commit? 00:52
flussence yeah, sure
00:53 xinming_ joined
dalek ast: 12508e2 | (Solomon Foster)++ | integration/advent2009-day18.t:
Switch dies_ok to eval_dies_ok, flussence++.
 00:54
00:54 felliott joined
colomon Hope that's okay. :) 00:54
flussence no problem :)
00:56 xinming left, felliott left 00:57 rgrau left 01:00 roen joined 01:05 fIorz joined, florz left 01:07 fIorz left, florz joined 01:08 mikehh left
snarkyboojum where does one get viv from these days? 01:22
all links pointing to github.com/perl6/mu/raw/master/src/perl6/viv 404
and hello :) 01:27
01:28 QinGW joined 01:29 s1n left
snarkyboojum ah - found it at github.com/perl6/std :) perl6.org needs updating 01:30
01:38 lamstyle left 01:41 lamstyle joined 01:42 bacek__ joined 01:46 bacek_ left 02:01 noganex_ joined
diakopter my latest push's alert here is .. slow from github 02:02
maybe I should convert it to sorear's push model
02:04 noganex left
dalek rixel: e4063bc | diakopter++ | sprixel/src/ (3 files):
made eval EXPR work inside methods & subs, enabling recursive
 02:07
colomon rakudo: my $a = "This gf gfg a g \t \t gfgf".subst(/\s+/, " "); say $a 02:09
p6eval rakudo de98d1: OUTPUT«This gf gfg a g gfgf␤»
colomon global! 02:10
rakudo: my $a = "This gf gfg a g \t \t gfgf".subst(/\s+/, " ", :g); say $a
p6eval rakudo de98d1: OUTPUT«This gf gfg a g gfgf␤»
02:14 rindolf left
diakopter rakudo: my $a = "This gf \n\n gfg a g \t \t gfgf".subst(/\s+/, " "); say $a 02:15
p6eval rakudo de98d1: OUTPUT«This gf ␤␤ gfg a g gfgf␤»
02:15 bacek_ joined
diakopter rakudo: my $a = "This gf \n\n gfg a g \t \t gfgf".subst(/\s+/, " ", :g); say $a 02:15
p6eval rakudo de98d1: OUTPUT«This gf gfg a g gfgf␤»
diakopter rakudo: my $a = "This gf \n\n gfg a g \t \t gfgf".subst(/\s+/, "", :g); say $a
p6eval rakudo de98d1: OUTPUT«Thisgfgfgaggfgf␤»
diakopter colomon: what about this gf? ;)
colomon gfgag! 02:16
It's funny how often I forget the :g in subst. I almost never want to do just a single substitution.... 02:18
dalek rixel: 48ae9c0 | diakopter++ | / (7 files):
implement the "_cc" special form for current continuation,

be ported back over from the old perlesque (maybe it won't). Also clean up the static/instance method resolution to use the runsharp library better.
02:19 bacek__ left
sorear the old perlesque didn't have proper continuations, only coroutines 02:21
flussence interesting rosettacode thing on /r/perl/ here, I might have a go at it... redd.it/evopg 02:25
diakopter sorear: oh
sorear: but.. I thought it had nearly all the full cloning necessary? 02:26
(maybe you didn't know about it)
sorear diakopter: no, it only cloned one deep 02:29
diakopter hm, oh 02:30
02:36 karb joined
snarkyboojum ah - just realised that installing STD off CPAN obviates the need to know where viv is installed for the purposes of building Niecza :P 02:37
snarkyboojum bangs head against the README he didn't read properly
02:38 masonkramer joined
flussence small question for anyone: what's the simplest way to go from (1,2,3,4,5,6) to (1=>2, 3=>4, 5=>6)? 02:41
02:42 yarp joined
masonkramer same thing 02:42
=> is a way of spelling ,
that happens to quote the LHS if it's a single token
if you mean you want a hash with 1 as the first key and 2 as the first value 02:43
my %h = (1,2,3,4,5,6)
hashes are created by assigning lists to a variable with a % sigil
flussence I'm trying to get an array of Pairs
02:43 jevin joined
masonkramer oh... 02:43
damn..wrong channel
flussence heh
masonkramer sorry man
:)
snarkyboojum rakudo: my @list = (1,2,3,4,5,6); say @list.hash.perl 02:44
p6eval rakudo de98d1: OUTPUT«{"1" => 2, "3" => 4, "5" => 6}␤»
flussence .hash outputs them in order on rakudo, but I don't think I can count on that behaviour
Oh. 02:46
.map(*=>*)
masonkramer [=>] <1 2 3 4 5 6>
colomon oh, bother. just got a Null PMC working on p5. :(
masonkramer perl6: say [=>] <1 2 3 4 5 6> 02:47
p6eval niecza v1-35-gc9d17a7: OUTPUT«===SORRY!===␤Action method prefix_circumfix_meta_operator__S_reduce not yet implemented at /home/p6eval/niecza/src/Niecza/Actions.pm line 55.␤Action method prefix_circumfix_meta_operator not yet implemented at /home/p6eval/niecza/src/Niecza/Actions.pm
..line 55.␤Che…
..pugs, rakudo de98d1: OUTPUT«1 2 3 4 5 6␤»
flussence masonkramer: close, but that's (1=>(2=>...))))))
masonkramer well, anyway, it works
is it?
I'll keep quiet now
02:47 florz left
jasonmay rakudo: (1,2,3,4,5,6).map(* => *).perl 02:47
p6eval rakudo de98d1: ( no output )
flussence interesting answer though :)
snarkyboojum that's cool :) 02:48
rakudo: say (1,2,3,4,5,6).map(*=>*).perl
jasonmay rakudo: (1,2,3,4,5,6).map(* => *).perl.say
oops
p6eval rakudo de98d1: OUTPUT«(1 => 2, 3 => 4, 5 => 6)␤»
snarkyboojum jinx
02:48 florz joined 02:51 bacek_ left 02:55 felliott joined, karb left
snarkyboojum flussence: by my reading of the synopses, it looks like .hash should work in order for a simple list like that 03:00
i.e the "hash contextualizer" section in S29 03:01
03:05 bacek_ joined
lue hello world o/ 03:08
snarkyboojum hi lue o/
colomon \o 03:09
colomon had a beautiful long example work well for masak's p5, but a simple test case blows up my code. 03:10
huh. Actually it seems to be the test which is blowing up. Weird. 03:14
omg 03:15
rakudo: my $a; say ">$a<"; say ">$a<"; 03:16
p6eval rakudo de98d1: OUTPUT«>Any() Any() Any()Any()<␤»
colomon rakudo: my $a = 10; say ">$a<"; say ">$a<";
p6eval rakudo de98d1: OUTPUT«postcircumfix:<{ }> not defined for type Int()␤ in 'Any::join' at line 1␤ in 'Any::join' at line 1462:CORE.setting␤ in 'List::Str' at line 2748:CORE.setting␤ in 'Parcel::Str' at line 1399:CORE.setting␤ in main program body at line 1␤»
colomon and there it is.
flussence wtf
colomon I think it's parsing that as $a<"; say "> 03:17
flussence oh...
colomon std: my $a = 10; say ">$a<"; say ">$a<";
p6eval std 625303c: OUTPUT«===SORRY!===␤Unable to parse quote-words subscript; couldn't find right angle quote at /tmp/BalF0CqATO line 1:␤------> my $a = 10; say ">$a<"; say ">$a<⏏";␤ expecting escape␤Parse failed␤FAILED 00:01 122m␤»
flussence yeah, now I remember masak's post about $var</a> doing that :) 03:18
lue wow. That's... a surprise.
colomon but Rakudo's error is LTA, and when the two say statements are twenty lines apart...
sorear I suggest using qs"" 03:20
colomon will that help?
In practice, I just switch say ">$a<" to say :$a.perl
the important bit was being able to see trailing whitespace.
std: my $a = 10; say qs">$a<"; say qs">$a<"; 03:21
p6eval std 625303c: OUTPUT«===SORRY!===␤Unable to parse quote-words subscript; couldn't find right angle quote at /tmp/5EWzcLrocN line 1:␤------> my $a = 10; say qs">$a<"; say qs">$a<⏏";␤ expecting escape␤Parse failed␤FAILED 00:01 122m␤»
lue std: my $a = 10; say ">{$a}<"; say ">{$a}<";
p6eval std 625303c: OUTPUT«ok 00:01 121m␤»
lue That's a possible workaround. 03:22
flussence I'm starting to have doubts about the usefulness of allowing interpolating <> directly inside "". The error's logical but it's very non-obvious...
lue I would vote for printing what's in the string literally if interpolation fails (e.g. say ">$a<" would come out ">10<\n") 03:23
rakudo: say "$a"
p6eval rakudo de98d1: OUTPUT«===SORRY!===␤Symbol '$a' not predeclared in <anonymous> (/tmp/bzL3932pwT:22)␤»
03:25 kraai_ joined
flussence that's the problem; the line colomon showed *is* a valid interpolation (even if the result is useless to most people) 03:26
lue what, the $a<"; say "> bit? I'd like to cry foul on that one, but I see why it happened. 03:28
03:30 Guest40868 joined
flussence someone could be using a weird hash key there, but there's no way for perl6 to know until it gets to that part of the code, and by then it's already decided how to parse it... 03:31
03:33 jhuni left, gbacon left
lue my $a = 10; say "⟩$a⟨"; say "⟩$a⟨"; 03:34
colomon it's easy enough to work around, but it's a heck of a surprise when it happens.
lue rakudo: my $a = 10; say "⟩$a⟨"; say "⟩$a⟨";
p6eval rakudo de98d1: OUTPUT«⟩10⟨␤⟩10⟨␤»
lue I'd say just use {} around the $a. 03:35
03:35 bacek__ joined
colomon right. 03:35
03:36 Guest40868 left
colomon okay, I have a working algorithm for masak's p5! Needs massive cleaning, though. 03:37
time for bed. 03:38
sorear bye
03:38 bacek_ left 03:39 yarp left 03:42 karb joined 03:44 helv joined 03:55 arlinius|3 joined 04:03 satyavvd joined, bacek__ left 04:12 kraai_ left 04:13 kraai_ joined
lopnor hello, 04:19
is there any performance profiler for rakudo?
04:20 bacek__ joined 04:23 Su-Shee_ joined 04:25 cottoo left, cottoo joined 04:27 Su-Shee left
sorear lopnor: I think there's a profiling runcore 04:33
lopnor sorear: thank you, how do i use it? is there any documents in rakudo-star tarball? 04:41
arlinius|3 dsa 04:42
lopnor hm, parrot-2.11.0/docs/dev/profiling.pod this seems to be 04:43
thank you, i will try this 04:44
04:56 Trashlord joined 05:05 karb left 05:12 roen left 05:19 coldhead left 05:21 xinming_ left 05:22 xinming joined
sorear niecza: sub foo($s) { constant %x = a => 1, b => 2, c => 3; return %x{$s} }; say foo('b') 05:23
p6eval niecza v1-35-gc9d17a7: OUTPUT«Unhandled exception: Cannot use hash access on an object of type Parcel␤ at (eval) line 1 (MAIN foo @ 2)␤ at (eval) line 1 (MAIN mainline @ 1)␤ at /home/p6eval/niecza/lib/SAFE.setting line 1064 (SAFE C381_ANON @ 1)␤ at /home/p6eval/niecza/lib/SAFE.setting line 1064
..(SAFE module…
sorear thought so.
05:25 helv left, roen joined 05:37 mberends joined
sorear hello mberends 05:39
mberends hi sorear, I've just discovered the MacSurf in Copenhagen Central Station ;) 05:40
masak and I plan to meet here today, he has just relocated to the part of Sweden nearest to here :) 05:41
sorear: how is niecza's eval() coming along? It seems once that has landed, we can begin spectesting niecza in earnest. 05:49
05:58 cjk101010 joined 06:02 arlinius|3 left 06:07 j-invariant joined
j-invariant are there any songs about perl6? 06:07
or perl
dalek p-rx: 89a7459 | cotto++ | src/NQP/ (2 files):
add ...
 06:09
p-rx/nom: 98b0b4f | bacek++ | src/ (3 files):
Remove usage of deprecated concat_s_s op
p-rx/nom: 2a73b89 | cotto++ | src/stage0/ (2 files):
fix deprecated concat_s_s usages
06:09 arlinius|2 joined
cottoo ... wasn't intentional 06:10
06:10 cottoo is now known as cotto, cotto left, cotto joined
dalek p-rx: 2f2bc9a | cotto++ | src/NQP/ (2 files):
Revert "add ..."

This reverts commit 89a7459bf15f084f034b4572e3121833f22aa6e3.		 06:11
cotto in nom, is "type object" another way of saying "class"? 06:15
diakopter bacek__: you'd also need to fix nqp-rx's pir emitter; I suspect it emits usages of that concat form 06:16
maybe not. 06:17
bacek__ diakopter, let me check it
cotto that was me trying to cherry-pick nom back into building 06:18
bacek__ but I did update nqp-rx when strip concat_s_s from parrot.
06:18 bacek__ is now known as bacek
cotto nom doesn't bootstrap atm but it does seem to build 06:19
diakopter bacek: why the removal of that concat form?
cotto immutable strings
bacek diakopter, leftover of "mutable string world"
diakopter oh.. 06:20
cotto concat_s_s assumes that you're mutating the first argument
diakopter I guess if you treat a string like a value type it would matter
cotto did nom previously work fine after bootstrap-files?
(for its own special value of "works") 06:21
sorear mberends: I don't know 06:29
mberends: v2 (Jan 31) will almost certainly be self-hosted; I expect eval will require significant rearranging, but I have no idea how significant 06:30
mberends sorear: ok, in that case I'll attempt to achieve some spectest coverage, without creating any high expectations. 06:32
sorear pmurias has been doing some interesting stuff. 06:33
06:35 bacek_ joined 06:38 bacek left
dalek ecza: 7e32c33 | sorear++ | v6/NAMOutput.pm6:
[v6] Translate NAMBackend
 06:52
06:55 wtw joined 06:59 bacek__ joined 07:03 bacek_ left 07:04 masak joined
masak good morning! 07:04
masak stretches, cat-like
sorear niecza: say +[ map { $_, $_ }, 1, 2 ] 07:08
p6eval niecza v1-35-gc9d17a7: OUTPUT«2␤»
sorear gah
07:08 kaare joined 07:09 kaare is now known as Guest88087
dalek ecza: 25fdee0 | sorear++ | lib/SAFE.setting:
Fix &take losing list character of items
 07:12
sorear hello masak. 07:13
masak rakudo: say +[ map { $_, $_ }, 1, 2 ] 07:16
p6eval rakudo de98d1: OUTPUT«4␤»
masak o.O
oh, right.
diakopter rakudo: say ~[ map { $_, $_ }, 1, 2 ] 07:17
p6eval rakudo de98d1: OUTPUT«1 1 2 2␤»
diakopter rakudo: say ~[ map { $_ ** $_ }, 1, 2 ] 07:18
p6eval rakudo de98d1: OUTPUT«1 4␤»
diakopter er
rakudo: say ~[ map { $_ zZ $_ }, 1, 2 ]
p6eval rakudo de98d1: OUTPUT«===SORRY!===␤Unable to parse blockoid, couldn't find final '}' at line 22␤»
diakopter hee
rakudo: say +[ map { $_ xx $_ }, ^99 ] 07:19
p6eval rakudo de98d1: OUTPUT«4851␤»
07:22 agentzh joined
sorear phenny: tell moritz_ p6eval is 46 commits behind niecza head, did I break the build again? D: 07:25
phenny sorear: I'll pass that on when moritz_ is around.
moritz_ sorear: there were local changes, which prevented automatic pulling. Sorry for that, fixed now 07:28
phenny moritz_: 07:25Z <sorear> tell moritz_ p6eval is 46 commits behind niecza head, did I break the build again? D:
moritz_ niecza: say '\o/'
p6eval niecza v1-80-g25fdee0: OUTPUT«\o/␤»
07:28 Guest88087 is now known as kaare_ 07:39 kjeldahl left
diakopter moritz_: who made the local changes.. 07:39
07:41 justatheory left, kjeldahl joined 07:42 mberends left, jaldhar left, jaldhar joined
moritz_ diakopter: me, after pmurias asked me to try something when we had a build failure 07:45
at least I think that's what happened
dalek ecza: e941d29 | sorear++ | v6/ (2 files):
[v6] Start a stopgap test harness
 07:48
ecza: f7b008c | sorear++ | v6/ (3 files):
[v6] More fixes; harness can do "Hello, world" now
sorear very exciting 07:50
masak is it for the spectests?
sorear masak: No, I have the Perl 6 version of Niecza able to run a Hello World program. Sort of. 07:51
masak ah :)
nice to see things progress. 07:52
sorear next will be Niecza::Actions, so I will be able to give input in Perl 6
then the optimization passes
then the driver... hopefully pmurias will have finished the driver refactor by then
then a new build systme 07:53
then ... done
07:54 kanishka joined
masak \o/ 07:56
08:01 mberends joined 08:03 daxim joined 08:04 Su-Shee_ is now known as Su-Shee 08:08 kfo joined 08:11 kfo_ left 08:24 noganex_ is now known as noganex 08:32 Vlavv` left 08:33 Vlavv joined 08:35 DAAYG1R joined
mberends afk & # meeting masak++ 08:38
08:38 mberends left 08:40 kensanata joined 08:43 explorer__ left, dual left, knewt2 left, JodaZ left, robinsmidsrod left, aloha left, sjohnson left, flatwhatson left, elb0w left, PacoLinux left, sunnavy left, hillu left, toebu left, jerome left, phenny left, nsh_ left, sbp left, barika_ left, oha left, __sri left, DarthGandalf left, wtw left, felipe left, Gothmog_ left, tomaw left, kanishka left, noganex left, stepnem left, aindilis left, Bucciarati left, ggoebel left, agentzh left, jenk left, arlinius left, spq1 left, IllvilJa left, literal left, j-invariant left, satyavvd left, wagnerf left, larsen_ left, VXZ left, Sarten-X left, Tene left, zorgnax left, daxim left, lamstyle left, Helios left, jmmills left, estrabd left, shabble left, patch left, jaldhar left, bacek__ left, felliott left, jevin left, envi left, c9s_ left, Patterner left, mtk left, plobsing left, roen left, yves_ left, starcoder left, f00li5h left, p6eval left, nadim left, diakopter left, charsbr_ left, meteorjay left, cibs left, kfo left, kjeldahl left, kst left, meraxes left, jarfhy left, Alias left, pothos left, chitragupt left, Intensity left, cotto left, masonkramer left, jdv79 left, lue left, bartolin left, rhr left, thepler left, kaare_ left, Su-Shee left, tadzik left, cxreg left, dju_ left, PerlJam left, Tedd1 left, Gahrian left, mtve left, avuserow left, pochi left, TimToady left, fith left, simon left, arlinius|2 left, moritz_ left, takesako left, sol1tude left, rjbs left, colomon left, jedai left, arthur-_ left, ponbiki left, drbean left, szbalint left, puge left, kensanata left, cjk101010 left, pumphaus left, hcchien left, shachaf left, c1sung left, shortcircuit left, doddo left, hatseflats left, arnsholt left, zb left, sECuRE left, cotto_work left, masak left, florz left, QinGW left, Chillance left, jjore_ left, dsp_ left, araujo left, Eevee left, awwaiid left, y3llow left, mathw left, IRSeekBot left, tty234 left, ascent_ left, xinming left, kraai_ left, ruoso left, orafu left, cognominal left, dukeleto left, Lorn left, mdxi left, itz left, vert2_ left, flussence left, kcwu left, edenc left, BinGOs left, krakan_ left, perigrin left, nothingmuch left, snarkyboojum left, _ilbot left, huf left, nrr left, jnthn left, spacebat left, kthakore left, peters_mops left, revdiablo left, Vlavv left, frooh left, [particle] left, Juerd left, sorear left, flatwhatson_ left, cataska left, gfldex left 08:48 gfldex joined, lestrrat joined, cataska joined, flatwhatson_ joined, sorear joined, Juerd joined, [particle] joined, frooh joined, Vlavv joined, DAAYG1R joined, literal joined, IllvilJa joined, spq1 joined, arlinius joined, jenk joined, explorer__ joined, dual joined, knewt2 joined, JodaZ joined, robinsmidsrod joined, aloha joined, sbp joined, sjohnson joined, flatwhatson joined, elb0w joined, PacoLinux joined, hillu joined, sunnavy joined, toebu joined, jerome joined, phenny joined, nsh_ joined, barika_ joined, oha joined, __sri joined, DarthGandalf joined, tomaw joined, Gothmog_ joined, ggoebel joined, felipe joined, Bucciarati joined, aindilis joined, stepnem joined, noganex joined, wtw joined, agentzh joined, kanishka joined, kensanata joined, kfo joined, daxim joined, jaldhar joined, kjeldahl joined, kaare_ joined, bacek__ joined, cotto joined, arlinius|2 joined, j-invariant joined, cjk101010 joined, roen joined, Su-Shee joined, satyavvd joined, felliott joined, florz joined, jevin joined, masonkramer joined, lamstyle joined, QinGW joined, envi joined, Chillance joined, moritz_ joined, patch joined, shabble joined, estrabd joined, jmmills joined, Helios joined, takesako joined, sol1tude joined, c9s_ joined, kst joined, pumphaus joined, rjbs joined, jjore_ joined, tadzik joined, cxreg joined, dju_ joined, Patterner joined, PerlJam joined, mtk joined, meraxes joined, plobsing joined, dsp_ joined, cibs joined, doddo joined, hcchien joined, araujo joined, Tedd1 joined, wagnerf joined, cotto_work joined, sECuRE joined, zb joined, arnsholt joined, hatseflats joined, shortcircuit joined, c1sung joined, shachaf joined, yves_ joined, Gahrian joined, starcoder joined, mtve joined, zorgnax joined, Tene joined, Sarten-X joined, VXZ joined, larsen_ joined, jdv79 joined, Eevee joined, jarfhy joined, colomon joined, awwaiid joined, jedai joined, arthur-_ joined, Alias joined, f00li5h joined, ponbiki joined, simon joined, y3llow joined, pothos joined, avuserow joined, pochi joined, p6eval joined, nadim joined, TimToady joined, diakopter joined, drbean joined, chitragupt joined, mathw joined, lue joined, IRSeekBot joined, charsbr_ joined, tty234 joined, puge joined, leguin.freenode.net sets mode: +v p6eval, bartolin joined, meteorjay joined, fith joined, rhr joined, Intensity joined, thepler joined, ascent_ joined, szbalint joined, clkao joined, pnu joined, LoRe joined, kolibrie_ joined, allbery_b joined, nperez_ joined, ingy joined, cschimm1 joined, sjn joined, hudnix joined, tylerni7 joined, cookys joined, szabgab joined, jrockway joined, jql joined, wolverian joined, Maddingue joined, TiMBuS joined, jasonmay joined, cls_bsd joined, yahooooo joined, PZt joined, broquaint joined, pjcj joined, rokoteko joined, mux joined, zostay joined, Grrrr joined, cosimo joined, spinclad joined, wooden joined, betterworld joined, Exodist joined, HarryS left, kensanata` joined, xinming joined, kraai_ joined, ruoso joined, orafu joined, cognominal joined, dukeleto joined, itz joined, Lorn joined, mdxi joined, vert2_ joined, flussence joined, kcwu joined, frettled joined, saaki joined, jpr5 joined, buubot joined, edenc joined, BinGOs joined, krakan_ joined, perigrin joined, nothingmuch joined, snarkyboojum joined, _ilbot joined, huf joined, nrr joined, jnthn joined, spacebat joined, kthakore joined, peters_mops joined, revdiablo joined, masak joined 08:50 kensanata left, HarryS joined, HarryS is now known as 13WAAYULM
j-invariant perl6: 3 == * 08:59
p6eval pugs, rakudo de98d1, niecza v1-82-gf7b008c: ( no output )
j-invariant perl6: say(3 == *);
08:59 Miguelanxo joined
p6eval rakudo de98d1: OUTPUT«_block124␤» 08:59
..niecza v1-82-gf7b008c: OUTPUT«Sub()<instance>␤»
..pugs: OUTPUT«␤»
moritz_ rakudo: if ?(Q:PIR { %r = box 3␤}) { say "alive" } 09:02
p6eval rakudo de98d1: OUTPUT«alive␤»
moritz_ rakudo: if (Q:PIR { %r = box 3␤}).Bool { say "alive" }
p6eval rakudo de98d1: OUTPUT«alive␤»
j-invariant rakudo: say(3 == *); 09:03
p6eval rakudo de98d1: OUTPUT«_block124␤»
moritz_ rakudo: say (3 == *).(5) 09:04
p6eval rakudo de98d1: OUTPUT«Bool::False␤»
moritz_ rakudo: say (3 == *).(3)
p6eval rakudo de98d1: OUTPUT«Bool::True␤»
j-invariant is there something equal to everything? 09:05
moritz_ no 09:06
but there's something that smart-matches True against everything
rakudo: say 0 ~~ True
p6eval rakudo de98d1: OUTPUT«===SORRY!===␤Smartmatch against True always matches; if you mean to test the topic for truthiness, use :so or *.so or ?* instead at line 22, near " True"␤»
moritz_ ... and warns :-) 09:07
rakudo: say 0 ~~ Mu
j-invariant cool!
p6eval rakudo de98d1: OUTPUT«1␤»
09:10 dakkar joined 09:24 roen left 09:31 molaf joined 09:32 Kovensky joined 09:46 bacek_ joined 09:47 yarp joined 09:49 bacek__ left 10:16 kensanata` is now known as kensanata, kensanata left, kensanata joined, sol1tude left 10:20 wamba joined
colomon masak: Do you have an interface in mind for p5? Something like take two lines and long for their longest substring? 10:27
moritz_ colomon: see base-test 10:28
iirc it's exactly that
colomon oh. Is there a base-test in every directory that I've been ignoring?
moritz_ seems like :-) 10:29
10:29 tzhs joined
colomon "Look at that," as my little guy would say. :) 10:29
moritz_++
moritz_ cleverl little guy, I'd say :-) 10:30
10:30 QinGW left, yarp left
colomon you can tell when he is excited by something new, because he'll say it three or four times a minute. :) 10:31
moritz_ probably because those stupid adults fail to show the proper excitement after the first time 10:33
10:57 bacek_ left 11:02 bacek_ joined 11:06 yarp joined 11:09 muixirt joined
muixirt hi colomon 11:09
colomon o/ 11:10
muixirt why did you revert jnthn patch?
11:10 IllvilJa left
moritz_ it caused spectest failures 11:12
muixirt rakudo is quite buggy as a hole, and tests like t/spec/S14-roles/parameter-subtyping.rakudo where broken before
moritz_ which jnthn++ couldn't test, because parrot was b0rked at that point
muixirt moritz_: is the pir :main fixed? Or gets fixed in next days? 11:13
moritz_ muixirt: it's "fixed", which means the message disappears. Instead now the main body of a Perl 6 program is executed twice. 11:14
muixirt didn't see anything in rakudo / Commit History
moritz_ parrot changed
see yesterday's #parrot logs 11:15
search for pmichaud++ appearing
muixirt too bad to see a performance increase slip away
ok moritz_
colomon I'm sure jnthn will get around to fixing his patch soon, and we'll get it back. 11:16
It just wasn't quite ready for prime time yet.
11:16 kanishka left
muixirt colomon: good to hear 11:16
I don't want to harp on it to much but rakudo perl could *really* use some speed enhancements 11:18
it's heartbreaking slow and it's annoying that parrot get often in the way 11:19
sorry for stating the obvious 11:21
11:23 Util joined
Util First cut of Rakudo binary package for Mac OS X: 11:30
s3.datasaw.com/Rakudo_Star_2010-12_...ment_1.dmg
Testers wanted!
s/Rakudo binary/Rakudo Star binary/ 11:34
11:35 oha left 11:36 j-invariant left 11:37 lamstyle left
colomon Good grief, my masak code is failing the base tests miserably. 11:42
11:43 gbacon joined 11:47 lamstyle joined
colomon okay, p3 fixed. 11:55
11:59 roen joined
Util p3 was my favorite 11:59
p4, not so much 12:00
arnsholt p3 had the advantage of being pretty simple 12:01
colomon I really am kind of disliking the base-test tests.
particularly in p4.
arnsholt Yeah. The ones for p3 are pretty normal, but p1 and p4 are weird
Haven't looked much at p2 or p5 12:02
colomon I'm grumpy about having to add a bunch of error checking which has nothing to do with the problem in case of grossly illegal input.
arnsholt I'm pretty sloppy with my input validation, TBH 12:03
colomon But I can't begin to fathom why base-test insists the computer makes a certain move in situations where every move is exactly equivalent.
arnsholt Yeah, those tests I really don't like 12:04
I think I'll just ignore base-test for p4
12:10 kanishka joined
muixirt moritz_: do all nowspectest pass ? 12:12
t/spec/S14-roles/parameter-subtyping.rakudo hangs again
12:13 satyavvd left
colomon muixirt: context? It works fine for me with latest Rakudo and its default Parrot. 12:13
muixirt Rakudo Perl 6, version 2010.12-16-gde98d15 built on parrot 2.11.0 RELEASE_2_11_0-659-gf1c0156 12:14
colomon Platform? 12:15
muixirt linux amd64
lots of other test fails
hence the question if the spectest is supposed to pass with latest parrot/rakudo or if I oberlooked something again 12:16
colomon yeah, that test runs fine with those versions on OS X. 12:17
I'm building on 64-bit linux now.
muixirt lots of 'More than one plan found in TAP output' 12:18
neat: # Looks like you planned 21 tests, but ran 42 12:19
well moritz_ said: Instead now the main body of a Perl 6 program is executed twice. 12:20
He meant it! :-)
<zum mäuse melken> 12:21
12:22 Eevee left, phenny left 12:23 Eevee joined, nsh_ left, sbp left 12:24 sbp joined
colomon oh, I'm not using that version. hmmm? 12:24
This is Rakudo Perl 6, version 2010.12-16-gde98d15 built on parrot 2.11.0 RELEASE_2_11_0-478-gd69dbbc
Guess that's not the latest Parrot?
12:25 agentzh left, nsh joined 12:27 yarp left
muixirt where do these gde98d15 gibberish come from again? 12:27
gfldex a missing git pull maybe? 12:28
colomon It's g plus the first 7 characters of the hash tag for the git commit.
gfldex kill parrot_install while you are on it
colomon gd69dbbc is what current Rakudo calls for. 12:29
12:29 masonkramer left 12:30 masonkramer joined, bacek_ left
muixirt according to github.com/parrot/parrot f1c015663e153d35a528 is the latest commit 12:30
gfldex check build/PARROT_REVISION 12:31
in rakudo
colomon right, Rakudo's not on the latest Parrot yet.
12:31 nsh left
muixirt ok, I don't do --gen-parrot, sorry for the confusion 12:32
12:32 nsh joined
colomon muixirt: right, it appears the correct answer is that Rakudo does not yet work with the latest Parrot. 12:35
12:35 avuserow left
muixirt colomon: so rakudo works and spectests passes with d69dbbc? 12:35
colomon yes
muixirt ok, thanks colomon 12:36
12:36 _twitch joined 12:37 avuserow joined
colomon rakudo: my @a = "4,5'.eval; say @a.perl 12:40
p6eval rakudo de98d1: OUTPUT«===SORRY!===␤Confused at line 22, near "my @a = \"4"␤»
colomon rakudo: my @a = "4,5".eval; say @a.perl
p6eval rakudo de98d1: OUTPUT«[4, 5]␤»
moritz_ trac.parrot.org/parrot/ticket/1932 is my bug report for rakudo on current parrot 12:41
12:42 kanishka left, nsh left 12:44 sbp left, mtk left 12:45 [sbp] joined 12:46 mtk joined 12:48 sbp joined
muixirt "... and can't be easily reverted. " Erm, what? 12:48
colomon hmmm? 12:49
12:49 phenny joined 12:50 nsh joined
muixirt colomon: see the comment of bacek to moritz_ 's ticket 12:50
colomon ah.
muixirt we should request from the parrot devs that it executes main twice in to seperate threads :-) 12:56
moritz_ threads are deprecated :/ 12:57
12:58 yarp joined
moritz_ but I don't understand... there's a merge commit, that means it can't have been --no-ff 12:58
but maybe we're talking about different branches
muixirt or we could sell rakudo to the pentagon: SAFE programming through redundant computation 12:59
12:59 Miguelanxo_ joined, spacebat left 13:01 takadonet joined 13:02 Miguelanxo left 13:03 kraai_ left
takadonet morning all 13:04
tadzik morning takadonet
colomon o/
13:05 Miguelanxo_ left
wagnerf hi all, is there a way to profile some code with rakudo ? 13:15
moritz_ yes, add 'say time' statements to it and parse / analyze the output
colomon wagnerf: There is a Parrot profiler, last I heard it was working with Rakudo. But it's tricky to use.
moritz_ no automized way though
wagnerf ok 13:16
13:18 MayDaniel joined
muixirt i like perl6 --stagestats 13:18
moritz_ but it doesn't give profiling information
muixirt if you wonder how long it takes to parse your program 13:19
colomon now has all but one base-test passing from p1-p4, and he'd argue the failing test is wrong. 13:21
muixirt moritz_: but i would argue that a profiler for rakudo isn't very helpful for rakudo perl *users*
moritz_ colomon: which one do you think is wrong?
13:22 sftp joined
wagnerf well, maybe it's a good opportunity to try playing with the meta object protocol :) 13:23
colomon moritz_: Test 5 on p4. All first moves are exactly equivalent, but it assumes the computer will always choose 0, 1
moritz_ colomon: I've argued with masak over that one too. He meant I was free to ignore that test 13:24
colomon: and then I started to implement the solution, and found that it was easier for me if I special-cased the first move anyway :-)
13:25 [Coke] joined
moritz_ because after the first move, I can ignore the periodic boundary conditions, which suited me well 13:25
colomon moritz_: huh. I didn't special case the first move.
And what if the human player doesn't do something to break the periodic boundary? 13:26
moritz_ then I do a small coordinate transformation
aka .rotate()
[Coke] yawns. 13:27
arnsholt The first move in p4 is irrelevant anyways =) 13:30
13:33 tzhs left 13:35 V15170R joined
V15170R hi :) 13:35
i'm a newbie programmer, and i was reading about MONADS.... and learned about roles on perl6. 13:36
takadonet V15170R: welcome
V15170R i don't understand monads completely, yet, but it seems that they are "similar" in the way they both "compose" behavoir.... 13:37
are roles inspired in some way in "monads" ?
13:38 MayDaniel left 13:39 arlinius|2 left 13:43 yarp left
shortcircuit One thing I really, really like about the Perl 6 community over the Perl community; it's not a sniping gallery in here. 13:44
tadzik what do you mean?
13:46 smash joined
smash hello everyone 13:46
shortcircuit My last couple experiences in #perl have been, in reverse order, responding to someone's suggestion that a code example be fixed, and (a year or so ago) being treated like a n00b (in a negative way) as I was trying to get the JSON module working on a broken operating system. 13:47
The more recent event had a guy who wanted to see the code fixed, a guy who didn't care, except to see the site go away, and a guy who provided a nopaste with corrected code. 13:48
I've not seen non-constructive aggressiveness around here, and I've been treated subjectively well when I haven't known what I'm doing.
I like that about this place. :) 13:49
muixirt I second that, but of course #perl6 is very small community 13:50
daxim I can troll you hard to restore the universal balance and natural order in the perliverse
13:52 yarp joined
daxim that reminds me, why is that reddit novelty account that used to come up during the last year not active anymore? :( was fun 13:53
13:57 vastak joined 14:06 V15170R left 14:11 Miguelanxo joined 14:12 Miguelanxo left 14:17 satyavvd joined 14:19 yarp left 14:23 zorgnax left 14:28 risou joined 14:29 fhelmberger joined 14:30 fhelmberger left, pmurias joined, fhelmberger joined
colomon Okay, I've got all the base-tests passing on all problems, except for that one on p4. 14:31
14:36 cjk101010 left 14:45 felliott left 14:48 risou left, molaf left 14:53 kaare_ left 14:54 wamba left, wamba joined 14:58 pyrimidine joined
takadonet pyrimidine: hey 14:58
pyrimidine o/
takadonet long time no see
pyrimidine took a well-needed break 14:59
14:59 khisanth__ joined 15:00 khisanth__ is now known as Khisanth
takadonet good to hear 15:01
15:05 WonTu joined 15:06 WonTu left 15:08 drbean left 15:09 orafu left, orafu joined 15:13 drbean joined 15:17 hercynium joined 15:20 MayDaniel joined 15:24 wtw left 15:25 kaare_ joined, florz left 15:26 florz joined 15:27 tzhs joined 15:31 IllvilJa joined 15:38 PerlPilot joined 15:39 frodwith joined
PerlJam good * #perl6 15:39
15:42 Kovensky left
colomon o/ 15:42
15:48 felliott joined 15:55 Kovensky joined 15:56 orafu left, orafu joined 15:58 plobsing left 16:09 gottreu joined 16:11 PerlJam left 16:24 roen left, _twitch left 16:28 tzhs left 16:33 kanishka joined, molaf joined 16:36 muixirt left 16:42 silug joined, shi joined 16:49 daxim left 16:51 satyavvd left 16:54 ggoebel left 16:59 ggoebel joined 17:02 plobsing joined, c1sung left 17:03 cogno joined 17:04 justatheory joined 17:07 cdarroch joined, cdarroch left, cdarroch joined 17:09 c1sung joined 17:13 jaldhar left 17:16 c1sung left 17:23 kent\n joined 17:24 c1sung joined
kent\n I'm looking for "resumable exceptions", but the synopsis code is a bit daunting. Essentially, I have a place where it may be useful to report a warning, but it may also be a good place to throw an exception, and I don't want to care which it is, and just throw an exception, and let the calling context decide "nah, thats alright, keep going" and return to statement line after the exception was thrown. 17:27
flussence try "fail()"
17:28 c1sung left
kent\n pseudocode: try { x ; fail(); y } catch { $_->resume() } # x, fail(), resume(), y 17:28
flussence rakudo: sub a { fail('foo') }; try { my $v = a(); CATCH { say 'fail' } }; my $w = a(); say 'alive'; say $v; 17:29
p6eval rakudo de98d1: OUTPUT«===SORRY!===␤Symbol '$v' not predeclared in <anonymous> (/tmp/gvnnsF3Rjs:22)␤»
flussence rakudo: sub a { fail('foo') }; try { my $v = a(); CATCH { say 'fail' } }; my $w = a(); say 'alive'; say $w;
p6eval rakudo de98d1: OUTPUT«alive␤»
pmurias kent\n: warnings are implemented as "resumable exceptions" 17:30
flussence hm, someone else probably knows it better than I do
kent\n code sample is doing my head in -_- 17:31
kent\n hasn't mastered p6 yet, its just something I concluded/realised p5 will probably never handle and want to see how/if p6 does it in a way I can understand
17:34 PerlPilot is now known as PerlJam
PerlJam no one has mastered Perl 6 yet. 17:34
kent\n I shall rephrase: I haven't really yet written anything in p6 yet =p 17:35
pmurias kent\n: something like $!.resume() will resume the exception
kent\n awesome. 17:36
17:37 kensanata left
kent\n toys with the idea of writing his module in p6 for the lulz 17:37
TimToady well, if you write it as a warning, it will default to resuming, and you can write a CONTROL exception catcher to kill it
kent\n Ah!. the light brightens. 17:38
TimToady All the parts of Perl 6 are very well designed, except for the parts that aren't... 17:39
PerlJam some parts have sharp edges (don't cut yourself!) Other parts have fractal edges. some parts are tooled to a fine precision using electron microscopes; other parts were marked with chalk and cut with an axe. :-) 17:43
oh, and these parts I've just described are not all mutually exclusive. 17:45
kent\n bets some of those edges are imaginary, or only exist in certain dimensional planes. 17:46
TimToady rakudo: say 42i.Bool 17:47
p6eval rakudo de98d1: OUTPUT«Bool::True␤»
flussence rakudo: say (now R- now).perl # was just playing around in the REPL, found the numbers amusingly weird for some reason
p6eval rakudo de98d1: OUTPUT«Duration.new(397/1809)␤»
17:48 M_o_C joined
Util Many things that were dull edges in Perl 5 are now tight dovetails in Perl 6. I plan a "dovetail" category when I start blogging. 17:49
VXZ rakudo: say i**.5 17:50
p6eval rakudo de98d1: OUTPUT«0.707106781186548 + 0.707106781186547i␤»
flussence rakudo: say (0.18547485973194).Rat.perl # oh wow, didn't know this worked
p6eval rakudo de98d1: OUTPUT«166/895␤»
TimToady rakudo: say pi.WHAT
p6eval rakudo de98d1: OUTPUT«Num()␤»
TimToady rakudo: say pi.Rat.perl
p6eval rakudo de98d1: OUTPUT«355/113␤»
17:51 cogno left
TimToady most of the sharp edges in Perl 6 are out there where in Perl 5 you're falling off the edge of the world 17:51
flussence
.oO( maybe "pi.perl" should output a continued fraction sequence )
 17:52
moritz_ rakudo: say pi.Rat(1e-6).perl
p6eval rakudo de98d1: OUTPUT«355/113␤»
moritz_ rakudo: say pi.Rat(1e-8).perl
p6eval rakudo de98d1: OUTPUT«103993/33102␤»
TimToady my @pi-digits := ... 17:53
Util flussence: a continued fraction object type would be interesting. Module welcome :)
17:53 envi left
[Coke] p6eval: say pi.Rat 17:54
TimToady you broked it
moritz_ rakudo: say pi.Rat
p6eval rakudo de98d1: OUTPUT«3.14159292035398␤» 17:55
PerlJam interesting
TimToady seems like one could get a better Rat of pi, even in 32 bits 17:56
[Coke] didn't like to be addressed by name. I was trying to /msg it anyway.
Util (Repeating announcement from this morning)
Mac OS X users needed for testing of Rakudo Star binary package (first cut) : s3.datasaw.com/Rakudo_Star_2010-12_...ment_1.dmg
moritz_ TimToady: .Rat uses a predefined precision in rakudo
rakudo: say pi.Rat(1e-10) 17:57
p6eval rakudo de98d1: OUTPUT«3.14159265361894␤»
moritz_ rakudo: say pi.Rat(1e-10).perl
p6eval rakudo de98d1: OUTPUT«312689/99532␤»
moritz_ rakudo: say pi.Rat(1e-12).perl
p6eval rakudo de98d1: OUTPUT«4272943/1360120␤»
TimToady the default seems rather lossy 17:58
PerlJam Any reason we don't truncate the digits beyond the Rat's precision?
17:58 impious joined
moritz_ PerlJam: the NYI reason, probably 17:59
PerlJam okie
TimToady also, with 64-bit denom we could pretty much convert most Nums directly to a Rat
and I still think lazy gcd is a good idea
there's no reason to do gcd when adding up a bunch of, say, dollars and cents. 18:00
you just want everything to be /100
though we should probably support Rat100 etc. types anyway, unless they're called Fix100 18:01
but it's kinda convenient that powers of ten don't intersect powers of two
18:02 c1sung joined
TimToady but probably it's just a role, anyway: Fix100 := Fix[1/100] 18:02
VXZ Where do I report problems with that os x binary mentioned above? 18:03
moritz_ VXZ: talk to Util
TimToady if it's a role, we can also have fixed points larger than 1, like Fix[1_000_000_000] # gdp scale :) 18:04
18:04 c1sung left 18:05 jaldhar joined, c1sung joined
VXZ Util: trying that os x binary...dyld complaining about not being able to find /sw/lib/libgmp.3.dylib...let me know what I can do to be more informative 18:06
arnsholt VXZ: Do you have fink installed? 18:07
kent\n are there any static-builds of rakudo that can just be downloaded and run? waiting for compile and getting it to work is such a barrier to entry
VXZ arnsholt: Nope. Didn't know it was needed. Perhaps the install doc should mention it?
kent\n ( for .nix that is ) 18:08
arnsholt VXZ: It shouldn't be a requirement =)
But it could be that Util is using it and missed a dependency 18:09
18:09 hanekomu joined
PerlJam kent\n: I think you can get .deb and .rpm packages 18:09
moritz_ kent\n: the rakudo homepage has links to some linux binary packages
kent\n PerlJam: I was just wanting a tar.bz2/tar.gz I could untar in ~/ and have it JustWork
18:09 c1sung left
Util kent\n: We recognized exactly that point in a #phasers meeting last month. This OS X binary is the first of several platforms; good to know that at least one person would want it on Linux too. 18:10
moritz_ rakudo and parrot aren't relocatable
Util VXZ: Thanks! That is just the kind of info I was looking for!
VXZ Util: Np. :) Yeah, it's a nearly-pristine install of 10.6. So...I don't even have a /sw 18:11
18:11 c1sung joined
Util moritz_: I have partially solved the relocation problem, at least on OS X. 18:11
18:11 wamba left
arnsholt goes to try the OS X binary as well 18:11
I'm pretty sure I have a libgmp, but in /opt/local and not /sw (since I use MacPorts) 18:12
Util arnsholt: You may need to ln -s /opt/local/lib/libgmp.3.dylib /sw/lib/libgmp.3.dylib 18:13
arnsholt Yeah, probably =)
colomon #phasers in 46 minutes? 18:14
arnsholt You're using fink or somesuch, right?
18:15 hanekomu left
kent\n wonders how to get -I lib # equivlaent in rakudo 18:15
arnsholt Util: Symlinking gave me a new error
Util A mix of fink and manual installs, most dated from before I started keeping notes.
kent\n 'man rakudo' # very unhelpful :(
moritz_ kent\n: contributions for a better manapge are welcome 18:16
arnsholt It finds the lib, but it's the wrong architecture
Your perl6 is i386 and my lib is x64
Util kent\n: BEGIN { @*INC.push("the/path/of/my/lib") }; 18:17
18:17 jaldhar left
Util (Not a permanent solution; `use lib` not implemented yet) 18:18
arnsholt There's also PERL6LIB=/path/to/stuff perl6
Util arnsholt: OK, so I must either bundle libgmp, or tell Parrot to build as if I lack it. 18:19
18:19 MayDaniel left
kent\n Util: thanks. 18:20
arnsholt Util: Yeah, I think so 18:21
Util I want to bundle, but creating the scaffolding to allow it will be complicated. I will rebuild without it for now, and add it back in later. 18:23
18:23 c1sung left 18:24 kanishka left 18:25 kanishka joined
kent\n Is there anything like Path::Class qw( dir file ) + MooseX::Types::Path::Class in P6 yet? 18:29
[Coke] Util: relocation problem - did you fix the "really linked against build dir" problem too? 18:30
18:32 plobsing left 18:33 dakkar left
[Coke] which syn defines Num? 18:33
colomon S32 18:35
what do you need to know?
Util [Coke]: `install_name_tool` lets me change the installed exes to look for the dylibs at their new locations. Probably not the long-term solution, but it works for now.
TimToady num is basically just a C double 18:36
Num is just the boxed version of that
[Coke] Util: ah, that's the trick that the port uses.
Util [Coke]: Which port is that?
[Coke] TimToady: ROODE is asking why Num includes .Rat(); trying to find a Syn to point him at.
Util: ... the macport
Util thx
TimToady and the Num api is in S32-setting-library/Numeric.pod, presumably 18:37
[Coke] colomon: S32 is the setting. Is the setting "core" ? (is the part of core you can write in perl6?)
TimToady we don't use the word "core" much 18:38
colomon I'm not quite sure what you mean by that.
Most of the Num operations are a direct call into Parrot.
TimToady the setting is the namespace that is outside your compilation unit
[Coke] I don't see the rat function in there, just the Rat type.
TimToady and I mean lexical namespace when I say that
18:39 Juerd left
[Coke] by core I mean: anything calling itself perl6 will have to have it. 18:39
18:39 Juerd joined
TimToady all classes are automatically recognized as conversion functions/methods 18:39
the precision to Rat() is likely undocumented 18:40
[Coke] ah, so that's really a cast between types?
(lossy)
18:40 pmichaud joined
TimToady yes 18:40
colomon Yes.
method Rat(Real $epsilon = 1.0e-6)
18:41 fhelmberger left
pmichaud is #phasers still at 18:30 utc? 18:41
[Coke] Thanks.
TimToady epsilon? doesn't seem very infinitesimal...
pmichaud or is it 19:30 utc?
18:41 cogno joined
[Coke] I predict his followup question will be: "Why is Rat a builtin type?". ;) 18:41
moritz_ pmichaud: 19:00 utc iirc 18:42
pmichaud moritz_++ # thanks
moritz_ ie in ~18min
pmichaud so, #phasers in 18
colomon yes 18:43
Util For future reference, the /topic in #phasers shows the meeting time.
pmichaud D'oh! Util++ 18:44
I rarely see the topics... but yes, it was there when I re-joined the chan.
colomon I'm always wondering how UTC corresponds to local time. ;) 18:45
[Coke] ... wierd. google comp.lang.parrot calendar, I'm not seeing
pmichaud one of my previous jobs meant doing nearly all of my work in utc, so I've learned to keep track of both :-) 18:46
Util You can do a bare /topic to force the topic to display as if it were a message; I have to do that to see long topics.
moritz_ colomon: do you know of date --utc ? 18:48
colomon doesn't work on the mac
though if I remembered it, I suppose I could always log into one of the linux boxes to check.
except I think their time is drifting again. :\
18:51 mtk left
pmichaud I often use "TZ=UTC0 date" 18:51
works in places where --utc isn't available
PerlJam date -u ?
18:51 mtk joined, rpr_ joined
PerlJam (where is that not available?) 18:51
colomon ah, date -u works 18:52
(on OS X)
TimToady the terminal should just highlight utc times with a popupable local time :) 18:53
like it does with already with URIs 18:54
18:56 stkowski joined
diakopter TimToady: have you thought any further about my whine for signature aliases? I realize you said "so I'd just go with Callable[:()] for now", but what I was asking for was a way to alias *those* signatures/annotations to a much shorter symbol. 19:02
TimToady all types can be aliased with subset
diakopter oh 19:03
TimToady #phasers, btw
diakopter I didn't understand that from irclog.perlgeek.de/perl6/2011-01-03#i_3145363
oh
19:04 estrabd is now known as estrabd_reading 19:05 c1sung joined
diakopter TimToady: but now I get it, usthinks 19:09
er, usthink
19:10 c1sung left 19:18 shi left 19:25 cogno left 19:27 dual left 19:31 Chillance left
masak colomon: skipping "that test" in p4's base-test is OK. especially if passing it would sacrifice elegance. 19:31
colomon: in fact, you're free to disregard the base-test files altogether... they're just there to make sure there is *some* common ground between my vision of the problem and people's solutions. 19:32
19:32 bacek joined 19:33 dual joined
arnsholt masak: I hope the secret test cases for p4 are bit cleverer though 19:34
If not, evaluating the submissions is gonna be a pain =)
masak s/If not, // 19:35
:)
arnsholt I suppose
colomon \o/
masak arnsholt: I would object, but then I'd have to explain why it ain't necessarily so. and I don't want to spoil the fun.
19:36 plobsing joined
arnsholt masak: Then we are two ;) 19:36
masak I'm glad to hear colomon, arnsholt and Util are all working on solutions. 19:37
I'm slightly *surprised* to see you discussing your solutions in the backlog. but I forget that you are trusting, non-conspiratory people :) 19:38
arnsholt "Working on a solution" might be overselling a bit, but I've done the research necessary for work at least =)
masak arnsholt: ok. you realize that the deadline is in 10 days, right?
arnsholt Heh. It's hard not to discuss it for me. I quite like discussing this kind of thing
colomon +1 19:39
arnsholt Yeah, it pains me greatly that I don't have time to complete all the tasks =)
Especially p4 was fun discovering some cleverness for
masak yeah, but don't you like to sneak in the best solution, win the books, and THEN talk about it? :P
arnsholt Well, that would be fun. But I've been spending most of the christmas holidays doing absolutely nothing, so I don't have the time to do that =) 19:40
19:41 mtk left
arnsholt I'm pretty sure I'm out on number of completed tasks 19:41
Tene masak: For me, and I expect several others, if I wanted books, I'd just buy them. A competition like yours is fun in itself, rather more than just the value of books.
colomon +1
PerlJam Tene++
arnsholt Yeah, it's mostly the chance to do stuff with Perl 6
colomon Well, I wouldn't mind a few books. ;) 19:42
arnsholt True, true. I did say mostly =D
Tene masak: as well, only one person will enjoy winning, but everyone can enjoy talking about it together.
masak :)
colomon But I'm doing it for the craic.
arnsholt colomon: craic?
Tene So if I were trying to optimize my expected utility, I'd anticipate more incremental expected fun from a conversation than from trying to keep secrets.
colomon arnsholt: en.wikipedia.org/wiki/Craic
pmichaud I even thought about contributing to increase the size of masak++'s prize.... or to allow masak++ to declare a second winner that I'd fund
masak I'm going to host a context next year. and I'll spend all year thinking of a way to get you guys to be more secretive and avaracious :P 19:43
pmichaud since I'm not planning to compete :)
arnsholt masak: I suggest mind controlling space satelites 19:44
colomon arnsholt: normally I'd spell it crack, but I thought that would just cause more confusion.
Tene masak: Next year you should solicit contributions to the reward.
I'd have contributed.
TimToady always enters contexts
pmichaud maybe I'll just offer to fund an "honorable mention" prize if masak++ would like to award one
pmichaud always loses contexts
masak arnsholt: how come, it didn't help this year. er. I mean "interesting idea". heh.
19:44 stkowski left
colomon is context-free 19:44
19:44 stkowski joined
arnsholt colomon: Nifty concept. Never encountered that before 19:45
masak pmichaud: I think that might become very much needed. we have 18 contestant, someone is bound to be honourable and mentionworthy. :)
Tene masak: speaking of optimizing expected utility, HP:MoR is updating again.
masak colomon: maybe you're already on craic.
pmichaud masak: plan on being able to award a second prize then. you can also decide the amount
masak Tene: \o/
pmichaud (presumably it would be for less than the first prize :)
er 19:46
rephrase
masak pmichaud: :)
pmichaud (presumably it would not be greater than the first prize)
(it could be ==)
masak depends on the honourableness, I guess.
Tene I'm rather entertained by greater reward for second place. 19:47
dalek tpfwiki: (Herbert Breunung)++ | www.perlfoundation.org/perl6/index....tor_tablet
Tene Greater reward for second place would rather *encourage* sharing and cooperation.
Which I rather like. 19:48
masak I missed #phasers today, because a bus driver thought I said "middle of nowhere", not "$home".
Tene: think of the first price as the second price, then. :P
flussence masak: that's because $HOME is case-sensitive :P 19:49
flussence ducks
masak s:g/price/prize/
d'oh!
I should have shouted at him. I'll do that next time I see him.
arnsholt This channel is pleasingly random 19:52
pmichaud $home would've worked in csh 19:53
but it's probably the only thing that works in csh :)
Su-Shee good evening everyone. 19:54
pmichaud Su-Shee! o/
colomon \o
19:55 mtk joined
Util Now configured without gmp, pcre, opengl, zlib, gettext, icu, libffi, or readline : 19:55
s3.datasaw.com.s3.amazonaws.com/Rak...ment_2.dmg
masak Su-Shee: \o 19:57
tadzik ~~
Su-Shee ~~ should have been called surf operator.. it looks like riding the wave.. ;) 19:58
tadzik :)
masak the pair-o'-worms operator. 20:00
20:01 estrabd_reading is now known as estrabd
Su-Shee wiggle op would also have been nice, because you could say something like "just wiggle it!" 20:02
arnsholt Util: That looks good to me
otool -L show no weird libs being referenced as well
masak Su-Shee: from now on, I will tell people to just wiggle it. 20:03
PerlJam Su-Shee: except that I call it "squiggle" 20:04
;)
Su-Shee works as verb as well :)
20:05 plobsing left
Util arnsholt: thanks! 20:08
20:09 c1sung joined 20:10 shi joined 20:13 c1sung left 20:17 stkowski left, stkowski joined 20:19 gabhi joined, molaf left, tylercurtis joined 20:22 coldhead joined
tylercurtis Hi, #perl6. 20:24
20:25 amkrankruleuen left
colomon o/ 20:25
20:26 amkrankruleuen joined
TimToady I kinda think of it as the "aloha" operator 20:26
tadzik tylercurtis: o/
snarkyboojum: it's Niecza, not Niezca (twitter) :) 20:28
20:29 y3llow_ joined, pothos_ joined
masak tylercurtis: hi! 20:29
20:30 kanishka1 joined, y3llow left, y3llow_ is now known as y3llow, pothos left, pothos_ is now known as pothos 20:31 kanishka left 20:35 gabhi left 20:36 c1sung joined 20:40 c1sung left 20:47 c1sung joined 20:54 c1sung left 20:59 MayDaniel joined 21:00 c1sung joined 21:01 vastak left 21:03 plobsing joined 21:05 dukeleto left, dukeleto joined 21:06 bacek left 21:08 takadonet left, c1sung left
VXZ Util: that one works good here too...any tests or anything I should run? 21:16
21:18 V15170R joined 21:19 bacek joined 21:21 c1sung joined
Util VXZ: thanks for testing! I am not looking for any particular functional tests yet, just "it runs on my system" and "I don't understand X, so explain X better" 21:22
21:25 lue left
VXZ Util: nope, seems to work fine. Follows the usual mac install process and packaging format and everything, ought to be easy for anyone to figure out. Maybe put an alias to perl6 with a pretty icon in the root? *shrug* 21:26
21:29 bacek_ joined 21:30 c1sung left
Util VXZ: The "pretty icon" is in the works; Camelia is modelling for icon and for the background of the root folder. 21:32
I am not sure what you mean by "alias to perl6", though. Can you describe it more precisely?
21:33 bacek left
VXZ An alias in os x is like a shortcut in windows. 21:34
perl6 being bin/perl6
[Coke] Util: via the finder instead of the shell. 21:36
Util Do you mean a symlink?
[Coke] symlink is for the shell. Alias is for the Finder.
I don't particularly care for aliases, myself. not sure if it's because I'm using using the command line. 21:37
VXZ I agree, I'm just looking for ways to make it more friendly for the mac gui people.
Util OK, let's approce this from the other direction. What is this supposed to *do*? Double-click on it, and it opens a Terminal session with the $PATH set correctly? I am just fuzzy on the goal of the suggestion. 21:38
s/approce/approch/
VXZ Also, you could change the structure of the folder to be a .app (with is just a folder with a special bit set and a particular format that gives os x some ideas of what to run when you click on it, which file types it should be associated with, etc)
Trying to make it easier for less-knowledgable people to play with it, or for gui-inclined people to integrate it into their desktop and workflow. 21:39
arnsholt While we're at it, I'd suggest changing the _ in the folder name to a space as well
VXZ And yeah, I was thinking it could just launch perl6...you can already double-click on it directly in the finder, and it opens a terminal for you automatically, nothing to do there. 21:41
Util I am considering .app format, just because it provides auto-support of bundled libgmp, etc. I had rejected .app previously because it is *not* an app; it has no sensible base function. I will look harder at this; I see that it is what users would expect, at the least. 21:42
arnsholt: Noted, thanks! # underscore in name 21:43
VXZ Util: plus, then, you can do the filetype associations, in case a script doesn't have a shebang, I think 21:44
21:45 alester joined
Util VXZ: filetype associations; noted, thanks! 21:47
21:49 tylercurtis left
VXZ Pretty sure .app allows you to do some other stuff, too, but not sure what. Not a big mac fan. I just own one. 21:49
Util: btw, thanks for doing this! Anything that makes p6 more accessible to more people is, in my mind, a Good Thing(tm). 21:50
21:51 Patterner left 21:53 Psyche^ joined, Psyche^ is now known as Patterner 21:59 M_o_C left 22:06 lamstyle left 22:09 amkrankruleuen left, amkrankruleuen joined 22:16 shi left 22:17 c1sung joined 22:19 bacek__ joined 22:22 lamstyle joined, bacek_ left 22:26 Franek joined 22:27 drake1 joined
drake1 can I please have those video presentation links about perl6 again? 22:28
they were fairly good
22:29 shi joined 22:30 c1sung left
masak drake1: sorry, we only hand out one video presentation link per customer. 22:31
drake1 oh
masak if you lose it, you have to wait until our next Special Offer.
drake1 I thought they functioned like ads
masak we like to keep scarcity in the picture. 22:32
drake1 fine
extremely feature rich
PerlJam drake1: they're on perl6.org aren't they?
masak drake1: which video presentations do you mean? szabgab's?
PerlJam masak: there are *more* than just szabgab's? 22:33
drake1 the one about argument aignatures
signatures
masak PerlJam: ssh! let me handle the PR here :P
drake1 *giggle*
masak drake1: szabgab.com/blog/2010/08/perl6-subr...ators.html 22:34
22:34 molaf joined
masak der 22:34
drake1 great masak
masak drake1: don't listen to PerlJam, there's lots and lots of Perl 6 instruction videos! :)
we have to ask people to slow down making them.
22:35 stkowski left
PerlJam masak: careful what you wish for :) 22:35
masak hm, true. 22:36
szabgab++ # for making those videos
today's autopun spotted in the wild: "Apple Cancels Press Conference on iPhone Alarm Glitch after Steve Jobs Oversleeps" 22:37
PerlJam notes he's spent a good portion of his time with Perl on IRC telling people "that book/tutorial/whatever is crap, see perl.org"
I'd hate to have to repeat that with Perl 6
(although, I suppose I'll spend a good bit of time telling people instead, "that's for Perl 5, what you want it perl6.org") 22:38
drake1 the transcript is really helpful. thanks
sub fn(Int $i where { $i < 500 } ) { ... 22:39
that's pretty obvious restraints on the signature 22:40
masak yes, it sure beats no signatures, as with Perl 5.
or less-than-awesome signatures, as with most other languages :P
drake1 I still love to shift and ( $a, $b, $c ) = @_ # list context 22:41
tadzik I don't :)
drake1 hmm
huf as always, there's a module for that on CPAN :) 22:42
tadzik but I started to like Perl 5 less as I use more 6
huf but that still doesnt make it awesome-signatures, only less-than-awesome
drake1 perl6 seems perlfect
22:43 pyrimidine left
drake1 in C it's pretty standard to add the signature like in perl5, for the `ed' scripts 22:44
when they're no all int's
the sematic tokens, however, have always had room for it 22:45
semantic
tadzik goo.gl/BQCxh (SCNR) 22:49
22:51 kaare_ left
drake1 great thing about perl6, it's now with extra tasteful cast operators 22:52
TimToady or at least tasty
drake1 yeah
22:53 plobsing left
drake1 Perl6 is actually in the Future Directions of www.gangsterfreak.com/vcfed 22:54
22:55 c1sung joined, _twitch joined
drake1 can the similarity of "4k" and "fork" be a straight coincidence? 22:56
Alias Quick Question: Will Perl 6 allow for fast subroutine calls?
In Perl, the limiting factor for many things seems to be that the underlying minimum cost of a subroutine call is rather high
(Methods even more so) 22:57
22:59 c1sung left
drake1 you can't call them from your assembly codes either 23:03
or maybe? 23:04
pmurias Alias: according to sorear subroutines are already faster in niecza than they are in Perl 5 23:05
drake1 Perl has always been a bit slow but incredibly handy
Alias neat 23:06
pmurias s/subroutines/subroutine calls/
23:06 bacek__ left
snarkyboojum I wonder how that would be/has been benchmarked 23:07
pmurias calling a sub 1000x times? 23:08
23:09 _twitch left, hercynium left
Alias snarkyboojum: Some of the benchmarks in those online multi-language shootout claim to test something or another, but they require all languages implement the code directly as a copy of the C code 23:09
snarkyboojum: And when you make it as sub-call heavy as the C version it seems to end up being 10 or more times slower than if you could just inline the most intensive ones 23:10
snarkyboojum Alias: ah - language comparison is tough I guess 23:12
Alias Especially when they insist things be implemented as if you are writing C 23:13
TimToady ***************
obviously, we just need a Perl 6 compiler that's a little smarter about inlining and unrolling... :) 23:14
Alias right! :) 23:15
Actually, I'm seeing something similar when I'm writing OpenGL code
I'm having to be completely paranoid about avoiding function calls anywhere I can
Anything that is O(Frame * Object) is going to be hit 60k times a second, and the function/method call overheads are starting to add up to a non-trivial percentage of total CPU consumption 23:17
drake1 I like the ease of hash associative sub routines in Perl
Alias Although that's probably being exadurated quite a bit by the profiler
TimToady drake1: not sure what you mean by that
drake1 then I don't have to link up the "hal" object that I wrote in m4 $^|as -o $< 23:18
TimTody: like spaghetti declarations
nested sub's in hash definitions 23:19
TimToady that's mostly just a consequence of having first-class functions 23:20
snarkyboojum has Niecza implemented loops yet?
23:20 icwiener joined 23:21 MayDaniel left
TimToady niecza: .say for 1..10 23:21
p6eval niecza v1-82-gf7b008c: OUTPUT«===SORRY!===␤Unhandled statement modifier for at /tmp/mGgoJgOKLu line 1 (EOF):␤------> .say for 1..10⏏<EOL>␤Can't call method "statement_level" on unblessed reference at /home/p6eval/niecza/src/Niecza/Actions.pm line 2383.␤Check
..failed␤»
TimToady niecza: for 1..10 { .say }
p6eval niecza v1-82-gf7b008c: OUTPUT«===SORRY!===␤Can't call method "statement_level" on unblessed reference at /home/p6eval/niecza/src/Niecza/Actions.pm line 2383.␤Attribute (do) does not pass the type constraint because: Validation failed for 'Op' with value undef (not isa Op) at
../home/p6eval/niecz…
TimToady niecza: for 1,2,3 { .say } 23:22
p6eval niecza v1-82-gf7b008c: OUTPUT«===SORRY!===␤Can't call method "statement_level" on unblessed reference at /home/p6eval/niecza/src/Niecza/Actions.pm line 2383.␤Attribute (do) does not pass the type constraint because: Validation failed for 'Op' with value undef (not isa Op) at
../home/p6eval/niecz…
TimToady niecza: loop { say "hi"; last }
p6eval niecza v1-82-gf7b008c: OUTPUT«===SORRY!===␤Action method statement_control__S_loop not yet implemented at /home/p6eval/niecza/src/Niecza/Actions.pm line 55.␤Check failed␤»
pumphaus hum, it seems the rakudo star build system seems to be broken
23:22 bacek__ joined
snarkyboojum doesn't look like it eh 23:22
pumphaus it assumes an already exisiting perl6 binary in $PREFIX
TimToady snarkyboojum: the answer appears to be a close approximation of "no"
well, there's always recursion... 23:23
23:23 Grrrr left
snarkyboojum which might not be the best way for benchmarking a number of subroutine calls :D 23:23
TimToady niecza: map { .say }, 1,2,3 23:24
p6eval niecza v1-82-gf7b008c: OUTPUT«===SORRY!===␤Can't call method "statement_level" on unblessed reference at /home/p6eval/niecza/src/Niecza/Actions.pm line 2383.␤Attribute (do) does not pass the type constraint because: Validation failed for 'Op' with value undef (not isa Op) at
../home/p6eval/niecz…
TimToady niecza: my @x = map { .say }, 1,2,3
p6eval niecza v1-82-gf7b008c: OUTPUT«===SORRY!===␤Can't call method "statement_level" on unblessed reference at /home/p6eval/niecza/src/Niecza/Actions.pm line 2383.␤Attribute (do) does not pass the type constraint because: Validation failed for 'Op' with value undef (not isa Op) at
../home/p6eval/niecz…
TimToady niecza: my @x = map -> $x { $x.say }, 1,2,3
p6eval niecza v1-82-gf7b008c: OUTPUT«1␤2␤3␤»
TimToady there's a loop!
snarkyboojum building niecza has improved quite a bit on OS X over the last couple of months anyway - Just Worked the other day 23:25
23:25 Grrrr joined
TimToady niecza: my @x = for 1,2,3 -> $x { $x.say } 23:25
p6eval niecza v1-82-gf7b008c: OUTPUT«===SORRY!===␤Word 'for' interpreted as a listop; please use 'do for' to introduce the statement control word at /tmp/k3QkznPnEQ line 1:␤------> my @x = ⏏for 1,2,3 -> $x { $x.say }␤Unexpected block in infix position (two terms in a
..row) at /tmp/k…
TimToady niecza: my @x = do for 1,2,3 -> $x { $x.say }
p6eval niecza v1-82-gf7b008c: OUTPUT«1␤2␤3␤»
TimToady ooh, there's another loop
masak tadzik: the more I use Perl 6, the more I like and respect Perl 5, too. 23:26
snarkyboojum specifying the series explicitly is no fun :P
TimToady niecza: my @x = do for 1 .. 3 -> $x { $x.say }
p6eval niecza v1-82-gf7b008c: OUTPUT«␤Unhandled Exception: System.Exception: Unable to find lexical &infix:<..> in ANON␤ at Niecza.CLRBackend.NamProcessor.ResolveLex (System.String name, System.Int32& uplevel, Boolean core) [0x00000] in <filename unknown>:0 ␤ at
..Niecza.CLRBackend.NamProcessor.RawAccessLex (System.St…
TimToady niecza: my @x = do for 1 ... 3 -> $x { $x.say }
p6eval niecza v1-82-gf7b008c: OUTPUT«␤Unhandled Exception: System.Exception: Unable to find lexical &infix:<...> in ANON␤ at Niecza.CLRBackend.NamProcessor.ResolveLex (System.String name, System.Int32& uplevel, Boolean core) [0x00000] in <filename unknown>:0 ␤ at
..Niecza.CLRBackend.NamProcessor.RawAccessLex (System.S…
23:29 pmurias left
snarkyboojum I'll place a feature request with sorear asking for loops :) 23:30
23:31 felliott left 23:33 Grrrr left 23:34 Franek left 23:36 felliott joined, Grrrr joined 23:37 felliott left
drake1 Perl6 can run in a Windows 7 terminal, right? 23:38
23:39 felliott joined
drake1 might be a little bit awkward 23:40
TimToady I can't think of any reason why not, as long as it can actually find the executable
drake1 set the path with the path command I guess 23:41
actually I think I'll just copy my Cygwin to it 23:42
cygwin.dll or something like that
no
TimToady or just install a fresh cygwin
drake1 I will simply use the ftp client to transfer the program to a proper hosted execution environment 23:43
23:43 felliott left 23:46 Franek joined 23:47 plobsing joined
drake1 does anyone know how to: cat > $name on windows? 23:49
flussence s/cat/type/
drake1 ok
that should be fine 23:50
sjohnson pigwyn
cotto_work jnthn: ping
drake1 maybe copy type cat hehe heh
23:51 shi left
sorear good * #perl6 23:52
tadzik good star sorear
drake1 then it's quick to only update a part of the script before it's concatenated on the host
flussence if you want to `cat` files on windows, it's "copy file1 + file2 + fileN file.out" 23:53
drake1 ok
then I wont copy type cat
thanks for the tip 23:54
does anyone have a shell account with perl6 ?
ed and cat? 23:55
flussence (of course, the last time I used windows was some horrible WinCE embedded thing, so all this is a bit fuzzy)
drake1 eventually through a VPN telnet
sorear -> backlog 23:56
V15170R: monads and roles are not related 23:57
23:57 felliott joined, gbacon 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!