Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2012-10-31

Perl 6 | Reference Documentation | Rakudo

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:06 dayangkun joined #perl6
00:38 sivoais joined #perl6
00:44 rjbs What a day!  Cheer me up, #perl6!  What's new?
00:48 anuby joined #perl6
01:01 arnsholt phenny: tell masak Friday works for me
01:01 phenny arnsholt: I'll pass that on when masak is around.
01:05 benabik joined #perl6
01:14 ChoHag joined #perl6
01:15 arnsholt phenny: tell jnthn Friday work for you? Thursday's bad for masak
01:15 phenny arnsholt: I'll pass that on when jnthn is around.
01:16 hypolin joined #perl6
01:21 stocks joined #perl6
01:28 VXZ joined #perl6
01:37 fgomez joined #perl6
01:38 [Coke] r: new.WHAT
01:38 p6eval rakudo 34febe: OUTPUT«===SORRY!===␤CHECK FAILED:␤Undefined routine '&new' called (line 1)␤»
01:43 am0c joined #perl6
01:45 diakopter r: say (&new).WHAT
01:45 p6eval rakudo 34febe: OUTPUT«Nil␤»
01:45 diakopter r: say (&ne).WHAT
01:45 p6eval rakudo 34febe: OUTPUT«Nil␤»
01:45 diakopter r: say (&booger).WHAT
01:45 p6eval rakudo 34febe: OUTPUT«Nil␤»
01:45 diakopter eh
01:46 sorear r: say &booger
01:46 p6eval rakudo 34febe: OUTPUT«Nil␤»
01:52 [Coke] rjbs: there's a new Star Wars movie coming out in 2015.
01:52 [Coke] rjbs: I am provisionally cheered.
01:53 rjbs I am curious as to what it will be.
01:53 sorear VII
01:54 [Coke] sorear: I think he was going slightly deeper.
01:54 rjbs Yeah.  What's the plot?
01:55 rjbs I'm hoping it will be a romantic comedy taking place in a suburb moon of Coruscant.
01:56 sorear "last I heard (which was about 15 years ago) it's going to involve Luke's son"
01:58 [Coke] rjbs: seth green is working on a comedy set on coruscant, I believe.
01:59 rjbs starring action figures?
02:00 [Coke] rjbs: http://www.ign.com/articles/2012/08/24/firs​t-footage-shown-from-seth-green-and-matt-se​nreichs-star-wars-detours-animated-series
02:00 rjbs oh yeah, heard about that
02:01 [Coke] ah, I lied about the setting.
02:11 benabik joined #perl6
02:13 gootle joined #perl6
02:21 sudokode joined #perl6
02:21 [Coke] jnthn: in nqp, how can i tell if something is a sixmodelobject?
02:22 [Coke] (hace to split use of clone && repr_clone) :(
02:22 [Coke] *have
02:22 sorear nqp: 5 + pir::new('FixedPMCArray')
02:22 p6eval nqp: OUTPUT«Error while compiling block : Error while compiling op add_n: ResizableStringArray: Can't shift from an empty array!␤current instr.: '' pc 43971 (src/stage2/QAST.pir:15393) (src/stage2/QAST.nqp:2331)␤»
02:22 sorear nqp: 5 + pir::getinterp__p()
02:22 p6eval nqp: OUTPUT«Error while compiling block : Error while compiling op add_n: Cannot infer type from ''␤current instr.: '' pc 43971 (src/stage2/QAST.pir:15393) (src/stage2/QAST.nqp:2331)␤»
02:23 sorear nqp: 5 + pir::getinterp__P()
02:23 p6eval nqp: OUTPUT«get_number() not implemented in class 'ParrotInterpreter'␤current instr.: '' pc 45 ((file unknown):154158703) (/tmp/nKXoJLKlm0:1)␤»
02:23 sorear nqp: pir::getinterp__P().moo
02:23 p6eval nqp: OUTPUT«Method 'moo' not found for invocant of class 'ParrotInterpreter'␤current instr.: '' pc 48 ((file unknown):38) (/tmp/Em44dtINkc:1)␤»
02:23 sorear nqp: pir::getinterp__P().WHAT
02:23 p6eval nqp: OUTPUT«Can only use get_what on a SixModelObject␤current instr.: '' pc 42 ((file unknown):38) (/tmp/2k6E8q3T4u:1)␤»
02:23 sorear nqp: say(pir::isa(pir::getinter​p__P(),'SixModelObject'))
02:23 p6eval nqp: OUTPUT«Error while compiling block : Error while compiling op call: ResizableStringArray: Can't shift from an empty array!␤current instr.: '' pc 43971 (src/stage2/QAST.pir:15393) (src/stage2/QAST.nqp:2331)␤»
02:23 sorear nqp: say(pir::isa__IPS(pir::getin​terp__P(),'SixModelObject'))
02:23 p6eval nqp: OUTPUT«0␤»
02:24 sorear nqp: say(pir::isa__IPS(5,'SixModelObject'))
02:24 p6eval nqp: OUTPUT«0␤»
02:24 sorear nqp: say(pir::gettype__SP(5))
02:24 p6eval nqp: OUTPUT«error:imcc:syntax error, unexpected SREG, expecting '(' ('$S5001')␤ in file '(file unknown)' line 38␤␤»
02:25 sorear nqp: say(pir::typeof__SP(5))
02:25 p6eval nqp: OUTPUT«Integer␤»
02:25 sorear nqp: my $x = 5; say(pir::typeof__SP($x))
02:25 p6eval nqp: OUTPUT«Assignment ("=") not supported in NQP, use ":=" instead at line 2, near " 5; say(pi"␤current instr.: 'panic' pc 13089 (src/stage2/gen/NQPHLL.pir:4619) (src/stage2/gen/NQPHLL.pm:324)␤»
02:25 sorear nqp: my $x := 5; say(pir::typeof__SP($x))
02:25 p6eval nqp: OUTPUT«Integer␤»
02:25 sorear nqp: 5.WHAT
02:25 p6eval nqp: OUTPUT«Can only use get_what on a SixModelObject␤current instr.: '' pc 43 ((file unknown):38) (/tmp/0j7pSKMD9K:1)␤»
02:25 sorear nqp: [].WHAT
02:25 p6eval nqp: OUTPUT«Can only use get_what on a SixModelObject␤current instr.: '' pc 43 ((file unknown):38) (/tmp/6LXEyWDw0C:1)␤»
02:26 sorear nqp: say(pir::isa__IPS(NQPMu,'SixModelObject'))
02:26 p6eval nqp: OUTPUT«1␤»
02:26 sorear nqp: say(pir::isa__IPS(5,'SixModelObject'))
02:26 p6eval nqp: OUTPUT«0␤»
02:26 sorear [Coke]: ^ might work until you get a Real Answer(tm) from jnthn
02:32 plobsing joined #perl6
02:58 sudokode joined #perl6
03:10 stocks_ joined #perl6
03:15 orafu joined #perl6
03:29 japhb o/
03:39 timotimo joined #perl6
04:35 infosophy joined #perl6
04:42 havenn joined #perl6
04:46 stocks joined #perl6
04:48 [particle] joined #perl6
04:59 kaleem joined #perl6
05:19 _jaldhar joined #perl6
05:31 birdwindupbird joined #perl6
05:31 fgomez joined #perl6
05:36 _jaldhar joined #perl6
05:37 telex joined #perl6
05:44 diakopter how adorkable
05:48 cognominal joined #perl6
06:27 dayangkun joined #perl6
06:28 SamuraiJack joined #perl6
06:29 GlitchMr joined #perl6
07:12 sudokode joined #perl6
07:31 SamuraiJack_ joined #perl6
07:35 jnthn phenny: tell arnsholt yes, Friday works for me also
07:35 phenny jnthn: I'll pass that on when arnsholt is around.
07:35 phenny jnthn: 01:15Z <arnsholt> tell jnthn Friday work for you? Thursday's bad for masak
07:41 domidumont joined #perl6
07:43 diakopter o/
07:44 sorear o/
07:45 shachaf sorear: Does Perl 6 have something like lenses and traversals? :-)
08:05 domidumont joined #perl6
08:30 arnsholt Heh. Preemptive phenny reply
08:30 phenny arnsholt: 07:35Z <jnthn> tell arnsholt yes, Friday works for me also
08:39 Gothmog_ joined #perl6
08:42 tadzik good morning
08:45 rindolf joined #perl6
09:00 hoelzro is the following valid Perl6? Int sub foo { return 0; }
09:00 hoelzro I'm wondering if Rakudo just doesn't parse it
09:01 tadzik std: Int sub foo { return 0; }
09:01 flightrecorder joined #perl6
09:01 p6eval std edd7715: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Two terms in a row (listop with args requires whitespace or parens) at /tmp/mV8FOqTv7E line 1:â�¤------> [32mInt [33mâ��[31msub foo { return 0; }[0mâ�¤    expecting any of:â�¤  infix or meta-infixâ�¤    infixed functionâ�¤       statement modifier loopâ�¤P…
09:01 tadzik yeah, invalid Perl 6
09:01 hoelzro what's std? just a parser?
09:01 tadzik std: sub foo returns Int { return 0; }
09:01 p6eval std edd7715: OUTPUT«ok 00:00 42m␤»
09:01 tadzik yes
09:01 hoelzro ah, returns Int, that's good to know
09:01 tadzik std: sub foo(--> Int) { return 0; }
09:01 p6eval std edd7715: OUTPUT«ok 00:00 44m␤»
09:01 * hoelzro is improving vim-perl for Perl 6
09:01 tadzik or this
09:02 diakopter parser + some semantic analysis/checking
09:02 moritz mostly related to symbol tables
09:02 hoelzro std: sub foo ($bar) returns Int { return 0; }
09:02 p6eval std edd7715: OUTPUT«Potential difficulties:â�¤  $bar is declared but not used at /tmp/YUKeTqkY9l line 1:â�¤------> [32msub foo ([33mâ��[31m$bar) returns Int { return 0; }[0mâ�¤ok 00:00 44mâ�¤Â»
09:03 hoelzro std: sub foo (Int $bar) returns Int { return $bar + 0; }
09:03 p6eval std edd7715: OUTPUT«ok 00:00 46m␤»
09:03 hoelzro \o/
09:05 * diakopter just lifted his Thinkpad by the corner of the front, and heard/felt some cracks
09:05 tadzik use duct tape. It fixed mine
09:06 diakopter probably something to do with dropping it on that corner the other day
09:06 brrt joined #perl6
09:06 hoelzro my Thinkpad faceplanted into a coffeeshop floor a few months ago
09:07 hoelzro still running =)
09:07 hoelzro (although the HDD needed replacement)
09:07 tadzik did you replace it with the ssd?
09:07 hoelzro no, actually
09:08 tadzik do consider it next time, it's awesome :)
09:08 hoelzro I'm thinking of moving my desktop to an SSD when I move back to the US, though
09:08 hoelzro I like the extra space of a conventional drive ;)
09:08 * diakopter has the ssd, and is running very low on space
09:09 diakopter clearly $work's IT can't afford the big ssds
09:11 tadzik uh-oh
09:11 tadzik rakudo/Q just started swapping on my ssd
09:11 tadzik do not want
09:11 bonsaikitten hoelzro: use a SSD as cache for a harddisk
09:11 bonsaikitten that way you get a good compromise of both
09:11 hoelzro bonsaikitten: in a laptop, though?
09:12 bonsaikitten hoelzro: hybrid drives ... but yeah, laptop is a bit cramped
09:12 wk joined #perl6
09:13 tadzik jnthn, jnthn
09:15 hoelzro is there a way I could run std on my local machine, so I don't need to come here and spam the channel with questionable Perl6? ;)
09:15 tadzik you can always /query p6eval
09:15 diakopter well, you can msg p6eval directly
09:15 hoelzro fair enough =)
09:15 tadzik but yes, I think you can install viv locally
09:15 diakopter git checkout it from perl6/std
09:15 diakopter make
09:16 diakopter (doesn't work on windows at all)
09:16 diakopter (cygwin included - I tried for hours)
09:16 muixirt joined #perl6
09:17 muixirt hi
09:17 diakopter hi
09:17 muixirt does anyone know why rakudo star for windows downloads spiked?
09:18 diakopter linked from twitter? dunno
09:20 jnthn tadzik, tadzik!
09:20 tadzik jnthn: are you aware of the memory problem of rakudo/Q?
09:20 tadzik I mean, 4G not being enough to build it anymore
09:21 kresike joined #perl6
09:21 jnthn tadzik: Whoa...really? :/
09:21 kresike hello all you happy perl6 people
09:21 tadzik jnthn: yep, I'm afraid so :(
09:21 tadzik my laptop didn't make it
09:21 jnthn Odd...on my box it builds no slower than nom
09:21 tadzik no slower, maybe :)
09:22 jnthn then, it's 16 GB of RAM
09:22 tadzik I originally doubled ram on my laptop to be able to build rakudo in early nom days
09:22 tadzik not sure if my motherboard can handle more :)
09:23 jnthn Well, it aint' gonna merge with a memory usage regression like that...
09:23 muixirt_ joined #perl6
09:23 diakopter merge it with fire
09:23 tadzik I wanted to work on Pod formatting codes for moritz, now that there's nibbler and awesomeness generally
09:25 cognominal joined #perl6
09:25 moritz nibbler, quibbler, babble, ...
09:26 jnthn Blame STD for those :P
09:26 jnthn I can think of two things it could be
09:26 tadzik I'm bisecting Q now, maybe that'll help
09:27 jnthn either 1) we're somehow capturing a LOAD more match objects, or 2) the derived quote languages are insanely costly.
09:27 moritz or 3) we have some weird memory corruptions which the moon phases of Q trigger, but nom does not
09:27 diakopter or tadzik's machine is finicky
09:28 diakopter "I don't like star lately"
09:28 brrt anyone got a good git bisect guide?
09:29 ChoHag joined #perl6
09:29 hoelzro brrt: like, how to use it, or...?
09:29 brrt i want to figure out why building nqp on darwin-x86-64 segfaults
09:29 daxim joined #perl6
09:29 brrt so, yes, how to use it
09:30 tadzik well, you git bisect start
09:30 tadzik then you have to find two revisions, good and bad. You do git bisect good and git bisect bad on another
09:31 tadzik then git will automagically jump to various revisions which you then test and classify (git bisect good|bad)
09:31 tadzik and at some point it will stop and say "now. This is the commit that broke stuff"
09:31 brrt ah
09:31 brrt that seems simple enough
09:31 moritz it's a binary search
09:31 moritz except that you do things like skip commits
09:32 moritz which makes it a bit more complex under the hood
09:32 brrt well, with a bit of work i'll figure out the nqp regression then
09:33 bonsaikitten brrt: if you have a test case that can be run in a shell script you can automate it
09:33 bonsaikitten which means you let the computer do the heavy lifting :)
09:34 marloshouse_ joined #perl6
09:34 moritz somehow I never manged to get that script right
09:34 hoelzro is std *supposed* to spit out funky symbols, like the symbol for the eject button?
09:34 moritz though when I tried, it were rather complex cases
09:34 moritz hoelzro: yes
09:34 hoelzro interesting...
09:34 hoelzro I always thought that was an issue with irssi
09:34 * hoelzro shrugs
09:35 moritz the eject symbol marks the spot where a syntax error was found
09:37 brrt bonsaikitten, the test if any is 'make'
09:38 bonsaikitten brrt: and that gives sane return values
09:38 Rix joined #perl6
09:44 brrt fun
09:45 cognominal_ joined #perl6
09:46 tadzik jnthn: 467a479850ce70c11b7905e618c6879fc1ee1d22 is the first bad commit
09:46 tadzik Switch single quoted strings to new quote parser.
09:47 tadzik the first one that exceeds 20% of my memory on stage parse and then goes up to 11
09:49 brrt compiling QAST is a bit slow, though
09:49 brrt but it seems to work
09:51 patspam joined #perl6
09:53 mr007x joined #perl6
09:54 grondilu joined #perl6
09:54 grondilu rn: proto f($ where * > 0) {*}; multi f($n) { $n*2 }; say f -1;
09:54 p6eval rakudo 34febe, niecza v22-15-gc600005: OUTPUT«-2␤»
09:55 brrt i have the evil commit
09:55 * grondilu thought multis inherit constraints from their proto
09:56 brrt Eliminate use of lexpad_pull cheat.
09:57 moritz grondilu: they should
09:57 moritz grondilu: I think that's a bug
09:58 moritz r: proto f(Int $) {*}; multi f($) { 42 }; say f 'foo'
09:58 p6eval rakudo 34febe: OUTPUT«42␤»
09:59 jnthn r: proto f(Int $) { {*} }; multi f($) { 42 }; say f 'foo'
09:59 p6eval rakudo 34febe: OUTPUT«Nominal type check failed for parameter ''; expected Int but got Str instead␤  in sub f at /tmp/VLCNua6c9F:1␤  in block  at /tmp/VLCNua6c9F:1␤␤»
09:59 jnthn ah...onlystar inlining is cheating and getting caught I suspect
10:00 jnthn Above is workaround anyway
10:00 moritz that's probably why my proto-arity branch didn't slow down rakudo :-)
10:00 brrt jnthn: I have the commit which breaks nqp on mac systems
10:01 brrt its probably a compiler difference now that i think of it
10:01 tadzik moritz: I think custom-lib actually broke panda, to some extend
10:01 tadzik it no longer finds .panda/state on rebootstrap
10:01 jnthn brrt: Oh?
10:02 brrt the [Coke] problem
10:02 tadzik or it now wants it to be in ~/.perl6/ somewhere?
10:02 jnthn brrt: Ah, same commit he found?
10:02 brrt i'll push it on github
10:02 brrt i don't know
10:02 brrt he also ran git bisect probably :-)
10:05 rindolf Hi all.
10:05 hoelzro ahoy
10:05 rindolf hoelzro: what's up?
10:05 brrt i'll backlog to see what he found
10:05 hoelzro nothing really
10:05 hoelzro I started adding some improvements to the Vim syntax file for Perl 6 last night, thought
10:05 hoelzro *though
10:06 hoelzro I supppose that's significant =)
10:06 cognominal_ joined #perl6
10:06 jnthn .oO( vimprovements )
10:06 rindolf jnthn: heh.
10:06 diakopter is there a cpu or os instruction that says "please don't preempt me during the next X instructions"?
10:07 jnthn not afaik
10:07 diakopter because that would be nice
10:07 jnthn for DoS attacks!
10:07 diakopter anything's bad for untrusted code
10:08 hoelzro if I want to check Perl 6 code for standard correctness on my local machine, I'm guessing I should just run std/tools/tryfile on the file in question?
10:08 hoelzro that, or just type some code into STDIN?
10:09 moritz sounds right
10:10 brrt can i post the diff somewhere?
10:10 jnthn gist?
10:10 hoelzro that's interesting, because the example I posted earlier is ok according to std/tools/tryfile...
10:11 hoelzro (Int sub foo { return 0; })
10:11 moritz diakopter: I can imagine that real-time systems might have such a mechanism
10:11 brrt good idea
10:11 muixirt_ joined #perl6
10:11 moritz r: Int sub foo { return 0; }
10:11 p6eval rakudo 34febe: OUTPUT«===SORRY!===␤Confused␤at /tmp/ejIsTVfn_h:1␤»
10:11 moritz r: my Int sub foo { return 0; }
10:11 p6eval rakudo 34febe:  ( no output )
10:11 hoelzro I know Rakudo doesn't parse it
10:12 hoelzro std: Int sub foo { return 0; }
10:12 p6eval std edd7715: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Two terms in a row (listop with args requires whitespace or parens) at /tmp/fntRXARyIH line 1:â�¤------> [32mInt [33mâ��[31msub foo { return 0; }[0mâ�¤    expecting any of:â�¤  infix or meta-infixâ�¤    infixed functionâ�¤       statement modifier loopâ�¤P…
10:12 hoelzro and it gets angry here, too...
10:12 hoelzro and I have the same commit
10:12 grondilu rn: multi f($n where * %% 2) { $n }; multi f($n) { &?ROUTINE($n+1) }; say f 3;
10:12 hoelzro I'm running this:
10:12 hoelzro echo 'Int sub foo { return 0; }' | ./tools/tryfile
10:13 p6eval niecza v22-15-gc600005: OUTPUT«(timeout)»
10:13 p6eval ..rakudo 34febe: OUTPUT«maximum recursion depth exceeded␤current instr.: 'print_exception' pc 89115 (src/gen/CORE.setting.pir:42750) (src/gen/CORE.setting:9098)␤called from Sub 'infix:<+>' pc 54896 (src/gen/CORE.setting.pir:26231) (src/gen/CORE.setting:2941)␤called from Sub 'f' pc 272 ((f…
10:13 brrt https://gist.github.com/3986248
10:13 grondilu shouldn't &?ROUTINE redispatch a multi?
10:13 grondilu (if you see what I mean)
10:14 jnthn grondilu: No, it means the curernt routine you're in
10:14 grondilu so I need to be explicit and use the actual name of the function?
10:14 jnthn Yes
10:14 jnthn Well
10:15 jnthn You can get to the proto from &?ROUTINE I guess...
10:15 jnthn &?ROUTINE.dispatcher iirc
10:15 grondilu rn: multi f($n where * %% 2) { $n }; multi f($n) { &?ROUTINE.dispatcher.($n+1) }; say f 3;
10:15 p6eval rakudo 34febe: OUTPUT«4␤»
10:15 p6eval ..niecza v22-15-gc600005: OUTPUT«Unhandled exception: Unable to resolve method dispatcher in type Sub␤  at /tmp/YACv5Qeo7n line 1 (f @ 4) ␤  at /tmp/YACv5Qeo7n line 1 (mainline @ 5) ␤  at /home/p6eval/niecza/lib/CORE.setting line 4215 (ANON @ 3) ␤  at /home/p6eval/niecza/lib/CORE.setting …
10:16 grondilu works with rakudo indeed
10:18 hoelzro huh...so my example fails with tloop
10:20 hoelzro oh, ok, I figured it out
10:20 hoelzro tryfile doesn't read Perl6 from STDIN if you provide no files; it reads a *list* of files from STDIN which it tests
10:25 moritz hoelzro: now that you mention it, I remember that TimToady++ piped in `git log --stat` from the tests to syntax-check the changed test files, or something
10:28 azawawi joined #perl6
10:33 hoelzro that makes sense, I guess
10:42 moritz nqp::ctxcaller(nqp::ctx());
10:43 moritz I'd expect that to be nqp::ctxcaller(nqp::interp())
10:43 moritz but then I don't know anything about how nqp::ctxcaller works
10:43 jnthn It takes a ctx and gets its caller
10:43 jnthn So you can walk a chain of contexts
10:43 brrt … any chance there is a zero somewhere there?
10:44 diakopter nqp: say(nqp::ctxcaller(nqp::ctx()).gist)
10:44 p6eval nqp: OUTPUT«Method 'gist' not found for invocant of class 'CallContext'␤current instr.: '' pc 56 ((file unknown):40) (/tmp/Q2E_cLXs94:1)␤»
10:46 jnthn brrt: I'd hope any nulls become proper Parrot nulls...
10:46 diakopter nqp: say(nqp::ctxcaller(nqp::ctxcaller(nqp::ctxcal​ler(nqp::ctxcaller(nqp::ctxcaller(nqp::ctxcal​ler(nqp::ctxcaller(nqp::ctxcaller(nqp::ctxcal​ler(nqp::ctxcaller(nqp::ctx())))))))))).gist)
10:46 p6eval nqp: OUTPUT«(signal SEGV)»
10:46 diakopter whee.
10:46 brrt jnthn: evidence is they don't, in the general case
10:46 jnthn Then, Parrot bug...
10:47 diakopter jnthn: ^^ <snicker>
10:47 diakopter (not snicker at nqp; just snicker at the code that breaks it)
10:48 jnthn diakopter: You don't need the .gist
10:48 brrt jnthn: thats not really fair, the ops are implemented in nqp
10:48 brrt in c
10:48 brrt they can introduce the bug easily
10:49 moritz then look at the PIR it's compiled to
10:49 jnthn brrt: What ops?
10:49 azawawi hi
10:49 jnthn brrt: I'm calling getattribute on a context.
10:49 jnthn It's not even a dynop that NQP has added.
10:49 brrt hmm
10:53 azawawi moritz: is there a way to catch a bad-file-name open and not die from it? :)
10:53 diakopter nqp: my $_ := nqp::ctx(); $_ := nqp::ctxcaller($_) while 1;
10:53 p6eval nqp: OUTPUT«(signal SEGV)»
10:53 azawawi moritz: try { open 'bad-file-name'; CATCH { say "Caught!" } }    # "caught" and not "Caught\nUnable to open filehandle from path 'bad-file-name'"
10:54 bloonix joined #perl6
10:54 diakopter phenny: tell masak masakbot :) nqp: my $_ := nqp::ctx(); $_ := nqp::ctxcaller($_) while 1;
10:54 phenny diakopter: I'll pass that on when masak is around.
10:54 moritz azawawi: sure. add a handler for the exception inside the CATCH block
10:57 moritz azawawi: if you write a CATCH block yourself, you have to write a 'when' (or a 'default') block that matches the exception
10:57 moritz azawawi: on then is it marked as handled. If not, it's rethrown when you exit the CATCH block
10:58 azawawi moritz: got it... thanks
10:58 tadzik do we have a way to capture both stdout of a process and its error code?
10:58 brrt i can't see where it is defined
10:59 moritz tadzik: maybe with open :p ?
11:00 tadzik I didn't know we have such thing :)
11:01 tadzik p6prove so far: https://gist.github.com/3986456
11:02 moritz \o/
11:03 ingy joined #perl6
11:04 doy joined #perl6
11:04 dbr joined #perl6
11:04 dalek panda: fb6ced2 | tadzik++ | bootstrap.pl:
11:04 dalek panda: Add a friendly message to bootstrap.pl, now that bin is in a new place
11:04 dalek panda: review: https://github.com/tadzik/panda/commit/fb6ced21ec
11:05 ruoso joined #perl6
11:07 flussence hoelzro++ # fixing vim-perl
11:08 hoelzro flussence: I'm just adding syntax folding; I'm not really fixing anything =P
11:08 hoelzro but I'm going to work with petdance to get rid of some small issues, maybe
11:12 birdwindupbird joined #perl6
11:12 sivoais joined #perl6
11:19 gugod joined #perl6
11:26 cognominal joined #perl6
11:36 gootle joined #perl6
11:40 dmiu joined #perl6
11:40 dmiu left #perl6
11:49 kaleem joined #perl6
11:56 dayangkun joined #perl6
11:57 SmokeMachine joined #perl6
12:08 Woodi diakopter: about cracking Thinkpad: T40's had known bug that base of laptop was too elastic and when you lift it when it is hot then graphic procesor soldering was damaged and you get blue screens from now on... probably seence [TX]60's base is more solid but your cracking sound looks similiar
12:09 Woodi afternoon v6 ppls :)
12:11 diakopter yeah it sounded like plastic that was already cracked was splintering
12:11 azawawi moritz: does 'require' in Perl 6 work in the same way that it works in Perl?
12:12 moritz azawawi: no
12:13 jnthn Well, it loads things at runtime. The difference is mostly in the import semantics.
12:13 spider-mario joined #perl6
12:13 azawawi jnthn: interesting
12:13 jnthn Which I guess is a pretty huge difference ;)
12:13 cognominal joined #perl6
12:13 diakopter rn: require YourMom
12:14 p6eval niecza v22-15-gc600005: OUTPUT«Unhandled exception: System.NotImplementedException: The requested feature is not implemented.␤  at Anon.1.Run.MAIN.C0mainline (Niecza.Frame ) [0x00000] in <filename unknown>:0 ␤  at Niecza.Kernel.RunCore (Niecza.Frame& cur) [0x00000] in <filename unknown>…
12:14 p6eval ..rakudo 34febe: OUTPUT«Could not find YourMom in any of: /home/p6eval/nom-inst/lib/parrot/​4.8.0-devel/languages/perl6/lib, /home/p6eval/nom-inst/lib/parrot/4.8​.0-devel/languages/perl6/vendor/lib, /home/p6eval/nom-inst/lib/parrot/4.​8.0-devel/languages/perl6/site/lib, /home/p6eval/.perl6/2…
12:14 jnthn Your mom is not implemented in Niecza...
12:14 moritz azawawi: the difference is that stuff in Perl 6 is imported into the lexpad, which is immutable at run time
12:15 moritz azawawi: so you must supply a list of symbols you want to import in the require statement
12:15 azawawi im refactoring things in farabi6... and wanted to delay loading File::Spec until being used in the HTTP actions/routes...
12:15 azawawi so i deleted those use File::Spec and move require into the HTTP GET action sub...
12:15 azawawi and now it is not finding '&Spec'...
12:16 azawawi http://paste.debian.net/205225/
12:19 moritz r: require A::B:; say A::B; BEGIN exit;
12:19 p6eval rakudo 34febe: OUTPUT«===SORRY!===␤Confused␤at /tmp/6RzdGeiT50:1␤»
12:22 azawawi r: sub foo { require File::Spec; say File::Spec.catdir('a','b'); }; foo;
12:22 p6eval rakudo 34febe: OUTPUT«Could not find File::Spec in any of: /home/p6eval/nom-inst/lib/parrot/​4.8.0-devel/languages/perl6/lib, /home/p6eval/nom-inst/lib/parrot/4.8​.0-devel/languages/perl6/vendor/lib, /home/p6eval/nom-inst/lib/parrot/4.​8.0-devel/languages/perl6/site/lib, /home/p6eval/.perl…
12:22 jnthn Try with star
12:22 azawawi star: sub foo { require File::Spec; say File::Spec.catdir('a','b'); }; foo;
12:22 p6eval star 2012.10: OUTPUT«Could not find File::Spec in any of: /home/p6eval/star/lib/parrot​/4.6.0/languages/perl6/lib, /home/p6eval/star/lib/parrot/4.​6.0/languages/perl6/vendor/lib, /home/p6eval/star/lib/parrot/4​.6.0/languages/perl6/site/lib, /home/p6eval/.perl6/2012.10/lib, /home/p6eval/.pe…
12:22 azawawi rakudo star 2010 is here?
12:22 moritz yes
12:23 azawawi i didnt see any announcement about it
12:23 azawawi cool
12:23 moritz http://rakudo.org/2012/10/29/​rakudo-star-2012-10-released/
12:24 moritz r: require A::B; INIT { exit }
12:24 p6eval rakudo 34febe:  ( no output )
12:24 moritz r: require A::B; say A::B; INIT { exit }
12:24 p6eval rakudo 34febe:  ( no output )
12:24 moritz r: require A::B; say A::B;
12:24 p6eval rakudo 34febe: OUTPUT«Could not find A::B in any of: /home/p6eval/nom-inst/lib/parrot/​4.8.0-devel/languages/perl6/lib, /home/p6eval/nom-inst/lib/parrot/4.8​.0-devel/languages/perl6/vendor/lib, /home/p6eval/nom-inst/lib/parrot/4.​8.0-devel/languages/perl6/site/lib, /home/p6eval/.perl6/2012…
12:24 moritz r: require A::B; say A; INIT { exit }
12:24 p6eval rakudo 34febe: OUTPUT«===SORRY!===␤CHECK FAILED:␤Undefined routine '&A' called (line 1)␤»
12:25 moritz azawawi: it seems that require doesn't (yet?) declare the symbol you declare
12:25 moritz azawawi: you can try to look it up at runtime with ::('File::Spec')
12:25 azawawi moritz: see this strange behavior http://paste.debian.net/205228/
12:25 azawawi moritz: rakudo star 2012.09
12:26 diakopter r: exit &YourMom
12:26 p6eval rakudo 34febe:  ( no output )
12:26 jnthn azawawi: That's not entirely strange
12:27 jnthn Whther Foo::Bar is a type name or a call is decided at compile time.
12:27 jnthn require doesn't import things until runtime
12:27 moritz and it doesn't import the type name currently
12:27 moritz which might or might not be a bug
12:27 jnthn The later lines in the REPL are compiled in the context of the previous scope
12:27 moritz anyway, use ::('File::Spec') for now
12:28 moritz jnthn: but why doesn't it find sub foo on lines 4/5 ?
12:29 azawawi star: ::('File::Spec')
12:29 p6eval star 2012.10:  ( no output )
12:29 azawawi star: ::('File::Spec')
12:29 p6eval star 2012.10:  ( no output )
12:29 diakopter r: say say::~say
12:29 p6eval rakudo 34febe: OUTPUT«␤␤TrueTrue␤»
12:30 azawawi what about this one http://paste.debian.net/205229/ ?
12:30 MayDaniel joined #perl6
12:30 moritz azawawi: well, you didn't require File::Spec, did you?
12:31 moritz anyway, it shouldn't give the second error message
12:31 azawawi moritz: true but why the different error message on the second one
12:31 moritz bug
12:31 azawawi maybe it is the hidden AI inside perl6 :)
12:32 moritz related to serialization, probably
12:32 azawawi http://paste.debian.net/205230/    # same bug apparently
12:33 * azawawi installs rakudo star 2012.10
12:38 nyuszika7h joined #perl6
12:38 [Coke] sorear++ # works.
12:39 [Coke] Seems ugly, but I figure I'll go for functional and have pmichaud and jnthn  troll it for ugly after.
12:40 [Coke] http://blogs.perl.org/users/leo_susa​nto/2012/10/garbage-collection.html Nice article about perl GC.
12:43 hash_table joined #perl6
12:55 azawawi is there a plan to bundle an editor  with rakudo star? hint: farabi6?
12:57 azawawi Some languages come with a built-in editor.. why not perl6?
12:59 FROGGS joined #perl6
13:05 wk joined #perl6
13:06 nyuszika7h joined #perl6
13:10 Psyche^ joined #perl6
13:12 atrodo joined #perl6
13:14 [Coke] I would probably not use it myself, but seems potentially useful
13:14 moritz azawawi: what exactly is farabi6?
13:15 azawawi moritz: https://github.com/azawawi/f​arabi6/blob/master/README.md
13:16 azawawi moritz: basically im using a web-based interface with a perl 6 backend
13:16 PerlJam azawawi: I don't think "explode in your face" jibes with Rakudo Star  :)
13:16 azawawi PerlJam: :)
13:20 azawawi moritz: basically i want to integrate perl6 tools here and there with farabi6
13:20 azawawi moritz: and make it easier to write Perl 6
13:21 azawawi moritz: basically an IDE in a browser... but you control the server...
13:21 * hoelzro starts to install Farabi
13:25 bluescreen10 joined #perl6
13:27 spider-mario (...)/pkg/usr/bin/perl6 --target=pir --output=(...)/pkg/usr/lib/parrot/4.8.​0/languages/perl6/lib/SVG/Plot/Pie.pir (...)/pkg/usr/lib/parrot/4.8.0/la​nguages/perl6/lib/SVG/Plot/Pie.pm
13:27 spider-mario use of uninitialized value of type Any in string context  in  at src/Perl6/ModuleLoader.pm:256
13:28 spider-mario it happens repeatedly when precompiling modules
13:28 spider-mario is that normal?
13:28 tadzik doesn't occur to me
13:28 spider-mario I’m building Rakudo Star 2012.10
13:29 cognominal joined #perl6
13:29 spider-mario that line is “if $setting_name ne 'NULL' {”
13:30 spider-mario maybe it could be changed to: if defined $setting_name and $setting_name ne 'NULL' {
13:30 spider-mario could it not?
13:30 tadzik I'm interested in why does it happen to be called with an uninitialized value
13:39 PacoAir joined #perl6
13:41 Khisanth joined #perl6
13:44 jerome joined #perl6
13:57 ChoHag joined #perl6
13:59 flussence `panda install DBIish` doesn't work on my laptop - 37/40 sqlite tests fail. It worked fine on my desktop...
14:02 bowtie joined #perl6
14:03 moritz flussence: a detailed test output would be very helpful
14:03 flussence I'll get to that just as soon as I figure out where panda puts everything now...
14:04 tadzik %*CUSTOM_LIB<site> probably
14:04 tadzik I discovered that today too :)
14:07 stopbit joined #perl6
14:07 benabik joined #perl6
14:07 moritz %*CUSTOM_LIB<site>/panda/source or so
14:11 [Coke] r: say %*CUSTOM_LIB
14:11 p6eval rakudo 34febe: OUTPUT«("perl" => "/home/p6eval/nom-inst/lib/parro​t/4.8.0-devel/languages/perl6", "vendor" => "/home/p6eval/nom-inst/lib/parrot/4​.8.0-devel/languages/perl6/vendor", "site" => "/home/p6eval/nom-inst/lib/parrot/​4.8.0-devel/languages/perl6/site", "home" => "/home/p6eval/.per…
14:12 [Coke] r: say keys %*CUSTOM_LIB
14:12 p6eval rakudo 34febe: OUTPUT«perl vendor site home␤»
14:12 tokuhiro_ joined #perl6
14:16 alester joined #perl6
14:27 flussence ok, it segfaults on test #4.
14:28 pmurias joined #perl6
14:28 flussence dmesg says "perl6[1316] general protection ip:b360251d sp:bfcebfd0 error:0 in libsqlite3.so.0.8.6[b359d000+aa000]", which doesn't tell me much...
14:29 benabik joined #perl6
14:31 pmurias jnthn: hi
14:31 moritz flussence: doesn't tell me much either :(
14:31 flussence oh wait, I have gdb on here.
14:32 * flussence uses
14:32 jnthn o/ pmurias
14:33 pmurias if I have XYZ::foo();module XYZ {our sub foo() {...}} how can sub foo access lexicals from a block which hasn't been initalised yet?
14:34 pmurias can it only access static lexicals till we execute the module block?
14:34 [Coke] jnthn: hio. Is there a better way than pir::isa to see if something is a sixmodelobject?
14:34 flussence moritz: still doesn't tell me much, maybe someone can make sense of it: https://gist.github.com/3987365
14:37 jnthn pmurias: Only the static ones make any sense
14:37 moritz flussence: I can't :(
14:38 jnthn pmurias: The others will have some type object or similar I guess
14:38 jnthn Rakudo probably gets things righter than NQP on this front presently.
14:40 PerlJam flussence: Did you try to re-install NativeCall first?
14:41 jnthn flussence: What platform?
14:41 jnthn Looks like a 32-bit from the trace...
14:45 jnthn [Coke]: No, there's no
14:45 * PerlJam got an email from a Texas land surveyor named "Lawrence Wall"  :)
14:45 jnthn *not
14:46 PerlJam sorry, "Laurence"
14:46 xinming_ joined #perl6
14:46 hash_table joined #perl6
14:46 rjt_ joined #perl6
14:49 rjt_ Hello. Is this the best place to come to ask about Rakudo feature implementation progress, and how I can help?
14:49 moritz hello rjt_. Yes, it is.
14:50 rjt_ Ok, good. I'm very interested in shaped arrays, and compact arrays to a lesser extent. I'm also quite capable of helping out design/implementation/whatever wise.
14:50 moritz \o/
14:50 jnthn nice :)
14:50 rjt_ :-)
14:51 moritz I think so far that's been on jnthn's TODO list, which is rather full :-)
14:51 nyuszika7h joined #perl6
14:51 jnthn Yes, task-stealing from my TODO list is always welcome ;)
14:51 rjt_ I can imagine. :-) So I guess my next question is, how can I help, and where do I start?
14:52 rjt_ (with clever use of comma operator to combine two questions, that is)
14:52 moritz well, you can help by reading the spec, writing tests, designing and implementing the features
14:52 moritz that's the easier part :-)
14:53 jnthn Devil's in the details, as always.
14:53 moritz aye
14:53 moritz so, where to start? Probably by coming up with a design for how to store the stuff
14:54 pmurias jnthn: what do all the LexInfo pir things in post_serialize do? setup lexicals for post_serialisation?
14:54 jnthn At the moment, we use Parrot's resizable PMC array for the low-level storage. I've been expecting that we move over to using a 6model representation thingy for that instead. A bit like arnsholt++ did for CArray representation tht we use for native call.
14:54 mucker joined #perl6
14:55 jnthn pmurias: I'd have to go look at the code to be sure, but I think they just install the static values.
14:55 jnthn "this WVal is the static lexical for Foo"
14:55 rjt_ Ok, I can focus on that. I'll still need to spend some time coming up to the speed on some of the higher level points, but that shouldn't take too long.
14:56 flussence bit distracted for a minute there, sorry
14:56 flussence PerlJam: yes, NativeCall installed fine
14:56 pmurias jnthn: static lexical for Foo while executing post_serialisation?
14:56 moritz rjt_: may I ask what timezone you're in?
14:57 pmurias post_serialisation is just an array of stmts instead of an block, so it setups it's lexical variables that way?
14:57 moritz and if so, what timezone are you in? :-)
14:57 jnthn pmurias: No, it's setting up the static lexicals for all the other blocks
14:58 jnthn That's what the .const thing does - gets the object representing the static lexicals for the block that's named
14:58 pmurias and how does it setup static lexicals for itself?
14:58 * jnthn doesn't get teh question
14:59 * pmurias rechecks the ast dump
14:59 jnthn post_deserialization stuff is run once, right after we deserialize objects
15:00 pmurias yes
15:00 jnthn It doesn't map to any user code
15:00 jnthn It's just doing some setup work
15:00 jnthn Before we ever run any real code.
15:00 pmurias but it refers to lexicals
15:00 pmurias but it doesn't have a QAST::Block to declare those
15:00 jnthn It refers to the static lex info
15:00 jnthn Well, we're on the runtime side of the fence here
15:00 jnthn QAST::Block is the compile time thing
15:01 pmurias post_deserialize contains code that puts a BVal into a package
15:02 pmurias but the BVal refers to lexicals like $?PACKAGE
15:02 pmurias how should I resolve them, find a different place the BVal is used and find it's parrent?
15:02 cfi joined #perl6
15:03 kaleem_ joined #perl6
15:03 cfi hi perl6
15:03 pmurias or do the LexInfo pir things set it up?
15:04 jnthn You're not actually invoking the block during post_deserialize, just installing a reference to it somewhere.
15:05 PerlJam cfi: greetings!
15:05 jnthn When compiling a QAST::Blcok you end up with the code itself, and also a hash mapping lexicals to default values.
15:05 cfi seq 1 5|perl6 -ne '.say'; error!!Option 'e' in grouped options '-ne' needs a value, but does not have one
15:05 cfi what's the value?
15:06 jnthn The LexInfo is that hash
15:06 jnthn to a first approximation
15:07 PerlJam cfi: that looks like a bug to me, but you want: seq 1 5 | perl6 -n -e '.say'
15:08 cfi PerlJam: thanks! we can't use -ne just like perl 5anymore?
15:09 geekosaur see "that looks like a bug"
15:09 jnthn It probably should work with -ne eventually also.
15:09 moritz I thought I had fixed that
15:09 cfi PerlJam: thanks! that's a big bug.
15:10 moritz I'd call it a small bug :-)
15:10 cfi moritz: I can't do that. -ne
15:12 moritz it's in nqp/src/HLL/CommandLine.pm
15:13 cfi moritz: hope you can fix that.
15:13 moritz oh
15:13 moritz it works if you don't add a space between the -e and the '
15:14 jnthn moritz: the -ne ?
15:14 pmurias jnthn: to make the question clearer: in line 1110 of http://paste.ubuntu.com/1321001/ how do I resolve $?PACKAGE?
15:14 moritz jnthn: yes
15:15 GlitchMr joined #perl6
15:17 cfi moritz: you are right.but looks like a bug.
15:17 cfi thanks! all perl6
15:17 mr007x joined #perl6
15:18 PerlJam moritz: too much $i in that code :)
15:18 thou joined #perl6
15:18 moritz PerlJam: yes, my local patch renames one of them :-)
15:19 jnthn pmurias: Oh, I think this dump may be misleading...
15:19 jnthn pmurias: A BVal contains a reference to a QAST::Block somewhere else in the tree
15:19 jnthn pmurias: It's not that the block is lexically inside of deserialize. It's elsewhere, we're just referring to it.
15:20 jnthn I can understand why you're confused now :)
15:21 pmurias so i should just take the static lexicals from the other place
15:23 jnthn Yes. I guess you have a table somewhere of cuuid to whatever you turn a QAST::Block in to.
15:23 jnthn So you'd compile it to a lookup into there
15:23 colo-work joined #perl6
15:24 jnthn (though maybe you have another table for the static lexicals too...)
15:25 jnthn pmurias: Are you working on deserializing the binary blob over in JavaScript too, ooc?
15:25 fhelmberger joined #perl6
15:25 cfi joined #perl6
15:26 cfi rn:my Int @a = 'a', 'b';say @a;
15:26 cfi rn: my Int @a = 'a', 'b';say @a;
15:26 p6eval rakudo 34febe, niecza v22-15-gc600005: OUTPUT«a b␤»
15:27 jnthn cfi: That's a known not implemented thing just yet.
15:27 cfi jnthn: thanks.
15:27 jnthn On the todo list. Like many things :)
15:38 birdwindupbird joined #perl6
15:39 pmurias jnthn: binary blob? I'm serialising object myself rather the using the serializer
15:42 pmurias jnthn: by binary blob you mean what the serializer turns objects into?
15:42 jnthn pmurias: Yes.
15:42 jnthn pmurias: I don't think re-inventing that is desirable.
15:42 jnthn I specified the format in hope of that ;-)
15:43 pmurias I'll look into that
15:43 jnthn (in docs in nqp repo)
15:43 jnthn Anyway, up to you really. But my gut feeling is that re-inventing this bit will make problems later.
15:50 pmurias currently I serialising a few types of things in an slightly adhock manner
15:51 pmurias jnthn: I think if I'll serialize a similiar set of info but don't use (the exact) binary format it won't cause much trouble
15:55 flussence I'm out of ideas with this sqlite thing... tried downgrading sqlite to the previous version, other nativecall stuff seems to work fine... I'm just gonna ignore it and hope the other DBDs work
15:58 kresike bye folks
15:59 hoelzro flussence: what's the nature of the test failures?
16:07 moritz hoelzro: flussence posted a gist URL in the backlog
16:07 cognominal joined #perl6
16:08 ChoHag joined #perl6
16:09 hoelzro ah, serves me right for half-assing things ;)
16:11 jnthn decommute, bbl
16:13 sudokode joined #perl6
16:39 HarryS joined #perl6
16:46 * jnthn back
16:47 japhb o/
16:48 fgomez joined #perl6
16:48 jnthn hi jvalphb
16:50 jnthn pmurias: It's more that I'm not sure to what degree you can get at all the things it puts into the serialization (like, REPR data) though I guess it can be recreated...
16:51 * japhb chuckles at 'jvalphb'
16:51 japhb (homemade pumpkin oat bread)++
16:52 jnthn ooh :)
16:53 arnsholt o/
16:56 japhb Good *, arnsholt
17:06 colomon \o
17:10 p6eval joined #perl6
17:12 jnthn tadzik: Wow...rakudo/Q is indeed rather into RAM...
17:12 jnthn 5.5GB :D
17:13 TimToady can you figure out whether it's the nibbler or the derivation that is leaking?
17:14 TimToady I assume you did the deletes in nibbler that STD does...
17:14 TimToady (though my gut tells me this has to be something major going wrong on the order of rederiving languages)
17:14 jnthn TimToady: I pondered it on the train home and I think it may be never.
17:15 TimToady it?
17:15 jnthn TimToady: The memory use
17:15 jnthn er
17:15 jnthn *may be neither
17:15 jnthn :)
17:15 TimToady that...makes a bit more sense :)
17:15 jnthn Long day... :)
17:15 jnthn Cursor initialization transcodes things to fixed width.
17:16 jnthn I fixed a bug where LANG was confusing orig (the original thing) and target (what it actually scans)
17:16 jnthn And it was re-transcoding every time we switched langauge
17:16 TimToady that might well do it
17:17 TimToady esp since STD punts and keep orig in a global
17:18 jnthn The size of the setting multiplied out for every single quote construct would add up quite a bit, I imagine.
17:19 snearch joined #perl6
17:22 * TimToady should try keeping orig in a high-level cursor, and providing access via a method, to see what the performance would be...
17:25 jnthn Yes, fixing that makes a huge difference :)
17:27 TimToady whew!
17:28 azawawi joined #perl6
17:28 azawawi hi #perl6
17:32 Pleiades` joined #perl6
17:33 dalek rakudo/Q: fb38835 | jnthn++ | src/Perl6/Grammar.pm:
17:33 dalek rakudo/Q: Fix nibble to also pass on target.
17:33 dalek rakudo/Q: review: https://github.com/rakudo/rakudo/commit/fb38835997
17:34 dalek nqp: acae25d | jnthn++ | src/ (2 files):
17:34 dalek nqp: LANG passes on target and orig.
17:34 dalek nqp:
17:34 dalek nqp: This avoids re-transcoding every single time we switch language, and
17:34 dalek nqp: fixes the memory use regression in rakudo/Q.
17:34 dalek nqp: review: https://github.com/perl6/nqp/commit/acae25d275
17:34 dalek nqp: 579f3d0 | jnthn++ | src/6model/knowhow_bootstrapper.c:
17:34 dalek nqp: Give KnowHOW a type check cache.
17:34 dalek nqp: review: https://github.com/perl6/nqp/commit/579f3d0990
17:34 jnthn phenny: tell tadzik that memory issue in Q is gone (need latest NQP too).
17:34 phenny jnthn: I'll pass that on when tadzik is around.
17:34 jnthn tadzik++ # noticing :)
17:36 jnthn TimToady: Actually we're not doing the :delete's yet, so I expect there's a win to be had from that too.
17:36 TimToady tadzik++ # not committing suicide :)
17:36 TimToady maybe the [*-1] could be .pop
17:36 jnthn The thing that this is replacing wasn't either, mind...
17:37 jnthn I think NQP and STD must have the match/cursor relationship a little different.
17:37 jnthn Well, factored a little different.
17:37 jnthn The cursor just keeps a stack of cursors to hand
17:38 Chillance joined #perl6
17:38 jnthn And they're arranged into the right "shape" when we make the match object
17:38 jnthn Thus elimiating stuff from the match object doesn't throw away the various cursors.
17:38 jnthn Maybe that's true in STD also though...
17:39 jnthn hmmm
17:39 SamuraiJack_ joined #perl6
17:40 * jnthn spots a potential improvement...
17:49 skids joined #perl6
17:50 gdey joined #perl6
17:50 benabik joined #perl6
17:51 mr007x joined #perl6
17:55 benabik_ joined #perl6
17:57 jnthn Hm, not so easy. :)
18:00 azawawi jnthn: any update on the web-based perl6-debug api?
18:00 benabik joined #perl6
18:01 benabik joined #perl6
18:03 jnthn azawawi: My tuit supply is a bit tight at the moment, and I'd like to use the handful I have to try and get the quote stuff I'm working on done. So, won't get to it for a little bit yet.
18:03 azawawi jnthn++
18:07 dalek rakudo/Q: 1065be2 | jnthn++ | src/Perl6/Grammar.pm:
18:07 dalek rakudo/Q: Keep less matches around from nibbler.
18:07 dalek rakudo/Q:
18:07 dalek rakudo/Q: Different (simpler ;-)) factoring than STD, but same win.
18:07 dalek rakudo/Q: review: https://github.com/rakudo/rakudo/commit/1065be23f8
18:27 wk joined #perl6
18:39 hash_table joined #perl6
18:41 SamuraiJack__ joined #perl6
18:53 masak lol hi! #perl6! I again come to here.
18:53 phenny masak: 01:01Z <arnsholt> tell masak Friday works for me
18:53 phenny masak: 10:54Z <diakopter> tell masak masakbot :) nqp: my $_ := nqp::ctx(); $_ := nqp::ctxcaller($_) while 1;
18:53 masak nqp: my $_ := nqp::ctx(); $_ := nqp::ctxcaller($_) while 1;
18:53 p6eval nqp: OUTPUT«(signal SEGV)»
18:53 masak woo
18:54 masak but I don't know enough about nqp failure modes to know if this is fine or not.
18:54 zby_home_ joined #perl6
18:54 masak maybe we are OK with SEGVs in NQP-land?
19:00 sudokode joined #perl6
19:00 * azawawi farabi6 new UI... incoming... :)
19:03 fhelmberger joined #perl6
19:03 sorear masak!
19:03 masak sorear! \o/
19:04 * masak has done three days of teaching and is now rewarding himself with... Angband
19:04 azawawi http://feather.perl6.nl:3030/  # New farabi6 UI... Hopefully the server will stay up :)
19:05 masak azawawi++
19:06 azawawi masak: try "Syntax Check (F6)"
19:06 azawawi masak: it uses std's
19:07 masak cool.
19:07 masak it looks really sleek.
19:07 azawawi twitter bootstrap...
19:08 masak oho.
19:11 masak hm, seems I've forgotten most keybindings from the roguelike keyset... :)
19:12 * azawawi thinks, 'perl6-debug menu, integrate rakudo/niecza, add a REPL' and we have a winner here :)
19:12 masak azawawi: wow. yes.
19:13 azawawi and we then drop it to feather3... and keep restarting it :)
19:16 azawawi masak: please let know if you have any cool ideas
19:16 * azawawi sleepy
19:16 masak :)
19:16 * azawawi sleep &
19:21 jnthn o/ masak
19:21 masak jnthn! \o/
19:21 jnthn masak: The segfault isn't good
19:21 masak jnthn: I'm a level-2 warrior with a broad sword.
19:21 jnthn masak: I think it's reproducable with pure PIR, though...
19:22 jnthn Need to be sure of that.
19:27 * masak submits it to the perl6/nqp issue queue
19:52 rurban left #perl6
19:55 sudokode joined #perl6
19:59 benabik joined #perl6
20:00 birdwindupbird joined #perl6
20:03 arnsholt jnthn, masak: Since Friday seems to work for both of you, I'll see you then =)
20:10 kurahaupo joined #perl6
20:11 ashleydev joined #perl6
20:14 masak arnsholt: looking forward to it!
20:31 jnthn arnsholt: Nice. In Malmo?
20:32 * jnthn guesses evening...
20:32 jnthn Friday is quite flexible for me so whatever can work :)
20:32 masak Friday is a half day, too :)
20:33 jnthn Oh. :)
20:33 jnthn Oh, I know that
20:33 jnthn That's why tomorrow is le stress.
20:34 arnsholt Yeah, I'll try to figure out train times and such tomorrow
20:35 arnsholt Meeting up at the station (or somewhere nearby) work for you? =)
20:35 arnsholt Why's Friday half day, BTW?
20:35 masak arnsholt: allhelgona
20:36 jnthn .oO( all hel gona break lose... )
20:36 jnthn *oose
20:36 arnsholt Oh, right
20:36 jnthn er, *loose
20:36 * masak .oO( when it comes to spelling, you loose )
20:37 arnsholt *giggle*
20:39 sorear what is allhelgona?
20:39 masak phenny: sv en "allhelgona"?
20:39 phenny masak: "All Saints" (sv to en, translate.google.com)
20:39 arnsholt All hallow's eve
20:40 masak I associate it with lighting a candle lantern at some loved one's grave.
20:42 flussence .oO( I usually associate it with "it's cold and dark outside now when I leave the office"... )
20:42 * PerlJam wishes it were cooler outside.
20:43 jnthn Same
20:44 jnthn (Then the weekend's boring rain would be pretty snow)
20:45 _jaldhar joined #perl6
20:50 rindolf joined #perl6
20:58 dalek joined #perl6
21:01 moritz rain is boring, yes
21:02 Pleiades` joined #perl6
21:05 * masak .oO( that's why some rocks have holes )
21:07 leont joined #perl6
21:08 * jnthn stones masak
21:08 moritz masak is now stoned.
21:09 flussence jnthn's lost his marbles
21:09 masak jnthn: ow!
21:09 sjn you guys rock!
21:09 * masak is not giving back those pebbles to jnthn
21:09 Tene man, I wish it would rain here
21:09 * jnthn hopes we won't be charged with basalt
21:11 sjn jnthn: I gravel before your pun-ability
21:12 * jnthn is, once again, karst as a punster
21:13 sjn those are true tailings, they are!
21:14 [Coke] Takes real grit to put up with all these puns.
21:14 sjn boulder get over the fact you're a punster, jnthn
21:14 jnthn I'm sure I shale
21:15 masak your joking about this serious matter is not coal.
21:15 tadzik jnthn++ #savior!
21:15 phenny tadzik: 17:34Z <jnthn> tell tadzik that memory issue in Q is gone (need latest NQP too).
21:16 * tadzik updates ALL THE THINGS
21:16 tadzik I can work on Q tomorrow then :)
21:16 masak maybe we should just wipe the slate and move on.
21:16 sjn jnthn: ore you doubting yourself? I think you should crust in your insticts about this. :)
21:16 * masak .oO( thought your mileage may variolite )
21:17 jnthn Anyone want to chalk up some more puns? :D
21:17 tadzik my head asplodes by now
21:18 jnthn Just don't fall over; it's suck if your asphalt sore...
21:18 jnthn *it'd
21:21 sjn sand that would hurt a lot, believe me! :)
21:22 * masak .oO( no need to rub it in )
21:22 sjn masak: earth you not happy with our puns? :)
21:24 masak they are dirt to me.
21:24 masak except the slightly moist ones, which are mud to me.
21:24 tadzik stahp
21:24 masak :P
21:24 tadzik I'm not capable of catching up now :)
21:25 sjn masak: I'm sure hummus be happy with *some* of the puns!
21:25 masak sjn: ...no sense of hummur...
21:25 skids I think we have finished mining this vein of conversation.
21:25 masak we've hit rock bottom.
21:25 jnthn Not my fault.
21:26 masak *groan*
21:26 sjn hehehe
21:26 skids There had to be a nugget eventually.
21:28 masak "nugget". could be a nice name for a programming language.
21:28 jnthn "Tastes like chicken"
21:35 _jaldhar joined #perl6
21:38 kiyan joined #perl6
21:40 mdhafen joined #perl6
21:41 mdhafen Hello
21:42 tadzik hello
21:42 mdhafen I'm looking for help with perl6 syntax.  Anyway game?
21:43 jnthn Most likely :)
21:43 pmurias joined #perl6
21:44 mdhafen ok then.  What I'm trying to do is a hash of hashes.  In perl5 I could do '%h = ( 1 => { name => 'horse', size => 12 }, 2 => {name => 'dog', size => 5 } )'  How do I do that in Perl6?
21:45 flussence r: my %h = ( 1 => { name => 'horse', size => 12 }, 2 => {name => 'dog', size => 5 } ); say %h.perl
21:45 p6eval rakudo 34febe: OUTPUT«("1" => {"name" => "horse", "size" => 12}, "2" => {"name" => "dog", "size" => 5}).hash␤»
21:46 jnthn Just like that :)
21:46 TimToady don't need the parens though
21:46 jnthn r: my %h = 1 => { name => 'horse', size => 12 }, 2 => {name => 'dog', size => 5 }; say %h.perl # hah, TimToady++ beat me to it :)
21:46 p6eval rakudo 34febe: OUTPUT«("1" => {"name" => "horse", "size" => 12}, "2" => {"name" => "dog", "size" => 5}).hash␤»
21:47 mdhafen I'll try it.
21:47 fgomez joined #perl6
21:49 mdhafen hmm, maybe my problem was actually something else.  figures.  I had life-span => 6x365, looking at it that is obviously wrong.
21:54 mdhafen yeah, I fixed those statements and the error is gone.  Thanks for the help.
21:58 mdhafen left #perl6
22:05 _jaldhar joined #perl6
22:07 kaare_ joined #perl6
22:09 bluescreen10 joined #perl6
22:14 jaldhar_ joined #perl6
22:15 arlinius joined #perl6
22:16 masak cool, someone we don't know is using Perl 6 :P
22:16 diakopter it escaped irc?
22:17 sivoais joined #perl6
22:17 jnthn We already have a darkpan... :P
22:17 diakopter hopefully it won't be a flash in the pan
22:18 masak flashes are seldom dark.
22:25 sudokode joined #perl6
22:43 kiyan left #perl6
22:45 masak 'night, #perl6
23:01 stocks joined #perl6
23:06 tokuhiro_ joined #perl6
23:11 jaldhar_ joined #perl6
23:18 japhb pmichaud, It was finally the Lego robotics class in which they started teaching the visual programming language.  It's ... a little odd to work in, but not (quite) as horrid as I was afraid.  I can understand why your son really wanted to just program in text, though.  :-)
23:20 sudokode joined #perl6
23:21 araujo joined #perl6
23:21 araujo joined #perl6
23:24 jnthn sleep &
23:38 tokuhiro_ joined #perl6
23:49 jaldhar_ joined #perl6
23:51 daniel-s_ joined #perl6
23:53 Pleiades` joined #perl6

| Channels | #perl6 index | Today | | Search | Google Search | Plain-Text | summary

Perl 6 | Reference Documentation | Rakudo