Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2013-09-15

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:03 pdurbin joined #perl6
00:12 Psyche^ joined #perl6
00:14 Mouq wtf? How do you get the length of Str?
00:15 Mouq kk
00:16 Mouq r: "a→b".encode("utf8").bytes.say
00:16 camelia rakudo 67c6fd: OUTPUT«5␤»
00:17 Mouq Good call, Perl 6. Still wish it was more obvious
00:18 geekosaur length is not as obvious as you think. bytes? codepoints? graphemes?
00:18 geekosaur you must declare which you mean because all attempts at figuring it out met very sad ends in bewildering behavior
00:19 Mouq I know. It would be nice if Perl would tell me as much when I try to do "abcd".length or something
00:20 Mouq So the "Oh, duh" moment I just had would be automated :)
00:20 geekosaur mm. that might be fileable as an LTA bug
00:22 Mouq r: "a→b".codes
00:22 camelia rakudo 67c6fd:  ( no output )
00:22 Mouq r: "a→b".codes.say # Yay
00:22 camelia rakudo 67c6fd: OUTPUT«3␤»
00:24 BenGoldberg r: "a→b".graphemes.say
00:24 camelia rakudo 67c6fd: OUTPUT«No such method 'graphemes' for invocant of type 'Str'␤  in block  at /tmp/BvmkZcdM22:1␤␤»
00:26 BenGoldberg r: my @a = ("foo", "bar"); @a[1] := @a; print @a
00:26 camelia rakudo 67c6fd: OUTPUT«(timeout)foofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoof…
00:26 cognominal joined #perl6
00:27 BenGoldberg How would one go about counting the number of graphemes in a string?
00:27 lee_ r: "a→b".chars.say
00:27 camelia rakudo 67c6fd: OUTPUT«3␤»
00:27 lee_ maybe that?
00:29 BenGoldberg How does .codes differ from .chars?
00:31 benabik joined #perl6
00:32 lee_ they appears to be implemented exactly the same in Cool.pm
00:37 Mouq rnp: 1.say
00:37 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f, pugs: OUTPUT«1␤»
00:38 Mouq rnp: "a" ~~ tr/a/b/
00:38 camelia pugs: OUTPUT«*** Can't modify constant item: VStr "a"␤    at /tmp/FbNGtW2VCN line 1, column 1 - line 2, column 1␤»
00:38 camelia ..rakudo 67c6fd: OUTPUT«Cannot modify an immutable value␤  in block  at /tmp/5MqozMZ8FM:1␤␤»
00:38 camelia ..niecza v24-95-ga6d4c5f: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤Action method quote:tr not yet implemented at /tmp/NkLUtWdi4w line 1 (EOF):â�¤------> [32m"a" ~~ tr/a/b/[33mâ��[31m<EOL>[0mâ�¤â�¤Unhandled exception: Check failedâ�¤â�¤  at /home/p6eval/niecza/boot/lib/CORE.setting line 1502 (die @ …
00:39 Mouq rnp: (my $s = "a") ~~ tr/a/b/
00:39 camelia niecza v24-95-ga6d4c5f: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤Action method quote:tr not yet implemented at /tmp/3AyTsSIKuJ line 1 (EOF):â�¤------> [32m(my $s = "a") ~~ tr/a/b/[33mâ��[31m<EOL>[0mâ�¤â�¤Potential difficulties:â�¤  $s is declared but not used at /tmp/3AyTsSIKuJ line 1:â�¤------> …
00:39 camelia ..rakudo 67c6fd, pugs:  ( no output )
00:43 ssutch joined #perl6
00:56 Mouq BenGoldberg: I may have your issue solved, give me some time to test the code
00:59 benabik joined #perl6
01:05 slavik joined #perl6
01:11 dalek evalbot: d0cc1d3 | Mouq++ | / (2 files):
01:11 dalek evalbot: Formatting of output doesn't belong in executor
01:11 dalek evalbot:
01:11 dalek evalbot: Moved output formatting from EvalbotExecuter to Evalbot. The strange
01:11 dalek evalbot: placement of formatting in EvalbotExecuter made the code confusing and
01:11 dalek evalbot: caused Issue #2.
01:11 dalek evalbot: review: https://github.com/perl6/evalbot/commit/d0cc1d345f
01:14 Mouq If that code didn't break it, it fixed it :p I presume moritz hosts the evalbot and will have to update it manually
01:23 BenGoldberg Mouq++
01:27 colomon t/spec/S32-str/substr-rw.t failing in Rakudo JVM?
01:38 camelia joined #perl6
01:38 diakopter r: Mouq
01:38 diakopter /o\
01:39 camelia rakudo 67c6fd: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/THa5MkibxIâ�¤Undeclared name:â�¤    Mouq used at line 1â�¤â�¤Â»
01:39 diakopter rn: say 1..100
01:39 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«1..100␤»
01:39 diakopter rn: .say for 1..*
01:40 camelia rakudo 67c6fd: OUTPUT«(timeout)1â�¤2â�¤3â�¤4â�¤5â�¤6â�¤7â�¤8â�¤9â�¤10â�¤11â�¤12â�¤13â�¤14â�¤15â�¤16â�¤17â�¤18â�¤19â�¤20â�¤21â�¤22â�¤23â�¤24â�¤25â�¤26â�¤27â�¤28â�¤29â�¤30â�¤31â�¤32â�¤33â�¤34â�¤35â�¤36â�¤37â�¤38â�¤39â�¤40â�¤41â�¤42â�¤43â�¤44â�¤45â�¤46â�¤47â�¤48â�¤49â�¤50â�¤51â�¤52â�¤53â�¤54â�¤55â�¤56â�¤57â�¤58â�¤59â�¤60â�¤61â�¤62â�¤63â�¤64â�¤65â�¤66â�¤67â�¤68â�¤69â�¤70â�¤71â�¤72â�¤73â�¤74â�¤75â�¤76â�¤77â�¤78â�¤79â�¤80â�¤81â�
01:40 camelia ..niecza v24-95-ga6d4c5f: OUTPUT«(timeout)1â�¤2â�¤3â�¤4â�¤5â�¤6â�¤7â�¤8â�¤9â�¤10â�¤11â�¤12â�¤13â�¤14â�¤15â�¤16â�¤17â�¤18â�¤19â�¤20â�¤21â�¤22â�¤23â�¤24â�¤25â�¤26â�¤27â�¤28â�¤29â�¤30â�¤31â�¤32â�¤33â�¤34â�¤35â�¤36â�¤37â�¤38â�¤39â�¤40â�¤41â�¤42â�¤43â�¤44â�¤45â�¤46â�¤47â�¤48â�¤49â�¤50â�¤51â�¤52â�¤53â�¤54â�¤55â�¤56â�¤57â�¤58â�¤59â�¤60â�¤61â�¤62â�¤63â�¤64â�¤65â�¤66â�¤67â�¤68â�¤69â�¤70â�¤71â�¤72â�¤73â�¤74â�¤75â�¤76â�¤77â�¤78â�¤79â
01:40 diakopter Mouq: didn't seem to fix it
01:44 BenGoldberg Just for fun, who can guess what this function is? :)
01:44 BenGoldberg r: sub A($m is copy, $n is copy) { state %c; my $key = $m => $n; while ( $m > 3 ) { return %c{$m => $n} if %c{$m => $n}; $n = $n ?? A($m, $n-1) !! 1; --$m }; %c{$key} = $m == 3 ?? 2**($n+3)-3 !! $m == 2 ?? 2*$n+3 !! $n+$m+1; }; say A(4,2)
01:44 camelia rakudo 67c6fd: OUTPUT«200352993040684646497907235156025575044782547556975141926501697371089405955631145308950613088093334810103823434290726318182294938211881266886950636476154702916504187191635158796634721944293092798208430910485599057015931895963952486337236720300291696959215610876…»
01:55 BenGoldberg Ok, so either it's not sufficiently fun, or I've optimized to the point of being unrecognizable :)
01:55 BenGoldberg Here's a slightly less optimized version of the mystery A function:
01:55 BenGoldberg r: sub A($m is copy, $n is copy) { state %c; my $key = $m => $n; while ( $m > 1 ) { return %c{$m => $n} if %c{$m => $n}; $n = $n ?? A($m, $n-1) !! 1; --$m }; %c{$key} = $n+$m+1; }; say A(3,$_) for 0..12
01:55 camelia rakudo 67c6fd: OUTPUT«5␤13␤29␤61␤125␤253␤509␤1021␤2045␤4093␤8189␤16381␤32765␤»
02:00 BenGoldberg r: sub A($m is copy, $n is copy) { while ( $m > 0 ) { $n = $n ?? A($m, $n-1) !! 1; --$m }; $n+1; }; say A(3,$_) for 0..12
02:00 camelia rakudo 67c6fd: OUTPUT«(timeout)5␤13␤29␤61␤125␤253␤509␤»
02:03 BenGoldberg r: sub A($m, $n) { $m > 0 ?? A( $m-1, $n ?? A($m, $n-1) !! 1 ) !! $n+1; }; say A(3,$_) for 0..12
02:03 camelia rakudo 67c6fd: OUTPUT«(timeout)5␤13␤29␤61␤125␤253␤509␤»
02:03 diakopter fizzbuzz?
02:04 BenGoldberg ?
02:16 sizz_ joined #perl6
02:17 ingy BenGoldberg: http://cpansearch.perl.org/src/INGY/Language-SNUSP-0.0.3/examples/fizzbuzz.snusp :)
02:17 ingy I should port Language::SNUSP to p6
02:19 BenGoldberg No, it's not fizzbuzz... it's the ackermann function :)
02:22 Mouq rn: .say for 1..200
02:23 BenGoldberg r: 'alive'.say
02:23 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«1â�¤2â�¤3â�¤4â�¤5â�¤6â�¤7â�¤8â�¤9â�¤10â�¤11â�¤12â�¤13â�¤14â�¤15â�¤16â�¤17â�¤18â�¤19â�¤20â�¤21â�¤22â�¤23â�¤24â�¤25â�¤26â�¤27â�¤28â�¤29â�¤30â�¤31â�¤32â�¤33â�¤34â�¤35â�¤36â�¤37â�¤38â�¤39â�¤40â�¤41â�¤42â�¤43â�¤44â�¤45â�¤46â�¤47â�¤48â�¤49â�¤50â�¤51â�¤52â�¤53â�¤54â�¤55â�¤56â�¤57â�¤58â�¤59â�¤60â�¤61â�¤62â�¤63â�¤64â�¤65â�¤66â�¤67â�¤68â�¤69â�¤70â�¤71â�¤72â�¤73â�¤74â�¤75â�¤76â�¤77â�¤78â�
02:23 camelia rakudo 67c6fd: OUTPUT«alive␤»
02:23 BenGoldberg Mouq++
02:24 Mouq Are you guys also getting Unknown Character (diamond question mark) symbols?
02:24 flussence yeah
02:25 flussence r: say '»'.encode
02:25 BenGoldberg The << symbol shows up ok
02:25 camelia rakudo 67c6fd: OUTPUT«Buf:0x<c2 bb>␤»
02:25 flussence r: say '␤'.encode
02:25 camelia rakudo 67c6fd: OUTPUT«Buf:0x<0a>␤»
02:25 Mouq Yeah, I messed something up with "…"
02:25 flussence .u ␤
02:25 yoleaux U+2424 SYMBOL FOR NEWLINE [So] (␤)
02:25 flussence r: say "\x2424".encode
02:25 camelia rakudo 67c6fd: OUTPUT«Buf:0x<e2 90 a4>␤»
02:25 flussence it's cutting that off, that'd be it
02:26 flussence off-by-3, seems like
02:26 Mouq Ohhhh, thank you. Hmm
02:30 BenGoldberg r: say "â".encode
02:30 camelia rakudo 67c6fd: OUTPUT«Buf:0x<c3 a2 c2 90>␤»
02:31 BenGoldberg r: say "…".encode
02:31 camelia rakudo 67c6fd: OUTPUT«Buf:0x<e2 80 a6>␤»
02:40 Mouq StackOverflow++ http://stackoverflow.com/questions/10953069/
02:48 lue ingy: I've already written a SNUSP in Perl 6. IIRC I just need to touch it up a bit in the usability department...
02:50 BenGoldberg Oh, before I forget...
02:50 BenGoldberg r: say map { $_ % 15 ?? $_ % 3 ?? $_ % 5 ?? $_ !! 'buzz' !! 'fizz' !! 'fizzbuzz' }, 1..100
02:50 camelia rakudo 67c6fd: OUTPUT«1 2 fizz 4 buzz fizz 7 8 fizz buzz 11 fizz 13 14 fizzbuzz 16 17 fizz 19 buzz fizz 22 23 fizz buzz 26 fizz 28 29 fizzbuzz 31 32 fizz 34 buzz fizz 37 38 fizz buzz 41 fizz 43 44 fizzbuzz 46 47 fizz 49 buzz fizz 52 53 fizz buzz 56 fizz 58 59 fizzbuzz 61 62 fizz 64 …»
02:54 BenGoldberg preflex: karma perl
02:54 preflex perl: 270
02:54 BenGoldberg preflex: karma perl6
02:54 preflex perl6: 11
02:54 dalek evalbot: 9ca77f7 | Mouq++ | evalbot.pl:
02:54 dalek evalbot: Truncate utf8 compatibly
02:54 dalek evalbot: review: https://github.com/perl6/evalbot/commit/9ca77f7d16
02:54 camelia : ( no output )
02:54 BenGoldberg preflex: karma rakudo
02:54 preflex rakudo: 108
02:54 camelia : ( no output )
02:55 BenGoldberg preflex: karma n
02:55 preflex n: 158
02:55 camelia : ( no output )
02:55 BenGoldberg preflex: karma r
02:55 Mouq To be proper, ikegami++
02:55 preflex r: 25
02:55 camelia : ( no output )
02:56 BenGoldberg Next issue.... make evalbot ignore bots :)
03:04 Mouq It already ignores dalek
03:06 Mouq hugme: hug r: 1.say
03:06 * hugme hugs r:
03:06 Mouq hugme: hug r:1.say
03:06 * hugme hugs r:1.say
03:06 Mouq Yeah
03:07 dalek evalbot: dbc13cc | Mouq++ | evalbot.pl:
03:07 dalek evalbot: Ignore the preflex and yoleaux bots
03:07 dalek evalbot: review: https://github.com/perl6/evalbot/commit/dbc13cccb4
03:07 * sorear leers at BenGoldberg
03:09 preflex_ joined #perl6
03:13 Mouq rn: .say for ^200
03:13 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«0␤1␤2␤3␤4␤5␤6␤7␤8␤9␤10␤11␤12␤13␤14␤15␤16␤17␤18␤19␤20␤21␤22␤23␤24␤25␤26␤27␤28␤29␤30␤31␤32␤33␤34␤35␤36␤37␤38␤39␤40␤41␤42␤43␤44␤45␤46␤47␤48␤49␤50␤51␤52␤53␤54␤55␤56␤57␤58␤59␤60␤61␤62␤63␤64␤65␤66␤67␤68␤69␤70␤71␤72␤73␤74␤75␤76␤77␤
03:13 Mouq …interesting
03:14 geekosaur sometimes I regret the xchat plugin I did that translates embedded control chars...
03:15 Mouq rn: ^200.say
03:15 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«200␤»
03:15 Mouq rn: (^200).say
03:15 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«0..^200␤»
03:15 Mouq rn: (^200).eager.say
03:15 camelia niecza v24-95-ga6d4c5f: OUTPUT«Unhandled exception: Unable to resolve method eager in type Range␤  at /tmp/ku1D_1WOdC line 1 (mainline @ 3) ␤  at /home/p6eval/niecza/lib/CORE.setting line 4583 (ANON @ 3) ␤  at /home/p6eval/niecza/lib/CORE.setting line 4584 (module-CORE @ 576) ␤  at …»
03:15 camelia ..rakudo 67c6fd: OUTPUT«0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 9…»
03:16 Mouq rn: .say for ^200
03:16 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«0␤1␤2␤3␤4␤5␤6␤7␤8␤9␤10␤11␤12␤13␤14␤15␤16␤17␤18␤19␤20␤21␤22␤23␤24␤25␤26␤27␤28␤29␤30␤31␤32␤33␤34␤35␤36␤37␤38␤39␤40␤41␤42␤43␤44␤45␤46␤47␤48␤49␤50␤51␤52␤53␤54␤55␤56␤57␤58␤59␤60␤61␤62␤63␤64␤65␤66␤67␤68␤69␤70␤71␤72␤73␤74␤75␤76␤77␤
03:17 geekosaur are you missing a ... or dd my plugin screw up?
03:17 Mouq rn: (0 «+« ^200).say
03:17 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 8…»
03:18 Mouq I'm mising a "…". Something's going very wrong with the truncation...
03:19 BenGoldberg rn say join ' ', ^200
03:19 BenGoldberg rn: say join ' ', ^200
03:19 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 8…»
03:21 Mouq rn: say join '→', ^200
03:21 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«0→1→2→3→4→5→6→7→8→9→10→11→12→13→14→15→16→17→18→19→20→21→22→23→24→25→26→27→28→29→30→31→32→33→34→35→36→37→38→39→40→41→42→43→44→45→46→47→48→49→50→51→52→53→54→55→56→57→58→59→60→61→62→63→64→65→66→67→68→69→70→71→72→73→74→75→76→77→
03:21 Mouq rn: say join 'λ', ^200
03:21 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«0λ1λ2λ3λ4λ5λ6λ7λ8λ9λ10λ11λ12λ13λ14λ15λ16λ17λ18λ19λ20λ21λ22λ23λ24λ25λ26λ27λ28λ29λ30λ31λ32λ33λ34λ35λ36λ37λ38λ39λ40λ41λ42λ43λ44λ45λ46λ47λ48λ49λ50λ51λ52λ53λ54λ55λ56λ57λ58λ59λ60λ61λ62λ63λ64λ65λ66λ67λ68λ69λ70λ71λ72λ73λ74λ75λ76λ77λ78λ79λ80λ81λ8…»
03:25 Mouq rn: say join '????', ^200
03:25 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«0????1????2????3????4????5????6????7????8????9????10????11????12????13????14????15????16????17????18????19????20????21????22????23????24????25????26????27????28????29????30????31????32????33????34????35????36????37????38????39????40????41????42????43????44????45????46????47????48????49????50????51????52????53????54????55????56????57????58????59????60????61????62????63????64????
03:25 BenGoldberg Is it my imagination, or is truncation happening based on the number of bytes of the character you're joining on?
03:25 Mouq Well, it's something to do with it. I just want it truncate near a high-byte character
03:26 geekosaur I suspect it's trying to not break in the middle of a multiyte character and getting it wrong when it needs to adjust?
03:26 Mouq rn: say '????' x 200
03:26 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????
03:29 dalek evalbot: f6af675 | Mouq++ | evalbot.pl:
03:29 dalek evalbot: Do truncation stupider.
03:29 dalek evalbot: review: https://github.com/perl6/evalbot/commit/f6af675f1f
03:31 Mouq Oh, wait… This might be a fencepost error :/
03:34 Mouq rn: say '????' x 200
03:34 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????
03:43 Mouq rn: say '????' x 200
03:43 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????
03:47 Mouq rn: say 'a' x 200, '????'
03:47 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa????␤»
03:47 Mouq rn: say 'a' x 300, '????'
03:47 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa…»
03:47 Mouq rn: say 'a' x 250, '????'
03:47 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa…»
03:47 Mouq rn: say 'a' x 220, '????'
03:47 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa????␤»
03:47 Mouq rn: say 'a' x 221, '????'
03:47 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa????␤»
03:47 Mouq rn: say 'a' x 222, '????'
03:47 camelia rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa????␤»
03:47 Mouq Sorry, #perl6, I'm kinda spamming you here
03:48 geekosaur I believe evalbot responds to /msg
03:48 geekosaur (to camelia, per the /topic)
03:49 Mouq '< Thank you. I'm sorry
03:54 ingy o_0
04:25 Mouq rnp: "\c[ALPHA]".say
04:25 camelia rakudo 67c6fd: OUTPUT«(timeout)»
04:25 camelia ..niecza v24-95-ga6d4c5f: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤Unrecognized character name ALPHA at /tmp/aKilrvmqh8 line 1:â�¤------> [32m"\c[ALPHA[33mâ��[31m]".say[0mâ�¤â�¤Unhandled exception: Check failedâ�¤â�¤  at /home/p6eval/niecza/boot/lib/CORE.setting line 1502 (die @ 5) â�¤  at /home/…»
04:25 camelia ..pugs: OUTPUT«Unknown charname 'ALPHA' at (eval 2) line 1␤\c[ALPHA]␤»
04:28 Mouq rnp: "\c[SYMBOL FOR NULL]".say
04:28 camelia pugs: OUTPUT«\c[SYMBOL FOR NULL]␤»
04:28 camelia ..rakudo 67c6fd, niecza v24-95-ga6d4c5f: OUTPUT«␀␤»
04:34 btyler joined #perl6
04:46 SamuraiJack__ joined #perl6
05:04 daniel-s_ joined #perl6
05:40 FROGGS[mobile] joined #perl6
05:47 SamuraiJack joined #perl6
05:51 FROGGS good morning
05:53 sorear o/ FROGGS
05:54 FROGGS hai sorear :o)
05:55 FROGGS sorear: sad that you are not here
05:55 sorear i'm right here, no?
05:55 FROGGS *g*
05:55 FROGGS somehow
05:56 sorear oh, is this that internals workshoppy thing I heard about by emial?
05:56 FROGGS yes
05:56 FROGGS and jnthn++ is the most awesome teacher I've ever had
05:59 FROGGS k, I need to get going... see you
06:17 ingy heya sorear
06:37 xinming joined #perl6
06:39 sorear o/
06:41 sqirrel joined #perl6
06:41 sorear o/ sqirrel
06:42 sqirrel o/ sorear
06:45 iSlug joined #perl6
06:56 timotimo ohai
06:57 * [Coke] yawns
06:57 dalek panda: da441fd | tadzik++ | rebootstrap.pl:
06:57 dalek panda: Bla bla bla chocolatecake. It's early
06:57 dalek panda: review: https://github.com/tadzik/panda/commit/da441fddcc
07:00 [Coke] tadzik++
07:01 * [Coke] switches partcl-nqp to try to build off nqp-jvm instead of nqp-parrot.
07:01 tadzik are you able to run nqp-jvm outside of nqp-jvm directory?
07:02 FROGGS joined #perl6
07:03 [Coke] I'm updating gen-nqp to work more like rakudo's, so I'll have an install-jvm dir, and then it won't matter.
07:03 tadzik ossum
07:03 [Coke] ... except I can't git from the back row!
07:04 tadzik not even through https?
07:04 [Coke] .to diakopter - ping us if you need directions.
07:04 yoleaux [Coke]: I'll pass your message to diakopter.
07:06 FROGGS ribasushi: the ast dumper will be invoked when you all `nqp --target=ast -e "say(1)"` for example
07:07 lizmat joined #perl6
07:09 woolfy joined #perl6
07:09 [Coke] hoelzro: trouble maker! ;)
07:10 timotimo i'm not a fan of american-sized softdrinks, but 0.2l is really not much
07:10 ingy good morning lizmat!
07:10 lizmat morning ingy
07:11 ingy timotimo: you need 2.0l
07:11 [Coke] US standard is 0.6l
07:11 woolfy Hey, diakopter is still not here (with physical us in Ökohaus). Anybody saw him?
07:11 hoelzro heh
07:11 [Coke] (for a single serve) though we are seeing smaller and smaller serving sizes in those, too.
07:14 woolfy jnthn started 10 minutes ago with day 2 of the RaNiW course in Frankfurt.
07:14 woolfy 16 people attending
07:15 [Coke] r: say "RaNiW".flip
07:15 camelia rakudo 67c6fd: OUTPUT«WiNaR␤»
07:16 timotimo r: say "RaNIW".comb.pick(*).join: ""
07:16 camelia rakudo 67c6fd: OUTPUT«RWIaN␤»
07:16 timotimo r: say "RaNIW".comb.pick(*).join: ""
07:16 camelia rakudo 67c6fd: OUTPUT«NIaRW␤»
07:16 timotimo r: say "RaNIW".comb.pick(*).join: ""
07:16 camelia rakudo 67c6fd: OUTPUT«NaWRI␤»
07:17 tadzik NaRNia
07:17 timotimo RAWRRAWR
07:28 timotimo Error while compiling op if (source text: "@args[1].has_compile_time_value {\n                $hint := pir::repr_hint_for_i_p_p_s(@args[1].comp..."): Error while compiling op bind (source text: ":="): ResizableStringArray: Can't shift from an empty array!
07:28 timotimo hooray! errors in the compiler!
07:28 iSlug joined #perl6
07:28 tadzik HyperWhatever, it sounds like a most boring superhero ever
07:29 hoelzro tadzik: you saw that too? =)
07:29 tadzik yep :)
07:32 lizmat .NYI also
07:32 lizmat an unimplemented boring superhero :-)
07:33 hoelzro that's what caught my eye
07:33 hoelzro something to work on =)
07:36 [Sno] joined #perl6
07:38 woolfy yeah diakopter is awake
07:38 timotimo http://questhub.io/realm/perl/quest/52356423a7bb212f76000048 <- :)
07:39 tadzik hah!
07:40 timotimo if you're interested, i can "invite"
07:40 tadzik I "Like"d it
07:40 timotimo \o/
07:41 hoelzro ditto
07:41 hoelzro I don't the have NQP-fu to help (yet)
07:42 tadzik you wan't have that excuse after today :>
07:43 hoelzro heh
07:43 hoelzro that's why I say "yet"!
07:43 timotimo the optimizer isn't super hard to use, fortunately
07:45 timotimo oooh, dynamic variables inside tokens are available in action methods, still
07:45 timotimo that's cool
07:45 timotimo "keinen deckel drauf tun"
07:49 timotimo maybe somebody should build a quest for removing .node from QAST::Node and putting it into QAST::Stmt (and removing all related fallout)
07:50 moritz why?
07:50 moritz it helps to have a line number for calls, for example
07:50 moritz and a call is a QAST::Op
07:50 timotimo jnthn hinted it may be sensible; in the jvm backend apparently it's only ever used ni Stmt
07:50 * hoelzro .oO( QAST::Away? )
07:51 timotimo WILSOOOOOOOOoooooooon~
07:53 timotimo moritz: so having it in the Stmt node around it will not be enough?
07:53 timotimo i suppose that's valuable if you have a top-level call that is split into multiple lines?
07:53 * [Coke] wonders if he can just lift ConfigureJVM.pl from rakudo for use in partcl
07:54 [Coke] (as long as the copyright stays in place)
07:54 xenoterracide_ joined #perl6
07:54 moritz timotimo: correct. If you have foo(\n bar()), the bar() call would show a wrong line number
07:55 timotimo i still need some way to differentiate vars that are from signatures from other vars; at least in nqp, but ideally also in rakudo
07:55 timotimo (this is for the lexical-to-local transformation i would like to make work)
08:00 [Coke] weird. getting http://irclog.perlgeek.de/perl6/2013-05-23 using the jvm nqp for partcl (but it works fine for rakudo-jvm) ... will dig after class.
08:01 timotimo did you mean to link to a specific time in that log? or to something else entirely?
08:03 [Coke] the first line.
08:03 [Coke] same thing that japhb was experiencing at http://irclog.perlgeek.de/perl6/2013-05-23#i_7105647
08:03 timotimo i think install/nqp is deprecated
08:03 timotimo install/bin/nqp is the way to go iirc
08:05 [Coke] timotimo: there is no install/bin/nqp on jvm here.
08:05 timotimo hm, i'll go hunt
08:06 [Coke] er, no install-jvm/bin/nqp
08:06 * timotimo has to re-make
08:07 [Coke] I am fairly sure that I should throw out the nqp-rx (parrot style) nqp code that was translated to nqp (current) style, and just start from scratch.
08:07 timotimo huh, it actually does say :decl(param)
08:09 denis_boyun joined #perl6
08:13 timotimo oh
08:13 timotimo you have to pass an absolute value to --gen-nqp if you use --gen-nqp
08:13 timotimo that may be your problem, [Coke]
08:14 timotimo but i think it only makes a difference in the actual source of the nqp file, you should be able to make it look like this: exec java -Xbootclasspath/a:/home/timo/build/jakudo/nqp/install:/home/timo/build/jakudo/nqp/install/nqp-runtime.jar:/home/timo/build/jakudo/nqp/install/asm-4.1.jar:/home/timo/build/jakudo/nqp/install/jline-1.0.jar:/home/timo/build/jakudo/nqp/install/nqp.jar -cp /home/timo/build/jakudo/nqp/install nqp "$@"
08:15 dalek v5: e89c54f | (Tobias Leich)++ | / (2 files):
08:15 dalek v5: move Perl5::Terms loading to v5's grammar, jnthn++
08:15 dalek v5: review: https://github.com/rakudo-p5/v5/commit/e89c54f25c
08:16 tadzik what if you don't pass any value to --gen-nqp? Then the default is rong
08:18 timotimo woolfy: is there a way to acquire more coca cola for people who don't like coffee, like me?
08:19 woolfy In your book, go forward 13 pages...  'That's all for today":  "Today we've covered a lot of ground"..."We'll take these on tomorrow."  Next page: "Welcome back"
08:19 woolfy timotimo: I will ask
08:19 timotimo that is 13 A4-pages, right? not 13 slides? >_<
08:19 timotimo thank you very much :)
08:20 woolfy Indeed, physicla pages
08:20 timotimo i'm much less tired than yesterday, but that doesn't mean too much ;)
08:20 tadzik the material is a bit more hardcore, imo
08:20 timotimo currently it's chewy, it'll get much harder soon i fear
08:20 timotimo at least to me, i've had lots of exposure to QAST from my optimizer adventures
08:21 woolfy I;ve seen jnthn talk about QAST before and it puzzled me then and it puzzles me now.
08:21 timotimo the thick green ink for the titles on the slides is rubby
08:24 moritz woolfy: it's one of those things you just have to use for a while to get used to
08:24 moritz woolfy: like monads, I guess (haven't ever gotten used to them)
08:24 timotimo it helps to look at lots of qast dumps from programs
08:24 timotimo monads are just like burritos
08:26 moritz timotimo: and that's the problem. I don't know how to program with burritos :-)
08:26 PZt joined #perl6
08:27 moritz .oO( burrito-driven development )
08:27 diakopter moritz: lots of sour cream
08:27 yoleaux 07:04Z <[Coke]> diakopter: - ping us if you need directions.
08:27 SamuraiJack_ joined #perl6
08:29 kaare_ joined #perl6
08:32 timotimo colomon: may i suggest an enhancement to the smoker web app?
08:32 [Coke] diakopter: you coming over?
08:32 timotimo it would be neat if the leftmost column would display a number instead of a letter to indicate how many days the status has been the status of the column one to the right
08:33 diakopter brt
08:33 [Coke] diakopter: \o/
08:33 timotimo because with the speed of change in the ecosystem, it seems like 14 days is not that much
08:34 diakopter heh
08:34 diakopter it moves cyclicly
08:34 [Coke] java.lang.RuntimeException: java.lang.ClassCastException: org.perl6.nqp.sixmodel.reprs.P6strInstance cannot be cast to org.perl6.nqp.runtime.EvalResult
08:35 [Coke] :(
08:36 woolfy joined #perl6
08:40 lizmat joined #perl6
08:43 timotimo saw that, too ... for some reason :|
08:43 timotimo hold tight, i'll post a keyword token for php
08:45 timotimo http://pastebin.com/dytfHWhg
08:46 FROGGS hmmm, sizeof if missing...
08:47 timotimo oh?
08:47 timotimo i suppose predefined functions are missing >_>
08:49 FROGGS well, it is listed as an alias to count()... but w/e, it doesn't surprise me that there is no complete list of built-ins :o)
08:50 timotimo yes.
08:50 timotimo also, are those things case-insensitive?
08:53 [Coke] (some reason) jnthn pointed out I had an action referring to a sub-rule with a wrong name.
08:53 [Coke] Now I'm getting a NPE trying to run echo(3*2);
08:53 [Coke] ... because I'm an idiot. :)
08:54 [Coke] \o/
08:54 wtw joined #perl6
08:55 FROGGS *g*
08:55 FROGGS --ll-exception is your friend
08:56 woolfy diakopter is in the building
08:56 woolfy all welcome diakopter
08:57 wtw joined #perl6
09:00 wtw joined #perl6
09:02 hoelzro fun with dynamic variables in NQP: https://gist.github.com/hoelzro/6569041
09:02 hoelzro is this a bug? or am I Doing It Wrong?
09:02 timotimo i wonder if trying to split Stmts into individual Stmt nodes during optimization time will have a noticable effect on performance or quality of generated code
09:03 FROGGS hoelzro: if you redeclare it is is a new one
09:03 hoelzro ah
09:03 hoelzro so I should just do $*TOP = 0?
09:03 FROGGS yes, that should do
09:03 hoelzro cool, thanks!
09:03 timotimo yeah, you can do "temporary overrides" with a new my $*foo
09:04 timotimo that's the mechanism that's being used for the current block
09:04 FROGGS hoelzro: sine the new one is installed in the stmts before the actual code of the sub, only the new one is visible inside
09:04 diakopter woohoo hack the planet
09:04 FROGGS since*
09:04 hoelzro FROGGS: ah, that makes sense.
09:12 spider-mario joined #perl6
09:28 dmol joined #perl6
09:28 timotimo tadzik: NQP/Actions.nqp line 193
09:28 FROGGS tadzik: it is fine in nqp and rakudo
09:28 timotimo already fixed apparently
09:28 FROGGS yeah :o)
09:29 tadzik yeah, noticed that too :)
09:29 tadzik it's also in Perl6/Actions.pm, also $past both places
09:29 timotimo ok
09:29 timotimo maybe jnthn++ was like "$past is stupid, i'll replace it with $ast"
09:29 tadzik :)
09:29 diakopter ()pmohc
09:29 tadzik but not past the for statement
09:30 diakopter er, ()pohc
09:30 lizmat joined #perl6
09:30 timotimo i don't understand what that means :(
09:34 Rotwang joined #perl6
09:36 dalek rakudo/nom: 2230750 | (Elizabeth Mattijsen)++ | / (4 files):
09:36 dalek rakudo/nom: Step one of converting {} and [] access from method to sub calls
09:36 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/2230750d8b
09:36 timotimo oh cool!
09:36 timotimo not paying attention, liz? ;)
09:36 timotimo or did you do all this between sleep and breakfast? ;)
09:38 diakopter r: my Int $jnthn but Dog;
09:38 camelia rakudo 67c6fd: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/RMTmGjf60Uâ�¤Undeclared name:â�¤    Dog used at line 1â�¤â�¤Â»
09:39 timotimo so really you should have an Int for the tax number that you mix your dog into?
09:39 woolfy joined #perl6
09:39 tadzik std: my Int $diakopter but Cat;
09:39 camelia std 7c17586: OUTPUT«ok 00:00 43m␤»
09:40 diakopter O_O
09:40 timotimo mau
09:40 lizmat timotimo: actually last night, just epectested it now
09:40 timotimo not bad
09:40 diakopter my Int $tadzik but Rat;
09:40 timotimo you're not putting it up on questhub? ;)
09:40 tadzik hah
09:40 diakopter r: my Int $tadzik but Rat;
09:40 camelia rakudo 67c6fd: OUTPUT«Cannot mix in non-composable type Rat into object of type Int␤  in sub infix:<but> at src/gen/CORE.setting:15490␤  in sub infix:<but> at src/gen/CORE.setting:15466␤  in block  at /tmp/xtbKYsLxsg:1␤␤»
09:40 tadzik at least not a FatRat
09:40 diakopter damn
09:41 tadzik s/damn!/rats!/
09:43 diakopter why didn't it complain about the 2nd class Dog, but did on the 3rd
09:43 timotimo i guess it error'd out earlier? or something?
09:43 tadzik the 1st dog died
09:43 timotimo :(
09:43 diakopter yeah but its BEGIN-time should've completed
09:44 timotimo too soon
09:44 * timotimo mourns
09:44 tadzik Perl 6 is a middle-sized dog
09:44 diakopter tadzik: you belong on the back row
09:45 tadzik :(
09:46 timotimo but HOW
09:46 tadzik also, How, pronounced /hau/ is the sound that Dog makes in polish
09:46 timotimo how do you describe cat sounds?
09:46 tadzik miau
09:46 timotimo like in german
09:46 timotimo in php it's ... and in perl6 it's ~~~
09:47 diakopter I thought it was Mu
09:47 [Coke] meow
09:47 tadzik mu is the sound a cow makes
09:47 diakopter sea cows make the whales
09:48 [Coke] C cows?
09:48 timotimo i think i may steal more tuits today
09:48 timotimo i shall spread perl6 joy in karlsruhe :3
09:49 timotimo i hear maybe there'll be a karlsruhe.pm meeting some time soon
09:50 * diakopter wonders how in the world I was pushing code 6.5 hours ago
09:50 tadzik HOW
09:50 timotimo DiakopterHOW
09:50 timotimo Perl6::Metamodeler::DiakopterHOW
09:50 diakopter OMFG I thought cows lived in stables
09:51 timotimo HOWs live in stables
09:52 diakopter r: knowhow Tadzik is repr('Uninstantiable') { ... }
09:52 camelia rakudo 67c6fd: OUTPUT«===SORRY!===␤No such method 'is_composed' for invocant of type 'KnowHOW'␤»
09:52 tadzik I'm no composer
09:53 diakopter r: knowhow foo is repr('') { }; say foo.^methods
09:53 camelia rakudo 67c6fd: OUTPUT«No such method 'dispatch:<.^>' for invocant of type 'foo'␤  in block  at /tmp/r6F7DnTZ2r:1␤␤»
09:54 tadzik meth_type("not even once")
09:54 diakopter r: knowhow foo {}; print foo
09:54 camelia rakudo 67c6fd: OUTPUT«Nominal type check failed for parameter '$got'; expected Any but got foo instead␤  in method REIFY at src/gen/CORE.setting:7823␤  in method reify at src/gen/CORE.setting:6754␤  in method gimme at src/gen/CORE.setting:7164␤  in method print at src/gen/CORE.setti…»
09:55 diakopter expected Any but got foo instead
09:55 timotimo this is way too simple, jnthn must be hiding most of the secrets from us!
09:55 timotimo giiiive ussssssssss the secretsssssssss~
09:56 PacoAir joined #perl6
09:56 timotimo we should greet Greta with Greeter
09:57 tadzik there should be find_find_method_method which finds find_method methods
09:57 timotimo it's just the find_method method on the ClassHOW, because you build your ClassHOW out of that ... or possibly KnowhowHOW
09:58 diakopter tadzik: moarvm has that
09:58 tadzik moar find_methodz
09:59 diakopter someone should make a moarvm git branch that injects the word MOAR_ in every function and variable name in 1 or 2 places
09:59 timotimo how do you propose we call that branch?
09:59 timotimo moarvm_moar_moar?
10:00 tadzik http://www.reactionimage.org/img/gallery/5816489306.jpg
10:00 diakopter rotfl literally
10:00 tadzik lies, you still sit
10:00 timotimo nope
10:00 diakopter I finished rolling
10:01 timotimo i didn't hear anything
10:01 diakopter it's a frictionless surface
10:01 timotimo yeah, but were you laughing silently?
10:01 tadzik is this how you roll?
10:02 FROGGS not very impressive O.o
10:02 diakopter i'm enjoying my L of Coke
10:02 diakopter actually jnthn lied
10:03 diakopter moarvm's qast compiler DOES inject coercions for all those ops he said it didn't...
10:03 diakopter I, um, just copied parrot's behavior for that
10:03 tadzik boo
10:03 timotimo heh, great
10:03 FROGGS hehe
10:03 diakopter he must not've read that part
10:03 lizmat little white lies, though
10:03 hoelzro you don't need the know-how to write a KnowHOW?
10:03 FROGGS lets copy parrot \o/
10:03 timotimo at least you'll also benefit from the optimisation in nqp that turns _n into _i if applicable
10:04 FROGGS \o/ copy all teh code \o/
10:04 diakopter however, I didn't look at parrot's qast compiler to do it; I did it Journey's way
10:04 timotimo go on a train going anywhere?
10:04 timotimo get on a train going anywhere*
10:04 diakopter no, go your own way
10:04 [Coke] Enjoying my what now?
10:04 diakopter liter
10:04 timotimo EnjoyCokeHOW
10:04 [Coke] oh, whew.
10:05 [Coke] Responsibly.
10:05 diakopter *litter
10:05 diakopter at least I didn't say line
10:05 diakopter that was earlier this morningj
10:06 timotimo that's why you were late to the workshop
10:06 denis_boyun joined #perl6
10:06 diakopter yes, HOW as a kowt
10:06 timotimo that stretches it a bit too thin
10:08 diakopter I wonder at what point backlogging this channel will crowd out 100% of TimToady's time
10:08 diakopter mberends: stay awake
10:09 [Coke] did he say where we can find a list of what reprs are available and why we'd pick different ones?
10:09 timotimo timtoady will want to evolve to get a very quick heuristic to differentiate into different types:
10:09 timotimo 1) humour
10:09 timotimo 2) not worth reading
10:09 diakopter [Coke]: each has its own file in reprs/
10:09 diakopter each has a comment or two explaining it
10:10 diakopter each's .h file shows what fields its body has
10:10 diakopter so you can see what's stored in each
10:10 timotimo diakopter seems to know everything already!
10:10 diakopter well hey I did write the entire QAST compiler in moar..
10:10 diakopter .. more than half of the reprs
10:11 diakopter everything having to do with strings & encodings
10:11 diakopter .. everything having to do with IO until JimmyZ ripped it out and replaced it..
10:11 diakopter .. the bytecode validator, until not_not_gerd replaced it entirely
10:12 [Coke] I'm sensing a pattern.
10:12 dalek v5: 27b050e | (Tobias Leich)++ | / (4 files):
10:12 dalek v5: fix newly introduced slowdown by adding method .slang
10:12 dalek v5:
10:12 dalek v5: This is the entry-point when switching to our slang. When we switch back to Perl6
10:12 dalek v5: we still jump directly in statementlist sine Perl6 does not do anything special.
10:12 dalek v5: TimToady: Suggestions about a better method name are welcome :o)
10:12 dalek v5: review: https://github.com/rakudo-p5/v5/commit/27b050ef74
10:12 diakopter [Coke]: :D
10:12 tadzik PAY ATTENTION FROGGS
10:12 FROGGS gah
10:12 diakopter hugme: remove FROGGS from perl6
10:13 tadzik no code, no lol
10:13 FROGGS >.<
10:14 diakopter [Coke]: yes, there's a couple other things folks have replaced..
10:14 diakopter but no one will ever replace my unicode character database compiler
10:14 diakopter it's 1000 lines of TI-Basic in Perl 5 syntax
10:15 diakopter moarvm\tools\ucd2c.pl or something
10:15 timotimo m)
10:15 FROGGS it has a \t in its name?
10:15 diakopter *kick*
10:15 timotimo that's pretty terrible
10:15 timotimo ooooh, diakopter long legs
10:15 FROGGS yeah, \terrible
10:16 diakopter .oO( how do you autocomplete something with tabs? )
10:16 timotimo what do you mean by        errible?
10:16 tadzik what we're now doing is what the Squaak tutorial in parrot wanted to be when it grew up
10:17 diakopter true.
10:17 timotimo so .... it grew up and then wanted to be this?
10:18 diakopter [Coke]: we should port your tcl frontend to moarvm tomorrow
10:18 diakopter well.. er, update its nqp to te newness new stuff
10:18 diakopter the new qast
10:19 [Coke] I'm still trying to switch it to jvm.
10:19 woolfy new coffee, hot water for tea, cocacola, orange juice
10:19 [Coke] For now, I can copy p6 and have a ConfigureJVM, ConfigureMoar..
10:21 FROGGS ahh, I already see the upcoming tickets... "nqp doesnt build for JVM and rakudo bakend even when I specified it should configure more"
10:21 FROGGS backend*
10:21 diakopter actually not_not_not_not_gerd & I plan to finish up MoarVM tomorrow
10:21 timotimo %)
10:22 FROGGS ++diakopter, ++not_gerd !!!
10:22 FROGGS ossum!
10:22 timotimo "finish up moarvm" o_O
10:22 timotimo that's pretty cool
10:22 FROGGS I'm sitting in the train like 4 hours... I'll pop in when I can
10:22 timotimo so, NFG and nqp self-hosting and threading stuff?
10:23 [Coke] diakopter++ not_gerd++ !
10:23 diakopter well self-hosting doesn't depend on nfg or thradeing
10:23 timotimo that's true, but "finish up" does, at least to me :P
10:23 diakopter oh heh, yeah, joking.
10:23 FROGGS selfhosting++
10:23 FROGGS that will introduce many LHFs I can work on
10:24 timotimo so, how much are you actually going to finish? making nqp self-host and pass all (relevant) tests?
10:24 FROGGS the relevant ones, yes
10:24 timotimo this month's release will have moarvm for the first time? :)
10:24 FROGGS jnthn said yesterday that t/nqp/24 is not even needed to run rakudo
10:25 diakopter timotimo: c'mon don't push for commitments or expectactions; we're expectation-less around here, remember?
10:25 timotimo also, it seems like we'll make it to a jvm rakudostar :D
10:25 diakopter *expectations
10:25 timotimo okay, i take back my expectations and requests for commitment
10:26 FROGGS timotimo: maybe the nqp compiler release will contain it, but I won't announce that till it is stable and useful enough
10:26 timotimo fair enough.
10:30 lizmat joined #perl6
10:31 [Coke] hurm. jnthn's solution to excercise five doesn't let me echo $variables.
10:31 [Coke] (rakudo-jvm)
10:31 [Coke] er, nqp-jvm, whoops.
10:32 timotimo yeah, NPE, right?
10:33 timotimo don't even get a stack trace :(
10:38 woolfy joined #perl6
10:40 * hoelzro just got say('Hello!') working on Lua o/
10:41 timotimo oh, that was quick
10:41 timotimo when did you start?
10:41 hoelzro last night
10:41 hoelzro it's *extremely* simple at the moment
10:41 hoelzro but it's a start =)
10:41 timotimo i suspected as much :)
10:42 timotimo so you're actually turning qast into lua code (bytecode?)?
10:42 hoelzro mhmm
10:42 timotimo all you can do so far is SVal and Op(:op('call'))? ;)
10:42 hoelzro I'm sticking with Lua source, because the bytecode isn't compatible across versions
10:43 hoelzro pretty much =)
10:43 timotimo that's sensible
10:43 hoelzro and this way, it should work on LuaJIT =)
10:43 timotimo fastest hello world program so far? ;)
10:44 hoelzro hmm
10:44 hoelzro it's probably pretty snappy
10:44 hoelzro but I have no setting
10:44 hoelzro the "setting" is "say = print"
10:44 timotimo :D
10:45 spider-mario joined #perl6
10:51 Bzek joined #perl6
10:51 zakharyas joined #perl6
10:59 woolfy (everybody is working real hard and quiet on exercise 7)
11:01 tadzik hoelzro: did you see the old lua compiler for parrot?
11:01 hoelzro tadzik: I haven't looked at it, no
11:27 dakkar joined #perl6
11:41 EvanTeitelman joined #perl6
11:52 nebuchadnezzar joined #perl6
11:53 nebuchadnezzar joined #perl6
11:57 woolfy joined #perl6
11:58 lizmat joined #perl6
12:16 woolfy nice, nobody is writing about the secret thing
12:17 timotimo too excited to see jnthns secret stash instead
12:17 woolfy shhhht!
12:19 diakopter hoelzro: ping
12:19 FROGGS well, we just need to make sure TimToady doesn't read about that here^^
12:19 hoelzro diakopter: pong
12:20 diakopter you know I wrote an nqp-lua two years ago..
12:20 hoelzro oh, really? =/
12:20 hoelzro I wasn't aware of that
12:20 diakopter did everything but the regex compiler
12:20 diakopter (passed the whole nqp test suite except the regex ones)
12:20 hoelzro nice.
12:20 FROGGS where is it?
12:20 hoelzro but bummer for me =/
12:21 diakopter I kept running into lua limitations
12:21 hoelzro for example?
12:21 diakopter like number of locals, depth of nested blocks
12:21 hoelzro ah
12:21 diakopter number of "upvalues"
12:21 hoelzro I figured that the local/upvalue count would be a limit
12:21 hoelzro I had an idea on how to fix that, but it's hacky
12:22 diakopter yeah me too
12:22 diakopter FROGGS: nt sure
12:22 timotimo there's so many hash tables, yet recently i read that hash tables are super not-cool for language implementation
12:22 diakopter not sure
12:22 diakopter I might've buhleted it
12:23 hoelzro timotimo: that "Why Python is Slow" talk?
12:23 timotimo it was about ruby
12:23 timotimo the pypy people actually have darn clever shit that makes __slots__ completely obsolete
12:23 hoelzro hmm
12:23 diakopter hoelzro: it was an almost entirely direct transliteration/port of the 6model runtime jnthn wrote in C#
12:23 diakopter timotimo: we have that
12:23 timotimo all of their classes are actually super-tightly packed
12:23 diakopter we have that
12:24 timotimo i'm not sure if it does something similar for method lookup
12:24 diakopter we have that for the usual case..
12:24 timotimo p6opaque lays out attributes like a struct, yeah, we know that :)
12:24 tadzik ...the secret thing?
12:24 timotimo but what about methods?
12:25 diakopter hash lookups don't really ever need to be done unless the key being passed in is from an expression or variable.. anything that would do a hash lookup from a constant/known key at compile-time can be optimized away
12:25 timotimo oh
12:25 timotimo well, that's nice :)
12:25 timotimo (but the hash still has to be stored and likely wastes a lot of space)
12:25 timotimo (when compared to a tree of some kind, for instance)
12:25 diakopter not a ton
12:26 diakopter well most hashes are a tree of sorts
12:26 timotimo it really depends on how much empty buckets you instantiate
12:26 timotimo many*
12:26 diakopter some hashes don't use buckets
12:26 timotimo OK
12:26 diakopter (as arrays, anyway)
12:27 timotimo for some reason i associate "hashmap" with "an array plus a hashing function and either linear search or linking"
12:27 diakopter (the one moarvm has, uthash, uses linked lists for both layers/dimensions
12:27 diakopter yes
12:27 tadzik )
12:27 diakopter ()
12:27 timotimo thank you tadzik
12:27 tadzik there was a disturbance in the force
12:27 diakopter ((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((((
12:28 timotimo nooooooooooooooooooo
12:30 timotimo could the type check cache be binary searched through somehow? ordered by some internal ID that each class could get?
12:30 tadzik or a hash :)
12:30 * sjn wonders if #perl6 is turning into a lisp channel :-P
12:31 diakopter timotimo: yep, or a hash
12:31 diakopter .... or prime numbers and prime factorization.
12:31 timotimo what's the "NO LOL" equivalent of lisp?
12:31 timotimo "NO :)"?
12:31 tadzik '\o with a taste of your LISP I'm on a ride '\b
12:31 timotimo perhaps "NO '("
12:31 * sjn was mostly thinking about the ((((((((('s :)
12:31 diakopter prime numbers would be awesome
12:31 timotimo diakopter: ooooh
12:32 diakopter Mu gets 2
12:32 timotimo and then multiply them onto each other
12:32 diakopter Any gets 3
12:32 tadzik wow :)
12:32 timotimo i was thinking there could be just a bitfield, but theoretically there could be more than 64 classes
12:32 timotimo (who would ever need more than 63 classes!?!)
12:32 tadzik you may get so many classes that primes will go into bigints
12:32 diakopter what's the value of the first 64 prime nubmers multiplied
12:33 timotimo that's easy enough to do:
12:33 moritz diakopter: an even number :-)
12:33 * diakopter waits for the linear search
12:33 timotimo r: [*] ((2..Inf).map( *.is-prime )[^64])
12:33 camelia : ( no output )
12:33 timotimo r: say [*] ((2..Inf).map( *.is-prime )[^64])
12:33 camelia rakudo 223075: OUTPUT«0␤»
12:33 diakopter hee hee
12:33 timotimo i think i may have done it wrong
12:34 timotimo r: say [*] ((2..Inf).grep( *.is-prime )[^64])
12:34 timotimo %)
12:34 diakopter stop mkaing me laugh
12:34 camelia rakudo 223075: OUTPUT«195133959955058134502488637025552252876537200920889365778484014616008832503770745901542489344429915739687480227261852231729170␤»
12:34 tadzik yep. Big.
12:34 moritz yes, grep helps.
12:34 timotimo kinda
12:34 timotimo just a little
12:34 diakopter darn.
12:34 moritz r: say log([*] ((2..Inf).grep( *.is-prime )[^64])) / log(2)
12:34 camelia rakudo 223075: OUTPUT«416.205476738791␤»
12:34 moritz less than 512 bits
12:35 timotimo 512 bits may very well be smaller than a little hash of type objects
12:35 moritz r: ([*] ((2..Inf).grep( *.is-prime )).fmt('%b').chars
12:35 camelia rakudo 223075: OUTPUT«[31m===[0mSORRY![31m===[0m Error while compiling /tmp/NJdTP8yqtsâ�¤Unable to parse expression in parenthesized expression; couldn't find final ')'â�¤at /tmp/NJdTP8yqts:1â�¤------> [32mInf).grep( *.is-prime )).fmt('%b').chars[33mâ��[31m<EOL>[0mâ�¤    expecting any…»
12:35 diakopter haha
12:35 moritz r: ([*] (2..Inf).grep( *.is-prime )).fmt('%b').chars
12:35 camelia rakudo 223075: OUTPUT«(timeout)»
12:35 moritz :/
12:35 timotimo yikes
12:36 timotimo you forgot to limit
12:36 moritz aye :-)
12:36 moritz r: say ([*] ((2..Inf).grep( *.is-prime )[^64])).base(2).chars
12:36 camelia rakudo 223075: OUTPUT«417␤»
12:36 timotimo er, what's the benefit of prime numbers and multiplication over bits and adding?
12:36 diakopter none...
12:37 moritz timotimo: I think this was just an academic exercise
12:37 timotimo it happened only in our heads? :(
12:37 diakopter or de-academicizing
12:38 moritz aca-de-mize
12:38 timotimo .o(push the button!)
12:41 timotimo i warned you about stables, bro. i told you Dog.
12:41 dalek nqp: 0c7c701 | (Rob Hoelz)++ | docs/nqp-opcode.txt:
12:41 dalek nqp: Fix NQP op documentation for nqp::exist*
12:41 dalek nqp:
12:41 dalek nqp: It's existskey, not existkey
12:41 dalek nqp: review: https://github.com/perl6/nqp/commit/0c7c701892
12:41 dalek nqp: 54e960d | jimmy++ | docs/nqp-opcode.txt:
12:41 dalek nqp: Merge pull request #124 from hoelzro/master
12:41 dalek nqp:
12:41 dalek nqp: Fix NQP op documentation for nqp::exist*
12:41 dalek nqp: review: https://github.com/perl6/nqp/commit/54e960d5b9
12:42 diakopter hoelzro: you don't have an nqp commit bit?
12:42 hoelzro huh, I forgot I did =/
12:43 timotimo with all this knowledge we can re-invent nqp completely and re-do rakudo in about a month!
12:44 diakopter *wham* *wham* *wham*
12:44 diakopter (sound of rolling)
12:44 timotimo wake me up before you code-code~
12:44 timotimo i don't associate "*wham* *wham*" with rolling at all
12:45 timotimo just like "*pom*" for pondering was completely new to me recently
12:45 denis_boyun joined #perl6
12:46 woolfy timotimo: thanks for volunteering that re-doing!!! :-)
12:48 timotimo i'll do the 100 hours of work per week thing you recommended!
12:51 timotimo is rw for native types would allow my int $fo = 10; $fo++; right?
12:51 diakopter yes
12:54 timotimo that's quite a papercut, imo
12:54 timotimo (not volunteering to implement it right away ;) )
12:55 timotimo i'm beginning to wonder if it was a bad idea to show interest in the optimizer :P
12:56 timotimo i'm kinda sad that optimisations in flow-analysis etc can't really easily be done at the QAST level, so it has to be done at a backend-specific level
12:56 timotimo or at least that's my understanding, i'd happily be proven wrong
12:58 pupoque joined #perl6
13:00 rafl joined #perl6
13:01 [Coke] java.lang.RuntimeException: No REPR HashAttrStore
13:04 pmurias joined #perl6
13:05 hoelzro diakopter: if you find your nqp-lua impl, could you let me know? I'd like to take a look at it
13:05 pmurias timotimo: why can'f flow analysis be done on the QAST level?
13:05 timotimo it can, just not as naturally as with a bytecode-like format
13:05 timotimo at least that's my understanding so far
13:08 pmurias we could turn while/if using QAST into goto/conditional jump using QAST as preprocessing step before the flow analysis
13:08 timotimo qast has that? o_O
13:09 * [Coke] spends this exercize time grabbing a fresh copy of nqp to build a parrot based version.
13:11 pmurias timotimo: has what?
13:13 pmurias hmm, QAST doesn't seem to have goto
13:15 timotimo that's what i though
13:15 timotimo we have a musician in the first row :3
13:16 timotimo SQUERL
13:18 lizmat QAST::Op(call &postcircumfix:<[ ]>) [0]
13:18 lizmat QAST::Op(call &postcircumfix:<{ }>) <a>
13:18 lizmat :-)
13:19 lizmat one step closer to making [] and {} being multi sub rather than multi method calls
13:19 * JimmyZ wonders the reason
13:22 woolfy timotimo: I did recommend working less than 100 hours per week, and if you still do more than 100 hours per week, I did recommend not doing it for years in a row without holiday
13:22 woolfy but I understand that my words have to be twisted!
13:25 timotimo JimmyZ: we can compile-time-dispatch subs, but not method calls
13:25 timotimo (and inline)
13:27 lizmat indeed
13:27 lizmat remarkably few test failures so far
13:27 JimmyZ I saw Optimizer ROADMAP in rakudo/docs
13:27 JimmyZ 2 ***   inlining calls to methods
13:27 JimmyZ 3 ***   guard lifting for method call inlining
13:28 timotimo inlining calls to methods requires some guessing + fallback magic
13:29 timotimo so inline a call, but guard against a wrong guess and doing a full lookup in case of wrongness
13:33 pmurias timotimo: it's possible to turn goto/conditional jump back into loops: https://github.com/kripken/emscripten/tree/master/src/relooper
13:37 pmurias timotimo: IMHO flow analysis is definitely something that needs to be done on the shared across backends level
13:41 timotimo when will the hotel tomorrow throw us out?
13:41 timotimo oh, relooper
13:41 timotimo i heard good things about it
13:42 timotimo or maybe that's something else?
13:42 lizmat timotimo: I asked for late checkout (which is 13:00): please check with the desk if you want to be sure
13:42 timotimo 13:00 is nice
13:42 lizmat (for everybody afaik)
13:43 timotimo will we be thrown out of the common area, too?
13:43 timotimo will there be after-hotel-hackathon-stuff? or will we just disperse right after?
13:43 tadzik I leave FRA at about 13:40
13:43 timotimo aaw :(
13:47 tadzik so I disperse right after waking up, mostly :)
13:48 pmurias timotimo: relooper is used by emscripten
13:50 felher Any idea as to why this doesnt work:
13:50 felher my Array of Str sub s() { my Str @array = < 1 2 3 4 >; }; s;
13:50 felher r: my Array of Str sub s() { my Str @array = < 1 2 3 4 >; }; s;
13:50 camelia rakudo 223075: OUTPUT«Type check failed for return value; expected 'Array[Str]' but got 'Array[Str]'␤  in sub s at /tmp/WoG2qQWeCf:1␤  in block  at /tmp/WoG2qQWeCf:1␤␤»
13:50 woolfy lizmat and I don't have pressing matters back home, so we don't have to leave early.  But we have mberends with us and maybe he needs to be early.
13:51 lizmat timotimo, tadzik : discuss this further over dinner tonight
13:51 [Coke] diakopter & I are heading out to the airport together for 5pmish flights, but haven't decided when we're leaving yet.
13:52 felher Does a little bit like that old windows error messages: "An error occured: you have 8 MB ram, but you need at least 1 MB"
13:52 felher *Does read
13:54 timotimo "Can only use nqp_code_ref_to_sc to add a Sub" wat
13:56 lizmat decont?
13:56 lizmat missing decont?
13:56 timotimo i don't think i have made any change that would cause any error that's anything like that :(
13:59 ajr joined #perl6
14:04 hoelzro does return exit a sub early in NQP? a simple test implies that it doesn't =/
14:04 timotimo make doesn't exit a sub, but i think return ought to
14:04 timotimo can you paste the test here?
14:04 hoelzro sure
14:04 hoelzro incoming
14:04 timotimo nqp: sub a() { say "before"; return 1; say "after" }; say(a);
14:04 camelia nqp: OUTPUT«Unable to parse expression in blockoid; couldn't find final '}' at line 2, near "say \"befor"␤current instr.: 'panic' pc 14693 (src/stage2/gen/NQPHLL.pir:5223) (src/stage2/gen/NQPHLL.nqp:279)␤»
14:04 hoelzro https://gist.github.com/hoelzro/6571032
14:05 timotimo nqp: sub a() { say("before"); return 1; say("after") }; say(a);
14:05 camelia nqp: OUTPUT«␤»
14:05 timotimo nqp: sub a() { say("before"); return 1; say("after") }; say(a());
14:05 camelia nqp: OUTPUT«before␤1␤»
14:05 timotimo nqp: sub a() { say("before"); return; say("after") }; say(a());
14:05 camelia nqp: OUTPUT«before␤after␤1␤»
14:05 timotimo interesting!
14:05 hoelzro ah
14:05 hoelzro very
14:06 hoelzro so 'return;' doesn't work
14:06 hoelzro but 'return $value;' does
14:09 [Coke] that seems.. sub optimal
14:10 hoelzro indeed
14:13 GlitchMr nqp, where return in noop, unless called with a value.
14:13 GlitchMr (but seriously, I think it's a strange bug)
14:13 GlitchMr nqp: sub a() { say("before"); {return;}; say("after") }; say(a());
14:13 camelia nqp: OUTPUT«before␤after␤1␤»
14:14 GlitchMr nqp: sub a() { say("before"); return; say("after"); return 42; say("after after") }; say(a());
14:14 camelia nqp: OUTPUT«before␤after␤42␤»
14:15 hoelzro the QAST for that code is interesting
14:15 woolfy some sort of music from the back
14:15 hoelzro it looks like it doesn't parse as an actual return
14:15 hoelzro oh
14:15 timotimo unless it appears because of the new repr_hint_for, which i'm not sure gets used anywhere at all
14:15 hoelzro it inteprets it as a var, I think
14:16 timotimo hey GlitchMr :)
14:16 hoelzro nqp: sub a() { rob; say('hi'); }; a()
14:16 camelia nqp: OUTPUT«hi␤»
14:16 hoelzro mwahaha
14:17 avar joined #perl6
14:17 avar joined #perl6
14:17 hoelzro ah
14:17 lizmat nqp: rob
14:17 camelia : ( no output )
14:17 hoelzro return is a prefix rule
14:17 timotimo ah so if you don't have a value, it doesn't get parsed
14:17 timotimo interesting!
14:17 timotimo and of course the bare "return" is just NQPMu
14:17 timotimo (almost everything seems to come down to this ...)
14:18 hoelzro I'm going to see if I can fix it
14:18 hoelzro make myself useful =)
14:18 timotimo yays
14:18 timotimo i think a big loud error would be fine
14:18 timotimo or, of course, return NQPMu :P
14:18 hoelzro yeah
14:19 [Coke] nqp: foo; bar; baz; say('frob');
14:19 camelia nqp: OUTPUT«frob␤»
14:20 btyler joined #perl6
14:21 timotimo is there some variant of "lalaland" that we can use to describe CORE.setting making things wonky?
14:21 timotimo coocooland, perhaps
14:22 hoelzro neuland?
14:22 timotimo hahahahaha, you're an ass :D
14:22 hoelzro you assked for it.
14:22 timotimo i did. you deserve a cookie
14:24 [Coke] butt I thought he was an ass?
14:28 hoelzro sweet, fixed the return thing
14:28 hoelzro just going to make sure Rakudo still builds before pushing up
14:30 benabik joined #perl6
14:30 [Coke] hoelzro++
14:33 timotimo i don't know why i'm getting the silly sc thing
14:36 timotimo turns out "invar INT" for an op doesn't make sense; that's probably what got me b0rked?
14:36 timotimo (waiting half an hour for the compile now >_>)
14:36 timotimo (i miss my desktop)
14:39 timotimo jnthn just unveiled his new secret project perl76
14:39 spider-mario joined #perl6
14:40 JimmyZ 76?
14:40 timotimo yeah
14:40 timotimo from the far future
14:40 pmurias what's that?
14:40 timotimo we have no ideas
14:40 timotimo jnthn just let it slip that it exists
14:42 timotimo Perl6 + Perl5 + Moose = Perl76 :)
14:42 FROGGS wow, Perl6 + Moose
14:43 FROGGS can't wait :P
14:43 FROGGS damn, I'm sleepy
14:43 JimmyZ Isn't Is Moose part of Perl6?
14:44 JimmyZ or rakudo + rakudo-v5 + moarp5?
14:45 FROGGS JimmyZ: we are not that serious atm :o)
14:49 timotimo i got an nqp to compile
14:49 timotimo now i've made bootstrap-files and enabled the compile-time hint-precompilation
14:54 hoelzro hmm
14:54 timotimo if i manage to get this build going, can anybody benchmark my new nqp? i suppose compiling the rakudo setting with an "improved" nqp would be enough to see if it makes a difference in performance
14:54 hoelzro I don't think I'll be able to fix this return thing without breaking a bunch of other stuff
14:55 tadzik timotimo: sure
14:55 timotimo (my computer either takes 10 minutes or overheats at a random point in the middle and gives crazy noisy results)
14:55 timotimo ossum :)
14:55 timotimo let me build a commit and make it available
14:55 timotimo actually, have a diff instead
14:56 tadzik I'll time a regular build first
14:56 timotimo nc 192.168.0.79 1234 > nqp_hint_precomp.patch
14:56 timotimo i may have to turn off my firewall first?
14:56 tadzik I think so, I got an empty file
14:57 timotimo try again?
14:58 tadzik nope
14:58 timotimo aaw man :(
14:58 timotimo the connection didn't even go through
14:58 timotimo what does it output?
14:58 [Coke] gist, mebbe?
14:58 tadzik nothing
14:58 timotimo you are on the same network as me?
14:58 tadzik 192.168.0.96 is me
14:58 timotimo yeah, i'll gist it. i'm just sad i have to rely on a centralised service fro this :(
14:58 tadzik put it up again, I'll go for nc -vv
14:58 timotimo actually. hold on.
14:58 timotimo it is still up
14:59 tadzik nc: connect to 192.168.0.79 port 1234 (tcp) failed: No route to host
14:59 timotimo no route o_O
14:59 tadzik what sorcery is this
14:59 hoelzro do they mess with stuff on *this* network, too?
14:59 hoelzro timotimo: upload it to the proxy machine?
14:59 timotimo i scp'd it up to the server on the whiteboard
14:59 hoelzro then tadzik can scp it down
14:59 timotimo :)
14:59 hoelzro perfect =)
14:59 tadzik heh
15:00 timotimo hoelzro: they probably isolate all nodes on the network from each other
15:00 timotimo i guess they heard there will be hackers comming
15:00 hoelzro I think so
15:00 hoelzro bastards
15:00 tadzik timotimo: where did you put it?
15:00 timotimo ~/
15:00 timotimo oh
15:00 timotimo duh
15:00 tadzik there's only 'report' there
15:00 timotimo forgot the : at the end
15:00 tadzik classic
15:01 hoelzro no wonder it was so fast!
15:01 timotimo why doesn't it do anything at all? o_O
15:02 tadzik just put gzipped base64 on irc :)
15:02 timotimo because it's -P in scp and -p in ssh >_>
15:02 tadzik oh I hate this
15:02 timotimo my nqp built to completion <3
15:02 timotimo ef08484d985b10e56ca333acee9c937e3c099f9d <- i based my commit on this commit
15:02 timotimo my non-commit, actually
15:03 tadzik oh well, I'm on NQP_REVISION
15:03 tadzik let's see if it applies
15:03 timotimo it gives new bootstrap files
15:03 timotimo but it shouldn't be a problem
15:05 timotimo i can type ¢ by compose, |, c
15:06 JimmyZ by NFG?
15:06 JimmyZ :P
15:07 timotimo i don't know what you mean :|
15:08 JimmyZ you didn't see MoarVM ppt? o_O
15:08 timotimo i know what NFG is, i do not know why you bring it up right now
15:08 JimmyZ kidding
15:09 timotimo oke :)
15:11 zakharyas joined #perl6
15:11 kaare__ joined #perl6
15:12 krokite joined #perl6
15:12 krokite joined #perl6
15:15 timotimo OMG
15:16 woolfy icecreams 4 types
15:16 woolfy 16 icecreams in total
15:17 tadzik ooooo
15:17 woolfy horn with strawberry ; horn with nuts ; chocolate on a stick ; chocolate (with nuts) on a stick ; (4 of each)
15:21 [Coke] before I came here, I pronounced QAST as KWAST. CAST sounds odd to me.
15:21 btyler joined #perl6
15:21 tadzik timotimo: without: 332.56s user 3.76s system 99% cpu 5:39.14 total
15:21 tadzik with: 323.97s user 3.75s system 99% cpu 5:30.53 total
15:22 tadzik that's nqp without --optimize, fwiw
15:22 timotimo nqp doesn't take optimize yet, sorry
15:22 timotimo so yay, it's a tiny bit faster!
15:22 tadzik oh, parrot does
15:22 tadzik so no difference, yay
15:22 timotimo .... huh?
15:22 tadzik I mean, I didn't do it wrong :)
15:22 hoelzro I thought it was qast with a uvuluar stop
15:23 hoelzro er, uvular
15:23 tadzik it is a bit faster, yes :)
15:23 timotimo just a tiny bit, though :(
15:23 diakopter hoelzro heh lua
15:23 timotimo i was hoping for at least 50% change
15:23 timotimo or maybe 99.999%, but that's just unlikely
15:23 [Coke] Q'ast?
15:23 [Coke] Qapla'
15:24 timotimo :)
15:24 hoelzro heh
15:24 pmurias timotimo: 50% for which optimalization?
15:25 timotimo precalculating the HINT stuff for private attribute access
15:26 timotimo now i don't know what the negate thing is for, why does it say ' in there?
15:26 timotimo on the slide
15:27 timotimo and i don't dare ask because i kind of didn't pay attention for the duration of that slide >_<
15:27 diakopter if someone wants a really mind-bending challange, try to figure out why the quant() thing in NFA.pm needs all the extra edges it has in addition to the minimum you might expect... AND ALSO TO FIGURE OUT HOW IT EVEN WORK THE WAY IT IS
15:27 pmurias timotimo: I wouldn't expect much speed up on parrot for minor things
15:27 timotimo ;_;
15:28 diakopter warning: it's diakopter code
15:28 pmurias ;_; ?
15:28 timotimo tadzik: can you try compiling nqp without actually overwriting the bootstrap files? i don't know if i need to check that in, but my build time is terrible
15:28 timotimo pmurias: crying emoticon
15:28 kivutar joined #perl6
15:29 timotimo won't the icecream get damaged by lying in this not-too-cold room?
15:29 diakopter yeah it all melted
15:29 diakopter so I ate it all
15:29 timotimo nooooooooo :(
15:29 lizmat very likelty
15:29 [Coke] .U+1F63F
15:29 iSlug joined #perl6
15:29 timotimo is that one of the cat faces?
15:29 diakopter .u 1f63f
15:29 yoleaux U+1F63F CRYING CAT FACE [So] (????)
15:29 tadzik timotimo: how do I do that?
15:29 [Coke] aw, not in this font.
15:30 diakopter .u FATFACE
15:30 yoleaux No characters found
15:30 diakopter .u OBAMAFACE
15:30 yoleaux No characters found
15:30 pmurias timotimo: the whole parrot machinery is slowing things down, if we turned a hash lookup into a super fast jitted memory offset lookup the speedup should be noticable, but if we have a super slow do-a-ton-of-parroty-things on top of not so much
15:30 timotimo tadzik: either unapply the patch, open the patch in a text editor, kill all the lines that belong to these files, or just use git to reset these files selectively
15:30 timotimo :(
15:32 tadzik timotimo: two files remained with changes, rebuilding
15:32 diakopter woolfy: can I take an ice cream
15:32 timotimo perfect
15:33 woolfy diakopter & everybody : please take an icecream
15:33 timotimo cue loud noises
15:33 [Coke] someone throw me one!
15:34 krunen mmmmmmmmmmmm....
15:34 timotimo mmmmmmmm....
15:35 lizmat nomnom
15:36 woolfy i take the last one (my second one) -> double nom
15:37 timotimo i'll probably have to build extra code to handle nqp::getattr(...) and nqp::setattr(...)
15:38 [Coke] woolfy++ lizmat++
15:39 pmurias timotimo: a good plan for optimalizations would be to talk with jnthn what info he could make best use of while emitting code in the JVM backend
15:40 tadzik timotimo:
15:40 tadzik it's the same time
15:40 timotimo the jvm backend is already fast-ish; i'd like the parrot backend to be usable as well
15:40 timotimo tadzik: oh, i was mostly interested in: can nqp actually build?
15:40 timotimo but thanks, that's excellent :)
15:40 tadzik oh, sure
15:40 timotimo it's like 4%!
15:40 tadzik yep
15:40 timotimo maybe another 4% if nqp::getattr and friends learn to create HINT, too
15:40 tadzik r: 9/332
15:40 camelia : ( no output )
15:41 tadzik r: say 9/332.Num
15:41 camelia rakudo 223075: OUTPUT«0.0271084337349398␤»
15:41 timotimo (don't calculate it! ;_;)
15:41 tadzik ...sorry
15:41 tadzik but I approximated :)
15:45 pmurias timotimo: you want make siginifcant progress on the parrot backend without changing parrot itself
15:46 GlitchMr rn: use Cat if False
15:46 tadzik woolfy: http://cache.gawkerassets.com/assets/images/17/2010/12/500x_10-yahoo-answers.jpg
15:46 camelia niecza v24-95-ga6d4c5f: OUTPUT«Unhandled exception: System.Exception: Unable to locate module Cat in /home/p6eval/niecza/lib /home/p6eval/niecza /home/p6eval/.config/niecza␤  at /home/p6eval/niecza/boot/lib/CORE.setting line 1502 (die @ 5) ␤  at /home/p6eval/niecza/src/NieczaPathSea…»
15:46 camelia ..rakudo 223075: OUTPUT«===SORRY!===␤Could not find Cat in any of: /home/p6eval/nom-inst/lib/parrot/5.5.0-devel/languages/perl6/site/lib, /home/p6eval/nom-inst/lib/parrot/5.5.0-devel/languages/perl6/vendor/lib, /home/p6eval/nom-inst/lib/parrot/5.5.0-devel/languages/perl6/lib, /home/p6…»
15:46 GlitchMr I cannot have conditional "use"?
15:47 woolfy tadzik: there is no hope for the human race....
15:47 [Coke] rn: if 0 { use Barf }
15:47 camelia rakudo 223075: OUTPUT«===SORRY!===␤Could not find Barf in any of: /home/p6eval/nom-inst/lib/parrot/5.5.0-devel/languages/perl6/site/lib, /home/p6eval/nom-inst/lib/parrot/5.5.0-devel/languages/perl6/vendor/lib, /home/p6eval/nom-inst/lib/parrot/5.5.0-devel/languages/perl6/lib, /home/p…»
15:47 camelia ..niecza v24-95-ga6d4c5f: OUTPUT«Unhandled exception: System.Exception: Unable to locate module Barf in /home/p6eval/niecza/lib /home/p6eval/niecza /home/p6eval/.config/niecza␤  at /home/p6eval/niecza/boot/lib/CORE.setting line 1502 (die @ 5) ␤  at /home/p6eval/niecza/src/NieczaPathSe…»
15:47 tadzik woolfy: there's also a nice flash-y video made, with cavemen having exactly this conversation
15:47 [Coke] similar to p5, no.
15:48 tadzik it's really accurate
15:48 woolfy tadzik: I repeat my statement (...)
15:48 tadzik :)
15:48 hoelzro tadzik: you don't mean http://www.explosm.net/comics/3297/, do you?
15:48 tadzik hoelzro: close ;)
15:49 GlitchMr r: use if $ENV{DEV}, "warnings"
15:49 camelia rakudo 223075: OUTPUT«===SORRY!===␤Could not find sub &DEV␤»
15:50 GlitchMr r: use if %*ENV<DEV>, "warnings"
15:50 camelia rakudo 223075: OUTPUT«===SORRY!===␤Could not find if in any of: /home/p6eval/nom-inst/lib/parrot/5.5.0-devel/languages/perl6/site/lib, /home/p6eval/nom-inst/lib/parrot/5.5.0-devel/languages/perl6/vendor/lib, /home/p6eval/nom-inst/lib/parrot/5.5.0-devel/languages/perl6/lib, /home/p6e…»
15:50 diakopter OBAMAFACE
15:51 iSlug joined #perl6
15:51 tadzik diakopter: http://ragefac.es/not-bad-obama ?
15:51 huf ~/obamaface.../~ sung to the tune of "all by myself"
15:52 moritz
15:52 [Coke] hugme: hug moritz
15:52 * hugme hugs moritz
15:53 * moritz hugs [Coke]
15:56 hoelzro huh
15:56 hoelzro apparently I *don't* have a commit bit for nqp?
15:56 * [Coke] wonders why this install of nqp can only be run from inside the install dir. wtf.
15:57 tadzik [Coke]: yep. Sucks. Or try passing the absolute path to something's Configure.pl
15:57 [Coke] tadzik: already using an absolute path, methinks.
15:58 [Coke] config.status contains: /Users/willcoleda/perl5/perlbrew/perls/perl-5.16.0_THREAD/bin/perl Configure.pl "--prefix=/Users/willcoleda/sandbox/partcl-nqp/install-jvm" $*
15:59 ajr joined #perl6
15:59 tadzik fair enough
16:01 [Coke] arglebargle. I have an install-jvm/nqp and an install-jvm/bin/nqp
16:01 [Coke] one works. :)
16:01 woolfy My first impression of this course: it needs to be done in 3 days.
16:01 hoelzro woolfy: definitely
16:01 moritz not a week? :-)
16:01 tadzik it'd be safer for the brains :)
16:02 woolfy And 17 people is not a bad number.  I feel jnthn gave enough attention to anybody in need / want of attention.
16:03 timotimo looking forward to all the commits :D
16:03 timotimo pmurias: you have a kind of defeatist attitude towards parrot :(
16:03 woolfy moritz: no, not a week, jnthn even had time to do a lot of examples in a terminal and we did a lot of exercises where he went into the room to stand by people doing the examples...
16:04 dalek nqp: 7e2aa75 | moritz++ | Configure.pl:
16:04 dalek nqp: [Configure.pl] turn prefix into an absolute path
16:04 dalek nqp: review: https://github.com/perl6/nqp/commit/7e2aa754ed
16:04 timotimo jnthn seems to want to do more whiteboard use next time around
16:05 woolfy he has 2 flipovers at his disposal!
16:05 timotimo :D
16:06 timotimo "what do you want to do?" "what, besides rip off two flipover charts at the same time?"
16:06 woolfy "three flipovers!"
16:09 [Coke] hurm. no not/bnot ops in nqp?
16:09 timotimo what's that? binary negation?
16:10 timotimo i think i've seen many ?? !! in there
16:12 ajr_ joined #perl6
16:14 moritz [Coke]: there are bitneg_i and bitneg_I nqp:: ops
16:14 timotimo oh, bitwise negation, yes
16:14 xenoterracide_ joined #perl6
16:14 moritz and nqp::not_i
16:14 timotimo and nqp::no_u
16:15 [Coke] moritz: ah, those are not in the docs. danke. I'll add them
16:15 [Coke] what about getting the handles for stdout, err, in?
16:17 [Coke] also, the thing that used to be parrot config - is there an nqp equiv?
16:19 timotimo pmurias: i'm convinced nqp, as well as parrot, does too much too often
16:20 timotimo right now i may be working on parrot-specific code, but soon enough i'll return to the optimizers and hopefully improve codegen across all backends that way
16:20 [Coke] timotimo++
16:21 dalek nqp: db32fa1 | coke++ | docs/nqp-opcode.txt:
16:21 dalek nqp: add refs to more ops
16:21 dalek nqp:
16:21 dalek nqp: moritz++
16:21 dalek nqp: review: https://github.com/perl6/nqp/commit/db32fa1015
16:21 xenoterracide joined #perl6
16:21 krunen Diod alcoholaidd yw cwrw. Grawnfwyd wedi ei eplesu ydyw. Cynhyrchir cwrw trwy ei fragu (e.e. mewn bragdy).
16:22 timotimo the what now? :|
16:22 timotimo er, the .WHAT.HOW?
16:22 krunen http://cy.wikipedia.org/wiki/Cwrw
16:22 btyler joined #perl6
16:23 timotimo is that so :)
16:23 colomon "Drink is alcoholic beer . Cereal is fermented it is. Beer is produced by brew (eg in a brewery )." according to google translate.
16:23 diakopter Chrome crashed when it tried to translate that page
16:24 timotimo mhhh fermented cereal
16:24 kivutar joined #perl6
16:24 timotimo ah, since we have an appointment at the restaurant at 7, this seems to be working out almost perfectly fo rtime
16:25 timotimo (but we did skip the boring parts of the jvm and did not do the 9th exercise, which would be exploring NFAs and such)
16:25 ajr_ joined #perl6
16:25 colomon I am vain, I will drink,
16:25 colomon I g'wilydd see my country,
16:25 colomon I spend like crazy man;
16:25 colomon Fi get drunk, disgusting medicine.
16:25 colomon This google translation is AWESOME.
16:25 timotimo hahaha
16:27 timotimo # TODO mkdir jit
16:28 [Coke] didn't jnthn++ say there was an nqp slurp?
16:29 * FROGGS @attendees>>.slurp( [#]P )
16:29 hoelzro nqp::slurp, right?
16:29 hoelzro or maybe it's just the builtin
16:29 hoelzro the built works, I think
16:29 [Coke] ah, not an op.
16:29 timotimo there are slurpy parameters ;)
16:29 FROGGS [Coke]: there is an nqp::readallfh op
16:30 FROGGS nqp: nqp::readallfh()
16:30 camelia nqp: OUTPUT«Error while compiling block : Error while compiling op readallfh (source text: "nqp::readallfh()"): The 'readallfh' op expects one operand␤current instr.: '' pc 52004 (src/stage2/QAST.pir:18966) (src/stage2/QAST.nqp:3186)␤»
16:30 FROGGS right
16:30 timotimo it'll probably take a filehandle
16:30 timotimo :P
16:30 FROGGS don't you say :o)
16:30 timotimo cpt. obvious signing off o>
16:30 Guest5678 joined #perl6
16:30 Guest5678 hello
16:31 FROGGS o/
16:31 timotimo a guest!
16:31 * timotimo decks the table
16:31 Guest5678 hi
16:31 FROGGS ossum!
16:31 Guest5678 no bolivard
16:31 Guest5678 fg
16:32 FROGGS w/e
16:32 timotimo :\
16:32 Mouq joined #perl6
16:32 timotimo mouq mouq :)
16:33 hoelzro if I have a PMC for the WHAT of an object, how can I print its name?
16:34 hoelzro I've been digging forever yielding nothing =/
16:35 Mouq o/
16:35 timotimo lizmat: "ss" is not even a unicode character! :o
16:35 [Coke] FROGGS: say(nqp::slurp($file)); was fine. Danke.
16:36 FROGGS \o/
16:36 FROGGS (I fiddled with the IO ops on MoarVM quite a bit)
16:37 benabik joined #perl6
16:37 lizmat group photo
16:38 timotimo .o(should have noticed it in the first email tbs)
16:38 timotimo tbh*
16:38 timotimo bbl
16:39 colomon group photo!
16:56 Mouq r: my $f = "a\tb\tc,d\te\tf"; $f ~~ tr/ \, / \n /; $f.say
16:56 camelia rakudo 223075: OUTPUT«a b       cnd     e       f␤»
16:56 Mouq r: my $f = "a\tb\tc,d\te\tf"; $f ~~ tr/ \, / "\n" /; $f.say
16:56 camelia rakudo 223075: OUTPUT«a b       c\d     e       f␤»
16:57 tba joined #perl6
17:00 isBEKaml joined #perl6
17:09 EvanTeitelman joined #perl6
17:09 kivutar joined #perl6
17:41 timotimo hooray, ssh works in this restaurant
17:42 colomon \o/
17:43 moritz timotimo! colomon!
17:44 moritz Mouq!
17:49 colomon moritz!
17:51 Mouq moritz: moritz!
17:52 Mouq I considered saying omitomit earlier, but it's not nearly as fun as "quoM quoM in podicem"
17:53 ajr joined #perl6
17:54 Mouq Oh, nope, got quom and quod mixed up :/
17:55 ajr_ joined #perl6
17:58 dalek rakudo-js: 69a7e62 | (Pawel Murias)++ | run_tests:
17:58 dalek rakudo-js: Add to nqp/t/serialization/01-basic.t run_tests.
17:58 dalek rakudo-js: review: https://github.com/pmurias/rakudo-js/commit/69a7e62e7a
17:58 dalek rakudo-js: 2c7325f | (Pawel Murias)++ | runtime/reprs.js:
17:58 dalek rakudo-js: add a compose method to P6int repr
17:58 dalek rakudo-js: review: https://github.com/pmurias/rakudo-js/commit/2c7325f62e
17:58 dalek rakudo-js: cfcbc97 | (Pawel Murias)++ | runtime/bootstrap.js:
17:58 dalek rakudo-js: set the type check cache in KnowHOW
17:58 dalek rakudo-js: review: https://github.com/pmurias/rakudo-js/commit/cfcbc97416
17:58 dalek rakudo-js: f176a5d | (Pawel Murias)++ | run (5 files):
17:58 dalek rakudo-js: Pass nqp/t/serialization/02-types.t.
17:58 dalek rakudo-js:
17:58 dalek rakudo-js: Implement nqp::scsetcode, nqp::markcodestatic.
17:58 dalek rakudo-js: Do more thing inside P6opaque REPR .compose.
17:58 dalek rakudo-js: Mark the contents of the 6MODEL_CORE sc as coming from that sc.
17:58 dalek rakudo-js: review: https://github.com/pmurias/rakudo-js/commit/f176a5d990
18:01 kivutar joined #perl6
18:03 timotimo pmurias++ \o/
18:04 xenoterracide joined #perl6
18:06 iSlug joined #perl6
18:08 pmurias serialization of closures is still left to do :/
18:10 timotimo seems like you're up to the task, though :)
18:18 ingy mmm serialization :)
18:27 rindolf joined #perl6
18:35 ajr joined #perl6
18:43 pmurias joined #perl6
18:43 cogno joined #perl6
18:44 pmurias if a nqp script is invoked as 'node foo.js 1 2 3' what should @ARGS contain?
18:44 moritz [1, 2, 3]
18:44 moritz well
18:45 moritz ["1", "2", "3"] to be precise
18:45 btyler joined #perl6
18:45 moritz and foo.js should be in $PROGRAM_NAME
18:45 moritz $*PROGRAM_NAME
18:45 moritz r: say $*PROGRAM_NAME
18:45 camelia rakudo 223075: OUTPUT«/tmp/TbiNJmtl2M␤»
18:46 pmurias moritz: this isn't the case in nqp-parrot
18:50 sqirrel joined #perl6
18:51 zakharyas joined #perl6
18:54 timotimo moritz: do you have some experience with pir-level hacking? i'm trying to improve the way the compiler handles ops like bindattr_i and i'm not doing much progress
18:54 mberends_ joined #perl6
18:55 mberends_ left #perl6
18:55 moritz timotimo: I "speak" some PIR, if that's what you want to know
18:55 FROGGS joined #perl6
18:55 FROGGS \o/
18:55 timotimo oh i'm making progress!
18:55 timotimo now the error message is actually understandable
18:56 colomon \o/
18:56 colomon understandable++
19:03 hoelzro wooo! RaNIW party!
19:04 sorear RaNIW?
19:04 sorear oh
19:04 sorear Rak and Nqp Int'l Workshop
19:04 sorear ?
19:04 tadzik yep
19:09 kivutar joined #perl6
19:12 nwc10 is anyone still awake enough for a survivors' party?
19:13 mberends joined #perl6
19:13 timotimo yeah, it's not that late yet
19:13 * mberends is surviving :)
19:14 FROGGS yeah, it is just like quarter past o'beer
19:14 dalek rakudo-js: 5a85a38 | (Pawel Murias)++ | src/QAST/Compiler/JavaScript.nqp:
19:14 dalek rakudo-js: Implement @ARGS and sub MAIN(*@ARGS) {...}
19:14 dalek rakudo-js: review: https://github.com/pmurias/rakudo-js/commit/5a85a3885e
19:14 timotimo that's a quarter to rock o clock ;)
19:15 timotimo FROGGS: i need your help! :)
19:15 timotimo No applicable candidates found to dispatch to for 'Int'.
19:16 timotimo current instr.: 'dispatch' pc 6303 (src/stage2/gen/NQPCORE.setting.pir:2876) (src/stage2/NQPCORE.setting:341) <- so probably a result from my probably b0rked bindattr_i :)
19:19 sorear o/ mberends
19:19 mberends yo sorear
19:20 timotimo hey sorear :)
19:26 mberends_ joined #perl6
19:27 moritz http://www.fudco.com/chip/deconstr.html # 10 years old, but I read it the first time today, and enjoyed it
19:28 kaare__ joined #perl6
19:33 colomon moritz: 20 years old according to HN...
19:38 moritz yes, I can't subtract
19:38 moritz sorry
19:39 moritz and the HN discussion made me read https://en.wikipedia.org/wiki/Sokal_affair
19:42 [Sno] joined #perl6
19:43 colomon I associate Chip Morningstar with Perl, not sure why.
19:43 timotimo gnite
19:46 colomon moritz++ # read this ages ago, but it's a good read and the Internet was otherwise being boring this afternoon
19:48 sorear colomon: probably because you're mixing him up with Chip Salzenburg, who you almost certainly met in austin and is a fairly active p5per
19:48 colomon sorear: could be, I suppose
19:51 arnsholt I'm not entirely sure postmodernism quite deserves the bad rap it's gotten (especially in technical circles)
19:51 arnsholt PZ Myers had a post on just that recently, actually
19:55 krokite joined #perl6
20:05 arnsholt (http://freethoughtblogs.com/pharyngula/2013/09/10/can-we-rehabilitate-post-modernism-please/)
20:13 cognominal joined #perl6
20:16 colomon sorear: actually was thinking of Chip Morningstar, it's just that he was a Commodore 64 guy rather than a Perl guy.   :)
20:22 dalek perl6-roast-data: 7dbc10a | coke++ | p (2 files):
20:22 dalek perl6-roast-data: today (automated commit)
20:22 dalek perl6-roast-data: review: https://github.com/coke/perl6-roast-data/commit/7dbc10a675
20:26 lizmat joined #perl6
20:31 rurban joined #perl6
20:32 rurban diakopter: what did you say at your kiev lightning talk about p5 on p6 performance? when will p5 on p6 bypass p5p? I want to add it to wikipedia.
20:34 lizmat rurban: I think diakopter is getting some shuteye
20:34 masak evening, #perl6
20:34 rurban At the YAPC Europe conference 2013 in Kiev Matthew Wilson announced that within a few years Perl5 on Perl6 will bypass Perl performance...
20:35 masak rurban: I'm semi-sure that lightning talk is now online, if you want to research it yourself.
20:35 dalek rakudo-js: 16a0fe8 | (Pawel Murias)++ | / (4 files):
20:35 dalek rakudo-js: Make nqp-compiler.nqp take the source file as command line argument.
20:35 dalek rakudo-js:
20:35 dalek rakudo-js: Make nqp-compiler.nqp work with real serialization.
20:35 dalek rakudo-js: review: https://github.com/pmurias/rakudo-js/commit/16a0fe8906
20:35 masak should be able to find it on YouTube.
20:35 pmurias rurban: why do you want to add it to wikipedia?
20:36 tadzik rurban: I it was not quite an announcement, afair
20:36 tadzik it was more like "we can make it happen with your help"
20:40 rurban Because I like the idea and it was a pretty important annoucement :)
20:40 rurban Is it too dangerous?
20:41 rurban So far I just added: In late 2012 and 2013 several projects for alternative implementations of Perl 5 started: Perl5 in [[Perl6]] by the Rakudo Perl team, ''moe'' by Stevan Little and friends, ''p2'' by the Perl11 team under Reini Urban and ''gperl'' by goccy.
20:42 rurban predicted is a better term
20:44 ssutch joined #perl6
21:11 [Coke] He'll probably be online in the morning, CET.
21:13 btyler joined #perl6
21:19 kivutar joined #perl6
21:22 rurban I didn't add it
21:27 kivutar joined #perl6
21:36 timotimo you wanna drive home in my bahn? in my bahn?
21:37 timotimo hey masak :)
21:48 timotimo https://gist.github.com/timo/67bd469e5a2317046856 <- this is giving me "No applicable candidates found to dispatch to for 'Int'." when compiling even the simplest thin, like a say("hello world"), but also when compiling nqppregex for stage2
21:48 timotimo i would love to wake up to a solution :)
21:49 timotimo i'll try for a bit o fsleep
21:50 kivutar joined #perl6
21:53 cognominal joined #perl6
22:00 dalek nqp: 289685c | coke++ | docs/nqp-opcode.txt:
22:00 dalek nqp: more ops
22:00 dalek nqp: review: https://github.com/perl6/nqp/commit/289685c143
22:01 Mouq rn: 'aa' ~~ / 'a'+ {note "⇒ a+"} | 'aa' {note "⇒ aa"} /
22:01 camelia rakudo 223075: OUTPUT«⇒ a+␤»
22:01 camelia ..niecza v24-95-ga6d4c5f: OUTPUT«⇒ aa␤»
22:01 Mouq rn: 'aa' ~~ / 'aa' {note "⇒ aa"} | 'a'+ {note "⇒ a+"} /
22:01 camelia rakudo 223075, niecza v24-95-ga6d4c5f: OUTPUT«⇒ aa␤»
22:01 cognominal joined #perl6
22:01 Mouq niecza is correct here, right?
22:03 FROGGS joined #perl6
22:04 fhelmberger joined #perl6
22:08 pmurias [Coke]: have you seen that there is a new docs/ops.markdown file?
22:11 cognominal joined #perl6
22:13 [Coke] pmurias: ah. it was short, I assumed it was the wrong one to add to.
22:13 [Coke] I just wanted to document the existence of these ops I found out about today.
22:16 [Coke] we can still keep adding more detail descriptions to the new file.
22:24 cognominal joined #perl6
22:37 pmurias [Coke]: the .txt is fild
22:38 pmurias [Coke]: the .txt is filled with a lot of old and incorrect stuff, so I think it's best to put everything in the new one
22:44 kivutar joined #perl6
22:54 KiWi__ joined #perl6
23:13 ssutch joined #perl6
23:18 [Coke] ok. I'd rather not have two going. I'll try to spend some time this week combining the two. (IME it's more missing data in the old one)
23:18 [Coke] first, though, getting partcl back.
23:21 [Coke] nqp-jvm has an eval-client but not a server?
23:38 btyler joined #perl6
23:41 xenoterracide joined #perl6
23:44 Mouq joined #perl6
23:50 BenGoldberg joined #perl6

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

Perl 6 | Reference Documentation | Rakudo