»ö« | perl6.org/ | nopaste: paste.lisp.org/new/perl6 | evalbot usage: 'perl6: say 3;' or rakudo: / pugs: / std: | irclog: irc.pugscode.org/ | UTF-8 is our friend!
Set by Juerd on 28 August 2009.
00:06 dukeleto joined 00:08 SmokeMachine left, hercynium left 00:10 nihiliad joined 00:12 rhr_ joined 00:13 synth joined 00:16 rhr left 00:19 ihrd joined 00:23 SmokeMachine joined, freenose left 00:28 arthur-_ left 00:33 ihrd left 00:40 __ash__ joined 00:41 sri_kraih_ joined 00:45 drbean left, nErVe joined 00:46 elton_ left 00:48 sri_kraih left, sri_kraih_ is now known as sri_kraih 00:51 sri_kraih left
colomon rakudo: class Foo { has $.x; multi method bar($baz = 'default') { say $baz; }; }; my $a = Foo.new(x => 2); $a.bar; 00:52
p6eval rakudo 446d49: ( no output )
colomon std: class Foo { has $.x; multi method bar($baz = 'default') { say $baz; }; }; my $a = Foo.new(x => 2); $a.bar; 00:53
p6eval std 28215: ( no output )
colomon std: class Foo { has $.x; multi method bar($baz = 'default') { say $baz; }; }; my $a = Foo.new; $a.bar;
p6eval std 28215: OUTPUT«ok 00:03 40m␤»
colomon rakudo: class Foo { has $.x; multi method bar($baz = 'default') { say $baz; }; }; my $a = Foo.new; $a.bar; 00:54
p6eval rakudo 446d49: ( no output )
00:54 sharada left, rfordinal left 00:55 rfordinal joined, arthur-_ joined 00:56 sri_kraih joined
colomon Anyone out there know if "is export" is actually implemented yet? 00:58
__ash__ should be, the tests aren't fudges for rakudo, so i would assume they work 01:00
t/spec/S11-modules/export.t has the relevant tests
colomon thanks, I'll take a look 01:02
01:04 M_o_C left
colomon __ash__++ 01:05
__ash__ do you know if you can drop down to PIR code in perl6?
colomon Yes, see for example src/setting/Num.pm
Mind you, I don't understand any of that, but Q:PIR is all over that file. 01:06
01:07 SmokeMachine left
Confield Is @_ going to be in Perl 6? 01:07
__ash__ cool, thanks, didn't know about that colomon 01:08
colomon++
Confield yeah, its in perl 6
colomon __ash__: None of the tests in export.t actually test the method XXX is export format which is all over Num.pm. (Which is why I had that file open.)
01:11 mikehh_ joined 01:13 Whiteknight left
Confield __ash__: what are some ways around using @_? As a beginner I hate it. Maybe this way? www.shadowcat.co.uk/archive/confere...9/whyperl/ 01:15
01:17 rfordinal left
__ash__ using @_ isn't really required if you want to use named params or set up your function to take a set number of params 01:18
01:19 hercynium joined 01:20 wayland76 left
Confield ok 01:20
colomon Wow, I've so little used @_ in Perl 6 I forgot that was what it did! :)
01:20 wayland76 joined
lisppaste3 colomon pasted "method is export test" at paste.lisp.org/display/86839 01:21
colomon __ash__: That paste shows how I think is export should work on class methods. But it doesn't, the third test gets "Could not find non-existent sub bar". 01:23
Of course, I may have no clue how is export actually works.
01:25 mikehh left
__ash__ do Bar::bar 01:25
last line should be is(Bar::bar($a)... etc.
indirect method calling doesn't work on class methods currently 01:26
eventually doing bar $a: ; should be valid (which is the indirect way of calling bar)
colomon Ah. 01:28
Well, that let's me set up a test that works, so I guess I should check it in. But it doesn't help with my actual problem at all. 01:29
__ash__ what is the problem? I am not sure i understand whats wrong
pugs_svn r28216 | colomon++ | [t/spec] Add tests of "is export" on class method. 01:30
colomon I've been trying to track down why the $angle.num($base).sin('degrees') tests don't work in trig.t. 01:32
It works fine if I just call it here...
rakudo: say 360.0.sin("degrees");
p6eval rakudo 446d49: OUTPUT«-2.44921270764475e-16␤»
colomon But it fails with "too many arguments passed (2) - 1 param expected" in trig.t. 01:33
I thought it might have something to do with "is export" behaving oddly with default params, but the tests I just checked in work fine.
__ash__ my $angle = 180; $angle.num(10).sin('degrees'); right? throws that error 01:35
colomon rakudo: my $angle = 180.0; $angle.in('degrees'); 01:36
p6eval rakudo 446d49: OUTPUT«Method 'in' not found for invocant of class 'Num'␤»
colomon rakudo: my $angle = 180.0; $angle.sin('degrees');
p6eval rakudo 446d49: ( no output )
__ash__ so it only happens when you chain the methods? 01:37
colomon Assuming that last test was the error, it looks like calling from a variable is part of the issue?
__ash__ rakudo: my $angle = 180.0; say $angle.sin('degrees'); 01:38
p6eval rakudo 446d49: OUTPUT«1.22460635382238e-16␤»
colomon good catch.
rakudo: my $angle = 180.0; say $angle.Num.in('degrees');
p6eval rakudo 446d49: OUTPUT«Method 'Num' not found for invocant of class 'Num'␤»
colomon Dang it, cut n paste issues.
rakudo: my $angle = 180.0; say $angle.Num.sin('degrees'); 01:39
p6eval rakudo 446d49: OUTPUT«Method 'Num' not found for invocant of class 'Num'␤»
__ash__ rakudo: my $angle = 180.0; say $angle.Num::sin('degrees');
p6eval rakudo 446d49: OUTPUT«1.22460635382238e-16␤»
__ash__ shouldn't that be 0?
colomon Round-off error, no big deal.
__ash__ ah, got ya 01:40
If you want to explicitly call a function on an object you do $var.Class::method(); btw
01:41 rhr joined
colomon Right, I was trying to convert $angle to a Num and then call sin on it. 01:41
Thought it was probably a Float instead of a Num.
__ash__ rakudo: say 1.5.WHAT; 01:42
p6eval rakudo 446d49: OUTPUT«Num()␤»
__ash__ i don't think perl6 has a generic float, i think its called Num
colomon There is a Float type that pops up, I think it's from Parrot. Numbers are still all funky in Rakudo.
__ash__ yeah, Num does inherit from parrot's Float 01:43
colomon I'm very tempted to implement Num::Num just so I don't have to see that message any more... 01:45
__ash__: Can you try S32-trig/trig.t if you've got a minute or two? I'd like to see if you get the same problem I'm getting. 01:47
01:48 sri_kraih left, sri_kraih joined
__ash__ alright, i am running the tests in it 01:48
colomon That should work, then delete line 301 #?rakudo skip "method .sin plus base doesn't seem to work?" and try again? 01:49
__ash__ k, it fudged them so i got 0 errors, i'll unfudge that line 01:50
line 302 died on me, too many arguments passed 01:52
colomon Okay, so it's not just me. __ash__++
01:53 rhr_ left 01:54 nErVe left
__ash__ my $a = 5; $a.Num.sin('degrees'); 01:55
rakudo: my $a = 5; $a.Num.sin('degrees');
p6eval rakudo 446d49: ( no output )
__ash__ can you tell what version p6eval is? 01:56
cause that last line i wrote causes the error for me
but apparently not for p6eval
01:57 rhr_ joined
colomon 446d69 is a Mon Sep 7 check in to git. 01:59
__ash__ colomon: try my $a = 5; $a.Num.sin('degrees'); and then try my $a = 5; say Num::sin($a.Num, 'degrees'); 02:00
colomon I'm actually using a more recent rakudo than that. :)
__ash__ me too
02:00 rhr left 02:02 justatheory left 02:03 alester left
__ash__ rakudo: my $a = 5; say $a.Num.sin('degrees'); 02:03
p6eval rakudo 446d49: OUTPUT«too many arguments passed (2) - 1 param expected␤in Main (/tmp/0ttIgQbEQc:2)␤»
__ash__ rakudo: rakudo: my $a = 5; say Num::sin($a.Num, 'degrees');
p6eval rakudo 446d49: OUTPUT«Confused at line 2, near ": my $a = "␤in Main (src/gen_setting.pm:3469)␤»
__ash__ rakudo: my $a = 5; say Num::sin($a.Num, 'degrees'); 02:04
p6eval rakudo 446d49: OUTPUT«0.0871557427476582␤»
__ash__ so, for whatever reason its when it converts the Int to a Num I think
thats causing the problem
colomon rakudo: say 360.Num.sin('degrees'); 02:05
p6eval rakudo 446d49: OUTPUT«too many arguments passed (2) - 1 param expected␤in Main (/tmp/EkJU7yD2vm:2)␤»
__ash__ vs
colomon rakudo: say (360 * 1.0).Num.sin('degrees');
__ash__ rakudo: say 360.0.sin('degrees');
p6eval rakudo 446d49: OUTPUT«Method 'Num' not found for invocant of class 'Float'␤»
rakudo 446d49: OUTPUT«-2.44921270764475e-16␤»
colomon hey, there's the float sneaking in. :) 02:06
rakudo: say (360 * 1.0).sin('degrees');
p6eval rakudo 446d49: OUTPUT«too many arguments passed (2) - 1 param expected␤in Main (/tmp/E9hh0aqmQZ:2)␤»
__ash__ 360.0.sin('degrees'); works, but 360.Num.sin('degrees'); doesn't;
rakudo: say (1.0 * 2).WHAT; 02:07
p6eval rakudo 446d49: OUTPUT«Num()␤»
colomon rakudo: say 360.0.WHAT
p6eval rakudo 446d49: OUTPUT«Num()␤»
__ash__ thats weird how that float came in
colomon So, this is both enlightening and mystifying. :) 02:11
02:14 rhr joined 02:15 buubot left, buubot joined
colomon reckon I should submit that as a bug? 02:15
__ash__ yeah 02:16
it would be worth noting
colomon okay, I'm going to repeat some here just to get a clean sample. :)
__ash__ i am looking at the PAST to see if there is any problems with that, i don't think so, i think it might be related to how it converts an Int to a Num
colomon thanks! 02:17
rakudo: say 5.Num.sin('degrees');
p6eval rakudo 446d49: OUTPUT«too many arguments passed (2) - 1 param expected␤in Main (/tmp/ikAjKOkV3I:2)␤»
colomon rakudo: say 5.0.sin('degrees');
p6eval rakudo 446d49: OUTPUT«0.0871557427476582␤»
colomon rakudo: say (5 * 1.0).sin('degrees');
p6eval rakudo 446d49: OUTPUT«too many arguments passed (2) - 1 param expected␤in Main (/tmp/DZhXZHeJmV:2)␤»
colomon rakudo: say 5.sin('degrees'); 02:19
p6eval rakudo 446d49: OUTPUT«0.0871557427476582␤»
__ash__ it has to be related to that 02:20
colomon The PAST, you mean?
(whatever that is?)
__ash__ i know doing 5.Num.WHAT says NUM but looking at the trace=1 its calling the sin in 2 different ways
colomon interesting. 02:21
__ash__ do perl6 --trace=1 some_file.pl where some_file.pl only contains the failing test (lots of junk is spit out
)
colomon Bug is reported, RT #69076.
__ash__ --trace=1 basically prints the opt codes before they are executed 02:22
i see, when you do 5.Num its passing a float around inside and calling the float's version of sin which doesn't have a 2nd parameter for degrees 02:23
(I think)
yup, if you look at the last line before the exception is thrown in the trace output its calling sin on P8 (for me) which is a Float not a Num, so its doing the wrong lookup 02:25
colomon __ash__++ 02:26
__ash__ and in parrot/src/pmc/float.pmc line 536, float's sin doesn't take any parameters but itself
02:27 rhr_ left 02:28 alester joined
colomon Any clue how to fix it? 02:29
__ash__ yeah, lines 5~7 need reworking in rakudo/src/setting/Int.pm 02:30
colomon Int::Num is wrong? 02:31
__ash__ yeah, its not being cast to the right type 02:32
doing $N0 = self converts the self object into a PASM number which is a Float, not a Num, i know calling 5.Num.WHAT returns Num for whatever reason, but its looking in the Float namespace for the sin function instead of the Num namespace 02:35
here is another example of what I am referring to 02:39
rakudo: say 5.5.can('sin');
p6eval rakudo 446d49: OUTPUT«sin␤»
__ash__ rakudo: say 5.Num.can('sin');
p6eval rakudo 446d49: ( no output )
__ash__ that should of said 'sin'
rakudo: say 5.0.can('sin'); 02:40
p6eval rakudo 446d49: OUTPUT«sin␤»
02:41 cdarroch left 02:45 stephenl1 is now known as stephenlb
colomon Any idea how we should take this from here? 02:46
02:48 payload left
__ash__ not sure exactly, i am looking around to see if anything sparks any ideas 02:49
ooo, look at this
rak
rakudo: say 5.5.sin('degrees'); 02:50
p6eval rakudo 446d49: OUTPUT«0.095845752520224␤»
__ash__ oops, i mean:
rakudo: say 5.5.*sin('degrees');
p6eval rakudo 446d49: OUTPUT«0.095845752520224␤»
__ash__ rakudo: say 5.Num.**sin('degrees');
p6eval rakudo 446d49: OUTPUT«Confused at line 2, near ".**sin('de"␤in Main (src/gen_setting.pm:3469)␤»
__ash__ rakudo: say 5.Num.*sin('degrees');
i can't type right now
p6eval rakudo 446d49: OUTPUT«get_iter() not implemented in class 'NCI'␤in Main (/tmp/6VOTeeo6yg:2)␤»
02:54 huf_ joined 02:56 sevvie_ joined 02:59 sevvie left
colomon I need to be getting to bed. But you've made awesome progress here, thanks! 03:00
__ash__ no problem, i'll update the bug report with what i found to see if its useful 03:02
colomon I'm predicting this will be fixed by someone tomorrow. :)
03:06 sm` left, sm` joined 03:17 sharada joined 03:23 ab5tract left 03:29 hercynium left 03:31 zpmorgan left 03:35 JimmyZ joined 03:41 mepplock joined
__ash__ rakudo: say 1 / 0; 03:46
p6eval rakudo 446d49: ( no output )
__ash__ say 1.0 / 0;
rakudo: say 1.0 / 0 ;
p6eval rakudo 446d49: ( no output )
spinclad [backlogging from 'second system' discussion...] 'elephantine' is an old word for ivory... so imagine Perl 6 a procession of feature-laden elephants, carved in a single tusk. 03:55
(q: but is this properly 'monstrous'? a: it is indeed a great wonder to behold!) 03:56
JimmyZ rakudo: (1/0).say;
p6eval rakudo 446d49: ( no output )
03:56 meppl left
spinclad rakudo: Inf.say 03:56
p6eval rakudo 446d49: OUTPUT«Inf␤»
04:00 nihiliad left 04:09 synth left 04:12 synth joined
__ash__ how does one add stuff to a bug report in perl bug ? 04:18
04:23 drbean joined
dukeleto __ash__: what do you mean? 04:24
__ash__ i figured it out, nevermind
04:26 asciiville joined
asciiville I wish #perl6 was on cable tv... 04:29
04:29 sharada left
pmichaud it isn't? 04:29
asciiville :) 04:30
04:30 tak11 left
asciiville I am thinking about creating a long-winded post for the parrot-dev mail list to consider SQLite3 to be the default, embedded DB for Parrot, kind of like it is for PHP, and later Pythons 04:36
__ash__ rt.perl.org/rt3/Ticket/Display.html?id=69076 anyone have any comments on this? 04:38
how do you convert from one type to another explicitly? 04:39
asciiville I thought casting was an evil side effect of Java. :) 04:41
er, uh, until generics of course 04:42
__ash__ but what if you want something to be a specific type?
04:43 stephenlb left
asciiville autoboxing wouldn't help there I assume? 04:43
__ash__ rakudo: say "5.5".sin('degrees'); 04:44
p6eval rakudo 446d49: OUTPUT«Method 'Num' not found for invocant of class 'Str'␤»
__ash__ rakudo: say 5.5.sin('degrees');
asciiville oh well :)
p6eval rakudo 446d49: OUTPUT«0.095845752520224␤»
__ash__ rakudo: say 5.Num.sin('degrees');
p6eval rakudo 446d49: OUTPUT«too many arguments passed (2) - 1 param expected␤in Main (/tmp/Y7vDSDQGfj:2)␤»
__ash__ stuffs not being cast to the right data types in those conversions 04:45
thats why i wana know how to explicitly cast :P
asciiville rakudo: say (5.5).sin('degrees')
p6eval rakudo 446d49: OUTPUT«0.095845752520224␤»
asciiville rakudo: say (5.5).sin('radians') 04:46
p6eval rakudo 446d49: OUTPUT«-0.705540325570392␤»
asciiville rakudo: say "5.5".sin('radians')
p6eval rakudo 446d49: OUTPUT«Method 'Num' not found for invocant of class 'Str'␤»
pmichaud rakudo: say "5.5".Num
p6eval rakudo 446d49: OUTPUT«Method 'Num' not found for invocant of class 'Str'␤»
pmichaud seems like we ought to fix that. 04:47
asciiville Java do it with a "Something.valueOf"
__ash__ Int::Num isn't working like you'd exepct either i think
pmichaud rakudo: say 5.Num
p6eval rakudo 446d49: OUTPUT«5␤»
__ash__ rakudo: say 1.Num.*sin('degrees'); 04:48
pmichaud rakudo: say 5.Num.WHAT
p6eval rakudo 446d49: OUTPUT«get_iter() not implemented in class 'NCI'␤in Main (/tmp/I8Q7je26EJ:2)␤»
rakudo 446d49: OUTPUT«Num()␤»
asciiville NCI?
__ash__ rakudo: say 1.0.*sin('degrees');
p6eval rakudo 446d49: OUTPUT«0.0174524064372835␤»
pmichaud oh, the fact that .* isn't working has to do with the fact that the sin() method appears to have been written in PIR
checking
__ash__ only 1 version of sin
pmichaud that's enough to cause a problem.
__ash__ there is a sin in Num and a sin in Parrot's Float
s1n and a s1n in #perl6 :) 04:49
asciiville :)
__ash__ I don't get why 1.Num.sin('degrees') calls the Float sine not the Num sine too
asciiville "5.5".s1n('radians')
carp
pmichaud rakudo: say 1.Num.PARROT
p6eval rakudo 446d49: OUTPUT«Float␤»
asciiville rakudo: "5.5".s1n('radians')
p6eval rakudo 446d49: OUTPUT«Method 's1n' not found for invocant of class 'Str'␤»
pmichaud That's why.
1.Num is returning a Parrot Float. 04:50
s1n OUTPUT«-0.705540325570392␤»
pmichaud and calling .sin on a Parrot Float gets us the Parrot .sin method
s1n :)
__ash__ ah, well, i didn't know about that..., any way to coerce it a Num?
pmichaud sure
in Int.pm
asciiville rakudo: (Num).WHAT
p6eval rakudo 446d49: ( no output )
pmichaud change line 7 from
%r = box $N0 04:51
to
%r = new ['Num']
%r = $N0
asciiville that's from inline PIR?
pmichaud yes
the box opcode is returning the Float
and since Float has a .sin already, it gets used in preference to the one we've written in Num 04:52
(and that one apparently doesn't dispatch properly using .*)
ideally Parrot's type mapping would get box to automatically return Rakudo Num objects instead of Parrot Float... but we haven't wanted to make that switch yet 04:54
__ash__ pmichaud: do you want to commit that change? that fixed some fudged tests in S32-trig/trig.t
pmichaud I'm in the middle of another change; I can commit it tomorrow morning if nobody beats me to it
__ash__ alright, just checking
asciiville Can't wait till rakudo gets a pythonish introspection: dir(Num) 04:57
__ash__ rakudo: say Num.HOW.methods(Num); 05:01
p6eval rakudo 446d49:
..OUTPUT«StraseccotanhacotanhsechACCEPTSatanasechacostanhasinatanhcoseccoshsuccacoshperlWHICHcotanatan2Scalarsectancossinpredacosecsinhasinhcosechacotanacosechendchompreduceindexceilingp5chomp:dpairs:e:fcombunpolarordchopintcharsrootsucreversekeysisaucfirstsubstrcosfmtbytespickevalfilejoint…
__ash__ rakudo: say ~Num.HOW.methods(Num);
p6eval rakudo 446d49: OUTPUT«asinh cosech acotan acosech Str asec cotanh acotanh sech ACCEPTS atan asech acos tanh asin atanh cosec cosh acosh succ WHICH perl cotan atan2 Scalar sec tan cos sin pred acosec sinh 1 1 end 1 reduce 1 ceiling p5chomp 1 pairs 1 1 comb unpolar ord chop 1 1 roots uc reverse keys
..isa u…
05:02 __ash__ left
asciiville dang! 05:03
rakudo: say Num.HOW.methods(Num); 05:04
p6eval rakudo 446d49:
..OUTPUT«sinhWHICHcotancosechatan2acotansecaseccotanhtansechatancosacossintanhasincoshsuccasinhperlScalaracosechStracotanhACCEPTSasechpredatanhcosecacosecacoshcanlcfirstrootsreduceComplextrimmap:dcis:e:flogminrandmaxtruncatebytessortsinsqrtIntgrepabsenducceilingp5chompunpolarsamecasefloorsu…
asciiville that is awesome!
doesn't roll trippingly of the tongue however. :) 05:06
05:09 Bzek joined
asciiville rakudo: say Num.HOW.methods(Int) 05:15
p6eval rakudo 446d49:
..OUTPUT«IntpredsuccNumACCEPTSperlWHICHScalarabsRatStrreversekeysisaucfirstsubstrcosfmtbytespickevalfilejointrimchrfloorrandtruncateroundsortrindexsqrtsplitmatchgrepwordsvaluescanlcfirstComplexsrandpolarmapciskvsamecaselogmincapitalizemaxtransfirstsinflipIntdoesp5choplcsubstabselemschompend…
asciiville k, i'm confused again ;)
rakudo: say join("\n", Int.HOW.methods(Int)) 05:19
p6eval rakudo 446d49:
..OUTPUT«WHICH␤perl␤ACCEPTS␤Scalar␤2␤Rat␤Str␤Int␤pred␤succ␤Num␤can␤lcfirst␤uc␤srand␤1␤keys␤map␤1␤cis␤1␤1␤min␤max␤pick␤first␤1␤trans␤sin␤1␤join␤2␤1␤1␤end␤1␤1␤ceiling␤trim␤p5chomp␤pairs␤unpolar␤ord␤rand␤1␤1␤1␤1␤1␤grep␤values␤1␤roots␤reverse␤isa␤ucfirst␤kv␤samecase␤cos␤capitalize␤fmt␤bytes␤fli…
asciiville a little clearer with the satellite characters 05:20
rakudo: "xyz".reverse().say; 05:21
p6eval rakudo 446d49: OUTPUT«xyz␤»
asciiville rakudo: "xyz".uc 05:22
p6eval rakudo 446d49: ( no output )
asciiville rakudo: "xyz".uc.say
p6eval rakudo 446d49: OUTPUT«XYZ␤»
asciiville rakudo: " hello ".trim.say 05:23
p6eval rakudo 446d49: OUTPUT«hello␤»
asciiville (((" hello ").trim()).say()) 05:25
rakudo: (((" hello ").trim()).say())
p6eval rakudo 446d49: OUTPUT«hello␤»
asciiville nice
05:27 nickgibbon joined
asciiville rakudo: say( (((" hello ").trim()).say()) == " hello ".trim.say) 05:27
p6eval rakudo 446d49: OUTPUT«hello␤hello␤1␤»
05:38 agentzh joined 05:41 synth left 05:49 synth joined 05:58 Confield left
JimmyZ rakudo: Num.HOW.methods(Str).join(' ').say; 06:04
p6eval rakudo 446d49: OUTPUT«pred encode succ ACCEPTS perl WHICH sprintf Scalar flip 1 does p5chop lc 2 1 1 end 1 reduce 1 ceiling p5chomp 1 pairs 1 1 comb unpolar ord chop 1 1 roots uc reverse keys isa ucfirst 1 cos fmt bytes pick 1 join trim chr floor rand 1 1 round 1 1 split 1 grep words values can
..lcfirst …
JimmyZ rakudo: ~Num.HOW.methods(Buf)..say; 06:05
p6eval rakudo 446d49: OUTPUT«say requires an argument at line 2, near ";"␤in Main (src/gen_setting.pm:2565)␤»
JimmyZ rakudo: ~Num.HOW.methods(Buf).say;
p6eval rakudo 446d49: OUTPUT«newpostcircumfix:[
..]decodelistofcharsrootsucreversekeysisaucfirstsubstrcosfmtbytespickevalfilejointrimchrfloorrandtruncateroundsortrindexsqrtsplitmatchgrepwordsvaluescanlcfirstComplexsrandpolarmapciskvsamecaselogmincapitalizemaxtransfirstsinflipIntdoesp5choplcsubstabselemschompendi…
06:40 orafu left 06:41 orafu joined
KatrinaTheLamia oh BTW--decided to start work on Jerl 6 in Octobre. It will be pronounce like "Girl 6", BTW ~.^ 06:44
araujo KatrinaTheLamia, ?
KatrinaTheLamia just need to freshen up on compiler design and working with the JVM--also I have another project that I need to work throughin Septembre
araujo, oh--a little while back, I suggested a version of Perl 6 for the JVM ^.^ 06:45
I mean Rakudo is good and all--it really is. However the JVM has Enterprise support, and has existed for a while. Plus the sheer blashphlemy of Jerl 6 is hilarious in and of itself. 06:46
Matt-W It's a good idea 06:52
Gives Perl 6 a potential foothold in Java houses, they can start writing bits of their systems in it and still use their existing components
06:55 alester left 07:00 iblechbot joined 07:15 xinming left
KatrinaTheLamia Matt-W, part of why I am decently looking into it with Jerl 6... I dunno, maybe if Jerl 6 goes decently, I may make a Jerl 5 07:15
I think we may need to giver Camela a barrista make over once Octobre comes ~.^ 07:16
07:16 zloyrusskiy left 07:17 asciiville left 07:22 rfordinal joined 07:23 rfordinal left 07:28 nickgibbon left 07:33 rgrau joined
araujo well 07:37
that'd be an interesting thing to see
i guess
though, KatrinaTheLamia , why not to continue helping with rakudo? 07:38
KatrinaTheLamia araujo, oh I prolly will. There will be more than a little crosspolination between the two projects. I also have another project known as SSBZ, that I has decided will _also_ be done in Perl 6, and that will require some way for Perl 6 to handle the graphics libraries available to Parrot and the JVM without much differences in APIs 07:40
I mean, writing Jerl 6 will likely give a lot of helpful information on decent ways to implement and improve Rakudo--and vica versa. 07:41
I am not writing Jerl 6 to complete with Rakudo--more compliment it, is all. 07:42
moritz_ diversity is a good thing
but you also must understand that writing a Perl 6 compiler is no small task
no matter in what environment
KatrinaTheLamia moritz_, yush, I am aware. I've read a few things, and have messed around with figuring out how to implement a Postscript interpreter. I've also looked around the Spidermonkey code, and have read some stuff on embedded Python. I am mostly just going over the Dragon Book to compliment this other knowledge, which I prolly should have committed to memory _already_--don't know why it isn't in my mind. 07:45
moritz_, if compilers really were easy to design we'd have 1006 of them being made all the time ^.^ 07:46
arnsholt KatrinaTheLamia: Iæd say you donæt have it committed to memory because it's quite complicated stuff
Er, s/æ/'/ 07:47
KatrinaTheLamia arnsholt, it is yush--there is lots to consider when making a compiler too. Though, I generally tend to just _get_ these sorts of things... generally the only time I have issues figuring stuff out is when I think I am going to have issues figuring stuff out. 07:49
araujo o_0
07:51 cotto left 07:53 agentzh left 07:54 sm` left, sm` joined 07:59 payload joined
colomon __ash__, pmichaud: %r = new ['Num'] %r = $N0 change doesn't seem to work for me, I still get the "too many arguments passed (2) - 1 param expected" after converting an Int to a Num and calling sin. 07:59
moritz_ colomon: just write %r = box $N0 08:00
colomon moritz_: That's what the old code that definitely doesn't work did. The two-step thing was pmichaud's proposed solution. 08:02
moritz_ colomon: what did it not work? I think it's used in quite some setting methods, with success
colomon If you go through the log, there's a long discussion in two parts on this. 08:03
moritz_ ok, I'll backlog
colomon basically, it's Int's Num function, and the %r = box $N0 is making some sort of weird float object that pretends to be a Num rather than a real Num. 08:04
so that
rakudo: say 360.0.sin('degrees');
p6eval rakudo 446d49: OUTPUT«-2.44921270764475e-16␤»
colomon works but
08:04 pmurias joined
colomon rakudo: say 360.Num.sin('degrees'); 08:04
p6eval rakudo 446d49: OUTPUT«too many arguments passed (2) - 1 param expected␤in Main (/tmp/CXJkarTx2u:2)␤»
colomon The latter is calling a different sin function. 08:05
pmurias KatrinaTheLamia: there are loads of compilers being written all the time... ;) 08:06
moritz_ ok, that's beyond my scoope
*scope
pmurias KatrinaTheLamia: the hard part with targeting jvm is that you have to find a way to map (most of) semantics correctly and vaguely fast 08:10
colomon moritz_: anyway, this is the mysterious error in sin I was worried about over the weekend. __ash__++ tracked it down to lines 5-7 of Int.pm, and that was pmichaud's proposed solution, but as I said, it doesn't seem to work for me. 08:12
moritz_ colomon: have you tried something really simple like adding 0.0 or multiplying with 1.0 or so? 08:13
colomon rakudo: say (360 * 1.0).sin('degrees'); 08:14
p6eval rakudo 446d49: OUTPUT«too many arguments passed (2) - 1 param expected␤in Main (/tmp/xguaXaW4aE:2)␤»
colomon rakudo: say (360 + 0.0).sin('degrees');
p6eval rakudo 446d49: OUTPUT«too many arguments passed (2) - 1 param expected␤in Main (/tmp/70qHTY6QkC:2)␤»
colomon hadn't actually tried the + 0.0 before. :) 08:15
08:32 synth left 09:00 masak joined
masak o/ 09:00
moritz_ \o
pmurias \ö/ 09:01
moritz_ >>ö<<
masak "The Metamorphosis" 09:02
Matt-W oh hai masak 09:07
09:07 rbaumer left
masak saluton, Matt-W 09:07
Matt-W just got back from the vet
don't know how long until the cat forgives me 09:08
pmurias untill it gets hungry enough ;)
Matt-W That's usually the case, yes 09:09
He was fairly well-behaved really
and his sulking gives me a chance to do something vaguely useful with the rest of my morning off 09:10
JimmyZ oh hi #perl6 09:11
masak oh hai JimmyZ 09:13
so, I almost got Squirrel working yesterday. 09:15
enough for it to be pretty exciting.
I think I'll do some debugging today to see why it didn't get all the way.
09:19 mepplock left 09:22 masak left, masak joined 09:34 rbaumer joined
masak wants to see the slides of conferences.yapcasia.org/ya2009/talk/2238 09:38
09:39 alexe joined, ihrd joined
masak people are buzzing about it on Twitter... :) 09:39
09:39 ihrd left 09:40 meppl joined 09:43 alexe_ joined 09:44 sharada joined, sevvie_ left, Grrrr left, kst` joined, Grrrr joined 09:47 sevvie joined, Aisling_ joined, Eevee joined
Matt-W masak: I want to see those slides too 09:48
it looks like great propaganda material
masak even the title is great. 09:49
09:50 cblaptop joined 09:52 payload left 09:54 Aisling left 09:55 alexe left 09:59 kst left 10:01 Eevee_ left 10:03 ssm left 10:04 Whiteknight joined 10:13 Whiteknight left 10:20 M_o_C joined
moritz_ it does sound great for a mostly perl 5 audience 10:30
10:30 Gothmog_ left
moritz_ s/<after great>/, especially / 10:30
10:30 Gothmog_ joined
masak that's what you'd have at any YAPC, no? 10:31
moritz_ aye 10:32
10:46 JimmyZ left 10:47 alexe_ left
masak is re-reading the Apocalypses 10:51
this part about killing off the $pkg'var syntax made me giggle: "[...] we won't make the mistake of reintroducing a syntax that drives highlighting editors nuts. We'll try to make different mistakes this time."
moritz_ it's time for better highlighters 10:52
see: perl6.vim
masak 对 10:53
commuting & 10:54
10:54 masak left 10:57 xinming joined 11:06 payload joined 11:08 finanalyst joined 11:10 SkyRocknRoll joined
SkyRocknRoll hi will perl6 support cpan? 11:10
moritz_ the question is more, will cpan support Perl 6? 11:11
I'm working (a bit) on it
SkyRocknRoll moritz_: can i use all cpan library in perl 6?
moritz_ others work on alternative solutions
SkyRocknRoll: hopefully, yes 11:12
rakudo: say eval('3+4', :lang<perl5>)
p6eval rakudo 446d49: OUTPUT«␤»
moritz_ uh, hm 11:14
if blizkost is installed properly that works
11:21 ruoso left
M_o_C Is Blitzkost a slavic word? Or more generell what does it mean if it has a meaning? 11:22
11:22 finanalyst left
moritz_ that's explained in the README, iirc 11:22
M_o_C Where exacty is Blitzkost hostet? 11:25
Google yields nothing useful, neither does github.com/search?q=blitzkost&x=0&y=0
moritz_ see use.perl.org/~JonathanWorthington/journal/ for the announceemnt
M_o_C: it's Blizkost, not Blitzkost 11:26
M_o_C Oh, rue
*true
>_>
11:27 cblaptop left 11:31 cblaptop joined, sm` left, sm` joined 11:40 SkyRocknRoll left 11:41 Rint joined 11:50 veeru joined, MoC` joined 11:52 veeru left 11:54 ab5tract joined 12:01 rgrau left 12:06 masak joined 12:08 M_o_C left 12:15 JimmyZ joined
takadonet morning all 12:18
masak morning, takadonet. 12:21
hey, which synopsis talks about the semantics of term:<*>? 12:25
12:26 Bzek left
moritz_ S02 iirc 12:26
look for Whatever :-)
masak yep, found it. 12:27
rakudo: (* + 24).(24).say 12:29
p6eval rakudo 446d49: OUTPUT«48␤»
masak rakudo: ([*] 1..4).flip.say 12:30
p6eval rakudo 446d49: OUTPUT«42␤»
masak \o/
12:34 meppl left, finanalyst joined 12:36 jaffa8 joined 12:37 synth joined
jaffa8 perl6: print "uuu"􏿽xE9 12:38
perl6: print "uuu";
p6eval pugs: OUTPUT«*** ␤ Unexpected "\65533"␤ expecting term postfix, operator, ":" or ","␤ at /tmp/KBQ3LZhXMn line 1, column 12␤»
..elf 28216: OUTPUT«Parse error in: /tmp/srnQMDYlte␤panic at line 1 column 0 (pos 0): Can't understand next input--giving up␤WHERE: print "uuu"�␤WHERE:/\<-- HERE␤ STD_red/prelude.rb:99:in `panic'␤ STD_red/std.rb:76:in `scan_unitstopper'␤ STD_red/std.rb:224:in `comp_unit'␤ STD_red/std.rb:210:in
..`_U…
..rakudo 446d49: OUTPUT«Confused at line 2, near "\ufffd"␤in Main (src/gen_setting.pm:3469)␤»
elf 28216, pugs, rakudo 446d49: OUTPUT«uuu»
jaffa8 pugs: /../ 12:39
p6eval pugs: OUTPUT«Error eval perl5: "if (!$INC{'Pugs/Runtime/Match/HsBridge.pm'}) {␤ unshift @INC, '/home/evalenv/pugs/perl5/Pugs-Compiler-Rule/lib';␤ unshift @INC, '/home/evalenv/pugs/third-party/Parse-Yapp/lib';␤ eval q[require 'Pugs/Runtime/Match/HsBridge.pm'] or die $@;␤}␤'Pugs::Runtime…
12:40 asciiville joined 12:42 asciiville left
jaffa8 pugs: $e=~/../; 12:44
p6eval pugs: OUTPUT«*** ␤ Unexpected "=~/../;"␤ expecting "::"␤ Variable "$e" requires predeclaration or explicit package name␤ at /tmp/KSauahGsMK line 1, column 3␤»
jaffa8 pugs: $e~~/../;
p6eval pugs: OUTPUT«*** ␤ Unexpected "~~/../;"␤ expecting "::"␤ Variable "$e" requires predeclaration or explicit package name␤ at /tmp/dIYLIaiB5u line 1, column 3␤»
jaffa8 pugs: my $e~~/../; 12:45
p6eval pugs: OUTPUT«Error eval perl5: "if (!$INC{'Pugs/Runtime/Match/HsBridge.pm'}) {␤ unshift @INC, '/home/evalenv/pugs/perl5/Pugs-Compiler-Rule/lib';␤ unshift @INC, '/home/evalenv/pugs/third-party/Parse-Yapp/lib';␤ eval q[require 'Pugs/Runtime/Match/HsBridge.pm'] or die $@;␤}␤'Pugs::Runtime…
jaffa8 pugs: my $e; $e~~/../; 12:47
p6eval pugs: OUTPUT«Error eval perl5: "if (!$INC{'Pugs/Runtime/Match/HsBridge.pm'}) {␤ unshift @INC, '/home/evalenv/pugs/perl5/Pugs-Compiler-Rule/lib';␤ unshift @INC, '/home/evalenv/pugs/third-party/Parse-Yapp/lib';␤ eval q[require 'Pugs/Runtime/Match/HsBridge.pm'] or die $@;␤}␤'Pugs::Runtime…
masak jaffa8: Pugs regex matching seems to have bitrotted. 12:48
moritz_ aye
pugs compiles regexes to perl 5 regexes 12:49
and then uses the perl 5 bridge to run them
maybe that broke with perl-5.10
masak sounds likely.
jaffa8 I see 12:51
DO you know how to translate \N into Perl 6? 12:52
masak there's a \N in Perl 6 as well. 12:53
moritz_ but it means something different
\N{LATIN SMALL LETTER A} translates to \c[LATIN SMALL LETTER A], I think
rakudo: say " \c[LATIN SMALL LETTER A]" 12:54
p6eval rakudo 446d49: OUTPUT« a␤»
moritz_ but in perl 5 it depends on 'use charnames qw(:full)' to be loaded
12:54 payload left
jaffa8 rakudo: say "\c[LETTER A]" 12:55
p6eval rakudo 446d49: OUTPUT«perl6regex parse error: Unrecognized character name LETTER A at offset 18, found 'L'␤in Main (src/gen_setting.pm:3469)␤»
jaffa8 rakudo: say "\c[LATIN LETTER A]"
p6eval rakudo 446d49: OUTPUT«perl6regex parse error: Unrecognized character name LATIN LETTER A at offset 18, found 'L'␤in Main (src/gen_setting.pm:3469)␤»
jaffa8 rakudo: say "\c[LATIN BIG LETTER A]" 12:56
p6eval rakudo 446d49: OUTPUT«perl6regex parse error: Unrecognized character name LATIN BIG LETTER A at offset 18, found 'L'␤in Main (src/gen_setting.pm:3469)␤»
moritz_ it's s/BIG/CAPITAL/ 12:57
jaffa8 rakudo: say "\c[LATIN CAPITAL LETTER A]"
12:57 ruoso joined
p6eval rakudo 446d49: OUTPUT«A␤» 12:57
jaffa8 rakudo: say ">>\c[LATIN CAPITAL LETTER A]<<<" 12:58
p6eval rakudo 446d49: OUTPUT«>>A<<<␤»
ruoso bom dia #perl6 12:59
masak ruoso: bom dia!
moritz_ is that portugese? 13:00
ruoso yes 13:01
"sim"
masak likes portuguese 13:03
both of them.
mikehh_ rakudo (5960161) builds on parrot r411780 - make test / make spectest (up to r28216) PASS - Ubuntu 9.04 amd64 (g++)
rakudo - t/spec/S03-operators/arith.rakudo - TODO passed: 120, 131-132
colomon mikehh_: Oh, interesting. There's only one TODO passing on arith.t on my MacBook. Wonder if it's a 32-bit versus 64-bit thing? 13:05
moritz_ sees the asme TODO passes as mikehh_
pmurias ruoso: hi 13:07
13:07 icwiener joined
moritz_ /asme/same/ 13:09
jaffa8 p6eval returns strange characters. 13:10
Do you know that
?
moritz_ s/strange/unusual/
13:11 iblechbot left
jaffa8 it looks bad 13:12
colomon moritz_: I'm definitely only getting TODO passed: 131 Admittedly that's with Monday's rakudo, but I don't see any sign there have been relevant changes since....
masak jaffa8: are you referring to the ␤ character? that's a feature, not a bug. 13:13
ruoso pmurias, hi
13:13 elton_mcz joined
jaffa8 it is not displayed correctly here, 13:13
I am not sure what you see.
moritz_ colomon: could be a platform difference, yes
jaffa8 masak,yes
moritz_ jaffa8: does your IRC client properly support UTF-8? and your terminal? is it in the font?
jaffa8 no 13:14
moritz_ well
then it won't work
masak jaffa8: see topic: "UTF-8 is our friend!"
moritz_ the IRC logs also display it correctly, if your browser understands UTF-8
jaffa8 What character is that? 13:15
masak jaffa8: ␤
ruoso jaffa8, newline
represented as a NL inside a dotted box here
masak I only have a diagonally laid-out N and L here, without the box. 13:16
colomon moritz_: 0x80000000 div 1 == 0x80000000 is test 120, I'm getting -0x80000000 here.
jaffa8 it does not degrade gracefully.... 13:18
mikehh_ when I last tested on i386 I only got test 131 passing 13:19
13:19 __ash__ joined, mikehh_ is now known as mikehh
moritz_ then the tests should be moved to overflow.t 13:19
jaffa8: suggestions for better solutions are welcome
MoC` Btw, does someone know what hl "." does? It seems that it's a command on linux which is executed during a test, however Windows doesn't have a hl.exe 13:20
masak jaffa8: how does it degrade for you? if all you see is a box with numbers, I'd say that's fairly graceful.
13:20 MoC` is now known as M_o_C
masak M_o_C: which Perl 6 project does this, you say? 13:21
moritz_ anyway, there's no way non-ASCII-chars can degrade gracefully if you don't understand the character encoding
jaffa8 I see this: rakudo 446d49: OUTPUT«A␤»
masak jaffa8: looks perfectly fine here. :)
moritz_ that looks perfectly fine for me
13:21 finanalyst left
jaffa8 I see this ,appr. OUTOUTA<<a xA>> 13:21
I see this ,appr. OUTOUTA<<Aa xA>> 13:22
masak jaffa8: we have basically reserved the right to be slightly inconsiderate with encodings, in order to urge evolution in the right direction. we respect it if you don't get a better client/console, but you might not meed with a lot of sympathy.
s/meed/meet/
moritz_ masak++
M_o_C Sry, I was mistaken, it seems like it tries to execute "." and Windows complains that there is no command or programm named ".". Happens in make spectest: t\spec\S02-magicals\env.rakudo ................................. Failed 2/14 subtests 13:23
jaffa8 sympathy seems to be reare community where I live.
colomon moritz_, mikehh: occurs to me that it could also be a difference in Parrot versions. 13:24
moritz_ poor jaffa8
jaffa8 What about you?
Do you meet sympathy often? 13:25
moritz_ yes
jaffa8 How often?
masak actually, yes. I think so.
13:25 JimmyZ left
moritz_ I pick my acquaintances that way 13:25
jaffa8: sadly too seldom. I see my girlfriend only on weekends
masak I live in a fairly socialist state. its whole infrastructure runs on sympathy. 13:26
jaffa8 Where are you?
masak Sweden.
moritz_ hugme: hug jaffa8 and masak
hugme hugs jaffa8
moritz_ hugme: hug masak 13:27
hugme hugs masak
masak current gov't is right-wing, but that hardly matters. 50 years of socialist roots go deep.
jaffa8 I am surprised
masak hugme++
Matt-W Apparently I live in a state with a communist health service 13:28
I haven't yet figured out why that's supposed to be a bad thing...
jaffa8 WHat state?
IN sweden, what can people understand? 13:31
masak jaffa8: I'm not sure I understand; what do you mean? :)
moritz_ and I think Matt-W is UK based
jaffa8 IN sweden, what can people be sympathetic about?
Matt-W I'm in the UK, yes
masak jaffa8: a lot of things, I think. it's a very non-specific question. 13:32
13:33 nihiliad joined
mikehh moi aussi 13:33
moritz_ "everything they like" would be a possible answer
13:33 payload joined
jaffa8 yes, I see that 13:33
moritz_ but it's kind of a non-answer as well
ikwiki has an interesting concept - it stores its data (for example) in a git repository 13:34
moritz_ thinks about a possible November backend
mikehh The NHS here has fault yes - but compared to the US system where millions can not afford health care there is no comparrison
jaffa8 land of freedom. 13:35
mikehh freedom to starve
13:35 patspam joined
Matt-W yeah I can always go to the doctor and get some treatment 13:36
it might be a bit slow and it might be a bit primitive compared to the state of the art
masak checks out ikwiki
Matt-W but a lot of the time it will be good and timely and work
mikehh and I always thought the so-called justice system - well ...
moritz_ masak: I particularly like that I can edit the wiki off-line 13:37
jaffa8 moritz_,is it possible that your country has strong sense of community than others
?
moritz_ jaffa8: I don't think mine has 13:38
masak's, maybe
Matt-W mikehh: the justice system is increasingly worrying here
moritz_ Matt-W: here too
Matt-W and education
management culture infecting the police
moritz_ but in Germany the social situation depends very much on the region you live in 13:39
masak we have _some_ sense of community... but then we look at our neighbours, in frettled's country, and then we realize that we have virtually nothing. :)
Matt-W whole groups of kids who don't see why they should learn things and get a good job
mikehh yeah - a lot lot of things are worrying
masak we've had what amounts to a youth revolt here, what with the Pirate Bay situation and all.
I've even been to a demonstration myself. 13:40
frettled masak: Now you're _really_ scaring me.
Matt-W mikehh: Having friends in the police does give me a possibly gloomier picture than I might otherwise have, of course
masak frettled: which part?
frettled masak: about that sense of community thing
13:40 __ash__ left
mikehh it's the media :-} 13:40
moritz_ masak: well, the pirate party made it into parliament, no?
I think that's a great success
I don't think we'll see that in Germany too
masak moritz_: indeed.
frettled I see, BTW, that there has been progress with the Date/Time/whatnot specifications in S02 and S32. Goodie! 13:41
Accidentally, I've been a bit out of sorts, so I haven't even had time to read up on things :(
masak frettled: just compare May 17 to June 6 any year, I dare you.
frettled masak: ah, that's not community, that's national pride ;) 13:42
mikehh Matt-W: getting too close to a problem tens to give a distorted picture
masak frettled: good point.
mikehh tends
jaffa8 Matt-W, what problems can you see?
Matt-W mikehh: indeed, and when you work with people who've never worked and never will and have no concept of 'other people' other than as obstructions, and you do this all the time, it's bound to make you pessimistic 13:43
frettled masak: well, good soundbite, anyway
mikehh I think the problem is we live in a society that considers not working BAD and working GOOD 13:44
Matt-W But do we
Some parts they're completely indifferent to it 13:45
jaffa8 WHat is that a problem, mikehh?
Matt-W They just get benefits if they're not working, and they're happy
in fact, they're not sure why they should bother to work
KatrinaTheLamia pmurias, ah, so to you, a goal of Jerl 6, you would fast running binary? Or by that, did you mean fast compiling time? I was thinking aiming for a fast running binary, with a small amount put on the small memory foot stamp (with no real care on the actual binary itself)--however it occurred to me that people may see Jerl 6 as to be interpreted. So I may need to work on both fast compile time and fast execution time. Which means I
may not be able to anything with footprint or binary size. Thank you for your input ^.^
with no real care on the actual binary file size itself*
mikehh well working does not in general benefit them 13:46
and who are you working for anyway
jaffa8 themselves
?
mikehh most people who work for themselves have other motivations 13:47
Matt-W most people work for someone else who keeps all the product and gives them a salary in return 13:48
jaffa8 What are you trying to say?
there could be benefit in it.
mikehh sorry that was a sligt pun 13:49
slight
in the UK funds received from the state are called benefits 13:50
jaffa8 it boils down to how somebody is conditioned, I guess. 13:53
what belief system one gets.
Matt-W yes there's a lot to that
and we've got conflicting belief systems, which isn't very helpful
mikehh the puritan work ethic has pervaded Western society 13:54
Matt-W some of it, at least
jaffa8 Which is?
mikehh I ain't talkin' nabout them social misfits :-}
13:55 Muixirt joined
Muixirt hi 13:55
Invalid charset number '33' specified
what 's that supposed to mean?
moritz_ where did you get that error message from?
mikehh Matt-W: well maybe we were 13:56
Muixirt invoking a small perl6 program with rakudo
masak Muixirt: could you nopaste it? 13:57
Muixirt and '33' is nowhere in that program
masak, i fumble arounf with nci 13:58
mikehh Gotta get my grandkids from school in about 10 minutes
moritz_ ah well, parrot stores charset numbers in conjunction with strings
with things like 0 = binary, 1 = latin-1, 2 = Unicode or so 13:59
Muixirt 33 = ??
moritz_ invalid
14:01 patspam left 14:02 patspam joined
M_o_C Do any plans for making "make smoke" available to Rakudo exist? 14:03
Muixirt moritz_, so this pbc format is a fast moving target 14:04
KatrinaTheLamia M_o_C, I suppose that would be a method of incinerating Rakudo? (where there is smoke, there is fire ~.^)
moritz_ M_o_C: no, it needs a champion who takes care of that 14:05
Muixirt: what's pbc being a fast moving target has to do with your error message?
M_o_C I'm just asking because last week someone at least created a Rakudo project on smolder. 14:06
Muixirt moritz_, mixed up an installed parrot with rakudo
14:07 __ash__ joined
moritz_ Muixirt: that's not just because of .pbc - I don't think parrot makes any stability guarantuees for ABIs, only APIs 14:09
and compiled pmcs and dynops aren't pbc
Muixirt ok
14:11 vmbrasseur joined 14:13 M_o_C left
KatrinaTheLamia oh BTW--I has officially been played... a cheap copy of the Dragon Book that I acquired apparently was from The People's Republic of China. I should prolly be able to learn something here >.> 14:15
I personally found it hilarious though... piracy++
serves me the right ^.^
masak KatrinaTheLamia: fitting also that the "Dragon Book" comes to you from China. 14:16
KatrinaTheLamia (It should be noted, that like _most_ pirates, I am going to buy most of the stuff I pirated anyways--once my funds allow for the stuff) 14:17
masak, very fitting indeed ^.^
masak 龙书
14:19 justatheory joined 14:20 abra joined 14:23 hanekomu left 14:31 patspam left
pmurias KatrinaTheLamia: binary? you mean a .class? 14:39
14:42 jiing joined, __ash__ left 14:43 __ash__ joined
pmurias KatrinaTheLamia: i wouldn't worry about the compilation time at the start 14:44
14:45 payload left, Tene joined 14:50 Psyche^ joined
KatrinaTheLamia pmurias, I mean the bytecode/machinecode that the JVM is capable of reading and understanding. 14:50
14:52 Patterner left, Psyche^ is now known as Patterner
pmurias isn't machinecode what the CPU understands as opposed to bytecode which the JVM understands? 14:55
masak yay, second Squirrel test passes! can now create a table, insert a row, and read it back out. \o/ 14:57
Tene masak: Sorry, life keeps coming up... :(
phenny Tene: 09 Sep 21:26Z <masak> tell Tene that I've made nice progress on Squirrel. I've pushed it so that you can see (in the Web.pm repo). try running t/squirrel/sqlite-write.t -- the second test fails for me, even though I think it ought to pass. it's a mystery. will look at it tomorrow with fresh eyes.
masak Tene: no worries, I'm experiencing progress here. :) 14:58
Tene masak: Let me know if there's anything specific you want/need me to work on.
:)
pmichaud Good morning, #perl6
masak pmichaud: morning! 14:59
Tene AFK teaching
masak Tene: as usual, I'll need you most if I run into low-level trouble. but right now the path is pretty clear. all I need is lots and lots of tuits. :)
KatrinaTheLamia pmurias, well, technically, there are some CPUs (mostly cellphones) designed to run JVM bytecode as though it were their own instruction set 15:00
masak user-defined .{} syntax in Rakudo would be nice, too. but not a strict requirement.
15:03 patspam joined, alester joined
jaffa8 KatrinaTheLamia,what does the mean? 15:06
pmurias masak: isn't there an inferior embedded language called squirrel? 15:10
__ash__ you mean squirrel-lang.org ? 15:11
KatrinaTheLamia jaffa8, it means, that in some cases JVM bytecode, can be in fact machinecode for certain chipsets. 15:12
15:12 payload joined
pmurias __ash__: yes 15:14
masak pmurias, __ash__: I'm open for suggestions for other names than "Squireel". I chose that name because we'd been discussing SQL that day, and "Sequel" and "Squirrel" were apparently two different ways people chose to pronounce SQL.
TimToady squerl 15:15
jaffa8 KatrinaTheLamia, your name....
what does your name mean? 15:16
TimToady it's just like KermitTheFrog, except the the ends 15:17
masak TimToady: cute :)
TimToady s/the/for
__ash__ jaffa8: arm processors (like whats in the iphone) have a special set of instructions called Jazelle that let them run JVM byte code natively 15:19
diakopter __ash__: all arm processors? or since a certain time? 15:23
15:24 sahadev joined
__ash__ any with the Jazelle instruction set, not all ARM processor have it, but i know the iphone does (Cortex-A8 is what the iphone has) 15:24
15:25 ejs joined 15:26 frederico joined
TimToady oh, wow, actually finished backlogging before dinnertime, for a change... 15:27
I obviously woke up too early today
15:28 patspam left 15:31 Confield joined 15:32 patspam joined 15:33 hercynium joined
__ash__ is there a perl6 vim anywhere already? 15:34
diakopter __ash__: yes, in svn.pugscode.org/pugs/util/perl6.vim 15:35
pioto __ash__: yes, github.com/petdance/vim-perl/tree/master
diakopter or that
masak letter probably newer.
s/e/a/
KatrinaTheLamia jaffa8, I am a Coral Snake Lamia named Katrina ^.^--my other nick of FullMetalHarlot prolly isn't appropriate for FreeNode >.>
alester Yes, the latter is newer 15:36
and is going into the vim core
__ash__ alright, thanks
if you drop down to PIR code, what is the %r i keep seeing, is that the return value of the PIR? 15:42
masak looks up "lamia" and decides to be extra nice to KatrinaTheLamia from now on...
KatrinaTheLamia masak, wait... wha? why? 15:44
masak KatrinaTheLamia: "1. Classical Mythology. one of a class of fabulous monsters, commonly represented with the head and breast of a woman and the body of a serpent, said to allure youths and children in order to suck their blood." 15:47
KatrinaTheLamia: "2. a vampire; a female demon." 15:48
15:48 elton_mcz left
KatrinaTheLamia masak, meh, most of that is a eon old smear campaign done by Hera... you'd think Hera wouldn't get nearly as jealous about her husband sleeping around. I mean, it is pretty clear that Hera is the _only_ one that doesn't realise it is an open relationship. 15:49
masak KatrinaTheLamia: gotcha. 15:50
KatrinaTheLamia yeah, my top half is human female, and my lower half snake. But, I really don't enjoy sucking blood. I am mostly a vegetarian. And I don't think I am a vampyr... though, I do sparkle in the sunlight... which is somewhat suspect. But yeah, I am technically a demon... but prolly more closer to one of Eris' Diamone (which I cannot find any information about the Diamone in this timeline >.>) 15:52
jaffa8 Are you male, female or both?
15:53 sm` left
masak name suggests female. 15:53
KatrinaTheLamia jaffa8, female ^.^
15:53 Guest36141 left
jaffa8 your name is not very friendly 15:53
KatrinaTheLamia what? :(
what is so unfriendly about it :O 15:54
pmichaud "My name is plenty friendly, and I'll sever the head of anyone who disagrees with me on that!" :-) :-)
jaffa8 vampire?
15:54 fredrecsky joined
pmichaud KatrinaTheLamia: I think your name is fine. :) 15:54
masak me too!
KatrinaTheLamia jaffa8, not as far as I know... I mean, I don't desire drinking blood. And I can go out into the sun... though I find it odd that I start to sparkle >.> (only a sign of _how_ much exactly is _wrong_ with the world :O ) 15:55
jaffa8 KatrinaTheLamia, you am not talking about you... just about the name you use.
masak it's obviously a very well chosen name.
now that I understand it better, I feel I know KatrinaTheLamia better, too.
KatrinaTheLamia thankies masak 15:56
rjh hahahaha
KatrinaTheLamia huggles masak
masak squeals huggedly
rjh i feel like i'm at a sci-fi convention
KatrinaTheLamia rjh, well it is #Perl6... so you aren't far off in the channel about TEH LANGUAGE OF TEH FUTURAH! 15:57
masak rjh: what do you expe... what KatrinaTheLamia said. :)
KatrinaTheLamia anyways... I think that last sentence I said dropped the IQ level of the room a few levels >.> 15:58
rjh i'm going back to #perl where everyone is a jerk to each other
:)
KatrinaTheLamia rjh, ah... but if we work things out, we may start a group hug...
KatrinaTheLamia actually censored that above sentence from something potencially more offensive >.>
jaffa8 I am thinking about a new name.
masak jaffa8: don't fix it if it ain't broken :) 15:59
TimToady jaffa9?
masak 哈哈
TimToady hey, I'm good with names :)
pmichaud after that would be jaffb0
jaffa8 What about IamnicemonsterIwillnoteastyoutoday? 16:00
pmichaud too long
TimToady where's the west of your name?
rjh jaffacake
pmichaud how about "SatiatedFriendlyMonster" 16:01
KatrinaTheLamia jaffa8, well, do you have a phenotype?
jaffa8, any particular movie or music interests--or interests of other kinds?
masak JaffaTheIncubus
KatrinaTheLamia I mean TimToady is obvious a fan of Perl, and that is where his (I am assuming) name came from.
hercynium JaffaKree!
pmichaud TimToady likes Perl? 16:02
TimToady you could pick a name that starts with 'p' like everyone else
jaffa8 TimToady, do you use Perl?
TimToady I hate Perl, but I use it twice a day...
KatrinaTheLamia huggles TimToady
hercynium Perl is overrated. I'm waiting for Jade7
KatrinaTheLamia clearly, we have not brainwashed TimToady well enough. 16:03
TimToady oh, wait, that's Listerine
masak :)
KatrinaTheLamia sends TimToady back to the "rehabilitater" for repair ^.^
jaffa8 TimToady, what do you use Perl for?
hercynium said by somebody @ Boston.PM the other day: "I write perl at work to get things done but nobody knows it yet" 16:04
pmichaud hercynium: I hear that a lot
hercynium (he works at a heavy python-centric place (I used to work there too))
pmichaud "I use Perl, but please don't tell my boss."
KatrinaTheLamia "I use Perl 6, as it is Eris' favourite language ^.^" 16:05
pmichaud mmm lunchtime
Tene I'm always glad to see another discordian in the channel. ^^
jaffa8 I made a code formatter in Perl. 16:06
as far as I am concerned.
pmichaud I made a Perl code de-formatter in Parrot. :-)
Tene fdisk /dev/pmichaud
jaffa8 deformatter?
hercynium just the other day, another former co-worker emailed me with a perl 6 question... 16:07
pmichaud Yes. It takes properly formatted Perl 6 code and turns it into an executable.
hercynium he wanted to know if there was a working "http object" in perl6 yet
I'm guessing he meant some sort of user-agent 16:08
pmichaud I know we have mod_parrot and mod_perl6
I don't know if we have a user agent yet. Shouldn't be too hard to create one.
Tene HTTP client is one of the too-many items on my TODO
hercynium I am curious as to what the state is of IO related stuff 16:09
jaffa8 IS it the same as the current perl -> pir -> pbc -> exe path?
pmichaud some of it is ready, some is blocked, and some is closed. :-)
hercynium (sockets, files, pipes... asynch, etc?)
pmichaud jaffa8: yes, that's basically how it works :)
TimToady jaffa8: I once wrote a Perl 6 parser in Perl... 16:10
16:10 cmv joined
KatrinaTheLamia TimToady, heh... here is a challenge: Perl 6 parser in Perl 0 ^.^ 16:10
hercynium if I can do asynch work with sockets I think I have a few perl6 mini-projects I'd like to start
TimToady Well, Perl 0 was turing complete
jaffa8 p6eval, Is it different from that?
pmichaud I think Parrot (and thus Rakudo) is a little weak on async I/O
TimToady unfortunately it was also classified, so I don't have a copy
pmichaud anyway, lunchtime here. bbiaw 16:11
jaffa8 pmichaud, Is it different from that?
hercynium ok.. (was going to also ask... wis that planned for Rakudo *?)
KatrinaTheLamia TimToady, I am certain if you propose to Larry that you intend to write a Perl 6 parser in Perl 0, he would grin madly at your insane request and grant you a copy of Perl 0 ^.^ 16:12
pmichaud I don't think async I/O has been specifically targeted for Rakudo * yet
diakopter tee hee
hercynium OK. good to know :)
Tene hercynium: I'm also blocking on async IO for some projects... I was working on getting threading support in Rakudo last night, and got farther than I did last time, but I ran into some nasty broken plumbing in the bowels of Parrot. 16:13
hercynium KatrinaTheLamia: I just asked Clark Kent to let Superman know I've got an extra ticket to the U2 concert next week.
rjh hahaha
KatrinaTheLamia TimToady, I am also certain that such a insane request, if successfully pulled off, would prolly either ensure Perl 6's place in the world... or doom it for all eternity. 16:14
pmichaud ...or both.
KatrinaTheLamia missed hercynium's joke completely... but has an idea of what he may have suggested >.>
diakopter pmichaud: :}
hercynium :)
__ash__ Tene: what kind of problems does parrot have with threading? I have been curious on that, not that i could help any in all likely hood
16:14 frederico left
diakopter wonders what perl0 was codenamed. maybe FOIA could help. 16:15
Tene __ash__: Parrot's threading works well in restricted cases, as far as I can tell. The problem is that it falls on its face when cloning the interpreter when certain things are true.
TimToady hercynium: I'm going to have to reprogram your brain to forget something; sorry about that.
KatrinaTheLamia ah, so TimToady is none of then Mr. Wall. 16:16
Tene __ash__: Things like "has a class in a nested namespace" or "has ever loaded any dynamic libraries" so far.
KatrinaTheLamia meh, neat ^.^
hercynium looks at the neuralyzer
*where am I??*
KatrinaTheLamia hercynium, I'll tell you where you are not... safe ^.^
__ash__ So, cloning the interpreter seems to not be cloning deep enough? 16:17
Tene is reminded of yudkowsky.net/other/fiction/prospiracy-theory
hercynium I'm never safe when I'm alone
KatrinaTheLamia oh, you're not alone...
Tene __ash__: It's not that it doesn't clone some things, it's that it crashes badly when it does the wrong thing when trying to clone them.
KatrinaTheLamia because look, sprout, we appear to have visitors.--oh shoot, I've forgotten the rest of this TFS DBZ Abridged line of quotes >.> 16:18
16:19 frew__ left, frew_ joined
hercynium Tene: I can't begin to pretend I understand how/why parrot does what it does... but is the interpreter cloned in-process when a thread is requested? 16:19
Tene hercynium: Yes. 16:20
KatrinaTheLamia Tene, wouldn't that make it behave more like a fork, than an actual thread? 16:21
hercynium seems a bit heavy to me... it makes sense if you consider the interpreter as another "CPU" in the VM though, I guess
Tene KatrinaTheLamia: I don't actually understand the issue well-enough to say; sorry.
16:22 rfordinal joined
hercynium hunts down chromatic to sate his curiosity! ;-) 16:22
__ash__ Tene: should you be using a Task? docs.parrot.org/parrot/latest/html/...y.pod.html rather than an interpreter clone? 16:24
masak Tene: speaking of TODO... do you think we'll see a decent Rakudo REPL by April?
KatrinaTheLamia Tene, how I'd likely handle it, is merely have each thread have its own process inside the VM. Where it goes through both instruction stacks with the same otherwise interpreter states elsewhere. About the only thing local to those would be the relevant scope. I mean outside the actual instruction stack and local scope, it should see everything else the other threads see. If you are really nice, a possible feature to be able to re 16:25
nice different threads could be added as well ^.^
but then, serves me right for reading the UNIX 6 source code, as I've prolly screwed lots up >.> 16:26
16:27 sm` joined
__ash__ Parrot has 2 different ways of concurrency that I see currently, creating a "Task" and "ParrotThreads" which create a clone of the Interpreter but I am not sure which is the right thing to use in rakudo 16:29
the Task doesn't fully clone the Interpreter
16:29 masak left
__ash__ but like I said, I haven't really looked into how they work in much detail, so I am not sure what the big differences are 16:30
16:30 cotto joined
__ash__ gotta go to lunch bbl 16:30
16:30 __ash__ left
KatrinaTheLamia yeah, since it would be a thread, and not an actual fork, I'd prolly just put the instruction stack and local scope of a thread into a seperate area of memory, and just switch which instruction is done based on the various nice levels of the threads. 16:31
so we'd have main: 1 tick; thread 1: 1 tick; thread 2: 1 tick; main: 1 tick;--and so on... or something to that effect.
prolly getting to the point to making Parrot able to be nicer to some threads than others, based on some manner of command. 16:32
possibly have the niceness range from -23 to 23, with them defaulting to a nice value of 1 16:33
16:34 icwiener left
KatrinaTheLamia but then, I am likely messing a _lot_ up here >.> 16:34
this is how I'd approach it without even looking at the current code. 16:35
16:35 abra left
KatrinaTheLamia again each "Kind::Thread" as I guess I could call it, would only contain the instruction stack and variables it can see at its scope--but can view various scopes of other namespaces... like global or main >.> 16:37
just how I'd approach it.
so that we don't have Kind::Threads eavesdropping on other Kind::Threads, as that is not proper behaviour, we'd most likely have their scope in the program not writable or readable by other Kind::Threads. Except for possibly the main thread, with would likely end up with some form of Semaphore files or something like that. 16:38
you are more than welcome to tell me how wrong I am at any given time, BTW ~.^ 16:39
Tene KatrinaTheLamia: Threads and Tasks in Parrot haven't been worked on in a very long time. If you're interested in trying to work on this and make it a bit more sane, it would be very welcome. :) 16:40
hercynium knows far too little about it to say she's wrong :)
KatrinaTheLamia Tene, I may look into it ^.^ 16:41
Tene KatrinaTheLamia: If you do, feel free to ask me about anything.
KatrinaTheLamia thankies Tene ~ 16:43
Tene, I'll assume parrot is on Github?
KatrinaTheLamia is still a level 1 Knave at Perl 6 :O
PerlJam KatrinaTheLamia: parrot is an svn project still 16:45
Tene KatrinaTheLamia: Parrot uses svn, but I always access it through git-svn, because I can't use svn without screwing things up. 16:46
(I learned that the hard way)
PerlJam KatrinaTheLamia: svn.parrot.org/parrot/trunk # BTW
Tene trac.parrot.org/parrot/wiki/git-svn-tutorial is an intro to using Parrot with git-svn. 16:47
KatrinaTheLamia thankies~
PerlJam Just so I don't have to read throughg scrollback ... you guys told KatrinaTheLamia about #parrot, right? 16:48
KatrinaTheLamia on of these days, I will have finished my genocides of Slimes, and become a Level 2 Knave at Perl ^.^
there doesn't appear to be anybody in #parrot BTW >.>
PerlJam KatrinaTheLamia: irc.perl.org
KatrinaTheLamia: different network
KatrinaTheLamia I knew that >.> 16:49
Tene AFK teaching 16:50
KatrinaTheLamia PerlJam, does irc.perl.org have SSL? 16:51
hmm.. apparently not >.> 16:52
oh and thankies PerlJam ^.^ 16:55
16:57 patspam left
PerlJam KatrinaTheLamia: hey, if you're volunteering to do stuff... thank *you* :) 16:58
KatrinaTheLamia I expect to be paid in cuddles~ ^.^
cookies would be nice too~ ^.^
PerlJam hugme hug KatrinaTheLamia 16:59
hugme hugs KatrinaTheLamia
16:59 jan_ joined 17:03 Bzek joined, stephenlb joined 17:04 M_o_C joined 17:11 M_o_C left
pmichaud phenny: tell masak I expect us to have a decent REPL by December. Probably sooner than that. 17:13
phenny pmichaud: I'll pass that on when masak is around.
17:22 frederico_br joined 17:34 sevvie left 17:41 fredrecsky left 17:43 __ash__ joined 17:46 Zloyrusskiy joined 17:48 Zloyrusskiy left 17:49 M_o_C joined 17:50 rindolf joined
M_o_C Interesting: When Java can't extract the files to the installation target because there are also files with the same name the german version will quit with the error: "Can't compress core files.". Very helpful diagnostic message, Sun, well done. :x 17:52
17:59 frederico_br left 18:05 patspam joined 18:12 Bzek left 18:13 ruoso left 18:23 rindolf left 18:26 payload left 18:31 mberends joined, patspam left, patspam joined 18:35 snearch joined 18:36 snearch left 18:38 sahadev left 18:40 sevvie joined 18:43 sevvie left, sevvie joined 18:47 __ash__ left, patspam left, zloyrusskiy joined 18:48 patspam joined 18:52 zloyrusskiy left 18:56 jrtayloriv joined 19:04 cdarroch joined 19:05 ejs left 19:08 MoC` joined, japhb left 19:15 rbaumer left 19:19 patspam left, patspam joined 19:20 sm` left, sm` joined, patspam left 19:21 patspam joined 19:23 jrtayloriv left 19:24 patspam left, patspam joined 19:25 M_o_C left 19:28 pdcawley joined 19:35 sevvie left 19:37 rbaumer joined 19:40 vmbrasseur left 19:43 Zloyrusskiy joined 19:49 __ash__ joined 19:54 takadonet left 19:56 rbaumer left 20:02 mberends left 20:07 alester left, Zloyrusskiy left 20:09 Zloyrusskiy joined 20:21 patspam left, patspam joined 20:24 stephenl1 joined
colomon __ash__: Did you get pmichaud's patch for the Int.Num method to work last night? It didn't seem to work when I tried it. 20:27
__ash__ it works for some tests, but not all of them
there still seems to be something returning a float when it should be returning a Num
the fact that its a for loop kinda messes it up because some of the tests in the loop works, while other dont 20:28
20:29 patspam left, patspam joined
colomon Hey, actually that's really interesting. Let me take another look at it... 20:29
__ash__ oh 20:32
if you want to see what the parrot data type is do .parrot
rakudo: say 1.Num.PARROT
p6eval rakudo 446d49: OUTPUT«Float␤»
__ash__ ^^ float, not num, there are a few other conversions that return floats that should be returning Nums
20:33 icwiener joined
colomon I'll bet it has something to do with the arithmetic operators. 20:34
__ash__ probably, look into Num division
thats where i'd check nexct 20:35
next
rakudo: say (1 / 1.5).PARROT
p6eval rakudo 446d49: OUTPUT«Float␤»
__ash__ ^^
there you go
thats another place its doing the wrong datatype
colomon Give me a minute or two and I hope to have it narrowed down nicely. :) 20:36
20:36 stephenlb left
__ash__ kk 20:36
it has to do with in the Q:PIR { } segments where it boxes the values 20:39
the boxing returns a float in some cases 20:40
20:44 Zloyrusskiy left
__ash__ docs.parrot.org/parrot/latest/html/...C,_in_NUM) there, when it calls box it should be converting that into a Num but currently its converting the value into a float, pmichaud said there may be a way to get the box to return a Num which would be a nicer way of solving the problem because as it stands if you ever box a num it always returns a float where it should be returning a Num 20:44
colomon I can confirm: with the patch in there, 5.Num.sin works in any base. 20:45
However, (5.Num * 2.0).sin('radians') fails.
__ash__ rakudo: say (5.Num * 2.0).PARROT;
p6eval rakudo 446d49: OUTPUT«Float␤»
pmichaud anytime we ask parrot to automatically box a numeric ($N0) into a PMC, it will currently make it into a Float
we can change it so that it always makes a Num, but we take a speed hit from doing that 20:46
__ash__ i'd guess all arithmetic will return a float
pmichaud if you'd like, instead of changing all of the "box" opcodes, we could try turning on the type mapping and see what happens
just a sec
(creating patch)
dalek kudo: 8ea834f | pmichaud++ | build/PARROT_REVISION:
Bump to 41181, to (hopefully) get some speed improvements.
 20:47
colomon What's happening with the trig.t test is the angles are all specified in Int degrees, and then when we convert to whatever we're actually testing we turn it into a Float by accident. 20:48
Except degrees, which should work after the Int.Num patch. 20:49
pmichaud I have a different patch to try
doing a quick test on it now
colomon k
20:50 payload joined
pmichaud rakudo: say 3.Num.PARROT 20:50
p6eval rakudo 446d49: OUTPUT«Float␤»
pmichaud okay. just a sec
gist.github.com/184809 # patch to automatically convert all Float PMCs into Num objects in Rakudo 20:51
on my system with that patch, I get
pmichaud@plum:~/rakudo$ ./perl6
> say 3.Num.PARROT
Num
lambdabot Not in scope: `say'Not in scope: data constructor `Num.PARROT'
pmichaud and "say 3.sin" works
try that patch, see what fails, and let's see if it makes things incredibly slow 20:52
(I can't do it now, have to help kids with homework)
colomon pmichaud: Actually that one should work because I got Int.sin working a few days ago.
I'll try the patch asap here and report back.
20:53 sharada left
colomon Going to take a few minutes, as I've just updated to the latest rakudo, and want to make sure it works as-is before patching it. 20:54
__ash__ i am rebuilding rakudo right now with the patch applied 20:56
20:57 cmv left 20:58 frew__ joined, frew_ left
__ash__ it worked for me, trig passes all the tests from the seciton that was failing 20:59
colomon \o/
__ash__ now its a complex number problem, but the Num ones all pass 21:01
for me
colomon If the Num stuff works, I can probably have Complex working pretty quickly, I think.
Unless there's some sort of hard-coded Float inside it...
__ash__ well at least you know how to figure out if its a Float and a good idea where to fix that 21:03
colomon Yup. Once again, __ash__++ and pmichaud++. :)
21:04 sevvie joined 21:06 icwiener left 21:07 krunen left 21:13 l3thal joined 21:14 japhb joined 21:18 l3thal left
frettled unsnoozes, «complex,», he heard. 21:26
:)
__ash__ colomon was working on some of the trig functions that use complex
frettled yes, I saw that was mentioned :) 21:27
21:31 jaffa8 left
pmichaud on my system, my patch causes all tests to pass and doesn't seem to hurt performance too much. Shall I apply it? 21:32
pmichaud applies. 21:33
frettled pmichaud++, I hope.
pmichaud applied, pushed. 21:34
__ash__ pmichaud it worked for me too, i didn't clock it for a comparision however, but it does fix some bugs... 21:35
pmichaud on my system it seemed to cost about 5-10 seconds total -- that's noise level as far as I'm concerned 21:36
dalek kudo: d0355a5 | pmichaud++ | src/classes/Num.pir:
Enable hll_map for Float->Num. This enables us to pass some tests
 21:37
21:39 ruoso joined
hercynium is all happy, his camelia shirt already shipped! 21:40
pmichaud hercynium: let me know how it looks when you get it -- I have no idea what the quality will be. where did you order from?
hercynium spreadshirt - looking around online they seem to have better ratings in regard to quality 21:41
pmichaud that's what I had heard as well. But I don't know if that's for their vector-based shirts only or for the dot-matrix ones also
21:42 tak11 joined
hercynium I'll be sure to post my opinion when it comes, maybe after it's first washing :) 21:42
21:47 huf_ is now known as huf 21:48 rbaumer joined 21:52 patspam left
ruoso hercynium, can others reues the shirt you ordered? 21:52
hercynium reues? re-use? 21:53
ruoso reuse, yea
but I already found it
hercynium heh
maybe my girlfriend, but I generally don't pass my clothes around, schwern-style ;-) 21:54
ruoso heh
pmichaud I can make other custom designs/slogans if anyone wants them 21:55
ruoso I didn't get the phrase, tho
pmichaud the "second system" phrase? 21:56
hercynium ruoso: en.wikipedia.org/wiki/Second-system_effect
pmichaud "What's your problem" has at least three meanings in English (more) 21:57
hercynium however, Larry has (as usual) found multiple meanings for the quip,
what pmichaud said :)
pmichaud (1) "describe the problem you're having, so that I can help" 21:58
(2) "that's what is wrong with me... what's wrong with you?"
(3) "Yes, I'm a second system. Do you have a problem with that!?!"
21:58 arthur-- joined
pmichaud (latter is usually said sarcastically or in a threatening tone) 21:59
ruoso it's cool... the only problem of having it in a tshirt is that few people know what "Second System" menas 22:00
pmichaud sure, it's a bit of an inside joke
we'll take other taglines as they come
I'm still thinking of "Less code. More Perl 6."
ruoso pmichaud, have you seen Tim Bunce's today post? 22:01
pmichaud or perhaps even: "The rumors of my death have been greatly exaggerated."
moritz_ ruoso: the outsiders still see a nice butterfly ;-)
pmichaud ruoso: yes, I did see the post (I think)
moritz_ which post? 22:02
hercynium I will tell people that it's Mothra's adolescent form
moritz_ saw nothing
pmichaud moritz_: bit.ly/15VKCh
ruoso Perl: a great language for getting your job done, for the last 20 years, and the next 20
hercynium (perhaps python is godzilla's younger form?)
moritz_ pmichaud: thanks 22:03
lisppaste3 colomon pasted "MacBook Pro spectest fails before Float -> Num patch" at paste.lisp.org/display/86895 22:05
22:07 payload left 22:08 synth left
pmichaud colomon: hmmmm 22:10
colomon pmichaud: just to be clear, that's the before. I had those failures in my previous build as well. 22:11
starting the after spectest run now, should have more info in a bit.
afk
22:13 hercynium left 22:20 arthur-_ left, PZt joined 22:25 nihiliad left 22:39 synth joined, MoC` left 22:44 vmbrasseur joined, Muixirt left 22:46 sm` left 22:51 vmbrasseur left 22:55 ruoso left, pmurias left 23:05 hercynium joined
colomon wow, the spectests are taking a long time these days... it's been running solid since my last comment here. 23:08
lisppaste3 colomon annotated #86895 "post-pmichaud Float -> Num fix" at paste.lisp.org/display/86895#1 23:12
23:16 sevvie left 23:26 frew_ joined
colomon Num.sin and Num.cos tests that were previously skipped work just fine now. 23:27
I've got to run out and buy cat and dog food, but I'll take a look at the failing Complex tests (currently fudged) when I get back. 23:28
23:41 meppl joined 23:49 payload joined 23:50 __ash__ left 23:51 frew_ left 23:59 tak_ joined, tak_ left

Powered by Cro and the Raku® Programming Language.

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

Thank you!