»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend!
Set by sorear on 4 February 2011.
TimToady sorear: the spec only mentions int64, not Int64 00:06
dalek ecza: f95532e | sorear++ | / (5 files):
[mmd] Implement bindability checks
 00:07
ecza: 70fa410 | sorear++ | / (2 files):
[mmd] Implement redispatching to candidate list
TimToady but I view storage types as primarily constrained values of a larger type
sorear TimToady: this was prompted by thundergnat looking for :rotate 00:08
I'm not sure how :rotate is expected to discover the type size
on +> / +<
00:09 ymasory joined
TimToady I suppose we could do :rotate(64) 00:10
or we could somehow mark ops whose unboxed types are known at compile time 00:12
that might take some type inference though 00:13
bbl &
00:15 sftp left, sftp_ joined 00:16 starcoder left, starcoder joined 00:21 whiteknight joined 00:32 icwiener left 00:33 benabik joined
Tene TimToady: any thoughts on sorear's assertion that *&foo in a signature is unimplementable? 00:34
00:34 justatheory left
sorear *&foo appears to be a relic from the Perl5 '&' signature 00:34
00:43 madalu joined 00:46 wallberg left 00:48 madalu left 00:52 ymasory left 01:11 justatheory joined 01:19 whiteknight left 01:23 leprevost joined 01:27 mtk left 01:28 gantrixx left, gantrixx joined 01:35 mtk joined 01:44 kanishka left 01:49 silent_h_ joined 02:00 Alias joined 02:01 Woody2143 left, larsen_ left, edenc left, hillu_ left, dukeleto left, aesop left, athomason left, szbalint_ left, Grrrr left, PerlJam left, [Coke] left, sunnavy left, BinGOs left, Raynes left 02:02 Sarten-X left, ggoebel left, xinming left 02:03 Sarten-X joined, ggoebel joined, xinming joined, amkrankruleuen joined, Woody2143 joined, LoRe joined, huf joined, larsen_ joined, hillu_ joined, dukeleto joined, aesop joined, athomason joined, Grrrr joined, PerlJam joined, [Coke] joined, sunnavy joined, BinGOs joined, Raynes joined 02:04 edenc joined, szbalint joined, nsh_ joined 02:11 silent_h_ left 02:26 agentzh joined 02:48 takadonet1 joined
takadonet1 hey 02:48
02:48 gantrixx left
takadonet1 anyone know what the equivalent is for Perl 5 $/ in Perl 6? 02:49
02:51 ymasory joined 02:53 woosley joined 02:57 leprevost left
sorear takadonet1: S28:182 02:58
takadonet1 sorear: looks NYI :( 03:01
03:06 hudnix left
tylercurtis std: $/ 03:08
p6eval std 4608239: OUTPUT«ok 00:01 118m␤»
tylercurtis ah, right.
takadonet1 sorear: thx 03:20
night all
03:20 takadonet1 left 03:21 JimmyZ joined
JimmyZ TimToady: ping 03:21
03:23 Su-Shee_ joined 03:27 Su-Shee left 03:33 Exodist left, nymacro joined 03:38 JimmyZ left
sorear rakudo: multi foo(Int $a, Int $b) { say 1; nextsame }; multi foo(Any $a, Int $b) { say 2 }; multi foo(Int $a, Any $b) { say 3 }; foo(5,6); 03:43
p6eval rakudo 4bf132: OUTPUT«1␤Null PMC access in clone()␤ in 'foo' at line 1␤ in main program body at line 22:/tmp/M1X43YJANG␤»
sorear What?
03:50 JimmyZ joined, MindosCheng left, JimmyZ left 04:01 gantrixx joined 04:02 JimmyZ joined, orafu left 04:04 orafu joined 04:08 plobsing_ left 04:18 agentzh left 04:29 ymasory left 04:30 cibs joined, kst left 04:52 ymasory joined 05:17 sftp_ left 05:32 Khisanth left 05:36 arlinius left 05:37 justatheory left 05:48 arlinius joined 05:53 Guest84973 joined 05:55 wtw joined 05:58 Guest84973 is now known as Khisanth
sorear today's .NET pet peeve: arrays use .Length while lists use .Count for no obvious reason 05:58
06:00 woosley left, woosley joined 06:08 donri joined
dalek ecza: fb61e2b | sorear++ | lib/Kernel.cs:
[mmd] Implement candidate sorting
 06:14
ecza: 2dd3fec | sorear++ | lib/Kernel.cs:
[mmd] Implement tie checking
sorear 'multi method' is now fully functional
tomorrow I'll work on 'multi sub' 06:15
also tomorrow will cache dispatch lists 06:16
06:16 kaare_ joined 06:20 birdwindupbird joined, birdwindupbird left 06:21 birdwindupbird joined 06:27 birdwindupbird left, birdwindupbird joined, noganex_ is now known as noganex 06:36 fhelmberger joined 06:45 Su-Shee_ is now known as Su-Shee 06:53 shi joined 07:01 kanishka joined 07:04 fhelmberger left, fhelmberger joined 07:17 wallberg joined 07:18 mj41 joined 07:33 kanishka left, Mowah joined 07:43 woosley1 joined 07:45 woosley left 07:46 mtoma joined, mtoma left 07:47 mtoma joined, mtoma is now known as miso, miso left 07:48 drbean_ joined 07:50 miso2217 joined
miso2217 perl6: say 3; 07:51
p6eval pugs, rakudo 4bf132, niecza v4-26-g2dd3fec: OUTPUT«3␤»
moritz good morning
miso2217 perl6: if (1){say 3;} 07:52
p6eval pugs: OUTPUT«*** ␤ Unexpected ";}"␤ expecting "_", fraction, exponent, term postfix, operator, ":", "," or "}"␤ at /tmp/egQTdujPvN line 1, column 13␤»
..rakudo 4bf132: OUTPUT«===SORRY!===␤Missing block at line 22, near ""␤»
..niecza v4-26-g2dd3fec: OUTPUT«===SORRY!===␤␤Any()Missing block at /tmp/bxdGIZlF1n line 1 (EOF):␤------> if (1){say 3;}⏏<EOL>␤␤Parse failed␤␤»
miso2217 perl6: if (1) {say 3;}
p6eval pugs, rakudo 4bf132, niecza v4-26-g2dd3fec: OUTPUT«3␤»
moritz perl6: if 1 { say 3 } # no need for those parens
p6eval pugs, rakudo 4bf132, niecza v4-26-g2dd3fec: OUTPUT«3␤» 07:53
JimmyZ std: if (1){say 3;} 07:55
p6eval std 4608239: OUTPUT«===SORRY!===␤Missing block at /tmp/zM2w3TW8cL line 1 (EOF):␤------> if (1){say 3;}⏏<EOL>␤Parse failed␤FAILED 00:01 119m␤»
moritz if( is parsed as a function call 07:56
thing{...} is parsed as a postcirumfix
JimmyZ and if ( ?
moritz JimmyZ: as a conditional
an 'if', where the '(' is a circumfix
JimmyZ so if (1){ ... } ? 07:57
moritz (which is usually a no--op
JimmyZ: that's parsed as if (1).postcircumfix:<{ }>( ... }
and then there's no block left for the if 07:58
which is why STD complains about missing block
miso2217 It was somwhat surpsising as I always used the if (1){ form, the error message is somewhat not too clear. Should be somewhere in the doc that the space before the { is needed, I think
07:58 pjcj left
miso2217 Always used in perl5 of course 07:58
moritz miso2217: the spec is not user-level documentation. It's in the spec, but not very explicit
07:59 Rotwang joined
moritz we really need to have user-level documentation that explains such stuff 07:59
JimmyZ std: if (10){} { say 'hi'; } 08:00
p6eval std 4608239: OUTPUT«ok 00:01 120m␤»
moritz it will be a run time error of course
JimmyZ std: if (10){say 'hi'} { say 'hi'; }
p6eval std 4608239: OUTPUT«ok 00:01 120m␤»
moritz but it's valid syntax
miso2217 Yes of course, it is but as there is not yet a user level documentation yet I checked the specs.
perl6: if (10){say 'hi1'} { say 'hi2'; } 08:01
moritz miso2217: true. But adding user-level docs to the spec is the wrong fix
p6eval rakudo 4bf132: OUTPUT«hi1␤»
..niecza v4-26-g2dd3fec: OUTPUT«hi1␤Unhandled exception: Cannot use hash access on an object of type Num␤ at /tmp/6U8fM8nLma line 1 (MAIN mainline @ 1)␤ at /home/p6eval/niecza/lib/CORE.setting line 1277 (CORE C535_ANON @ 2)␤ at /home/p6eval/niecza/lib/CORE.setting line 1278 (CORE module-CORE @
..39)␤ at /home/p…
moritz and there is some doc, just incomplete
p6eval ..pugs: OUTPUT«hi1␤*** Odd number of elements found where hash expected: VInt 10␤ at /tmp/X42di0Ts0J line 1, column 4-20␤»
JimmyZ so, the right things is 08:05
08:05 drbean_ left
JimmyZ rakudo: if (1){ say 3;} {} 08:05
p6eval rakudo 4bf132: OUTPUT«3␤»
moritz miso2217: github.com/perl6/book/issues/53 I'v submitted a bug report against the book in progress, so that we don't forget it
JimmyZ that's ugly 08:06
moritz JimmyZ: it's a known bug in rakudo that hash-indexing a string is not an error 08:07
miso2217 I don't really get it, what is the if (1){ say 3;} {} form suposed to do? Call the {say 3;} code block on the (1) list and evaluat that as a bool? 08:08
JimmyZ also a std bug?
moritz miso2217: the { say 3; } is parsed as a hash index 08:09
miso2217: just like in %hash{ $key }
JimmyZ: nope. STD doesn't actuall run the code, just parses it
JimmyZ: so it can't know it's an illegal operation
jnthn morning o/ 08:11
moritz \o
JimmyZ good morning
jnthn: where are you now? sweden ？
08:12 bacek left
jnthn JimmyZ: yes 08:12
miso2217 moritz: so it thries to evaluate (1) as a hash, ok, it's clearer now, thx. 08:13
08:13 drbean joined, pjcj joined
JimmyZ jnthn, I just got the link. conference.perlchina.org/bjpw2011/ 08:19
08:24 miso2217 left
jnthn JimmyZ: Ah, OK. One day in June. 08:25
08:25 JimmyZ_ joined 08:27 miso2217 joined, JimmyZ left, JimmyZ_ is now known as JimmyZ 08:35 domidumont joined 08:36 ab5tract joined 08:38 Rotwang left 08:44 toebu left 08:54 awoodland joined 09:03 jaffa4 joined
jaffa4 hi 09:04
09:13 mtk left 09:17 autin joined 09:19 JimmyZ left 09:21 pmurias joined 09:22 awoodland left, mtk joined 09:27 ab5tract left 09:29 kanishka joined
pmurias phenny: tell sorear can we drop log from the unit in nam? 09:36
phenny pmurias: I'll pass that on when sorear is around.
09:45 gantrixx left
miso2217 hello, I have two versions of rakudo, rakudo-star-2011.01 and the latest trunk from git. I have a smoll test script which I execute using rakudo-star-2011.01/perl6 ent it works. But if I try with the latest trunk I have an error mesasge: PARROT VM: Could not load bytecodeCould not load oplib `perl6_ops'. This is not really a serious idssue as I only copied the modules directory from rakudo star into the latest rakudo from git. The real problem is that 09:51
when after this I try to execute my script with the perl6 in rakudo-star I get a segfault. Any ideas?
moritz miso2217: have separate install locations for both versions. Always run 'make install'. 09:54
miso2217 moritz: what do you mean by "have separate install locations for both versions"? does the 'make install' command try to write comewhere else than the original rakudo directory? I downloaded both vesions in two separate directories "~/TEMP/rakudo-star-2011.01" and "~/TEMP/rakudo" and I think i run make install in both of them. But I did not specify any other install location as I tough I was only writing to the original directories. 10:00
moritz miso2217: how did you configure the rakudos? 10:01
miso2217: if you configure the current dev version with 'perl Configure.pl --gen-parrot', it will then install into the parrot_install subdirectory 10:03
miso2217: if you do that, and don't have the other perl6 executable in $PATH, you should be safe 10:04
miso2217 I run: perl Configure.pl --gen-parrot in both directories
I'll check my $PATH 10:05
I have only ~/TEMP/rakudo/perl6 and ~/TEMP/rakudo-star-2011.01/perl6 none of them is in the $PATH 10:10
moritz where did the star install to? 10:11
miso2217 in the star directory only: ~/TEMP/rakudo-star-2011.01/, at least this where it generated the perl6 executable ant the parrot-3.0.0 directory 10:14
moritz that's where it generated it, but that's not where it installed it
miso2217 ~/TEMP/rakudo-star-2011.01/install/ 10:17
moritz then I don't see how the installations could interact :(
miso2217 Neither me, i'm checking if the star installation didnt put the parot exe somewhere else, but I dont think so. 10:20
10:22 autin left
miso2217 And it did not :( strange. Re-running make install in the star directory solves the problem. 10:23
10:26 Mowah left 10:38 dayangkun_ left 10:41 jaffa4 left 10:46 am0c^ joined 11:05 hudnix joined 11:14 dakkar joined 11:17 noganex_ joined 11:21 noganex left 11:23 satyavvd joined, plobsing joined 11:53 Patterner left 11:55 Psyche^ joined 11:56 Psyche^ is now known as Patterner 11:57 satyavvd left 12:01 pmurias left
takadonet morning all 12:02
12:07 mingzhi joined
mingzhi echo "fff" 12:08
what is this??? 12:09
moritz not Perl 6.
12:10 noganex_ is now known as noganex
tadzik good day everyone 12:12
mingzhi Is there anyone know how to write a shell like in the unix? 12:14
moritz mingzhi: which language do you want to write it in?
mingzhi C or C++ with python 12:15
moritz then you're off topic here
but as a small hint: there are already many open source shells in C
look at their code for inspiration
mingzhi what one?
moritz gondor.apana.org.au/~herbert/dash/ 12:16
mingzhi thanks a lot 12:17
12:19 kanishka left 12:32 sftp joined 12:33 am0c joined 12:35 _twitch joined, Sharon joined, JimmyZ joined
Sharon is it legal to use a "powered by perl6" with the logo of perl6 on a perl6 powered website? 12:36
are there any restrictions?
JimmyZ you have site powerd by perl6 ? 12:37
moritz Sharon: Larry said that you can use the Perl 6 logo fairly liberally with anything connected to Perl 6
github.com/perl6/mu/raw/master/misc/camelia.txt
The Camelia image is a trademark of Larry Wall, and permission is 12:38
granted for non-exclusive use to label anything related to Perl 6,
provided the image is labeled as a trademark when used as a main logo
on a page.
Sharon I just plan to use it on the footer on the page just to advertise perl6
moritz since advertising perl6 is surely "related to Perl 6", that's OK :-) 12:39
Sharon ok great :)
12:40 plainhao joined 12:42 [Coke] left
tadzik Sharon: how are you using Perl 6 to power your site, out of curiosity? 12:44
12:46 mingzhi left 12:47 burning_aces joined 12:50 _twitch left 12:54 Sharon left 12:57 bacek joined 12:58 Holy_Cow joined 13:01 jaffa4 joined
jaffa4 hi 13:01
What is the state of Perl 6?
moritz it's in development, as you well know 13:02
jaffa4 you measure its state somehow 13:03
I guess
e.g. completion state
in %
moritz no
13:03 shi left
jaffa4 how many features left? 13:03
moritz since we don't know how the "final" (whatever that means) thing looks, we can't know how close we are
13:03 ingy joined
tadzik aleph 0 13:03
13:03 shi joined
jaffa4 How come you do not know? 13:04
moritz jaffa4: counting features would need some kind of quantization, which doesn't exist
JimmyZ jaffa4: use it before ask :)
moritz jaffa4: the spec is still changing, in reponse to implementation concerns and user feedback 13:05
13:05 shi left
jaffa4 Are there any serious users? 13:05
moritz is serious
jaffa4 with $$$ 13:06
13:06 geekosaur joined
moritz some of us earn dollars, yes 13:06
though I'm more of a EUR person
jaffa4 with Perl 6?
Is there a debugger for Perl 6 now? 13:07
13:07 silent_h_ joined 13:08 shi joined 13:09 leprevost joined
tadzik not that I know of 13:17
oh, there is one for yapsi
moritz does it actually work?
tadzik dunno
13:19 pmurias joined, gantrixx joined
mathw One of the stated purposes for yapsi is the development of a time-travelling debugger 13:24
13:26 woosley1 left 13:29 am0c^ left 13:31 shortcircuit is now known as mikemol
moritz that doesn't imply that it works now :-) 13:41
13:42 hanekomu joined 13:43 hanekomu left, hanekomu joined
pmurias sorear: ping 13:49
13:56 MindosCheng joined 14:04 jaldhar left 14:05 shi left 14:09 silent_h_ left, am0c^ joined 14:17 xinming_ joined
jaffa4 time travelling debugger? 14:17
14:19 xinming left
mathw jaffa4: yes, I believe the idea is that you can rewind the program state to any point in its execution history. 14:21
jaffa4 Interesting concept 14:22
could be useful
14:23 am0c^ left 14:26 justatheory joined, kaare_ left 14:27 hanekomu left 14:29 justatheory left 14:32 mikehh joined
sorear jaffa4: ? 14:33
phenny sorear: 09:36Z <pmurias> tell sorear can we drop log from the unit in nam?
14:35 kanishka joined
pmurias sorear: hi 14:38
sorear pmurias: currently log is what is used to handle run-time stash merging. There is no easy way to get rid of it
pmurias it's described as Mostly unused vestige of last stash system
sorear: how is it used? 14:39
sorear: does it make sense for be to write the loading for the current nam format for the hoop optimiser? or will it change soon? 14:40
sorear pmurias: each entry is converted into code which is added to the BOOT sub 14:41
pmurias: er, if there's no loader, how do you plan to use it?
pmurias i could wait for the new format 14:43
14:46 envi joined, birdwindupbird left 14:52 JimmyZ left
sorear new format design is a ways off.. weeks, at least 14:52
I think tonight I'll try to document the stash system
perl6: class Doubler { multi method double($x) { $x * 2 }; multi method double(Str $x) { $x ~ $x } }; say Doubler.double(35); say Doubler.double("abc") 14:53
p6eval rakudo 4bf132, niecza v4-26-g2dd3fec: OUTPUT«70␤abcabc␤»
..pugs: OUTPUT«3535␤abcabc␤»
moritz pugs really faked types 14:54
pmurias sorear: converting the nam into a typed haskell data structure might even help to specify the format more rigorously
sorear moritz: I'll say! 14:55
sorear out
14:55 kanishka left 14:57 am0c left 14:59 [sbp] joined 15:05 kst joined 15:12 am0c joined 15:14 wtw left 15:17 kst left 15:22 uvtc joined
uvtc Python recently had their Pycon conference, and many of the talks were videotaped. 15:25
I wasn't at the conference, and it was very cool to watch many of them.
Learning some new things as well as putting faces with names.
Would be great if perl6.org/community/ had some links to videos of Perl 6 talks.
TimToady +1 15:32
jaffa4 What is multi method? 15:33
TimToady having multiple methods of the same name that are differentiated by the types of their arguments to pick the best one to call 15:35
operator overloading is a specific instance of it 15:36
jaffa4 Is it like overloading functions in C++? 15:37
TimToady a little, except it's dispatched at run time, not decided in advance like in C++ 15:38
(well, we can decide some parts of it at compile time too, like what the candidate list is.)
short answer, just like C++, except correct :)
we also discourage people from overloading operators for unrelated purposes, unlike C++ 15:39
instead we encourage people to use an appropriate unicode operator
you won't see us using < for IO frinstnce 15:40
or + for concat
(though we did steal p5's . operator for method calls, so switched concat to ~) 15:41
15:42 frooh_ left
jaffa4 ~ from D? 15:43
moritz
.oO( we could have used -> for concat :-))
uvtc '~' looks looks stringy ... like you're weaving strings together 15:44
TimToady no, we just picked ~ because it was available, not too hard to type, works as a prefix, and looks like a bit of string :)
15:45 frew joined, frew left
TimToady is not really familiar with D 15:45
so think of it as convergent evolution :)
15:45 frew joined 15:46 ymasory_ joined 15:47 mj41 left 15:51 frew left 15:53 rdesfo joined 15:54 kaare_ joined 16:00 ggoebel left 16:01 justatheory joined 16:08 fhelmberger left 16:09 MayDaniel joined
uvtc What is the current purpose of the github perl6/mu project? Are bits of it gradually being teased out and put into their own dedicated github perl6/foo projects? 16:11
moritz uvtc: mu are those parts of the former pugs repo that don't have enough activity to warrant extracting things into separate repos (yet) 16:13
in fact the one file showing non-zero activity during the last months was dalek-conf.json :-) 16:15
uvtc moritz: ah, the irc bot 16:16
moritz uvtc: right
16:20 risou joined 16:22 uvtc left 16:24 burning_aces left 16:34 pmurias left, kanishka joined
tadzik how does one install Rakudo on windows these days? 16:36
not neceserilly Star, rather a New Rakudo
moritz just like on linux 16:37
you build it
tadzik with some specific compiler, msvc or something? 16:38
tadzik is completely green about windows
moritz msvc or cygwin or mingw, iirc
jnthn MSVC is most actively tested on, I expect 16:39
tadzik jnthn: with MSVC the makefile and everything Just Works? 16:40
jnthn tadzik: I build with ActiveState Perl and MSVC and it associated tool chain, and it all Just Works. 16:44
The makefile works with nmake (the MS make implementation)
tadzik nice, thanks
16:53 M_o_C joined, Chillance joined
tadzik wow, cygwin can even install Rakudo and Rakudo Star, shame it's 201007 16:54
benabik Ug, Cygwin. I used to use it a lot but decided a VirtualBox Debian or Ubuntu install was just easier. 16:57
tadzik I'm in a "Virtualbox Windows" situation :P 17:00
benabik Details, details. 17:01
17:03 rdesfo left 17:07 dakkar left 17:14 M_o_C left
tadzik IAL. My signicant other: „Rakudo, that's this «way of the something»?” 17:15
17:17 M_o_C joined
PerlJam no, it's the way of whatever ;) 17:18
tadzik ;)
17:19 Eevee left 17:20 Eevee joined 17:21 ab5tract joined 17:25 masak joined
masak greetings, #perl6 folks. 17:25
jnthn o/ maska
gah
masak
masak :)
moritz \o
jnthn
.oO( it was a typo, but I coulda claimed accidental accusative case... )
 17:26
masak j'accuse!
jnthn :P
tadzik hi maska! 17:28
masak :P 17:30
17:30 cdarroch joined, cdarroch left, cdarroch joined
masak in my mother tongue, sv:'maska' means en:'stitch'. 17:30
tadzik in my mother tongue it's "mask" 17:31
phenny: "maska"?
phenny tadzik: "Mask" (sl to en, translate.google.com)
masak tadzik: was just going to say that :)
tadzik: or 'hood', right?
tadzik not really, if that's the hood I think about. Oh, the car hood?
masak dunno.
tadzik Not the Robin Hood :)
masak I'm doing Google Translate research :) 17:32
many things that cover/conceal, it seems.
mask/hood/vizor/guise.
jnthn nom time, then back to nqp :) 17:34
tadzik I thought nqp is a prerequisite for nom :) 17:35
17:35 wallberg left
jnthn :P 17:35
Last night, it turned out fixing Parrot's register allocator was a prerequisite for nqp :/
Thankfully, I haz a viable patch.
masak wow, niecza has multi methods. 17:36
jnthn++ # viable patch
17:39 risou_ joined 17:43 risou left 17:44 nnunley joined 17:45 nnunley left, nnunley joined 17:46 nnunley left
masak backlog: yes, the debugger for Yapsi works. it's called Tardis. it travels in time. 17:56
that said, it's limited by what Yapsi can run, which is... limited. 17:57
jaffa4: hi! \o 17:59
jaffa4: when you ask about the degree of completion of Perl 6, what is it that you really have in mind? do you plan to start trying it out when it hits 100%? ;) 18:00
if I told you "Perl 6 is at 65%" without providing a source for that number, would that be sufficient? 18:01
I'm not trying to be insolent, just wondering what you're going to use the percentage for.
18:01 icwiener joined, mj41 joined 18:02 mikehh left 18:03 M_o_C left
jaffa4 masaK: hi 18:03
I have to think about the answer
masak 18:04
masak that's fine.
jaffa4 I have already tried it couple of times
by the way
masak I know you have. :)
but that only means that you should be able to estimate what's there already and what isn't yet. :)
PerlJam 85% of statistics are made-up on the spot though 18:05
jaffa4 there is another project
they make a C++ compiler
arnsholt masak: But which estimator should he use? Maximum Likehood, Maximum A Posteriori? I don't know anymore!
jaffa4 there I could see what features of C++ compiler are implemented
arnsholt (Says the man up to his ears in statistical gobbledy-gook)
jaffa4 when I asked that is what I ahd in my mind
masak arnsholt: crystall ball, preferably. 18:06
moritz already explained the difference about frozen spec
jaffa4 look here what I mean: clang.llvm.org/cxx_status.html
masak jaffa4: oh, ok. well, the closest thing we have is the spec (in its current state), and the amount we implement of that.
jaffa4: BUT (as moritz points out), the spec is still changing in response to the implementation digging further into Perl 6. 18:07
jaffa4 ok
arnsholt masak: Indeed. I wish I had a CRF crystal ball. Would save me enormous amounts of time waiting for numerical approximations to converge ^_^
masak jaffa4: and (perhaps more important), we *expect* it to do so.
jaffa4 you can still have a snapshot
masak jaffa4: we have lots of snapshots :)
it's called git :)
jaffa4 a snapshot of completion
masak jaffa4: point is, we're currently more interested in making a good spec than having a frozen spec. 18:08
at least that's true for some parts of the spec. some are already quite stable.
if you want a Perl 6 that doesn't change much at all, I recommend traveling about 30 years into the future.
jaffa4 that is not a problem if it is changing 18:09
I just would like to see what is its state now
masak if you're fine with it not changing: welcome! we're having fun using Perl 6 already, and you're free to join! :)
there are a few drawbacks:
* documentation is sometimes difficult and/or incomplete
* blog posts grow old quickly, and aren't always kept up-to-date 18:10
* sometimes the only way to learn something is to ask here
#perl6 is definitely the prime source of knowledge at this point. not some book or online tutorial.
that said, the Advent Calendar is very good, and probably very up-to-date. 18:11
moritz advantage: you can blog multiple times about the same topic
masak :P
one of the biggest advantages for me is that I get to be a (small) part of creating an awesome language.
jaffa4 What Advent Calendar?
masak perl6advent.wordpress.com/ 18:12
(enjoy!)
PerlJam boggles that someone is on #perl6 and missed the advent calendar
(but I guess that's normal because it is a seaonal thing after all)
jaffa4 good idea, the advent calendar 18:14
takadonet people blog about perl6? cool 18:15
:)
tadzik :)
I almost didn't catch it on time :)
18:16 envi left
jaffa4 has anyone made a compiler for parrot? 18:18
has anyone made a compiler for parrot bytecode?
masak yes. 18:19
jaffa4 last time I checked something like that
it did not work
I wonder anything has changed since then.
benabik Compiler for bytecode? Like bytecode to i386, or language to bytecode? 18:20
jaffa4 bytecode to assemly or machine language or to executable
benabik I don't think there's a program that does that currently. There's a JIT prototype, but nothing functional at the moment. It does allow you to make a small executable wrapper (ref: "perl6" binary in Rakudo) 18:21
tylercurtis Wait. Was there Perl 6 advent calendar last year? 18:22
tylercurtis somehow forget to read it, I think.
flussence_ yep
masak tylercurtis: happy reading! :D 18:23
tadzik (: 18:27
18:28 Eevee left
tylercurtis masak: thanks. I'm sure I will enjoy reading them eventually. 18:28
jaffa4 ok, bye all
masak bye, jaffa4 \o 18:30
18:34 MayDaniel left 18:42 nymacro left 18:58 fhelmberger joined, fhelmberger left 18:59 fhelmberger joined, awoodland joined 19:02 kanishka left 19:03 Eevee joined 19:04 kst joined 19:06 ymasory left, ymasory_ is now known as ymasory, risou_ left 19:18 kanishka joined 19:32 proller joined 19:35 justatheory left
dalek p/ctmo: ad4b814 | jonathan++ | src/ (3 files):
Make native types declared in the setting be lexically scoped.
 19:43
p/ctmo: 89440ca | jonathan++ | build/PARROT_REVISION:
Bump Parrot revision to get lexical register allocation fix.
19:43 fhelmberger left
dalek p/ctmo: 9a54b8e | jonathan++ | src/Regex.pir:
Remove debugging comment accidentally left in.
 19:45
sjohnson so, whats new in the perl 6 world
tadzik sjohnson \o/ 19:48
sjohnson yo tadzik
PerlJam is waiting for more Perl 6 proposals for GSoC
tadzik sjohnson: you don't have this long face unicodeart somewhere near, do you? :) 19:50
19:51 aesop left
sjohnson i do! 19:52
different varieties too
（　｀ー´）
fat face
（´ー｀　）
worried fat face
（　° ー°）
fat face after realizing there is porn on the internet
dalek rors: 7e7bf7c | moritz++ | notes.pod:
add notes on checking for specific excpetions
 19:53
masak sjohnson: I always thought it was the face of a pug :> 19:56
19:59 stkowski joined
sjohnson well, interpretation is in the eye of the beholder 20:00
mr wall thought it was a sleeping cat, which is cute
20:01 frew joined
masak frew! \o/ 20:02
frew ...?
PerlJam Stop saying "mr wall". It always takes me a second to figure out who you're talking about ;)
pyrimidine Pink Floyd's Mr. Wall
PerlJam wonders how the bricks got there though 20:03
Did he eat them?
20:03 BenPoole joined
TimToady is usually just known as "that guy cowering over in the corner" 20:03
20:03 BenPoole left
masak TimToady: because of the bricks? 20:04
pyrimidine is trying to think of a finish to the line "We don't need no ..." w/o starting a language war
20:04 kanishka left
PerlJam pyrimidine: we're fairly language-agnostic here ... I mean they're all just subsets of Perl 6 anyway ;) 20:05
20:05 mj41 left
PerlJam s/subset/slang/ 20:05
pyrimidine PerlJam: very true
20:08 plainhao left
sjohnson Larold Wallstonbury 20:10
20:10 aesop joined 20:12 y3llow_ joined, pothos_ joined 20:14 y3llow_ is now known as y3llow 20:15 pothos_ left, pothos joined 20:21 leprevost left 20:23 PerlJam left, PerlJam joined 20:26 [Coke] joined 20:30 MayDaniel joined 20:39 MayDaniel left 20:47 testin joined, Mowah joined
testin hola? 20:47
jnthn cerveza? 20:48
testin adios..may the force be with youQ
cerveza porfavor
20:48 testin left
jnthn Well, that's my Spanish practice for the year. :) 20:50
tadzik "a mi tambien", he said, laughing in the face of grammar rules 20:53
20:53 Holy_Cow left 20:59 pmurias joined 21:01 Khisanth left
pmurias masak: re frozen spec, even if we had a frozen spec, it wouldn't help estimating completion much 21:01
masak no, probably not. 21:02
freezing the spec, IMO, is a solution looking for a problem.
tadzik no need for freezing the spec w/o freezing the compiler behaviour 21:03
21:12 mj41 joined 21:14 Mowah left, justatheory joined
pmurias why are people interested so much when 100% feature completeness will happen? 21:18
tadzik they have a different attitude, I can understand that 21:19
masak easiest to think of this in terms of early adopters vs late adopters, perhaps. 21:23
asking for 100% feature completeness is a late-adopter question. 21:24
21:26 Guest51323 joined
masak rakudo: my @a = ""; @a = @a X~ "foo", "bar"; say @a.perl 21:28
p6eval rakudo 4bf132: OUTPUT«["foo", "bar"]␤»
masak rakudo: my @a = ""; @a X~= "foo", "bar"; say @a.perl
p6eval rakudo 4bf132: OUTPUT«[""]␤»
masak rakudo: my @a = ""; @a X~= ("foo", "bar"); say @a.perl 21:29
p6eval rakudo 4bf132: OUTPUT«[""]␤»
masak bug?
pmurias masak: but 100% feature completeness isn't important for production readiness, things like library support and performance are much more important 21:30
masak that's another thing. people want something, but they think that something is called "100% completeness". 21:31
pmurias according to google g++ isn't 100% complete
google meaing stuff found by a search not some offical statement from the company ;) 21:34
21:34 proller left
pmurias * meaning 21:34
21:35 Guest51323 is now known as Khisanth
Eevee I don't think "100% complete" means the same to you and them 21:37
I suspect a lot of people asking when it's "done" just mean "when can I do all the things you've promised"
masak that's a good point.
maybe never. we've promised so many crazy things. 21:38
jnthn We've delivered quite a few of them though :)
21:38 icwiener left
masak yeah. 21:38
jnthn Though with quite a few still go to :)
masak but looking good! 21:44
S09 up next, with luck.
I still believe S09 will be an important one for showing off Perl 6 from the good side.
aesop it'll be done when it's done - lwall 21:45
masak and that goes for any piece of software. regardless of what deadlines or rules mere mortals come up with. 21:48
Eevee on the bright side, everyone who's bothered that it's not "done" can only be bothered because they want to use it 8) 21:50
21:52 kaare_ left
dalek ecza: 9f14024 | pmurias++ | hoopl/n (2 files):
[hoopl] started of loading nam opcodes
 21:54
ecza: 7d6f610 | pmurias++ | hoopl/nam.hs:
[hoopl] loads all opcodes in "say 1"
21:55 eternaleye joined
masak and we should be saying "we're working on it as fast as we can. help us or step aside." ;) 22:08
22:09 ymasory left
masak today's cute little script: this piece of code generates all the possible strings that match /<[hm]> <[ea]> [nk|n|nn] [ing|ell|all]/: gist.github.com/908829 22:10
in a way, it could be seen as running the regex "in reverse", spitting out all possible matching strings rather than accepting a string and saying if it matched or not. 22:12
cognominal_ jnthn, trying to build the current ctmo branch with --gen-parrot and got : Checking out Parrot RELEASE_3_2_0-100-g7e9553d via git... error: pathspec 'RELEASE_3_2_0-100-g7e9553d' did not match any file(s) known to git. Command failed (status 256): git checkout RELEASE_3_2_0-100-g7e9553d
flussence_
.oO(quantum computer regex engine!)
masak with this strategy, we'd run into some depth-first problems with quantifiers... but most other things would work fine. 22:13
22:13 jaffa4 left
masak it should be possible to find a lazier formulation of lines 12..14. suggestions welcome. 22:14
22:16 Su-Shee left
masak rakudo: my @a = "foo", "bar"; my @b = "baz", "quux"; .say for [X~] "", @a, @b 22:16
p6eval rakudo 4bf132: ( no output )
masak rakudo: my @a = "foo", "bar"; my @b = "baz", "quux"; .say for [X~] "", [@a], [@b]
p6eval rakudo 4bf132: ( no output )
masak oh well. 22:17
'night, #perl6
22:17 masak left 22:23 proller_ joined 22:29 ab5tract left
TimToady niecza: my @a = ""; @a [X~]= ("foo", "bar"); say @a.perl 22:29
p6eval niecza v4-28-g7d6f610: OUTPUT«["foo", "bar"]␤» 22:30
TimToady \o/
22:37 thundergnat joined
thundergnat rakudo: our multi sub infix:«op»( $a, $b, :$foo, :$bar ) { .say for $a, $b, $foo, $bar }; infix:«op»( 45,3,:foo); 22:39
p6eval rakudo 4bf132: OUTPUT«Could not find sub &infix:«op»␤ in main program body at line 22:/tmp/rBdw0eLNPE␤»
thundergnat Arg.
rakudo: our multi sub infix:«op»( $a, $b, :$foo, :$bar ) { .say for $a, $b, $foo, $bar }; infix:<op>( 45,3,:foo);
p6eval rakudo 4bf132: OUTPUT«45␤3␤Bool::True␤Any()␤»
thundergnat sigh. Is there any other way to quote infix sub names? Rakudo apparently doesn't like guillemot quotes when searching for infix subs. 22:41
Makes it extremely difficult to use alternate calling conventions for infix:«+>»( ){ } among others... 22:43
rakudo: my $INTSIZE = (4294967297.Int == 1) ?? 32 !! 64; $INTSIZE.say; 22:45
p6eval rakudo 4bf132: OUTPUT«64␤» 22:46
22:46 dayangkun_ joined
thundergnat ^^ ugly hack to determine native int size. 22:46
22:52 awoodland left
flussence_ rakudo: say max (1..128).grep({ ('0b1'~'0' x $^i - 2).Int.sign == 1 }) 22:53
p6eval rakudo 4bf132: OUTPUT«64␤»
22:55 am0c left, Chillance left 22:56 mj41 left 23:03 whiteknight joined
thundergnat Anyway, I wrote the logic to implement :rotate and :unsigned for bit shift operators. It works correctly under 32 and 64 bit linux and Windows. Gist here: gist.github.com/908965 23:05
23:08 am0c joined 23:10 cdarroch left 23:24 pmurias left 23:37 thundergnat left 23:38 zer0o joined 23:48 cbk 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!