Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2009-12-30

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:00 eiro wow ... 'll try to find it ... somewhere in the house
00:00 eiro thanks a lot
00:00 dryden nice
00:00 eiro 'night all
00:00 dryden good night erio
00:01 dryden if i might ask sjohnson, where do you work?
00:02 sjohnson small software company in BC Canada
00:02 arnsholt eiro: In my copy of the Camel book, page 197 "The Little Engine That /Could(n't)?/"
00:02 sjohnson i try to use perl as often as i can for work tasks
00:02 arnsholt phenny: tell eiro In my copy of the Camel book, page 197 "The Little Engine That /Could(n't)?/" for backtracking
00:02 phenny arnsholt: I'll pass that on when eiro is around.
00:04 dryden that is sweet, i'm glad you're working in the field you seem to enjoy. us programming enthusiasts are lucky in the sense that our hobby can also be our profession...not everyone has that luxury.
00:05 sjohnson well my job is alright, not a dream job really as I mostly do PHP and javascript which i hate
00:05 sjohnson but better than doing manual labour as i used to do
00:06 dryden i suppose it is better than manual labor, which is sort of what my parents do.
00:07 arnsholt Speaking of JS, I'm a bit ambivalent. It's a cool language, but there are some icky parts
00:07 arnsholt Like the optional ;
00:07 dryden I get the feeling that there isn't a lot of perl in the so called "industry", which sort of makes me sad. My "dream job" would be to work for Google, and sadly they heavily use Python.
00:11 dryden Gah...Python.
00:12 araujo http://i.imgur.com/1gF1j.jpg
00:13 dryden haha
00:14 sjohnson heh great photo
00:14 arnsholt dryden: I think the important thing to remember WRT Python is that there's a difference in philosophy
00:15 sjohnson darn
00:15 sjohnson was hoping perl was on this chart
00:16 arnsholt I especially liked the Lisp fan row. Describes my Lisp fan friend quite well =D
00:16 dryden yeah me too. and yeah..i arnsholt, i understand they have a different philosophy and i suppose it works fairly well, i'm still not a fan.
00:17 sjohnson i looked at Python code, and didn't really care for it too much
00:17 sjohnson perl makes me feel like i can program with both sides of the brain
00:17 * colomon can't see the graphic because most of his screen is currently dedicated to fiddle videos for his little boy...  :)
00:17 sjohnson kind of like a lightning storm of creativity in my brain.  that's how i feel anyway
00:17 arnsholt The indenting thing drives me spare for some reason. But the main problem is that it's not as easy to do list stuff with Python
00:18 arnsholt Things like map and grep are a lot more accessible in Perl
00:18 arnsholt sjohnson: Very nice description. Fits me rather well too, I fell
00:18 dryden i really dislike their standard api, it's very ridiculous.
00:18 sjohnson i would hate to have to import re in python everytime i got fancy with teh r3gexz
00:19 sjohnson i do like their standard libs though, kinda cool that you dont have to rely on CPAN for a number of some common tasks
00:19 arnsholt Yeah. And it's possible to use it without going completely insane
00:19 dryden there's no string.length() in python..why!? there's a static len() function, that's just silly.
00:19 arnsholt For example I had to interface with a C lib a bit back, which only had Python and Ruby bindings
00:20 arnsholt Easier to use Python than implement the Perl bindings (even though I'm considering of doing the Perl bindings no that I'm done with the project =)
00:20 arnsholt dryden: Yeah, there are some wrinkles in the language
00:21 arnsholt But then again, Perl has a couple of warts too =)
00:21 colomon dryden: we realized the same thing with perl 6's eval unction last week (no method version, which felt inconsistent) and within about thirty minutes we'd added it to Rakudo.  :)
00:21 sjohnson araujo: im going to redo that to include Perl heh
00:21 colomon s/untion/function/
00:21 araujo sjohnson, hehe
00:29 payload joined #perl6
00:36 dryden i'm sorry i was away, my girlfriend called. but that's awesome. you guys quick to eliminate minor language inconsistencies.
00:36 dryden are* quick
00:38 dryden it's been awfully fun, you guys are awesome. I should get going. bye everyone.
00:38 dryden left #perl6
00:38 colomon dryden: actually, the fun thing about it was it was just a couple of us normal guys here, not the high powered language designers or implementors.
00:40 colomon Perl 6 has a very relaxed approach to making changes.  :)
00:41 colomon gotta run bathe my boy.  back later.
00:41 sjohnson cya
00:50 * arnsholt smacks forehead
00:50 arnsholt I was wondering what the attribute $last was for in the Match object in nqp-rx
00:51 arnsholt It wasn't $last. It was $!ast, which makes a lot more sense
00:54 ihrd1 joined #perl6
00:54 ihrd1 left #perl6
01:07 justatheory joined #perl6
01:11 Baggio_ joined #perl6
01:28 eiro arnsholt, the "perl6 book" (page 30) explains :, :ratchet and token nicely !
01:28 phenny eiro: 00:02Z <arnsholt> tell eiro In my copy of the Camel book, page 197 "The Little Engine That /Could(n't)?/" for backtracking
01:29 eiro (i couldn't sleep without knowing)
01:29 arnsholt Know the feeling =)
01:29 arnsholt (And I really should get around to reading the Perl 6 book)
01:30 eiro this is a very good book (i just started)
01:35 eiro so now i can sleep :-) cya all
01:37 alester joined #perl6
01:40 agentzh joined #perl6
01:48 zykes- joined #perl6
02:16 sjohnson araujo: OT - do you know much about 4chan pics?
02:16 sjohnson having a real tough time finding the last piece for the language chart Perl puzzle
02:17 xinming_ joined #perl6
02:17 sjohnson its the pic with the guy leaning back in his chair with the white hair and moustache
02:26 zykes- when's perl6 gonan be released ?
02:27 rjbs You can download it right now.
02:27 zykes- yeh, but what modules are available ?
02:27 zykes- how "ready" is it contra perl5
02:27 rjbs I wouldn't use it for performing my job.
02:28 zykes- i don't get the whole, pugs, rakudo thing ?
02:28 rjbs Same as gcc, icc, scc, etc.
02:28 rjbs Different implementations have different strengths.  Also, the developers of each implementation tend to work on different parts of the langauge, which strengthens the design.
02:29 rjbs Sprixel, for example, is written in JavaScript, which means you may end up being able to run Perl 6 in the browser.
02:29 zykes- how many different compilers are ther ?
02:29 zykes- for perl6 i mean
02:30 rjbs http://perl6.org/compilers/ <-- might not be complete
02:30 diakopter rjbs: 'cept sprixel is now sprixel#, in C#... but it may get ported back to javascript too
02:30 rjbs diakopter: time to tell soembody to update /compilers :)
02:30 diakopter meh.
02:30 diakopter no need for hype on vapor/darkware
02:30 rjbs nod
02:31 JimmyZ joined #perl6
02:31 zykes- so to run perl6 from the commandline
02:31 * diakopter <-- king of throwaway code and rewriting
02:31 zykes- rakudo is the correct thing ?
02:31 diakopter zykes-: sure, for now
02:31 rjbs yeah, I think it's almost certainly the most complete implementation atm.
02:32 rjbs (and it's sort of the most mainline one, for various historical and other reasons)
02:32 zykes- heh, when's perl6 gonna me "mature" or done ?
02:32 diakopter could be never
02:32 diakopter :)
02:32 diakopter but rakudo* in April is expected to be stable-ish
02:32 zykes- heh, doesn't seem to be any use in writing perl6 code atm then
02:33 rjbs Why not?
02:33 rjbs It's fun.
02:33 diakopter and a decent subset of the Synopses
02:33 rjbs It's educational.
02:33 zykes- well is there modules avaiable for it now then ?
02:33 diakopter zykes-: it's a great language in which to learn more programming techniques
02:33 rjbs zykes-: Sure, but "modules" isn't a very useful measure.
02:33 zykes- i'm kind of addicted to POE atm
02:33 zykes- how come rjbs ?
02:34 rjbs because what if there were 912031238 modules and each one was lame and useless?
02:34 rjbs There is no database module.  There is no event-driven programming library as far as I know.  There is no crypto.
02:34 rjbs (I think.)
02:34 rjbs but there are HTTP and HTML libraries
02:34 rjbs and other stuff
02:35 zykes- where can one see the "available" modules?
02:35 diakopter well, pugs has all of those, since pugs is linked to libperl (5)
02:35 zykes- pugs is yet another perl6 compiler ?
02:35 diakopter the first major/large implementation
02:35 diakopter written largely in Haskell
02:36 rjbs zykes-: Yes, but it's mostly dormant and outdated.
02:36 rjbs but it was very, very influential
02:37 diakopter imho the "use v5; { }; use v6" capability is extremely important, but many around here disagree with me on that.
02:37 zykes- well i kind of like the modules, so i guess i need to wait then until stuff get's written.
02:37 rjbs Sure, suit yourself!  We'll still be here.
02:37 diakopter :)
02:38 rjbs Perl 6's core functionality is much greater than Perl 5's, so learning the core language can help you learn a lot of interesting programming concepts, fwiw.
02:38 zykes- well i guess one needs to write alot more to get the same stuff if there are no modules available
02:38 rjbs ...btu as I said, I wouldn't use Perl6 to get my work job done.
02:38 rjbs People aren't really using Perl 6 to implement enterprise web applications, so they're not doing the same stuff.
02:39 rjbs You can't look at Perl 6, as it stands now, as competing with Perl 5.
02:39 zykes- what is perl6 getting used for then ?
02:39 zykes- toying ? :p
02:39 rjbs Perl 6 is being designed and constructed.
02:39 rjbs Software implemented with it is often very useful, but it is, as far as I see things, a proving ground, not a factory.
02:39 JimmyZ use v6; { use v5; ... }  ...
02:40 rjbs People are saying, "Can I do X with Perl 6?" and if they can't, they figure out whether it's a problem beyond not having the right core written.
02:40 diakopter JimmyZ: exactly :)
02:40 rjbs zykes-: Did you read the Perl 6 Advent Calendar?
02:40 zykes- wasn't perl6 gonna have some event library ?
02:40 diakopter std: use v6; { use v5; ... };  ...;
02:40 p6eval std 29401: ok 00:01 104m␤
02:40 rjbs I'm sure Perl 6 will have many even tlibraries.
02:41 rjbs I don't know if a standard, core one was discussed.
02:41 rjbs It woudl probably fall under concurrency, which isn't quite specified yet.
02:41 diakopter TimToady's been toeing around the concurrency spec for a coupla years
02:44 diakopter rjbs: re "Perl 6's core functionality is much greater than Perl 5's", well, yes, especially since it by definition subsumes all of Perl 5's core functionality, since Perl 6 must *be* Perl 5 as well.
02:45 rjbs diakopter: Yeah, we'll see how that turns out... ;)
02:45 zykes- the decade that happens :p
02:56 diakopter rjbs: pugs did it... :(
02:56 diakopter also, pmurias did it for [some past version of] mildew
02:58 sjohnson araujo: http://imgur.com/LPPaL.jpg <-- now with perl
02:59 araujo sjohnson, hahaha
03:00 araujo hehe
03:01 Baggio__ joined #perl6
03:04 araujo man, I can't stop laughing watching Java .. as seen by... Haskell
03:04 go|dfish Ruby as seen by Perl
03:06 JimmyZ araujo: ni hao a
03:07 araujo JimmyZ, ni hao
03:08 JimmyZ ni zai nali a
03:09 diakopter sjohnson: wow.
03:11 diakopter JavaScript is definitely in the Perl family, imo.
03:26 Exodist joined #perl6
03:37 Transformer joined #perl6
03:40 Transformer left #perl6
03:57 _jaldhar joined #perl6
04:04 Baggio_ joined #perl6
04:25 meppl joined #perl6
04:57 meppel joined #perl6
05:56 Baggio_ joined #perl6
06:01 mberends sjohnson++ and others++ # http://irclog.perlgeek.de/​perl6/2009-12-30#i_1876551
06:13 Baggio_ joined #perl6
06:25 msmatsko joined #perl6
06:26 colomon joined #perl6
06:29 cls_bsd joined #perl6
06:39 nookie joined #perl6
07:02 ppperlerrrr joined #perl6
07:14 racheldaBROWN joined #perl6
07:16 Baggio_ joined #perl6
07:21 pugs_svn r29419 | mberends++ | [vill] interim commit of src/graph_traverse, the spider that will do the low level work in test/yaml_compose_roundtrip (a lite YAML composer+serializer)
07:25 kaare joined #perl6
07:52 kaare joined #perl6
08:04 eternaleye_ joined #perl6
08:19 ejs joined #perl6
08:20 Su-Shee joined #perl6
08:20 Su-Shee good morning
08:21 diakopter good night
08:21 eiro joined #perl6
08:21 iblechbot joined #perl6
08:24 synth joined #perl6
09:17 ejs joined #perl6
09:18 Chillance joined #perl6
09:18 ejs1 joined #perl6
09:29 ejs2 joined #perl6
09:34 Baggio__ joined #perl6
09:39 sjohnson diakopter: JS thing is like Perl sure, but not Java so much
09:40 Ash_ joined #perl6
09:42 mberends sjohnson: that is also why V8 is a pretty decent runtime engine for Perl 6 implementations
09:42 sjohnson hopefully didnt offend anyone by the Perl x and y lines
09:42 sjohnson those were the best things i could think while at work :)
09:46 Baggio___ joined #perl6
09:48 mberends I didn't understand the "Perl as seen by Ruby fans" image
09:51 sjohnson just tried to make a joke about ruby taking the good parts of Perl 5
09:51 sjohnson in a thief-like context
09:52 mberends ah, a sort of white-collar PHP
09:52 sjohnson well, "stealing" the great regex engine was kinda what i was imagining
09:54 sjohnson one of my friends changed the jesus img in the perl / perl thing to http://dl.dropbox.com/u/347510/Untitled-1.jpg
09:54 sjohnson which i also think works quite well
09:54 sjohnson as far as 2000+ uses goes
09:55 mberends less potentially offensive, fersure
09:58 sjohnson i feel both ways about Perl that's for sure :)
09:58 sjohnson anyways, bed time.. cya
10:01 meneldor joined #perl6
10:01 meneldor morning guys
10:02 mberends hi meneldor
10:08 Baggio_ joined #perl6
10:10 cognominal joined #perl6
10:24 kangu joined #perl6
10:41 Student joined #perl6
10:47 colomon_ joined #perl6
11:16 payload joined #perl6
11:36 pmurias joined #perl6
11:39 wolverian there's no builtin subrule for delimited lists, right?
11:40 pmurias mberends: seen code.google.com/p/v8/issues/detail?id=413?
11:44 Tene wolverian: you shoul dbe able to do like: /<digit> ** ','/, but it seems to be NYI
11:44 Tene oh, looks like it works in ng
11:45 Tene ng: say "foo bar 1,2,3 baz" ~~ /<digit>**","/
11:45 p6eval ng 4e4d82: 1,2,3␤
11:45 wolverian oh, **. I looked for subrules only.
11:45 wolverian thanks!
11:45 Tene np
11:48 Baggio_ joined #perl6
11:49 cognominal joined #perl6
11:49 pmurias diakopter: i don't think mildew ever had use v5
11:50 pmurias diakopter: mildew has use SDL:from<perl5> tho
11:52 mberends pmurias: thanks for the suggestion, I'll try the 27-Dec patch at the bottom
11:57 pmurias mberends: did it help?
11:58 mberends not at first attempt, now checking SConstruct manually
12:02 _Jedai_ joined #perl6
12:04 mberends pmurias: the patch works only after also doing export GCC_VERSION=44
12:05 hejki ng: say "." ~~ :s
12:05 p6eval ng 4e4d82: 0␤
12:05 hejki rakudo: say "." ~~ :s
12:05 pmurias does export GCC_VERSION=44 work without the patch?
12:05 p6eval rakudo db84bc: 4096␤
12:06 hejki rakudo: say qx{ls -l|tail -1}
12:06 p6eval rakudo db84bc: operation not permitted in safe mode␤in Main (file /home/p6eval//p2/lib/parrot/1.9.0-​devel/languages/perl6/lib/Safe.pm, line 24)␤
12:06 hejki :)
12:06 hejki i already thought it had something very severe stuff going on :P
12:07 hejki how about unlink("."); ? :)
12:07 hejki (cares not to try)
12:07 mberends pmurias: I'll first complete the build and install the generated v8 files, then try re-building without the patch. still building..
12:08 hejki oh.. use Safe; is actual mode :O
12:08 hejki nice :)
12:11 cognominal joined #perl6
12:16 dug joined #perl6
12:19 mberends pmurias, ok, the v8 files are installed as per JavaScript::V8 perldoc. Now running in cpan, 'make JavaScript::V8' gives "Cannot open 'xspp --typemap=typemap.xsp JavaScript-V8-Context.xsp |': No such file or directory in WithV8Context.xs, line 21"
12:20 douglashunter joined #perl6
12:22 MaCkeR joined #perl6
12:22 MaCkeR left #perl6
12:25 mberends pmurias: "Warning: prerequisite ExtUtils::XSpp 0.01 not found" that'll be it :)
12:29 Baggio_ joined #perl6
12:31 masak joined #perl6
12:32 masak o/
12:32 mberends \o
12:32 pmurias \o/
12:33 masak :)
12:34 mberends pmurias: \o/ JavaScript::V8 installed doing a manual make, but not from sudo cpan
12:34 hejki hmm.. can i overload datatypes? they're classes after all, right?
12:35 pmurias mberends: how did it fail when installing from sudo cpan?
12:36 ruoso joined #perl6
12:36 hejki ohai masak
12:36 masak o hei hejki
12:37 hejki yet another beautiful day to explore perl6? :)
12:37 masak naturally.
12:37 mberends pmurias: "Could not make: Unknown error" :(
12:37 masak hejki: actually, I did some beautiful exploration last night. I didn't run into one single bug, but I had a lot of fun with OO.
12:37 pmurias that's bad :(
12:38 hejki masak: nice.. i had severe brainfarts last night and basically just wasted my time :P
12:38 hejki masak: thought i found a bug, but it actually was just me doing things oh-so-wrong :P
12:38 masak that's alright too.
12:38 hejki sure
12:38 pmurias mberends: it failed after you installed ExtUtils::Xspp?
12:38 hejki at least i did something :)
12:38 masak hejki: been there, sone that.
12:38 masak s/sone/done/
12:38 hejki heh y
12:39 hejki btw.. how much ng differs from rakudo?
12:39 masak lots.
12:39 hejki hmm..
12:39 hejki ng: say "." ~~ :s
12:39 hejki .<
12:39 p6eval ng 4e4d82: 0␤
12:40 masak it's, like, the biggest branch, ever.
12:40 hejki if it uses even remotely same kind of definitions for those operators my patches could be directly applied
12:40 mberends pmurias: it failed before that. actually, now in 'look JavaScript::V8', it *does* configure and make correctly. There must have been a previous incomplete build getting in the way.
12:40 masak hejki: I'll leave it to you to find out. :)
12:40 hejki heh :P
12:40 hejki so i thought :P
12:41 hejki how's ng with current parrot btw (i won't test it nao, i'm at work :P)
12:44 tsv joined #perl6
12:45 mberends pmurias: \o/ the JavaScript::V8 synopsis code runs ok :)
12:46 pmurias make test works?
12:46 mberends yes it also worked
12:46 mberends 99 bottles of beer is quite a lot of output to check ;)
12:48 * pmurias realised he never run the synopsis code ;)
12:49 hejki masak: i blame you for my slacking @ work
12:49 * hejki is currently cloning and skimming ng branch :P
12:49 masak hejki: sure, if that makes you feel better... :)
12:49 hejki heh..
12:50 hejki doing free-time stuff on paid working hours makes me feel better :)
12:51 cognominal joined #perl6
12:51 hejki mm.. Mu() is the new Failure()?
12:51 huf there's Mu()? fuck yeah.
12:51 hejki ng: say .WHAT
12:51 p6eval ng 4e4d82: Mu()␤
12:51 pmurias Mu() is the new undef is the new Object
12:52 colomon_ hejki: Mu is not Failure, but it is used for "undef" failures.
12:53 hejki i should buy some of the neat perl6 books tbh
12:53 masak hejki: one of them will publish around April.
12:54 masak (the one we're building on #perl6book)
12:54 mberends left #perl6
12:54 hejki oreilly?
12:54 * hejki prefers pdf
12:55 mberends joined #perl6
12:55 hejki is there a draft somewhere? :)
12:56 colomon_ http://github.com/perl6/book
12:57 masak there's a PDF under Downloads...
12:57 masak ...but it omits some of the newer chapters, which can still be read in source form.
13:02 mberends pmurias: perl mildew -Bjs -e'say "V8 V8 ";' also works :-)
13:04 hejki nice
13:04 hejki masak: none of the ~~ :{d,f,l,s} is in ng yet.. where should i put these? (clearly some of the PIR part :))
13:05 colomon_ hejki: where did you put them in rakudo master?
13:05 hejki builtins/any-str.pir
13:05 hejki so Any.pir or Str.pir? :>
13:05 hejki (i'd use Str.pir)
13:05 colomon_ sounds reasonable to me.
13:06 hejki or hmm.. by definition should i be able to: 3 ~~ :d ?
13:06 hejki i mean.. i can have a file named 3
13:06 hejki but is it interpreted (in human terms) as a string or any? :)
13:06 colomon_ well, what does the :d code do?
13:06 hejki -d
13:06 hejki checks if it's a directory
13:06 colomon_ no, I mean what does it take as an argument?
13:06 hejki hmm.. a string :)
13:07 hejki so Str.pir it is
13:07 colomon_ :)
13:07 hejki think i'll add the :d and :f as well from master now when i'm on it :P
13:07 hejki (and might even actually use the patch guideline instead of pastebinning stuff for masak - unless this is more favorable ;))
13:10 masak hejki: I'd go with sending the patches by me. quicker, more efficient. :)
13:10 masak unless I'm away for some reason.
13:10 hejki well you're most often around when i'm around
13:11 masak seems like it, yes.
13:11 hejki which timezone?
13:11 y3llow joined #perl6
13:11 colomon_ unobe just forked rakudo on github and sends pull requests there when he has a patch.  that seems to work well.
13:12 hejki well i'll pick on masak until i might be subject for push permissions ;)
13:12 takadonet joined #perl6
13:12 takadonet morning all
13:12 masak takadonet: \o
13:12 colomon_ o/
13:13 takadonet how are you guys doing today?
13:14 masak pret-ty good. :)
13:15 mberends happy to have kick-started a V8 :)
13:15 masak takadonet: whoz op with you?
13:15 takadonet masak: no idea hehe why?
13:15 Student joined #perl6
13:16 masak takadonet: sorry, I meant 'how are you today?'
13:16 takadonet masak: good good.
13:17 pmurias mberends: make test-js in mildew works?
13:20 abra joined #perl6
13:23 mberends pmurias: some subtests not run, looking like viv errors, ok up to while.mildew..
13:25 pmurias nopaste?
13:26 mberends waiting until all tests done..
13:40 mberends pmurias++: Files=56, Tests=213, 1005 wallclock secs # slow but impressive :)
13:40 lisppaste3 mberends32 pasted "make test-js output for pmurias++" at http://paste.lisp.org/display/92811
13:42 pmurias joined #perl6
13:43 mberends .oO( knocked the laptop off the table in excitement? )
13:43 pmurias mberends: the p5 integration tests failed
13:45 pmurias mberends: it seems a bug of some sort that they got run by the test harness
13:45 mberends ok, not very serious
13:46 pmurias mberends: you are testing on EEE PC?
13:46 mberends yes, Atom N270 1.6GHz processor
13:50 pmurias EEE is your main development box?
13:53 orafu joined #perl6
13:54 mberends pmurias, almost  50% of the time, especially when commuting etc. Today there are other machines available, but the portability is useful around home as well. Not many tasks require a lot of brute force, and the keyboard and 10" LCD size are quite comfortable after getting used to them
13:56 Baggio_ joined #perl6
13:56 mberends if the Motorola Droid could run Rakudo, I'd probably shrink into that ;)
14:00 masak what's that, a vacuum cleaner?
14:00 takadonet hehe
14:02 mberends oh, Motorola Milestone in the UK http://en.wikipedia.org/wiki/Motorola_Droid
14:04 Alias joined #perl6
14:05 bluescreen joined #perl6
14:07 avar '/w 13
14:09 * masak hugs avar, nice irssi user
14:09 mberends .oO( a vacuum cleaner running Rakudo :)
14:09 masak man, that would suck :P
14:09 mberends lol lol
14:09 mberends rotfl
14:10 masak god thing you've vacuumed the floor.
14:10 pmurias joined #perl6
14:10 masak s/god/good/
14:10 masak (thanks, I'll be here all week)
14:10 mberends thank good for that
14:11 masak divinity-based Freudian slips are uncanny.
14:12 masak fortunately, they don't happen allah that often.
14:12 mdxi o/` what if god was one of us? / vacuum ro-bot like one of us? o/` # with apologies to everyone
14:13 masak :)
14:29 colomon joined #perl6
14:31 takadonet colomon: welcome back
14:35 colomon thanks
14:37 payload joined #perl6
14:52 rgrau_ joined #perl6
14:52 * masak catches himself writing `a ? b ! c` in a Perl 5 program
14:54 mberends over here, that would result in a spec addition and implementation within 30 minutes ;)
14:55 mberends agreed by a quick show of hands
14:55 masak well, I wouldn't mind if someone told the Perl 5 parser to detect the above half-Perl6ism...
14:55 mberends quite
14:56 mberends there'd be no breakage of existing code
14:56 masak oh, I'm not out to change the way things are done in Perl 5.
14:56 masak but it would be an excellent place to detect unwarranted futurism and report intelligently on it.
14:56 mberends heh, that's why they put us in this padded cell
14:57 masak :)
14:57 Baggio__ joined #perl6
14:57 Su-Shee left #perl6
14:57 takadonet how can you determine if an object has a method before calling it? i.e $object.can('drink_beer')?
14:57 masak takadonet: there's $object.?drink_beer if that's enough for your purposes.
14:58 masak otherwise what you wrote should work, with the addition of a '^' before 'can'.
14:58 takadonet thanks guys. '^' with can will work
14:59 takadonet hmmm Test.pm in rakudo main directory does not have a 'can_ok' test function
15:00 masak patches welcome.
15:03 PacoLinux joined #perl6
15:06 pdcawley joined #perl6
15:08 takadonet nuts....
15:11 masak almonds....
15:14 patspam joined #perl6
15:40 Baggio_ joined #perl6
15:44 bluescreen joined #perl6
15:48 ash_ joined #perl6
15:55 Psyche^ joined #perl6
15:55 fridim_ joined #perl6
15:56 rgrau_ joined #perl6
16:02 masak std: sub foo(?$) {}
16:02 p6eval std 29401: [31m===[0mSORRY![31m===[0m␤Malformed parameter at /tmp/ZpqnCTjB6P line 1:␤------> [32msub foo([33m⏏[31m?$) {}[0m␤    expecting any of:␤      name␤   parameter␤  signature␤      whitespace␤FAILED 00:01 107m␤
16:02 masak hm, are optional anonymous parameters not allowed?
16:05 arnsholt std: sub foo($param?) {}
16:05 p6eval std 29401: ok 00:01 106m␤
16:05 masak oh :/
16:05 masak std: sub foo($?) {}
16:05 p6eval std 29401: Potential difficulties:␤  Illegal to use ? twigil in signature at /tmp/GNWAks0cJM line 1:␤------> [32msub foo([33m⏏[31m$?) {}[0m␤ok 00:01 106m␤
16:06 arnsholt It'd need a name too, no?
16:06 masak arnsholt: thanks. I was momentarily confused.
16:06 masak arnsholt: I'm actually in a situation where giving it no name makes sense.
16:08 arnsholt What kind of situation is that? Now I'm curious =)
16:09 * masak considers where to start
16:10 masak I'm doing a diagram renderer.
16:10 masak I have different objects: so far 'image', 'text', 'box', and 'arrow'.
16:10 synth joined #perl6
16:11 masak they are defined by structural recursion, so a box can contain any object among the above ones.
16:11 masak an arrow can have texts (or anything else) above or below itself.
16:12 masak to make the rendering work, I make cute little methods that call each other.
16:12 masak for example, for a box to find out how big it is, there are a width-of and a height-of which take a Box as its first argument.
16:13 masak (a bit like methods, but these are multi subs)
16:13 arnsholt Right
16:13 masak the width-of for a Box calls the width-of of the thing it contains, and returns that with a bit of added padding.
16:14 masak now,
16:14 arnsholt And for some types the return value is the same for all objects of the class?
16:14 masak I'm adding a new type of multi called allocate-region, so I only have to calculate all the sizes and positions once, and not all the time through the traversal of the raster image.
16:15 masak this new multi takes relative $x and $y coordinates, so that parent elements can send in offsets to the allocation of their child elements.
16:15 masak these are optional, because they have default values.
16:15 masak so that's where the optional part comes from.
16:16 colomon joined #perl6
16:16 masak for the anonymous part: I have a top-level class called Object from which all the image element types inherit.
16:16 masak and I make one multi of each type for Object, just to catch calls to multis I haven't written yet.
16:17 masak all these do is die with a nice error message.
16:17 masak so I never use the values of the parameters, and I might as well leave them as anonymous, just giving the sigil.
16:17 masak there you go. :)
16:17 * masak exhales
16:18 pmurias joined #perl6
16:18 masak I can't wait to show you the images that this tool spits out. :)
16:18 arnsholt That makes sense. Thanks
16:18 KyleHa joined #perl6
16:18 vi joined #perl6
16:18 vi ?
16:19 KyleHa !
16:19 masak ¿
16:20 vi where browse if i want to download perl ?
16:20 masak vi: Perl 5 or Perl 6?
16:20 vi perl 6
16:20 masak vi: perl6.org
16:21 masak vi: also, check out the Perl 6 Advent Calendar if you haven't already.
16:21 KyleHa I was looking at the P6 Advent Calendar yesterday, and I was glad I did.
16:21 vi i am newbie, i want to learn about perl....
16:22 perlygatekeeper joined #perl6
16:22 masak vi: then the Advent Calendar is for you! http://perl6advent.wordpress.com/
16:23 masak wow, there's a spike going on in the visitor statistics! we're up to above 1k again.
16:23 masak seems a Czech site gives us a lot of hits right now: http://www.abclinuxu.cz/zpravick​y/prvni-perl-6-adventni-kalendar
16:24 vi advent calendar for what ?
16:25 masak vi: each day during December, one of us wrote about some Perl 6 feature.
16:25 masak vi: but it's easier to just show you than to explain: http://perl6advent.wordpress.com/
16:30 vi if you want to use perl, I just need to download perl at perl.org? or other....
16:31 KyleHa vi: If you know nothing about Perl, you might want to start by learning the difference between Perl 5 and Perl 6 to decide how to proceed.
16:31 vi if I wanted to use perl, I just need to download perl at perl.org? or other...
16:32 masak vi: what KyleHa said. in this channel, which is called #perl6, we're working on Perl 6, a not-yet-feature-complete language in the Perl family.
16:32 masak vi: there's also the much more established Perl 5, which you will indeed find at perl.org
16:42 vi thank u
16:42 masak u are welcome.
16:42 masak 你是中人吗?
16:43 masak (just a guess)
16:43 masak or maybe Czech?
16:44 mberends masak++ # polyglot welcomebot
16:44 masak vi: nie mamy czas?
16:44 orafu joined #perl6
16:45 vi i not understand
16:45 masak vi: I was just curious as to where you're from.
16:45 masak but then we've ruled out China and the Czech republic, I guess.
16:46 vi indonesian
16:46 masak oh, cool.
16:47 masak not sure I have the fonts for that, though.
16:47 masak and definitely not the polyglottery. :P
16:47 masak vi: why did you get interested in Perl?
16:48 vi for my thesis
16:48 masak wow
16:49 vi but i cannot to use perl
16:49 vi i must learn about perl
16:50 sbp there's always http://en.wikipedia.org/wiki/Perl of course
16:50 masak vi: Perl is big. is there something in particular you'd like to learn?
16:51 xinming vi: Perl if fun to learn. :-)
16:51 sbp perl6 is basically perl, but with s/camel/butterfly/
16:51 frettled Or s/camel/camelia/
16:51 frettled ia = intelligent architecture ;)
16:51 KyleHa I think there's a little bigger difference than the mascot.  8-)
16:52 sbp well, the mascot is symbolic of all the rest though
16:52 frettled Beware intelligence, it can bite you in the a...
16:52 frettled sbp: psychedelic, psychotic, brain-eating monster?
16:53 vi for tagging the text....
16:53 masak vi: so, you have some text coming in and you want to do something with it?
16:54 vi ya
16:54 vi can u speak indonesian ?
16:54 sbp there's also a #perl note
16:54 masak vi: sorry, not yet.
16:55 masak vi: Saya tidak bisa berbahasa Indonesia. :/
16:56 vi :) oooo
16:56 vi I just need to download perl 6 and it can be used directly
16:57 masak with caveats, yes. :)
16:57 vi my english is not fluently
16:57 meteorjay joined #perl6
16:57 masak if you're new to Perl, then probably you should try Perl 5 first, not Perl 6.
16:57 sbp jangan khawatir, Anda jauh lebih inggris fasih bahwa Indonesia
16:58 masak vi: your English is good enough for me to understand you.
16:58 vi ooooo .....;)
16:59 vi tq...
17:00 ash_ vi: the Beginning Perl book http://www.perl.org/books/beginning-perl/ is really good and It should also be in indonesian
17:05 vi thank u...:)
17:05 ash_ having interactive perl is also really useful if your just learning. you can use: perl -ple '$_=eval'    if you have perl 5 installed
17:06 payload1 joined #perl6
17:16 mdxi i don't speak indonesian, but i love rendang :)
17:17 diakopter pmurias: oh
17:17 diakopter (re: use v5)
17:19 mberends oh hai diakopter
17:20 eric256 joined #perl6
17:21 diakopter hi
17:22 cdarroch joined #perl6
17:22 vi mdxi : i know rendang....:)
17:23 justatheory joined #perl6
17:23 cognominal TimToady donne de la voix  :)
17:24 masak TimToady est un voix-donateur.
17:27 cognominal litteraly "donner de la voix" means shouting.
17:28 masak oh, I see.
17:28 cognominal bad pun, as usual
17:28 masak in English, one sometimes 'gives voice to' concerns, but not always by shouting.
17:28 masak cognominal: you're certainly in the right company. :)
17:29 TimToady maybe I was just throwing my voice
17:29 cognominal do you mean that's a exception?
17:30 TimToady I have Failure to communicate...
17:30 eric256 the #perl channel isn't very open to talking about perl 6. lol at least not the few times i've been there when it comes up
17:30 eric256 :)
17:30 * masak is reminded of http://paikea066.vox.com/library/post/t​hrowing-shoes-other-global-insults.html
17:31 ash_ thats odd
17:31 masak eric256: I agree. but then again, many of the people on that channel have a vested interest in Perl 5.
17:31 TimToady so do we
17:32 masak yes, but...
17:32 * masak starts over
17:32 ash_ i may be wrong, but perl 6 might not exist without perl 5, but then again, shouldn't people plan for the future?
17:32 takadonet Perl 5 protects me from doing Java projects
17:32 eric256 its just odd to me cause i move back and forth between 5/6 and consider them intertwined...so their anti 6 feeling was a bit un-perlish to me
17:32 eric256 lol
17:32 pmurias diakopter: it's mostly a matter of changing STD to get that working
17:32 takadonet eric256: I'm in the same boat
17:33 * diakopter switches to JEDI mode.  that's Just Effing Do It mode.
17:33 masak the people on #perl (both of them) often feel they don't have to take Perl 6 into account, at least not yet. some of them feel irritated because people from outside of the community ask about Perl 6 as if it had a larger mindshare within the Perl community than it actually does.
17:34 masak eric256: I move back and forth a lot too. I'm writing a program right now which sends data back and forth. both Perl 5 and Perl 6 has individual strengths.
17:35 diakopter masak: I agree completely. That's due to the inaccurate marketing of Perl 6 and its implementations over the years.
17:35 masak diakopter: yes.
17:35 diakopter where "That" is "the incorrect perception about the larger mindshare"
17:36 masak diakopter: maybe it's not possible to keep the right expectations on a thing like Perl 6 for 10 years.
17:36 ash_ TimToady: how do you write 'when not'?
17:36 cognominal Perl6 is growing exponentially, it is just starting very small. And the perl5 crowd  will be the first crowd that will be gobbled up.
17:36 * eric256 doesn't see much gobbling for a while :P
17:37 eric256 i was just suprised by the antipathy towards it, like a red headed step child. hehe.
17:37 TimToady the marketing has always been impossible: "Keep excitement up while not overpromising"
17:39 TimToady ash_: depends on the test; but usually I try to write the cases in the correct order so they're all positive
17:40 ash_ is there a convenient way to do a when not in a given statement? I can only come up with doing like: when $_ !~~ ...
17:40 ash_ hmm, okay
17:40 TimToady when none(42)   when :!r
17:41 diakopter .oO( Real excitement doesn't need propped up; capability is louder than words.  Words can sell books, but software can sell more books. )  /me struggles to enter JEDI mode :/
17:41 TimToady we could add a negative form, but I'm inclined against it since cases are generally more understandable if expressed positively
17:41 TimToady and on any boolean expression, 'when not' already works
17:42 ash_ you can do like a macro pretty easy i imagine if you want to fix that, just do like: macro prefix:<when not> { when $_ !~~ ... } (insert appropriate macro stuff)
17:43 TimToady we've thought about that, but it seems like a smell to have multi-word keywords, and also violates the expectations of a user who is thinking prefix:<not>
17:43 TimToady I'd rather see whenn't  :)
17:44 * masak has never felt the need for 'when not'
17:45 cognominal masak, about the page you gave the url... "In France, of course, it’s a waffle. Throw a waffle at someone and you have said, in essence: “I loathe you. You are scum. Your people are donkey traders."  We don't throw waffle in France...
17:45 supernovus joined #perl6
17:45 masak cognominal: you don't?
17:45 eric256 lol cog, i'm guessing they dont throw tanks in russia either. lmao
17:45 masak cognominal: what do you throw?
17:45 diakopter 'when not' -> 'except'
17:46 cognominal I don't know anything thrown specific to french.
17:46 frettled cognominal: cows
17:46 frettled fecher la vache!
17:46 cognominal fait chier!
17:47 eric256 now go away or i shall taunt you a second time!
17:47 cognominal but that the NYT, what do they know...  that _was_ a  reputable paper
17:48 cognominal *that's
17:49 frettled cognominal: you seem to be missing the satire
17:49 cognominal probably
17:49 frettled cognominal: please read the rest of the article, it ought to be self-evident.
17:50 frettled if not, I shall be forced to throw a stave church (filled with the heads of my enemies) at you
17:50 ruoso joined #perl6
17:51 cognominal I don't accept religious arguments how heavy they are :)
17:52 cognominal *no matter
17:52 masak would a hyper function call syntax fly? something like foo>>( @array ) ?
17:52 frettled hee-hee
17:53 colomon masak: er?
17:54 orafu joined #perl6
17:54 masak that way, I wouldn't have to write 'map' all the time. :)
17:54 * masak is hyper-lazy
17:54 colomon Oh, foo is the function, and it takes one argument?
17:54 masak right.
17:55 eric256 that would be crazy. lol
17:55 colomon I proposed that back when I was working on the Mandelbrot Advent post, and it didn't fly very well.
17:55 colomon but I didn't use that syntax.
17:56 colomon I was using a similar argument -- to lazy to type hyper map.
17:57 TimToady hyper (foo($_) for @array) is about the best we have currently
17:58 TimToady there is the conjectural foo(each(@array))
17:59 TimToady but unlikely to be in 6.0
17:59 * diakopter reads pugs/docs/notes/docs_evil_plan.txt
18:01 TimToady but we might consider each() to be a junction that responds to list context rather than boolean
18:01 pugs_svn r29420 | diakopter++ | [sprixel] Add new dir for non-machine-readable writings on sprixel
18:01 TimToady and get autothreading for free
18:02 colomon masak: isn't it terrible that map is starting to seem too clumsy to us?   it's still vastly nicer than anything I have available for my C++ code (for instance).
18:03 masak colomon: convenience is addictive.
18:03 masak TimToady: I thought the death of each() wasn't greatly exaggerated.
18:04 Alias_ joined #perl6
18:05 TimToady S09:1080 still conjectures an each(), though it's mixed up with ordinary junction semantics a bit more than a pure hyperizing junction would be
18:06 masak ah, ok.
18:06 masak std: / \b /
18:06 p6eval std 29401: ok 00:01 105m␤
18:06 TimToady in my thinking above I was assuming no values would ever be thrown away
18:07 masak TimToady: switching tracks a bit.. why do both PGE and STD.pm accept \b, when S05 says it's been replaced?
18:07 TimToady so basically each lets you autothread lists while all lets you autothread sets (assuming we allow all as the only junction that can actually represent a set)
18:07 masak that sounds nice.
18:08 TimToady I think \b would match a backspace
18:09 TimToady std: /\b/
18:09 p6eval std 29401: ok 00:01 105m␤
18:09 TimToady but arguably it should be disallowed
18:09 masak rakudo: say "  foo " ~~ / \b foo \b /
18:10 p6eval rakudo db84bc: foo␤
18:10 masak PGE doesn't interpret it as a backspace.
18:15 orafu joined #perl6
18:24 masak in a multimethod which was the most specific one among a number of candidates, can I do 'callsame' to get the next most specific one?
18:25 masak rakudo: multi f(Num $x) { say "Num" }; multi f(Int $x) { callsame; say "Int" }; f(42)
18:25 p6eval rakudo db84bc: Null PMC access in clone()␤in Main (file <unknown>, line <unknown>)␤
18:26 ash_ isn't that nextsame?
18:26 masak only if I don't want to return.
18:26 masak in my particular case, I do.
18:27 ash_ so... it would return twice?
18:27 masak but I'm asking because I suspect that the answer to my question is "no", and that I don't fully grok 'callsame' et al. yet.
18:27 ash_ how would that work? (logically speaking) which value would be the return value?
18:28 masak ash_: foo(Int) would call foo(Num). foo(Num) would return to foo(Int), and foo(Int) would return as usual.
18:28 masak again assuming that it should work at all.
18:30 xomas_ joined #perl6
18:35 ash_ masak: after looking at S06 and S12 i don't know how that should work... *off to check for a spec test to see if those make more sense*
18:35 * arnsholt does not grok the nqp-rx code
18:39 ash_ masak: i think callsame is really for inheritance not for multi-methods specifically
18:39 TimToady it's for both
18:39 masak ash_: well, it's also used when doing calls with .* and .+
18:39 masak ash_: and for wrappers.
18:40 masak TimToady: so, what should happen in my multi above?
18:40 TimToady it's essentially just thrown a 'next' control exception to whatever dispatcher is running down a list
18:40 hejki hmm.. WTB: class Foo { method bar { … } } :)
18:40 masak I have a vague feeling I've been discussing this with jnthn++ before...
18:41 TimToady that should say Num and then Int, I think
18:41 masak the problem, I have a vague feeling he explained, is that the multi-dispatcher has already done away with the looser candidates.
18:41 masak but I might mis-remember.
18:41 * masak submits rakudobug for now
18:41 TimToady it can't have, or it's not a Perl 6 dispatcher :)
18:41 masak :)
18:42 TimToady at minimum, it has to be able to reproduce the tail of the list
18:44 pugs_svn r29421 | lwall++ | [STD] redirect /\b/ to use a p6 word boundary assertion instead
18:48 ash_ TimToady: if a role defines a method with ^ in it and is composed into a class, would you expect that class to have the ^method? eg. role foo { method ^bar { } }; class Car does foo {}; Car.^bar; ?
18:50 pmurias joined #perl6
18:51 pmurias TimToady: what would i take to add 'use v5' to STD.pm?
18:52 TimToady pmurias: finish turning STD_P5.pm into a Perl 5 grammar :)
18:52 hejki misread turding :)
18:53 pmurias diakopter: want to help make STD support 'use v5'?
18:53 TimToady ash_: I would expect any method defined in a role to work identically to how it works in the class it's composed into
18:53 TimToady you should know that STD_P5.pm is still a complete mess in the non-regex bits
18:54 TimToady but we at least need to get it to the point where it can find the end of a P5 block
18:55 pmurias it being a complete mess increases the propability of my changes improving it ;)
18:55 TimToady STD_P5.pm is an attempt to get to a P5 grammar by cutting down the P6 grammar and adding in the special P5 bits
18:55 TimToady it's currently partway through the cutting down stage
18:56 TimToady but in theory we can at least get it to parsing as well as PPI does
18:56 * justatheory does not parse
18:56 TimToady that's because it's too dark inside of you
18:56 seeker42 joined #perl6
18:56 TimToady errands &
18:56 justatheory exactly
19:05 eternaleye joined #perl6
19:08 seeker42 left #perl6
19:09 diakopter pmurias: no, but thanks; I'm reimplementing STD/Cursor/gimme5...
19:13 pmurias not again... ;)
19:20 Student joined #perl6
19:23 rgrau` joined #perl6
19:24 pmurias diakopter: how many parser engines did you write already?
19:29 Lorn joined #perl6
19:31 colomon pmurias: I'm sure once you've done your first dozen, it starts getting easy.  ;)
19:31 diakopter pmurias: just 1, but many, many times.
19:31 eric256 Practice makes perfect
19:31 eric256 what is it parsing?
19:33 mberends eric256: Perl 6 (or Perl 5) source code
19:35 eric256 is that all? lol.. i was trying to figure out what the name STD/Cursor/gimme5 means...
19:36 payload joined #perl6
19:39 ash_ joined #perl6
19:43 hejki hmm.. do i need to stuck stuff into some export-line in builtins on ng branch?
19:46 Tene hejki: eh?
19:46 hejki like with master-branch of rakudo ..
19:47 Tene I don't understand what you're asking.
19:47 hejki lemme rephrase
19:48 hejki the master-branch src/builtins/*.pir define a !EXPORT-line to add method-calls into so they're exported. is this still done in the same way in ng-branch?
19:50 Tene I don't see it happening anywhere in src/builtins/*.pir in ng.
19:50 mberends pmurias: what is your YAML::XS version? since changing 0.23 here to 0.32 this morning, STD.pm does not build, "YAML::XS::Load Error: Invalid leading UTF-8 octet was found at document: 0"
19:50 hejki Tene: neither do i. how is stuff exported then? :)
19:50 hejki i'm trying to add methods but i fail hard
19:51 Tene Methods shouldn't need to be exported anywhere...
19:52 Tene So, I think I'm missing something.  What are you trying to do that doesn't work?
19:52 hejki make a method for Str-class via src/builtins/Str.pir
19:53 Tene can you post the diff?
19:53 hejki well mainly it just adds a test method (to first try out how-to make methods properly) but just a sec
19:54 abra joined #perl6
19:54 Tene Make sure it's in the Str namespace and marked with :method
19:58 hejki it's supposed to be C<.sub ':s' :method> and used in C<Str ~~ :s>. maybe this is why i'm failing
19:59 hejki ?
20:01 zamolxes joined #perl6
20:03 Tene I don't see any indication that that's supposed ot be a method of Str
20:04 hejki heh.. we talked this already over once today :P
20:04 Tene Look at the pir it compiles to.
20:04 hejki not you and me, but anyways :P
20:05 hejki if i call it .sub 'schlumorp' instead of ':s' and call Str.schlumorp it does work just fine (sherlock!)
20:05 Tene {"foo" ~~ :s} compiles to: &infix:<~~>("foo",1)
20:05 Tene currnetly
20:05 Tene Yes, of course it works.
20:05 hejki hmm..
20:06 hejki so i need to define a proper call for ~~ :s
20:06 sjohnson TimToady: as far as the "when not" thing is concerned, i thought it would be a cool addition as I use it sometimes like the if / unless pair
20:06 sjohnson fwiw
20:06 Tene I don't know where the correct place to do that is, but it's not currently making a Pair object there, like I thought it was supposed to.
20:07 hejki eww
20:08 abra joined #perl6
20:11 sjohnson starting to use unless quite a bit, actually
20:11 sjohnson also like the while / until pair
20:11 Tene Rakudo master compiles it to approximately: infix:~~("foo", infix:=>('s',1))
20:13 plainhao joined #perl6
20:13 abra left #perl6
20:18 abra joined #perl6
20:25 sjohnson wow someone else uses weechat here
20:25 sjohnson i suppose that makes two!
20:27 ash_ does anyone in here have OS X? and not 10.6?
20:38 orafu joined #perl6
20:40 colomon ash_: yes
20:40 colomon (OS X, 10.5.8 I think)
20:40 ash_ does parrot's examples/opengl/shapes.pir work for you?
20:41 colomon no idea.
20:41 colomon okay, I'm in that directory.  how would I run it?
20:42 colomon ./parrot examples/opengl/shapes.pir
20:42 colomon ?
20:42 colomon Could not find a suitable GL shared library!
20:42 Tene works for me!
20:45 colomon afk for grocery shopping, back in a bit if more OS X experiments needed.  :)
20:48 ash_ colomon: k, thanks, its doing that for me to
20:48 ash_ Tene: are you on OS X? i think this is an OS X specific issue
20:50 Tene ash_: No, I'm not.  I'm not really being helpful here. :)
20:51 ash_ Tene i know it works for me in linux, so just trying to track down the issue in OS X and get a proper bug report for parrot (although they have one, maybe i'll just do a patch instead...)
20:51 arnsholt I'd guess it's related to how OS X packages libraries
20:52 arnsholt With frameworks and such
20:53 arnsholt Also, FWIW, I get the same error in 10.6
20:53 ash_ well, i think its looking in the wrong folder, actually, i know it is on 10.6, i am just trying to figure out why its configured the way it is, thats why i was looking for someone with not 10.6, maybe 10.4 has a different layout, one that follows what parrot currently does, so yeah. /thinking outloud
20:54 pmurias mberends: 0.27
20:56 mberends thanks pmurias, be careful upgrading viv dependencies
20:59 pmurias upgraded YAML::XS and i ge the same error
20:59 mberends :(
21:01 pmurias we need to move towards using only YAML::XS or YAML::Syck
21:02 ash_ Syck, so you know, isn't really maintained much anymore
21:02 mberends using YAML::Syck alone did not work 3 days ago :(
21:02 pmurias what happended?
21:02 ash_ whytheluckystiff (the guy that made syck) is gone
21:03 ash_ and deleted all his sites, and github account, and twitter account, etc.
21:03 mberends the YAML file format changed in subtle ways, anchor numbers were out of sequence and there were empty lines after [] sequences
21:05 mberends pmurias, see comments in viv lines 11-17
21:06 pmurias what programs did it confuse?
21:06 pmurias nothing sensible should depend on anchor ordering
21:08 mberends true. it may have been only vill, but now I don't remember exactly whether mildew was affected.
21:11 * pmurias gets rids of YAML::Syck and sees what happens
21:12 * ruoso just found out that Python doesn't really support threads...
21:12 mberends pmurias: see http://irclog.perlgeek.de/​perl6/2009-12-27#i_1870482 through to 07:22
21:14 ash_ ruoso: python has a GLI, just like ruby, although they are system threads so they get scheduled by the OS which is good
21:15 ruoso but it's not really threading support...
21:15 ash_ ruoso: if you want true concurrency in python look up Multiprocessing, it uses sub-process and shared memory to get real parallelism
21:15 ruoso pretty much what threads::shared does
21:17 ruoso but threads::shared uses threads
21:17 ruoso it's like POE
21:17 ruoso Python is
21:19 ash_ stackless python has no GIL, and i know unladen swallow (an experimental python fork integrated with the llvm) is in the process of removing the GIL
21:19 pmurias YAML::XS seems to require something none UTF-8
21:20 mberends that's right, it's looking for a byte order mark
21:21 pmurias please don't use the BOM
21:22 pmurias what i meant is that it mangles ¢ into ¢
21:22 mberends no, not using it at all, just getting the error message "Invalid leading UTF-8 octet". oh wait, that's not a BOM.
21:26 pugs_svn r29422 | pmurias++ | [gimme5] remove some noop code
21:27 pugs_svn r29423 | pmurias++ | use only YAML::XS, it might be spewing out strange things but at least
21:27 pugs_svn r29423 | it should be compatible with itself
21:28 pmurias mberends: i switched STD to pure YAML::XS
21:29 racheldaBROWN joined #perl6
21:29 mberends ok, trying to rebuild STD
21:29 ruoso pmurias, http://github.com/skaphan/stmmap this is something I didn't know... and seems to fit quite well with smop
21:30 * pmurias checks it
21:33 ruoso Ok... time to go..
21:33 lisppaste3 mberends32 pasted "r29423 STD build error for pmurias" at http://paste.lisp.org/display/92839
21:39 pmurias mberends: it works here
21:43 mberends pmurias: it could not YAML::XS itself over here, but something that YAML::XS depends on.
21:43 mberends *could be
21:45 mberends most Perl modules on this computer are Ubuntu versions, some are now CPAN upgraded. It would make sense to CPAN upgrade them all.
21:48 orafu joined #perl6
22:04 eternaleye joined #perl6
22:10 xenoterracide joined #perl6
22:11 eternaleye joined #perl6
22:21 pmurias mberends: could you check the what does the program i nopasted do?
22:22 pmurias http://nopaste.info/e0c370dd5c.html
22:25 benkim joined #perl6
22:26 mberends pmurias: one success, one failure as predicted
22:26 lisppaste3 mberends32 pasted "utf8 test for pmurias" at http://paste.lisp.org/display/92844
22:28 mberends cpan is busy upgrading many modules while the test runs
22:28 kangu_ joined #perl6
22:34 pmurias mberends: so it seems YAML::XS does like utf8 marked as such
22:38 mberends ouch
22:40 pmurias so what i plan to do is to make sure utf8 doesn't get mangled anywhere and unmark it before passing to YAML::XS
22:40 pmurias OTOH we could just fix YAML::XS
22:40 pmurias nothingmuch: ping
22:41 mberends pmurias: whatever you think is best
22:41 pmurias one is easier the other is more correct
22:46 eric256 left #perl6
22:48 pugs_svn r29424 | pmurias++ | try unmarking text sent to YAML::XS
22:48 pmurias mberends: see if that works for you
22:48 mberends ok, right away
22:50 eternaleye joined #perl6
22:53 ejs joined #perl6
23:00 mberends pmurias: \o/ STD make and ./tryfile ran ok
23:18 soupdragon joined #perl6
23:25 jnthn phenny: any messages for me?
23:26 jnthn Aww. :-)
23:26 jnthn <- alive, back "properly" in a bit.
23:27 mberends phenny, tell jnthn welcome back
23:27 phenny mberends: I'll pass that on when jnthn is around.
23:28 ash_ jnthn: welcome back?
23:28 pmurias mberends: seems the hack works
23:29 mberends pmurias: yes, doing mildew test now :)
23:59 colomon joined #perl6

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

Perl 6 | Reference Documentation | Rakudo