»ö« | perl6.org/ | nopaste: paste.lisp.org/new/perl6 | evalbot usage: 'perl6: say 3;' or rakudo: / pugs: / std: , or /msg p6eval perl6: ... | irclog: irc.pugscode.org/ | UTF-8 is our friend!
Set by wolfe.freenode.net on 30 October 2009.
00:03 ash__ joined
zaslon lolperl6adventhazblogged! perl6advent++ 'Day 18: Roles': perl6advent.wordpress.com/2009/12/1...-18-roles/ 00:06
00:07 _jaldhar joined
diakopter pmurias: sorry (was afk) 00:09
00:10 nihiliad left
diakopter pmurias: let's wait on that idea; keep at the translation of STD output, I'd recommend... 00:10
quiet the past hour... 00:19
00:19 ash__ left
pmurias diakopter: doing git svn dcommit then ;) 00:19
00:20 jaldhar left
carlin jnthn++ # 'does' good advent posts 00:23
00:23 lestrrat is now known as lest_away 00:25 justatheory left 00:27 snearch_ left
pmurias mildew-js now only fails to more tests then before moving to new STD (because the newer STD cannot parse those) ;) 00:40
s/to/2/ 00:41
cotto_w0rk who's in charge of the advent calendar? The code needs some indentation love. 00:43
diakopter cotto_w0rk: you? :) 00:46
00:46 lest_away is now known as lestrrat
cotto_w0rk definitely not me 00:47
arnsholt It'll have to be someone with the right permissions in Wordpress, I'd bet =) 00:48
chromatic will send the release announcement later tonight 00:50
00:51 lestrrat is now known as lest_away 00:53 justatheory joined 00:55 justatheory left 00:59 pmurias left 01:05 dduncan joined, dduncan left 01:07 chromatic left 01:09 ShaneC left 01:12 payload joined 01:15 jferrero left 01:26 TiMBuS left 01:31 JimmyZ joined 01:35 lest_away is now known as lestrrat 01:52 tsbtmn joined 01:53 tjc joined 02:00 colomon left 02:15 patspam left 02:18 tjc left 02:22 RichiH left 02:23 agentzh joined 02:24 RichiH joined 02:27 kst` is now known as kst 02:30 patspam joined 02:33 colomon_ joined 02:38 tarski joined 02:42 mikehh left 02:43 mikehh joined 02:49 gbacon left 02:52 stephenlb left 02:54 patspam1 joined 02:55 patspam left 02:56 gfx joined 02:57 RichiH left, RichiH joined 03:00 tarski left 03:08 colomon joined, colomon_ left 03:13 justatheory joined 03:17 justatheory left 03:20 colomon_ joined, colomon left, colomon_ is now known as colomon 03:21 cdarroch left 03:22 meppl left
colomon phenny: tell jnthn Somehow all the code formatting had disappeared from your post. I just did my best to get it back, but you might want to double-check it to make sure I got it more or less the way you would like it. 03:28
phenny colomon: I'll pass that on when jnthn is around.
03:35 colomon left 03:45 constant left, ssm left, parduncia left 03:50 constant joined, ssm joined, parduncia joined 03:53 ssm left, parduncia left, constant left 04:02 constant joined, ssm joined, parduncia joined
PerlJam succeed/proceed +1 04:15
04:24 zaphar_ps left 04:49 TimToady sets mode: +vvv buubot dalek hugme, TimToady sets mode: +vvv ilbot2 ilogger2 IRSeekBot, TimToady sets mode: +vvvv lisppaste3 mubot p6eval phenny, TimToady sets mode: +vv pointme zaslon 04:50 patspam1 left 05:03 meppl joined 05:04 tsbtmn left 05:34 matsuzine_ joined 05:44 mz_ joined, mz_ left
japhb nqp: my $foo; sub say_foo () { say($foo) }; INIT { $foo := "Foo"; }; say_foo() 05:50
p6eval nqp: ␤
japhb How do I make that work?
TimToady nqp: state $foo = "Foo"; sub say_foo () { say($foo) }; say_foo(); # maybe? 05:52
p6eval nqp: Confused at line 1, near "state $foo"␤current instr.: 'parrot;HLL;Grammar;panic' pc 500 (src/cheats/hll-grammar.pir:197)␤
TimToady # maybe not...
obviousl you need a "so quite perl" 05:53
*ly
japhb well, sure. :-) 05:54
Of course, since I'm trying to write something to help install SQP, I'm kinda stuck. ;-)
05:54 justatheory joined 05:57 rgrau` left
diakopter std: so so 1 06:12
p6eval std 29369: Undeclared routine:␤ 'so' used at line 1,1␤ok 00:01 104m␤
diakopter std: not 1 06:13
p6eval std 29369: ok 00:01 105m␤
diakopter std; true True
std: true True
p6eval std 29369: ok 00:01 105m␤
diakopter std: so not True
p6eval std 29369: Undeclared routine:␤ 'so' used at line 1␤ok 00:01 105m␤
diakopter std: use ValleyGirl; 06:14
p6eval std 29369: Potential difficulties:␤ Can't locate module ValleyGirl at /tmp/LvcGxV390P line 1:␤------> use ValleyGirl⏏;␤ok 00:01 103m␤
06:15 justatheory left, hicx174_ left
diakopter std: that is, like, so like, so not True 06:16
p6eval std 29369: Undeclared routines:␤ 'is' used at line 1␤ 'like' used at line 1,1␤ 'so' used at line 1,1␤ 'that' used at line 1␤ok 00:01 106m␤
06:16 hicx174 joined
TimToady and stuff 06:19
06:28 matsuzine_ left 06:40 Caelum joined
diakopter rakudo: say (!!7).WHAT === (Bool::True).WHAT === (true 7).WHAT 06:46
p6eval rakudo 7f1c3f: 1␤ 06:47
diakopter pugs: say 3 ?? !!4 !! 5
p6eval pugs: 1␤
diakopter rakudo: say 3 ?? !!4 !! 5
p6eval rakudo 7f1c3f: 1␤
moritz_ std: say 3 ?? !!4 !! 5 06:48
p6eval std 29369: ok 00:01 105m␤
diakopter rakudo: say 3 ?? !!0 !! 5
p6eval rakudo 7f1c3f: 0␤
diakopter rakudo: sub so { !!$^a }; say (so not True).WHAT 06:52
p6eval rakudo 7f1c3f: Bool()␤
06:52 Su-Shee joined 06:59 silug left
diakopter twitter dieth 06:59
oooo www.flickr.com/photos/ceronman/4194674390/
I'd tweet that url, but ... 07:03
moritz_ hugme: tweet hugmetest test
hugme moritz_: Sorry, I don't have access to twitter account 'hugmetest' 07:04
moritz_ hugme: tweet hugme_test test
hugme hugs moritz_; tweet delivered
TimToady std: so so ... 07:06
p6eval std 29370: ok 00:01 107m␤
TimToady std: not so so so ...
p6eval std 29370: ok 00:01 105m␤
moritz_ so you think that's funny? :-)
TimToady std: not so true 07:07
p6eval std 29370: Undeclared routine:␤ 'true' used at line 1␤ok 00:01 105m␤
TimToady std: so say my @friends 07:08
p6eval std 29370: ok 00:01 106m␤
TimToady std: [1...] 07:09
p6eval std 29370: ===SORRY!===␤Missing term at /tmp/2LLfdQa8Wx line 1:␤------> [1...⏏]␤ expecting any of:␤ prefix or term␤ standard stopper␤ term␤ terminator␤ whitespace␤FAILED 00:01 105m␤
diakopter oh, they recovered twitter quickly 07:10
TimToady std: 1...... 07:12
p6eval std 29370: ok 00:01 104m␤
TimToady std: .........
p6eval std 29370: ok 00:01 104m␤ 07:13
diakopter std: ... ... ... ...
p6eval std 29370: ok 00:01 104m␤
diakopter std: ...: # SLTA error 07:14
p6eval std 29370: ===SORRY!===␤Illegal use of colon as invocant marker at /tmp/lavfQDCEsF line 1:␤------> ...:⏏ # SLTA error␤FAILED 00:01 104m␤
TimToady what do you think it should say? :)
something like Huh? 07:15
diakopter "not everyone gets the colon"
"I think you need a colonectomy"
"..or perhaps a colostomy, or maybe just a colonoscopyo" 07:16
"..or perhaps a colostomy, or maybe just a colonoscopy"
sorry 07:17
TimToady std: so not funny
p6eval std 29370: Undeclared routine:␤ 'funny' used at line 1␤ok 00:01 105m␤
diakopter std: so so !! any 1 but self 07:19
p6eval std 29370: ok 00:01 105m␤
TimToady std: not so eager
p6eval std 29370: ok 00:01 105m␤
TimToady std: not so hyper # bedtime 07:20
07:20 IllvilJa left
p6eval std 29370: ok 00:01 105m␤ 07:20
moritz_ good night
07:20 meppl left
diakopter moritz_: nite 07:20
I mean
heh
moritz_: g'morning
moritz_ :-) 07:21
07:26 mariuz joined
diakopter std: class must does must does must does must does must does must { } 07:28
p6eval std 29370: ok 00:01 104m␤
07:32 kaare joined 07:33 kaare is now known as Guest81964 07:39 gfx left 07:47 iblechbot joined
spinclad std: so not hyper 07:55
p6eval std 29370: ok 00:01 106m␤
07:55 chromatic joined
spinclad std: not so True 07:56
p6eval std 29370: ok 00:01 105m␤
spinclad std: so not True
p6eval std 29370: ok 00:01 105m␤
spinclad std: so antonymous! 07:57
p6eval std 29370: ===SORRY!===␤Negation metaoperator not followed by valid infix at /tmp/IQ8QDSLV9V line 1:␤------> so antonymous!⏏<EOL>␤ expecting any of:␤ infix or meta-infix␤ infix stopper␤ standard stopper␤ terminator␤Undeclared routine:␤
..'antonymous' used at lin…
spinclad std: so not the same 07:58
p6eval std 29370: Undeclared routines:␤ 'same' used at line 1␤ 'the' used at line 1␤ok 00:01 105m␤
chromatic Do I have posting permissions on rakudo.org?
diakopter prob not 07:59
chromatic Can someone with permissions post the release announcement?
diakopter I think only alester and pmichaud
and jnthn
moritz_ and me 08:00
and masak
and all previous release managers :-)
chromatic Well volunteered!
moritz_ either rakudo.org or my internets are slow 08:02
rakudo.org/node/61 08:03
chromatic Thanks!
moritz_ chromatic++ # managing one of the most difficult releases so far
chromatic Especially on a basketball night. 08:04
diakopter quietfanatic++ helped me talk out how to manage declarative API for labels/gotos in declarative MSIL generation (runsharp.googlecode.com) 08:06
08:07 masak joined
diakopter masak: hm 08:07
masak good morning.
moritz_ lolitsmasak 08:09
is there a book release today?
Tene anyone willing to comment on preferred style for p6advent posts?
masak moritz_: yes, I have a feeling there is.
Tene Oh, it's masak! I bet I can con him into doing it!
masak Tene: 'preferred style'? 08:10
we're anarchists!
Tene masak: I'm talking about Grammars... think there's any value in including the long output of .perl on a match object?
232 lines.
Or just snip it?
masak eurgh.
Tene Maybe just print parts of it... 08:11
masak keep it short, I think.
moritz_ Tene: you could include a visualization
masak moritz_++
Tene ... ><
moritz_ Tene: I've written a moduel for that
Tene Orly?
moritz_ pointme: SVG::MatchDumper 08:12
pointme Sorry, I don't know anything about SVG::MatchDumper
Tene I wrote a little program once to generate graphs of STD.pm and Rakudo's grammar. That was fun.
moritz_ github.com/moritz/svg-matchdumper
I need to check if it bitrotted 08:13
Tene Hmm. i'd need to install SVG.pm too. 08:14
Maybe I just make you generate the svg for me.
lisppaste3 tene pasted "untitled" at paste.lisp.org/display/92302 08:15
moritz_ I don't know if it handles multi-line input... 08:19
masak wohoo! the first Rakudo release with a negative number of additional passing spectests! \o/
moritz_ nope, not useful :(
diakopter cool; I got the declarative labels/gotos to work: mono.pastebin.com/d37392572
masak ooh, also (.grab)++ 08:22
TimToady: should there be an Array.grab as well?
(for the times when one wants to mutate the array as well as return the picked items) 08:24
08:24 colomon joined 08:25 JimmyZ_ joined
masak ooh, and (succeed/proceed)++! only good changes this mroing! :) 08:26
08:35 JimmyZ left 08:42 Chillance joined
Tene Up to 112 lines long blog post! Almost halfway to the length of that match object. ;) 08:42
Hmm. This looks to be rather longer than most of the existing posts.
Anyone willing to review it so far? 08:43
08:45 Chillance left, Chillance joined 08:46 Chillance left
masak reviews it 08:46
mathw OH HAI
Tene: what's it about?
lisppaste3 tene pasted "p6advent WIP" at paste.lisp.org/display/92303 08:47
08:47 Chillance joined
Tene mathw: grammars and action methods. 08:47
08:47 Chillance left, Chillance joined
masak Tene: no riff on 'today's gift' in the beginning? :) 08:47
colomon wait, where is it to be reviewed?.... oh, that's freaky, the paste came up here just as I finished typing the first half of the sentence.
08:48 Chillance left
Tene I'm about halfway done. 08:48
08:48 bl joined, Chillance joined 08:49 Chillance left
masak Tene: why token + <.ws> instead of rule? 08:49
08:49 Chillance joined 08:50 Chillance left, Baggio_ joined, Chillance joined
Tene masak: I didn't really want to get into regex/token/rule any more than I did, and I wanted to be precise about whitespace because it's used to differentiate answer/header, and I wanted a reason to mention leading ., and to be more idiomatic 08:50
s/more idiomatic/because that's how I was initially inclined to write it/ 08:51
masak :P
fair enough. not a big issue; I was just a tad surprised.
Tene nods. 08:52
I thought about it.
masak $_<header><text> 08:53
08:53 silug joined
masak better written as .<header><text> 08:53
colomon oooo, oooo, I want to see what comes next!
masak also, if it's not too late to change, the class name 'Answer' feels a tad confusing. would it be better to name it 'Alternative' or 'Option' or 'Item'? 08:55
Tene Those are a bit too vague, IMO.
You've got your Correct Answers and Incorrect Answers.
masak it's just that Question::Answer feels like a wonky type name. 08:56
but oh well.
maybe type $.correct as a Bool?
Tene I'm a bit opinionated about that right now. I've spent the past week trying to keep similar things in this semantic space different, so...
Oh, did I not include that?
JimmyZ_ masak: chinesepod.com/
Tene Yes, let's. 08:57
08:57 JimmyZ_ is now known as JimmyZ
masak JimmyZ_: 太好了! 谢谢. 08:57
Su-Shee hm. why do I still have this font rendering problem in certain ranges.. 08:58
JimmyZ masak: I don't know any more about it. Just find it.
mathw Tene: looks okay so far
masak JimmyZ: thanks for thinking of me. 08:59
JimmyZ :)
Tene masak: opinions on iterating over $<answer> and $<question> in the action method vs >>.ast ?
We did already have a post about >>.
masak Tene: if it's without side effects, >>. is fine.
IMHO. 09:00
Tene Yes, I thought that too.
mathw yeah >>. is great if you're not worried about ordering or interleving 09:03
09:04 Baggio_ left
masak ordering in the temporal sense. 09:05
the order of the sequence that comes out corresponds to the order of the sequence that goes in. 09:06
sjohnson yo masak 09:07
masak 'sup sjohnson
mathw masak: yes, I was referring more to order of execution
well, entirely, actually
masak :) 09:08
09:08 chromatic left
sjohnson masak: just drinking some wine and checkin the internet 09:09
masak sjohnson: well, here we are. the Internet. and we're looking back at you. :) 09:10
mathw mmmmm wine 09:11
I doubt my manager would like it if I started drinking wine in the meeting we're having in a bit
Although he'd probably be happy to join in
Maybe I should go to the shop... 09:12
Tene oh man, this prose really needs some review. I'm not very confident in it. 'sec, lemme post. 09:13
sjohnson just buy a nice looking bottle
and say it's to celebrate sometihng
much better idea than what i did, which was bring a 6 pack to work!!! 09:14
09:14 lestrrat is now known as lest_away
sjohnson <--- learned his lesson 09:14
lisppaste3 tene pasted "more p6advent WIP" at paste.lisp.org/display/92304
Tene I'm particularly interested in feedback on anything after /As you can guess/ 09:16
JimmyZ wow,more long
Tene JimmyZ: too long for p6advent? 09:17
JimmyZ no, It's better.
spinclad Tene: s/:action/:actions/
Tene spinclad: only in ng, not in master. 09:18
I'll add a note that it will be different eventually, though. Thanks.
spinclad ah, i remembered the transition but forgot when it happened. my bad. 09:19
Tene yeah, it tripped me up too.
masak is this the first mention of the ng branch in an advent post? or did someone mention it before?
Tene enoclue 09:20
day 14 mentions it
"One interesting development which has not made it into the main Rakudo build yet is decimal numbers in the source are now spec’d to be Rats. Luckily this is implemented in the ng branch, so it is possible to demo how it will work once it is in mainstream Rakudo"
spinclad are the implicit {*}'s in master? they also transitioned in.
Tene spinclad: they are not. 09:21
masak Tene: "Named and Positional access (<> and []) without a variable name go to $/." -- I know what you mean, but I'm not sure a reader would.
Tene Yeah, that's not great.
masak perhaps concrete examples would be better. 09:22
Tene "Named and Positional access without a variable name ($<foo>, $[0]) go to ..."
?
masak something like "$1 translates to $/[1] and $<foo> translates to $/<foo>"
moritz_ $<foo> is a shortcut for $/<foo> and $1 a shortcut for $/[1]
masak yes.
Tene Yes, I like that. Thank you. 09:23
"Named and Positional access without a variable ($<foo> and $[1]) are translated into access to $/ ($/<foo> and $/[1])." ? 09:24
masak by the way, Happy 22th birthday! \o/
Tene Whose?
masak Tene: Perl's. 09:25
Tene Ah!
masak s/th/nd/
09:25 JimmyZ left
masak Tene: I'd slap on a ", respectively" in the 2nd parenthesis, but that's mostly because I like the word. :) 09:26
Tene ", masakspectively"
09:27 rfordinal joined
masak erm. In the Wikipedia article for Perl, under section "Future", it says "Main article: Perl 6". 09:27
wonder what the #perl people would think about that... 09:28
moritz_ :-)
masak I think I'll go and ask them. :)
09:32 bl left 09:33 krakan joined
masak I was quickly told that Perl 6 is not released yet. maybe next year. 09:33
09:33 JimmyZ joined
frettled Haha :) 09:35
09:35 frettled sets mode: +o masak
wayland76 Well, they must be fine with it, right? 09:36
Su-Shee this bickering between the perls could stop some day. ;)
wayland76 yes -- as soon as we decide it's casting our perls before swine :) 09:37
mathw I want to know, ignoring any concerns about implementation and migration, why there are some people who say they don't like Perl 6 the language 09:40
As in, what don't they like about it
Why do they think Perl 5 is better?
Are there technical arguments, or is it just people not wanting to deal with something different? 09:41
Tene mathw: I have a friend who HATES it, apparently because it disallows dropping a space between 'if' and parens around a condition.
wayland76 Some of them it's just CPAN, but others, I don't know
Ah, yes, there are those
Su-Shee perl 5 is there, fast, usable, with a vast landscape of modules.
frettled mathw: Ah. There are plenty of reasons for that, rationally and irrationally. Rationally, Perl 5 works today, it's reasonably quick, its core features don't change every week, and yes, CPAN.
wayland76 I don't like that either (the if-space thing), but I can live with it to get the rest
09:42 pnu left
frettled Yeah, the significant whitespace-thing is definitely something that you have to be extremely careful with, also in mentioning. :) 09:43
mathw Tene: but... you can drop the brackets completely!
Tene mathw: that's even worse, according to him.
mathw ... 09:44
I can't comprehend that
I look at Perl 6 code and it feels clean and airy
spinclad Tene: (re draft): I know I have a hard time, in exposition like this, achieving a tone that is light and still precise, complete but not heavy. This seems a very good draft: a good choice of toy, well explained. Perhaps in polishing I would look to maintain an awareness of the setting (opening presents under the tree) and let it help with lightness of tone.
mathw Largely because of that
wayland76 Well, I've been writing Perl and C for the last 14 years, and doing it that way. I can understand not wanting to have to re-learn
Tene mathw: it has to do with hoe he chunks code as he reads it. 09:45
Su-Shee mathw: yeah, that's all great and pretty and shiny - and what exactly do I use in the office for web programming with a data base and SSL? or for network programming? or for parsing 200 gig logfiles _per day_?
Tene how
Su-Shee you may insist on believing it's about "not wanting to learn" - it's not.
(which is a pretty nasty assumption on the side of the perl6 folks, btw.) 09:46
mathw Su-Shee: I specifically said this is apart from migration issues, current development state, performance etc. 09:47
It's obvious that people are going to be reluctant to move away from a proven system
Su-Shee mathw: so, what exactly is left then?
09:47 frettled sets mode: +o Su-Shee
mathw The language design itself 09:47
moritz_ dislike for the language, community, learning process, websites
Su-Shee "use Perl 6 - if you don't need speed, don't have to migrate, need no module and don't need a complete implementation"? :)
yeah and yet you all care so very much what the perl 5 folks think about you... ;) 09:48
mathw I'm not saying those things aren't important, but I understand why people raise them, so that's why I was specifically excluding them 09:49
But there are people posting things in which they say they don't like the language itself, and that's what intrigues me
lisppaste3 tene pasted "Full p6advent draft" at paste.lisp.org/display/92305
Su-Shee mathw: so try to understand that for many people _those_ _are_ the issues. no matter how nice the language itself might be.
spinclad sure, the language has a lot of growing up to do. 09:50
mathw Su-Shee: And as I said, I'm not curious about those people because I understand those objections. 09:51
Su-Shee I'm rather undecided up until now wether or not I really like Perl 6 as long as I constantly bump into some limits. and the "it'll be there' is a little old for someone like me who already did Perl in '96 and really waited from 2000 over pugs until today. ;)
frettled heh
mathw Perl 6 the language is distinct from Perl 6 as you can currently run it, and I'm interested in why people might dislike the language. 09:52
frettled I recall that at NPW in 2006, there was something about pugs, but it wasn't terribly interesting.
Rakudo and ng, however, give me positive vibes.
Su-Shee because it's typing hell on a non english keyboard. because the operator zoo is hell of confusing. for starters.
Tene Su-Shee: can you explain non-english keyboard problems with Perl 6? 09:53
moritz_ Su-Shee++ # talking turkey (if I my trust my dictionary)
mathw The operator zoo?
Su-Shee moritz_: sounds christmassy :)
moritz_ Su-Shee: american englisch for "Tacheles" 09:54
frettled mathw: Well, as I said, the significant whitespace thingy is annoying. There are other things in P6 which are similar enough to other languages on the surface, but which surprise you when you dig into them. The specification is unfinished, and it keeps changing. It's a language for theorists, not for programmers, since there is no implementation to test it on. Etc.
Su-Shee Tene: I already hated the sigils anyway and I have small hands, I have to grab keys like @ % $ for example with annoying finger acrobatics.
frettled And yes, it's a bother on a non-english keyboard, but Perl 5 isn't much better there. 09:55
Su-Shee which is why I would prefer languages with very few special chars.
mathw An unfinished spec and a lack of implementation are not objections to the actual language, they're reasons not to use it yet.
frettled mathw: Yes, actually they are. 09:56
Su-Shee {} plainly sucks on a german keyboard.
frettled mathw: The unfinished spec most definitely is an objection to the actual language.
mathw Su-Shee: is Python more popular in Germany then?
frettled mathw: The lack of an implementation makes it an _impractical_ language.
Su-Shee mathw: perl isn't popular anymore at all, most folks switched to ruby, python, java, php. there's a functional fashion wave, though.
mathw Functional programming is an eminently sane way to think about certain kinds of system 09:58
Despite what Guido might think
Su-Shee anyway. I have my reasons why I like both Perls, but there's many hurdles Perl 6 has yet to take.
Guido?
frettled Yep.
mathw Guido, the Python man 09:59
Su-Shee ah, that guido.
I don't care, I know some python, but I don't like it. it's not sexy even though it's great to type.
mathw There was some scuffle over getting rid of a load of functional constructs from the core, and an ongoing thing about tailcall optimisation
Python is... not nice
frettled I like Perl 6 for the learning experience. I like it for fixing some of the annoyances with Perl 5. And I like it for simplifying a bunch of things. :)
mathw It's got nice bits, but I find its inability to have strict variable declarations or parameter type annotations is a great weakness in readability and maintainability 10:00
oop
mathw -> boring meeting
Su-Shee well I want write code and get projects out and use my favorite language in the office and earn money with it and do open source and like the language and the community. if I don't get that, I ignore the language, no matter how cool or clean or nice or xy-complete its design might be. 10:02
masak Su-Shee: I think that's one of the chief (valid) criticisms of Perl 5 people. Perl 5 puts food on their family, why should they care about Perl 6? 10:03
Su-Shee: of course, people who think like that don't hang out on #perl6. 10:04
well, unless they can see the other side of the coin, that is.
10:04 am0c left
Su-Shee I think, Perl (5) people need a certain dynamic in the development - they all like shiny, crazy projects and practical modules and such. and that's something Perl 6 isn't there yet. 10:06
frettled Perl 5.10 and 5.12 is upsetting the stability of Perl! ;)
s/is/are/
Darn sentence rewriting thingy in my brain forgets to fix such things. 10:07
masak frettled: could be a precedence thing. :P
Su-Shee and #perl6 has a very academic taste sometimes, even I have difficulties to actually see a future for us common folks. ;) 10:10
masak Su-Shee: right. it's a bit like enduring the lowered living standards while people are redoing your kitchen. Perl 5 people say "why the heck would we want to do have breakfast at a table all wrapped in ugly plastic? our house is fine as it is, why should we switch to one which isn't even finished yet just for the promise of some great faraway future?"
I think p6l has a very academic taste. #perl6 used to intimidate me, but I think I've grown used to the jargon here. 10:11
10:12 payload left
Tene Okay, my p6advent post for day 21 is scheduled. 10:13
If someone else could confirm that it's set to be posted at the right time, that would be great.
frettled Tene++ 10:14
Tene I really have no taste for the christmas-themed flavor stuff, so anyone who thinks it needs some of that, feel free to add it. 10:15
10:15 rfordinal left
Su-Shee well, unless the futuere is actually the presence, Perl 6 is competing with many other great languages one could try who doesn't have the vast CPAN environment. 10:15
10:16 snearch joined 10:17 payload joined
masak give a Perl 6 implementation sufficient speed and sufficient access to Perl 5/CPAN, and it doesn't matter what subset of Perl 6 it implements. people will love it. 10:17
spinclad Tene++ # as of the last pasted draft, anyway. reads well. no access to the pending posts, will worry about that another time. 10:18
10:19 colomon left
Tene masak: Given an appropriate viewing angle, you just described Perl 5. ;) 10:20
masak Tene: I'm aware of that. :) 10:21
10:21 vamped joined
masak which is why I use Perl 5 quite a lot, too. 10:21
spinclad masak: not all subsets make a whole I would love for its sixness...
masak spinclad: true. I think I must have exaggerated. 10:22
but the fact is that it doesn't take very much.
vamped do I understand correctly, that perl6 will be backward compatible with perl5, and will run perl5 code?
masak I remember using a very reduced subset 18 months ago and loving it.
vamped: that is the idea, yes.
spinclad but speed and access to fiveness are _big_ plusses
masak vamped: not backward compatible, though. just able to bridge the language gap seamlessly. 10:23
frettled vamped: as in-line Perl 5 code
vamped wow. seems like that will be tough to pull off.
masak it's not trivial. 10:24
frettled It's a bit worse than in-line assembly in C. :)
I also suspect that this is not a _top_ priority.
BinGOs if you can do that, you can leverage existing CPAN assuming you have a suitable supporting toolchain for installation etc.
spinclad to bed & ~~ 10:28
RichiH hmm, i just had an idea.. how about an option that can be set which enables verbose signal handling? i.e., by default, it would say something like 'Caught SIGTERM, exiting' or 'Caught SIGHUP, doing nothing'. 10:30
10:30 JimmyZ left
Tene sleep 10:31
masak RichiH: sounds like a good job for a module.
RichiH masak: or that. though i am not sure how to access that info properly
might be a weekend project if someone gives me a hint or two :)
masak S17 mentions signals, but is admittedly underspec'd. 10:32
RichiH: don't want to discourage you, but I fear it might be too early. Rakudo doesn't do signals yet; nor does any other implementation I know of. 10:33
RichiH masak: heh, that might be a bit of a showstopper
where should i put this idea so it does not get lost?
other than my ever-growing todo list?
Su-Shee RichiH: on the other todolist. ;) 10:34
RichiH Su-Shee: but where is it? 10:36
Su-Shee RichiH: it's logged now. 10:38
RichiH Su-Shee: link?
or do you mean 'in irc log'?
which is pretty much useless
Su-Shee (maybe one of the bots should recognize "idea" commands which are written to some list available on the web.." 10:39
RichiH: yes, there's an extensive irc log available.
RichiH Su-Shee: maybe a 'proposed feature' list is a tad more useful 10:40
Su-Shee I don't care how it's called :)
proposed features sounds like "someone should do it" - ideas are "I'll do it when I have component xy available" :) 10:41
wayland76 RichiH: Maybe you could add it to your "list of things to do someday" :)
RichiH wayland76: 11:33:47 < RichiH> other than my ever-growing todo list? 10:42
i actually _do_ plan to pursue this, but a way to say 'hey, if you get around to implement signal handling, please give me a poke because i want to implement a proper module for handling it comfortably with perl6' would be nice 10:43
as, realistically speaking, i will not know of that fact within less than a month or six
10:43 iblechbot_ joined 10:44 iblechbot left
wayland76 RichiH: I'd say make a note to check back in about a year. S17 is getting no love at the moment, and until it does, it won't be happening 10:44
RichiH wayland76: so it will be ready for christmas? 10:45
weee!
:p
wayland76 That's the general idea :) 10:46
10:56 vamped left 10:57 Gothmog_ left, Gothmog_ joined
frettled :) 11:01
11:05 agentzh left 11:09 iblechbot_ left 11:13 payload left 11:14 payload joined
masak hah -- RT moves so slowly that the spec overtakes it sometimes. :) 11:16
has anyone ever figured out why RT ticket updates are so slow in reaching p6c?
jnthn morning 11:19
phenny jnthn: 03:28Z <colomon> tell jnthn Somehow all the code formatting had disappeared from your post. I just did my best to get it back, but you might want to double-check it to make sure I got it more or less the way you would like it.
jnthn wtf.
phenny: tell colomon Thanks for clearing that up, looks good. :-) 11:20
phenny jnthn: I'll pass that on when colomon is around.
jnthn thanks, phenny
11:20 pdcawley joined
masak jnthn: I didn't point it out at the time, but when reviewing the post I saw that you had used <code>...</code> around the code snippets. might have been related to that. 11:20
unlike <pre />, <code /> is free-form just like normal HTML, so newlines and many spaces all get collapsed down to one space in rendering. 11:21
moritz_ on perlmonks <code> preserves whitespaces
which adds to confusion
jnthn masak: oh, hmm 11:22
it looked fine in the preview, which is all the more odd.
Anyway, thanks for fixes and question ansewrs.
masak: Your reply on precedence isn't quite right - we are parsing it as a trait mod there. 11:23
masak oh.
please feel free to correct me in the comments.
jnthn the does in $x does Foo; (mix-in) and the does in class Foo does Bar { } are parsed differently.
First is an operator, second is trait.
Will post :-)
masak jnthn: but how would `does A, B` get parsed? 11:24
lunch & 11:27
11:27 masak left 11:28 IllvilJa joined
moritz_ phenny: tell masak `does` is tighter than comma, so that `my @a = 1 does False, 0 does True` DWYM 11:28
phenny moritz_: I'll pass that on when masak is around.
11:28 rodi joined 11:33 Baggio_ joined
jnthn std: role A { }; role B { }; class C does A, B { } 11:36
p6eval std 29372: ===SORRY!===␤Unable to parse class definition at /tmp/C7MPpi8XBb line 1:␤------> role A { }; role B { }; class C does A⏏, B { }␤ expecting any of:␤ trait␤ whitespace␤FAILED 00:01 106m␤
jnthn phenny: tell masak it doens't get parsed at the moment :)
phenny jnthn: I'll pass that on when masak is around.
11:40 colomon joined
jnthn ooh! new laptop has been shipped! 11:45
11:47 snearch left 11:52 payload left 12:13 colomon left 12:14 colomon joined 12:25 colomon left 12:26 colomon joined
moritz_ it will arrive just after you leave 12:30
moritz_ has just handed in his diploma thesis, and dances with joy
it's a bit hard to type while dancing
wayland76 congrats! :)
moritz_ thanks
wayland76 have to alternate then
Does that mean that you'll suddenly not have time to do P6 stuff any more? 12:31
moritz_ it means that I'll suddenly have more time to do P6 stuff, at least until I find a job :-)
wayland76 ...but you should be spending all that time on job searching :) 12:32
Su-Shee moritz_: no oral exam? 12:33
moritz_ Su-Shee: done them already 12:34
Su-Shee moritz_: then we made diplom the same day. :) it's nice that short before christmas.
moritz_ Su-Shee: this year? :-) 12:37
frettled moritz_: oh, congratulations, this is good news! 12:38
12:40 pnu joined
Su-Shee moritz_: haha, not even this decade :) 12:41
jnthn moritz_: Congrats! :-) 12:42
moritz_ \o/ 12:43
12:49 pmurias joined 12:51 meneldor joined, JimmyZ joined
jnthn is happy that his current post attracted some more discussion than his pervious one. :-) 13:07
Some replies later...lunch! 13:08
13:11 plainhao joined
takadonet morning all 13:15
13:15 JimmyZ left 13:16 JimmyZ joined
colomon moritz_: \o/ 13:19
phenny colomon: 11:20Z <jnthn> tell colomon Thanks for clearing that up, looks good. :-)
colomon jnthn: I wonder how it happened? The code samples looked great in your draft... 13:20
13:22 Wolfman2000 left
frettled colomon: <code>…</code> instead of using <pre>…</pre> 13:24
(or so masak suggested)
colomon ah. definitely was <code>...</code>. I switched them to [sourcecode][/sourcecode] and then re-indented. 13:25
frettled For my blog on blogspot, I use <pre>…</pre> for greater control.
I always edit in source mode, though.
13:26 cosimo joined 13:33 ssm left, parduncia left, constant left 13:34 constant joined, ssm joined, parduncia joined 13:37 hanekomu joined, masak joined
masak moritz_: yes, but... 13:37
phenny masak: 11:28Z <moritz_> tell masak `does` is tighter than comma, so that `my @a = 1 does False, 0 does True` DWYM
masak: 11:36Z <jnthn> tell masak it doens't get parsed at the moment :)
masak moritz_: ...that's what I answered in the comment, and jnthn said it wasn't applicable because the 'does' in the class decl isn't an operator. 13:38
jnthn masak: I repliedto your post. 13:42
masak jnthn: looks good. thanks.
I like the other answer even better. mostly because it's a slightly provocative question. :) 13:44
we need that. all this advent stuff has made outsiders too nice to us. :P
13:45 ssm left, parduncia left, constant left
jnthn masak: It's a fair enough quesiton. :-) 13:45
masak oh, definitely.
jnthn The bigger innovation is in the composition mechanism, whereas I think the asker was thinking more of the parametric part. 13:46
masak it was the combination of a knowledgeable question and the 'but isn't this just...' tone that I liked. :)
jnthn "but isn't that just..." is often a good question to ask anyway.
13:46 constant joined, ssm joined, parduncia joined
masak yes. 13:46
jnthn I often (at lesat mentally) ask it when I see new stuff.
OK, @house-chore for a bit...family are coming back tomorrow morning! 13:47
masak I think there's a class of people who prefer to learn that way. I'm one of them, too.
13:48 abra joined, ssm left, parduncia left, constant left, constant joined, ssm joined, parduncia joined 13:49 iblechbot joined 13:50 t4n4be joined
wayland76 Goodnight all :) 13:52
13:52 orafu left 13:53 orafu joined
JimmyZ goooooood evening. 13:54
14:00 rgrau left
masak moritz_++ # building parrot branches with Rakudo and sending emails when they fail 14:01
that way, I won't have to be the one to discover the build fail. :)
14:06 Guest81964 left 14:07 colomon_ joined, colomon left, colomon_ is now known as colomon
moritz_ hey 14:08
I just had a "great" idea
I could couple the IRC logs with karma tracking
i can haz 2.6M lines of IRC logs 14:09
meneldor Hello all, what i need to be able to test perl6 on my WinXP without Administrative rights (i cannot install anything on my pc) ?
moritz_ meneldor: I'm not a windows user, so I have no idea what works... 14:10
meneldor: but maybe the packages from sourceforge.net/projects/parrotwin32/ can be installed without admin rights
meneldor: or if you already have a compiler and perl installed, you can compile rakudo from source 14:11
meneldor im not windows user too but right now im on guest Win OS so i've got some limitations :)
[particle] strawberry perl is a compiler and perl
masak PerlJam: did you follow the docs/release_guide.pod when you released the book last month?
PerlJam would go with strawberry perl
masak: pretty sure I did. why? 14:12
14:12 gabiruh joined
masak PerlJam: well, I was curious about that months announcement... 14:12
[particle] download and install strawberry perl, then download rakudo and build from source
masak PerlJam: I find it here rakudo.org/node/60 but not in the repo itself. 14:13
PerlJam doh! I just just checked my local repo. I never committed it.
meneldor tnx i will try parrotwin32
masak PerlJam: I'll add that step to the release guide. :) 14:14
PerlJam committed and pushed 14:15
masak PerlJam++
meneldor dammit :) it wants to write to the registry and i dont have permissions
masak meneldor: I only know one Windows Rakudo user, and that's jnthn. pretty sure he has root, though. :) 14:16
meneldor if i was at home i could install it on my ubuntu but right now.... 14:17
jnthn I also tend to build from source.
masak :)
jnthn Which you can do without admin rights if the right tools are already installed.
Which is probably the sticking point. :-( 14:18
pmichaud good morning, #perl6
dalek ok: 680b5b8 | duff++ | docs/announce/2009-11:
belated checkin of Nov release announcements
colomon o/
dalek ok: a6ee18a | duff++ | src/preface.pod:
Merge branch 'master' of [email@hidden.address]
jnthn morning, pmichaud
masak mornin', pm. 14:19
meneldor hmm rakudo has been installed successfuly in My documents, but PATH is not set
colomon Tene: I'd add another word or two about .ast where you introduce it -- I mean, I just figured out how it works (I think), but it took me a few reads to get there. 14:20
14:20 patspam joined
pmichaud "<ws> is a builtin regex, like <alpha>, <digit>, and friends, that matches whitespace." <-- this phrasing always bugs me a little 14:22
PerlJam heh, I was just reading that very bit 14:23
masak it's wrong.
pmichaud "postfix:<>" and "postfix:[]" are also a bit on the wrongish side. I'd eliminate the colons. 14:24
14:25 rfordinal joined
masak ...and replace them with spaces... 14:25
pmichaud yes
thanks
(I meant that but didn't say it)
masak :) 14:26
meneldor what means: error:imcc:syntax error, unexpected IDENTIFIER, expecting $end ('my')
pmichaud it also shouldn't be necessary to instantiate an actions object
moritz_ meneldor: what did you do to get that error message? 14:27
pmichaud Tene++ # *excellent* explanation of why .comb is better than .split
PerlJam wishes those {*} weren't there 14:28
pmichaud yeah, this is a case where the ng branch would've been a lot more useful
PerlJam yep
pmichaud sorry about that, folks :-| 14:29
meneldor path\to\parrot.exe path\to\perl6_test_code.pl
PerlJam meneldor: show the code :)
jnthn meneldor: Ah, that's why. :-)
meneldor: You need to run perl6.exe
Not parrot.exe 14:30
meneldor ah i see :)
dalek ok: c40a1dd | masak++ | docs/release-guide.pod:
[docs/release-guide.pod] instructions to push announcement
meneldor but what parrot.exe does then?
jnthn Parrot expects a Parrot assembly file or bytecode file.
meneldor tnx
PerlJam meneldor: Parrot is the underlying VM that Rakudo runs upon. (still, show us your perl 6 code :) 14:31
PerlJam is not multi-tasking very well this morning
14:31 payload joined
meneldor im a bit confused with all these words rakudo, parrot, pugs etc :) 14:31
im not sure what to use to be able to test perl 6 code
PerlJam Rakudo is an implementation of Perl 6. 14:32
Rakudo executes on top of Parrot
masak meneldor++ # confused but persistent!
PerlJam pugs is an obsolete, unmaintained implementation of Perl 6 that uses Haskell instead of Parrot
meneldor now im more confused :D
moritz_ meneldor: I forgot to mention, you also need the perl6 (or rakudo) addon from the parrotwin32 page 14:33
setup-parrot-1.9.0-rakudo-24.exe
meneldor where is the old #!/usr/bin/perl :D
PerlJam meneldor: get used to it! I always feel confused myself. I think it's a natural state on #perl6 :)
meneldor ok tnx moritz ill try it
moritz_ and then there'll be a perl6.exe
which you can use to run your Perl 6 programs 14:34
meneldor im perl5 fan from 7 years but i did not try perl 6
[particle] we forgive you.
meneldor :D tnx for that
14:36 mikehh left, mikehh joined 14:41 t4n4be left
dalek ok: 69a026a | masak++ | docs/announce/2009-12:
[docs/announce/2009-12] created announcement
 14:42
14:45 rfordinal left
meneldor setup-parrot-1.9.0-rakudo-24.exe wants admin right to be installed :( 14:46
moritz_ :( 14:47
meneldor ok ill install it on my ubuntu later
14:47 Chillance left
meneldor thanks anyway moritz_ 14:47
PerlJam meneldor++ tenacity is an excellent trait.
PerlJam has discovered that Hershey's dark chocolate goes quite well with Dr Pepper 14:48
14:49 rgrau joined 14:52 anonperler joined
jnthn multi trait_mod:<tenacity>($amount) { ... } 14:52
eiro cd 14:54
sorry
anonperler Hi. Just wanted to say thanks for making Rakudo self contained and easy to set up. Just built it using the instructions on day 1 of the advent calendar, and even after `make install` the whole thing is still nicely contained in its own `rakudo` directory. 14:55
Makes upgrading later on very easy (that is, `rm -fr rakudo` and then just download and build again)3.
pmichaud anonperler: you're welcome :-)
PerlJam anonperler: glad you like it.
pmichaud anonperler: glad to hear it worked out for you
moritz_ anonperler: thanks for telling us! Usually people only come here when they have problems with the build :-)
BinGOs shame rakudo ate all my memory when I tried to build it 14:56
anonperler You're welcome. Great stuff!
PerlJam BinGOs: on feather?
BinGOs On my NetBSD 3.1 box.
14:57 anonperler left
BinGOs rather parrot tried to take all my system memory 14:57
14:57 drbean joined
BinGOs Hostname: canker.bingosnet.co.uk - OS: NetBSD 3.1/i386 - CPU: 2 x Intel Pentium III (686-class), 897.32 MHz, id 0x68a - Processes: 53 - Uptime: 376d 2h 39m - Load Average: 0.24 - Memory Usage: 66.60MB/255.00MB (26.12%) - Disk Usage: 22.88GB/109.56GB (20.89%) 14:58
pmichaud 256MB ?
BinGOs ( Only one processor incidently ).
Yes, 256MB RAM
pmichaud yeah, that's a little on the smallish side for building Rakudo at the moment. 14:59
BinGOs oh meh.
I build blead on here all the time. but meh.
PerlJam Has anyone ever figured out what the memory requirements are t obuild rakudo?
pmichaud PerlJam: not yet 15:00
BinGOs Definitely more than 256MB then >:)
pmichaud I think 512MB is generally enough
BinGOs Well ....
I tried on my Ubuntu box as well ...
PerlJam pmichaud: maybe. maybe not. I'm not so sure.
BinGOs seemed to kill that as well ... but I am running vmware server on there as well.
pmichaud BinGOs: what step eats up all of the memory?
BinGOs after parrot is built, running 'make', something to do with a pmc? 15:01
pmichaud ...with a pmc? that would be odd, normally those are pretty lightweight. 15:02
I would've expected it to eat up all of the memory when compling the setting.
BinGOs actually, during the perl6.pir stage I think
15:02 payload left
pmichaud that also is supposed to be light-ish 15:02
BinGOs I am not running it again to find out exactly I'm afraid. 15:03
pmichaud no problem, I can set up a vm to try it out :)
BinGOs I do have a lot of stuff going on on this box, database server, various bots, hourly rsyncs of CPAN, etc. 15:04
BinGOs goes back to lurking. 15:05
frettled BinGOs: You have my permission to upgrade to a post-2000 computer. ;) 15:06
BinGOs Oh hilarious. 15:07
This little box was a). free b). quiet enough that I can have it on all the time without annoying the family. 15:08
15:09 drbean_ left
BinGOs the other little box has only 128MB of RAM but handles all my email server needs. 15:09
PerlJam quiet boxes should never be underestimated.
I used to have a little linux server with a fan that occasionally rattled. It kept waking us up at night. 15:10
15:11 Tene left 15:12 Chillance joined, Tene joined 15:16 ash___ joined 15:23 rfordinal joined 15:24 mj41 left
ash___ does anyone else get an error when they run rakudo ng branch with --trace=1 ? 15:24
masak new, fresh book release: rakudo.org/node/62 # moritz_++
moritz_ masak++ for doing actual release 15:25
ash___ masak++ moritz_++ *adds to my to read list*
mathw everyone++ 15:26
masak nobody-- 15:27
ash___ hmmm rakudo ng with --trace=3 is very random... it works one time, then bus error, then seg fault, then bus error, etc. thats odd
masak pmichaud: I changed the ticket to succeed/proceed, but I forgot to cc p6c.
pmichaud masak: ah. 15:28
masak s/changed/already changed/
pmichaud yeah, I just went from the mailing list.
anyway, masak++
JimmyZ ng: (1..999).grep({$_ % 3 == 0}).say 15:29
p6eval ng b9eec4: Method 'grep' not found for invocant of class 'Range'␤current instr.: '_block14' pc 29 (EVAL_1:0)␤
JimmyZ rakudo:(1..999).grep({$_ % 3 == 0}).say
15:29 rfordinal left
PerlJam ng still doesn't do ranges, does it? 15:30
JimmyZ so slow!
colomon ash___: rakudo ng is like that without trace at the moment, too. I've got S32-lists/end.t 1/3 working here -- except that's 100% working 1/3 of the time, and crashing during the second test the other 2/3rds of the time.
PerlJam: no real ranges.
frettled BinGOs: There's nothing wrong with free hardware. If you'd been living in Norway, I'd have a _bunch_ of old hardware (P4 and newer) that you could get for free, even some dual Xeon thingies. 15:31
colomon PerlJam: series operator works for N1 ... N2
masak JimmyZ: I believe you need a space after 'rakudo:'
15:32 mj41 joined
JimmyZ rakudo:(1..999).grep({$_ % 3 == 0}).say # why is it so slow? 15:32
real0m7.866s
15:33 rfordinal3643 joined
colomon ng: (1...999).grep({$_ % 3 == 0}).say 15:33
p6eval ng b9eec4: ( no output )
JimmyZ rakudo: (1..999).grep({$_ % 3 == 0}).say # why is it so slow?
p6eval rakudo 7f1c3f: ( no output )
colomon ng: (1...9).grep({$_ % 3 == 0}).say
p6eval ng b9eec4: Method 'grep' not found for invocant of class 'Array'␤current instr.: '_block14' pc 29 (EVAL_1:0)␤
15:33 FalseVinylShrub joined 15:34 frettled sets mode: +oo colomon masak
colomon yeah, that's what I thought.... 15:34
diakopter rakudo: (1..9).grep({$_ % 3 == 0}).say
p6eval rakudo 7f1c3f: 369␤
diakopter rakudo: (1..29).grep({$_ % 3 == 0}).say
p6eval rakudo 7f1c3f: 369121518212427␤
diakopter rakudo: (1..59).grep({$_ % 3 == 0}).say
p6eval rakudo 7f1c3f: 36912151821242730333639424548515457␤
diakopter rakudo: (1..99).grep({$_ % 3 == 0}).say
p6eval rakudo 7f1c3f: 369121518212427303336394245485154576063666972757881848790939699␤ 15:35
JimmyZ rakudo: ([+] (1..999).grep({$_ % 3 == 0})).say
p6eval rakudo 7f1c3f: ( no output )
15:35 chromatic joined
diakopter timed out 15:35
rakudo: (1..199).grep({$_ % 3 == 0}).say
15:35 frettled sets mode: +o diakopter
p6eval rakudo 7f1c3f: 369121518212427303336394245485154576063666972757881848790939699102105108111114117120123126129132135138141144147150153156159162165168171174177180183186189192195198␤ 15:35
15:35 Patterner left
diakopter rakudo: ([+] (1..299).grep({$_ % 3 == 0})).say 15:35
p6eval rakudo 7f1c3f: 14850␤ 15:36
diakopter (didn't time out)
rakudo: ([+] (1..499).grep({$_ % 3 == 0})).say
p6eval rakudo 7f1c3f: 41583␤
diakopter (didn't time out)
masak who wants to do the January release of the book?
diakopter rakudo: ([+] (1..699).grep({$_ % 3 == 0})).say
p6eval rakudo 7f1c3f: 81783␤
JimmyZ time ./perl6 -e '([+] (1..999).grep({$_ % 3 == 0})).say' 15:37
real0m29.449s
user0m19.497s
sys0m0.300s
pmichaud JimmyZ: we know that there are some very slow parts of Parrot and Rakudo, especially when creating large numbers of objects 15:38
JimmyZ: these are being worked on
JimmyZ: if you want a more detailed answer as to why this particular case is so slow, I don't know the answer right now.
I suspect it has something to do with GC, though. 15:39
JimmyZ pmichaud: I see.
15:40 Psyche^ joined, Psyche^ is now known as Patterner 15:41 rfordinal3643 left
diakopter masak: consider your question bumped :) 15:41
pmichaud I'll do the jan book release, if nobody else takes it :) 15:42
masak diakopter: which question?
ash___ isn't there a GC updating in parrot around the corner?
diakopter masak: lol
masak ah, the release question. :)
moritz_ yes, but we don't know how cles the corner is :-)
pmichaud ash___: I don't know about "around the corner".
Parrot recognizes that gc is currently a huge problem. But I'm not sure who will step up to work on it. 15:43
chromatic Why, whiteknight and myself, of course.
pmichaud +1 to that!
moritz_ chromatic++ and whiteknight++ for that
diakopter <<64 to that! 15:44
chromatic: are you considering diving right into generational? 15:45
ash___ pmichaud: well, i have seen people talking about it in the mailing list is what i mean, its a known issue, see ( trac.parrot.org/parrot/wiki/Develop...Priorities ) since it specifies parrot 2.0, and i think thats the next scheduled release, but then again i could be mistaken on the a number of area's so take that with a grain of salt
pmurias do any perl6 benchmarks exist? 15:46
diakopter dunno, but it would be nice if p6eval could report elapsed millis if you, say, appended a _b to the run target name 15:47
chromatic I doubt we'll go straight to generational. We have an idea which could double the speed of the current system; it's a sweep-free, garbage-friendly system.
pmichaud ash___: that page was a working/planning document 15:48
chromatic The idea is to find the most minimal change.
diakopter quips flippantly: minimal changes lead to, well, local maxima 15:49
ash___ pmichaud: see, i told you i am not a reliable source of whats going on in parrot (or perl)
pmichaud ash___: anyway, since the Parrot 2.0 release is 33 days away, I'm not sure there will be a new gc by 2.0 :)
pmurias diakopter: the time for snippets in the evalbot would be dominated by compile/startup time 15:50
ash___ pmichaud: yeah, there is that whole time constraint on things, you never have enough it seems, in my experience anyway
pmichaud ash___: yeah, I was going to do a study on that topic, but I ran out of time.
chromatic If we can improve freeze/thaw and constant PBCs, startup time may improve measurably for 2.0. 15:51
pmichaud chromatic: I tried switching exception handlers to constant PBCs in nqp-rx, and ran into serious FAIL
chromatic There's a good possibility that a current Parrot branch can provide a 20% runtime improvement.
diakopter pmurias: ok.. but baselines for the startup times could be subtracted away
chromatic I'd like to see the bug report there (I don't remember if you filed one).
We may have to evaluate the static-scope exception handling idea sooner. 15:52
pmichaud The :immediate subs remained subs -- they didn't become ExceptionHandler arguments
diakopter pmurias: (for comparing on the same impl)
pmichaud s/arguments/objects
chromatic Okay, I remember that ticket. I'll work on that test.
pmichaud I couldn't reproduce it down to a small test case, though.
the small test cases I came up with all worked fine.
I'm going to go ahead an implement the caching strategy in past for exception handlers, that will give us a sizable win to begin with. when constants in pbcs become available, it'll be a trivial change from there. 15:53
15:53 jferrero joined
pmichaud I'm about to file some messages noting that parrot-nqp doesn't run in the context_unify3 branch, nor can I build rakudo-ng nor nqp-rx 15:53
diakopter pmurias: but including the startup/compile time is good for comparing across impl 15:54
moritz_ pmichaud: nor can I even install context_unify3
pmichaud moritz_: hmm, that part seems to work for me. What did you use as the prefix argument ?
I just discovered this morning that --prefix to Parrot's configure.pl requires a full path name
moritz_ pmichaud: --prefix=/home/moritz/rakudo/parrot_install/ 15:55
pmichaud hmmm
I dunno then. 'make install' worked for me once I got the arguments right.
(fsvo "right")
moritz_ pmichaud: I think it's actually a build problem
pmichaud it could be. 15:56
moritz_ pmichaud: that was revealed during 'make install'
pmichaud: I ususally run parallel builds, which makes it a bit harder to see if the build failed
chromatic Coke had trouble with context_unify3. I think it's a modest change to his PMCs. 15:57
pmichaud well, since nqp-rx doesn't have custom pmcs, it must be something else happening there.
chromatic Interesting. We'll take a look. 15:58
jnthn fwiw, once we do get to the point of being able to try building Rakudo on top of that branch, I expect there will be difficulties.
Well
pmichaud and I'm a bit concerned about how it will impact rakudo, because rakudo has a custom dispatcher
jnthn there may be.
pmichaud so it may be a bit more involved than partcl :)
jnthn I fear we rely on contexts and call info being separate.
15:58 abra left
jnthn I brought this up when it was being discussed, and said I had no objects *provided* somebody from the Parrot team could work out how to patch Rakudo up to work with the changes. 15:59
pmichaud jnthn: rely on the design having them separate, or just the implementation?
jnthn pmichaud: I know that we, for example, rely on being able to do multiple invocations with the same callsig.
I'm not sure how that will work out if contexts and callsigs are merged. 16:00
It may or may not be a problem - I'm not familiar with what the branch has done. 16:01
pmichaud me neither.
jnthn Thus why I'd rather somebody who is write the patch.
chromatic Multiple invocations with the same call sig, how does that look? 16:02
jnthn chromatic: perl6multisub.pmc does it, iirc 16:04
chromatic: checking...
(will get you a line #)
chromatic: in master, starting line 448 16:05
In ng I think this code has changed relatively little.
ash___ ng: role A { }; say A.^methods; # is this a known issue? I can try to look into it more (although that seems like parrot stuff that i am not very familiar with) 16:06
p6eval ng b9eec4: No such attribute 'parrotclass'␤current instr.: 'perl6;RoleHOW;methods' pc 4835 (src/metamodel/RoleHOW.pir:47)␤
jnthn chromatic: It's largely similar.
ash___: I suspect just a tad incomplete. 16:07
chromatic Suppose you had a way both to clear a Context/Sig of its return values (returning them or gathering the returns) and a way to reuse a Context/Sig for a subsequent call.
Would that suffice?
16:07 JimmyZ left
jnthn chromatic: I *think* that might be enough. 16:08
chromatic: See the comment starting XXX for what I had in mind for a future change. 16:09
But I don't think that'll be a problem either.
ash___ jnthn: classes work, so i figured it was something like
jnthn ash___: Yeah, Parrot roles and Rakudo roles are having an ongoing divorce... 16:10
16:10 nihiliad joined
jnthn ash___: The only thing that the Parrot role PMC is really used for now is (1) storing the methods and (b) making punning work. 16:11
diakopter chromatic: would that break non-tailcall recursive invocations? 16:14
chromatic I'm not sure. 16:15
We might have to enable cloning.
(if it doesn't already work)
back in a while 16:16
diakopter (which would have the same (but less efficient) effect as keeping them (callsig/context) non-merged) 16:17
16:23 gbacon joined, rgrau left, justatheory joined 16:27 meneldor left 16:30 FalseVinylShrub left 16:34 colomon_ joined, colomon left, colomon_ is now known as colomon, chromatic left
ash___ nobreak is now proceed? 16:36
PerlJam aye
break is now succeed.
masak \o/
PerlJam and I like it!
jnthn nobreak I didn't like. I like succeed much more :-) 16:37
masak who doesn't? you can't dislike 'succeed'! ;)
PerlJam eliminates the break/continue confusion, make sense, and uses words of the same length with same suffix (for mnemonic help)
ash___ yeah, nobreak was confusing
PerlJam and having when block succeed is far more appealing than having them break ;) 16:38
16:38 jferrero left
masak is conspiring with mst again 16:39
16:39 jferrero joined
mdxi (they have the same *root*, not the same *suffix* :) ) 16:39
PerlJam whatever, I'm not a linguist, nor do I play one on IRC :) 16:40
16:40 Baggio_ left
ash___ for try's is there a finally mechanism? 16:41
masak there's the LEAVE phaser. 16:42
jnthn LEAVE iirc
But you can use it in any block.
Just as with CATCH.
[particle] why isn't break now secede?
masak concede, recede...
[particle] or maybe that should be leave....
ash___ isn't there already a leave? 16:43
16:43 wlod joined
[particle] aye. but maybe it should be called secede :) 16:43
16:45 wlod left 16:49 Tene left
ash___ i think succeed and proceed make sense, given $letter { when "A" { say 'its an "A"'; proceed } when "Ã" { say "its also unicode"; proceed } }; 16:50
proceed makes sense
moritz_ but "A" ne "Ã" :-) 16:51
ash___ well i never said it made sense
moritz_ 17:50 < ash___> i think succeed and proceed make sense 16:53
*SCNR*
16:55 Tene joined
mdxi do_the_next_chunk_too_i_reckon is kinda wordy, but hey, maybe dangerous things should be huffman coded that way :) 16:56
moritz_ maybe :-)
pmurias who killed svnbot? 17:04
masak some ardent committer, perhaps? 17:05
moritz_ can't remember which host it is supposed to run on 17:06
feather3?
does anyone remember its host mask?
diakopter heh 17:07
pugs_svn you mean? 17:08
pmurias propably
diakopter yes, feather3 17:09
pugs_svn [n=SVNBot@feather3.perl6.nl] has joined #perl6
17:11 pugs_svn joined 17:12 pugs_svn left, pugs_svn joined
diakopter pugs_svn: owe high 17:15
pmichaud k 17:16
ww
moritz_ mmanges too many bots on too many hosts with too low truck number
17:17 colomon left
PerlJam moritz_: you just need a family of bots to manage themselves 17:17
pmichaud sounds like it could be a new sitcom
"The Bots!"
moritz_ PerlJam: no, I need more co-admins 17:18
17:19 _jaldhar left
masak pmichaud: "Join us next week for another episode of botty humour!" 17:19
17:19 SmokeMachine joined
TimToady you guys have bots in the belfrey 17:20
moritz_ wonders of botty humour involves autopunning
diakopter our dingbots are dingy
pmichaud <mommybot> I don't know what to do.... lambdabot is always speaking out of turn!
<daddybot> Yes, but at least lambdabot isn't as annoying as purl! 17:21
diakopter ..and purl has Tourette's
Juerd moritz_: If the truck number is low, at least document what's there and how to restart them
masak <phenny> Why does no-one ever notice what I'm doing? Because I'm quiet, that's why!
pmichaud <hugme> Anyone need a hug? 17:22
masak *lol*
moritz_ hugme: hug phenny
hugme hugs phenny
masak :D
TimToady phenny: hug hugme
moritz_ phenny: tell hugme hug me 17:23
phenny moritz_: I'll pass that on when hugme is around.
moritz_ hugme: hug phenny
hugme hugs phenny
phenny hugme: 17:23Z <moritz_> tell hugme hug me
masak almost...
moritz_ too bad it didn't recurse :-)
masak that can be arranged...
PerlJam TimToady++ for succeed/proceed. that was an excellent change. 17:24
diakopter phenny: tell phenny tell phenny tell phenny
phenny Hey, I'm not as stupid as Monty you know!
PerlJam TimToady: not so sure about "so" though :)
masak PerlJam: ssh! it's better than 'true'!
PerlJam It's *shorter* than 'true'
TimToady I agree that it's so so... 17:25
diakopter but not as short as !!
PerlJam I'm not quite sure what criteria make it better
TimToady the problem was the existence of true, not the absence of so
PerlJam diakopter: eh? 'so' is shorter than '!!'
pmichaud I was thinking perhaps we need "iznot" and "izso"
diakopter std: so 3; !!3
p6eval std 29373: ok 00:01 107m␤
pmichaud "iznot!" "izso!"
diakopter PerlJam: shorter to type, ok, but longer on the screen 17:26
jnthn std: so WHAT
p6eval std 29373: ok 00:01 105m␤
PerlJam diakopter: okay, I'll grant you that.
diakopter what you want is an abbrev for 'whether' 17:27
PerlJam diakopter: but the op still needs to be wordy to be a 'not' parallel
pmichaud "really"
if really 3 + 4 ....
PerlJam huffman defenestration. 17:28
pmichaud :)
jnthn if its 3 + 4 ...
pmichaud there's always "NOWAI" and "WAI"
jnthn uh, it's :-)
Which we can have. ;-)
masak 'truely', as in 'Im truely sorry for their lots'... 17:29
pmichaud "truly" isn't too bad :)
if truly $a == 5 { ... }
diakopter "verily"
masak if truly $sorry { for $their.lots { ... } } 17:30
pmichaud @their, probably
masak oh, dang. :)
if truly $sorry { for @their.lots { ... } }
PerlJam I would think @lots would make more sense. 17:31
diakopter "iff" kinda has the "whether" connotation
pmichaud my @result = if truly .sorry for @their>>.lots;
PerlJam diakopter: I like that.
diakopter: but it's not useful elsewhere.
dalek kudo/ng: dd78a40 | jnthn++ | src/builtins/ (2 files):
Code does Callable, List does Positional.
kudo/ng: 041178b | jnthn++ | src/ (2 files):
Couple of role related fixes.
diakopter (and "iff" also is the direct complement to "not")
dalek kudo/ng: 21558a9 | jnthn++ | src/Perl6/Compiler/Signature.pm:
Re-enable sigil type checks in parameter lists.
pmichaud "iff" sounds two-way-ish-for-me
diakopter well
when used in a unary sense... like I said, 'kinda'
TimToady iff is spelled ?& in Perl 6
jnthn So...does ng do hashes yet...
arnsholt iff is eq for logicians =)
pmichaud my mind keeps wondering if there's a latin root or word here 17:33
PerlJam forget latin ... what about other languages? :)
zaslon lolmasakhazblogged! masak++ 'The wish, the tuning fork, and the one true heir': use.perl.org/~masak/journal/40037?from=rss
17:33 mubot left
diakopter not vs. ver 17:33
ver, for verily, verify 17:34
pmichaud alas, "ver" tends to think "version"
arnsholt I can't quite make out from the scrollback, but what are you looking for a word for?
PerlJam or "fi" for "fidelis" ;>
pmichaud arnsholt: replacement for "true"/"so"
i.e., the logical opposite of "not"
diakopter arnsholt: trying to find a better word for the combination of "bikeshed" and "yakshave" and "elephant.ignore" 17:35
pugs_svn r29374 | pmurias++ | [mildew-js] add --profile option to print out the time execution and
r29374 | loading the setting took
arnsholt Well, truth in latin is veritas
PerlJam yeah, I was going more for "spirit" than "letter"
pmichaud and truth in government is Nil. Or Mu. 17:36
arnsholt PerlJam: Well, my vocabulary digs out letter before spirit
pmichaud if vino ....
arnsholt But the logical opposite of not just nothing, no?
pmichaud arnsholt: we'd like a low-precedence form of prefix:<?> 17:37
"not" is the low-precedence form of prefix:<!>
jnthn mmm...vino :-)
17:38 Chillance left
arnsholt A right, now I see 17:38
Juerd pmichaud: ita 17:39
arnsholt Some kind of copula might be appropriate
Juerd pmichaud: (latin)
arnsholt "is" is taken I believe, but since someone mentioned Latin, "est" is short and means the same
PerlJam Hmmm "ta"
Juerd arnsholt: "est" declares equivalence, not being so. 17:40
arnsholt In a declarative sentence, yes
But boolean context is essentially question
"Is this expression true?"
Juerd false est false.
pmichaud arnsholt: the point is to be able to get a boolean value when not in a boolean context
Juerd I'd see "est" as something inherently binary. For me, it doesn't work as some prefix thing (lowprec unary) 17:41
ita $foo eq $bar works
arnsholt True. My analogy is essentially from Prolog
Juerd But I don't think it'll ring many bells.
arnsholt Yeah, there's that
17:41 mariuz left
pmichaud "ton" 17:41
(not backwards :-) 17:42
Juerd One positive thing about ita is that it is as long as not.
pmichaud: Hmmmmm
masak 'indeed'
pmichaud "notnot"
arnsholt ton might be the best suggestion yet
PerlJam most of these "cute" ones will require far more explanation than should be warranted IMHO
ergo "so" is still the winner
Juerd masak: indeed looks weird as a contrast to negation. if (not ...) { ... } ... if (indeed ...) { ... }
pmichaud PerlJam: if you say so
PerlJam IMHO again
arnsholt Juerd: You wouldn't need it in a conditional though 17:43
Juerd PerlJam: Agreed
masak Juerd: also, you don't need parens. :)
arnsholt The if forces boolean context
Juerd arnsholt: Would for stress, as documentation
arnsholt True
Juerd Dutch has a word for this and it's used all the time.
arnsholt What is it?
Juerd It's hard to explain the value of having such a word :)
"wel"
masak 'akshually'
Juerd It's the opposite of "niet" (not) 17:44
forum.wordreference.com/showthread.php?t=148276
PerlJam google translate doesn't think that :)
Juerd "It is not blue" "Wel!" (Often as "welles!" or "jawel!") 17:45
In English you wuold say "Yes it is"
Or "But it is"
pmichaud or "Is so!"
Juerd Yes.
PerlJam I've seen "jawel" before.
pmichaud or "Is too!"
Juerd But it's harder to explain how "wel" is used in the context of there not being context :) 17:46
"I've decided. I will *wel* go to that party."
"I will so go" would imply enthusiasm, I think.
Tene 07:22 <@pmichaud> "<ws> is a builtin regex, like <alpha>, <digit>, and friends, that matches whitespace." <-- this phrasing always bugs me a little 17:47
PerlJam "wel" seems to capture the emphasis nicely, but not "truthiness" (but maybe I'm bent)
Tene pmichaud: any recommendations on something better to day there?
Juerd To express the same thing in English I think you'd say "I am going to that party", with "am" stressed.
pmichaud I'd leave the <ws> out of the example entirely
Juerd PerlJam: "not" doesn't capture falsiness.
pmichaud just use \s*
PerlJam Juerd: I guess you're right about that too 17:48
Juerd I'd almost suggest "bool" but it'd look too much like typecasting. (Which it is...) 17:49
arnsholt Or truth
17:49 cdarroch joined
arnsholt But truth is long 17:50
Juerd "truth" is ugly
pmichaud afk (kid pickup)
Juerd my $foo = truth ...;
if not ... { ... } ...; if truth ... { ... }
PerlJam Juerd: and that's why american prefer beauty to truth ;)
Juerd truth wants the opposite "lie". 17:51
if lie ... { ... } ...; if truth ... { ... }
That looks better :)
arnsholt Well, "lie" is the same length as "not" =)
Juerd Yes, but we already had a not.
I think we really shouldn't rename that one :) 17:52
arnsholt Indeed
But I've got to pop out for a bit as well
PerlJam maybe we should
if true ... { ... } and if untrue ... { ... }
:)
is has the right relative huffmanization :)
TimToady you forget why we killed true in the first place 17:53
PerlJam No, I'm just twiddling knobs I haven't twiddled before in hopes that something useful will come out.
masak if concede ... { ... }; if deny ... { ... } 17:54
Juerd is happy with "so"
It's as short as !! :)
If you'd translate "wel" via Esperanto, you'd end up with "yes" 17:55
"Mi ja faros tion" - "I will *yes* do that" 17:56
Er, s/ja/jes/
frettled My goodness, Perl 6 changed again while I blinked! ;)
Juerd Which is different from "Jes, mi faros tion" - "Yes, I will do that"
frettled What about "so true" and "make it so"?
Juerd frettled: It's just renaming fortunately :) 17:57
No big operator shifting
frettled Juerd: Yeah, but each change kills a baby COBOL programmer! :D
Juerd thus $foo == $bar
masak just found this: www.perlmonks.org/?node_id=663566 17:59
17:59 nihiliad left
masak to what extent can today's Perl 6 be said to be statically parseable. 17:59
I don't recognize any of the problems merlyn brings up. 18:00
TimToady masak: it's statically parseable to the extent that STD does it. :) 18:01
since STD can't run BEGIN blocks
masak I should have expected that answer. :)
but what does not being able to run BEGIN blocks mean, in terms of not being able to statically parse Perl 6?
TimToady but it does fail to parse 6 .t files currently, and some of those depend on dynamic code
masak ok. 18:02
is there some sense in which static parsing of Perl 6 is 'better' than static parsing of Perl 5?
18:02 colomon joined
masak I mean, apart from the greatly cleaned-up grammar. 18:02
18:04 nihiliad joined
Tene masak: also, you need to run BEGIN to get macros. 18:04
masak hm, yes. 18:05
TimToady some of the failures are also failure to recognize 'use v5;' which is arguably not statically parseable :)
well, the 'use v5;' is statically parseable, but what comes after...
moritz_ is just as statically parseable as Perl 6 18:06
Tene also some traits
I made the changes pmichaud recommended. Did anybody else have changes that I didn't notice in the log?
TimToady std: so quietly use v5; 18:07
p6eval std 29374: ok 00:01 107m␤
PerlJam Tene: I thought the section where you introduce the actions could use some more explanatory prose, but since it wasn't a strong feeling, I didn't mention it earlier. 18:08
colomon Tene: I suggested a bit more explanation on .ast when it is introduced.
PerlJam Tene: I only mention it now in case you want to "polish" something, that would be my vote.
colomon (based totally on my own initial confusion reading that section. :)
Tene hmm. My feeling there was that it was better to get into the example ASAP, and then discuss it more in review. 18:09
I'll definitely re-work that section, though.
thanks.
frettled masak++ - excellent entry. 18:10
masak thanks. mst++ too.
Tene What did masak write where? 18:11
masak use.perl.org/~masak/journal/40037
Tene Yes, agreed. 18:12
Tene afk work 18:13
masak Tene: did you consider replacing the instances of $_<something><something> with .<something><something>?
it's slightly easier on the eye.
PerlJam masak: I am reminded of how love isn't a zero-sum game. The more you give, the more you create. We lovers of Perl are spreading the love :) 18:14
Tene I did. I expected explicit $_ to be a bit easier to understand. There's already a lof of new things in that post, so I tried to keep it down in a couple of ways.
PerlJam masak: (the "rule the kingdom with twice as many resources" is what triggered that thought) 18:15
masak PerlJam: yes, 'zero-sum game' was a term that came up in my discussion with mst today.
Tene PerlJam: I very much agree, and I have some very strong opinions on exactly that topic.
masak PerlJam: and that phrase was my re-wording of same. :)
18:15 ChanServ sets mode: +o Tene
masak Tene: fair enough. great post! 18:15
Tene all the thanks should go to procrastination; I couldn't have done it without her. 18:16
PerlJam Tene++
(and masak++ and mst++ for continuing the conversation between the two halves of the Perl family)
masak heh. it's not like I have any choice. :P 18:17
Tene successfully surpressed the urge to discuss political topics; really afk driving to work now.
moritz_ just updated the Whatever post and scheduled it for tomorrow 18:19
colomon \o/ 18:22
18:23 justatheory left
lisppaste3 colomon pasted "Benchmarking Mandelbrot" at paste.lisp.org/display/92324 18:25
mathw masak++ # blog post, peacemaking, hugging 18:26
pmichaud something bugs me about "save for a few CPAN modules in the Perl6:: namespace, nothing usable had trickled back into the Perl community from the Perl 6 crowd."
I don't think that's quite true, depending on the timeframes.
18:27 supernovus joined
masak pmichaud: hm, that's true. 18:27
pmichaud The biggest example being, of course, Moose.
18:27 mberends joined
masak when I wrote that, I was thinking circa 2004. 18:27
pmichaud but quite a few other ideas trickled back into Perl 5 from Perl 6.
masak mst wouldn't give an exact date for the birth of Modernlightennaissance Perl. 18:28
pmichaud circa 2004 is fine, but it should probably be explicit. did any of Perl Rennaissance, Modern Perl, or Enlightened Perl really begin in 2004, though? 18:29
masak no.
but the year is significant in that that was when I got interested in Perl 6, and mst had already lost hope.
masak adds 'circa 2004' to the post 18:30
pmurias mberends: mildew doesn't need an old version of itself to compile it's setting
18:30 crythias joined
mberends pmurias: are you leaving the m0ld behind completely? 18:31
supernovus has anyone has anyone reported that the 2009-12 release of the Perl 6 book is missing the new chapters that the blog post talks about? The Makefile hasn't been updated to include them.
pmichaud speaking from a strictly personal perspective, the fork wasn't nearly as split or bifurcated as this post seems to make it. But perhaps I'm a special case.
masak supernovus: oh dear. 18:32
supernovus: thanks for the heads-up.
pmichaud I agree that to the vast majority of Perl 5 folks, Perl 6 was at best a dream, it certainly didn't have a significant positive impact on their reality
masak moritz_: hm, that didn't occur to me at all :/
PerlJam pmichaud: you're speaking in past tense.
colomon pmichaud: My impression is that the "fork" has mostly occurred as Rakudo actually gets close to producing something?
moritz_ masak: I didn't include it because I wanted to discuss some options first
pmurias mberends: mildew is still using m0ld 18:33
moritz_ masak: but so far my attempts to discuss it haven't had any resonance
masak: and then I forgot
masak oh, ok.
pmurias mberends: having trouble with getting cabal?
masak let's leave it as it is, then.
and include it in the next month's release.
pmichaud PerlJam: yes, I am speaking in past tense, because the paragraph is "circa 2004"
colomon That is to say, for a long time there Perl 6 was sort of Perl 5's wacky uncle with the crazy ideas, and Perl 5 took the better ideas and ignored everything else. 18:34
PerlJam okay.
masak pmichaud: the conflict grows with the strength of the Perl 6 community.
pmichaud masak: that's a reasonable point
colomon But with Perl 6 starting to look like it might actually produce results, suddenly it's a threat rather than a curiosity.
threat to the status quo, that is.
pmichaud I'm just not sure I want to present history as being as bifurcated as this sounds. it didn't seem that way to me.
PerlJam colomon: Hmm. 18:35
pmichaud but again, I'm a minority case.
because I'm well on the inside looking out
masak pmichaud: I didn't see it that way until I started spending time with mst.
there's a whole other world out there! :P
they call themselves Perl 5 people, and their hats look funny. 18:36
pmichaud well, if the "fork" views makes all of this more palatable for the p5 side of things, I'm willing to go along wit hthat
*view
masak <mst> absolutely nothing in there to complain about
<mst> I think we're getting better at this :) 18:37
pmichaud I'd just want to avoid any sense of trying to designate one heir as the "one true heir"
masak well, that's the point, isn't it?
Su-Shee and it would be really nice if folks like me who actually live and work in _both_ perls not get the feeling to be forced to one side or the other by getting constant teasing from both sides.
masak it's the "one true heir" thinking that creates friction.
PerlJam masak: were you on #perl earlier when I mentioned how Perl 6 treats the "fat comma"?
masak PerlJam: #perl on freenode? 18:38
PerlJam aye
masak I was, but I must have missed it.
masak backlogs
Su-Shee (and this gossiping here isn't really much better than the same over there...)
masak Su-Shee: sure it is! we have more hugs! :) 18:39
even the bots hug each other today.
supernovus I use Perl 5 and Perl 6 both (although currently in seperate projects), and once Blizkost supports XS modules, I'll probably start using them together. :-)
Su-Shee yeah. and all this butterflies and flowers and hugging really doesn't hide certain assumptions what perl 5 folks might think not very well. ;) 18:41
18:41 PZt left
PerlJam masak: how would you interpret LoeNerd's reaction? (if you're really backlogging) 18:41
18:41 IMeMyself left, PZt joined, IMeMyself joined
masak PerlJam: I see you in various places in the backlog, but didn't find anything on fat commas.... 18:41
18:41 rodi left 18:43 kst left, PZt left, kst joined
patspam douglas crockford made an interesting comment recently about how they're deliberately not giving the next planned version of javascript a version number (yet at least) 18:44
and why he doesn't like html5 being called "5" yet because it hasn't been recognised by any standards bodies yet 18:45
moritz_ it's a bit too late for that in Perl 6 :-)
masak Haskell Prime suffered, according to some #haskell folks I've talked to, from the fact that it wasn't a significant improvement on 98. mainly because the latter is really good.
patspam yeah, very different political sphere, but interesting to compare 18:46
masak PerlJam: no, I give up. I can't find the place you're referring to. I've found LeoNerd and fat commas, but I don't find you in it.
18:46 PZt joined
masak ah, found it! 18:47
Su-Shee masak: you'll find the more or less same discussion for Lisp under the slogan "worse is better"
masak PerlJam: ok, here's what I think: LeoNerd belongs to the set of people who want to make a point out of Perl 5 and Perl 6 being "different languages". since I could successfully argue for them being both different languages and the same language, I don't find making such a point important. but he does. maybe he even feels that you're on his lawn or something, due to talking Perl 6 on #perl. 18:49
18:50 payload joined
frettled He's Texan? 18:51
masak PerlJam: I'm no psychologist, but perhaps we are considered a threat by some, and they react by hostility. maybe they're afraid that we'll flood #perl with academic theory and vapourware releases. :) 18:52
Su-Shee (or maybe they just see the academic tendencies here and the abstractness and compare it to their dynamics and plainly stick with theirs.. ;)
frettled If I'd felt comfortable banging rusty nails with my wibbly bits, I would probably ask for a link to the IRC log in question. But I don't think I do. :) 18:53
Su-Shee it's really interesting that you all think that the perl 5 folks feel threatened instead of maybe just once assuming that they have a good reason for their view.. ;)
masak frettled: I'm not sure there are IRC logs for #perl. 18:54
I was backlogging in my client.
frettled masak: aha
You know what I like about #perl6?
masak Su-Shee: it's a typical from-the-inside rationalization, I guess. 18:55
frettled: no, what do you like about #perl6?
moritz_ Su-Shee: I am interested in why Perl 5 folks dislike Perl 6. Just when the only answer is "because it's not there yet" there's nothing I can do with that answer
Su-Shee: and that includes infrastructure and libraries
18:55 xinming_ joined
Su-Shee moritz_: you could take it it is meant - that someone just doesn't care about Perl 6 until it's really there and has a basic set of libs. 18:56
frettled Every other online community that I've joined, I've had to start out as an utterly deferential n00b who doesn't speak up, because, well, you're always a friggin' n00b every time you join a new online community, nevermind how much experience you have in the field or tech in question.
But not so here.
18:56 nihiliad left
masak I can proudly and without hesitation say that Perl 6 has one helluva development process, with spec/spectests/implementation reacting faster than any language process I know of. 18:56
moritz_ Su-Shee: so what would be my conclusion to draw from accepting that?
Su-Shee: that I should leave poeple allone?
Su-Shee moritz_: you don't have to conclude anything from it, just take it as they say it and leave it at that. 18:57
PerlJam Su-Shee: I'm a "perl 5 folk" and I don't understand those *other* perl 5 folk who seem to react as if they are being threatened.
Su-Shee I'm half of a perl 6 folks and my doubts are growing every day and I understand very much what the fivers talk about.
masak PerlJam: but you're Perl 6 infected, too.
PerlJam Su-Shee: just like I don't understand those people that think that there is some sort of competition between Perl and Ruby or Perl and PHP or Perl and Python, etc.
pmurias moritz_: "because it's not there yet" is a valid reason 18:58
PerlJam Su-Shee: In my mind it doesn't have to be "Perl xor Ruby", but it can (and sometimes should) be "Perl *and* Ruby"
moritz_ pmurias: I know.
masak PerlJam: there should be a term for that kind of disease. maybe zerosumgameitis.
PerlJam masak: heh
Su-Shee PerlJam: but this channels enters exactly this competetition every day over and over again by implying the 5ers feel threatened or by joining the other channel and teasing around. 18:59
19:00 nihiliad joined 19:01 xomas left
frettled Su-Shee: If that's how it appears, then we should amend our ways. 19:01
Usually, though, I think that we're quite constructive, both in terms of community, and in terms of technology. 19:02
masak we definitely have our myths and misconceptions about the other side. 19:03
no question.
frettled And looking at how other people react as well as how we _think_ that other people react is an important part of the first.
Su-Shee actually, I think that there is no conflict and that all this is artificial anyway, because we're talking 5 folks in one channel and 5 in another - what the silent masses really think we just don't know.
frettled We are different enough within this channel that we can at least enlighten ourselves a bit.
Su-Shee: well, we know that they're either being silent here or noisy somewhere else ;)
masak Su-Shee: I disagree.
Su-Shee: Perl 6 has had, and still has, directly provocative slogans that would enrage Perl 5 people if we weren't so insignificant. 19:04
"the next major version of Perl"
Su-Shee it would annoy mst and three others. ;)
pmurias and what is the next major version of Perl 5 supposed to be named? 19:05
frettled pmurias: that's a contradiction in terms
supernovus Perl 5.12 :-)
TimToady it is certainly the case that Perl culture has *always* been about teasing, even back at Perl 1 :)
frettled pmurias: "5" is the major version
5.10.1 => major version 5, minor version 10, patch level 1 19:06
masak Su-Shee: it's an intriguing idea that the whole conflict phenomenon is limited to five people on each side. that's not congruent with my experience, however.
pmichaud 18:58 <masak> PerlJam: there should be a term for that kind of disease. maybe zerosumgameitis.
Jim Collins (author) calls it "Tyranny of the 'Or'"
frettled samiam
supernovus I'd say in the case of Perl 5.10.1, 5 is the language varient, 10 is the 'major version' and '1' is the minor version. That's just my own opinion though :-)
frettled pmurias: not the xor? :D
masak pmichaud: cool.
Su-Shee masak: I think, you're implicitly counting the public status of the 5 (let it be 12, then.. )
frettled supernovus: and that's (deliberately?) misunderstanding what "major version" means 19:07
19:07 ssm left, parduncia left, constant left
pmichaud the opposite of which is "Genius of the 'And'". He uses it to refer to business and organizational settings where people get trapped in an either/or mentality instead of finding ways to acheive all goals simultaneously 19:07
masak Su-Shee: I agree to the extent that the high-profile people are the most vocal in the debate. but I have no reason to believe that they are not expressing concerns that are more widespread.
moritz_ maybe we need to rename both Perl 5 and Perl 6 to completely diffent names :-) 19:08
frettled pmichaud: ah, so it's also sort of covering NIH?
masak moritz_: funny, that's what mst said. :)
frettled moritz_: white perl and black perl?
pmichaud "Strawberry Perl" "Rakudo Perl"
frettled The Programming Language Formerly Known As Perl 5
pmichaud "Pugs" 19:09
masak Funny Glyph Perl.
19:09 mikehh left, mikehh joined
supernovus Perl 5 becomes Perl -> and Perl 6 becomes Perl * ... that would go over well methinks ;-) 19:09
frettled Perl star?
19:09 japhb left
moritz_ supernovus: not at all 19:09
Su-Shee masak: well if I ask the three dozen perl fellows I know which are the john and jane perl in berlin, they mostly tend to have a friendly view towards a new Perl, but they plainly wait until it's tangible for _them_ and their expectations. mr shee will try perl 6 - when it has mysql DBI and can parse his 200 gig per day. 19:10
masak "Wanted: Perl star programmer"
supernovus haha
frettled I think we should have a PostgreSQL DBI before that ;) 19:11
19:11 xinming left
Su-Shee will answer to "perl starlet wanted" ;) 19:11
frettled: well that excludes most ISP offering LAMP
masak Su-Shee: perhaps it all comes down to the fact that people engage differently in what's commonly thought of as 'the community'. my feeling is that the Perl 5 community has woven some sort of Perl 6 aversion/dislike/mistrust into itself. 19:12
frettled Su-Shee: ";)" is a winking smiley, which means...?
Su-Shee frettled: in my old school internet smiley dictionary it's "irony"
PerlJam masak: a vocal subset of the Perl 5 community ...
pmichaud PerlJam: +1
19:12 payload left, ShaneC joined
masak PerlJam: yes, but a subset which sets the tone. 19:12
pmichaud ... a subset which sets a tone that we're attuned to :) 19:13
Su-Shee which sets the tone in a very few places.
masak like conferences.
PerlJam Su-Shee: cite your use of "few"
pmichaud oh, I tend to agree with Su-Shee. I hear from a lot of folks that use p5, are interested in p6 "when it's ready", but don't at all feel threatened by it
PerlJam I've seen it on IRC, perlmonks, twitter, blogs, encountered it in real life conversations, etc.
pmichaud they're just not vocal in their "I'm not threatened"-ness
Su-Shee masak: honestly, do you think the swedish perl community really cares as a whole what the 10 grand old dogen of perl 5 think about perl 6? 19:14
frettled Su-Shee: In my smiley list from 1985, it's the dwarf version of ;-), which is «Winky smiley. User just made a flirtatious and/or sarcastic remark. More of a "don't hit me for what I just said" smiley.»
pmichaud: mm
Su-Shee frettled: well I mean it in irony and/or sarcastic way. and yes, something I actually do pun on being female.
19:15 mikehh left
frettled Su-Shee: but in either case, it was clear that I wouldn't mean what was written _seriously_. 19:15
Su-Shee: both to you and other readers.
19:15 mikehh joined
pmichaud anyway, I don't think think any Perl 6 aversion is completely woven into the Perl (5) community. It's there among many of the more visible and vocal members... but of course we don't hear from the people who aren't really threatened by Perl 6 19:15
frettled Su-Shee: I think I'll take a break now, because I can see that we're on the verge of hostility. Sorry about that. 19:16
colomon pmichaud++
Su-Shee frettled: uhm what?
masak Su-Shee: I think the reason non-Perl people think that Perl 6 === DNF is due to bad or confused messages. I think there are similar reasons for Perl 5 people (not all of them, but some) not liking Perl 6.
Su-Shee frettled: I'm lying on the sofa with chocolate and hot tea and candles and chat about the politics and dynamics of programming communities... if that's hostile.. :) 19:17
PerlJam Su-Shee: you're female?!? ;)
ash___ in parrot, is there an easier way to print a hash in a way to see its contents? using say just did: Hash[0xdeadbeef]
Su-Shee masak: "DNF"?
masak Duke Nukem Forever.
Su-Shee ah :)
masak inherent in the talk about 'setting the tone' is the idea that if our tone is good enough, we get to set it to some degree.
pmichaud ash___: see "dumper.pir"
ash___ pmichaud: thanks
pmichaud ash___: I generally do _dumper(hash)
19:17 constant joined, ssm joined, parduncia joined
PerlJam Su-Shee: or, put another way ... on IRC, no one knows you're *not* male. :) 19:17
ash___ pmichaud: would _dump($P2) function the same? 19:18
19:18 ssm left, parduncia left, constant left
pmichaud ash___: do you mean as opposed to _dumper($P2) ? I don't think so 19:18
but _dumper(...) is Parrot's generic data dumper
Tene I've been known to be not-male on IRC a few times. Haven't worked out how to pull it off outside of IRC yet though.
Su-Shee PerlJam: well I'm not really hiding it but I don't have to wear it on my shirt.. a wait I just said I want a rakudo starlet shirt.. ;)
ash___ oops i meant _dumper($P0), just wanted to make sure it worked with the $P0 style registers 19:19
19:19 pmurias left
pmichaud ash___: it's supposed to be able to dump any PMC type 19:19
(if the PMC type has the appropriate methods defined, which Hash, Array, etc. do)
ash___ pmichaud: alright, thanks
masak heads home
Tene bye masak
masak I'll be happy to talk more about communities some other time! 19:20
it's a discussion worth having.
19:20 masak left 19:21 constant joined, ssm joined, parduncia joined
PerlJam wonders if there's any divisiveness between "old school" perl 5 hackers and those that use things like Moose and Modern::Perl. I haven't noticed any, but that doesn't mean it's not out there. 19:21
supernovus PerlJam: I have seen a little bit, but not as much as Perl 5 versus Perl 6. 19:22
ash___ in pir code, is the first param always self for a method sub? 19:24
Tene ash___: 'self' is a special register, always pointing to the invocant. it's not passed as a param.
supernovus At work, we have a few people who consider Moose, Modern::Perl and/or perl5i to be "slow, bloated overkill", and don't want us to use it in work related projects.
Su-Shee PerlJam: there are. I know mostly "no moose, no poe, no catalyst" perl people. 19:25
supernovus Having never used POE or Catalyst, I guess I fall into some subset of that demographic, but I do use Moose (via MooseX::Declare) and my own branch of perl5i in a lot of projects. 19:26
moritz_ supernovus: I think the question is not if you use those things or not, but if you reject them 19:27
ash___ Tene: do you know what would be passed to something if you did role A { }; A.^methods; as the first parameter?
moritz_ ash___: the type object
A.^methods is short for A.HOW.methods(A)
Tene A.^methods is the same as A...... yeah, what he said.
Su-Shee supernovus: I don't use Catalyst for example, I'm a qmail-mojo-unixoid-kiss-yagni-kind-a-girl. 19:28
PerlJam Everytime I think of some useful feature of Perl being "slow", it reminds me of the time I spent benchmarking the difference in speed between Perl 4 and Perl 5. Ultimately I came to realize that it didn't matter so much that Perl 5 was slower than Perl 4 because Perl 5 was just so much *better* :)
19:30 pmurias joined
ash___ moritz_: okay, well, there is an issue in ng's RoleHOW methods method src/metamodel/RoleHOW.pir#157 it doesn't have the attribute 'parrotclass' 19:30
moritz_ ash___: I'm not familiar with that code :( 19:31
ash___ ng: role A { method a { return 'b'; }}; say A.^methods;
p6eval ng 21558a: No such attribute 'parrotclass'␤current instr.: 'perl6;RoleHOW;methods' pc 4866 (src/metamodel/RoleHOW.pir:47)␤
pmichaud ash___: the problem is that very few of us know how type objects (A) get mapped internally 19:32
ash___ moritz_: you don't need to, i am just talking outloud mostly, trying to figure our whats going on to see if i can make a patch
moritz_ ash___: great :-)
pmichaud so, if you define role A { ... } -- I think only jnthn++ could say precisely what its type object ends up looking like 19:33
(internally, that is)
I can say that it's not going to be exactly the same as the type object for a class, and so I don't know its internal details 19:34
supernovus Well, in the case of Catalyst, I've never had the need for a MVC Web framework. For simple scripts, "use CGI::Simple" is all I really need.
ash___ pmichaud: jnthn++ said it was possibly only partially implemented, so i am seeing if i can figure it out thats all 19:35
19:36 japhb joined
Su-Shee supernovus: I'm undecided on the whole MVC thing yet... 19:36
still?
Tene "still" 19:37
jnthn Role type objects aren't so interesting, really. Mostly, they just need to know what their metaclass is, handle punning if somebody tries to call a method on them, and so forth. :-) 19:44
ash___ i see whats wrong
jnthn The actual info is in the metaclass.
At the moment we still somewhat base that on a Parrot role, but soon we can (I expect) stop doing so.
ash___ that methods method lists RoleHOW's methods not the specific Role's methods
jnthn Ah 19:46
That's a bug then.
19:47 jonasbn joined
ash___ ng: role A { method a { return 'b'; }}; say A.HOW.^methods; 19:47
p6eval ng 21558a: sh: ./perl6: No such file or directory␤
19:48 jonasbn left, jonasbn joined
ash___ on my computer that returns the list of methods in RoleHOW, but for some reason A.^methods also gets dispatched to that function 19:48
19:48 xenoterracide_ is now known as xenoterracide
ash___ jnthn so, thats not the method that should be handling the a specific role's ^methods method? 19:48
moritz_ ttp://lastofthecarelessmen.blogspot.com/2009/12/now-with-grammar-and-tests.html 19:54
the tests can be greatly simplfied
with given $match<ABC::broken_rhythm> { is .<<note>[0]<pitch><basenote> ... } 19:55
erm, .<note>...
sadly I don't manage to post code in the comments
jnthn ash___: Doing SomeRole.^methods should call SomeRole.HOW.methods(SomeRole)
ash___: And SomeRole.HOW should return an instance of RoleHOW 19:56
colomon moritz_: post the link to the irc log there? :)
jnthn So yes, it's the right method.
the method is probably buggy.
moritz_ colomon: or I just hope the author lurks here :-)
likewise regex barline { '|' | ':|' | '|:' | ':|:' | '::' }
looks like it would need LTM to work
reordering the longest alternatives to the front would help 19:57
wanted to post a link to that blog post, but now it doesn't respond at all when press the preview or submit button 20:05
20:06 jferrero left 20:08 jferrero joined 20:10 kaare_ joined
ash___ jnthn: how would you access the method list of a Perl6Role from pir? 20:23
if you know, if not i can try to find out 20:24
jnthn ash___: Currently, those are stored in the underlying Parrot role object (which is stored as an attribute in the metaclass instance), so you can get that and then do meth_list = inspect that_thing, 'methods' or so. 20:32
ash___ jnthn: k, i'll try that 20:33
TimToady heh, today's /. quote: Nothing is easier than to denounce the evildoer; nothing is more difficult than to understand him. - Fyodor Dostoevski 20:39
moritz_ how fitting 20:40
PerlJam Are you implying that Perl 5 people are evil? ;-> 20:41
moritz_ no :-) 20:42
jnthn afk for a while 20:43
TimToady note that it is part of my belief system that a non-evil person can nevertheless be an evildoer. 20:46
20:48 reqamst joined
TimToady same goes for companies :) 20:49
diakopter yelp won't help google stay less evil 20:50
TimToady and, from time to time, language designers...
PerlJam TimToady: There's a little yin and yang in all of us.
cj TimToady: karma.colliertech.org/~cjac/tmp/isurvived.png 20:55
moritz_: ilbot.colliertech.org/ubuntu-us-wa/today 20:56
moritz_ cj: thanks 20:57
cj moritz_: you bet. thanks for your help. /lastlog moritz_: will get you a couple of deps you should add to your HOWTO 20:58
21:00 justatheory joined
moritz_ cj: Bot::BasicBot and Calendar::Simple - did I miss any other? 21:00
21:01 Su-Shee left 21:05 pmurias left, pmurias joined 21:13 takadonet left
cj moritz_: let me check my logs 21:19
Date::Simple 21:20
frettled wants Dating::Simple
cj Bot::BasicBot v0.81 (newer than lenny) 21:21
those seem like the only ones I mentioned in the channel
21:25 kaare_ left 21:32 yahooooo left 21:40 patspam left 21:41 patspam joined 21:46 uva joined
wayland76 wants Dating::TheRightPerson :) 21:52
ash___ is that a perl 5 module? i don't see it in cpan 21:54
21:59 meppl joined
TimToady wrong cpan, you want the Complementary Person Analysis Network for that one 22:00
wayland76 Is that Open Source? :) 22:01
mberends has Dating::More been written yet? 22:02
IllvilJa What is the best way, non-blocking way to check if the user has pressed a key? (Considering a Perl6 program running in a terminal session)
22:02 patspam left
IllvilJa Oh, and of course, if a key is pressed... what key as well :-). 22:02
22:07 crythias left
frettled wayland76: Anyway, that's NYI 22:07
22:08 lazybird left
mberends IllvilJa: there is no way to read keys without blocking. Even the mature Perl 5 ways are said to not always work, afair. 22:08
moritz_ there's Term::ReadKey, but it's *nix only, and has no real tests 22:09
(for Perl 5, that is) 22:10
TimToady your best bet for Perl 6 right now is probably to write a helper program to read the terminal and send in the keystrokes as events via a socket 22:11
moritz_ but we don't have non-blocking socket reads either :Z/ 22:12
s/Z//
PerlJam blocking--
TimToady blocking is okay if there's only one source of events, and the problem is event-driven... 22:13
in general, polling solutions are going to bite you sooner or later
frettled especially the electronic ones 22:14
ash___ how do you get the type of a pmc in pir? class $P0, $P1 seems to work, is that what that function does? or am i crazy
22:18 Wolfman2000 joined
moritz_ ash___: maybe #parrot would be more responsive to PIR questions 22:18
ash___ thats on the perl irc server right?
22:22 jonasbn left
moritz_ yes 22:22
irc.perl.org
wayland76 ash___, moritz_: No, it's on irc.parrot.org 22:23
moritz_ wayland76: that's a DNS alias
$ host irc.parrot.org 22:24
irc.parrot.org is an alias for irc.perl.org.
wayland76 Ah, ok :)
So now I've simplified my IRC config :) 22:25
ash___ ng: say ''.^methods; 22:30
p6eval ng 21558a: Null PMC access in inspect_str()␤current instr.: 'perl6;ClassHOW;methods' pc 3898 (src/metamodel/ClassHOW.pir:372)␤
ash___ i think there are some problems with introspection at this moment in ng 22:31
IllvilJa Ok, if I do a blocking read of a single keypress... is there a way for that? 22:32
22:33 JodaX joined
IllvilJa (But no non-blocking read of keypresses... that would be disastrous if I would attempt to reimplement good old unix 'hunt' in perl 6 ;-) ) 22:33
Tene IllvilJa: I've done some Curses stuff with rakudo. It likely won't work anymore without some updates to rakudo's inter-hll loading, though.
22:34 Gruber joined, d4l3k_ joined, r0bby_ joined, nothingmuch_ joined, szabgabx left, wayland76 left, r0bby left, p6eval left, Woody2143 left, Grrrr left, Infinoid left, charsbar left, elmex left, szabgab left, _Joda_ left, renormalist left, yves left, breinbaas left, Solarion left, buubot left, pmichaud left, nothingmuch left, simcop2387 left, IRSeekBot left, dalek left, nothingmuch_ is now known as nothingmuch, elmex joined, d4l3k_ is now known as dalek 22:35 yves joined, p6eval joined 22:36 simcop2387_ joined 22:37 Infinoid joined, IRSeekBot joined 22:38 buubot joined 22:41 simcop2387_ is now known as simcop2387 22:43 stephenlb joined 22:47 wayland76 joined, ashizawa left
ash___ moritz_: $a.^methods; translates to $a.HOW.methods($a) or $a.HOW.methods($a.WHAT) ? 22:53
22:55 nihiliad left 22:57 mikehh left
ash___ nm, i checked S12, its the first one $a.HOW.methods($a) 22:58
found another bug then
22:58 mikehh joined 23:01 supernovus left
zaslon lolmoritzhazblogged! moritz++ 'Publicity for Perl 6': perlgeek.de/blog-en/perl-6/publicit...erl-6.html 23:01
mberends moritz_: it has been an incredibly successful Perl 6 Advent blog. Well done! 23:05
23:06 ssm left, parduncia left, constant left
uva voice test 23:06
23:07 constant joined, ssm joined, parduncia joined
ash___ ng: ''.^methods; 23:09
p6eval ng 21558a: Null PMC access in inspect_str()␤current instr.: 'perl6;ClassHOW;methods' pc 3898 (src/metamodel/ClassHOW.pir:372)␤
ash___ should i put ng branch bugs in the perl 6 bug tracker? 23:10
pugs_svn r29375 | mberends++ | [pugs/src/perl6/vill] introducing vill, a viv+llvm mashup 23:11
23:11 ssm left, parduncia left, constant left, constant joined, ssm joined, parduncia joined 23:13 ssm left, parduncia left, constant left, constant joined, ssm joined, parduncia joined 23:14 ssm left 23:15 parduncia left, constant left, constant joined, ssm joined, parduncia joined 23:18 ash___ left 23:19 patspam joined
PerlJam All you perl6advent people look at my draft for Day 20 and send me some commentary. (I know it needs work) 23:20
Gotta go now, but I will backlog (or you can tell phenny)
23:21 meteorjay left 23:24 am0c joined 23:27 Exodist left, mberends left 23:30 iblechbot left
Tene PerlJam: s/xpress/express/ 23:32
sjohnson oh, the bliss of Perl -> PHP serializers 23:53
23:55 uva 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!