»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.
00:06 ryan_ joined
zaslon lolperl6adventhazblogged! perl6advent++ 'Day 14: Going to the Rats': perl6advent.wordpress.com/2009/12/1...-the-rats/ 00:06
00:11 lestrrat is now known as lest_away 00:14 yahooooo joined 00:16 ryan_ left 00:22 IllvilJa left 00:26 payload joined 00:31 IllvilJa joined 00:32 payload left, payload joined 00:55 kcwu left 01:00 kcwu joined 01:02 jrockway_ is now known as jrockway 01:14 Ryan52 joined 01:22 pmurias left 01:23 payload left 01:25 agentzh joined 01:39 jferrero left 01:40 JimmyZ joined 02:19 c9s left 02:24 c9s joined 02:36 ihrd joined 02:44 zaslon left, mubot left, pointme left, envi^office joined 02:45 pointme joined, zaslon joined 03:04 abra left 03:13 coke joined
coke so, how would one return from a try block? 03:13
best I've got so far is to set a $flag and then have a conditional post-try. 03:14
diakopter pugs: sub a() { try { return 4 }; return 5 }; say a; 03:19
p6eval pugs: 4␤
diakopter I dunno, honestly 03:20
(I don't know whether that's up-to-spec)
coke doubt it.
rakudo: sub a() { try { return 4 }; return 5 }; say a;
p6eval rakudo 7ef386: 5␤
coke the problem is that return is a CONTROL exception...
hurm. though now that I think about it.. 03:21
03:21 _jaldhar left
diakopter mildew-js: sub a() { try { return 4 }; return 5 }; say a; 03:22
p6eval mildew-js: Can't open perl script "mildew": No such file or directory␤
coke (doesn't help that I'm actually using NQP)
diakopter o :)
arnsholt Reading about git makes me suspect it's not actually intended to be used by humans >.< 03:25
03:26 Wolfman2000 joined
Wolfman2000 ...I forgot. I'm supposed to post to the calendar tomorrow, aren't I? 03:26
coke diakopter: hurm. I think pugs might be right on this one.
carlin Wolfman2000: looks like it, dev.pugscode.org/browser/misc/perl6...9/schedule 03:28
Wolfman2000 thanks for the reminder.
Wolfman2000 has been kept busy the past few days
carlin Wolfman2000: did you see that someone attempted your Caesar cipher challenege? 03:29
Wolfman2000 carlin: I got a few email posts about it. didn't see a reason to reply.
Wolfman2000 was also busy with wrapping up college stuff...stupid tuition and almost lost health records. 03:30
03:31 nihiliad left 03:33 ihrd left 03:35 nihiliad joined 03:38 _jaldhar joined 03:45 meppl joined 03:46 envi^home joined 03:59 Intensity left 04:03 lest_away is now known as lestrrat
Wolfman2000 reminder please: I know Perl 6 has .perl that should be the toString() equivalent. Are there any other similar commands? 04:10
carlin Wolfman2000: .perl returns the data structure. .Str, .Int, .Num etc. type-cast 04:17
Wolfman2000 ah, that explains that. 04:18
carlin rakudo: my %h = { foo => 1, bar => 2}; say %h.perl; say %h.Str; say %h.Int;
p6eval rakudo 7ef386: ( no output )
carlin argh, evalbot can't do hashes
rakudo: my @h = { foo => 1, bar => 2}; say @h.perl; say @h.Str; say @h.Int; 04:19
p6eval rakudo 7ef386: ( no output )
carlin rakudo: my @h = <Foo bar baz>; say @h.perl; say @h.Str; say @h.Int;
p6eval rakudo 7ef386: ["Foo", "bar", "baz"]␤Foo bar baz␤3␤
Wolfman2000 ...I forgot. Is there a quick way to shuffle the elements of a list around?
carlin my @a = <foo bar baz>; say @a.perl; @a .= pick(*); say @a.perl 04:20
rakudo: my @a = <foo bar baz>; say @a.perl; @a .= pick(*); say @a.perl
p6eval rakudo 7ef386: ["foo", "bar", "baz"]␤["baz", "bar", "foo"]␤
Wolfman2000 ...holy crap
now if only I understood what pick(*) is supposed to do with that 04:21
04:21 justatheory joined
Wolfman2000 ...along with .= 04:21
I thought ~ is the new concatenation operator. What does .= do now?
carlin @a .= pick(*); is the same as writing @a = @a.pick(*) 04:22
Wolfman2000 so @a.pick(*) picks the elements in random order...and then stored in @a again.
Where * means "all you can" in this context
carlin pick picks a number of elements at random 04:23
rakudo: my @a = 1..5; say @a.pick(2)
p6eval rakudo 7ef386: 53␤
Wolfman2000 I know it picks parameter elements
I'm writing on that tomorrow
carlin yes, * basically means everything, so it picks everything in a random order 04:24
Wolfman2000 ...and someone else is going to talk about the * soon. Oh well: sneak preview 04:25
carlin Wolfman2000: you definitely 'picked' a good topic :-) 04:28
Wolfman2000 ...urge...to groan...rising...
carlin: my post is ready for previewing. 04:30
diakopter coke: a 04:46
coke: ah 04:47
silly wireless keboard drops characters
keyboard
04:51 meppel joined 04:56 justatheory left 04:59 meppl left 05:22 dryden joined
dryden how goes the project? 05:24
Wolfman2000 which project? There are many in Perl 6 05:25
dryden I should have been more specific. I don't actually know enough to fully understand the progress of the Perl6 project in its entirety. I can say that I am very excited. 05:28
Wolfman2000 There are many implementations out there, with some farther along than others. 05:32
Primarily, this room focuses on Rakudo, with some also devoting time to...I think it's called Mildew 05:33
dryden I haven't heard of Mildew, I have heard of Rakudo. I've been meaning to try it out. I haven't yet, I'm concerned about performance since it runs on parrot (as far as I know). 05:37
Wolfman2000 dryden: the goal is to make it work first, then speed it up later.
dryden I definitely see your point, and completely agree with you. I'm starting to lose sight of the point in trying to make it work. Especially with Ruby having become so popular, and the recent release of stable Ruby 1.9 05:39
Wolfman2000 ...perhaps the best ones to answer your concerns are the developers of Rakudo
TimToady, pmichaud, jnthn: either of you guys here?
dryden Yikes, that's alright, I wouldn't want to disturb them. They're probably very busy. 05:41
[particle] they're probably sleeping 05:47
by the end of march, you'll see a release of rakudo named "Rakudo *" 05:48
it will be a usable subset of Perl 6, one with which you can solve everyday problems.
it may not be very fast, but it'll get the job done.
dryden That sounds really exciting, I suppose as wolfman said, step one is to get it working, and we can worry about performance later. 05:51
05:54 Astoria` joined
dryden I think performance in the long run will still be an issue, simply because the set of "everyday problems" or applications in general are becoming more networked or web-based. And that's one area where performance still matters mostly because of server limitations and a massive number of users. 05:57
By the time, Perl6 is out, stable, and performing well. Newer arguably "faster developing" languages like Ruby and Ruby based systems might have a stronghold. 05:59
I'm not trying to depress anyone. I'm worried. I love Perl. 06:00
06:03 thowe left
dryden I love what you're all doing. I'm so sorry for talking so much. 06:07
06:10 dryden left 06:16 coke left
diakopter hm 06:17
I admit; I think dryden's comments have validity (dryden if you're reading the irclogs, feel free to return or otherwise continue the discussion). 06:20
wayland76 diakopter & dryden; have you read mst's post on Perl 5 vs. Perl 6? 06:24
06:25 nihiliad left
wayland76 www.shadowcat.co.uk/blog/matt-s-tro...ck-perl-6/ 06:25
Ryan52 mst is great 06:26
06:38 coke joined
wayland76 Well, I could do without the profanity, but he makes a good point 06:42
hejki yup 06:45
perl5 is great :)
06:51 kaare joined, kaare is now known as Guest31990, envi^home left 06:54 mberends joined 06:58 Astoria` left
mberends we need to hug our neighbours in #perl more - even yesterday a visitor asked what they thought of Perl 6 and the regular members just trashed it. I cannot find the log reference, but it was phrased ["Perl" 6]. 06:59
07:01 masak joined
mberends masak! 07:01
masak \o/
phenny masak: 11 Dec 18:57Z <TimToady> tell masak if you call it Rocket Art you can say it's not rocket science...
masak: 12 Dec 02:46Z <carlin> tell masak november and uri (owned by ihrd) have a lib/URI.pm
masak: 12 Dec 03:01Z <carlin> tell masak ... but it looks like that is the last conflict \o/
hejki mberends: heh.. i'm a perl5 fanatic, but i tend to like perl6 instead of bashing it - after all it solves lots of stuff that's not that trivial in perl5 (at least without any additional modules :P) 07:02
masak carlin: I believe (but investigation needed) that it's "the same" URI.pm, possibly with improvements in the standalone repo.
hejki also i think it's a common misbelief that perl6 is even close to ready :P
masak hejki: the concept 'ready' is a mirage. 07:03
07:03 Supaplex joined
hejki well i don't think perl6 is even close to usable 07:03
atm :)
masak I respectfully disagree.
that said, I wouldn't use it for everything. 07:04
hejki sure, but i think that spending several tens of seconds to do for 1..10_00 -> $i is rw { $i++ } means it's not usable
masak yes, I know what you mean.
hejki (not sure if several tens of seconds are even enough for that)
ye
this is why i consider it being unusable :)
masak we seem to have different criteria, is all. 07:05
hejki yup. i love how all stuff is subjective :)
masak I wouldn't use Perl 6 for speed right now.
Supaplex let's all go back to dos 3.3.
masak then again, I wouldn't use Ruby for speed either.
hejki masak: well i'm crawling through project euler.net atm
masak: and most of those do require some speed to be comfortable to test :P
masak hejki: trust me, I feel your pain. 07:06
hejki masak: (i've improved my algorithm skills a lot tho ! :))
ye :P
masak hejki: wait until you run up against the complexity barriers, too. :/
hejki and the most fsced up thing is that i'd really want perl6 to be good and competetive language - cause i really do like it and the concept
masak hejki: patience. :)
hejki masak: heh.. using gaher { .. } with a for ^Inf -loop didn't work out as intetended either tbh ;> 07:07
masak no, Rakudo isn't lazy yet...
there's a branch going on where it will be.
hejki ye.. i'm patient, but i think that the amount of fuzz around perl6 makes other people impatient (also they might think that perl6 sucks, since they aren't aware of it's current state)
masak merge ETA one month.
hejki ahh
:P
masak (that's my subjective estimate, based on current speed) 07:08
hejki: we try to target early adopters, and people who can fill in the blanks. :)
naturally, in almost any comparison against (say) Perl 5, Perl 6 is going to come up short right now. that's because no-one's ever tried this route before. 07:10
people who want the polished, well-tested product will have to hold their breath for a few years while people who are willing to improve the imperfect product get to work. :) 07:11
07:11 youwin joined
hejki :) 07:12
masak thing is, we need more of the latter kind -- hence the active PR.
hejki i'm willing to do stuff if needed, i've never developed a language tho. :) 07:13
mberends hejki: that's the -Ofun here :-)
Supaplex lol 07:14
mberends and the most awesome language too :)
er, *the* 07:15
Supaplex Larry likes my vanity license plates
mberends ... with what text ? 07:16
masak hejki: when I started getting interested in Perl 6, I had never developed a language either.
Supaplex perl6
mberends Supaplex: :-) 07:17
masak Supaplex: I remember you. you've been here before, haven't you?
Supaplex ohhh probably.
Supaplex is still a Supaplex after all these years
masak I also remember enjoying the game 'Supaplex' many years ago on an i386 machine. 07:18
hejki masak: :) 07:19
07:19 Su-Shee joined
Su-Shee good morning. 07:19
mberends gm Su-Shee 07:20
masak Su-Shee: mornin'!
moritz_ good morning
Supaplex I still keep a 386sx16 for those good ole days myself. Has a ide to cf adapter with a 32mb CF in it. :)
masak tonight I dreamt that we were slashdotted a second time. 07:21
moritz_ :-) 07:22
07:24 rgrau left
masak mberends: re #perl6: I hang on #perl6-on-freenode whenever I'm on IRC, trying to detect when people mention Perl 6. I'm sad to hear there are people trashing the language still. some of the members are very understanding post-bridge. 07:26
mberends masak: yes, I was sad too, it passed by before I could "try to help". 07:27
masak right. no use bringing it up after the fact. we'll try to be there next time instead. :) 07:28
mberends +1
Su-Shee It'll pass when there's something really nice to show off with. 07:29
masak the tendency to write '"Perl" 6' is easily understood as an expression of the feeling that Perl 6 is encrouaching somehow.
s/oua/oa/
mindshare-wise if not feature-wise. 07:31
mberends we've been through the naming debate often enough. we have make the best of what we've got.
*have to
masak I'm in the "it's the right name, lay off it" camp... 07:32
hejki i guess the similarity between perl and perl6 is that both try to utilize every possible character in the syntax :) 07:33
moritz_ not quite
since Perl 6 is Unicode-based, there are many unused characters left :-)
mberends unicode gives even more characters ... what moritz_ said :) 07:34
Su-Shee yes, wonderful. I'm already wishing for a use English ;)
Supaplex just wait until buu adds the orgasm operator 07:35
Su-Shee which will be an opengl-based sparkling 3d animated true color thingie? ;) 07:36
mberends virtual fun! 07:37
Supaplex yes. and if you don't draw it correctly, it's a syntax error.
masak some people like to think of the history of programming languages as a struggle where one 'usurps' others: twitter.com/byxorna/status/6641461450
hejki :P 07:38
07:38 LeifD joined 07:39 LeifD left
Su-Shee well who cares.. either perl 6 will be adopted or it won't and that doesn't depend on the name but on the environment, examples and usefulness. 07:39
hejki well i guess it has something to do with the name as well - there are loads of perl-bashers that haven't written a single line of perl 07:40
wayland76 Su-Shee: To continue our previous discussion, I just found a real reason to not use an online office suite 07:45
It stops working when a truck drives past your house and tears the telephone line from the pole, as happened in the midst of our previous discussion :) 07:46
hejki :P
Supaplex hahaah
moritz_ ouch
Supaplex wayland76: so why do you live in a place like that? ;) 07:48
wayland76 Because it's nice out here in the country :)
07:49 cotto left
Supaplex cool 07:49
Su-Shee I know why I like the city.. ;)
wayland76 In relation to that link; is Perl 6 fast? 07:50
Supaplex because it has a different smell
hejki wayland76: not atm, but i hope it'll be :)
wayland76 ok, just checking -- that's what I thought :)
Maybe we should add it to the spec :) 07:51
moritz_ perl 6 is lightning fast - just no Perl 6 compiler so far :-)
07:53 TiMBuS joined, Alias joined
Su-Shee -> office. 07:54
TiMBuS Can't return outside a routine
in Main (file <unknown>, line <unknown>)
:I
ok but, what if i dont have a single return anywhere 07:55
moritz_ maybe a fail()?
rakudo: ...
p6eval rakudo 7ef386: Can't return outside a routine␤in Main (file <unknown>, line <unknown>)␤
hejki rakudo: foo(2); sub foo (Int $x) { $x; #returns 2 } 07:56
p6eval rakudo 7ef386: Unable to parse block; couldn't find final '}' at line 2, near ""␤in Main (file <unknown>, line <unknown>)␤
hejki heh.. comments ftw :>
JimmyZ ng: sub foo { return {say $^a ~ $^b x 2 ~ 'bar2000'} }; $_ = foo; $_('f', 'o');
p6eval ng 0307c2: foobar2000␤
JimmyZ ng: sub foo { return {say $^b ~ $^a x 2 ~ 'bar2000'} }; $_ = foo; $_('f', 'o');
p6eval ng 0307c2: offbar2000␤
JimmyZ hello all, Is it right? 07:57
mberends looks right
JimmyZ ng: sub foo { {say $^a ~ $^b x 2 ~ 'bar2000'} }; $_ = foo; $_('f', 'o');
p6eval ng 0307c2: Not enough positional parameters passed; got 0 but expected 2␤current instr.: '_block36' pc 260 (EVAL_1:113)␤
moritz_ TiMBuS: fail() and '...' both fall back to return() in rakudo 07:58
TiMBuS i see
JimmyZ pugs: sub foo { {say $^a ~ $^b x 2 ~ 'bar2000'} }; $_ = foo; $_('f', 'o');
p6eval pugs: *** Blocks with implicit params cannot occur at statement level␤ at /tmp/QTM8cu8d6P line 1, column 46-54␤
JimmyZ rakudo: sub foo { {say $^a ~ $^b x 2 ~ 'bar2000'} }; $_ = foo; $_('f', 'o');
p6eval rakudo 7ef386: Not enough positional parameters passed; got 0 but expected 2␤in Main (file src/gen_setting.pm, line 324)␤
TiMBuS im guessing its a fail() somewhere
08:04 colomon joined, Alias left, Alias joined 08:05 iblechbot joined
TiMBuS welp turns out it was a {...} i left somewhere. dang. 08:08
on a related note that old perl 6 irc bot i made went from using 70mb of ram to 107mb. O_o 08:09
08:12 hanekomu_ joined
carlin It'll increase more if you leave it running :-) 08:15
08:20 Baggio_ joined 08:22 jferrero joined
diakopter wayland76: yes, why? 08:27
(why do you ask, I mean) 08:28
wayland76 diakopter: If that's in reference to the link I posted before, I thought it might be related to what dryden was asking about :)
diakopter wayland76: yes, the question you directed at me, above 08:30
whether I read the scrant at that url
08:31 hirschnase left
diakopter I read dryden's use of "it" in "... starting to lose the point of trying to make it work.." as referring to "Perl 6 on Parrot" 08:32
since his only complaint was, precisely, "I'm concerned about performance since it runs on parrot" 08:33
his/her
(dryden's sex being unspecified) 08:34
Supaplex use Wahmbulance; 08:35
diakopter wayland76: you said "what dryden was asking about" - dryden's only question was "how goes the project?" 08:36
masak: who was trashing the language earlier? 08:37
wayland76 diakopter: I was thinking particularly of this dryden quote: "I'm not trying to depress anyone. I'm worried. I love Perl."
masak diakopter: according to mberends, someone on #perl6. I wasn't there.
um, #perl, that is.
sometimes the '6' sneaks in there, in writing as well as in speach... :/ 08:38
diakopter ahhhh, I see the double typo now :) sry
actually, triple, now
Supaplex www.bash.org/?5300 08:39
masak diakopter: I rest my case. :/
diakopter (the earlier typos were what threw me off :)
masak Supaplex: I really like that one.
Supaplex I agree :) It fits pretty well right now. 08:40
Su-Shee *hihi* :) 08:41
diakopter HI 08:42
08:45 mathw joined 08:49 wayland joined 09:05 wayland76 left
dalek ok: d5d9a2e | (Carlin Bingham)++ | src/ (2 files):
[Builtins] [Preface] Fix typos

s/architectual/architectural		 09:14
BooK ONOES progress on the perl 6 book will wake me up!
masak BooK: maybe /ignore dalek? 09:18
moritz_ BooK: time to get a case sensitive hilight :-)
09:22 colomon left
BooK moritz_: yeah :) 09:23
09:40 envi^office left
masak .oO( Let's finish the Perl 6 book quickly, so we don't wake BooK up unnecessarily! ) 09:42
moritz_ has to hand in his thesis on Friday. After that he might have more time for writing book stuff 09:43
09:44 pmurias joined
masak it's fun to say "book", isn't it? :) 09:45
when I write "book", it shows up highlighted like a nickname in my client.
moritz_ let's just come up with a good name for it, then we can call it by name 09:46
masak <superlative adjective> Perl 6
diakopter "Up the Perl 6 Space Elevator" 09:47
"Orbiting Rakudo Star"
"Blasting Off with Perl 6" 09:48
moritz_ what about something more modest? "Using Perl 6"? 09:49
masak that's a nice title. 09:50
not very exciting, but perhaps we don't need exciting.
moritz_ I think it captures that idea that we want usable examples, to enable people to write usable Perl 6 code 09:51
carlin Only problem is once you have Using Perl 6, Learning Perl 6, Programming Perl 6 ... it gets confusing as to which is which
hejki ye
diakopter "Using Perl 6: The first high is free"
hejki "Landing Manual for Planet Perl6"
moritz_ carlin: so far there's neither of {Learning,Programming} Perl 6, so we can ignore them if we want :-)
hejki "The Big Book of Perl6" 09:52
could it be a pun as well?
moritz_ that's definitively not what we're writing
diakopter well, sixperl call mentioned Damian+Larry ramping up on Camel6, and Stonehedge wrote a lot of LearningPerl6 already...
so, I disagree that those can be ignored 09:53
esp since they'll likely be O'Reilly
hejki "grammar Perl6 { .. }"
diakopter "Adopting Perl 6" 09:54
"Skinny-Dipping With Perl 6"
"Seducing Perl 6"
"Abusing Perl 6"
I mean, there are lots of comical tongue-in-cheek cutesy options
moritz_ diakopter: Learning Perl 6 is a sleeping project 09:55
diakopter "Implementing Perl 6" is already around 100,000 pages long
so I don't think that'll ever be printed 09:56
09:56 ejs joined
diakopter "Birthing Perl 6" will need to be travailed by TimToady 09:56
"Debunking Perl 6" is plenty long too 09:57
"Trolling #perl6" is never long enough :P
mberends "Sleeping with Perl 6"
diakopter on that note, "A Taste of Perl 6" wouldn't be all that bad. 09:58
hejki "Bashing Perl6" :)
diakopter so, "Tasting Perl 6"
"Wandering Perl 6"
moritz_ likes "A Taste of Perl 6", but it feels like it stays too much on the surface 09:59
diakopter I mean, "Exploring Perl 6"
"Drowning in Perl 6"
mberends "Diving for Perl 6"
diakopter oooooo
09:59 cotto joined
diakopter as in "Cracking the Perl 6 Oyster" 10:00
"This Oyster Tastes Like Onions"
hejki cracking the oyster sounds a bit perverted :) 10:01
diakopter oh. I thought it was the next one that was...
hejki hmm.. it would be if you'd s/Onions/Blood/;
"Primal Concrete Perl6" ? 10:02
diakopter "Perl 6, Thusly" 10:03
hejki "Lo and Behold - Perl6" 10:04
diakopter "Perl 6, Exploded. Just some of the Clusters, though" 10:05
hejki :P
10:05 hanekomu_ left
hejki "Practical Roundup for Perl6" 10:06
diakopter "Interring Perl 6"
masak "I made you a Perl 6, but I eated it"
diakopter or did I mean Entering
lol
10:07 Baggio_ left
diakopter "I CAN HAZ PERL6" 10:07
hejki "Perl6 makes the Cake real"
10:07 ejs1 joined
diakopter "Perl 6 Takes the Cake" 10:08
hejki "Perl6 ICT - Intensive Combat Training course with Chuck Norris"
carlin Iridescent Perl 6
Su-Shee Perl 6 Primer which we could shorten to p6p
hejki "Adolescent Perl6"
diakopter "Perl 6: A Piece of Cake"
hejki "Perl 6: A Cuckoo for Cac"
s/Cac/Caca/;
Su-Shee Diving into Perl 6 10:09
diakopter "The Nascent Perl 6". "The Latent Perl 6" "The Extant Perl 6" "The Burgeoning Perl 6"
mberends "Punning in Perl 6" - the #perl6 style guide
hejki :P
moritz_ Su-Shee: that's already the name of a blog about Perl 6
10:09 jferrero left
masak "The Truth About Perl 6" 10:09
Su-Shee moritz_: oh, never heard of it.
diakopter "Plumbing The Depths of Perl 6: No, this is not a Perl 6 Deep Dive" 10:10
hejki "Perl6.WHAT"
diakopter likes
"Perl6.HOW", and so forth, too
hejki ye
diakopter "Perl6.WHICH", even!
hejki "Perl6.ICANHASCHEEZBURGER - for advanced programmers" 10:11
carlin Nacreous Perl 6
diakopter "LOLPERL: formerly known as Perl 6"
carlin
.oO( CLAZZ foo { HAZ $.bar; }... )
diakopter std: SEZ "O HAI, WORLD" 10:12
p6eval std 29343: Undeclared name:␤ 'SEZ' used at line 1␤ok 00:01 106m␤
mberends ENEEDMOREGRAMMAR 10:13
carlin Pearlescent Perl 6
LOLCAT found where correct English expected
diakopter "The Splendiferous Perl 6" "Roundabout Perl 6" "Approximating Perl 6" "Approaching Perl 6" "Designing Perl 6" "Mutating Perl 6" 10:14
"Perl 6, ish" 10:15
"Perl 6, or somesuch"
hejki "Perl6-on-a-stick"
diakopter i kill me
10:15 Baggio_ joined
hejki "Perl6 - Schlumorp!" 10:15
moritz_ misses a minimum of seriousness
mberends afk & # serious things 10:17
10:17 ejs left
diakopter afk& # serious things like sleep 10:17
moritz_ now I scared them off :/ 10:19
diakopter o hai from sleep 10:20
"Perl 6, Delivered" 10:21
"Perl 6, Just In Time"
"Perl 6, Again" 10:22
"Perl 6, Yes, Really"
"Perl 6, We Mean It This Time For Realz Srsly" 10:23
hejki "Perl 6, Funky and Fresh" 10:28
masak "Perl 6, Seriously!" 10:30
10:31 JimmyZ left 10:33 c9s left 10:38 pmurias left
mathw "I HAZ A PERL 6" 10:43
masak it'll never make it past copy-editing. :P
apart from that, I like it. :) 10:44
wayland "Perl 6, and DNF -- well, one out of two"
masak "Perl 6 Using Your Hands" 10:45
10:45 wayland is now known as wayland76
masak "Dig into Perl 6" 10:45
carlin I was thinking Forever Perl 6
wayland76 The Perly Bird Get 6 worms
hejki :D
wayland76 ...and no butterflies
carlin Perl 6: Tales from the bikeshed
wayland76 carlin++ :) 10:46
masak :)
wayland76 'night all
mathw Well obviously, it's for the LOLSPEAK translation
10:46 TiMBuS left
hejki "Peeling Perl 6" 10:47
wayland76 Perl 6: Oh what a Peeling! :) 10:48
afk &
carlin Trawling for Perl 6 10:50
hejki "Crawing Perl6" 10:51
s/w/v/;
carlin Making a pun from pearl seems clever but also cliched 10:52
hejki same goes with Perlfect
hanekomu "Do They Know It's Perl6?"
carlin * Perl 6 10:57
to go with Rakudo *
mathw do cilre roda la perlxas. 11:02
Or something like that
my Lojban is terrible 11:03
11:07 agentzh left 11:09 Baggio_ left
jnthn o/ 11:18
mathw o/ 11:19
masak o/
carlin o/ 11:20
Infinoid o/
moritz_ \o 11:21
jnthn moritz++ # dares to be different :) 11:23
Friend who has been visiting for the weekend heads home later today...so will be back to hacking on Rakudo Stuff soon. :-) 11:24
mathw yay 11:25
11:26 TiMBuS joined 11:27 TiMBuS left, Baggio_ joined 11:39 envi^home joined 11:41 lestrrat is now known as lest_away 11:52 Baggio_ left 11:57 lest_away is now known as lestrrat 12:01 SmokeMachine joined 12:03 iblechbot left 12:10 Baggio_ joined 12:11 pnu left 12:13 Baggio_ left 12:15 Baggio_ joined, pnu joined 12:27 rgrau_`` is now known as rgrau 12:42 ejs2 joined 12:47 colomon joined 12:54 ejs1 left 13:00 pjcj_ joined, pjcj left 13:06 payload joined 13:08 colomon left 13:09 meppel left 13:11 meteorjay left
pmichaud "Perl 6: It's not for everybody (yet)... but you're not everybody." 13:12
13:12 rodi joined
moritz_ A slightly long title for a book :-) 13:13
masak "Hugs and kisses from the #perl6 crew." :)
pmichaud ...maybe a tagline more than a book title, then. Plus I stole it from the graduate studies program of the university I work at :)
mathw "Perl 6 for Adventuresome Types" 13:14
Wolfman2000 *yawn* Perl 6: The Revolution 13:15
I don't know
Wolfman2000 just recently woke up
pmichaud as counterpoint to "Perl 6 for Dummies", perhaps we should do "Perl 6 for Cool and Smart People" 13:17
masak could be seen as slightly elitist, I dunno... 13:19
pmichaud "Perl 6: The Adventure Begins"
mathw elitist is unfashionable
masak rakudo: class A {}; multi sub cmp(A $a, A $b) { 1 }; say (A.new, A.new).sort
pmichaud "Perl 6 Tamed"
p6eval rakudo 7ef386: ( no output )
masak locally, "Multiple Dispatch: No suitable candidate found for 'cmp', with signature 'PP->I'"
mathw "You have six onions" 13:20
masak submits rakudobug
pmichaud "Perl 6: Null PMC Access"
masak \o/
13:20 meppl joined
masak pmichaud: that one has my vote! 13:20
moritz_ pmichaud: but do we want to tame it?
pmichaud "Perl 6: Warp Speed" (probably can't get the rights from Paramount) 13:22
afk for a while... have lots of errands this morning :( 13:23
Su-Shee Rakudo and the six Perls 13:30
13:30 payload left 13:31 rodi left, meppel joined 13:36 jan` joined
masak std: for ^10 { break } 13:36
p6eval std 29343: ok 00:01 107m␤
masak TimToady: shouldn't that be a compile-time error?
13:36 jan` is now known as Guest43273
masak Do I correctly assume that 'augment class' changes a class within the current module scope, whereas 'augment slang' changes a slang only lexically? 13:37
moritz_ thought that 'augment' changes globally 13:38
Wolfman2000 I'll have to leave soon for winter break work. If there are any changes that need to be made to the (planned) last gift I have for the Advent Calendar, you have my permission to make them. 13:39
moritz_ hugme: tweet rakudoperl Perl 6 advent calendar day 14: Going to the Rats perl6advent.wordpress.com/2009/12/1...o-the-rats 13:41
hugme hugs moritz_; tweet delivered
masak moritz_: S02:2632
13:41 payload joined
moritz_ that doesn't seem to imply to me that the scope of 'augment' is limited to a package 13:43
masak what does it seem to imply, then?
13:43 meppl left
masak it's "lexically scoped". 13:43
moritz_ augment slang, yes 13:44
masak yes.
moritz_ sorry
I was not clear about that
I thought 'augment class' was global, 'augment slang' lexical
masak moritz_: ok. 13:45
so 'augment class' is not a 'language tweak' a la S11?
but introducing operators is?
Wolfman2000: great post! 13:46
Wolfman2000 No obvious errors this time? 13:47
masak Wolfman2000: '♥', '♣', '♦', '♠' -- have you considered using <>?
same with the 'A', '2'... above that.
that would solve the wrapping issue.
Wolfman2000 masak: ...right, should have remembered that. Good idea
masak also, I personally prefer Card.new(:$rank, :$suit) 13:48
otherwise, I see no nits to .pick :)
Wolfman2000 rakudo: @suits = <♥, ♣, ♦, ♠>; say @suits.join(" ");
p6eval rakudo 7ef386: Symbol '@suits' not predeclared in <anonymous> (/tmp/O39wFpA3Oh:2)␤in Main (file <unknown>, line <unknown>)␤ 13:49
Wolfman2000 rakudo: my @suits = <♥, ♣, ♦, ♠>; say @suits.join(" ");
p6eval rakudo 7ef386: ♥, ♣, ♦, ♠␤
13:49 mepplock joined
Wolfman2000 ...derp 13:49
moritz_ Wolfman2000: don't use , inside <...>
rakudo: my @suits = <♥ ♣ ♦ ♠>; say @suits.join("|");
p6eval rakudo 7ef386: ♥|♣|♦|♠␤
moritz_ ng: my @suits = <♥ ♣ ♦ ♠>; say @suits.join("|");
Wolfman2000 my post is updated
p6eval ng 0307c2: ♥|♣|♦|♠␤
moritz_ Wolfman2000++ 13:50
Wolfman2000 of course, I'm still surprised that this worked when I tested...
rakudo: my @dice = 1 5 10 20 50 100 500; say @dice.pick(10, :replace).join(" ");
p6eval rakudo 7ef386: Confused at line 2, near "5 10 20 50"␤in Main (file <unknown>, line <unknown>)␤
Wolfman2000 ...or I didn't copy it right
rakudo: my @dice = <1 5 10 20 50 100 500>; say @dice.pick(10, :replace).join(" ");
p6eval rakudo 7ef386: 500 20 20 500 10 20 1 1 500 10␤
Wolfman2000 alright, everything should be good from what I can tell. I can't guarantee I'll be here when the code is made public 13:52
Still, hopefully my lack of working directly with the building of Perl 6 wasn't a hindrance.
moritz_ don't worry, if something goes awefully wrong somebody else will fix it :-) 13:53
Wolfman2000 *nods*
mathw oh, moritz_, thanks for answering that question on my Junctions post
I was offline most of yesterday
moritz_ mathw: you're welcome
13:54 riffraff joined 13:55 JimmyZ joined 13:57 meppel left 13:58 ejs1 joined 14:01 ejs2 left 14:08 ejs2 joined 14:09 Exodist left, colomon joined, ihrd joined, ihrd left 14:10 ejs1 left 14:21 synth joined, synth left 14:22 Guest31990 left, lisppaste3 left, viklund left, hicx174 left, tarbo2 left, Guest69693 left 14:23 Guest31990 joined, Guest43273 left, Alias left, jrockway left, zostay left, frew left, slavik left, jlaire left, jan_ left, ashizawa left 14:24 viklund joined 14:25 iblechbot joined 14:26 viklund left, lisppaste3 joined, viklund joined, hicx174 joined, tarbo2 joined, viklund_ joined 14:27 Joda joined, jrockway joined, zostay joined 14:28 Joda is now known as Guest221, moritz_ left, moritz_ joined 14:29 tarbo2 left 14:30 c9s joined, tarbo2 joined 14:37 colomon left 14:51 elmex left 14:52 viklund left 14:56 synth joined, elmex joined, [synth] joined 14:57 mariano__ joined 15:02 alester joined 15:04 KyleHa joined, PacoLinux joined 15:06 [synth] left, jan` joined 15:07 jan` is now known as Guest11428 15:08 Guest31990 left 15:10 ejs2 left, takadonet joined
takadonet morning all 15:11
masak o/ 15:12
15:13 perlygatekeeper joined
takadonet masak: how are you? 15:13
masak takadonet: tired but happy. 15:14
takadonet: how are you?
takadonet masak: busy and want to hack on bioperl6
15:14 colomon joined 15:15 JimmyZ left 15:17 JimmyZ joined 15:20 Baggio_ left 15:24 colomon left 15:26 masak left 15:31 payload left 15:35 mariano__ left, mariano__ joined 15:39 Psyche^ joined 15:41 kaare joined, kaare is now known as Guest12394 15:43 Patterner left, Psyche^ is now known as Patterner, payload joined 15:46 beggars joined 15:50 jan_ joined 15:57 JimmyZ left, tylerni7 left 15:58 tylerni7 joined 16:11 nihiliad joined 16:13 astrojp left 16:22 jlaire joined, frew joined, ashizawa joined, slavik joined 16:30 SmokeMachine left 16:35 payload left, payload joined 16:36 zloyrusskiy joined 16:39 envi^home left 16:41 payload left 16:43 PerlJam joined 16:49 PerlJam left, PerlJam joined 16:52 masak joined
masak my computer crashed. 16:52
jeremiah yuck 16:53
TimToady that seems...suboptimal...
masak once again, it is an innocent victim of my Firefox tabs.
jeremiah All ten thousand?
TimToady but did it crash *fast*?
masak more like 2.5k.
jeremiah Which tab was the culprit?
PerlJam masak: www.google.com/chrome :-)
phenny PerlJam: 12 Dec 02:41Z <carlin> tell PerlJam there's a copy of your Benchmark.pm in perl6-examples lib and in its own repo, which causes problems for proto. Would you mind removing or renaming the perl6-examples copy?
masak PerlJam: :)
jeremiah Wow - you must have been reading a lot to get down to 2.5K
masak jeremiah: yes, and irreversible crashes brought it down a bit, too. 16:54
jeremiah :(
masak bah. win some, lose some :)
moritz_ masak: maybe you should just put less in your reading queue
16:54 payload joined
masak moritz_: but there are so many worthwhile and interesting things! 16:54
diakopter masak: how many of those open tabs are duplicates? :) 16:55
masak diakopter: some. not more than 0.5%.
moritz_ masak: but they are only interesting and worthwile if you actually get around to read them...
masak moritz_: well, I do. eventually.
jeremiah What about all the tabs that you don't even know about yet?
TimToady you sure you're not looking for "worthwhile OR interesting"? :P
moritz_ unless firefox crashes
masak jeremiah: whoa.
moritz_: firefox is good at bringing itself up with the tabs intact. 16:56
jeremiah Chrome does that too - but puts them in history
masak moritz_: as long as I unplug the ethernet cord while it tries to load them all in... :)
diakopter lol.
masak jeremiah: interesting. how does that work?
16:56 SmokeMachine joined
jeremiah So you have to click on a link to restore all your tabs 16:57
Which is nice because it avoids the firefox behavior
Which seems to be affecting your computer
16:57 _jaldhar left
masak jeremiah: only once every month or so. 16:58
jeremiah aha
16:59 _jaldhar joined
diakopter survey: if you were looking for an open-source/free-software garbage-collecting multi-stage/meta-programming multi-platform VM/compiler, which would you choose, and why? 16:59
moritz_ I'd ask diakopter :-) 17:00
jeremiah parrot - because it is the only one that fits that criteria
diakopter cheat answer :P
jeremiah ?
PerlJam jeremiah: uh ... no it doesn't.
jeremiah :)
PerlJam :)
jeremiah I was just guessing - I thought it was a trick question
diakopter you got that right
moritz_ probably depends on what you mean by "meta-programming" 17:01
PerlJam jeremiah: parrot would be my answer too (but only because I haven't looked at any other VMs)
diakopter runtime codegen in-process/in-memory
that can be invoked on the same working set
mberends Perl 5? 17:02
you didn't say multi-language :P
diakopter yeah, but any interpreter can be multi-language 17:03
:P
moritz_ Perl 5 is a very bad assembler language :-)
diakopter PerlJam: how doesn't parrot fit the criteria?
moritz_ parrot has garbage collection, but at times it's less-than-awesome 17:04
diakopter oh
PerlJam diakopter: I was being overly critical. If you twiddle the knob that says how successful parrot is at each of those criteria then parrot fits perfectly. 17:05
masak moritz_: parrot has speed, but at times it's less-than-awesome :)
TimToady parrot has *, but at time it's LTA. :) 17:06
*times
masak if Parrot has *, then it must be a perfect platform for Rakudo *. 17:07
TimToady I wonder if break would be easier to implement if we made it a statement_prefix... 17:08
diakopter V8 would fit the criteria, except it's not really a compiler stand-alone. it doesn't have a bytecode/IR to which it can persist/read. to persist state you have to build your script into an executable that links into libv8 using g++
masak swim &
17:09 masak left, solarion left
TimToady pictures masak testing a rock while swimming... 17:09
"yep, that one sinks fast!"
"I'd better let go of it..."
diakopter "we must be in the dead sea, b/c rocks float" 17:10
TimToady silly masak, you're testing pumice! 17:11
diakopter any other answers for my survey?
arnsholt I'm not at all familiar with the field, but the alternatives are more or less Parrot, JVM and CLR, no?
moritz_ mono 17:12
TimToady s/CLR/mono/ if you want FOSS
moritz_ you can look also in LLVM
17:12 Astoria` joined
moritz_ s/in/at/ 17:12
arnsholt Yeah, I'm not familiar with the terminology, but yeah the .Net VM thing
TimToady you could put it on top of Haskell... 17:13
arnsholt moritz_: Good point
mberends diakopter: since V8 is open source C++, could persisting the IR be added?
arnsholt TimToady: In that case you could add any Turing-complete language, no?
TimToady that was my next point
arnsholt Oh, right 17:14
Ignore my last then
diakopter mberends: yes, but there's no IR at all. JS source directly to machine code that runs upon loading. You'd have to bundle a g++ too
TimToady but Haskell turns most of those knobs pretty high
arnsholt Haskell is very cool indeed 17:15
moritz_ ... out of my reach
TimToady in spots
arnsholt moritz_: Same here. I've not managed to put in enough effort to really learn how to use it properly
diakopter new criterion: that doesn't require being able to write in Haskell, the Language of the Gods 17:16
TimToady does writing in pugs count?
diakopter :P
PerlJam I thought Perl 6 was the language of the gods?
moritz_ PerlJam: depends on the gods, really :-)
diakopter (kidding on the new criterion) 17:17
TimToady hemidemisemigods
takadonet xkcd.com/224/
PerlJam takadonet: exactly!
diakopter takadonet: oh yeah :D
re JVM runtime codegen/loading, how do folks do that? 17:21
arnsholt Re: Loading, that's handled by methods in java.lang.ClassLoader, AFAICT 17:26
As for compiling, I think you need either a compiler (if you want to compile at runtime, like JSP does) or a library that gives you a sane interface to the assembler language
17:26 payload left
arnsholt I don't think you have a byte code generation lib in the standard library, like C# has (from what I understand) 17:28
17:30 solarion joined
diakopter for .net/clr, there's an MIT-licensed addon library runsharp.googlecode.com that wraps even System.Reflection.Emit... it's what I've been experimenting with lately 17:31
clr/mono, I mean (I've been using mono) 17:32
for jvm, looks like www.gnu.org/software/kawa/ is a good tool 17:33
arnsholt Right 17:34
17:38 _jaldhar is now known as jaldhar
diakopter (specifically, the gnu.bytecode included in kawa) 17:40
obra TimToady: ping? I recall in some writing of yours, you've stated that one of your biggest regrets about p5 is that 'use strict' isnt' the default. Am I on crack? Can you think of a convenient citation? 17:41
mberends diakopter: did you need answers, or were you just teasing us to see if we also know what you already know?
TimToady I don't think I ever said that about P5, except insofar as P6 does it differently 17:42
obra ok
TimToady I maintained for many years that was good for people to pick righteousness explicitly :)
*that it
obra nods. 17:43
TimToady but p6 has enough new righteousnes to pick explicitly already :)
diakopter mberends: both
obra I'm in the midst of a discussion about p5 policy and "upgrading" the defaults on users in a newer release of Perl 5, such that code that previously worked needs a "use foo" statement to make it keep working. 17:44
PerlJam TimToady: Are you sure you don't want to herd people towards righteousness with a sword? That's been used with great success in the past. ;->
obra That...makes me sad.
diakopter the discussion makes you sad, or the proposed "upgrading" policy? 17:45
obra diakopter: actually, the arguments against my "we should do everything we can to make sure old code keeps working" policy :) 17:46
mberends it depends what kind of damage results from breaking the old code 17:48
obra mberends: you mean whether they come after us with pitchforks or with knives? 17:49
TimToady
.oO(It's the End of the World as we Know It, and I feel fine...)
obra grins at TimToady
mberends heh, perhaps just the cost and time to rewrite the old bad bits 17:50
TimToady "Hi, I see that you have a bad bit at line 42; rewriting it for you..."
diakopter yeah. unfortunately, I'm in agreement with you. The Slippery Slope mindset is what maintainers must by necessity listen to: "if they're introducting [backward-compatability-breaking-change X] now, that basically means perl is no longer supported." Otoh, I agree fully with the argument that the systems that have the "old" code will almost never be upgraded anyway, ever... But it's a branding/marketing issue, unfortunately (for those who want to improve
obra TimToady: That would be a lovely solution 17:51
TimToady well, except for the little matter of write access :)
arnsholt Isn't there also precedent for keeping backwards compatibility?
TimToady P5 specializes in keeping; P6 specializes in breaking it. :) 17:52
*keeping it
obra diakopter: I have people installing 17:54
software I wrote in 1997 on modern machines running RedHat Enterprise 5.3 with Perl 5.10 17:55
and it all works just fine :)
TimToady: I could see a runtime rewrite and a "recompiled code cache"
17:56 PacoLinux left
diakopter obra: I hope I was clear that I thought it was a branding issue (that backward compatibility must be maintained) 17:56
obra Nod
diakopter but from where are they obtaining the 1997 software? CD-ROMs? or cpan? or another website? 17:57
mberends the cost of backward compatibility accumulates over time; one solution is to sneak reporting instrumentation into the old infrastructure, using its trace logs to gradually eradicate/replace the undesired parts.
obra diakopter: upgrading existing systems. 17:58
diakopter oh... you said installing. 17:59
17:59 stephenlb joined
obra new system. new copy of the same code they were already running. on a new Perl with deps from cpan, etc. 17:59
diakopter I hope that's a paid support contract :) b/c I wouldn't want to have to worry about that otherwise; newer deps from cpan break compatibility (often unintended) often.. 18:01
obra shrugs 18:02
I still don't want to screw over end users. 18:03
diakopter still, I don't see why both modes can't be supported going forward...? a local/site flag?
obra Hurting people makes them not like you.
diakopter: the issue is entirely one of "what does perl do without you Explicitly saying you want a certain behavior"
diakopter (with the backward-compat mode still the default)
obra right. your parenthetical expression is the topic of the debate.
diakopter I'm referring to a system-wide default 18:04
moritz_ I've been submitting patches for dev.perl.org/perl6/ for a while, and now the question has come up if we should remove links to pugscode.org there
quick poll: [A] leave as is [B] remove most, but not all [C] remove them all 18:05
mberends B
moritz_ votes for B, but would love to hear other opinions too
mberends moritz_: it's ok, two makes a quorum ;)
diakopter B 18:06
(but first, I see only one link there anyway...)
obra diakopter: right. Should the systemwide default be the new "better" behavior or the behavior that existed before we "fixed" it? 18:07
moritz_ some have been removed already, some are hidden on subpages
diakopter incl the subpages
there's one at the bottom of the FAQ
actually, no, it's not even a hyperlink 18:08
<a name="www.pugscode.org/"
and then the 1 actual link on the Status page 18:09
architecture.html definitely needs mothballed 18:10
18:11 orafu left, orafu joined
diakopter the copy on status.html needs lots of tense changes 18:11
"Internals" on status.html needs rewording/rewritten 18:12
moritz_ if anybody wants to help, the sources can be obtained with svn co svn.perl.org/perl.org/docs/live/dev
18:14 PacoLinux joined 18:15 colomon joined
diakopter which pugscode link was proposed to remove? 18:16
18:19 ShaneC joined
moritz_ Leo asked if we should remove them all 18:20
diakopter does he want all the parrot links removed too? 18:21
moritz_ nope
let me give you a bit more context
somewhere on one of the pages it said "for latest news re perl6, visist [...] www.pugscode.org" 18:22
a patch of mine removes that, reasoning that there haven't been any news on pugscode.org for years
so he asked if we should remove all links to pugscode.org
diakopter ok, but I don't see that anywhere in the svn checkout 18:23
your patch was already committed?
moritz_ index.html, perl6/status.html, perl6/faq.pod
diakopter: don't know, I sent two patches, one was applied already, the other not 18:24
and right now I have no overview what was in which patch
diakopter on the live site (and in svn) there is just the 1 link to pugscode, and then the other fake link on FAQ
18:25 cotto_work joined
moritz_ I see two links in svn 18:25
index.html
51: <td><font size="+2"><a href="www.pugscode.org/">Pugs</a>...gt;</td
perl6/status.html
46:<a href="www.pugscode.org/">Pugs</a> - an experimental implementation of Perl 6. Run Perl 6 today!
diakopter ok 18:26
why would Leo want to remove the Pugs links but not the Parrot links?
moritz_ probably because I told him it contains no news 18:27
18:27 cdarroch joined
moritz_ anyway, Leo asked, I'll answer that not all links should be removed 18:27
18:28 zloyrusskiy left
coke apparently not. bother. 18:30
ww
18:30 coke is now known as Coke
diakopter moritz_: how shall I submit patches for the site? 18:33
moritz_ diakopter: subscribe at lists.scsys.co.uk/cgi-bin/mailman/l...rg-patches 18:36
and maybe look at lists.scsys.co.uk/pipermail/perl-or...hread.html for my first two patches
18:39 riffraff left 18:40 rgrau` joined 18:52 Coke left 18:55 justatheory joined
IllvilJa If I want to handle YAML docs in rakudo perl6, what are my best options? Is it already built into the language, is it a core library already included or should I install some Perl6 module from github or elsewhere? 19:02
moritz_ you should first write a YAML library, and then use it 19:04
PerlJam Didn't someone code a small YAML lib in Perl 6 already? 19:12
moritz_ I'm not aware of that 19:20
only of JSON
19:21 KeithWolters joined
PerlJam Maybe I'm thinking of the JSON one then (they tend to occupy the same locality in my brain) 19:23
moritz_ when you try to implement them, you'll notice very soon that they are very different beasts :-) 19:24
PerlJam aye, they're used in similar contexts which is why they're close in my brain. 19:25
moritz_ me attempts to teach people that data serialization languages and data exchange languages fill different niches don't seem to very effective 19:27
PerlJam heh
It doesn't matter what niche they were designed for or are most useful for; people will use them however they please. 19:28
19:30 rachelBROWN left
moritz_ aye 19:30
19:33 thowe joined
thowe Hello. 19:33
moritz_ hi thowe
Su-Shee JSON is YAML, btw. 19:39
moritz_ JSON is a tiny, well-defined subset of the huge multiverse of YAML 19:40
19:49 rjh_ left
diakopter does anyone know whether either/both of the last two State of the Onion talks (video, audio, transcript/slides) are available online (or offline!) anywhere??? 19:54
I'll keep asking every week until someone who knows replies... ;)
IllvilJa moritz_: PerlJam: I assume then that the currently available JSON implementation in Perl6 is the most "whippitupitudical" approach to get some basic management of structured data on storage as well as interchanged between processes [potentially via a storage media ;-) ]. 19:57
moritz_ IllvilJa: right. It's just limited to numbers, booleans, undef, arrays and hashes 19:58
IllvilJa moritz_: JSON should do then. I'm actually quite accustomed to XML, but I don't know how well THAT beast is supported in Perl6. 19:59
XML is a heavyweight monster that actually cooperates with me. 20:00
moritz_ there's an XML grammar somewhere on github, but to be fair proper XML support consists of far more than just a grammar 20:01
20:01 Exodist joined, riffraff joined 20:02 vamped joined
mberends IllvilJa: the reference to YAML that you recall is probably about proto, which uses a teeny subset of the format in a very limited way 20:04
and that subset is not the same subset as JSON :( 20:05
vamped rakudo: my @a = 10 ... * +1,20; @a.perl.say 20:07
p6eval rakudo 7ef386: [10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20]␤
vamped rakudo: my @a = 1.0 ... * +.1,2.0; @a.perl.say
p6eval rakudo 7ef386: [1, 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9]␤
vamped anyone know why there is a difference here, above: ? 20:08
moritz_ vamped: probably floating point/rounding errors
colomon presume it's the good old... what moritz said.
moritz_ rakudo: my @a = 1.0 ... * + 1/10 , 2.0; @a.perl.say 20:09
p6eval rakudo 7ef386: [1, 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9]␤
colomon rakudo: my @a = 1 ... * + 1/10, 2; @a.perl.say
Tene rakudo: my @a = 1.0 ... * +.1,2.001; @a.perl.say
p6eval rakudo 7ef386: Multiple Dispatch: No suitable candidate found for 'cmp', with signature 'PP->I'␤in Main (file <unknown>, line <unknown>)␤
rakudo 7ef386: [1, 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2]␤
moritz_ that did it
colomon that was Tene's I think.
series operator hates me. :)
vamped I was thinking that was a possibility, but ideally I would expect it to work as it was.
moritz_ vamped: it will, in a few months 20:10
vamped: we have a branch where 1.0 and 0.1 are Rats too
colomon vamped: floating point numbers just don't work that way.
vamped ok. cool. speaking of ng?
moritz_ it just needs a lot of other things being worked on before we can merge it
vamped: yes
diakopter ng: my @a = 1.0 ... * +.1,2.0; @a.perl.say
p6eval ng 0307c2: too few positional arguments: 3 passed, 4 (or more) expected␤current instr.: 'perl6;Code;new' pc 11596 (src/builtins/Positional.pir:142)␤
colomon though actually, it seems to me that series operator ought to include its last argument in the list always, no matter what? 20:11
diakopter: that's NYI in ng.
if you're bored, you're welcome to implement it. :)
vamped and do I understand correctly that ng will probably not be merged by this month's release?
moritz_ colomon: yes, but if the accumulated value is larger than the limit, it won't be included
vamped: pmichaud hasn't decided finally on that (afaik), but I think it's highly unlikely 20:12
heck, we don't even have working hashes or ranges in ng
colomon to make it reasonable, ng would have to pass about 30,000 more tests in the next two days.
PerlJam colomon: it could happen :)
diakopter I thought LTM in nqp-rx was the holdup 20:13
the next big thing, I mean
Tene there are some things blocking on lists API refactor.
moritz_ diakopter: that's postponed for when it's actually needed
colomon no hashes, can't pass by array, lists don't really work, ranges don't work, no list assignment... 20:14
PerlJam colomon: I've known pmichaud to not sleep for really long periods of time ... it could still happen :) 20:15
colomon PerlJam: I can maybe believe he could take care of all the big blockers, but that still leaves rebuilding the core (aka setting) and I don't think you can run spectest fast enough to make all the needed changes in 48 hours. :) 20:16
RichiH hmm, today's advent calendar does not go into details about what causes `say pi.Rat` and `say pi` to print different values 20:17
PerlJam I'm going to continue to wear my optimist hat this week regardless.
colomon RichiH: it's implied, sorry. 20:22
basically, if you say pi.Rat, what you get is an approximation of pi as a Rat which is accurate to 1e-6 (I think). 20:23
RichiH colomon: well, because one is Rat and the other one a mere Num
colomon so the first six digits or so of pi.Rat and pi should match, and then they will diverge.
RichiH but i would have thought that pi is stored in a way that yields the same result in both contexts
colomon I guess the quick answer is, it definitely isn't. 20:24
vamped rakudo: say pi.Rat.perl
p6eval rakudo 7ef386: 355/113␤
colomon in master, pi is a Num.
as spec'd and in ng, pi will be a Rat. 20:25
(At least, I think the spec implies pi is a Rat.)
I guess once pi is a Rat, then pi.Rat will exactly equal pi. :)
(though pi isn't defined at all in ng yet.) 20:26
vamped rakudo: say pi.Rat(1e-14).perl
p6eval rakudo 7ef386: 80143857/25510582␤
colomon yeah, that might do it for accuracy. 20:27
rakudo: say pi.Rat(1e-14)
p6eval rakudo 7ef386: 3.14159265358979␤
vamped RichiH: I think that part of the post explains it. Yes?
colomon rakudo: say pi;
p6eval rakudo 7ef386: 3.14159265358979␤
sjohnson rakudo: my @a = <1 2 3>; for @a { $_ = 'x' }; say @a.perl 20:28
p6eval rakudo 7ef386: ["x", "x", "x"]␤
vamped RichiH: Rat defaults to 1e-6
RichiH vamped: hmm, true
sjohnson what is the best way to avoid "writing back" to the @ var? anyone? :] 20:29
vamped is very excited about decimal numbers being represtented as Rats 20:30
moritz_ sjohnson: use an explicit signature
rakudo: my @a = <1 2 3>; for @a -> $a { $a = 'x' }; say @a.perl
p6eval rakudo 7ef386: Cannot assign to readonly variable.␤in Main (file <unknown>, line <unknown>)␤
sjohnson hmm 20:32
moritz_: is this new to p6?
moritz_ sjohnson: yes 20:33
sjohnson moritz_: the p5 way is just doing $tmp = $_ i suppose 20:34
that is kinda cool though 20:36
moritz_: what about (in p6) if you wanted to actaully do the modifications i did, but not "write back" to the foreach array?
but just change in side the { } scope willy nilly 20:37
i suppose $tmp = $_ is the only way in P6 too? 20:41
moritz_ no
'is copy
rakudo: my @a = <1 2 3>; for @a -> $a is copy { $a = 'x' }; say @a.perl
p6eval rakudo 7ef386: ["1", "2", "3"]␤
20:43 dakkar joined
sjohnson wow, this is pretty cool 20:43
moritz_++
moritz_ not my idea :-) 20:44
sjohnson always ++ the messenger is my P6 motto 20:45
looks like i really can't think up any good ideas that haven't already been done to p6
20:47 am0c left 20:49 coke joined
sjohnson moritz_: is there a safe and proper way, to "is copy", and use the $_ var without too much extra writing? 20:49
will just "is copy" be safe + smart?
moritz_ $_ is copy works 20:50
sjohnson rakudo: my @a = <1 2 3>; for @a is copy { s/./x; }; say @a.perl 20:52
p6eval rakudo 7ef386: Confused at line 2, near "is copy { "␤in Main (file <unknown>, line <unknown>)␤
sjohnson rakudo: my @a = <1 2 3>; for @a $_ is copy { s/./x = 'x' }; say @a.perl
p6eval rakudo 7ef386: Confused at line 2, near "$_ is copy"␤in Main (file <unknown>, line <unknown>)␤ 20:53
sjohnson rakudo: my @a = <1 2 3>; for @a {$_ is copy} { s/./x = 'x' }; say @a.perl
p6eval rakudo 7ef386: Confused at line 2, near "is copy} {"␤in Main (file <unknown>, line <unknown>)␤
vamped sjohnson: for @a -> $_ is copy {...}
sjohnson is it impossible to implement an for @a is copy syntax? 20:55
to omit writing $_ if it's $_ ?
moritz_ the question is not if it's impossible
but if it's really worth the trouble
20:56 Su-Shee left
vamped maybe a .copy method should be invented, so: for @a.copy {...} 20:57
colomon std: my @a = <1 2 3>; for @a is copy { s/./x; }; say @a.perl
p6eval std 29343: ===SORRY!===␤Missing block at /tmp/MwjdXT7nB4 line 1:␤------> my @a = <1 2 3>; for @a ⏏is copy { s/./x; }; say @a.perl␤ expecting any of:␤ infix or meta-infix␤ infix stopper␤ parameterized block␤ standard stopper␤ terminator␤FAILED
..00:01 110m␤
20:58 KeithWolters left
sjohnson it is kind of cool when you can omit the $_ ... but that's just me i guess 20:58
coke is .panic a standard thing, or only NQP?
moritz_ STD.pm uses that too 21:00
so it's quasi standard
diakopter I think panic is in the synopses somewheres 21:02
21:02 gabiruh_ is now known as gabiruh
diakopter oh wait, or mabye it was removed 21:03
coke is this legal? nopaste.snit.ch/19081 21:04
if I try to put a variable in the .panic, I get an error during compilation about the angle bracket. 21:05
diakopter what does ** mean there
moritz_ separated by
coke: I think it should work, but maybe it's an nqp limitation 21:06
diakopter I knew that, last week :)
coke moritz_: any suggestions on alternate spellings that might work? =-)
moritz_ coke: if interpolation doesn't work either... no idea 21:08
coke is "$<error>" the right way to spell that? 21:09
(that didn't work either.)
moritz_ it is
coke I can only get literal strings to work. (but comma separating /them/ works fine.)
21:10 riffraff left
coke with "$<extra>", it compiles, but then doesn't actually interpolate at runtime. 21:10
coke finds the nqp-rx wishlist...
21:15 riffraff joined 21:17 jferrero joined
coke moritz_: added to wiki.github.com/perl6/nqp-rx/requests, danke. 21:17
21:18 SmokeMachine left, SmokeMachine joined 21:19 riffraff left 21:22 jaldhar left
vamped to install Rakudo on a computer (Ubuntu) without an internet connection, can I install on another computer (Ubuntu) and copy directory tree over? Are there any limitations to this method? 21:27
moritz_ vamped: if it's the same architecture (i386 vs. amd64) it should work
wayland76 Are they the same version of Ubuntu? That helps too. 21:28
Packaging systems do essentially what you're talking about
vamped will be the same version of Ubuntu. I don't know now if the other computer is same architecture.
wayland76 We can't guarantee it will work, but it's got a good chance
[particle] there are parrot packages for debian 21:29
are there rakudo packages for debian yet?
wayland76 vamped: It's also possible to just download the debian .deb file, and then use "dpkg -i <packagefile>"
moritz_ [particle]: in experimental, I think 21:30
vamped wayland76: there's a rakudo deb? where? 21:32
moritz_ packages.debian.org/rakudo 21:34
2009-09 - ancient already :-)
vamped :( I think I'll just try copying directory tree (if it's same arch, otherwise ??) 21:35
moritz_ otherwise: rsync or copy rakudo and parrot tree over, and build from source
instead of using --gen-parrot, you can also 21:36
cd parrot/; perl Configure --prefix=$path_to_rakudo/parrot_install; make install
cd ..
perl Configure.pl
make install
vamped vielen Dank, moritz.
moritz_ bitteschön :-) 21:37
21:45 Guest12394 left 21:48 Intensity joined, tarski left 21:49 meppel joined
coke cries, as the machine that he had unicode all setup on is now dead, so he's stuck with ?'s again. =-) 21:55
moritz_ coke: perlgeek.de/en/article/set-up-a-cle...nvironment unfinished and unreviewd, but maybe still of help 21:56
diakopter coke: ?? ? ???? ?????? ?
(kidding) afk&
21:58 mepplock left
coke moritz_: eh. I'll focus energies on resurrecting the dead machine. 21:58
but keep that for future reference, thank you! 21:59
but *will keep...
wayland76 I don't know that there is one. I guess I just assumed there was 22:03
vamped: ^^
22:03 mepplock joined
wayland76 Coke: Þát´s ñó góød 22:04
(Sorry :) )
moritz_ the IRC logs show such characters usually correctly, so if he's *really* curious he just needs a working browser 22:05
Tene I guess I've just been spoiled by fedora... 22:06
vamped wayland76: tháńks änÿwæy
wayland76 Well, but if the browser doesn't support unicode either... :) 22:07
Oh, and s/Þat/Ðat/, sorry :) 22:08
vamped @Tene: I'm too spoiled by Ubuntu to go back to Fedora ;) -- but that's a topic for another forum :D 22:09
wayland76 I'd argue that all the packaging systems are rubbish :)
moritz_ if you don't want downgrades, dpkg/apt is great :-) 22:11
22:11 mariano__ left
wayland76 I want a packaging system that works on every OS, so that my perl programs can be packaged once, run anywhere 22:15
I want a packaging system that can be relied upon by other software; mozilla plugins? Use a package. <other software> plugins? Use a package. These pieces of software should just be able to call upon a library function to generate themselves a plugin manager 22:16
moritz_ if you write one, I'll test it for you :-)
aye, plugin systems have been reinvented far too many times 22:17
diakopter the cpan client needs a special mode where it detects being run as root, and detects whether it's set to use the system package systems, and (if possible) builds/installs packages for the distributions it's requested to install/upgrade
dpkg/yast/yum/pkgadd
port on macs
moritz_ automatically shelling out to CPANPLUS::Dist::*
22:18 pmurias joined, SmokeMachine left, meppel left
wayland76 Oh, yeah, I also want it modular; specifically, I want to be able to replace/modify the algorithm that chooses which packages to install 22:18
BinGOs I use pkgsrc on Darwin, good luck forcing me to use ports 22:19
moritz_ but I guess it'll work the other way round in the end
pmurias diakopter: re looking for a vm what are you planning to implement?
diakopter BinGOs: (or whatever)
moritz_ large applications will ship as VMs
erm, VM images
wayland76 Do I want to delay my software packages a week behind the releases? Can be done. Do I want to make one of my repos take priority over another, even if there's a more recent version? Can be done :)
BinGOs edge-cases are going to burn.
moritz_ and maintain their own environment
wayland76 Oh, yeah, did I say I also want to support both source and binary distros? :) 22:20
moritz_ and world domination, too!
diakopter wayland76: well yeah :P
wayland76 Well, *of course* world domination :)
diakopter integration of usually-bundled C lib dependencies would be nice too 22:21
while we're dreaming...
wayland76 diakopter: I don't understand 22:22
integration of C lib dependencies with what?
with the C lib dependencies of other packages? 22:23
moritz_ XML::LibXML depends on libxml-2.0
diakopter exactly
(what moritz_ said)
moritz_ -dev, of course :-)
wayland76 But are you saying you want to bundle libxml-2.0 into the same package?
diakopter I'm saying it would be nice if a distribution that depended on libxml-2.0 could tell cpan to use the system libxml-2.0 if available, to optionally install a system libxml-2.0 if available, and to optionally fallback to building/installing a bundled libxml-2.0 .tar.gz (optionally with custom patches & install destination) 22:25
moritz_ for me declaring the dependency would be enough
22:26 patspam joined
pmurias diakopter: bunlding a C lib is not very elegant 22:27
* bundling
diakopter yeah, but plenty do it
(already)
I'm trying to provide other options
moritz_ more elegant than dying with an error message that only geeks understand, if at all
pmurias but installing the appropriate package via the package manager would be great 22:29
slavik will parrot have a configurable library loader? 22:32
22:33 vamped left
diakopter slavik: there's this github.com/quietfanatic/link-c 22:36
slavik diakopter: something like that 22:37
22:37 alfieANDfred joined, alfieANDfred left
slavik so that it would be possible to define the loading policy and libraries 22:37
basically, parrot should be able to do what the jvm can do, but not use java as the language 22:38
wayland76 My idea is that CPAN should simply be another repository that declares dependencies, just like anything else. 22:39
diakopter moritz_: I agree with what you said about large applications..
pmichaud Right now I'm leaning towards not merging ng for the release 22:42
I'd rather do good work over time than rush to get things in
PerlJam pmichaud++
moritz_ sounds sensible 22:43
diakopter is LTM still planned for nqp-rx?
pmichaud planned, yes, but how much we do depends more on what we need for Rakudo *
diakopter oh
wayland76 pmichaud: Didn't I say something about February for LTM? :) 22:44
pmichaud wayland76: you probably did. If there's a LTM feature that I find we absolutely have to have for Rakudo*, I'll undoubtedly expedite it
right now my focus is more on getting ng landed and making sure we're covering the items in the ROADMAP 22:45
22:45 am0c joined 22:46 pmurias left
wayland76 Yeah, I did. You were saying that it would be done by the end of Summer 2009. I said February 2010 :) 22:47
pmichaud well, I have sufficient LTM for us to get protoregexes working
PerlJam wayland76: So ... which christmas will Perl 6 be "released"?
pmichaud PerlJam: 2008
wayland76 I can't tell until pmichaud makes a prediction. Then I'll double that :) 22:48
PerlJam tries to tap into wayland76's clear psychic abilities
wayland76: so, christmas 4016? :)
22:48 KyleHa left
wayland76 sufficient LTM for protoregexes counts as something 22:48
22:48 Astoria` left
wayland76 Ok, what I really mean is (wayland76 predicted time) = (pmichaud's predicted time) - (date of prediction) 22:49
whoops
Ok, what I really mean is (wayland76 predicted time) = [ (pmichaud's predicted time) - (date of prediction) ] * 2 + (date of prediction)
sufficient LTM for protoregexes counts for a lot, I meant. pmichaud++ 22:50
pmichaud: Are you meaning that -ng should be merged after the December release? 22:51
22:56 dakkar left
pmichaud wayland76: I'm simply meaning it likely won't happen before 22:59
moritz_ in git, how can I find the point where two branches were... well, branched? 23:02
or the first differing commit, or so
PerlJam moritz_: git merge-base A B
moritz_ PerlJam: thanks 23:03
PerlJam moritz_: try git merge-base master ng # for instance
or you can use SHA1s
23:12 iblechbot left
moritz_ I looked in rakudo master if there's anything worth cherry-picking to ng 23:19
didn't find anything really
most commits in master were made to the setting 23:20
and despite all its smartness, git doesn't seem to like to cherry-pick patches to renamed files
23:34 xenoterracide left 23:35 pmurias joined 23:40 KyleHa joined 23:47 KyleHa left 23:59 LionMade1fLions joined, Exodist left, cotto_work left, elmex left, Guest221 left, cotto left, Wolfman2000 left, kcwu left, IRSeekBot left, revdiablo left, carlin left, wolverian left, BooK left, rgrau left, r0bby left, mathw left, draxil42 left, bloonix left, ruz left, omega__ left, LionMadeOfLions left, sunnavy left, idemal left, pmichaud left, avuserow 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!