Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2010-04-06

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:00 lue frettled: good night
00:00 TimToady \o
00:00 * ruoso calms TimToady by remembering him Math::Symbols wouldn't be in the core
00:00 yinyin joined #perl6
00:00 TimToady Yes, well, horror flicks are optional too...
00:01 TimToady but I'll still have to look at all the trailers...
00:02 TimToady "This trailer approved for all audiences between the age of 11 and 14."
00:03 [particle] joined #perl6
00:03 diakopter hundred.
00:04 TimToady I guess that leaves out Columbus...
00:04 TimToady oh wait, he *was* the horror show back then...
00:04 lue This movie is rated WTHC, for it's utter boringness (must be a documentary or movie involving Rubies)
00:09 ruoso std: multi prefix:<∑>(*@items) { [+] @items }; say ∑ 1..10
00:09 p6eval std 30324: OUTPUT«ok 00:02 112m␤»
00:09 ruoso rakudo: multi prefix:<∑>(*@items) { [+] @items }; say ∑ 1..10 # lucky guess
00:09 p6eval rakudo c41cf3: OUTPUT«12345678910␤»
00:10 ruoso not quite as expected... but it didn't die
00:10 ruoso rakudo: multi prefix:<∑>(*@items) { [+] @items }; say ∑1..10 # maybe rakudo doesn't like the space
00:10 p6eval rakudo c41cf3: OUTPUT«12345678910␤»
00:10 ruoso nah...
00:11 TimToady looks like a precedence issue
00:11 ruoso rakudo: multi prefix:<∑>(*@items) { [+] @items }; say ∑(1..10)
00:11 p6eval rakudo c41cf3: OUTPUT«55␤»
00:11 ruoso ahá!
00:15 Guest83313 joined #perl6
00:16 lue I've been considering a class-based method of handling numbers (arranged like a dependency tree). It would probably exist only as a module, but still...
00:17 ruoso lue, nope... you don't want to think it class-based
00:17 ruoso because that's not how math works
00:17 ruoso you need to think in subtypes
00:18 ruoso applying roles is a saner way
00:19 TimToady maybe instead of Top and Bottom the type theorists should have grabbed the Truth and Beauty that were abandoned by the physicists...
00:21 * TimToady imagines the billboards: Come learn type theory, where everything is true, and nothing is beautiful!
00:21 lue ruoso: I mean in terms of types. i.e. My Int 3, My Complex 2+5i, etc. However you're supposed to implement types. :) [check the backlogs; it's described in more detail there]
00:22 lue menowthinks I don't want to know about type theory.
00:22 * TimToady apologizes for having his brane in sideways today.
00:22 sorear <3 type theory
00:22 sorear however it has no place in perl6
00:23 * justatheory loves to type "theory:"
00:23 justatheory s/://
00:23 ruoso TimToady, oh... so is it a good day for you to understand me ;) since I'm the one usually thiking sideways
00:23 TimToady I rather suspect we're out of phase
00:23 ruoso heheh
00:24 ruoso good exit
00:24 sorear We're trying to design a worthy successor to Perl here.
00:24 sorear not Haskell With Sigils
00:25 lue .oO(sought Haskell Sith Sigils)
00:25 TimToady someday I might retire to a monadic lifestyle...
00:25 * ruoso usually thinks Perl 6 as very innovative in terms of its type theory
00:26 ruoso it's not even written down yet ;)
00:26 TimToady its innovation is in not forcing it down the user's throat.  :)
00:26 ruoso we need someone to do research on Perl 6 typesystem and maybe sometime we find out what it is ;)
00:26 * ruoso not really in a productive da
00:26 ruoso *dau
00:26 ruoso *day
00:27 TimToady I think we're in phase on that one
00:28 TimToady in my case, I blame the backlog
00:28 TimToady or maybe the Temporal Lobe epilepsy... :)
00:29 lue Oh yeah, for clarity, I need to insert something into the backlog for yesterday. Be right back...
00:29 lue .oO(I'm back.)
00:29 TimToady wait, why is there frost all over my DeLorean?
00:30 xuser left #perl6
00:30 lue Because it hasn't been used in a while (my ship is the TARDIS, thank you very much)
00:30 * lue wanted to fly to Britain yesterday just to watch the new Doctor Who
00:31 lue They have to admit, they couldn't design the DeLorean without it knowing it'd be used in a sci-fi movie...
00:33 TimToady well, but they've known that since some time in the 1800s
00:35 lue .oO(if the speed of light lets you go forward in time faster 1sec:1sec, then logically speed of dark lets you go back in time faster that 1sec:1sec)
00:35 TimToady "for one so forward, you're far too backward"
00:36 TimToady from TROTGTSOTC
00:36 lue chrot-git-sot-k?
00:38 TimToady my goodness, google *doesn't know* about TROTGTSOTC!
00:39 * ruoso feels a bit less idiot ;)
00:39 TimToady http://en.wikipedia.org/wiki/The_Roar_of_the_​Greasepaint_%E2%80%93_The_Smell_of_the_Crowd
00:40 lue The Right of The Great Tim "So Overwhelmed" Toady Command?
00:41 colomon Funny, vaguely familiar with the phrase, but very startled to learn it's a Leslie Bricusse musical.
00:43 colomon and ditto for "Stop the World, I Want to Get Off".
00:43 TimToady I am actually in the small minority of people who have actually helped perform that musical...
00:43 TimToady (was just in the orchestra)
00:43 colomon I was just pondering asking if anyone ever produced it anymore....  ;)
00:44 lue Take all the people that have helped with a musical, and it's still a minority ( in relation to the whole planet :) )
00:44 TimToady I saw a production of it in LA about 15 years ago at, I think, CSUN
00:44 colomon Wait a minute (surfing wikipedia) Bricusse did a version of Cyrano with Frank Wildhorn?  How did I miss that?
00:45 como joined #perl6
00:47 TimToady obviously, you missed it by a nose
00:49 diakopter flare your nostrils next time
00:50 lue [this is a placeholder to indicate a joke was meant to be here. Unfortunately, attempt to retrieve the joke returned HTTP 501]
00:50 lue [laugh here]
00:56 colomon Correct non-nose-wit answer appears to be that it has only been produced in Japan so far.  :\
00:57 lue Japan++
00:57 lue buubot: karma Japan
00:57 buubot lue: Japan has karma of 3
00:57 lue :O
01:00 lue rakudo: say (* - 1) ~~ *
01:00 p6eval rakudo 3b1d34: OUTPUT«1␤»
01:01 TimToady * is very accepting
01:02 TimToady in fact, it's the very picture of a brane that has fallen out and landed on the concrete
01:02 lue rakudo: my $fakereal = (-Inf..Inf); say ($fakereal - 1 ~~ fakereal)
01:02 p6eval rakudo 3b1d34:  ( no output )
01:03 lue rakudo: my $fakereal = (-Inf..Inf); say ($fakereal »-» 1 ~~ fakereal) # I can has hyper?
01:03 colomon nope
01:03 p6eval rakudo 3b1d34:  ( no output )
01:06 TimToady rakudo: my %h = a => '42'; given %h { when :(Int :$a) { say "lol int" }when :(Str :$a) { say "lol string" } }
01:06 p6eval rakudo 3b1d34: OUTPUT«lol string␤»
01:07 TimToady std: my %h = a => '42'; given %h { when :(Int :$a) { say "lol int" }when :(Str :$a) { say "lol string" } }
01:07 p6eval std 30324: OUTPUT«[31m===[0mSORRY![31m===[0m␤Strange text after block (missing comma, semicolon, comment marker?) at /tmp/rvy2a97g3U line 1:␤------> [32mn %h { when :(Int :$a) { say "lol int" }[33m⏏[31mwhen :(Str :$a) { say "lol string" } }[0m␤    expecting horizontal whitespace␤FAILED
01:07 p6eval ..00:01…
01:07 lue that's not good :/
01:08 lue .oO(STD is not prone to error. It must be rakudo who's wrong)
01:14 lue rakudo: my $ℝ = 3; say "$ℝ is a real number!"
01:14 p6eval rakudo 3b1d34: OUTPUT«3 is a real number!␤»
01:14 pugssvn r30325 | colomon++ | [t/spec] Add tests for <=> on Reals.
01:16 lue \o/
01:24 pugssvn r30326 | colomon++ | [t/spec] Add tests for < on Reals.
01:34 ewilhelm joined #perl6
01:43 lue afk
01:44 molaf joined #perl6
01:51 molaf_ joined #perl6
01:56 Psyche^ joined #perl6
02:00 ash_ joined #perl6
02:04 colomon_ joined #perl6
02:07 agentzh joined #perl6
02:10 ewilhelm joined #perl6
02:14 molaf joined #perl6
02:20 pugssvn r30327 | colomon++ | [t/spec] Tests for prefix:<-> and infix:<-> on Real.
02:24 ash_ joined #perl6
02:42 snarkyboojum joined #perl6
02:45 takadonet left #perl6
02:50 ewilhelm joined #perl6
02:52 petdance joined #perl6
02:56 * araujo got qt4 compiled in 1 hour 20 minutes in his new machine
02:56 araujo let's see how goes compiling rakudo
02:56 araujo :P
02:59 justatheory joined #perl6
03:04 Sarten-X joined #perl6
03:44 miyagawa joined #perl6
03:48 snarkyboojum_ joined #perl6
03:57 colomon joined #perl6
03:58 lue phenny: tell masak it seems like I got into some *cough* trouble changing Temporal.pod on the mailing list. According to perlcabal.org it's unchanged, but in my svn it's still my version. Help appreciated :)
03:58 phenny lue: I'll pass that on when masak is around.
04:01 diakopter lue: I didn't see a reply on the mailing list
04:02 snarkyboojum joined #perl6
04:02 lue Could've been sent to just me. Apparently everyone else who worked on S32::Temporal is supposed to have known more about Date/Time than most software engineers ever will :/
04:03 diakopter I don't know how you could conclude that
04:03 diakopter I mean, I haven't seen a reply from you to the question on the list
04:04 lue The guy who sent it to me said that (must've gleamed my email from the pod)
04:04 s1n joined #perl6
04:04 lue (and I didn't reply, because I didn't download the old message)
04:04 diakopter said what
04:04 lue (just a second; retrieving)
04:05 diakopter I've read the original email
04:05 diakopter to the list
04:05 diakopter I'm saying, I haven't seen a reply from you to the list
04:05 diakopter replying to the email in which you are asked questions
04:05 colomon joined #perl6
04:06 diakopter ok, by "that" you meant your previous msg here?
04:06 diakopter "Apparently everyone else who worked on S32::Temporal is supposed to have known more about Date/Time than most software engineers ever will
04:06 diakopter "
04:06 lue yes. It was sent to me apart from the mailing list.
04:06 lue "I am writing off-line as it is possible you are unaware of the ramifications of your actions. The changes you made to the spec were copied to the perl6-language list. The changes will not be easily received by the recipients of the list, who may also choose to write to you directly."
04:06 diakopter you're saying that the guy who emailed the list said "Apparently everyone else who worked on S32::Temporal is supposed to have known more about Date/Time than most software engineers ever will"
04:07 lue It's one Richard Hainsworth.
04:07 diakopter ok, so he's trying to give you advice and do you a favor
04:08 diakopter by informing you that your changes might get replies, but that you won't receive them in case you aren't subscribed to the mailing list, but that you might get messages from folks directly also
04:08 lue I know, I'm not hating him or anything, just noting to masak that this happened, and wondering what is going on with S32::Temporal (website is the "old" version, my local copy is my version)
04:08 diakopter probably it failed to parse as pod
04:09 lue I am signed up to the mailing-list, but it was after I commited S32::Temporal. I had no way of knowing.
04:09 diakopter by the update script that updates those synopses as html
04:09 lue I thank him for his advice. I don't hate him :)
04:09 lue .oO(I suppose then it's good there was a parsefail (if there was one; I could "compile" it correctly), considering it's going to change again)
04:10 diakopter I'm just surmising that
04:10 diakopter perhaps the update script failed another way or is disabled
04:11 diakopter http://perlcabal.org/svn/pugs/revision/?rev=30293
04:12 lue The one thing I didn't like was his assumption that software engineers wouldn't know anything about date/time: "You will also discover that they probably know more about dates and times than most software engineers ever will."
04:12 lue (good thing I'm not a software engineer :P)
04:12 diakopter heh
04:12 diakopter what are you?
04:13 lue a very clever person.
04:13 diakopter so by process of elimination, you're "not a software engineer"
04:13 diakopter that's kindof meaningless
04:13 lue I am, but it's not a job yet, only a hobby :)
04:14 lue Actually, I'm not *most* software engineers :)
04:15 lue .oO(Ah well. Point is, I appreciate his advice. I don't hate him.)
04:15 Trashlord joined #perl6
04:16 diakopter I'm certain Hainsworth didn't expect his email to be discussed on the public record, but I'll do it anyway: from the tone & content of your commit to Temporal.pod, he surmised that you weren't aware of the individual & collective expertise of the folks on p6l (since you didn't consult them first before committing) :)
04:17 diakopter and yes, he was implying that collectively and perhaps individually the mailing list probably knows more about dates and times than you
04:17 diakopter but I don't think he meant it in a mean spirit
04:17 diakopter I dunno
04:18 diakopter maybe he _was_ offended by not being consulted prior to the commit; who knows :)
04:18 diakopter so, I don't know.
04:20 lue I never saw a reason to sign up to p6l; therefore, I could not know pugs commits would show up. And, to top it off, everyone on IRC was OK with it :)
04:20 * diakopter goes to look at the backlog
04:20 lue btw, according to the US gov't, the second you send an email, you give up your right to be protected from search & seizure.
04:21 snarkyboojum definitely worth subscribing to the perl6 mailing lists :)
04:21 snarkyboojum perl6-all will do :)
04:22 lue I am now :)
04:22 lue (only to p6l though)
04:22 snarkyboojum lue: it's good bedtime reading anyway :)
04:24 diakopter lue: where's the consensus about http://irclog.perlgeek.de/​perl6/2010-04-03#i_2189283
04:25 snarkyboojum lue; you want to see all masak++ (bot)'s emails too, surely ;) on perl6-compiler
04:25 diakopter I mean, is it in the irclog?
04:27 lue yes, but the OK was given (relatively) yesterday, and then I posted (relatively) today.
04:28 diakopter the commit was more than 48 hours ago?
04:28 lue (OK, wow; I could swear there was something, looking at IRClog)
04:30 snarkyboojum so which Temporal spec does this commit map to? http://github.com/rakudo/rakudo/commit/3​611f1baeaafd8c27d1d282e6c1435176792d3c5
04:32 lue The one not-yet-published :)
04:32 miyagawa left #perl6
04:32 lue certainly not my version :D
04:32 snarkyboojum implementation followed by spec :)
04:33 lue diakopter: I could swear I got a clear OK. However, on the last Hitchhiker's Holy Day (UTC), I was discussing my changes, and never got a "STOP! STOP STOP STOP STOP STOP! Don't DO THAT!!"
04:33 lue http://irclog.perlgeek.de/out.p​l?channel=perl6;date=2010-04-02
04:34 snarkyboojum http://www.tondering.dk/claus/cal/calendar29.txt looks very juicy :)
04:36 lue .oO(Can't wait for the next Holy Day -- Towel Day!)
04:37 lue Anyone happen to know what the spec is licensed under?
04:37 diakopter lue: I skimmed that entire day's log.  I saw absolutely no discussion/mention of your Temporal ideas
04:38 diakopter so, there was absolutely no occasion for anyone to respond
04:39 diakopter the only mention is "is quietly rewriting all of Temporal to his own design, toning it down so you have a chance of implementing it :)"
04:39 lue Oh well. My memory must be off; I did at least ask questions relating to the "old" spec, and somehow, in my mind, that implies I discussed my changing of Temporal.
04:39 lue I screwed up :)
04:39 diakopter ah, oh
04:42 snarkyboojum in other news, the status graph is showing some nice jumps recently http://www.mutteringmadman.com/perl6/status.png
04:42 snarkyboojum :
04:42 snarkyboojum :)
04:43 lue chesire smiley!
04:47 * diakopter just lost 1 hour to a VS2010RC bug
04:48 * snarkyboojum just lost 1 hours to daylight saving :|
04:48 snarkyboojum 1 hour even
04:48 lue I *HATE* Daylight Savings :(
04:49 mantovani joined #perl6
04:54 lue rakudo: macro doit ($word) {return quasi <say <<<$word>>>,<<<$word>>>;>}; doit('one');
04:54 p6eval rakudo 3b1d34: OUTPUT«Confused at line 11, near "macro doit"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
04:54 lue std: macro doit ($word) {return quasi <say <<<$word>>>,<<<$word>>>;>}; doit('one');
04:54 p6eval std 30327: OUTPUT«[31m===[0mSORRY![31m===[0m␤Bogus statement at /tmp/vDBi9SENHM line 1:␤------> [32mturn quasi <say <<<$word>>>,<<<$word>>>;[33m⏏[31m>}; doit('one');[0m␤FAILED 00:03 116m␤»
04:54 lue std: macro doit ($word) {say "$word $word";}; doit('one');
04:54 p6eval std 30327: OUTPUT«ok 00:01 108m␤»
04:54 lue rakudo: macro doit ($word) {say "$word $word";}; doit('one');
04:55 p6eval rakudo 3b1d34: OUTPUT«Confused at line 11, near "macro doit"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
04:55 lue .oO(darn)
04:56 lue rakudo: say [X] [<a b c>] xx 3
04:56 p6eval rakudo 3b1d34:  ( no output )
04:56 lue rakudo: say [X] [<a b c>] xx 2
04:56 p6eval rakudo 3b1d34: OUTPUT«a a a b a c b a b b b c c a c b c c␤»
04:56 lue why does xx 3 provide no output?
04:58 lue rakudo: say [X] [<a b c d>] xx 3
04:59 p6eval rakudo 3b1d34:  ( no output )
05:03 lue ah well. In every case, good night.
05:06 snarkyboojum joined #perl6
05:23 M_o_C joined #perl6
05:25 kfo_ joined #perl6
05:26 gfx joined #perl6
05:26 justatheory joined #perl6
05:31 am0c joined #perl6
05:56 kaare joined #perl6
06:08 uniejo joined #perl6
06:32 Su-Shee joined #perl6
06:32 Su-Shee good morning
06:32 diakopter hi
06:35 moritz_ good morning
06:47 agentzh joined #perl6
07:06 Psyche^ joined #perl6
07:07 sorear hmm, looks like Parrot has no concept of metaclass compatibility
07:11 mantovani joined #perl6
07:11 Lorn joined #perl6
07:13 Trashlord joined #perl6
07:16 riffraff joined #perl6
07:25 hanekomu joined #perl6
07:26 Trashlord joined #perl6
07:32 rgrau joined #perl6
07:32 mathw Morning
07:35 prism joined #perl6
07:45 sorear How does metaclass compatibility work in Perl 6?
07:51 prism joined #perl6
07:59 moritz_ what do you mean by "metaclass compatibility"?
07:59 moritz_ most things just desugar to method calls on the meta class
07:59 masak joined #perl6
08:00 masak morning, #perl6.
08:00 phenny masak: 03:58Z <lue> tell masak it seems like I got into some *cough* trouble changing Temporal.pod on the mailing list. According to perlcabal.org it's unchanged, but in my svn it's still my version. Help appreciated :)
08:01 moritz_ hm, maybe the HTML versions are not regenerated
08:01 masak could be.
08:02 masak lue: don't worry about the version in the SVN repo. it'd be beneficial if you did reply to Mark J. Reed's email, though.
08:10 JimmyZ joined #perl6
08:10 JimmyZ Good morning, Mai Gao
08:11 masak JimmyZ: \o
08:11 masak rakudo: multi prefix:<∑>(*@items) { [+] @items }; say ∑ 1..10
08:11 p6eval rakudo 3b1d34: OUTPUT«12345678910␤»
08:12 masak how would I make this prefix op looser than infix:<..> ?
08:12 moritz_ in rakudo, or in Perl 6?
08:13 sorear moritz_: metaclass compatibility is the property that A.isa(B) implies A.HOW.isa(B.HOW)
08:13 masak moritz_: in Perl 6.
08:13 moritz_ sorear: no idea then
08:13 masak sorear: why would one want that?
08:14 moritz_ masak: is looser(&infix:<,>)
08:14 masak sorear: sounds like 'covariance under the metaclass transformation' :P
08:14 masak moritz_: ah, nice.
08:14 moritz_ or at whatever level you want to have it
08:16 sorear masak: it's necessary in a lot of cases involving traits... ask #moose-dev for the details
08:16 masak sorear: I'll take your word for it for now.
08:16 sorear today I'm trying to design the ability for P6 classes to extend P5 bases
08:19 moritz_ sorear: is it worth to add blizkost back to rakudo on the evalbot?
08:19 sorear moritz_: attempting to load additional languages into rakudo-ng causes a crash
08:20 JimmyZ rakudo: ( 1, (2+2).say, 3, 4, 5).pick(1).say
08:20 p6eval rakudo 3b1d34: OUTPUT«4␤3␤»
08:20 JimmyZ rakudo: ( 1, {(2+2).say}, 3, 4, 5).pick(1).say
08:20 p6eval rakudo 3b1d34: OUTPUT«4␤»
08:20 sorear I submitted a parrotbug but nobody's looked into it yet
08:20 sorear #1542
08:21 masak JimmyZ: parens not necessary around the (2+2).say
08:21 masak JimmyZ: that expression will give you 1, most often.
08:21 JimmyZ masak: I am making it 'lazy'
08:22 * masak does not understand how that's 'lazy'
08:23 JimmyZ don't eval 2+2 until I pick it.
08:24 masak ah.
08:24 masak well, if you put it in a list like that, you still have to .() it manually.
08:24 moritz_ JimmyZ: that would be done by lazy { (2+2).say } if it were implemented
08:24 masak rakudo: my %h = a => '42'; given %h { when :(Int :$a) { say "lol int" }when :(Str :$a) { say "lol string" } }
08:24 p6eval rakudo 3b1d34: OUTPUT«lol string␤»
08:24 masak std: my %h = a => '42'; given %h { when :(Int :$a) { say "lol int" }when :(Str :$a) { say "lol string" } }
08:24 p6eval std 30327: OUTPUT«[31m===[0mSORRY![31m===[0m␤Strange text after block (missing comma, semicolon, comment marker?) at /tmp/rFyy8Ug5Hg line 1:␤------> [32mn %h { when :(Int :$a) { say "lol int" }[33m⏏[31mwhen :(Str :$a) { say "lol string" } }[0m␤    expecting horizontal whitespace␤FAILED
08:24 p6eval ..00:01…
08:24 * masak submits rakudobug
08:25 masak rakudo: $_=2;when 1{}when 2{say"oh noes"}
08:25 p6eval rakudo 3b1d34: OUTPUT«Unable to parse postcircumfix:sym<{ }>, couldn't find final '}' at line 11␤current instr.: 'perl6;Regex;Cursor;FAILGOAL' pc 1664 (ext/nqp-rx/src/stage0/Regex-s0.pir:907)␤»
08:25 masak rakudo: $_=2;when 1 {}when 2 {say"oh noes"}
08:25 p6eval rakudo 3b1d34: OUTPUT«Unable to parse blockoid, couldn't find final '}' at line 11␤current instr.: 'perl6;Regex;Cursor;FAILGOAL' pc 1664 (ext/nqp-rx/src/stage0/Regex-s0.pir:907)␤»
08:26 JimmyZ rakudo: rakudo: ( { say 'run 1';  2 ** 100 }, { say 'run 2'; 2 ** 1000 } ).pick(1)().say
08:26 * masak .oO( when does it work, and when doesn't it? )
08:26 p6eval rakudo 3b1d34: OUTPUT«Confused at line 11, near "rakudo: ( "␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
08:27 JimmyZ rakudo: ( { say 'run 1'; 2 **  100; }, { say 'run 2'; 2 ** 1000; } ).pick(1)().say
08:27 p6eval rakudo 3b1d34: OUTPUT«run 1␤1.26765060022823e+30␤»
08:30 agentzh joined #perl6
08:30 JimmyZ rakudo: ( ( say 'run 1'; 2 ** 100; ), (say 'run 2'; 2 ** 1000;)).pick(1).say
08:30 p6eval rakudo 3b1d34: OUTPUT«run 1␤run 2␤1.07150860718627e+301␤»
08:31 agentzh joined #perl6
08:31 agentzh joined #perl6
08:31 * JimmyZ couldn't follow moritz_ :(
08:36 moritz_ JimmyZ: where did you want to follow me?
08:37 * JimmyZ couldn't understant 'that would be done by lazy { (2+2).say } if it were implemented'
08:37 JimmyZ understand
08:38 agentzh joined #perl6
08:39 moritz_ lazy { code } returns a value that's lazily evaluated
08:39 moritz_ so the code only runs when the value is actually used
08:39 moritz_ but it's not yet implemented in rakudo
08:39 JimmyZ Yeah.
08:40 moritz_ and from looking at your example, I thought that was what you were looking for
08:40 JimmyZ Yeah.
08:40 JimmyZ rakudo: ( { say 'run 1'; 2 ** 100; }, { say 'run 2'; 2 ** 1000; } ).pick(1)().say # Isn't this one?
08:41 p6eval rakudo 3b1d34: OUTPUT«run 2␤1.07150860718627e+301␤»
08:41 agentzh joined #perl6
08:42 masak would be nice (and probably not too hard) to make this work under the new DateTime class in Perl 6: http://use.perl.org/~snarkyboojum/journal/40297
08:43 mathw o/ masak
08:43 masak (well, we'd need to port DateTime::Format::Strptime)
08:43 masak mathw: \o
08:43 agentzh joined #perl6
08:43 snarkyboojum masak: oh hi there - yeah, that'd be worthwhile I think
08:44 masak snarkyboojum: \o
08:44 snarkyboojum masak: o/ :)
08:45 moritz_ .oO( snarkyboojum uses indirect object notation in Perl 5 )
08:45 snarkyboojum moritz_: whoopsy ;)
08:45 snarkyboojum moritz_: I probably copied that out of the perldoc :O (no excuse I know)
08:46 moritz_ I mean there's nothing wrong with that, except that I don't grok the rules how it is parsed :-)
08:46 agentzh joined #perl6
08:48 agentzh joined #perl6
08:49 agentzh joined #perl6
08:50 snarkyboojum moritz_: I updated it anyway :)
08:51 snarkyboojum afk&
08:58 agentzh joined #perl6
09:02 masak phenny: tell lue, "Some of the hardest work [in lnaguage design] isn't making the changes, it's getting them past the maintainers or an ensuing bikeshedding discussion that goes nowhere."
09:02 phenny masak: I'll pass that on when lue is around.
09:04 jnthn o/
09:05 jnthn masak: I made it back home! \o/
09:05 masak jnthn: glad to hear it :)
09:08 moritz_ lue: http://irclog.perlgeek.de/​perl6/2010-04-06#i_2199834 there's still this thing called "courtesy" - even if you have the right to publish something, it might not be friendly towards the sender
09:10 agentzh joined #perl6
09:11 clintongormley joined #perl6
09:21 colomon joined #perl6
09:27 agentzh joined #perl6
09:29 pugssvn r30328 | colomon++ | [t/spec] Couple more infix:<-> on Real tests.
09:43 xabbu42 joined #perl6
09:44 agentzh joined #perl6
09:52 Uazz_Z joined #perl6
09:56 pmurias joined #perl6
09:57 mberends joined #perl6
09:58 * mberends whispers masak: ping from the classroom ;)
09:58 masak mberends: pong from the office :)
09:59 mberends shh!
10:00 mberends ( before enabling the DateTime in Makefile.in and spectest.data, I wonder whether the whole Temporal naming should be changed to DateTime. )
10:02 rbaumer joined #perl6
10:02 masak mberends: aye, that's probably a good idea for two reasons.
10:02 masak (1) less pompous
10:03 masak (2) indicates a clean break from previous stuff
10:03 * mberends likes *both* reasons
10:03 moritz_ I'm mostly out of touch with Temporal/DateTime...
10:03 mberends (3) frees up Temporal for some really fantastic modules
10:03 masak :)
10:04 masak (4) consistency with the class name and namespace
10:04 moritz_ I just hope that time() is supposed to return an Instant (or whatever it's called these days)
10:04 masak moritz_: yes, I think it will eventually.
10:04 moritz_ and not an epoch or so
10:04 moritz_ and that the object stringifies sensibly
10:04 masak moritz_: right now I have it spec'd as in Perl 5, but I think Instant is wiser.
10:04 proller joined #perl6
10:05 moritz_ I haven't heard a good reason to return a number
10:05 mberends also, let's keep the DateTime out of Makefile.in until it is more rounded out and in sync with S32
10:06 masak agreed.
10:06 moritz_ -1 # I want to experiment with it on p6eval :-)
10:06 mberends boo hoo!
10:07 masak moritz_ has a point.
10:07 masak it's kinda fun to play around with :)
10:07 jnthn +1 to that
10:07 mberends 2 days should be enough to anneal it for online testing
10:07 masak moritz_: I'd like DateTime.now to create an object representing the current date and time. it has to do this from information returned by time() or equiv.
10:08 masak moritz_: whether time() then returns a number or an instant is not too important to me.
10:08 masak aye. give us two days.
10:08 * moritz_ offers 5 hours :-))
10:08 * jnthn offers another 43
10:08 jnthn ;-)
10:08 masak phew :)
10:08 * mberends feels prodded
10:09 moritz_ masak: I tink there can be a more low level funtion with a longer name
10:09 masak moritz_: I'll do it in five hours, but I'd have to bill you :)
10:09 jnthn epoch() :-)
10:09 moritz_ I just want to cater to user expectations
10:09 mberends there are from_epoch() and to_epoch() already
10:09 masak I'm not too happy about calling 'the number of seconds since the computer epoch' just 'epoch' :/
10:10 moritz_ and from reading perlmonks I see that lots of people call time(), even if it's not the best function to use
10:10 masak aye
10:10 moritz_ then call the low level function seconds_since_start_of_epoch() and discourage its use by having a long name :-)
10:10 jnthn now() would be kinda nice
10:10 hejki yup
10:10 hejki or today()
10:10 masak people will always do things on a plumbing level if we don't put appealing SYNOPSIS examples in their way :)
10:10 hejki obv. being != than now() :>
10:10 jnthn Well, now() better captuers the time componenet too.
10:11 masak hejki: we haz today() :)
10:11 hejki oh we do
10:11 hejki i'm such a bunch of noobdom
10:11 * moritz_ kinda links Date::Simple on CPAN
10:11 masak hejki: in the unpublished draft, I mean.
10:11 hejki ahh ok
10:11 hejki :>
10:11 moritz_ hejki: don't worry, we all are :-)
10:11 masak moritz_: I'll have a look at that, then.
10:12 moritz_ it's dead simple date-only arithmetics
10:13 JimmyZ joined #perl6
10:13 moritz_ today() - date('2010-01-01') # number of days since first of January
10:13 JimmyZ Will Perl6 support 'use Foo as bar' syntax which likes python?
10:13 moritz_ today() - 1 # yesterday's date
10:15 masak lunch &
10:16 mberends moritz_: nice, but how would you implement that? custom infix:<-> ?
10:16 moritz_ mberends: sure
10:17 moritz_ it just takes all integers as duration in days
10:17 moritz_ lunch&
10:17 Juerd JimmyZ: How can syntax like Python?
10:17 Juerd or anything.
10:19 JimmyZ There is a modlue with looooooong mame, and I want give it an alias.
10:19 JimmyZ s/modlue/model/
10:19 mberends JimmyZ: that's possible, in S11
10:20 mberends but not in the roadmap for Rakudo *
10:20 * Juerd still doesn't understand how that likes Python :)
10:21 pmurias Python is so cute, surely everything likes it ;)
10:22 Juerd JimmyZ: Did you mean "syntax like Python" instead of "syntax which likes python"?
10:22 JimmyZ Yeah, poor English
10:22 mberends pythons like parrots. for breakfast.
10:23 JimmyZ mberends: I couln't find it.
10:24 Juerd JimmyZ: That's okay. It just took a while for me to understand that you meant something different :)
10:25 JimmyZ Juerd: :(
10:25 cosimo joined #perl6
10:25 Juerd s/':('/:)/
10:25 JimmyZ So how could I give it an alias?
10:25 Juerd Have you read S11?
10:25 Juerd http://p3rl.org/S11
10:26 JimmyZ Yeah
10:26 JimmyZ import_realias ?
10:27 Juerd Can't say I can find it in S11, though
10:27 jnthn use Pivo:name<Beer::Slovak>;
10:27 jnthn Where :name is the real name to look up.
10:27 Juerd Ah, :name
10:28 jnthn Don't think we'll support it in R* though.
10:28 Juerd Is whitespace allowed before the :?
10:28 jnthn afaik no
10:28 jnthn std: use lol_module:ver<42>;
10:28 p6eval std 30328: OUTPUT«Potential difficulties:␤  Can't locate module lol_module:ver<42> at /tmp/7nzSOfSMPF line 1:␤------> [32muse lol_module:ver<42>[33m⏏[31m;[0m␤ok 00:01 107m␤»
10:28 jnthn std: use lol_module :ver<42>;
10:28 p6eval std 30328: OUTPUT«Potential difficulties:␤  Can't locate module lol_module at /tmp/9KopHMmljN line 1:␤------> [32muse lol_module :ver<42>[33m⏏[31m;[0m␤ok 00:01 106m␤»
10:29 jnthn Oh, it seems it copes.
10:29 jnthn rakudo: use Test;
10:29 Juerd Foo:name<Bar> looks so compressed :(
10:29 p6eval rakudo 3b1d34:  ( no output )
10:29 jnthn rakudo: use Test:ver<42>;
10:29 p6eval rakudo 3b1d34: OUTPUT«Unable to find module 'Test' with version '42'.␤current instr.: 'perl6;Perl6;Module;Loader;need' pc 34820 (src/gen/role_pm.pir:807)␤»
10:29 jnthn rakudo: use Test :ver<42>;
10:29 p6eval rakudo 3b1d34: OUTPUT«Confused at line 11, near "use Test :"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
10:29 Juerd :(
10:29 * Juerd did not mean to confuse Rakudo
10:30 jnthn I thought I'd followed STD there, fwiw.
10:30 jnthn So not sure why Rakudo and STD have different views.
10:30 JimmyZ use Whiteness:from<perl5>:name<Acme::Bl​each>:auth<cpan:DCONWAY>:ver<1.12>;
10:30 jnthn Yeah, that's teh reference in s11.
10:30 JimmyZ std: use Whiteness:from<perl5>:name<Acme::Bl​each>:auth<cpan:DCONWAY>:ver<1.12>;
10:30 p6eval std 30328: OUTPUT«Potential difficulties:␤  Can't locate module Whiteness:from<perl5>:name<Acme::Bl​each>:auth<cpan:DCONWAY>:ver<1.12> at /tmp/XJM3DAIEzM line 1:␤------> [32mme::Bleach>:auth<cpan:DCON​WAY>:ver<1.12>[33m⏏[31m;[0m␤ok 00:01 105m␤»
10:31 JimmyZ std: use Thread.pm;
10:31 p6eval std 30328: OUTPUT«[31m===[0mSORRY![31m===[0m␤Confused at /tmp/EZhQ30JfJi line 1:␤------> [32muse Thread[33m⏏[31m.pm;[0m␤    expecting statement_control__S_064use␤Other potential difficulties:␤  Can't locate module Thread at /tmp/EZhQ30JfJi line 1:␤------> [32muse
10:31 p6eval ..Thread[33m⏏[31m.pm;[0m␤F…
10:31 JimmyZ std: use Thread;
10:32 p6eval std 30328: OUTPUT«Potential difficulties:␤  Can't locate module Thread at /tmp/UR3iDakMhd line 1:␤------> [32muse Thread[33m⏏[31m;[0m␤ok 00:01 105m␤»
10:34 vdo joined #perl6
10:35 vdo left #perl6
10:44 envi^home joined #perl6
11:30 mberends phenny, tell masak forgot to mention a very important reason to identify Perl 6 modules on CPAN with a "6" name: to avoid having Perl 5 users stumble into them unintentionally.
11:30 phenny mberends: I'll pass that on when masak is around.
11:40 masak mberends: I don't really see a qualitative difference between that and Perl 5.6 users stumbling on a 5.10 module unintentionally.
11:40 phenny masak: 11:30Z <mberends> tell masak forgot to mention a very important reason to identify Perl 6 modules on CPAN with a "6" name: to avoid having Perl 5 users stumble into them unintentionally.
11:41 moritz_ masak, mberends: as a courtesy to Perl 5 users all Perl 6 modules on CPAN should prevent indexing of all their files with a suitable META.yaml
11:41 masak good point.
11:45 M_o_C joined #perl6
11:46 Uazz_Z left #perl6
11:58 snarkyboojum joined #perl6
12:03 ruoso joined #perl6
12:03 bbkr joined #perl6
12:06 takadonet1 morning all
12:09 moritz_ rakudo: my $;
12:09 p6eval rakudo 3b1d34: OUTPUT«Malformed my at line 11, near "$;"␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
12:09 masak takadonet: \o
12:09 moritz_ std: my $;
12:09 p6eval std 30328: OUTPUT«[31m===[0mSORRY![31m===[0m␤Unsupported use of $; variable; in Perl 6 please use real multidimensional hashes at /tmp/QTWBjtp5dw line 1:␤------> [32mmy $;[33m⏏[31m<EOL>[0m␤FAILED 00:01 108m␤»
12:09 jnthn o/ takadonet
12:09 moritz_ std: my ($);
12:09 p6eval std 30328: OUTPUT«ok 00:01 107m␤»
12:09 moritz_ rakudo: my ($);
12:09 p6eval rakudo 3b1d34:  ( no output )
12:10 moritz_ rakudo: say my ($);
12:10 jnthn If anyone wants to put the special variable obsolecense warnings into Rakudo from STD, feel free, btw. It's LHF.
12:10 p6eval rakudo 3b1d34: OUTPUT«Any()␤»
12:10 moritz_ rakudo: say \my ($);
12:10 p6eval rakudo 3b1d34: OUTPUT«Capture()<0x5557b80>␤»
12:10 bbkr rakudo: $_="a"; $_ R~= "b"; # hmm, counterintuitive... i expected $_ to be "ba"
12:10 p6eval rakudo 3b1d34: OUTPUT«Cannot assign to readonly value␤current instr.: '&die' pc 17293 (src/builtins/Junction.pir:404)␤»
12:11 jnthn bbkr: Precedence bites. :-)
12:11 jnthn bbkr: [R~]= probably is what you want.
12:11 bbkr rakudo: $_="a"; $_ [R~]= "b"; say $_;
12:11 p6eval rakudo 3b1d34: OUTPUT«Confused at line 11, near "$_ [R~]= \""␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
12:12 jnthn std: $_="a"; $_ [R~]= "b"; say $_;
12:12 p6eval std 30328: OUTPUT«ok 00:01 107m␤»
12:12 jnthn bbkr: We still need to do that in Rakudo yet.
12:12 bbkr ok, thanks for clarification jnthn
12:12 jnthn In theory it shouldn't be too hard.
12:12 jnthn :-)
12:14 * jnthn -> nom shop
12:19 takadonet 3m 26 second build time :)
12:19 takadonet wow
12:19 mathw nice
12:29 gavgpv joined #perl6
12:29 gavgpv left #perl6
12:37 twigil joined #perl6
12:41 snarkyboojum yeah, I can almost remember what I was doing between builds now :)
12:41 moritz_ oh, it's quite simple - after a finished spectest run I just look into 'git diff'
12:42 moritz_ and if there are changes, I usually remember why I did those, and commit if the tests passed :-)
12:42 snarkyboojum :)
12:45 colomon funny, I usually commit before I run the spectests.
12:46 colomon that way, if the spectest works, all I need remember to do is "git push"
12:47 moritz_ I sometimes do that too
12:48 moritz_ but I don't like to git reset or git rebase all that often if I find that my changes didn't work after all
12:49 colomon sure, I only do that after the exploratory phase of development.
12:49 snarkyboojum wow, I just did a git pull, and a successful build and my ./perl6 is broken :(
12:49 snarkyboojum get 'Could not find sub !YOU_ARE_HERE' no matter what I run or how I run it
12:49 masak I find I do 'git reset --hard' quite often nowadays. I also have a tendency to do 10-minute experimental branches.
12:50 takadonet wow 1400 test failed....
12:50 moritz_ snarkyboojum: did you reconfigure rakudo?
12:50 jnthn snarkyboojum: You did re-con...yes, that.
12:50 snarkyboojum mortiz_: just doing that now
12:50 jnthn My commit message even said to. ;-)
12:51 snarkyboojum moritz_ even
12:51 moritz_ :-)
12:51 snarkyboojum oh! commit messages!
12:51 moritz_ snarkyboojum: just mo<tab> for you :-)
12:51 snarkyboojum moritz_: my goodness you're right :) heh
12:51 jnthn I should start putting jokes in my commit messages so people read them. ;-)
12:51 moritz_ oh wait, that's ambiguous, at least in my irssi
12:51 snarkyboojum jnthn: I just read the code in your commits, not the messages :P
12:52 moritz_ jnthn: but then you couldn't hide your passwords there... oh wait, that was README, a totally different matter :-)
12:52 snarkyboojum (and obviously don't understand much)
12:52 * masak .oO( is the IRC equivalent of 'first-name basis' writing out someone's nick in full instead of tabbing it? )
12:52 jnthn snarkyboojum: Well, the message tells you what the code is menat to do, the code tells you what it actually does. ;-)
12:52 snarkyboojum I can't believe I've been writing out names in full all this time :|
12:53 snarkyboojum jnthn: yeah, I'm not very clever eh? :)
12:53 moritz_ snarkyboojum: a surprising number of people don't know about this convenient feature of most IRC clients
12:53 moritz_ snarkyboojum: in particular I've told Tim Bunce (father of perl 5's DBI module) about it, and it was news for him :-)
12:54 jnthn masak: yes, but if you don't type enough letters you get the wr...oh my.
12:54 jnthn :-)
12:54 snarkyboojum moritz_: weird how the context switch happens between different apps - if I was running an IRC client in a shell I'd be tab-happy by default
12:56 mathw ah yes, the #perl6 tab-completion nightmare
12:56 mathw all these people who begin with m
12:57 jnthn The nickspace is crowded in m.
12:57 colomon and the damned co-s
12:57 frettled better just to say:
12:57 moritz_ mathw: all these people beginninig with 'ma' even
12:57 frettled Hey, dude.  Don't make it bad.  Take some sad code, and make it better.
12:58 twigil hello everyone! I learnt about GSoC deadline just today. Is it too late to try to apply?
12:58 masak gee, I hope not :)
12:58 * masak should really sit down and write that application
12:59 moritz_ twigil: it's only too late when the deadline closes :-)
12:59 mathw skribu!
13:01 pmurias masak: what will you apply for?
13:01 pmurias what i meant what is your application about?
13:01 moritz_ twigil: what are you interested in?
13:02 hanekomu joined #perl6
13:03 twigil moritz_: I looked through the list of perl 6 projects for 2010. contributing to test suite looks pretty interesting for me
13:03 masak pmurias: I'd like to work either on Rakudo or on the spectests. haven't decided yet.
13:03 moritz_ twigil: cool
13:03 masak if twigil takes the test suite, maybe I'll focus on Rakudo :)
13:03 moritz_ twigil: also if you want to write Perl 6 code, Rakudo is mostly written in perl 6 these days (or subsets thereof)
13:03 moritz_ heh :-)
13:03 jnthn \o/
13:04 masak it be fun, and a bit fitting, to focus on closing RT tickets. but I don't know if it's concrete enough.
13:04 masak and it's a bit hard to make concrete, because RT keeps changing around.
13:04 moritz_ masak: nah, you'd have to come up with a concrete topic and schedule
13:04 jnthn masak: Didn't you want to focus on static anal?
13:05 moritz_ anyway, whoever works on the test suite is likely to be stuck with me as a mentor :-)
13:05 jnthn Or other such thingies? Or gone off that idea? :-)
13:05 moritz_ masak: you could implement named enums as the first step of your gsoc project :-)
13:05 masak jnthn: I'll focus on static analysis regardless. maybe if you tell me about the possibilities of it in Rakudo, I can write something up.
13:06 masak moritz_: no, actually I couldn't :(
13:06 snarkyboojum is the plan to make Temporal understand time zones like 'Australia/Sydney'?
13:06 masak moritz_: I'm currently blocking there, with no realistic way to unblock without help.
13:06 masak snarkyboojum: not in itself, no.
13:06 moritz_ masak: then get help :-)
13:06 masak jnthn: help! :)
13:07 snarkyboojum masak: bugger - I liked that about Perl 5 DateTime
13:07 jnthn masak: Yeah, I know, I still need to do Stash
13:07 jnthn :-)
13:07 masak snarkyboojum: things having to do with 'words' of some sort are exported to modules.
13:07 masak snarkyboojum: same goes for 'January' and 'Tuesday'.
13:07 masak snarkyboojum: it's a bit narrow, but it's a simple line to draw.
13:07 masak and we can always extend it if it turns out to be safe to do so.
13:07 jnthn The good news is that there's already a module for November.
13:08 masak :P
13:08 masak I just don't want to start with too much. better to build up from below than to tear down.
13:08 snarkyboojum so you'd need to lookup a timezone using 'Australia/Sydney' using a different module, and then plug that in at contruction time?
13:08 masak or load a module and use the string.
13:10 moritz_ masak: so you also postponed your plans to parse dates in ISO-$something formats?
13:10 moritz_ hey
13:10 moritz_ speaking of GSOC
13:10 moritz_ and masak
13:10 moritz_ and temporal/datetime
13:10 snarkyboojum :)
13:11 moritz_ masak: don't you want to make *that* part of your gsoc project? :-)
13:11 masak no :)
13:11 masak I've retreated a bit there, I think.
13:11 masak it parses ISO8601, but only the full date/time form.
13:12 masak maybe I should sign up for finally implementing Buf, including reading binary files etc.
13:12 snarkyboojum masak: is there a current Temporal POD somewhere I can have a read of?
13:13 mathw masak: that would be interesting
13:13 jnthn masak: That'd also be an excellent project.
13:13 mathw or if you feel ambitious, how about a concurrency framework with STM? :)
13:14 moritz_ mathw: that would be *very* ambitious, given that parrot doesn't support any concurrency in connection with HLLs
13:14 snarkyboojum something along the lines of http://gist.github.com/355664 I guess?
13:14 ignacio_ joined #perl6
13:15 mathw moritz_: I did say 'if you feel ambitious' :)
13:15 moritz_ btw <http://www.perlmonks.org/?node_id=832796> (Perl 5, but about language extensions, so maybe of some small interest here)
13:15 masak snarkyboojum: this is what I have so far: http://gist.github.com/356555
13:16 masak snarkyboojum: please don't consider it official or done in any way, shape, or form.
13:16 masak two more days, and it will be.
13:16 snarkyboojum masak: cheers!
13:16 snarkyboojum masak++ :)
13:16 masak lue++ mberends++
13:21 pmurias ruoso: hi
13:23 snarkyboojum :timezone defaults to +0000 now (though I'm sure that'll get picked up)
13:25 masak indeed.
13:25 masak that's a change I already had planned to put in.
13:27 snarkyboojum cool stuff :)
13:27 snarkyboojum it's quite amazing to watch the pace of Perl 6 development
13:28 snarkyboojum so much going on
13:28 masak we're eager to reach the future :)
13:28 snarkyboojum and I guess Temporal will tell you when you get there etc :P awesome
13:29 * snarkyboojum is sure all the Temporal jokes have been done to death
13:29 snarkyboojum :)
13:30 jnthn Yeah, we don't have time for them any more.
13:30 snarkyboojum :D
13:31 masak luckily, we're probably changing the name of the module. :)
13:31 masak we're adding three more dimensions, and calling it SpatioTemporal.
13:31 * masak hides
13:32 jnthn Now, wait a minute...
13:33 masak did you mean: 'wait a meterminute'?
13:34 jnthn .oO( I'd be more interested if I metadate... )
13:34 masak what's that? two dates going on a date together?
13:37 pmichaud good morning, #perl6
13:37 colomon o/
13:37 moritz_ \o/
13:37 jnthn hi, pmichaud! o/
13:38 pmichaud oops, afk for a bit
13:39 masak hi, pmichaud
13:40 rgrau` joined #perl6
13:44 slavik w00t
13:48 pmichaud back again, briefly
13:48 mathw o/
13:50 colomon_ joined #perl6
13:50 ruoso hi pmurias
13:50 dalek joined #perl6
13:51 colomon_ rakudo: role Animal { multi method speak() { "quiet"; }; }; class Dog does Animal; say Dog.new.speak
13:51 p6eval rakudo 3b1d34: OUTPUT«quiet␤»
13:51 colomon_ :\
13:52 colomon rakudo: role Animal { multi method speak() { "quiet"; }; }; class Dog does Animal {multi method speak() { "woof" }}; say Dog.new.speak
13:52 p6eval rakudo 3b1d34: OUTPUT«woof␤»
13:52 colomon :\
13:52 jnthn Yeah, roles + multi-methods = not really done yet.
13:52 molaf joined #perl6
13:52 jnthn (that should die due to ambiguity)
13:54 colomon really?  how can Dog's speak override Animal's, then?
13:54 colomon (and locally I'm getting get_attr_str() not implemented in class 'Perl6MultiSub' when I try this sort of thing.)
13:55 jnthn colomon: Only works that way because of multi
13:55 jnthn Maybe there's some canonicalization rule that we'd go on in the future.
13:56 masak spectested; got a couple of failures in t/spec/S06-multi/syntax.rakudo
13:56 jnthn But certainly the alpha semantics were that you can all multi candidates from the class and the role.
13:56 moritz_ masak: yes, they are there since Saturday nigth or so
13:57 colomon jnthn: in general I'm getting really weird results here.  may just be too tired to code.
13:58 jnthn colomon: If you're doing anythign to do with roles + multis all bets are off.
13:58 masak moritz_: oh, ok.
13:58 jnthn I've done just about nothing there.
13:59 jnthn Anything that works atm is by accident.
13:59 PerlJam (happy accidents)++
13:59 PerlJam good morning btw
14:00 masak PerlJam: \o
14:00 m6locks good afternoon
14:01 * PerlJam was summoned to be part of a jury yesterday.
14:01 PerlJam Sat all day and was never called.
14:01 PerlJam But, they said I'm not off the hook until Wed @ 5pm
14:02 moritz_ bah, doesn't sound like fun
14:02 moritz_ and let me guess, you're not allowed to reject the call
14:02 PerlJam sitting all day in a room full of several hundred people was not at all fun.
14:03 PerlJam moritz_: I don't qualify for any of the exemptions.
14:03 moritz_ you need a music player, a laptop, a good book and thing to eat and drink
14:04 mathw bah
14:04 pmichaud afk again for a while :-\
14:04 masak moritz_: kuro5hin makes it sound like there are ways: http://www.kuro5hin.org/story/2008/9/8/0245/12132
14:04 mathw I imagine jury service is much more interesting if you actually get to be on the jury
14:04 PerlJam I guess this is karma or something though.  I haven't had a jury summons since I was in college (about 17 years ago)
14:05 moritz_ masak: probably depends on the legislation
14:05 mathw I've never had one
14:05 mathw ...yet
14:05 mathw I'm about as eligible as you get though
14:05 PerlJam while I was in there with the other cattle, I found that some people had been summoned multiple times within a years
14:05 PerlJam s/years/year/
14:05 mathw irritating
14:06 mathw fortunately if they did decide to summon me and they did the sensible thing and got me in my local courts, it's just about as easy as getting to the office
14:06 mathw but it'd likely end up being massive hassle
14:06 mathw and lots of boring
14:06 PerlJam definitely lots of boring
14:06 mathw PerlJam: did lots of other people end up doing nothing all day too?
14:07 PerlJam like I said, there were several hundred people in the room with me.
14:07 PerlJam about 1/3 thought ahead enough to bring something to do
14:07 mathw several hundred?
14:07 mathw man
14:08 mathw how many juries did they need?
14:09 PerlJam They formed 3 panels of 100 people, 1 of 80, 2 or 3 of 60, and a small one of 20  (that's all I noticed, I may have missed one or two panels)
14:10 PerlJam The very first group of 100 people got to leave after about 3 hours of sitting around.
14:10 PerlJam The rest of us weren't as lucky
14:10 mathw :(
14:10 nihiliad joined #perl6
14:11 * moritz_ hopes that IO is scheduled more efficiently than prospective jury members
14:16 PerlJam Hrm.  Not that many student proposals for GSoC this time around.
14:16 ignacio_ joined #perl6
14:17 pmichaud I went for several years w/o ever being summoned for jury duty.  Then I was called as an expert witness in several cases protesting the computer algorithm used to select juries.  After my testimony, the county immediately changed their jury selection software, and I was called to jury duty the following year.
14:18 pmichaud (the computer software used for jury selection was laughably bogus)
14:18 * masak .oO( .pick(*) )
14:18 takadonet ...
14:19 mathw rofl
14:19 jnthn wow!
14:20 pmichaud the team that created the jury selection criteria thought that hash pools were sufficient to randomize the candidates
14:21 pmichaud they also thought that wall clock time (1-second resolution) was a good source of entropy for generating random numbers
14:21 mathw it's... *slightly* better than no entropy at all
14:21 pmichaud note:  not for seeding a random number generator.... for actually generating the random numbers
14:21 mathw oh
14:22 mathw oh dear
14:22 pmichaud given that the program was processing approximately 100 "random" entries per second...
14:22 jnthn wtf.
14:22 jnthn That's...such a fail.
14:22 masak I've done that. but it was millisecond resolution, and it was only for randomizing banners on pages :)
14:22 pmichaud that meant that 100 "random candidates" were all being selected based on the same wall clock time
14:22 jaldhar joined #perl6
14:23 mathw so you'd get Patrick Michaud, and Paul Michaud, and Peter Michaud all at once!
14:23 jnthn .oO( If I can make this guy late to work, I don't have to do jury duty! )
14:24 mathw .oO( If I cut off the power for just the right amount of time, I can make my neighbour do jury duty and rob his house while he's out)
14:24 pmichaud mathw: something like that, yes.  The trial attorneys knew something was up when they started seeing jury pools where everyone knew each other already (because they had served on a jury together the previous year :-)
14:24 mathw that would tend to give it away
14:24 mathw I imagine it was more likely to clump SSNs or something
14:25 pmichaud worse than that, even
14:25 pmichaud the records were loaded into the database based on voter registrations
14:25 PerlJam pmichaud: I think their randomness is still highly suspect. There were a couple of ladies chatting behind me who had both been called for jury duty last month.
14:25 pmichaud and the source file of the voter registrations was alphabetical by last name
14:26 pmichaud so people were being clustered together by last name.  It wasn't quite as obvious as there being a bunch of people with the same last name on the jury, because the hash pools did distribute things somewhat
14:27 mathw That is awful
14:27 PerlJam unless perhaps being called but not selected puts you on the short list for getting called again  ...  (randomly :)
14:27 moritz_ .oO( ... )
14:27 pmichaud PerlJam: same courts?  It wouldn't surprise me if they were called for federal jury duty and then state/county
14:28 moritz_ even microsoft can't properly randomize lists :-)
14:28 pmichaud oh, and here's the other funny part
14:28 moritz_ (re: browser choices in random order in EU countries)
14:28 pmichaud let's say we have a set of jurors:  Alice, Bob, Carol, Donald
14:28 PerlJam pmichaud: yep, same courst
14:28 pmichaud and the algorithm picked "Bob" for jury duty
14:29 proller joined #perl6
14:29 pmichaud if Bob was flagged as ineligible for duty, then the program simply moved to the next available candidate after Bob
14:29 pmichaud in this case, if Bob was ineligible for duty, then Carol became twice as likely to be chosen as Donald
14:29 moritz_ so if you had a huge clan of ineligible people with the same last name
14:29 sbp so if you have a load in a row who were ineligible, the poor person after would have an Nx greater chance of being picked?
14:29 sbp ouch
14:30 pmichaud yes
14:30 CokeBot9000 woof.
14:30 CokeBot9000 I bet they use the same software out here. =-)
14:30 moritz_ they knew *nothing* about statistics, did they?
14:31 moritz_ sounds like the software was written by the cousin of the president of some court
14:32 moritz_ how knew a bit about visual basic
14:32 pmichaud no, the authors were just clueless
14:32 pmichaud it was _very_ interesting to watch the county information officer defend the system
14:32 [particle] is that texas justice?
14:33 pmichaud the county of course wanted to be very careful, because they didn't want to be re-trying a year's worth of cases based on jury bias
14:33 * moritz_ thinks it's a good sign that pmichaud++ has the leisure to rant about non-working PRNGs :-)
14:34 pmichaud alas, it's a false leisure
14:34 PerlJam moritz_: either that or it's an escape mechanism
14:34 moritz_ :/
14:34 pmichaud what it really means is that I had about 15 minutes while the physical therapist was here in which I could sit and chat
14:34 pmichaud she's gone now, and I need to do a few things with Paula again :-)
14:35 moritz_ then wish her the best of luck (even though I never met her)
14:35 pmichaud but yes, things are slowly-but-surely getting better around here, so that I should have more time to sit in front of the computer
14:35 pmichaud (alas, more "slowly" than "surely", but at least headed in the correct direction)
14:36 CokeBot9000 pmichaud: best wishes to Paula.
14:36 pmichaud CokeBot9000: Thanks. :-)
14:36 slavik CokeBot9000: you deal coke?
14:37 pmichaud afk again -- bbl
14:38 masak pmichaud: yes, thinking of you and Paula a lot here. best wishes.
14:42 uniejo joined #perl6
14:44 bluescreen joined #perl6
14:47 dalek rakudo: c06062d | (Solomon Foster)++ | src/core/ (3 files):
14:47 dalek rakudo: Add Numeric.log10 which simply calls self.log(10) -- remove resulting unneeded log10s from Complex and Num.
14:47 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/c​06062d46904933c547fca36cdb81e726cfd5503
14:48 masak colomon++
14:49 justatheory joined #perl6
14:55 sorear joined #perl6
14:57 meppl joined #perl6
15:01 CokeBot9000 slavik: no.
15:01 slavik :(
15:01 mberends \o travel Arnhem -> Amsterdam.pm! :-)
15:04 molaf joined #perl6
15:04 m6locks rakudo: class A { has $.avg; multi submethod BUILD(Num @a) { my $a = 0; for 1..@a.elems -> $i { $a += @a[$i] };  $.avg = $a / @a.elems; }} my $b = A.new((1,2,3,4)); say $b.avg
15:04 p6eval rakudo 3b1d34: OUTPUT«Confused at line 11, near "class A { "␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
15:05 jnthn need ; before the my
15:06 moritz_ there's an easier way to calculte the average :-)
15:06 sorear joined #perl6
15:06 moritz_ rakudo: my @a = 1, 2, 3; says ([+] @a) / @a
15:06 p6eval rakudo 3b1d34: OUTPUT«Could not find sub &says␤current instr.: '_block14' pc 29 (EVAL_1:0)␤»
15:06 moritz_ rakudo: my @a = 1, 2, 3; say ([+] @a) / @a
15:06 p6eval rakudo 3b1d34: OUTPUT«2␤»
15:06 Jedai joined #perl6
15:06 m6locks my point was to demonstrate that BUILD won't instantiate with array
15:06 m6locks or at least my copy of rakudo does not do it ;PP
15:06 moritz_ m6locks: there are several things that won't work in your script
15:07 moritz_ one is that you think arrays are 1-indexed
15:07 moritz_ the second is that if you want to pass positional arguments to a construct, you have to write a custom .new method, not a BUILD method
15:07 m6locks oh ok
15:07 [particle] my irc client thinks 1..@a.elems is an email address
15:08 moritz_ m6locks: http://perlgeek.de/blog-en/perl-6/objec​t-construction-and-initialization.html
15:08 m6locks thanks, will check it out
15:08 m6locks BUILD worked with some named parameters that were not arrays
15:09 m6locks but i think i'll write a new method
15:09 moritz_ rakudo: class A { submethod BUILD { say "yes" } }; A.new
15:09 p6eval rakudo 3b1d34: OUTPUT«yes␤»
15:10 moritz_ rakudo: class A { submethod BUILD(:@b { say ~@b } }; A.new(:b[3, 4, 5])
15:10 p6eval rakudo 3b1d34: OUTPUT«Malformed method at line 11, near "BUILD(:@b "␤current instr.: 'perl6;HLL;Grammar;panic' pc 500 (ext/nqp-rx/src/stage0/HLL-s0.pir:328)␤»
15:10 moritz_ rakudo: class A { submethod BUILD(:@b) { say ~@b } }; A.new(:b[3, 4, 5])
15:10 p6eval rakudo 3b1d34: OUTPUT«3 4 5␤»
15:11 moritz_ seems to work fine here even with arrays
15:12 m6locks hmm i must've confused the notation then, gonna try if it works
15:12 kaare joined #perl6
15:27 masak *sigh*. alester asks for help writing a Plack article on Twitter, which causes mst to suggest he resign instead, and hanekomu to call him clueless. sometimes community interactions make me sad. :(
15:27 alester Where's that?
15:27 masak on Twitter.
15:27 alester Suggest I resign from what?
15:27 masak perlbuzz.
15:28 moritz_ that's not nice[tm]
15:28 alester Wow.
15:28 masak alester: fwiw, I appreciate your efforts with perlbuzz. I think others do as well.
15:28 masak I may not find everything there relevant or interesting, but that's just how it is.
15:28 * moritz_ certainly does
15:28 masak no-one's paying you for it.
15:28 PerlJam masak, alester: I do.
15:29 alester Wow.
15:29 * masak hugs alester
15:30 * moritz_ hugs mst
15:31 masak for a while, perlbuzz was one of the very few blogs out there which had a design that didn't look crap. nowadays other blogs are picking up, but it was definitely setting the standard for quite a while.
15:32 PerlJam masak: mst's comment shows that he's operating under a false assumption.
15:33 masak which comment? and which false assumption?
15:34 PerlJam masak: about passing "the perlbuzz brand onto somebody who's actually paying attention"   It seems to imply that alester is a generator of buzz rather than a reporter of buzz.
15:34 masak it does.
15:35 alester And also the assumption that perlbuzz is some sort of community property.
15:36 masak my initial reaction is still the overriding one: that no matter what mst and hanekomu think of alester, they are making themselves look like jerks by writing what they do. I don't see why anyone rational would do that.
15:36 M_o_C joined #perl6
15:36 alester And it reminds me that I want nothing to do with mst projects.
15:37 alester Catalyst, DBIx, etc
15:37 alester I personally don't use Catalyst because I don't want to deal with a community that is actively hostile.
15:38 masak Plack is an mst project.
15:38 masak at least he's contributed.
15:38 masak I'm not sure the label 'mst project' is a useful one...
15:39 masak but I agree that parts of the Perl 5 community are overly hostile, and that it doesn't have to be that way.
15:39 PerlJam masak: thus perl 6  :)
15:39 masak among other reasons, yes.
15:40 PerlJam I think it has to do with the 3 virtues of a perl programmer.  Many people exhibit false-laziness.  I think there's also some misunderstanding about what "hubris" really means in the perl universe.
15:41 PerlJam (granted it's all Larry's fault for using terms in a confusing anti-sense way ;)
15:42 Jedai joined #perl6
15:42 Guest60947 joined #perl6
15:42 gbacon joined #perl6
15:42 meteorjay joined #perl6
15:42 japhb joined #perl6
15:42 cls_bsd joined #perl6
15:42 astinus joined #perl6
15:42 lisppaste3 joined #perl6
15:42 m6locks joined #perl6
15:42 mathw joined #perl6
15:42 takadonet joined #perl6
15:42 masak "There ain't nothin' in this world that's worth being a snot over." -- Larry Wall
15:42 masak I think that applies. :)
15:43 jnthn Sure, though we're all human and can accidentally the asshole sometimes.
15:43 alester yeah, but this wasn't accidental.
15:43 masak oh, indeed.
15:44 PerlJam alester: so ... did you run over mst's cat or something?  He seems particularly hostile towards you.
15:44 jnthn Well, yes, people can just be nasty too.
15:44 alester Who knows?
15:44 * jnthn never quite knows what to make of mst. :-) I've found I can cope with him just fine so long as there's plenty of beer too. :-)
15:46 masak I generally like mst. even when it's be he's bellowing at, strangely.
15:46 masak that doesn't mean I agree with everything he says, of course.
15:46 masak especially not when he's actually being an ass.
15:47 PerlJam masak: that sums up how I feel about mst too.  Generally I like him; sometimes he goes irrationally far.
15:47 IllvilJa joined #perl6
15:47 jnthn masak: Yes, that probably sums it up pretty well.
15:49 mdxi i typically find him enjoyably bellicose and curmudgeonly, and he's rounded on me a couple of times, too. he's one of those people who define the middle ground by clearly being well to one side of it.
15:49 masak aye.
15:51 PerlJam But I wonder what it takes to get mst to *always* be against you?  I mean, in the past I've poked him enough that he blew up and stormed off (on IRC) and he and I still get along just fine.
15:52 PerlJam I bet it's some sort of fallen-hero kind of thing.  mst used to look up to an idealized alester and when he found out that alester is human, it was too much.
15:52 PerlJam now he's completely anti-alester.
15:53 alester But I'm not human.
15:53 alester I am the Loc-Nar.
15:54 masak o.O
15:59 astrojp left #perl6
16:03 REPLeffect joined #perl6
16:17 Guest60947 joined #perl6
16:22 Su-Shee well, he's funny, passionate, productive, has an opinion, does stuff, is no idiot so what more does someone want? politeness? well-behavior? what for? :)
16:23 masak sure. still a difference between not being polite all the time, and being hurtful.
16:23 [particle] less poison.
16:23 masak exactly. I'm amazed they do not see that themselves.
16:24 [particle] this cocktail is delicious! you can barely taste the arsenic.
16:25 masak [particle]++
16:25 frettled heh :)
16:26 frettled alester: I meant to alester++ for the bobby-tables.com post earlier this year, but I forgot.
16:27 frettled I can do it again: alester++
16:27 alester Thanks.
16:27 alester Patches welcome on anything on there.
16:32 xinming joined #perl6
16:32 moritz_ rakudo: sub test($test is write) { }
16:32 p6eval rakudo c06062:  ( no output )
16:33 moritz_ bad rakudo, doesn't complain about unknown traits
16:33 masak rakudo: sub test($test is hippopotamus) { }
16:33 p6eval rakudo c06062:  ( no output )
16:33 * masak submits rakudobug
16:33 jnthn It does
16:33 jnthn Where they're actually wired up.
16:33 masak ...but?
16:33 jnthn rakudo: sub yulia is krassivaya { }
16:33 p6eval rakudo c06062: OUTPUT«Could not find sub &substr␤current instr.: 'perl6;Signature;_block21375' pc 463825 (src/gen/core.pir:72226)␤»
16:33 jnthn wtf.
16:33 * masak submits rakudobug
16:33 jnthn ffs.
16:33 jnthn Well, it fails anyway.
16:34 moritz_ masak: there's already a ticket
16:34 masak oh.
16:34 masak moritz_: for the first one?
16:34 jnthn Point is that parameter traits are just NYI.
16:34 moritz_ RT #60966 - Rakudo should throw errors on unknown traits
16:34 jnthn (never were in alpha either)
16:34 masak moritz_: ok.
16:34 masak still submitting the second one.
16:34 jnthn Yeah
16:34 jnthn second one is weird.
16:34 masak jnthn: Yulia just made RT :)
16:34 jnthn Cute.
16:35 moritz_ Yulia?
16:35 masak :)
16:35 * jnthn quickly re-treats to working on fuzzy searching stuff
16:35 jnthn er, EUNREQUIREDHYPHEN
16:36 moritz_ rakudo: sub f() as Int { '5' }; say f().WHAT
16:36 p6eval rakudo c06062: OUTPUT«Could not find sub &trait_mod:<as>␤current instr.: '' pc -1 ((unknown file):-1)␤»
16:36 jnthn That's more like it.
16:36 jnthn Well, sane error
16:36 jnthn Should work of course.
16:36 moritz_ aye
16:36 jnthn YANYI.
16:37 kaare_ joined #perl6
16:37 diakopter s/UNREQUIRED/UNNECESSARY/
16:38 diakopter or SUPERFLUOUS
16:38 IllvilJa UNCOMPULSARY ?
16:38 moritz_ or simply s{ '-' } = '';
16:38 diakopter NONCOMPULSORY
16:39 [particle] ECLOSINGHYPHENNOTFOUND
16:39 moritz_ NONPULSORY?
16:39 jnthn rakudo: say "re-treat" ~~ s[ '-' ] = '';
16:39 p6eval rakudo c06062: OUTPUT«Cannot assign to readonly value␤current instr.: '&die' pc 17293 (src/builtins/Junction.pir:404)␤»
16:39 jnthn oh, yeah.
16:39 jnthn rakudo: my $x = "re-treat"; $x ~~ s[ '-' ] = ''; say $x;
16:39 p6eval rakudo c06062: OUTPUT«retreat␤»
16:39 moritz_ rakudo: sub infix:<,> { 42 }; say 5, 5
16:39 p6eval rakudo c06062: OUTPUT«55␤»
16:40 moritz_ (yes, there's a ticket for that already)
16:40 masak :)
16:40 IllvilJa UNNONOPTIONAL?
16:40 masak num &
16:41 * IllvilJa quits trolling
16:41 moritz_ nom is a good idea, akshually
16:41 diakopter that reminds me of guys at my university who wore t-shirts proclaiming "holiness is not an option; it's a command"
16:42 * moritz_ groans
16:42 jnthn yeahbut following commands is optional. ;-)
16:42 diakopter I teased them by asking, "how are you supposed to obey the command if obedience is not one of the possible options?"
16:42 jnthn lol
16:42 diakopter (it should have said "holiness is not optional; it's a command"
16:42 diakopter )
16:45 IllvilJa Hm.  "Not an option means it is a command" => "An option means it is an UNcommand"?
16:45 diakopter "an option" means "one of the possible alternatives"
16:46 TimToady 2 options == 1 choice :)
16:46 IllvilJa choice == uncommand == choice then...
16:46 IllvilJa Just trying to enrichen (?) the English language.
16:48 * IllvilJa leaves for playing a game with the small kids... :-)
16:55 justatheory joined #perl6
16:56 pugssvn r30329 | moritz++ | [t/spec] fudge signature.t for rakudo
16:59 dalek rakudo: 4e1dd48 | moritz++ | t/spectest.data:
16:59 dalek rakudo: we pass signature.t
16:59 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/4​e1dd48caf76f188e7486ba76a0abb6e988c6ea2
17:06 cdarroch joined #perl6
17:06 cdarroch joined #perl6
17:15 moritz_ http://blog.llvm.org/2010/04/amazing​-feats-of-clang-error-recovery.html
17:16 jaldhar joined #perl6
17:16 dalek book: 5fc47e2 | moritz++ | docs/release-guide.pod:
17:16 dalek book: [docs] update release guide
17:16 dalek book: review: http://github.com/perl6/book/commit/5f​c47e202a3114db31be242832ea997660545c09
17:24 astrojp joined #perl6
17:26 wasy joined #perl6
17:27 rajeshsr joined #perl6
17:43 obra joined #perl6
17:54 PerlJam moritz_: you're releasing (have released) a copy of the book today?
17:58 slavik what book?
17:58 slavik where can I buy it?
17:59 spinclad right here at http://github.com
17:59 moritz_ PerlJam: yes
17:59 moritz_ slavik: you can't, not yet
18:00 wolf2k_ubuntu joined #perl6
18:00 slavik moritz_: what's it about?
18:00 moritz_ slavik: Perl 6
18:00 slavik :D
18:00 slavik I want it
18:00 slavik where is it?
18:00 moritz_ http://github.com/perl6/book/
18:01 moritz_ there's an (unpolished) PDF version on the download page
18:03 slavik moritz_: if I stalk you long enough, will you sign it? :)
18:04 PerlJam slavik: the other authoers might be hurt if you didn't stalk them too :)
18:04 moritz_ slavik: a single meeting in meat space will be enough :-)
18:04 PerlJam s/oe/o/
18:04 slavik ;)
18:04 moritz_ anyway, the plan is still to release around the same time as Rakudo *
18:07 PerlJam SDL and parrot are getting lots of love from the gsoc, but perl 6 only a little love  :(
18:07 slavik rakudo what version?
18:08 PerlJam slavik: Whatever version ships with Rakudo Star
18:08 slavik oh
18:08 jnthn Might not even map to an actual Rakudo release either, fwiw.
18:08 jnthn Though really should run on a particular Parrot monthly.
18:09 PerlJam jnthn: in that case, I think perhaps a "special release" might be in order :)
18:10 jnthn We'll see what happens. :-)
18:14 rbaumer joined #perl6
18:14 PerlJam Grr.   I see that mst and alester are still fencing  :(
18:14 IllvilJa PerlJam: Where?
18:14 PerlJam twitter
18:15 IllvilJa Oh.  Then I'm safe ;-).  I'm 'deaf' to tweets.
18:15 obra mst reports that he's stopped ;)
18:17 zamolxes joined #perl6
18:17 [particle] second-hand tweets are as bad for you as second-hand smoke
18:20 colomon "_bsr At Hands On with Perl 6 using the rakudo compiler, taught by Damian Conway. Great stuff happening."
18:20 colomon speaking of second-hand tweets.
18:20 colomon :)
18:20 colomon that's 20 minutes ago.
18:21 rba_ joined #perl6
18:27 hercynium joined #perl6
18:34 moritz_ phenny: ask masak where he installed pdflatex from, CokeBot9000++ wants to know
18:34 phenny moritz_: I'll pass that on when masak is around.
18:38 M_o_C joined #perl6
18:38 levengli joined #perl6
18:44 alester PerlJam: I think your assessment that " mst and alester are still fencing " is inaccurate.
18:53 ShaneC joined #perl6
18:53 ShaneC left #perl6
18:56 moritz_ std: $!{ { } }
18:56 p6eval std 30329: OUTPUT«[31m===[0mSORRY![31m===[0m␤Unsupported use of $!{ { } variable; in Perl 6 please use smart match against $! at /tmp/NDRiyrTed6 line 1:␤------> [32m$!{ { }[33m⏏[31m }[0m␤FAILED 00:01 108m␤»
18:56 moritz_ hah, LTA error message from STD (notice the wrongly nested brackets) :-)
18:56 moritz_ but that's cheating, I know :-)
18:58 TimToady given what the p5 idiom is, yes. :)
18:59 szabgab joined #perl6
19:09 bkeeler G'mornin #p6
19:09 jnthn o/ bkeeler
19:14 nihiliad joined #perl6
19:24 * moritz_ takes a shot at special variables deprecation warnings
19:25 jnthn moritz_++
19:25 colomon \o/
19:26 moritz_ it's mostly just copy&paste from STD.pm
19:26 jnthn :-)
19:26 moritz_ and removing those :: about which NQP-rx complains
19:27 moritz_ $ git diff | wc -l
19:27 moritz_ 209
19:28 moritz_ std: $@
19:28 p6eval std 30329: OUTPUT«[31m===[0mSORRY![31m===[0m␤Unsupported use of $@ variable as eval error; in Perl 6 please use $! at /tmp/XVoRrgT6FF line 1:␤------> [32m$@[33m⏏[31m<EOL>[0m␤FAILED 00:01 108m␤»
19:36 colomon jnthn: ping?
19:36 moritz_ ok, got my patch ready, waiting for spectest to finish
19:38 moritz_ rakudo: for (1..3)->$n {say $n}
19:38 p6eval rakudo 4e1dd4: OUTPUT«1␤2␤3␤»
19:40 moritz_ std: for (1..3)->$n {say $n}
19:40 p6eval std 30329: OUTPUT«[31m===[0mSORRY![31m===[0m␤Unsupported use of -> as postfix; in Perl 6 please use either . to call a method, or whitespace to delimit a pointy block at /tmp/xkCY23e7MW line 1:␤------> [32mfor (1..3)->[33m⏏[31m$n {say $n}[0m␤FAILED 00:01 107m␤»
19:42 moritz_ ah; the title of RT #65212 is wrong
19:43 moritz_ STD.pm expects <ws> before the '->' in a for statement, Rakudo doesn't
19:43 moritz_ rakudo: say ')->' ~~ / ')' <.ws> '->'/
19:43 p6eval rakudo 4e1dd4: OUTPUT«)->␤»
19:43 moritz_ a <ws> matches the empty string between two non-identifier characters
19:44 bluescreen joined #perl6
19:44 TimToady LTM goof, perhaps.  STD never lets a token cross <.ws>
19:45 moritz_ so, should it be allowed with whitespace?
19:45 TimToady with whitespace it's not a postfix
19:45 moritz_ *without
19:46 TimToady without whitespace, the expression parser should have picked it up as a postfix before for's -> ever gets a shot at it
19:46 moritz_ ok
19:46 moritz_ so STD.pm is correct
19:46 TimToady I very nearly said, "Of course."  :)
19:46 moritz_ :-)
19:50 dalek rakudo: 5094077 | moritz++ | src/Perl6/Grammar.pm:
19:50 dalek rakudo: port most "Unsupported use of $weird variable" from STD.pm.
19:50 dalek rakudo: TimToady++ for coming up with them in the first place
19:50 dalek rakudo: I'm not yet convinced they all work (for example I couldn't trigger the one
19:50 dalek rakudo: for $@, although the regex seems pretty obvious), but all in all it looks like
19:50 dalek rakudo: a net win.
19:50 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/5​094077aeddbbce19902fbcc3b63bbefedd18650
19:53 xabbu42 joined #perl6
19:57 levengli how does one read STD.pm?
19:57 moritz_ one line at a time
19:57 moritz_ .oO( grammar patches work better if spell "postfix" correctly )
19:58 levengli i'll rephrase, in what language is it eritten?
19:58 levengli written
19:58 moritz_ in Perl 6
19:59 moritz_ S05 is of special interest :-)
20:00 mariano__ joined #perl6
20:01 levengli you're killing me :)
20:05 moritz_ I admit it's quite a tough read
20:05 moritz_ and i'm far from understanding it all
20:05 moritz_ std: say 1 << 2
20:05 p6eval std 30329: OUTPUT«[31m===[0mSORRY![31m===[0m␤Unsupported use of << to do left shift; in Perl 6 please use +< or ~< at /tmp/khpNxjPw3r line 1:␤------> [32msay 1 << [33m⏏[31m2[0m␤FAILED 00:01 108m␤»
20:06 moritz_ TimToady: why is there still a <O(|%multiplicative)> behind the <.obs()> in infix:«<<»?
20:06 nick joined #perl6
20:06 moritz_ (line 3610)
20:08 TimToady it's just there for old time's sake, but it's unreached code really
20:08 moritz_ ok
20:09 TiMBuS joined #perl6
20:09 moritz_ should auld acquaintance be forgot...
20:09 clintongormley joined #perl6
20:10 nick has anyone made a deb packet for ilbot?
20:11 moritz_ nick: nope; it hardcodes too many paths right now; it would need a proper installer first
20:11 nick i've been trying to install it on ubuntu at the moment. so many dependicies :)
20:12 moritz_ aye, that too
20:12 moritz_ when I wrote I really only thought of logging this channel here
20:13 meppl joined #perl6
20:21 frettled alester: Haha, I forwarded the link to Bobby Tables, and got the question: why aren't C and Lisp mentioned?  Answer from someone else: because C-coders and Lisp-coders hardly need those examples.  ;)
20:21 alester I need to put that question up
20:21 alester "
20:21 alester W
20:21 alester y
20:21 alester "Why isn't language X here?"
20:21 frettled "Answer: because you're a lazy bum and haven't provided a decent example yet."
20:22 frettled You might want to find a more diplomatic way of phrasing that.
20:22 frettled You don't want to be accidentally confused with mst, after all.  ;)
20:24 moritz_ "Answer: because our resources are limited. You can help out by providing an example on your own, which you please sent to $email_address"
20:24 frettled "Please send spam to $other_email_address."
20:24 nick left #perl6
20:25 * frettled falls over.
20:25 frettled I'm on Planet Perl 6!  Something happenedededed!
20:26 frettled If I only had something worth posting…  Perhaps after a good night's sleep.  :)
20:27 moritz_ (two more <.obs()> commits coming up if spectest run is successful)
20:30 moritz_ rakudo: use Test; plan 1; isa_ok(1|2, Junction);
20:30 p6eval rakudo 509407: OUTPUT«1..1␤Null PMC access in get_iter()␤current instr.: '!DISPATCH_JUNCTION_CORE' pc 16360 (src/builtins/Signature.pir:151)␤»
20:31 jnthn rakudo: say 1|2 ~~ Junction
20:31 p6eval rakudo 509407: OUTPUT«1␤»
20:31 jnthn rakudo: say (1|2).isa(Junction)
20:31 p6eval rakudo 509407: OUTPUT«any(0)␤»
20:31 jnthn heh, isa auto-threads...
20:32 spinclad so does isa_ok, looks like
20:32 jnthn Yeah, though I've no idea why it should then blow up like that.
20:33 moritz_ rakudo: say (1|2).^isa(Junction)
20:33 p6eval rakudo 509407: OUTPUT«1␤»
20:37 payload joined #perl6
20:41 pugssvn r30330 | moritz++ | [t/spec] unfudge test for RT #65212, for (list)-> $sig { ... } should die without whitespace before ->
20:41 frettled rakudo: say 1 ~~ Junction
20:41 p6eval rakudo 509407: OUTPUT«0␤»
20:42 dalek rakudo: e54dc62 | moritz++ | src/Perl6/Grammar.pm:
20:42 dalek rakudo: detect obsolete use of -> as postfix; fixes RT #65212
20:42 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/e​54dc62b90a281b1223243dd64f4f77456aaf969
20:42 dalek rakudo: b75b2b3 | moritz++ | src/Perl6/Grammar.pm:
20:42 dalek rakudo: catch infix << and >> as obsolete bit shift operators
20:42 dalek rakudo: review: http://github.com/rakudo/rakudo/commit/b​75b2b30a3cd75ce37dc43e9d5f0627fd580df46
20:44 hercynium joined #perl6
20:45 * moritz_ has submitted a bug to debian about an LTA error message :-)
20:45 moritz_ $ sudo route del foo
20:45 moritz_ SIOCDELRT: No such process
20:46 moritz_ it's one I've been hating for quite some time
20:46 moritz_ but never really thought of submitting before being exposed to the Perl 6 culture of wanting good error messages :-)
20:47 moritz_ anyway, enough LHF rakudo hacking for today, some sleep is appropriate
20:47 jnthn moritz_++ :-)
20:47 jnthn night o/
20:57 nihiliad joined #perl6
20:58 Chillance joined #perl6
21:17 lichtkind joined #perl6
21:17 lichtkind why is object type Str immutable?
21:18 spinclad same reason numbers are, they are value types, i would think
21:19 spinclad parrot is currently seeing some benefits to making their strings immutable
21:20 lichtkind spinclad: yeah but that calue could change no?
21:20 jnthn lichtkind: How so?
21:20 spinclad 'changing' a value produces a new value
21:21 tri1 joined #perl6
21:21 spinclad $a++ doesn't change any numbers, it puts a new number in the $a container.
21:21 spinclad likewise for $a ~= 'more string'
21:22 lichtkind spinclad: ah thanks i now get it but i think ~= is no longer on p6
21:23 spinclad yes, ~ is concat, ~= is its op= form
21:23 spinclad note: this is a functionalist point of view i have, here
21:25 lichtkind spinclad: i now understand why scalar is mutable type
21:25 spinclad excellent
21:29 wknight8111 joined #perl6
21:31 bkeeler joined #perl6
21:32 lichtkind spinclad: are all types defined my classes or some even by roles?
21:34 jnthn There are some roles too.
21:35 jnthn Iterator and Failure should be roles, I think.
21:36 lichtkind jnthn: thanks
21:37 quietfanatic joined #perl6
21:41 nihiliad joined #perl6
21:42 * lichtkind added the other 2 sigiles and som more linking and first object types
21:46 justatheory joined #perl6
21:52 nihiliad joined #perl6
21:53 aesop lol put that shit o  shit on klulz
21:55 diakopter ?
22:12 colomon joined #perl6
22:18 spinclad i think someone mistook their ircpane for a phone
22:26 jnthn colomon: oh hai
22:26 jnthn colomon: Is the roles + multi methods stuff your Top Priority Thingy for me to take a look at?
22:42 tri1 left #perl6
22:51 envi^home joined #perl6
23:04 lue o hai dere
23:04 phenny lue: 09:02Z <masak> tell lue "Some of the hardest work [in lnaguage design] isn't making the changes, it's getting them past the maintainers or an ensuing bikeshedding discussion that goes nowhere."
23:07 lue "C<S32> has always been volatile for these reasons; C<S32::Temporal> doubly so." heh heh
23:08 pugssvn r30331 | lwall++ | [STD] better message on $!{} just for moritz_++
23:18 lue masak: and I should always remember: "Everybody wants the colon". :)
23:18 * jnthn would like to keep his colon to himself.
23:21 fridim joined #perl6
23:22 lue There is a corollary, involving smileys...
23:25 Limbic_Region joined #perl6
23:35 ruoso joined #perl6
23:48 * jnthn -> sleep
23:48 ruoso joined #perl6
23:49 lichtkind joined #perl6
23:49 lue good night, jnthn o/
23:49 lichtkind good night
23:51 lue lichtkind: are you going to bed, or seeing jnthn off? (In case you are going to bed, goodnight o/)
23:52 lichtkind lue: no i thought you were waving jnthn goodnight
23:52 lichtkind i reconnected recently and didnt see
23:55 lue I was :)
23:59 lichtkind i was chatting with jnthn before forced disconnect

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

Perl 6 | Reference Documentation | Rakudo