»ö« Welcome to Perl 6! | perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg camelia perl6: ... | irclog: irc.perl6.org/ | UTF-8 is our friend!
Set by moritz on 3 May 2013.
benabik The link added to S17-concurrency is a 404. ( www.seas.upenn.edu/~lipeng/homepage/unify.html ) 00:31
[Coke] didn't someone comment on the commit with a valid url? 00:47
benabik Not that I can see. 00:51
geekosaur It wasn't a comment on the commit, it was an email reply on the list 00:55
benabik Might I suggest someone put it in the repository so people not on the list can see it? ;-) 00:56
[Coke] looks like use v6; is already doing basically nothing. 00:57
So I'm not sure how it's failing in jvm. maybe I'll be better off getting backtrace working.
[Coke] ok, backtrace is failing because the Exception object doesn't have a backtrace method, and when I try to find out what type it is, it complains: 01:08
Unhandled exception: MapIter NYI on JVM backend 01:09
RedditAnalytics have any of you ever done rm * in the wrong directory before? 01:42
benabik has not-so-fond memories of doing `del *.com` in c: by accident. 01:43
flussence I once fdisked the wrong drive
RedditAnalytics ouch
was there anything important on the drive?
flussence nah, just my $dayjob stuff :) 01:44
RedditAnalytics i think every programmer has to experience it once to learn 01:45
raiph moritz: irclogs for #perl6 and #moarvm show no new lines for last few hours 01:58
raiph .tell moritz looks like irclog is broken 01:59
yoleaux raiph: I'll pass your message to moritz.
sorear good * #perl6 02:07
lue RedditAnalytics: that's why I usually do an `ls` first (such as `ls ./*.html`) and then when it's good s/ls/rm 02:15
RedditAnalytics yeah ... I was viewing contents in another directory like ls -la some/other/place and then thought I was there
luckily it was a Dropbox folder 02:16
sorear trying to make a QAST::CompUnit object as described at dinner. have no idea what I'm doing. 02:44
sorear \o/ t/qast/01-qast.t is now failing 1 test instead of 58 02:47
dalek p: f88d35c | sorear++ | src/vm/jvm/QAST/Compiler.nqp:
CompilerJAST needs a CompUnit, so use that for wrapping
 02:48
sorear we now have 5 failing *tests*
sorear last thing in 01-qast.t is a lot of "I'm not actually sure if this should work". it's creating a typed native array, then passing it :flat to a function 03:02
this appears to be requiring nqp to box the values, but typed native arrays won't ever box on read... 03:04
sorear reads qast.markdown 03:26
sorear does not immediately see why QAST::Stmt is needed
TimToady carry debugging info? assert stack emptiness? 03:29
[Coke] sorear++ 03:30
benabik In Parrot, it helps limit the scope of temporary variables for expressions.
[Coke] sorear: what's driving the nqp hacking? 03:32
sorear [Coke]: it's failing tests on jvm! 03:34
also, yapc 03:35
[Coke] sorear++ 03:42
PerlJam yapc++
sorear "Cannot stringify this" - bane of my existance 03:48
*everything* should be stringifiable >_>
[Coke] ah, tcl, I kind of miss you 03:51
RedditAnalytics yeah just base64 encode all the binary strings
PerlJam sorear: I don't know ... that sounds quite a bit like "everything should be pronounceable" which kinda sounds like "all axoimatic systems should be self-consistent" ;) 03:53
sorear RedditAnalytics: wfm 03:55
anything as long as my debugging code doesn't crah.
(and this is nqp, so don't bring up .perl)
TimToady obviously we need a .nqp method :P 03:56
ingy looks across the table at TimToady 04:07
sorear jnthn: do we really need 64-bit cursor positions? :p 04:09
TimToady
.oO("4 gigabytes should be enough for anyone...")
 04:21
[Coke] tries to remember how to fix STable conflict detected during deserialization. 04:23
an error message so ubiquitous, google finds in the nqp source and in my irc rantings. 04:24
[Coke] how to spell __END in nqp ? 04:31
sorear __END? 04:31
[Coke] er, __END__
lue __END__ as in Perl 5's __END__? 04:32
[Coke] Yes. 04:34
sorear TimToady: java limits strings to 4 (actually 2) GB anyway 04:35
there are (int) casts all over the place
storing as long seems... pointless.
lue That's spelled =begin finish in Perl 6
(formerly =begin END, which I liked better :/ )
sorear or =finish 04:36
sorear S26:1559 "A warning is issued if an explicit C<=end finish> appears within a document." 04:37
EMISSINGTHEPOINT
lue sorear: Are you suggesting the Pod parser should bother looking for an =end finish in the first place? Because I agree 04:39
The only reason for it would be to keep the wary new coder from accidentally believing =end finish returns to code, but then again you should only be using the finish block if you know what it does already 04:40
sorear lue: I think you may be missing a "not" in your first reply
lue yes, s/should/shouldn't/
(and I doubt someone would type =begin finish ... =end finish entirely by accident) 04:41
[Coke] =finish doesn't work in nqp 04:43
r: say(3);
camelia rakudo b2072f: OUTPUT«3␤»
[Coke] bah, this terminal isn't setup to copy that newline. - but nqp runs code that appears after =finish 04:44
Anyone who wants to try to find my nqp-thinko, [email@hidden.address] //nqp2 branch 04:45
lue nqp: =begin END␤say(4)␤=end nonsense_pod_line; 04:48
camelia nqp: ( no output )
sorear o/ FROGGS & sqirrel
lue [Coke]: Apparently NQP hasn't switched to =begin finish yet
FROGGS hi sorear
[Coke] lue: danke. 04:49
lue gerne 04:50
♞, #perl6 o/ 04:53
moritz \o 05:24
yoleaux 01:59Z <raiph> moritz: looks like irclog is broken
moritz _ilbot is back now 05:25
the problem was that it lost connection to the database server, and then died
remedy: while true; do perl ilbot2.pl $configfile; sleep 10; done
FROGGS jnthn: does that look familiar? gist.github.com/FROGGS/d2c4f56d7765d8c5cac3 05:34
raiph Does anyone else wonder if the loss from having #moarvm be separate from #perl6 outweighs the gains? 07:46
moritz what loss?
raiph activity evident to curious visitors 07:47
raiph i find all commits and serious discussion related to p6 exciting. i imagine the same is true for many visitors and log lurkers 07:48
moritz I wouldn't mind merging the channels 07:50
though of course we have to ask the #moarvm folks :-)
raiph i even wonder if it wouldn't be wise to invite the parrot folk to merge too 07:52
and p2 and moe... 07:53
:)
only half kidding
moritz raiph: after irclog.perlgeek.de/parrot/2013-06-06#i_7166627 I'm not sure if the parrot folks want that
(though my impression is that "the parrot folks" isn't cohesive; much like "the perl 6 folks" isn't cohesive either) 07:54
masak is back in Sweden 08:37
oh hai, #perl6
moritz welcome back 08:38
masak thank you 08:39
I'm going to repack, shower, lie down for a while, and then get on a train.
the weirdest and least traveler-friendly part of the trip back home, oddly enough, was the check-in. 08:41
moritz huh? 08:42
it's usually annoying (queueing/waiting), but not weird
masak at the airport, they put a check-in machine *in front of* a desk with an official. so the official stands behind the desk watching you punch things in. guys, I don't think you grok check-in machines. 08:43
bonsaikitten masak: that sounds very asian ;)
did that guy have a supervisor watching him?
masak turns out it's *necessary*, though, because the check-in machine found something like 5 things it didn't like with my travel arrangements.
the two things I remember is that the name on my ticket didn't exactly match the name on my passport, and the machine couldn't seat me on my connecting flight because other airline. 08:44
a very whiny machine.
moritz oh indeed
both are not at all exceptional
masak oh, and for the flight I got to choose seating on, the screen where I got to choose a seat didn't look enough like an interactive UI, just like an info screen. so I missed my chance to choose a seat. 08:45
and when I tried to override manually with the official, she said I already had a window seat, which by the time I got to the plane turned out to be not so.
the flight was uneventful, except that it got delayed >1 h on the ground, because a "highloader" (something that loads baggage) had done the hydraulic equivalent of a bluescreen and wouldn't budge from up against the plane. 08:47
masak bonsaikitten: it does sound very Asian. I didn't see any supervisor, but there was one official per machine. mine came around the desk and helped me a few times. 09:08
flaviusb r: say 1,2,4,8...* 11:04
camelia rakudo b2072f: OUTPUT«(timeout)»
flaviusb say 1,2,4,8...* [^3] 11:05
r: say 1,2,4,8...* [^3]
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Unable to parse expression in bracketed infix; couldn't find final ']'␤at /tmp/e7zNgv7Q4X:1␤------> say 1,2,4,8...* [^⏏3]␤»
moritz r: say (1, 2, 4, 8 ... *)[^10]
camelia rakudo b2072f: OUTPUT«1 2 4 8 16 32 64 128 256 512␤»
flaviusb moritz: Thanks 11:06
moritz flaviusb: you're welcome
masak oh hai from a moving train! \o/ 11:48
masak reads the hackathon backlog with interest 11:55
masak sorear: as a Swede, I take offense at the mention of "rutabaga" in the backlog :P 11:59
[Coke] O_o 12:00
masak or not offense, really. just curious how it came up... 12:01
hm, "curious" is not the word I'm looking for. "indifferent"; yeah, that's it. 12:02
:P
masak labster++ # blog.brentlaabs.com/2013/06/thankin...-yapc.html 12:08
*sniff*
labster: s:3rd/it\'s/its/, though. 12:09
masak raiph: re merging #moarvm into #perl6 -- one of the things I keep hearing about this channel is that people (especially newcomers) don't speak because they think they'll disrupt important ongoing discussion about autopuns, your-mom jokes, lolcats, ragefaces and category theory. 12:12
...just a 0.02 € into the keep-separate pile, I guess. 12:13
[Coke] I'm used to chat servers where there would be a #rakudo, a #niecza, a #roast... this channel is already crowded. :) 12:19
masak .oO( #perl6-category-theory o/ ) 12:22
JimmyZ
.oO( maskist )
 12:23
error, Masakist
masak no-one would ever create a channel #masakism, though. that's just too far-fetched. 12:24
speaking of which! less than two weeks until next time! github.com/perl6/mu/wiki/perl6-wor...-june-2013 sign up sign up sign up
[Coke] I'm holding my sign up as high as I can. It's not doing anything. 12:32
flaviusb [Coke]: sign? 12:33
masak flaviusb: punning on "sign up".
flaviusb masak: Right - auto puns. You even hung a lantern on it. 12:35
masak I don't see the "auto" in this pun, but OK ;) 12:36
I definitely hung the lantern. I think I am this channel's lantern-hanger, in fact. 12:37
flaviusb Pun on auto-pun - rather than a pun about cars, puns that make themselves. 12:38
masak in order to get somewhere as a pun these days, you have to be self-made.
flaviusb Something something Y Combinator 12:39
masak bravo!
masak tries to write a Y-combinating pun generator 12:40
[Coke] Wouldn't that require every pun to have a point? 12:41
[Coke] feels like an eliza bot, trolling wikipedia for words to cram into my punresponder. 12:42
masak no, you can write a point-free pun.
er, I mean
no, you can write a point-free pun
[Coke]: though fwiw, that would be an awesome eliza bot. 12:43
[Coke] it's just a question of effort?
ok, this article: en.wikipedia.org/wiki/Jack_Donaghy is waaaay too long. speaking of wikipedia eliza bot fodder. 12:46
masak r: sub Y(&f) { (sub (&x) { &f(&x(&x)) })(sub (&x) { &f(&x(&x)) }) }; sub g(&recurse) { say "oh the punnery!"; &recurse() }; Y(&g) 12:48
camelia rakudo b2072f: OUTPUT«maximum recursion depth exceeded␤current instr.: 'print_exception' pc 102819 (src/gen/CORE.setting.pir:46465) (src/gen/CORE.setting:9044)␤called from Sub '' pc 322 ((file unknown):195899166) (/tmp/XsBB1dIKJu:1)␤called from Sub '' pc 362 ((file unknown):195899189) (…
masak well, clearly it recurses... :/
oh, need to eta-expand the self-applications in call-by-value languages... d'oh. 12:50
I get the feeling that the humor is dead at this point, though. :)
nwc10 flogging a dead beefburger? 12:53
[Coke] it's supposed to be beef. just waiting for the EU inspectors t... horse. it's horse, sorry. 12:58
masak nwc10: I think we have driven that pun into the ground beef by now. 13:00
nwc10 and there was me thinking that it was still a runner 13:02
diakopter masak: way to steer that one astray 13:02
masak tries to stall while thinking of something 13:04
[Coke] We need to rein these puns in, guys. 13:10
masak nay. 13:14
unbridled puns! :D 13:15
masak .oO( masak has been punning so much, he's a little hoarse right now ) 13:16
lizmat good *, #perl6! 13:44
diakopter hi
masak lizmat! \o/ 13:48
lizmat All these puns about horses, I think they're sheep
[Coke] perhaps masak should go on the lamb. 13:51
masak .oO( pun automutton ) 13:52
lizmat meh
masak .oO( a woolfy in chip's clothing! ) 13:53
sorry, didn't mean to cry woolfy.
masak .oO( baaaaaaaad puns ) 13:54
lizmat I feel a reference to a Concorde coming 13:55
lizmat (www.montypython.net/scripts/lecture.php) 13:56
[Coke] pings pmichaud 14:00
[Coke] wonders if he's driving back to DFW already.
masak lizmat: denying the antecedent at the end, but still pretty funny ;)
nwc10 who is at the hackathon today? 14:01
lizmat [Coke]: pmichaud drove back late last night 14:02
he will not be physically attending the hackathon today
[Coke] profers his "weird" nqp error as a problem for people to hack on! ;) 14:03
lizmat afaik: arnsholt, colomon, jnthn, lizmat, chip
lizmat and possibly many others 14:03
[Coke] suspects that group in particular is busy with other things. (chip) ooh, is he hacking on sixy thing? 14:03
*things
FROGGS sqirrel: ping 14:07
nwc10 They're going to come back to the channel with a pun-bomb? 14:09
masak nwc10: point of order: would a pun-bomb detonate all the existing puns out of existence, or result in an explosion of new puns? 14:12
nwc10 I was meaning a herd of new puns 14:13
nwc10 stampeding across the channel 14:16
masak .oO( just ram the puns into the channel ) 14:18
now let's all turn to page 48, and sing "Amazing Graze". 14:19
TimToady you guys are gonna be the laughing stock of the world 14:20
nwc10 just steaking our claim to fame 14:22
masak just ribbin' ya 14:24
nwc10 lizmat: was that the entire list of *everyone* in the hackathon, or just the people sitting on the Perl 6 side of the electric fence? 14:28
lizmat perl6 side 14:34
off to the hackathon
diakopter nwc10: actually no one's *at* the hackathon yet except me, genehack, and I-forget-his-name-ii-guy 14:36
genehack_ diakopter: ducks 14:37
(i mean, that's his name)
TimToady diakopter: Glo is wondering whether to take a taxi, or if someone can pick her up at 4:19ish
lizmat sitting at a table with diakopter, jnthn and chip now 14:53
colomon and arnsholt are on their way back home
TimToady is there any coffee over there, or should I make a second cup of Wolfgang Puck? 14:57
TimToady while I wait for Quartus to awaken... 14:59
lizmat there is coffee here, and breakfast 15:04
lizmat well, for now 15:05
TimToady okay, I'll guess abandon Q then; if anyone finds him on a doorstep, please adopt him 15:06
timotimo greetings * 15:08
Util nwc10: I will be at the Hackathon soon, and I think that dukeleto will be in as well.
nwc10 Util: aha. I was more curious
wondering which chefs are cooking up the awesome today 15:09
timotimo i wonder if IO::Select still works 15:34
Robsn *this was only a test* 15:36
masak Robsn: ok 1 - *this was only a test* 15:39
Robsn: Result: PASS 15:40
[Coke] is reminded of his "convert his perl5 chat bot into an perl6 irc chat bot" project. 15:48
jnthn masak: Guess you got back safely ;) 15:49
sorear good * #perl6 16:07
masak: yay for being home 16:08
sorear masak: I am told that when they shoot crowd scenes in movies, you can reasonably simulate "lots of people talking about lots of random things" by having everyone say 16:10
"rutabaga " xx *
jnthn .tell FROGGS no, that doesn't look familiar... 16:14
yoleaux jnthn: I'll pass your message to FROGGS.
timotimo now panda gives off lots and lots and lots of deprecation warnings 16:29
timotimo tadzik: could you checkout IO::Select? it says pir::can missing a signature, but the line is if pir::can($fh, 'mode') 16:39
jnthn Try fixing it to be nqp::can 16:40
timotimo mhm 16:53
error:imcc:syntax error, unexpected PREG, expecting '(' ('$P5004') - not much better i guess? 16:54
masak sorear: that makes sense. I can see that ensembling into a nice murmur. 16:55
sorear: I've heard the same advice, but for a similar Swedish word: "rabarber" (en:"rhubarb")
lizmat masak: which is actually the same in Dutch 17:00
cognominal TimToady: in an Eisenstein book, there is an expression that is employed many times, is not the index, and apparently explained nowhere. "glad tidings". Is it specific to some version of the bible? Are they specific connotations ? I know this is unrelated to you preach in YAPC::NA that I have seen on video 17:05
that's in The Printing Revolution in Early Modern Europe
sorear diakopter: 77, I think 17:07
diakopter: can't be more than that
masak lizmat: probably a loan from those parts :)
lizmat: rhubarb is fairly old, culturally. imported along the Silk Road from China. the word in various languages is probably about as old. 17:08
FROGGS thanks jnthn 17:09
yoleaux 16:14Z <jnthn> FROGGS: no, that doesn't look familiar...
masak cognominal: doesn't the word "evangelium" mean "glad tidings"? en.wiktionary.org/wiki/evangelium -- maybe of some help, I dunno... 17:10
PerlJam stevan is explaining Perl 6 traits in the advanced Moose class
woolfy stevan++ 17:14
cognominal masak: ya the denotation seems to be gospel. That's the connotations I am after. 17:17
pmichaud good afternoon #perl6 17:28
lizmat pmichaud++ !! 17:29
jnthn o/ pmichaud
jnthn Hope you had a smooth journey home 17:29
pmichaud yes, no problems
pmichaud apologies to everyone that I didn't get a chance to say goodbye to yesterday/last night 17:33
the hackathon and conference were very productive 17:34
[Coke] \o/
masak I also want to give out a blanket apology to all I didn't say goodbye to. YAPC::NA was great, and if I met you and talked to you during it, chances are I think you're great as well :) 17:38
FROGGS pmichaud: \o/ 17:39
sorear masak: :D:D:D 17:46
pmichaud: congrats for the safe trip. also, *hugs*
masak seriously, you people rock, and you're not even rock stars :> 17:47
only rakudo stars :P
(you bots rock, too)
lizmat realises she has rocked many a chair into breakage 17:48
TimToady "glad tidings" may be a refernece to "I bring you glad tidings of great joy" in the King James version of the Christmas story
except it's "good tidings" there... 17:49
masak lizmat: you rock so hard! and if you do it without anyone noticing, you're a ninja, too!
TimToady also, the carol "We wish you a merry Christmas" has both "good tidings" and "glad tidings" variations 17:50
flussence
.oO( do you qualify as a rock star programmer if you trash the stage while doing a talk? :P )
 17:51
jnthn flussence: Yes. It's essential to smash your laptop on the ground after the talk to be a true rock star programmer.
masak flussence: not if you're programming country and western all day. 17:52
timotimo jnthn: it could very well be that the error message i'm getting from IO::Select comes from my somehow broken state that i haven't been able to fix yet on my laptop
can anyone panda install IO::Socket? (and perhaps replace pir::can with nqp::can in the only source file in lib/)
dalek rl6-roast-data: 076a19f | coke++ | / (4 files):
today (automated commit)
 18:05
FROGGS uhh, 29 fails in rakudo O.o 18:06
[Coke] cannot reproduce it on the command line. 18:07
FROGGS hmmm 18:08
I dont like these kind of fails
tadzik these test failures... to accurate for star people 18:10
pmichaud only imperial phptroopers are so precise.
jnthn lunch & 18:14
[Coke] you must still be in austin! ;) 18:14
sorear WOOH no more jvm fails in t/p5regex/01-p5regex.t 18:22
colomon o.
o/
colomon hello from michigan! 18:22
sorear now let's see if I broke something in Another File
sorear Hello colomon in michigan 18:22
!!
Trip went OKish?
masak hello michigan around colomon!
raiph hi #perl6. i kept getting disconnected from #perl6 from irclog.perlgeek.de/perl6/2013-06-06#i_7167812 thru irclog.perlgeek.de/perl6/2013-06-07#i_7167885 can someone post a gist of those #perl irc lines from their personal irc client log? (and same for moarvm would be perfect) TIA 18:23
sorear we are now failing 3 tests total on JVM 18:24
dalek p: 3887f45 | sorear++ | src/vm/jvm/ (2 files):
Fix compilation of \n on JVM
 18:26
colomon sorear: I was very cramped in the airplane seats, other than that everything went great. 18:28
raiph masak: "people (especially newcomers) don't speak because they think they'll disrupt" [chatter] 18:34
are you agreeing with irclog.perlgeek.de/perl6/2013-06-07#i_7168154 ? (i can't tell)
(it looks like you are saying folk are telling you the same as what wfm and i imagine to be true for others) 18:40
s/the same as/something consistent with/
masak raiph: yes, I agree with that. 18:43
masak I'm just questioning whether as much moarvm discussion would have happened without #moarvm. 18:44
if people want to be in both places, that's already trivially easy.
if we merge and people are only interested in one kind of exciting news, well, that's a harder problem. 18:45
benabik As someone who lurks on #perl6, #moarvm, and #parrot, I really prefer them being separate. Each one can occasionally get full of very specialized discussion and it's easier to swap windows than to separate out the relevant topics. 18:46
masak benabik++ said it better than I could. 18:47
benabik (Especially given differing amount of conversation. One discussion in #parrot could get easily lost just in the noise of people poking camelia.) 18:48
kbenson benabik++ (see my recent random comments of #moarvm for example) 18:50
s/of/on/
pmichaud afk, lunch 18:51
sorear colomon: (cramped) I can imagine!
I barely fit in them myself
maybe ill-considered 18:52
masak TimToady: I'd really like to see the wedding lightning talk be put online. please with sugar on top ;) a gist is enough. 18:53
btyler masak: re the june perl workshop -- what does that consist of? I'd love to take part, but I can only get away with a certain amount of off-task-ness while at work, and the time falls squarely during the work day for most USians 18:54
masak btyler: well, peopl who devoted 100% attention for 4 horus last time didn't seem to feel they had time to spare... :)
btyler: but I don't really have any minimum requirements on your degree or amount of attention/participation. 18:55
btyler how does the 'lesson' part work? any chance of recording the happenings so folks could go through it later? 18:56
masak IRC logs.
people do the lessons by reading them.
btyler oh! didn't realize it was all over irc
masak it's all distributed! :D
(and a bloody good thing, because people take different tracks and have different speeds.) 18:57
masak train -> hotel
btyler masak: cool, thanks!
timotimo sorear: for a moment i thought you were refering to the rakudo test suite and went WAT 19:00
TimToady masak: see wall.org/~larry/1cor13.txt 19:09
jnthn sorear++ # making more stuff pass
raiph btyler: irclog.perlgeek.de/masakism/ 19:10
dalek : c5a8e1f | (Tobias Leich)++ | / (3 files):
pull in changes from rakudo since 2013.05
 19:15
: e91ce43 | (Tobias Leich)++ | t/test_summary:
append perl6 and nqp version to table
FROGGS (wifi token changes)-- 19:15
ajr_ orear: "rhubarb" was the traditional mantra 19:18
RedditAnalytics o/ 19:24
FROGGS o/ 19:25
btyler raiph++, thanks! 19:30
FROGGS TimToady: so indirect object syntax is "just" about having a bareword between identifier/name and the arglist? 19:31
and I guess the arglist in this case needs/should have braces
okay, works without... 19:32
I'm trying that now
pmichaud back from lunch 19:33
github.com/perl6/specs/issues/50 # pointing it out for possible discussion from others 19:34
dalek ecs: b6ca88e | (Elizabeth Mattijsen)++ | S02-bits.pod:
Some more adverbial combinations that sort of make sense

And are implemented now		 19:35
masak TimToady++ # wall.org/~larry/1cor13.txt 19:37
PerlJam ah, that's right. TimToady++ actually did the "love talk" 19:40
timotimo pmichaud: "EVAL" is not explicit enough, how about "GAPING_SECURITY_HOLE"? 19:41
flussence hehe
timotimo or require eval "say 'foo';", :iknowwhatiamdoing; 19:41
Util sorear: www.icu-project.org/apiref/icu4c/uchar_8h.html 19:42
flussence use JUGGLING_CHAINSAWS;
Util "Note: There are several ICU whitespace functions. Comparison:" ...
benabik timotimo: You think people won't copy/paste :iknowwhatiamdoing from stackoverflow?
timotimo :D
benabik I refer you to gkoberger.github.io/stacksort/# 19:44
eternaleye benabik: Pffffft that's glorious 19:48
dalek kudo/nom: d690ac6 | (Elizabeth Mattijsen)++ | src/core/Any.pm:
Implement the whole list of valid adverb combinations from S02:2533
 19:50
lizmat now on to some more test 19:51
masak benabik: now we just need to do that, but for Wikipedia and puns :) 19:55
dalek p: 3382c85 | sorear++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java:
Use a more correct definition of CCLASS_WHITESPACE for the JVM
 20:03
sorear 1 test left
masak sorear++ 20:04
benabik Unicode: You are in a maze of character tables, all alike.
masak sorear: won't you have to break for dinner soon? :) 20:05
oh wait, it's only, like four in the afternoon.
FROGGS more like three on my clock
colomon sorear++ 20:06
masak yes, right. three. 20:07
when I ask my body clock, I get something like that back, too. also "why is the sun setting?" 20:08
PerlJam sorear++ 20:11
dalek rl6-bench: 142e5d2 | (Geoffrey Broadwell)++ | / (3 files):
Add parse-json test for NQP and Perl 6
masak 'night, #perl6 20:12
FROGGS gnight masak
lizmat night masak! 20:15
awwaiid gnight masak!
dalek kudo/nom: e769409 | (Elizabeth Mattijsen)++ | src/core/Any.pm:
Need to revert xor logic to be correct
 20:22
d: b29e940 | larry++ | STD.pm6:
Try out feeds as statement separators
 20:23
pmichaud r: say 1 ?^ 0 20:29
camelia rakudo b2072f: OUTPUT«True␤»
pmichaud r: say 1 !?^ 0
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Cannot negate ?^ because it is not iffy enough␤at /tmp/FSVL2vYFeU:1␤------> say 1 !?^⏏ 0␤»
pmichaud how iffy does it need to be? ;-)
std: say 1 !?^ 0 20:30
camelia std ea24a29: OUTPUT«===SORRY!===␤Cannot negate ?^ because additive operators are not iffy enough at /tmp/JNvi8whRnT line 1:␤------> say 1 !?^⏏ 0␤Parse failed␤FAILED 00:00 42m␤»
pmichaud .....hokay, I guess
japhb_ pmichaud, perl6-bench has NQP and Perl 6 JSON parsing tests for you 20:36
clogs down again?
lizmat .u question mark 20:39
yoleaux U+003F QUESTION MARK [Po] (?)
lizmat .u question
yoleaux U+003F QUESTION MARK [Po] (?)
U+00BF INVERTED QUESTION MARK [Po] (¿)
U+037E GREEK QUESTION MARK [Po] (􏿽xCD􏿽xBE)
lizmat r: say 1 ¿^ 0 # pmichaud: alternative to !?^ 20:40
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Two terms in a row␤at /tmp/LGzQvSIbrn:1␤------> say 1 ⏏¿^ 0 # pmichaud: alternative to !?^␤ expecting any of:␤ postfix␤ infix stopper␤ infix or meta-infix␤ statement end␤ state…
timotimo ?^ is boolean xor? 20:41
lizmat yup
timotimo +^ is bitwise xor, yes? what does ~^ do? :) 20:42
r: say "foo" ~^ "bar";
camelia rakudo b2072f: OUTPUT«␤»
timotimo r: say "foo" ~^ "foo";
camelia rakudo b2072f: OUTPUT«␀␀␀␤»
timotimo r: say "xyz" ~^ "aaa";
camelia rakudo b2072f: OUTPUT«␤»
timotimo oic, cool beans 20:43
lizmat timotimo: S03:998
timotimo say "abc" ~| "xyz"
timotimo r: say "abc" ~| "xyz" 20:43
camelia rakudo b2072f: OUTPUT«y{{␤»
benabik Bitwise per grapheme/codepoint/? ? 20:46
PerlJam ~^ is "string xor" like in Perl 5 IIRC 20:47
PerlJam r: say "abc" ~^ " "; # should case change 20:48
camelia rakudo b2072f: OUTPUT«ABC␤»
PerlJam r: say "ABC" ~^ " "; # similar thing
camelia rakudo b2072f: OUTPUT«abc␤»
pmichaud lizmat: well, meta-prefix:<!> is supposed to negate any existing operator, so I was curious about !?^ . Doesn't seem to work, though, as ?^ isn't declared 'iffy' (perhaps it should be). 20:49
*any existing iffy operator
lizmat: oh, and I (obviously) didn't get around to Hash.new last night... I'm hopeful for tonight though :) 20:50
flussence sub flipcase(Str $a) { ($a.comb X~^ ' ').join }; say flipcase 'abcDEF'
rn: sub flipcase(Str $a) { ($a.comb X~^ ' ').join }; say flipcase 'abcDEF' 20:51
camelia niecza v24-67-gb8f846e: OUTPUT«Unhandled exception: Buffer bitops NYI␤ at /home/p6eval/niecza/lib/CORE.setting line 1511 (die @ 5) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3466 (infix:<~^> @ 4) ␤ at <unknown> line 0 (KERNEL map @ 2) ␤ at /home/p6eval/niecza/lib/CORE.setting li…
..rakudo b2072f: OUTPUT«ABCdef␤»
lizmat pmichaud: any time you have, would be excellent
pmichaud: maybe because "say 1 ?^ 1" is Nil rather than False ?
flussence r: sub flipcase(Str $a) { ($a.comb X~^ ' ').join }; say flipcase Buf.new(0xA1..0xFF).decode('iso-8859-1')
camelia rakudo b2072f: OUTPUT«àáâãäåæçèéêëìíîïðñòóôõö÷øùúûüýþÿÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖ×ØÙÚÛÜÝÞß␤»
flussence r: sub flipcase(Str $a) { ($a.comb X~^ ' ').join }; say flipcase Buf.new(0xA1..0xFF).decode('iso-8859-1').perl 20:53
camelia rakudo b2072f: OUTPUT«àáâãäåæçèéêëìíîïðñòóôõö÷øùúûüýþÿÀÁÂÃÄÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖ×ØÙÚÛÜÝÞß␤»
flussence r: sub flipcase(Str $a) { ($a.comb X~^ ' ').join }; say (flipcase Buf.new(0xA1..0xFF).decode('iso-8859-1')).perl
camelia rakudo b2072f: OUTPUT«Invalid operation on binary string␤ in method perl at src/gen/CORE.setting:4587␤ in method perl at src/gen/CORE.setting:897␤ in block at /tmp/ILmWLylttn:1␤␤»
flussence r: sub flipcase(Str $a) { ($a.comb X~^ ' ').join }; say (flipcase Buf.new(0xA1..0xFF).decode('iso-8859-1')).WHAT 20:54
camelia rakudo b2072f: OUTPUT«(Str)␤»
flussence I don't get what "binary string" means in a perl 6 context, but it feels like it's behaving p5-ish right now... 20:55
(also if I accidentally spammed people with \g control chars I apologise) 20:56
colomon I was under the impression was the intent was to use these operators with Bufs?
flussence that'd make sense, but it's weird it only complains when .perl is called 20:57
timotimo r: my @a := hyper map { say "build $_"; $_ + 10 }, 1..10; say @a; 20:58
camelia rakudo b2072f: OUTPUT«Cannot call 'hyper'; none of these signatures match:␤:(\op, \a, \b, :dwim-left(:$dwim-left), :dwim-right(:$dwim-right))␤:(\op, \obj)␤:(\op, Associative \h)␤:(\op, Associative \a, Associative \b, :dwim-left(:$dwim-left), :dwim-right(:$dwim-right))␤:(\op, Associative…
timotimo why? :( 20:59
pmichaud lizmat: (Nil v False) could very much be the case
hyper NYI
or if it is, it doesn't know what to do yet.
timotimo aaw :(
didn't you show that off in your talk already?
pmichaud ummmmmm 21:00
colomon timotimo: that talk was discussing what's in the spec, not what's implemented.
timotimo aaw
lizmat pmichaud, TimToady: is "1 ?^ 1" returning Nil a bug?
pmichaud timotimo: pmichaud.com/2013/pres/yapcna-paral...lide5.html
timotimo how about the new algorithm you had thought up on that morning? :)
pmichaud see the sixth item in the above :) :)
timotimo aaw
pmichaud lizmat: yeah, I don't see anything in the spec indicating that ?^ should return Nil 21:02
that would be... weird
r: say 1 ?^ 1;
camelia rakudo b2072f: OUTPUT«Nil␤»
pmichaud and yet there it is. :-/ 21:03
doh!
yes, I call bug.
multi infix:<?^>(Mu \a, Mu \b) { a.Bool ^^ b.Bool }
that's... wrong.
lizmat where does that live? 21:06
pmichaud src/core/Bool.pm
line 62
lizmat I could spectest a change?
timotimo pmichaud: what do you mean with "doesn't know what to do yet"? does that have a deeper meaning or is it just your interpretation of what the output is?
pmichaud timotimo: I didn't even think there's a &hyper implemented already.
but apparently there is. it's wrong.
pmichaud wonders if he's to blame there. 21:07
timotimo it seems like the hyper that's implemented is what's being used interally to service >> and >>OP<<
pmichaud right... in which case it shouldn't be called "hyper"
or there needs to be another multi candidate.
timotimo mhm
not quite sure how i could implement that multi candidate to work together with map
unless i could specialcase "hyper map" in the grammar or something :| 21:08
i think that's what you meant when you said you didn't know yet how to make the context flow
timotimo was about to demonstrate hyper and race to a friend 21:09
lizmat pmichaud: multi infix:<?^>(Mu \a, Mu \b) { so( a.Bool ^^ b.Bool ) } ?? 21:10
colomon pmichaud: I seem to recall that 1 ?^ 1 being Nil is intentional
or is that just 1 ^^ 1 ?
pmichaud colomon: that's just 1 ^^ 1 21:11
pmichaud anything with a ? prefix ought to return a Bool. 21:11
colomon rn: say 1 ?^ 1
camelia rakudo b2072f: OUTPUT«Nil␤»
..niecza v24-67-gb8f846e: OUTPUT«False␤»
timotimo r: say perl "foo" +^ "FOO" 21:12
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Undeclared routine:␤ perl used at line 1␤␤»
colomon I stand corrected.
timotimo r: say ("foo" +^ "FOO").perl
camelia rakudo b2072f: OUTPUT«Cannot convert string to number: base-10 number must begin with valid digits or '.' in '⏏foo' (indicated by ⏏)␤ in method Int at src/gen/CORE.setting:10157␤ in sub infix:<+^> at src/gen/CORE.setting:3129␤ in sub infix:<+^> at src/gen/CORE.setting:3126␤ in block…
timotimo ah
pmichaud lizmat: I think that ?^ ought to head more towards low-level ops than add more high-level dispatches onto the sequence
(calling "so" adds another level of dispatch to finding the result of what ought to be.... really fast.)
pmichaud otoh, I don't have a faster / better mechanism in mind yet, so using "so" looks like the best option for now. 21:14
dalek p: 527419e | sorear++ | src/vm/jvm/runtime/org/perl6/nqp/ (3 files):
Allow flattening native-typed arrays into argument lists on JVM
 21:15
pmichaud and we need roast tests :)
lizmat pmichaud: isn't there a nqp::bool or so?
pmichaud nqp::p6bool
timotimo nqp::p6bool afaict, but it's the other way around, no?
lizmat I was thinking: multi infix:<?^>(Mu \a, Mu \b) { nqp::p6bool( a.Bool ^^ b.Bool ) } 21:16
pmichaud that works for me for now
lizmat that would be lower level
ok, I will spectest this
pmichaud ultimately it may want to be nqp::p6bool( nqp::mumble( a.Bool, b.Bool ) ) but I'd go ahead and go with ^^ for now :) 21:17
sorear for those of you not lurking in #tcc2.122: nqp-jvm is now passing *all* of its tests.
pmichaud \o/ # sorear++
colomon \o/ # sorear++ indeed! 21:17
timotimo sorear: now bring rakudo-jvm to 100%! :))
sorear timotimo: that's my next goal. 21:19
timotimo awesome! :) 21:19
lizmat pmichaud: nqp::p6bool(nqp::xor(True,False)) 21:20
colomon sorear++
pmichaud r: say nqp::xor(True, False); 21:21
camelia rakudo b2072f: OUTPUT«True␤»
pmichaud r: say nqp::xor(True, True);
camelia rakudo b2072f: OUTPUT«No such method 'gist' for invocant of type 'Undef'␤ in sub say at src/gen/CORE.setting:11047␤ in block at /tmp/tvqNJ3M_dX:1␤␤»
pmichaud r: say nqp::p6bool(nqp::xor(True, True));
camelia rakudo b2072f: OUTPUT«False␤»
pmichaud yeah, that will work.
lizmat pmichaud: the p6bool is needed apparently
pmichaud oh, I figured something would be needed, as nqp::xor implements the semantics of the high-level xor, not the low-level one. 21:22
r: say nqp::xor(0, False, "", 3, "", Nil)
camelia rakudo b2072f: OUTPUT«3␤»
pmichaud r: say nqp::xor(0, False, "", 3, "", Nil, 6) 21:23
camelia rakudo b2072f: OUTPUT«No such method 'gist' for invocant of type 'Undef'␤ in sub say at src/gen/CORE.setting:11047␤ in block at /tmp/3CGTGEgUxT:1␤␤»
pmichaud r: say nqp::xor(0, False, "", 3, "", Nil, 6) // "xyz"
camelia rakudo b2072f: OUTPUT«No such method 'defined' for invocant of type 'Undef'␤ in block at /tmp/ISon1qNjK_:1␤␤»
pmichaud bah
FROGGS :o)
pmichaud anyway... yes, there's a Parrot leak in nqp::xor. 21:24
(and that's okay.)
lizmat result of spectest in ~ 20 mins 21:30
sorear so I kind of just told jnthn that I'm going to implement gather/take infrastructure in nqp-jvm
timotimo go sorear++ :) 21:33
should i check back in an hour for results? :)
sorear timotimo: No 21:34
it's very... clever... but will take a while
timotimo hm, nqp-jvm passes all of its test, but it doesn't do any gather/taking?
sorear timotimo: Right, no gather/take in core NQP 21:36
timotimo: rakudo-parrot uses Q:PIR for implementing gather/take 21:37
timotimo ooooh, i see!
sorear I'm going to add (JVM-specific) nqp::jvm_coro_whatever ops that Rakudo can use when appropriate
although I'm starting to muse that "delimited one-shot continuation" may be a nicer abstraction 21:38
timotimo did you see how the pypy people imlement stackless? 21:39
they have something called a "continulet"
dalek rl6-bench: fa8c1b1 | (Geoffrey Broadwell)++ | TODO:
Remove a completed TODO
 21:50
rl6-bench: 5201975 | (Geoffrey Broadwell)++ | EXPECTED:
Minor fixes to EXPECTED
timotimo oh, couldn't hyper just take a ListIter and hyper-hyper it just like the >>. things do?
not sure actually how to test for that 21:54
dalek kudo/nom: 0dceace | (Elizabeth Mattijsen)++ | src/core/Bool.pm:
Booleanify ?^, as discussed with Pm

As shown at irclog.perlgeek.de/perl6/2013-06-07#i_7170609		 21:59
timotimo hm. i kind of think, that the "stub code executed" warning could totally be lifted to compiletime 22:10
hm ... no, that doesn't make sense 22:11
but we warn at check time if a multi call will not work with the given parameters
if we figure out at compile time that the method does nothing but ... we could explode, too 22:12
lizmat chaos enusing here at the hackathon 22:15
ensuing
timotimo the good or bad kind?
colomon lizmat: what's going on? 22:16
lizmat breaking up the hackathon, lots of cats without herders 22:17
timotimo are they climbing on shelves and sitting in boxes?
FROGGS *g* 22:21
they are playing with cords at least
lizmat sort off, signing off!
sorear &.
timotimo so, er, how do i copy a file in a non-deprecated way? 22:23
timotimo ah, i get it 22:26
nwc10 sorear++ # All tests successful. 22:34
raiph r: say ++sorear for 1..Inf # timotimosorear: now bring rakudo-jvm to 100%! :)) ... soreartimotimo: that's my next goal. 22:35
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Undeclared routine:␤ sorear used at line 1␤␤»
timotimo :3 22:36
nwc10 can we evaluate 1..Inf in finite time? If not, that feels horribly like an unintended DOS
timotimo with a zeno machine you can! 22:37
raiph well, i'm pretty happy that sorear is not declared routine
timotimo i have a patch that passes spectests and makes "eager map { ... }, @..." not explode, but the hyper-reordering doesn't work, because i had a braino 22:38
does that sound like something that rakudo should get?
maybe it should warn to stderr that hyper doesn't actually hyper yet in that case?
sorear good * #perl6 22:54
timotimo that's a hello or goodbye? 22:55
sorear a hello.
timotimo hello sorear :)
sorear it's paired with the &. from earlier. 22:56
timotimo ah, indeed
flussence I wrote some code that does nothing useful, but it's interesting to see how moving bits outside the inner loop one at a time speeds it up: gist.github.com/flussence/5732972
flussence (I'm messing around with putting "is pure" on random things in core to see if that has any effect) 22:58
timotimo good luck! 22:59
flussence it hasn't caused any spectest failures so far at least :)
nwc10 sorear: I think I preferred this random factoid: 23:00
22:59 <purl> well, all tests successful is <reply>That's great, Nicholas! Now the only bugs that remain are the ones you haven't tested for yet!
but the thing was - thanks massively for figuring out the fixes for the 5 failing NQP-JVM tests
(delusional) happiness is a passing testsuite 23:01
TimToady well, 'is pure' where you don't need it is likely to only hurt things the second time through, and most tests are once-through... 23:09
s/don't need it/need to not have it/
timotimo tadzik: i have happies for you :3 23:11
tadzik timotimo: happies? :) 23:12
timotimo on github
tadzik oh
tadzik look
good things :)
will merge them tomorrow, when I get to test it
timotimo oke
timotimo good call, it doesn't actually work yet 23:13
something is wrong ... 23:18
"==> Fetching File::Tools" ... nothing happens 23:19
perhaps due to the repository not having the right commit or something? 23:20
tadzik hmm 23:27
no idea
dalek ast: 688c39f | (Elizabeth Mattijsen)++ | S32-hash/delete-adverb.t:
Some more tests for adverbial pair combinations

Comments are still rough, will fix later		 23:31
lizmat dinner&
TimToady sorear: diakopter is meeting us in lobby now for dinner 23:32
timotimo tadzik: it's kind of hard to find thingg slike that :| 23:34
splitcells is it possible to launch binaries from a perl6 script? 23:38
colomon splitcells: yes
splitcells how?
timotimo r: shell "echo 'foo'"; 23:39
camelia rakudo b2072f: OUTPUT«shell is disallowed in restricted setting␤ in sub restricted at src/RESTRICTED.setting:2␤ in sub shell at src/RESTRICTED.setting:7␤ in block at /tmp/K8MDX7lif1:1␤␤»
timotimo (except not allowed in the evalbot)
splitcells ty 23:43
timotimo tadzik: i can't find any usage of .announce('fetching', but it displays the Fetching somethingsomething string, the fuck? 23:46
tadzik timotimo: um, is it not in Panda.pm?
timotimo oh ... facepalm, i guess 23:47
tadzik sleep time for me, see you tomorrows o/ 23:48
flussence okay, I tried rerunning that script I posted above after putting «is pure» on Buf.new, .decode and Str.comb. Doesn't break tests but also doesn't change speed at all. 23:53
(also I've noticed significant pauses every second or so in the loop)
(and .comb is by far the slowest part of the whole thing) 23:54
timotimo oh, it seems like it's just utterly crazy-slow 23:55
or maybe not.
flussence if I move the Buf.new(...).decode outside the loop, it goes from 7 to 5 seconds per 100 iterations - but moving .comb out too makes it go down to 1.5s per 100 23:56
timotimo yeah, comb is weirdly slow :( 23:58
flussence it's possible to improve it by hand: ($chars.substr($_, 1) for (^$chars.chars).roll(16)) 23:59
(that gets it down to 2s)

Powered by Cro and the Raku® Programming Language.

Please report any issues / comments / feature requests as an issue on App::Raku::Log.

Thank you!