»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.
masak 'night, #perl6 00:17
grondilu rn: sub postfix:<X²>(@a) { @a X @a }; say <foo bar> X² 01:39
camelia rakudo bfd850: OUTPUT«===SORRY!===␤Bogus statement␤at /tmp/TfncIzgyN4:1␤------> ix:<X²>(@a) { @a X @a }; say <foo bar> X⏏²␤ expecting any of:␤ postfix␤ infix stopper␤ infix or meta-infix␤ prefix or term␤ prefix o…
..niecza v24-48-g1d127e4: OUTPUT«===SORRY!===␤␤Bogus term at /tmp/bIx_A8yZzC line 1:␤------> ix:<X²>(@a) { @a X @a }; say <foo bar> X⏏²␤␤Parse failed␤␤»
TimToady you can't put a space before a postfix like that 01:39
grondilu rn: sub postfix:<²>(@a) { @a X @a }; say <foo bar>² 01:40
camelia rakudo bfd850, niecza v24-48-g1d127e4: OUTPUT«foo foo foo bar bar foo bar bar␤»
TimToady nr: say ((2**100) +| (2**50)).fmt('%x') 01:58
camelia niecza v24-48-g1d127e4: OUTPUT«10000000000004000000000000␤»
..rakudo bfd850: OUTPUT«0␤»
TimToady r: say (2**100).WHAT
camelia rakudo bfd850: OUTPUT«(Int)␤»
TimToady rakudobug, I guess
colomon oooo, rakudobug!
where's masak when you need him?
TimToady r: say 0x100000000000000000000000000000000 +| 0x7777777777777777777777 01:59
camelia rakudo bfd850: OUTPUT«340282366921082889801291235130506442615␤»
TimToady r: printf "%x", 0x100000000000000000000000000000000 +| 0x7777777777777777777777 02:00
camelia rakudo bfd850: OUTPUT«77777777»
TimToady r: (0x100000000000000000000000000000000 +| 0x7777777777777777777777).base(16)
camelia rakudo bfd850: ( no output )
TimToady r: (0x100000000000000000000000000000000 +| 0x7777777777777777777777).base(16).say 02:00
camelia rakudo bfd850: OUTPUT«100000000007777777777777777777777␤»
TimToady bug is in printf, I guess
TimToady grondilu: is that good enough for you? 02:01
TimToady bad net weather tonight... 02:04
TimToady wonders if this is just the typical upswing in cyberattacks at the start of a weekend... 02:04
colomon TimToady: any thoughts on what, say, Set.Str should return? 02:32
TimToady r: say Set.Str 02:38
camelia rakudo bfd850: OUTPUT«use of uninitialized value of type Set in string context in block at /tmp/Kd4oLASDtH:1␤␤␤» 02:38
TimToady that looks right to me
TimToady or are you asking about Set:D ? 02:38
TimToady I suspect it should be the same as $set.keys.Str in that case 02:39
colomon no, Set:U 02:44
so returning the empty string with a warning is okay?
the bits I found in the spec on this were very vague 02:45
TimToady that's how all type objects are supposed to behave 02:46
colomon works for me, I just wanted "official" confirmation. 02:48
rn: say Int.Str
camelia rakudo bfd850: OUTPUT«use of uninitialized value of type Int in string context in block at /tmp/MHSgfk353t:1␤␤␤»
..niecza v24-48-g1d127e4: OUTPUT«Int()␤»
TimToady rn: say Str + 42 02:49
camelia niecza v24-48-g1d127e4: OUTPUT«42␤»
..rakudo bfd850: OUTPUT«use of uninitialized value of type Str in numeric context in block at /tmp/PZnP1j7pFM:1␤␤42␤»
colomon rn: say Str 02:50
camelia rakudo bfd850, niecza v24-48-g1d127e4: OUTPUT«(Str)␤»
colomon rn: my $a; say $a 02:50
camelia rakudo bfd850, niecza v24-48-g1d127e4: OUTPUT«(Any)␤»
TimToady that .gist, not .Str
colomon TimToady: yes, I know
just trying to get an uninitialized variable warning. in practice, it seems like Niecza is hitting me with them all the time... 02:51
rn: my $a; say "$a" 02:52
camelia niecza v24-48-g1d127e4: OUTPUT«Use of uninitialized value in string context␤ at /home/p6eval/niecza/lib/CORE.setting line 1355 (warn @ 5) ␤ at /home/p6eval/niecza/lib/CORE.setting line 266 (Mu.Str @ 15) ␤ at <unknown> line 0 (ExitRunloop @ 0) ␤ at /tmp/tWVlH019n6 line 1 (mainline @ …
..rakudo bfd850: OUTPUT«use of uninitialized value of type Any in string context in block at /tmp/T94yzYeA7E:1␤␤␤»
colomon ah....
TimToady rehearsal & 02:55
colomon o/
colomon n: say Int.Str 03:08
camelia niecza v24-48-g1d127e4: OUTPUT«Int()␤»
dalek ecza: fa3b5ca | (Solomon Foster)++ | lib/CORE.setting:
Fix Set, KeySet, Bag, and KeyBag .Str methods.
 03:13
snoopy rakudo say [^] (2..10) 03:25
colomon snoopy: too big... 03:26
r: say [^] (2..10) 03:27
camelia rakudo bfd850: OUTPUT«one(2, 3, 4, 5, 6, 7, 8, 9, 10)␤»
timotimo_ hehe
colomon err, not. was thinking [**] (2..10)
PerlPilot some memes are hard to shake
colomon hmmm. 03:28
r: say one(2..10)
camelia rakudo bfd850: OUTPUT«one(2, 3, 4, 5, 6, 7, 8, 9, 10)␤»
PerlJam snoopy: where is your sopwith camel? 03:29
colomon PerlJam: my little boy has been talking about snoopy and the bunny red baron a lot recently... 03:30
PerlJam I'm not even sure my kids would even know snoopy 03:31
They haven't seen enough Peanuts
colomon I don't think we've ever shown him anything but maybe the Christmas special. But we've read a couple of the sopwith camel comic strips in the paper, and I guess he saw something with the red baron at Granny's. He seems to know the song, too... 03:34
colomon rn: say (for 1..10 { $_ if .is-prime; }).Str 03:50
camelia niecza v24-48-g1d127e4: OUTPUT« 2 3 5 7 ␤» 03:51
..rakudo bfd850: OUTPUT«2 3 5 7␤»
colomon rn: say (for 1..10 { $_ if .is-prime; }).perl
camelia niecza v24-48-g1d127e4: OUTPUT«(Empty, 2, 3, Empty, 5, Empty, 7, Empty, Empty, Empty).list␤»
..rakudo bfd850: OUTPUT«(2, 3, 5, 7).list␤»
[Coke] .tell TimToady - can we get a ruling on rt.perl.org/rt3/Ticket/Display.html?id=93978 and rt.perl.org/rt3/Ticket/Display.html?id=77146 so I don't have to migrate the ticketds? 04:15
yoleaux [Coke]: I'll pass your message to TimToady.
[Coke] rakudo: class Car { has @.wheels }; say Car.^attributes.perl 04:26
camelia rakudo bfd850: OUTPUT«(Attribute.new(),)␤»
[Coke] .tell masak please respond on rt.perl.org/rt3/Ticket/Display.html?id=78316 04:27
yoleaux [Coke]: I'll pass your message to masak.
[Coke] r: class A { ... }; class B is A { }; class A { }; say B.new 04:28
camelia rakudo bfd850: OUTPUT«No such method 'new' for invocant of type 'B'␤ in block at /tmp/TZO9Ec3Lz7:1␤␤»
dalek rl6-bench: 6be544e | (Geoffrey Broadwell)++ | TODO:
A couple more TODO items
 04:40
rl6-bench: ce2ed23 | (Geoffrey Broadwell)++ | bench:
Stop saying 'eager do for' now that for is eager again
rl6-bench: 015c246 | (Geoffrey Broadwell)++ | bench:
Minor cleanups and comment improvements
rl6-bench: 587238c | (Geoffrey Broadwell)++ | bench:
Add TIMINGS_DIR to existing COMPONENTS_DIR and use both more consistently throughout bench code
perl6-bench: db3c7c1 | (Geoffrey Broadwell)++ | bench:
perl6-bench: Refactor generating command line options flags from Pairs into as-options()
[Coke] RT-- # "Service temporarily unavailable. 04:42
[Coke] rakudo: module Soric4::Event; macro foo is parsed( 0 ) { } 04:57
camelia rakudo bfd850: OUTPUT«===SORRY!===␤Cannot call 'trait_mod:<is>'; none of these signatures match:␤:(Mu:U $child, Mu:U $parent)␤:(Mu:U $type, :rw(:$rw)!)␤:(Mu:U $type, :nativesize(:$nativesize)!)␤:(Mu:U $type, :hidden(:$hidden)!)␤:(Mu:U $type, *%fail)␤:(Attribute:D $attr… 04:58
TimToady yes, comparisons should return failure, not automatically thrown an exception. a Match object should return both array and hash keys in the absence of further guidance as to which is wanted (implementations tend to store the numeric keys in the hash in any case for convenience) 06:03
yoleaux 04:15Z <[Coke]> TimToady: - can we get a ruling on rt.perl.org/rt3/Ticket/Display.html?id=93978 and rt.perl.org/rt3/Ticket/Display.html?id=77146 so I don't have to migrate the ticketds?
sorear hi TimToady. do you have a few to talk about yapc::na? 06:04
diakopter sorear: www.yapcna.org/yn2013/wiki?node=Hackathons 06:05
most of the p6'ers are getting there Sat. afternoon/evening and departing Fri and some Thurs 06:06
TimToady sorear: I can talk about yapc 06:07
diakopter I don't mean to presume that's what you wanted to talk about with TimToady; just injecting potentially informative data
sorear diakopter: i see.
1. I'm trying to establish what is being done before and after that I might want to be present for / that others want me to be present for 06:08
2. I'm trying to find information about the job fair
concrete information, with dates and places and stuff
diakopter I can help with 2 definitely
TimToady can't
diakopter ERGH
sorear ERGH? 06:09
TimToady on 1, I'm not sure who's planning to work on what when, but some mighty interesting stuff is going down
diakopter the date/times for the jobfair aren't on www.yapcna.org/yn2013/job-fair.html *facepalm*
sorear if they were I wouldn't ask here :[
diakopter sorear: for instance, FROGGS & lizmat & I have a start of slushy/umimplemented thingies that I hoped to at least begin trying to wrap our heads around for an eventual cut of 6.0.0 spec release: ethercalc.org/i982ooyabg 06:10
TimToady has mostly be in denial about this summer, and how much traveling he has scheduled, his executive secretary having arranged all that while he was recovering from surgery... 06:11
diakopter that's probably 15-20% of eventual total count... :S
TimToady *been in
[Coke] .ask TimToady can you rule on rt.perl.org/rt3/Ticket/Display.html?id=88750 also ? 06:13
yoleaux [Coke]: I'll pass your message to TimToady.
diakopter sorear: I'm half-excited about going through S03 & S05 for that list, and half dreading it
I'm hoping others dive in
diakopter sorear: the job fair is on Tues midday 06:15
sorear diakopter: should I stab #yapc about the job fair stuff?
diakopter it's on the "Schedule" but not the page about it
I'll fix that tonight
the companies at the job fair: the sponsors on teh left bar silver & above
sorear so, what do we need to do to be involved? 06:16
diakopter in the job fair?
sorear yes.
diakopter ah; just show up at the room. companies will have tables and displays and swag and handouts and stuff 06:17
sometimes free books, equipment
sorear I was not clear. I am enquiring about the job fair on behalf of my current employer.
TimToady [Coke]: as far as I can see through my clouded branes, an our sub oughta be okay just about anywhere, insofar as it's just a global really. I wouldn't necessarily expect it to instantiate $?CLASS when composed though.
yoleaux 06:13Z <[Coke]> TimToady: can you rule on rt.perl.org/rt3/Ticket/Display.html?id=88750 also ?
diakopter ah
sorear We wish to recruit.
diakopter see this link:
www.yapcna.org/yn2013/levels-of-sponsorship.html 06:18
silver & above donation gets a table at the job fair
hrm, this must've not been well advertised on the [email@hidden.address] list 06:19
[Coke] TimToady++ 06:20
sorear I'm not on that list
[Coke] I will attempt to integrate that into some kind of ticket resolution.
that list hasn't see a lot of action, I don'tt hink.
diakopter sorear: however, the job fair room is kinda full now.. yet another thing that's missing info on the site - cutoff date for getting in the job fair and getting your logo on everyone's t-shirt
I'm not sure what that date is, but I'm guessing "next week" 06:21
sorear (it's possible that someone else at $employer is on the list)
TimToady this is the problem with having different people run the conference every year :)
[Coke] nom: class A { role B {} }; A::B.new; say "alive"
camelia rakudo bfd850: OUTPUT«alive␤»
sorear diakopter: The web site implied that non-sponsors could get a table for $200
[Coke] nom: role A { role B {} }; A::B.new; say "alive"
camelia rakudo bfd850: OUTPUT«Could not find symbol '&B'␤ in method <anon> at src/gen/CORE.setting:10027␤ in any at src/gen/Metamodel.nqp:2504␤ in any find_method_fallback at src/gen/Metamodel.nqp:2492␤ in any find_method at src/gen/Metamodel.nqp:939␤ in block at /tmp/FGHQnny8UI:1␤␤»…
diakopter :) yet another thing I didn't know
sorear: no one's taken advantage of the $200 offer yet this year 06:22
sorear should I subscribe to the yapc list? 06:23
diakopter when you register for the conf, you're auto-subscribed, but sure
sorear oh, huh
should I give up now re. job fair tables?
diakopter yet another thing that's missing from the page with the register link
TimToady [Coke]: on the other hand, I could argue that everything inside a role doesn't really exist until instantiate as a class 06:24
including another role
diakopter sorear: give up? no..
sorear if not, who should I talk to? 06:25
diakopter ah.
yes, join #yapc on magnet and ask there
I'll answer you there :P kidding, half. toddr or apeiron or others can answer too
to send the $200, just go to www.yapcna.org/yn2013/purchase and type 200 in the donation box 06:26
sorear you may notice I'm already on.
hmm. is it necessary to put the money in ASAP? I would like to wait until it's a surer thing 06:27
diakopter no 06:28
they can proceed provisionally
sorear what's the arrival dinner thing about 06:36
diakopter click Blog
there's a helpful posting on it
there's usually an "official"[ly arranged] arrival dinner, and various anti-[anti-[anti-]]]arrival dinners :) 06:37
it's certainly not expected that everyone goes; it's just yet another social event [like the entire conference] 06:38
[also, it's limited to 120 people there, but it seems only 18 people are currently paid up for it] 06:40
[Coke] rakudo: class A {}; class B {}; subset C of A & B; say C ~~ A; say C ~~ B # RT#70844
camelia rakudo bfd850: OUTPUT«True␤False␤»
TimToady I'm pretty sure that's parsing as (subset C of A) & B 06:41
TimToady is a little surprised it's not giving a "Useless use of & in sink context" 06:42
diakopter tho I think TimToady's handler/booking-agent is planning on them going.. 06:43
TimToady is too stupid to figure out what his login name is...
diakopter it starts with larr and ends with arry 06:44
TimToady and then when it asks me to change the password, it asks for the old password first...
[Coke] rakudo: class A {}; class B {}; subset C of (A & B); say C ~~ A; say C ~~ B # RT#70844
camelia rakudo bfd850: OUTPUT«===SORRY!===␤Confused␤at /tmp/XuBccnEAYS:1␤------> class A {}; class B {}; subset C of ⏏(A & B); say C ~~ A; say C ~~ B # RT#708␤ expecting any of:␤ statement list␤ horizontal whitespace␤ postfix␤ …
diakopter TimToady: it's a feature! security by confusion
TimToady [Coke]: note that the 'of' trait only allows a single name 06:45
you'd have to write: subset C of Any where A & B 06:46
[Coke] rakudo: class A {}; class B {}; subset C of Any where A & B; say C ~~ A; say C ~~ B # RT#70844 06:47
camelia rakudo bfd850: OUTPUT«False␤False␤»
TimToady subset isn't about inheritance, except kinda by accident
diakopter TimToady: seriously though, it ignores that field; try it! :D
TimToady if you want multiple inheritance, subset isn't the way to get it 06:48
[Coke] rakudo: class A {}; class B {}; subset C of A|B; say C ~~ A; say C ~~ B # RT#70844 06:50
camelia rakudo bfd850: OUTPUT«True␤False␤»
TimToady "useless use of | in sink context"
we haven't had junctional types in Perl 6 for most of the last 10 years 06:54
[Coke] ticket's from 2009
TimToady so someone was remembering the old days
junctional types were deemed to be impossible to understand, and a general public nuisance 06:55
on top of which, we stole the syntax
since signatures can contain | or & that don't mean junctions now 06:56
so I can pretty safely say that you'll never just be able to throw A&B or A|B in where were currently parse a single type 06:57
I'd just close that one as Not a Bug 06:59
[Coke] Done. 07:00
(along with these notes)
moritz [Coke]++, TimToady++
TimToady oops, how'd it get to be tomorrow already? I was supposed to turn into a pile of rubble an hour or so ago... 07:01
diakopter caught in a time loop for ages..
[Coke] oops. TimToady++
TimToady maybe I'm getting better... 07:02
and actually, & was always a problem as long as we could bind to a &func parameter 07:04
TimToady --> pile of rubble & 07:07
sorear do I need to print a ticket or anything like that? 07:12
diakopter nope; you pick up your badge/lanyard on Sunday afternoon or Monday morning
just say "my name is ____" and they'll believe you 07:13
moritz just come in and say "my name is Larry Wall", and they won't believe you (unless TimToady does it :-) 07:14
diakopter actually a fake Larry Wall did register .. (oh wait, that was me) 07:15
sorear hmm. how is ____ pronounced?
diakopter in gaelic?
sorear wonders if he should bring a passport in case somebody in texas gets the idea to reject a calif ID card 07:16
diakopter nah 07:17
it'd be a pain to carry everywhere. every traveler everywhere knows to keep it on your person at all times; they make special pouches with straps for inside clothing even, specifically for that 07:19
just so it's not accidentally lost/stolen 07:20
diakopter [I don't mean to imply you don't know this already] 07:20
masak good morning, #perl6 07:32
yoleaux 04:27Z <[Coke]> masak: please respond on rt.perl.org/rt3/Ticket/Display.html?id=78316
masak looks
moritz good oh-it's-really-still-morning, masak :-)
masak ;) 07:33
(need to catch a train, so I'm up early-ish) 07:34
[Coke]: yes, and here's the general rule: if the ticket says "weird error" and there's no longer a weird error, the ticket is closable ;)
masak resolves the ticket 07:35
sorear gooood moooooorning maaaaaasak! 07:36
masak *giggle* good night, Tessie. 07:39
(fun trivia question: from which well-known musical is that line?) 07:40
sorear i didn't think I was from anything in particular 07:41
I just express my excitement by stretching vowels
s/think I/knot it/
s/knot/know/
masak well, it clearly reminded me of that scene, so... :) 07:42
sorear but. I'm coming to yapc::na, and my employer may be about to take a last-minute sponsorship
so I'm excited :D 07:43
sorear haven't figured out yet whether I'm staying for either hackathon 07:43
moritz if you have the chance, do it
moritz hackathons are always great fun 07:43
masak sorear: that is exciting indeed. looking forward to meeting.
train & 07:44
sorear yeah, I can, pre-approved for that 07:50
diakopter sorear: tadzik will be there on Sunday but has to leave early for school exams 08:05
bed, ish&
tadzik assuming I get this damned visa 08:39
sorear ICE--
or is it CBP-- I never can remember who handles what 08:40
tadzik whoever they are, well, I'll express a more elaborate opinion _after_ they make their decision
but the idea of a legal phishing of my bank account credentials still makes me shiver 08:41
sorear our immigration policies are insane, and when people here say "reform" they mean "make it worse" 08:42
sorear tries to refrain from ranting 08:43
tadzik well, maybe one day Poland will be considered a civilized country too, and most of the mess could be skipped
sorear they don't even give painfree visas to UK folks 08:45
how many times is the List.sink bug going to be reported? 09:04
hoelzro_ is the Perl 6 reddit active? 09:06
hoelzro I was thinking about Perl 6 compilers last night and wanted to share my musings =) 09:11
tadzik hoelzro: I subscribe it. It's not too active though 09:17
most news end up on #perl too anywa
hoelzro =(
well, I would prefer more stuff going onto #perl
tadzik the perl 6 things submitted to #perl usually get upvoted as well 09:23
hoelzro so what I was wondering was this: when Perl 6 started, why didn't anyone seem to consider Perl 5 as a compilation target?
tadzik because we wanted to have good threading
I suppose
shop & 09:24
sorear hoelzro: it's been done several times 09:32
nwc10 you won't get good threading, and your Unicode will be O(n) 09:43
or O(n**2) for Graphemes
hoelzro hmm
I suppose it's a suboptimal target
dalek : 8d70c4c | (Tobias Leich)++ | lib/Perl5/Grammar.nqp:
disable token print for now and add "use base"-noop
 10:11
: 519ed8a | (Tobias Leich)++ | STATUS.md:
status update
 10:13
FROGGS pnr: say "runtime!"; import ENOSUCHMODULE; 10:39
camelia niecza v24-49-gfa3b5ca: OUTPUT«===SORRY!===␤␤Action method statement_control:import not yet implemented at /tmp/3O1mZeqM4A line 1:␤------> say "runtime!"; import ENOSUCHMODULE⏏;␤␤Unhandled exception: Unable to resolve method statement_level in type A…
..pugs: OUTPUT«runtime!␤*** No such subroutine: "&import"␤ at /tmp/PnHS9iRHwz line 1, column 17-37␤»
..rakudo bfd850: OUTPUT«===SORRY!===␤Could not find module ENOSUCHMODULE to import symbols from␤at /tmp/2Lh45cRBvR:1␤------> say "runtime!"; import ENOSUCHMODULE⏏;␤ expecting any of:␤ postfix␤ statement end␤ statement modifier␤ …
FROGGS looks like import dies a bit too early 10:40
moritz import is compilr-time
jnthn What moritz said :) 10:42
afternoon, #perl6
FROGGS ahh, so `require` will do the importing? 10:45
r: require Test:file("Test.pm")
camelia rakudo bfd850: OUTPUT«Could not find file 'Test.pm' for module Test␤ in any load_module at src/Perl6/ModuleLoader.nqp:123␤ in block at /tmp/29GgEpH2d3:1␤␤»
moritz require doesn't import anything 10:46
r: require Test:file("lib/Test.pm6")
camelia rakudo bfd850: OUTPUT«Could not find file 'lib/Test.pm6' for module Test␤ in any load_module at src/Perl6/ModuleLoader.nqp:123␤ in block at /tmp/R3mr61vcgc:1␤␤»
FROGGS "Importing via require also installs names into the current lexical scope by default, but delays the actual binding till runtime:" 10:48
r: require Test:file("Test.pm6")
camelia rakudo bfd850: OUTPUT«Could not find file 'Test.pm6' for module Test␤ in any load_module at src/Perl6/ModuleLoader.nqp:123␤ in block at /tmp/2PnpCz2gAW:1␤␤»
FROGGS meh
jnthn You have to declare those symbols as part of the require, iirc. 10:49
FROGGS yeah, like require "/home/non/Sense.pm" <common @horse>; 10:49
hmmm, maybe more like: require Test::file('Test.pm6') <&plan>; or so 10:51
nr: BEGIN "test".say 11:06
camelia niecza v24-49-gfa3b5ca: OUTPUT«Unhandled exception: Excess arguments to Mu.say, used 1 of 2 positionals␤ at /home/p6eval/niecza/lib/CORE.setting line 1735 (List.gist @ 5) ␤ at <unknown> line 0 (ExitRunloop @ 0) ␤ at /tmp/glRrfVcr6N line 1 (ANON @ 2) ␤ at <unknown> line 0 (ExitRunloo…
..rakudo bfd850: OUTPUT«test␤»
FROGGS nr: BEGIN "test".slurp
camelia niecza v24-49-gfa3b5ca: OUTPUT«Unhandled exception: Unable to resolve method slurp in type Str␤ at /tmp/cNZ_PFrBb0 line 1 (ANON @ 2) ␤ at <unknown> line 0 (ExitRunloop @ 0) ␤ at /home/p6eval/niecza/src/NieczaBackendDotnet.pm6 line 75 (downcall @ 4) ␤ at /home/p6eval/niecza/src/Niecz…
..rakudo bfd850: OUTPUT«===SORRY!===␤Method 'slurp' not found for invocant of class 'Str'␤»
FROGGS nr: BEGIN "test".^methods.say 11:07
camelia niecza v24-49-gfa3b5ca: OUTPUT«Unhandled exception: Unable to resolve method methods in type ClassHOW␤ at /tmp/AtI0kYuVgs line 1 (ANON @ 3) ␤ at <unknown> line 0 (ExitRunloop @ 0) ␤ at /home/p6eval/niecza/src/NieczaBackendDotnet.pm6 line 75 (downcall @ 4) ␤ at /home/p6eval/niecza/sr…
..rakudo bfd850: OUTPUT«BUILD Int Num chomp chop substr pred succ match ords lines samecase samespace trim-leading trim-trailing trim words encode capitalize wordcase trans indent codes tclc path WHICH Bool Str DUMP ACCEPTS Numeric gist perl comb subst split␤»
FROGGS nr: "test".^methods.say
camelia niecza v24-49-gfa3b5ca: OUTPUT«Unhandled exception: Unable to resolve method methods in type ClassHOW␤ at /tmp/ueayeq4xHu line 1 (mainline @ 4) ␤ at /home/p6eval/niecza/lib/CORE.setting line 4327 (ANON @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 4328 (module-CORE @ 582) ␤ at…
..rakudo bfd850: OUTPUT«BUILD Int Num chomp chop substr pred succ match ords lines samecase samespace trim-leading trim-trailing trim words encode capitalize wordcase trans indent codes tclc path WHICH Bool Str DUMP ACCEPTS Numeric gist perl comb subst split␤»
FROGGS hmmm, so it would be sane to delay the check for subs until runtime... however this could work 11:10
(for v5) 11:11
moritz FROGGS: yes, in p5 subroutine lookup must be delayed (unless it can be found at compile time) 11:45
FROGGS so I basically have to move these checks from Grammar to an QAST::Op.new( :op('call'), :name('CHECK_ROUTINE') ) or something like that 11:49
and this one will croak if &routine is not found in $*PACKAGE.HOW 11:50
hmmm, doesn't sound impossible 11:51
moritz well, there's another slight gotcha 11:55
perl 5 only has our-scoped subs
FROGGS well, that is good actually, no?
moritz so maybe the most intuitive approach is to look for a lexical &yoursub first, in case it was provided by an outer Perl 6 scope 11:56
FROGGS hmmm, true
moritz and if that's the case, you emit a normal QAST::Op.new(:call, QAST::Var.new( :name<&yoursub>, :scope<lexical> )
FROGGS right 11:57
same like variables
moritz and if not , instead of the inner QAST::Var, you have a function that looks up the function in the symbol table at run time, and dies if it fails
FROGGS right, sounds good to me 11:58
dalek p/rak-jvm-support: 8bb6406 | jnthn++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java:
Delegation to .^type_check method.
 12:37
p/rak-jvm-support: f6f5788 | jnthn++ | src/vm/jvm/runtime/org/perl6/nqp/sixmodel/SerializationReader.java:
Don't create empty type caches on deserialization.

It confuses the .^type_check fallback.
kudo/jvm-support: b087a8a | jnthn++ | src/Perl6/ (2 files):
Missing null checks.
kudo/jvm-support: e1a3422 | jnthn++ | src/vm/jvm/Perl6/Ops.nqp:
Be sure we multi-dispatch on flattened args.
colomon \o/ 12:41
grondilu tadzik: isn't there a command to remove(uninstall) a module? 13:30
grondilu frgot to mention he's talking about Panda
tadzik grondilu: nope 13:36
panda can't do that yet
grondilu ok 13:37
tadzik but I want it to be able to do that :)
and GC-like ability to remove unnecessary deps too
grondilu GC?
tadzik garbage collection 13:38
grondilu k
tadzik panda knows if it has installed a module as a dependency or not
so later it'd be able to remove all dependencies which are not referenced by anything
grondilu well, even cpan doesn't provide with any command to remove a module, so I understand it's probably tough to implement
BinGOs cpanp does. 13:39
tadzik it's not that hard I think
panda just needs some packlist-like thing
BinGOs we're going to be moving away from packlists so please don't implement that broken idea. 13:40
tadzik ok :)
I was mostly looking for inspiration in linux package managers
some of them just keep one big file with mappings between module names and paths on the FS 13:41
BinGOs the main problem is that once a distribution (ie. a collection of modules) has been installed there is a disconnect between the modules and the distribution they came in. 13:42
tadzik I guess you have to decide, if you want to remove modules or distibutions 13:43
FROGGS moritz: 14:13
r: say INDIRECT_NAME_LOOKUP( PseudoStash.new, '&say' )
camelia rakudo bfd850: OUTPUT«sub say() { ... }␤»
FROGGS r: say INDIRECT_NAME_LOOKUP( PseudoStash.new, '&foo' )
camelia rakudo bfd850: OUTPUT«No such symbol '&foo'␤ in method gist at src/gen/CORE.setting:10022␤ in method gist at src/gen/CORE.setting:893␤ in sub say at src/gen/CORE.setting:10920␤ in block at /tmp/Ofw3qE4INd:1␤␤»
FROGGS \o/
spider-mario we haven’t had a * release in a while 14:20
has 2013.04 been skipped too?
moritz there's a release candidate at pmichaud.com/sandbox/rakudo-star-20...rc2.tar.gz 14:21
FROGGS spider-mario: there is no release yet because there are problems with some modules 14:22
spider-mario oh.
FROGGS might be easy to fix though
spider-mario thanks for both of your answers
colomon moritz: any interest in going further with the smoke review web page, or should I just start from where you left off? I've got seven days of smoke results now, I'm getting eager to have a slick web page to study them! ;) 14:31
moritz colomon: I'm caught up in other stuff at the moment (refactoring the irc logging system), please go ahead yourself 14:32
colomon moritz: roger. Of course, I don't know when I'll have time myself, may depend on how Henry is at entertaining himself today. 14:33
moritz sure thing 14:35
dalek p/rak-jvm-support: 56c139c | jnthn++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java:
Better error for null method lookup.
 14:42
p/rak-jvm-support: be074c1 | jnthn++ | src/vm/jvm/QAST/Compiler.nqp:
Fix mis-compile in takedispatcher.
kudo/jvm-support: e71d566 | jnthn++ | src/vm/jvm/runtime/org/perl6/rakudo/ (2 files):
Start to fill out bindOneParam a bit.
kudo/jvm-support: af5712b | jnthn++ | src/vm/jvm/runtime/org/perl6/rakudo/ (3 files):
Scalar binding.
[Coke] skims through an old Modula-3 book he has lying around. 15:12
rjbs stealing some more ideas for p6? 15:15
nwc10 "get it adopted as a teaching language" seems to be the best idea. 15:16
dalek p/rak-jvm-support: 32a753e | jnthn++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/ (2 files):
Decontainerize on findmethod.

Needs care to make sure indy guard clauses are aware of decont also.		 15:17
p/rak-jvm-support: d6f3554 | jnthn++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java:
Don't blow up on failed named lexical lookup.
kudo/jvm-support: 6b1c80d | jnthn++ | src/vm/jvm/runtime/org/perl6/rakudo/RakudoContainerSpec.java:
Scalar fetch.
kudo/jvm-support: eba83b5 | jnthn++ | src/vm/jvm/runtime/org/perl6/rakudo/Ops.java:
Implement nqp::p6decontrv.
FROGGS omfg, doing a indirect lookup for all subs is pretty slow 15:28
jnthn FROGGS: Are you constructing the PseudoStash every time? 15:29
jnthn Hm, though that shouldn't be too bad... 15:30
Indirect lookups are pretty, well, indirect though :)
FROGGS yes, every time
maybe it just *feels* slow right now
dalek kudo/jvm-support: fbf5100 | jnthn++ | src/vm/jvm/runtime/org/perl6/rakudo/Ops.java:
Stub p6typecheckrv.
 15:32
kudo/jvm-support: d03c7cc | jnthn++ | src/Perl6/World.nqp:
Fix an nqp::join that should be join.
kudo/jvm-support: 7ef33ff | jnthn++ | tools/build/Makefile-JVM.in:
Bring in the next 4 files of CORE.setting.

First 3 just compile already, final one shows up the next thing to fix.
jnthn FROGGS: No, it probably is slow.
nwc10 jnthn: oooh, how far how through the setting? 15:33
jnthn Up to line 399.
nwc10 cool
jnthn Though with some cheating :)
nwc10 p: say 339/13000
camelia pugs: OUTPUT«0.0260769230769230769230769230769230769231␤»
FROGGS every day +100 lines *g*
moritz nah, the first lines are very slow to compile
and the later will go much faster
... until you get to IO and Exception
FROGGS I know
jnthn Yeah. I mean, getting all the stuff in place so you can actually call a trait_mod and have it execute successfully is a good bit of work. 15:34
The fact we're doing that means Perl 6 multi-dispatch must be working reasonably well.
trait_mod:<is> weighs in at over 30 candidates and it's picking the right ones... 15:35
dalek : 81c8181 | (Tobias Leich)++ | / (3 files):
do checks for subs at runtime, => +57 passing tests!
 15:42
TimToady npr: say 339/13000 15:45
camelia pugs: OUTPUT«0.0260769230769230769230769230769230769231␤»
..rakudo bfd850, niecza v24-49-gfa3b5ca: OUTPUT«0.026077␤»
TimToady so is there a use case for false precision?
[Coke] npr: say ~(339/13000) 16:00
camelia pugs: OUTPUT«0.0260769230769230769230769230769230769231␤»
..rakudo bfd850, niecza v24-49-gfa3b5ca: OUTPUT«0.026077␤»
[Coke] (just making sure no .gist weirdness :P)
TimToady well, it's really the .base we installed that is picking the precision there, so 2/3 of our .bases belong to us 16:02
jnthn r: sub foo(:bar($baz)) { }; foo(:bar) 16:05
camelia rakudo bfd850: ( no output )
jnthn r: sub foo(:bar($baz)) { }; foo(:baz)
camelia rakudo bfd850: OUTPUT«Unexpected named parameter 'baz' passed␤ in sub foo at /tmp/hBaRlLTjv0:1␤ in block at /tmp/hBaRlLTjv0:1␤␤»
TimToady (and it picks 6 as a minimum precision, just so 1/7 and such are recognizable)
r: say 1/7
camelia rakudo bfd850: OUTPUT«0.142857␤»
TimToady but it picks more precision as the size of the denominator grows 16:06
say 339/130000000000
npr: say 339/130000000000
camelia pugs: OUTPUT«0.0000000026076923076923076923076923076923␤»
..rakudo bfd850, niecza v24-49-gfa3b5ca: OUTPUT«0.0000000026077␤»
TimToady r: say 22/7 16:07
camelia rakudo bfd850: OUTPUT«3.142857␤»
TimToady r: say 22/7000000000000000 16:08
camelia rakudo bfd850: OUTPUT«0.00000000000000314␤»
TimToady it could be argued that it should give a little more there
it could also be argued that it shouldn't :) 16:10
sorear: The problem is not so much people rejecting your ID card because it's Californian, but rejecting you because you're Californian. :) 16:15
dalek kudo/jvm-support: 5e2751c | jnthn++ | src/core/traits.pm:
Add some missing required markers.

Probably helps all trait applications to routines on whatever platform.		 16:17
dalek p/rak-jvm-support: 845307c | jnthn++ | src/vm/jvm/ (2 files):
Start to scatter some more decont calls.
 16:19
TimToady hoelzro: the main problem with targetting Perl 5 is that the (lack of a) type system actively fights you, and if you bypass all the built-in allomorphic types to do everything in P5-ish OO, it's slow, and when you start trying to emulate laziness and sane regex, it's even slower 16:25
hoelzro I see 16:26
TimToady also, consider that STD/viv, which *does* in fact target Perl 5, turns 6320 lines of Perl 6 into 66290 lines of Perl 5...
GlitchMr Perl 5 is already slow. Perl 6 running on Perl 5 would be even slower. 16:27
GlitchMr Even when you would compile Perl 6 code to Perl 5. 16:27
GlitchMr (but Inline::Perl6 module would be sort of fun :-)) 16:28
TimToady much of the pugs effort was for targetting Perl 5 (it's even on the T-shirt), and it's one of the reasons pugs didn't go as far as we hoped
GlitchMr Then again, Inline::Perl6 already exists. 16:29
It's called Inline::Pugs and Inline::Rakudo. 16:30
TimToady another data point was when we backported some p6 features such as switch into p5, and discovered they really didn't make a lot of sense without p6's type system 16:31
spider-mario maybe Inline::Perl6 could exist and be to Inline::{Pugs, Rakudo} as YAML::Any is to YAML::{XS, Tiny, etc.}
TimToady in a nutshell, P6 allows its users to be confused about its type system, but P5 allows itself to be confused about its type system 16:33
dalek p/rak-jvm-support: 25157f7 | jnthn++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java:
nqp::rebless needs to decont its args.
 16:38
spider-mario another problem with switch (more specifically `given`) in p5 is p5’s handling of lexical $_ 16:40
p6 makes it the parameter of the current block if it is used, which seems to work very well
TimToady yes, the legacy scoping is also an issue throughout 16:41
moritz: though you can have lexical subs in p5, you just have to write them: my $sub = sub {...};
dalek kudo/jvm-support: 3c2aad1 | jnthn++ | src/vm/jvm/ (2 files):
Implement nqp::p6captureouters.
kudo/jvm-support: 9382421 | jnthn++ | src/Perl6/ (2 files):
Fix a few type-os.
kudo/jvm-support: a962c59 | jnthn++ | tools/build/Makefile-JVM.in:
3 more files we now parse/build AST for.

Note that this doesn't mean we can actually compile them, just that the parser and actions survive. Some will blow up if we could get any further (blocked by mysteries for now, though) due to use pir:: and so forth (which parses, but obviously wouldn't actually compile).
TimToady and you have to call them funny :) 16:42
jnthn Make it to 1270 now, with above important caveats of course :)
nwc10_ oooh, 10% 16:42
TimToady semantics, how hard could it be? 16:42
jnthn Well, it's nice that it at least parses. To get this far it has to dispatch to and run dozens of trait mods and the odd BEGIN block, so it is managing to compile those routines that it needs during the compile. :) 16:44
TimToady wonders how much of the design of P6 was subconsciously driven by me having written a Java/P5 integration lo these many years ago 16:44
jnthn wonders how many other JVM languages do BEGIN-time stuff... :) 16:45
nwc10 "lo"?
is this a RTFM question yet? :-) I take it that NQP doesn't have BEGIN blocks. Correct? 16:46
jnthn nwc10: It does but very constrained.
nwc10: The BOOTSTRAP is actually mostly a huge BEGIN block.
nwc10 constrained that the BEGIN block isn't allowed to have side effects outside of the VM?
jnthn nwc10: It pieces together the object system using MOP calls in a BEGIN block, whacks all it did in EXPORT, and then leaves the serializer to save its work :)
nwc10: More that it has very limited visibility of the outer scopes. 16:47
TimToady darn it, someone drank all my coffee again!
jnthn NQP's BEGIN support exists for the sole purpose of letting us writing Rakudo's type system bootstrap :)
TimToady looks around, sees no other coffee drinkers in the house... 16:48
nwc10 did the culprit commit the crime yesterday, or this morning?
ie are you sufficiently bootstrapped to go out shopping?
jnthn: ah right. That makes a lot of sense. 16:49
jnthn 1853... 16:50
nwc10 14% 16:51
jnthn hm, seems "does" traits don't explode it. Good. 16:54
jnthn starts making a shopping list for dinner stuff while seeing how far he can get the parser to go before a file it explodes on... 16:55
TimToady nwc10: this morning; fortunately, there's more coffee in the pot, and after that, the cupboard, and after that, the freezer :) 16:56
dalek kudo/jvm-support: 1c87ba0 | jnthn++ | tools/build/Makefile-JVM.in:
Add more files, up to the next failure.
 17:03
jnthn Found the next explosion. 2098. :) 17:04
jnthn Shopping, curry cooking, etc... 17:04
&
grondilu nr: sub f(\a, \b) { a+b, a-b }; my ($a, $b) = 1, 2; say ($a, $b) .= &f;
camelia niecza v24-49-gfa3b5ca: OUTPUT«Use of uninitialized value in string context␤ at /home/p6eval/niecza/boot/lib/CORE.setting line 1295 (warn @ 5) ␤ at /home/p6eval/niecza/boot/lib/CORE.setting line 266 (Mu.Str @ 15) ␤ at /home/p6eval/niecza/src/STD.pm6 line 3857 (ANON @ 3) ␤ at /home/p… 17:05
..rakudo bfd850: OUTPUT«===SORRY!===␤Undeclared routine:␤ a-b used at line 1␤␤»
TimToady curry explosions, yum!
grondilu nr: sub f($a, $b) { $a+$b, $a-$b }; my ($a, $b) = 1, 2; say ($a, $b) .= &f;
camelia niecza v24-49-gfa3b5ca: OUTPUT«Use of uninitialized value in string context␤ at /home/p6eval/niecza/boot/lib/CORE.setting line 1295 (warn @ 5) ␤ at /home/p6eval/niecza/boot/lib/CORE.setting line 266 (Mu.Str @ 15) ␤ at /home/p6eval/niecza/src/STD.pm6 line 3857 (ANON @ 3) ␤ at /home/p…
..rakudo bfd850: OUTPUT«Not enough positional parameters passed; got 1 but expected 2␤ in sub f at /tmp/zOW1s5n6Cu:1␤ in method dispatch:<.=> at src/gen/CORE.setting:1057␤ in block at /tmp/zOW1s5n6Cu:1␤␤»
grondilu nr: sub f($a, $b) { $a+$b, $a-$b }; say (1, 2).&f; 17:07
camelia niecza v24-49-gfa3b5ca: OUTPUT«Unhandled exception: No value for parameter '$b' in 'f'␤ at /tmp/u_yX48bRYT line 0 (f @ 1) ␤ at /tmp/u_yX48bRYT line 1 (mainline @ 4) ␤ at /home/p6eval/niecza/lib/CORE.setting line 4327 (ANON @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 4328 (mo…
..rakudo bfd850: OUTPUT«Not enough positional parameters passed; got 1 but expected 2␤ in sub f at /tmp/Hjkw2ByEJn:1␤ in block at /tmp/Hjkw2ByEJn:1␤␤»
grondilu nr: sub f(*[$a, $b]) { $a+$b, $a-$b }; say (1, 2).&f;
camelia rakudo bfd850, niecza v24-49-gfa3b5ca: OUTPUT«3 -1␤»
TimToady wonders if it would be useful or confusing (or both) to allow a-b to parse as a - b when both a and b are declared but a-b isn't... 17:11
geekosaur eeeeew 17:12
grondilu nr: say ((state $ )//0 + 1) xx 4
camelia rakudo bfd850, niecza v24-49-gfa3b5ca: OUTPUT«1 1 1 1␤»
nwc10 I think that it would be confusing, if only because you then wouldn't be able to look at that line of code and know what it did 17:13
TimToady nr: say (state $)++ xx 4
camelia rakudo bfd850, niecza v24-49-gfa3b5ca: OUTPUT«0 1 2 3␤»
TimToady grondilu: is that what you were looking for?
nwc10: but in context, it's usually quite obvious, because such programs are also full of a-1, a+b, etc 17:14
colomon TimToady: (re a-b being a - b) Just say no.
rn: say (for 1..10 { $_ if .is-prime; }).perl
camelia niecza v24-49-gfa3b5ca: OUTPUT«(Empty, 2, 3, Empty, 5, Empty, 7, Empty, Empty, Empty).list␤»
..rakudo bfd850: OUTPUT«(2, 3, 5, 7).list␤»
colomon TimToady: is Niecza in some sense right here, or should Nil go away before it can be seen in this sort of case? 17:15
colomon still trying to figure out how to fix niecza.
TimToady well, I don't think .perl should force list context, but maybe a list comprehension should do so automatically 17:16
Empty is supposed to just mean () 17:17
basically, it's the old definition of Nil
grondilu TimToady: I just wanted to see if niecza would give the same result as rakudo
TimToady and Nil itself isn't quite de-listified yet either 17:18
nodnod
n: say (for 1..10 { .is-prime ?? $_ !! Nil }).perl
camelia niecza v24-49-gfa3b5ca: OUTPUT«(2, 3, 5, 7).list␤»
TimToady I don't know what's removing Nil there 17:19
colomon Oh! so if "if" returned Nil instead of Empty, it would work?
TimToady well, yes, that's what it used to return, but spec is now that it returns () in that case 17:20
because we want Nil to represent the most minimal kind of failure, without disappearing like () in lists
colomon .... 17:21
nwc10 TimToady: it might work if there was an option to swtich the grammar between either "a-b is a() - b()" or "a-b is a-b()", and so the other interpretation is disallowed within that lexical scope.
TimToady so that Nil can be one of the distinguished values that indicate success/failure to smartmatching
colomon TimToady: so it should return () (or maybe Empty if that's the same thing) but () / Empty should disappear and Nil shouldn't?
TimToady right, so it's a bit backwards there for some reason 17:22
colomon like niecza has the empty / nil thing backwards... yes
TimToady the problem I ran into was that the regex engine uses Nil internally to indicate "no more cursors", and it needs to be changed to use () or Empty 17:23
nwc10: I see it more as a longest-token thing, recognize a-b as a single token if it can, otherwise recognize it as 3, with the onus on the programmer to install whitespace if a-b is ambiguous (which it almost never is) 17:29
dalek ast: 6d79206 | (Solomon Foster)++ | S32-container/stringify.t:
Actually test results of type object .Str.
 17:32
TimToady a-b is just one of the thorny issues with the sigilless style, of course; all the "choose your own quote" keywords get in the way as well 17:38
which RC entries mostly solve by using Unicode, but that's kind of ignoring the problem
colomon is not fond of sigilless style 17:39
TimToady it's much better looking for math stuff, I think 17:39
TimToady but simply telling the mathfolk that they can't use m or s or q as variable names is a bit antisocial, if there's some way to do the right thing for them 17:40
colomon TimToady: it's maybe a bit better looking for math stuff ... but it's also ignoring all the reasons sigils are great. 17:41
TimToady so we should force that on everyone, eh? 17:41
lizmat hello #perl6!
lizmat r: sub not-so { False }; say not-so() 17:41
camelia rakudo bfd850: OUTPUT«True␤» 17:42
lizmat and we need to fix these issues with "-"as well :-)
colomon TimToady: I'm sorely tempted to say yes.
timotimo_ r: say so(); 17:44
camelia rakudo bfd850: OUTPUT«False␤»
TimToady some of this is also symptomatic of bifurcating the processing of identifiers into keyword tokens vs user-defined identifiers with different scoping rules
colomon I like more than one way to do it; but then I don't like making the parser and language more complicated to allow something I'm guessing most style guides are going to end up telling you never to use.
moritz it's a bit weird to have a sigil-ful language support sigilless style 17:45
after all we have *reasons* to use sigils
and then not using them after all is a bit like admitting we are/were wrong all the time :-)
TimToady it's also the reason that drives a large portion of the population away screaming in horror
I'd rather have a story that says, "Yes you can, but here are the tradeoffs: if you define \m, you can't say m//" 17:46
colomon TimToady: yes, but is letting them skip the sigils when all the texts, documentation, and standard libraries use sigils really going to make them happy?
TimToady if you define \s, you can't say s///
moritz but providing an option for sigilless style isn't going to attract that part of the population, because they know that when they read other people's code, they'll see sigils and want to run away screaming in horror again 17:47
moritz you really can't have it both ways *and* sanity 17:47
TimToady well, but that's always the case, even with sigils
most people run away screaming from most people's code most of the time anyway 17:48
colomon moritz and I mean run away screaming from the language itself
raiph there is rationale for noun/verb markers; rationale for using symbols, including as markers; but depends on coder and domain 17:49
raiph and a coders' point on the learning curve 17:50
timotimo_ since the GPN is fast approaching (end of this month), i don't think i can actually hold a proper talk; maybe i'll come up with a lightning talk instead.
timotimo_ for instance, at the congress one of the hacker jeopardy topics was "perl operators" (including eskimo kiss, spaceship, goatse, ...) so maybe i'll point out how perl6 operators and such are a bit nicer? (due to enforced spacing sometimes and due to other things at other times?) 17:51
nwc10 moritz: it depends. It may well (a) attract enough of the the NIH crowd to be useful (b) may well be enough to get some of the doubters to try (c) might get people started who then change their mind (d) it is a marketing tickybox. And sadly those matter, even if no-one uses the feature
but yes, it has a cost in complexity and sanity
trade off. :-(
nwc10 the NIH crowd being the ones who roll their own CGI parameter parsing etc, and copy-paste it into each new project 17:52
TimToady well, the real insanity is that we have single letter keywords that are fighting with the mathematician's need for simple variable names 17:53
if this can be dealt with sanely by invoking LTM and predeclaration, why not?
colomon like i
TimToady we already have those concepts, it would just be tweaking the application slightly when people actually declare \m thingies 17:54
nwc10 TimToady: oh, you're specifically meaning one character variables when you're talking about a-b
not the general case of longer variables?
TimToady can argue it both ways, but I'd prefer not to make an exception
nwc10 or I'm missing the point that the crunch will likely be on the single letters?
colomon what about the case of $a-b where \b has been declared? 17:55
TimToady the point is that keywords fight for namespace, and they currently fight a bit unfairly
colomon: it would only try that if $a-b wasn't declared
right now identifiers can only take an all-or-nothing approach under LTM 17:56
TimToady under the rules I'm conjecturing, it might Just Work nearly all the time, and still catch 99% of typos 17:58
it's probably something we should just prototype and try out for a while to see if it causes difficulties 18:00
nwc10 yes, that does feel like a good way forwards 18:01
TimToady and let me be clear: my difficulty here is not actually the a-b issue, or math notation, but the fundamental extensibility of a language with a slightly insane keyword policy, which is what we have at the moment 18:02
woolfy1 exit 18:45
oh wrong window
hmmm why do I see only 20 people in this room? Liz is in the room too, she says, but I don't see her in the list. Split? 18:46
moritz TimToady: it's _ilbot 18:50
TimToady right
moritz if somebody wants to contribute some CSS + JS to the IRC logs: it would be nice to have a bit of javascript that collapses multiple join/leave lines into one, with some sort of indicate that they are collopsed, and an option to uncollopse them again 18:51
jnthn Well, that's one way to discover the appartment's smoke alarm still works... 19:27
TimToady hopefully that's one of the better ways
gfldex moritz: that would be for irclog.perlgeek.de? 19:27
moritz gfldex: correct, and sources at github.com/moritz/ilbot 19:28
jnthn Decided to have a crack at doing onion bahjis. The results were OK for a first attempt, but apparently the frying passed the magical smoke threshold... :)
nwc10 jnthn: oh, they do smoke alarms in Sweden? Austria appears not to have heard of them. 19:28
jnthn nwc10: Yeah, and I never set mine off while cooking before. 19:29
nwc10 has :-)
usually with a frying pan
jnthn yeah...that was it :)
Hm, my appartment smells like an Indian restaurant now... :)
nwc10 is that a bug or a feature? 19:30
jnthn Well, the food that produced it was certainly a feature... :)
TimToady being anti-smoke is too left-wing for places like Austria :) 19:31
moritz ... assuming you can still eat it, and don't have to phone the fire department first :-)
speaking of IRC logs -- is anybody using the text view? 19:33
diakopter there's a text view?
moritz I dimly recall that I added it some time ago for somebody who's mobile phone couldn't handle the full HTML
diakopter that'd be cool for other bots/sites to parse
moritz diakopter: irclog.perlgeek.de/text.pl?channel=...2013-05-11
diakopter: JSON would be much cooler for that :-) 19:34
diakopter text.pl needs a charset/encoding header
moritz anyway, mobile phones have advanced quite a bit in the last years 19:35
and since I'm refactoring the whole thing, I would like to remove unused features
diakopter your site still breaks on built-in safari on iphone for long pages
like, freezes & crashes safari
but iphone chrome handles it fine 19:36
moritz diakopter: submit bug report (for iphone safari) :-) 19:36
tadzik moritz: I do sometimes 19:36
moritz ok, then I'll keep it
diakopter I'll use it now that I know about it 19:37
tadzik I'm sometimes using it on a mobile when I'm crazy about data usage
diakopter would download lots more quickly
on mobile when on slow networks
moritz: feature request (which I'm happy to do if you approve): auto-extending updates to the page with ajaxy thingies 19:38
reloading on iphone every 30 seconds or so is really annoying
moritz diakopter: you mean on the "today" page?
diakopter yup
when my ssh app isn't working or I want to be able to click links on the log on mobile, I use chrome/html 19:39
moritz diakopter: I approve (but only in the branch 'nextgen', or after that branch has landed)
diakopter and it takes 5-10s to reload usually
on long pages
moritz you'll like that branch better anyway, it offers deployment via Plack 19:40
so you just says 'plackup ilbot.psgi', and have a local dev server
diakopter (note: by happy to do it, I mean in several weeks. ;) 19:41
so feel free to beat me to it if you're bored :D
moritz after the refactoring I'll tackle the transition to postgres
and some other stuff
so, not very likely
but I'll open a TODO ticket anyway 19:42
diakopter does the outputted page on 'today' ever change except for what's added? 19:42
moritz well, summary stuff changes 19:43
but that can be ignored for 'today', IMHO 19:44
diakopter I'd simply use the webirc thing in chrome, except it doesn't create hyperlinks and such 19:45
last I checked
moritz ok, one more question 19:46
currently text.pl actually produces HTML, with a <pre> tag and then text inside 19:47
should it stay that way, or produce proper text/plain?
nwc10 proper text/plain would be lower bandwidth, wouldn't it, as nothing would need escaping? 19:49
fractionally lower
moritz sure, but back in the days[tm] some mobile phone browsers then offered it for download, instead of displaying it
nwc10 aha.
moritz I'm all for text/plain myself 19:50
and I don't own a mobile phone that is connected to the internez, so I don't care either way :-)
diakopter I say text/plain is fine, just test a few until you get the optimum charset/encoding header(s)
er, coordinate with others to test ;) 19:51
moritz diakopter: you can test at irclog.perlgeek.de:5000/perl6/2013-05-11/text if you want :-) 20:04
raiph what do folk think of (asking for) a camelia realm on questhub.io ? 20:05
diakopter seems cool to me. I can link to it from camelia.org.. (for whatever that's worth.. ;() 20:06
dalek kudo/jvm-support: 2745850 | jnthn++ | src/vm/jvm/runtime/org/perl6/rakudo/Binder.java:
Bind self for methods.
 20:07
kudo/jvm-support: 202a1e5 | jnthn++ | src/vm/jvm/runtime/org/perl6/rakudo/Ops.java:
Implement nqp::p6bindassert.
diakopter moritz: seems to hang 20:09
moritz diakopter: try again please 20:10
(now plackup started in screen :-)
diakopter not marked as utf8? 20:11
(or is it?)
dagurval r: "123" ~~ /[0-9]+/; 20:12
camelia rakudo b1695c: OUTPUT«===SORRY!===␤Unrecognized regex metacharacter - (must be quoted to match literally)␤at /tmp/9Pm8AaOVZR:1␤------> "123" ~~ /[0⏏-9]+/;␤Unrecognized regex metacharacter [ (must be quoted to match literally)␤at /tmp/9Pm8AaOVZR:1␤---…
dagurval should I use .. instead? 20:13
moritz Content-Type: text/plain; charset=UTF-8
dagurval: '123' ~~ /<[0..9]>+/
diakopter moritz: ah, I see that now. hm, I guess those irc format codes are just like that 20:14
moritz std: /[0-9]+/
camelia std b33d8e0: OUTPUT«===SORRY!===␤Invalid regex metacharacter (must be quoted to match literally) at /tmp/KUz4ihjKFs line 1:␤------> /[0-⏏9]+/␤Potential difficulties:␤ [0-9] appears to be an old-school character class; digits should be matched with \…
moritz huh, I thought it would recognize old-style character classes, and warn about them
dalek kudo/jvm-support: 12528db | jnthn++ | tools/build/Makefile-JVM.in:
Add 10 more files to the setting build.
 20:15
dagurval what does < > imply?
jnthn r: /[0-9]+/
camelia rakudo b1695c: OUTPUT«===SORRY!===␤Unrecognized regex metacharacter - (must be quoted to match literally)␤at /tmp/9iX_zncds5:1␤------> /[0⏏-9]+/␤Unrecognized regex metacharacter [ (must be quoted to match literally)␤at /tmp/9iX_zncds5:1␤------> …
jnthn std: /[a-z]+/
camelia std b33d8e0: OUTPUT«===SORRY!===␤Invalid regex metacharacter (must be quoted to match literally) at /tmp/3N9KVQpuoY line 1:␤------> /[a-⏏z]+/␤Potential difficulties:␤ [a-z] appears to be an old-school character class; please use <[a..z]> if you mean…
diakopter TimToady: re a-b, I mean, it's just like backtracking in general into tokens,... but that wouldn't work for post-declared subs..?
moritz dagurval: <[...]> is a character class 20:16
dagurval: see perlcabal.org/syn/S05.html 20:17
dagurval moritz: ok, thanks!
dalek kudo/nom: 06eddc0 | coke++ | t/spectest.data:
Run sink test
 20:18
dalek p/rak-jvm-support: 9466f56 | jnthn++ | src/vm/jvm/QAST/Compiler.nqp:
First cut at compiling nqp::chain.
 20:53
jnthn nqp: nqp::say(nqp::substr('abc', -1)) 20:57
camelia nqp: OUTPUT«c␤»
dalek p/rak-jvm-support: d31030b | jnthn++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java:
-1 case of nqp::substr.
 21:00
FROGGS_ nqp: nqp::say(nqp::substr('abc', -4)) 21:03
camelia nqp: OUTPUT«Cannot take substr outside string␤current instr.: '' pc 40 ((file unknown):139521819) (/tmp/t5J3bolEQQ:7)␤»
lizmat jnthn: in Grammar.nqp, rule package_def, I see a :my $*DOCEE, followed by a <.attach_docs> subrule. I assume that $*DOCEE is visible inside the subrule?
jnthn lizmat: Yes 21:04
lizmat sorry, token <.attach_docs>
jnthn lizmat: subrule calls are just method calls, so it's the usual $*FOO rules of dynamic scoping.
lizmat odd thing is, everything compiles ok and even spectests ok if I remove all :my $DOCEE lines 21:05
( I was assuming they were superfluous for a while)
jnthn Oh...hmm
Weird, I'd ahve expected pod tests to fail 21:06
lizmat ok, then I'll check in that area...
jnthn tadzik++ knows the doc stuff best and may have some idea :)
tadzik I have ideas? :) 21:09
colomon TimToady: it looks to me like the problem is Niecza compiles itself using the "boot" version. But the boot version hasn't been updated since the last time sorear did it. Since you added "Empty", it boot version doesn't know about it.
TimToady: I don't know how to fix that, but at least I know where to look for clues now. :)
jnthn tadzik: lizmat suspects some dead code in docs related stuff
lizmat well, maybe, I don't think so anymore
jnthn :)
lizmat just that I didn't realize that my $*foo has dynamic scope 21:10
it sort of clashes with "my" in my head 21:11
sorear good * #perl6
jnthn o/ sorear 21:17
colomon oooo, sorear! 21:18
sorear indeed I am. 21:19
dalek p/rak-jvm-support: d348d72 | jnthn++ | src/vm/jvm/runtime/org/perl6/nqp/runtime/Ops.java:
Some more decont-ing.
 21:27
kudo/jvm-support: cc5ba85 | jnthn++ | src/vm/jvm/runtime/org/perl6/rakudo/Ops.java:
nqp::p6capturelex.
kudo/jvm-support: a633aef | jnthn++ | src/Perl6/Metamodel/EnumHOW.nqp:
Don't publish Parrot v-tables on JVM in EnumHOW.
grondilu nr: say set(<foo bar>) and set(<foo>) 21:34
camelia rakudo 06eddc, niecza v24-49-gfa3b5ca: OUTPUT«set(foo, bar)␤»
grondilu nr: say so set(<foo bar>)
camelia rakudo 06eddc, niecza v24-49-gfa3b5ca: OUTPUT«True␤»
grondilu nr: say set(<foo bar>) && set(<foo>)
camelia rakudo 06eddc, niecza v24-49-gfa3b5ca: OUTPUT«set(foo)␤» 21:35
dalek kudo/jvm-support: 932d4b7 | jnthn++ | tools/build/Makefile-JVM.in:
Add 6 more files to the setting build.
grondilu nr: say so set()
camelia rakudo 06eddc, niecza v24-49-gfa3b5ca: OUTPUT«False␤»
dalek p/rak-jvm-support: 0693eea | jnthn++ | src/vm/jvm/runtime/org/perl6/nqp/sixmodel/reprs/P6Opaque (2 files):
P6opaque container attr auto-viv.
 21:48
: 4f8af7e | (Tobias Leich)++ | / (4 files):
adhoc-declared vars get their container
 21:58
jnthn Past the 7000 line mark. 22:01
jnthn (With same caveats as earlier... :)) 22:02
lizmat_ cool!
lizmat_ jnthn: I'm looking at dissect_longname in Actions.nqp 22:03
and find that it is being called for "class" etc, but not for "use" 22:04
lizmat_ would these not need to be handled similarly? 22:04
jnthn lizmat_: I don't thing of them as being especially similar. 22:05
lizmat_: use parses a module_name iirc 22:06
lizmat_: It may be possible to handle it through dissect_longname, but I'm not immediately sure if it helps anything or not.
ah, it does <module_name>, but <module_name> uses <longname>. So yeah, it's do-able. 22:07
lizmat_ ok, then I'll sleep over that tonight and will bother you with some more questions tomorrow :-)
jnthn lizmat_: I guess it may let you write stuff like constant foo = 'bar'; use ::(foo); or so 22:09
lizmat_ r: use Test 22:16
camelia rakudo 06eddc: ( no output )
dalek kudo/jvm-support: ee0463a | jnthn++ | src/core/ (2 files):
Some pir:: -> nqp::.
kudo/jvm-support: 892f0d7 | jnthn++ | src/core/ (8 files):
nqp::concat_s => nqp::concat

They are aliases these days anyway.
kudo/jvm-support: 52b7a40 | jnthn++ | src/vm/jvm/runtime/org/perl6/rakudo/Ops.java:
Implement nqp::p6var.
rakudo/jvm-support: 175286d | jnthn++ | tools/build/Makefile-JVM.in:
rakudo/jvm-support: Add a bunch more files to the build.
rakudo/jvm-support: 22:17
lizmat_ r: use Foo:name<Test>
camelia rakudo 06eddc: OUTPUT«===SORRY!===␤Could not find Foo:name<Test> in any of: /home/p6eval/nom-inst/lib/parrot/5.2.0-devel/languages/perl6/site/lib, /home/p6eval/nom-inst/lib/parrot/5.2.0-devel/languages/perl6/vendor/lib, /home/p6eval/nom-inst/lib/parrot/5.2.0-devel/languages/perl6/lib, /…
jnthn Hmm :)
lizmat_ yes, indeed… plenty of work to do here 22:18
tomrrow
gnight jnthn, #perl6!
jnthn 'night, lizmat_ 22:19
dalek rl6-roast-data: bed4414 | coke++ | / (4 files):
today (automated commit)
 22:29
: f811524 | (Tobias Leich)++ | / (2 files):
added "use bytes" todo/noop
 22:33
TimToady rosettacode.org/wiki/Playfair_cipher 22:39
lue thought for a second TimToady created a new task just to solve another one, but then checked the edit history 22:42
TimToady yeah, someone else added the task without a solution :)
itz Can't open perl script "toolsbuildnqp-jvm-rr.pl": No such file or directory
make: *** [blib/Perl6/BOOTSTRAP.class] Error 2
hmmm did I go something wrong? 22:43
s/go/do
jnthn itz: Can you give me a little more context? 22:44
itz: Guess you're trying to build the jvm-support branch in Rakudo?
itz yes with nqp 22:46
itz oops 22:46
itz nqp rak-jvm-support I mean 22:46
jnthn Yeah, that's the right combo. 22:46
What platform?
itz 64bit linux 22:47
jnthn Oh...and I did something outstandlingly silly :)
jnthn *outstandingly, even 22:47
dalek kudo/jvm-support: e80fa2e | jnthn++ | src/vm/jvm/runtime/org/perl6/rakudo/ (2 files):
nqp::p6store and basic scalar store stuff.
 22:48
kudo/jvm-support: 1556388 | jnthn++ | src/core/operators.pm:
Another pir:: -> nqp::.
kudo/jvm-support: f3879cd | jnthn++ | src/vm/jvm/runtime/org/perl6/rakudo/Ops.java:
nqp::p6box_[sin]
kudo/jvm-support: 379467d | jnthn++ | tools/build/Makefile-JVM.in:
Fix slash-o; hopefully fixes build on Linux.

d42d0ea | jnthn++ | tools/build/Makefile-JVM.in: Uncomment all but IO bits of setting.
Means we now manage to parse and run BEGIN/traits/constant in 12850 lines worth of the core setting.
jnthn Aww 22:49
jnthn The second line of that commit message read: 22:49
Reported by itz++.
:)
itz :) I'll try again 22:49
FROGGS jnthn: backslashes are evil :o) 22:50
jnthn FROGGS: Yeah...was probably just a brain-o. 22:51
FROGGS good that windows takes slashes too 22:51
jnthn Sometimes ;) 22:52
FROGGS why only sometimes?
jnthn ./C:\consulting\rakudo>./install-jvm/nqp 22:53
'.' is not recognized as an internal or external command,
operable program or batch file.
jnthn :) 22:53
Doesn't there, for example
FROGGS ahh, okay, yeah
FROGGS but / does always work as a path sep, cd / and friends do not work though IMO 22:54
jnthn I'm not sure it's a matter of opinion :P 22:55
FROGGS '.' on unix is not a command or program either, it is just the case that cwd is not considered when executing commands, so you have to supply the path to an executable to run it 22:56
FROGGS jnthn: IMO is what I'd say in german... to say that I'm not sure about something 22:56
jnthn FROGGS: ah :) 22:57
jnthn 15:33 <FROGGS> every day +100 lines *g* 22:59
Just passed 11500 :P
FROGGS I've seen that but then thought it must mean something else :P 23:00
jnthn++
so, I can install it tomorrow? 23:01
FROGGS guesses that jnthn is already writing the final blog post
jnthn Oh, this is just parsing it 23:02
Doesn't mean we get beyond parsing :)
FROGGS I hope that I can gain some speed too when p5's require is working
ahh, okay
jnthn What'll happen after this is we hit all the places the setting does pir::, Q:PIR and so forth. 23:03
FROGGS sounds like not-so-fun work
jnthn Yeah.
Drawn out, I suspect
And of course, the binder needs finishing.
Thank heavens I ported the multi-dispatcher to NQP already, so that Just Works... 23:04
itz stupid question.. I have a jar and class file .. how do I call them to do "say 1+1" or whatever? 23:05
FROGGS yeah, proven by hundreds of tests
jnthn itz: You don't yet...at least, not the Perl 6 one. I'm still working madly on getting the built-ins building (included stuff like +) 23:05
itz: install-jvm/nqp -e "say(1 + 1)" works, though 23:06
itz ok ty I'll play with nqp
FROGGS r: EXPORT_SYMBOL() 23:13
camelia rakudo 06eddc: OUTPUT«===SORRY!===␤CHECK FAILED:␤Calling 'EXPORT_SYMBOL' will never work with no arguments (lines 1, 1)␤ Expected: :(\exp_name, @tags, Mu \sym)␤»
raiph jnthn++ # advancing parse of core setting at a tad more than 100 lines per day... 23:24
FROGGS r: class C { sub a { say 42 }; EXPORT_SYMBOL('&a', ['ALL', 'DEFAULT'], &a) }; 23:26
sneilan hi, there are humans that use perl? perl is amazing & i've just discovered it after using python for years. I hope that people still use it. 23:55
python has gotten soooo boring. 23:56
TimToady no humans here, we're all subhuman :)
sneilan robots?
TimToady that would be telling
FROGGS I sometimes feel that perl is using me :o)
sneilan lol :) 23:57

Powered by Cro and the Raku® Programming Language.

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

Thank you!