»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.
00:01 gdey joined 00:02 thou left, thou_ joined 00:03 FROGGS left, bonsaikitten is now known as DrEeevil 00:11 Gruber joined
Teratogen r: 1 while fork 00:11
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Undeclared routine:␤ fork used at line 1␤␤»
Teratogen drat
00:13 kbenson1 joined, kbenson left 00:16 colomon joined
Teratogen how do I write a fork bomb in perl 6? 00:19
00:19 jeffreykegler joined 00:25 ajr_ left
timotimo very carefully 00:26
Teratogen you can't do 1 while fork in perl 6? 00:29
what kind of language is this!?
timotimo ;)
you can Inline::C fork(); 00:30
or NativeCall fork
parrot may explode
00:32 btyler left
Teratogen I thought fork would stil be a builtin function 00:34
does perl 6 HAVE builtin functions?
built in
excuse me 00:35
timotimo r: say "hello"; # do you consider this "built in"?
camelia rakudo b2072f: OUTPUT«hello␤»
Teratogen r: say('heya');
camelia rakudo b2072f: OUTPUT«heya␤»
Teratogen I guess
timotimo *technically* it's in the setting, which is written in perl6 itself, so you *might* consider it not built in
r: nqp::say("testing")
camelia rakudo b2072f: OUTPUT«testing␤»
timotimo this, however, is totally built in
Teratogen r: say sin(.6); 00:36
camelia rakudo b2072f: OUTPUT«0.564642473395035␤»
Teratogen thank god, it works!
00:36 raiph left 00:37 raiph joined 00:38 jeffreykegler left
Teratogen r: say atan2(1,2); 00:39
camelia rakudo b2072f: OUTPUT«0.463647609000806␤»
Teratogen r: say ln(e); 00:40
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Undeclared routine:␤ ln used at line 1. Did you mean '&lc'?␤␤»
Teratogen r: say log(e);
camelia rakudo b2072f: OUTPUT«1␤»
Teratogen r: say e;
camelia rakudo b2072f: OUTPUT«2.71828182845905␤»
Teratogen cool
r: say 1+3i 00:41
camelia rakudo b2072f: OUTPUT«1+3i␤»
Teratogen very very nice 00:42
r: say pi; 00:43
camelia rakudo b2072f: OUTPUT«3.14159265358979␤»
Teratogen r: say phi;
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Undeclared routine:␤ phi used at line 1␤␤»
Teratogen hmmm
timotimo i'm going to bed 00:44
Teratogen does the golden ratio have a constant?
timotimo have fun with p6 :)
Teratogen sure thanks
00:52 benabik joined
RedditAnalytics r: say pi**e 00:54
camelia rakudo b2072f: OUTPUT«22.4591577183611␤»
00:55 sqirrel joined, tgt left 00:57 tgt joined 00:59 sqirrel left
[Coke] r: say 1, e, pi...* 01:02
camelia rakudo b2072f: OUTPUT«Unable to deduce sequence␤ in method sink at src/gen/CORE.setting:10169␤ in block at src/gen/CORE.setting:13701␤ in any coro at src/gen/CORE.setting:6040␤ in method reify at src/gen/CORE.setting:6022␤ in method reify at src/gen/CORE.setting:5761␤ in method r…
01:02 anuby joined
Teratogen r: say e**(pi*i) 01:04
camelia rakudo b2072f: OUTPUT«-1+1.22460635382238e-16i␤»
Teratogen WHAT!
shouldn't that be -1 exactly?
geekosaur not even p6 can fix floating point 01:05
01:10 tgt left 01:39 colomon left
sorear good * #perl6 01:42
01:42 colomon joined
sorear o/ 01:44
colomon \o
01:52 colomon left 01:53 colomon joined 01:56 colomon left, grondilu left 02:00 skids left 02:08 colomon joined 02:11 btyler joined 02:20 lustlife joined 02:23 lizmat joined
colomon \o 02:31
\o\o\o\o
lizmat o/
sorear /o/o/o
02:33 _jaldhar joined, colomon left 02:35 skids joined 02:50 _jaldhar left 02:53 _jaldhar joined 03:08 konundra left 03:17 logie joined 03:18 preflex left 03:19 preflex_ joined, ChanServ sets mode: +v preflex_, preflex_ is now known as preflex 03:25 labster joined
sorear o/ labster 03:30
labster \o sorear 03:31
just made it back to the dorms
03:32 lustlife left 03:35 konundra joined 03:37 logie left
labster aww, no one fixed nqp yet 03:38
I guess it's still a busy night out there, but I came back. I feel like quiet right now. Probably going to write a blog. 03:39
lizmat everybody is still in the restaurant, afaik
labster I guess I have to give a talk next year to get in with the cool kids.
sorear labster: nqp is broken? 03:41
labster yep. doesn't compile on my machine; complains about lack of ICU 03:42
03:44 FROGGS joined
FROGGS [Coke]: you could try --ll-exception 03:44
eternaleye Teratogen: You know, every time you confuse camelia, I half expect someone to jump in with "What are you doing to my baby?!" :P 03:46
labster well, it's correct in that I don't have ICU installed, but it's not a requirement.
eternaleye Teratogen: (Warning: my puns may be considered a sanity hazard)
03:48 adu joined 03:49 odoacre_ joined 03:50 atroxaper joined
lizmat labster: that would only mean a few more ICU tests skipped 03:53
labster not running tests though; nqp fails on turning src/stage2/gen/NQPP5QRegex.nqp to pir 03:54
so compile fails 03:55
03:55 Chillance left 04:17 lizmat left 04:26 Psyche^ joined
masak FROGGS: if you're interested: strangelyconsistent.org/blog/the-thing-with-nim 04:28
04:29 Patterner left, Psyche^ is now known as Patterner
FROGGS reads it 04:30
04:30 btyler left
FROGGS "removing a positive number of stones" :P 04:30
masak that's funny? :) 04:31
labster masak should have tried removing a negative amount of stones with people who didn't know the rules yet
lue "Oh dear, it seems I might have no way to win the game normally. I think I'll take a few negative stones to prevent that." 04:32
TimToady note also the possibility of contributing to rosettacode.org/wiki/Game_of_Nim, though that's a draft task that should probably have the "golf" part of the task description excised.
FROGGS masak: it is of course accurate, but it is funny to me 04:33
lue I agree with the talk page for that RC task. Also, "as short as possible" is very subjective, at least to me. 04:34
masak FROGGS: I see :) I might be more of a mathematician than I realize.
TimToady I note that the racket example doesn't try to golf
FROGGS I guess I'm going to bed now, otherwise I accidentially hit some keys with my forehead 04:35
TimToady o/
FROGGS :o)
TimToady it was nice to meet you, btw
FROGGS gnight
thanks sir
only can say they same..
TimToady: are you still around the next days?
lue
.oO(Try describing it as "adding a negative number of stones" to throw people off! Fun for the family!)
04:36
TimToady leaving Saturday
FROGGS me too
lue ♞, FROGGS o/
04:36 dylanwh joined
sorear ditto 04:36
welcome dylanwh :)
FROGGS I think I'll enjoy the hackathon
cool!
labster hackathon++
FROGGS gnight anyway
04:36 FROGGS left 04:37 xinming joined
dylanwh sorear: hi 04:39
masak 'night, FROGGS
jnthn sorear: I'm here on IRC for a moment if you have nqp-jvm questions 04:41
sorear jnthn: yay. trying to figure out the relationship beteen QAST::CompilerJAST and QAST::Var and where variable refs are compiled 04:43
jnthn QAST::Var is just an AST node
sorear I've golfed package variables out of 24-modules.t; it's doing weird stuff with shadowing
jnthn QAST::CompilerJAST is what translates it to something. Of note, the method compile_var.
sorear module foo { sub bar() { ... } } # in ..., $?PACKAGE gets compiled as $?OUTER::OUTER::PACKAGE and winds up referring to GLOBALish 04:44
d'oh. was seeing QAST::CompilerJAST as a Node subclass somehow... *blind* 04:45
jnthn No, it's not
masak it is suggestively named... :)
jnthn ;) 04:46
You expect me to be *consistent*? :P
lue
.oO(There was that vtablee...)
04:47
04:48 woolfy joined
jnthn o/ woolfy :) 04:48
04:49 colomon joined
sorear o/ colomon 04:50
04:50 lustlife1 joined
TimToady is always consistent, except when he's not 04:50
colomon \o
04:50 woolfy left
colomon sorear: I figure if I can sort out the smoker bug in the next five minutes, I still have time to save tonight's smoke run. :) 04:51
PerlJam TimToady: that's how you avoid the hobgoblins of foolish consistency!
04:52 sqirrel joined
masak as a proponent of the strangely consistent, I take exception to this making light of the foolishly consistent... :P 04:54
04:55 mucker left 04:57 zby_home joined
lue
.oO(The Inconsistently Consistent would like to have a word with you on the strangely but not the foolishly!)
04:59
04:59 birdwindupbird joined
masak lue: I schedule meetings constantly with the Inconsistently Consistent, but they never show up. 05:02
colomon gives up and goes to bed. 05:04
lue that's very consistent. Which is inconsistent for the Inconsistently Consistent. Which is (consistent&!consistent) with their mission statement. I am shocked and undisturbed by their behavior.
♞, colomon o/
05:05 birdwindupbird left
masak ok, this day is now sufficiently awesome. I'm going to cap the number of impressions by going to bed. 'night, #perl6 05:11
lue impressions? ♞ anyways, masak o/
sorear o/ masak indeed 05:12
05:14 sqirrel left
jnthn 'night, #perl6 05:19
lue ♞, jnthn o/ 05:20
05:25 adu left 05:35 raiph left 05:39 saxx11 joined 05:40 birdwindupbird joined 05:41 raiph joined 05:42 Jun` joined, Jun` left, Jun` joined 05:44 daniel-s joined 05:45 Jun` left
sorear nqp: my %h; %h<a>=1; %h<b>=2; say(%h) 05:48
camelia nqp: OUTPUT«Assignment ("=") not supported in NQP, use ":=" instead at line 2, near "1; %h<b>=2"␤current instr.: 'panic' pc 14721 (src/stage2/gen/NQPHLL.pir:5232) (src/stage2/gen/NQPHLL.nqp:279)␤»
sorear nqp: my %h; %h<a>:=1; %h<b>:=2; say(%h)
camelia nqp: OUTPUT«Hash[0x8ecf2f4]␤»
sorear nqp: my %h; %h<a>:=1; %h<b>:=2; for %h -> $a { say($a) }
camelia nqp: OUTPUT«a␤b␤»
05:50 tomyan joined 05:56 Jun` joined 05:58 Jun` left, lustlife joined 06:10 dmol left 06:11 sqirrel joined 06:22 zby_home left 06:23 saxx11 left 06:28 tomyan left, sqirrel left
sorear .ask jnthn dylanwh is kind of putting me on the spot here with a question whose answer is intuitively obvious but not easy to express ;) ;) Why exactly cannot NativeCall functionality be used in CORE.setting? (beyond the obvious "it's a module") 06:29
yoleaux sorear: I'll pass your message to jnthn.
06:30 crab2313 joined 06:40 SamuraiJack joined 07:02 kaleem joined 07:03 domidumont joined 07:07 arlinius left 07:12 flaviusb joined 07:17 flaviusb left 07:28 fhelmberger joined 07:32 flaviusb joined 07:36 domidumont left 07:37 domidumont joined, flaviusb left 07:38 crab2313 is now known as jinkela 07:39 jinkela is now known as SanDiego 07:41 SanDiego is now known as crab2313 07:44 labster left, flaviusb joined
flaviusb r: say 1,1,2,3,5...* 07:44
camelia rakudo b2072f: OUTPUT«Unable to deduce sequence␤ in method sink at src/gen/CORE.setting:10169␤ in block at src/gen/CORE.setting:13701␤ in any coro at src/gen/CORE.setting:6040␤ in method reify at src/gen/CORE.setting:6022␤ in method reify at src/gen/CORE.setting:5761␤ in method r…
07:48 flaviusb left 08:02 flaviusb joined 08:04 [particle] left 08:07 flaviusb left, arlinius joined 08:21 flaviusb joined 08:26 flaviusb left
tadzik colomon: oh, so much backlog :) 08:32
yoleaux 5 Jun 2013 19:53Z <colomon> tadzik: Line 45 of smoker (emmentaler) is $panda.ecosystem.project-get-state($x) ne 'absent' -- surely that should be eq instead of ne? Otherwise we only test modules which are not in the ecoysystem.
5 Jun 2013 19:54Z <colomon> tadzik: Also, something else seems to be borked in panda, because $panda.ecosystem.project-get-state($x) is returning "installed" for a module which panda does not report as installed.
5 Jun 2013 20:05Z <colomon> tadzik: or it's possible that I do have it installed, and reporting it not installed from the panda list command was the error? I am confused.
tadzik %log{$p} = gen-result($! // 'chocolate cake'); # look at gen-result, you'll understand :) 08:33
08:34 [particle] joined
tadzik colomon: no, ne 'absent' is alright, if it's already installed we generate a positive result 08:34
colomon: it's fine that smoker thinks something installed and panda doesn't. Things installed with smoker are removed afterwards 08:35
moritz pragprog.com/magazines/2013-06/prog...ing-elixir # another interesting language 08:36
08:38 wtw joined 08:41 flaviusb joined 08:45 flaviusb left, cosimo joined 08:47 dakkar joined
mathw Too many interesting languages! 08:51
mathw has been looking at Go a bit 08:52
08:52 crab2313 left 08:53 raiph left
mathw That page doesn't make me interested in Elixir though 08:54
08:54 crab2313 joined 08:58 snearch joined 09:00 flaviusb joined
mathw what makes me interested is the metaprogramming stuff and protocols 09:02
unfortunately I find ruby-like syntax something of a turn-off
09:05 flaviusb left 09:20 flaviusb joined 09:23 crab2313 left 09:25 flaviusb left, crab2313 joined 09:28 raiph joined 09:31 FieldsaBB left 09:32 araujo left 09:35 crab2313 left 09:39 flaviusb joined 09:44 nyuszika7h left, flaviusb left 09:45 nyuszika7h joined 09:52 araujo joined, araujo left, araujo joined 10:04 rindolf joined, Guest1337 joined 10:05 anuby left 10:19 raiph left 10:24 snearch left 10:30 xinming left 10:32 xinming joined
timotimo good * 10:35
r: role Bob { method foo {...} }; class Bleh does Bob { method foo { say "yup" } }; my Bob %h; %h<a> = Bleh.new; say %h<a>; 10:36
camelia rakudo b2072f: OUTPUT«Bleh.new()␤»
timotimo r: role Bob { method foo {...} }; class Bleh does Bob { method foo { say "yup" } }; my Bob %h; %h<a> = Bleh.new; say %h<b>;
camelia rakudo b2072f: OUTPUT«Method 'foo' must be implemented by Bob because it is required by a role␤ in any at src/gen/Metamodel.nqp:2207␤ in any compose_method_table at src/gen/Metamodel.nqp:2198␤ in any apply at src/gen/Metamodel.nqp:2216␤ in any at src/gen/Metamodel.nqp:2356␤ in an…
timotimo :\
i think the java people know about this problem, too 10:37
10:48 xinming left 10:54 Boohbah left 10:57 atroxaper left 10:58 benabik left 11:00 crab2313 joined 11:08 cognominal left 11:27 gdey left 11:28 crab2313 left 11:32 xinming joined 11:33 gdey joined
timotimo oh, no talks to watch today 11:33
11:36 gudahtt joined
nwc10 maybe there will be a webcam streaming the hackathon 11:41
timotimo not as interesting if nobody is there to talk about what's going on ;) 11:42
and i could potentially get that on irc
(if only through commit notifications)
nwc10 depends if they are talking around a whiteboard
timotimo ah 11:43
11:47 arlinius left 11:48 Pleiades` left, lue left, kshannon left 11:49 lue joined, kshannon joined, Pleiades` joined 11:56 skids left 12:03 atroxaper joined 12:05 cibs left
colomon wishes jnthn were not currently in his timezone 12:09
12:12 cibs joined
atroxaper What do you think about Intellij Idea plugin for Perl6, #perl6? I think it may attract a lot of interest Perl6. 12:17
timotimo what would it offer beyond perhaps syntax highlighting and proper automatic indentation? 12:18
jump to definition seems doable
12:19 arlinius joined
colomon has anyone managed to build the Rakudo debugger recently? 12:19
atroxaper timotimo: Of course jump to definition is supported. confluence.jetbrains.com/display/In...ge+Support 12:21
timotimo colomon: i've failed multiple times on my laptop :(
error:imcc:loadlib directive could not find library `perl6_group' 12:22
atroxaper: well, first you'll have to develop it, no? ;) 12:23
atroxaper timotimo: :)
timotimo as far as i can tell, intellij idea is pretty darn good
atroxaper yes, it's pretty good.
12:24 konundra left
atroxaper timotimo: i frozen at the start - grammar definition xD 12:25
I just don't know what use for perl scripting in windows... 12:28
Ulti Padre? 12:31
colomon tadzik: I can definitely tell you that smoker hasn't run correctly for me in days...
atroxaper Ulti: is it really useful? 12:32
Ulti what do you mean by useful?
I find text editors useful enough 12:33
Perl5 suffers from not having function signatures so its not like autocomplete or something is any good, which is why a Perl6 IDE could be really awesome there is more information to scam from the language 12:34
padre has all the docs integrated which is useful
atroxaper :) Ok. I'll try Padre. 12:35
Ulti I think it does moose class completion stuff but that might be my imagination.... my vim setup does that anyway :3 12:37
12:37 guru joined
Ulti not that I use it because I write disgusting perl5 for doing research quickly 12:37
12:38 guru is now known as Guest2778, Guest2778 is now known as ajr_
[Coke] .tell frogs ll exception gives me 💈 ./perl6 --ll-exception -e 'use v6;' 12:38
yoleaux [Coke]: I'll pass your message to frogs.
12:39 konundra joined
[Coke] ugh. 12:40
.tell FROGGS ll-exception gives me : gist.github.com/coke/5721191
yoleaux [Coke]: I'll pass your message to FROGGS.
12:53 gdey left 12:55 Guest1337 left 13:01 mucker joined
colomon tadzik: think I've got the problem. failure is return X::AdHoc, not X::Panda 13:03
13:10 rhr joined, ggoebel left, konundra left, vaelxon left, bloonix left, ironcame1 left, rhr_ left, Guest61852 left, bakedb left, rom1504 left, konundra joined 13:11 ironcamel joined, tomaw_ joined, rom1504 joined, ingy joined, Guest61852 joined 13:12 Guest61852 is now known as felher, tomaw left, nyuszika7h left, gudaht joined, x2 joined 13:13 nyuszika7h joined
colomon tadzik: somewhere along the line you lost the X::Panda.new from Builder.pm and Tester.pm -- but all the code that calls them still expects it to be there. 13:13
13:14 tomaw_ is now known as tomaw, gudahtt left 13:15 sivoais_ joined, kshannon_ joined 13:16 dagurval joined 13:18 bloonix joined, sorear left, kshannon left, GlitchMr left, bakedb joined, sorear_ joined 13:19 sivoais left 13:20 p5eval_ left, p5eval joined, szbalint_ left, szbalint joined 13:25 sftp left, GlitchMr joined
masak morning, #perl6 13:26
about to head for $airport
timotimo hey masak
13:26 sftp joined
timotimo no hackathoning for you today? 13:26
masak unfortunately not.
timotimo well, have a good flight! :) 13:27
colomon masak: safe home!
13:27 JimmyZ joined
JimmyZ Good evening, #perl6 13:28
masak timotimo, colomon: thank you. 13:32
diakopter masak: want a ride to the airport? 13:34
masak diakopter: ooh!
diakopter k i'll be right there
masak diakopter: meant to hail a cab, but ride is better :>
diakopter++ 13:35
$weekend is likely to be full of AFK, so I don't think I'll resurface much until Monday. we'll see.
13:36 adu joined
masak I've had a completely wonderful YAPC::NA. I'm really happy I could come here. 13:37
hoelzro masak: how was the Perl 6 content received?
pmichaud's talk last year is what piqued my interest, so I'm curious =) 13:38
masak hoelzro: well, the new kid on the block this year was MoarVM, no doubt.
I heard it being whispered about in the hallway track.
hoelzro it's promising
I was excited to read about it!
masak jnthn++ gave two nice talks about Perl 6; one brain-twister, one about his cool debugger. 13:39
hoelzro I'm hoping to look at it more in depth
masak pmichaud++ gave a talk about what needs to be done with parallelism. more "what don't we know" than "what do we know". still very interesting.
japhb++ gave two nice talks: one about module loading, and one about performance.
FROGGS++ gave a cool talk about his v5 stuffs. 13:40
13:40 chayin left, chayin joined
moritz are the talks online somewhere? 13:40
13:40 gdey joined
timotimo any information on what kinds of things are being done to perl6 and the perl6 ecosystem in the hackathons? 13:40
hoelzro are the talks online this year?
masak it was definitely the most Perl 6 conf I've been to. 13:41
13:41 araujo left
hoelzro moritz: took the words right from my mouth =) 13:41
masak I don't know anything about talks online.
you people seem to know more about streaming and urls to that than I do.
moritz masak: how did the non-6ers react to so much 6?
masak I'm just physically present :P
13:41 araujo joined
masak moritz: well, those who were interested showed up for those talks. 13:42
that's all, I think :)
there were 80 speakers and 90 talks.
we figured out that between jnthn, japhb and me, we represented half of that >1 multiplicity :) 13:43
oh, more nice speakers: lizmat++ gave a lightning talk about "is Perl 6 'Perl'?"
diakopter masak: I'm downstairs 13:44
masak and TimToady gave a touching keynote and also a lightning-talk "poem", a modern adaptation of Corinthians.
diakopter: I'll be down right away. 13:45
&
moritz have a safe flight.
13:46 skids joined
timotimo i'm having trouble compiling older versions of rakudo: "operation setcontspec requires 3 operands, but got 4" 13:47
maybe --gen-nqp didn't get a sufficiently old version?
moritz gen-nqp doesn't compile an old nqp for you if a newer is available 13:48
13:52 woolfy joined, labster joined
timotimo tried an older parrot, too 13:52
13:53 yeltzooo left
timotimo tries* actually 13:53
13:53 yeltzooo joined
[Coke] timotimo: when I am doing things that are version specific, I typically do a complete wipe of nqp and parrot subdirs (which git clean -xdf does not do for you, it lies), and get whatever dep versions rakudo wants. 13:54
13:54 btyler joined
timotimo it may be my fault to try parrot on master branch and nqp on master branch, too 13:56
jnthn morning, #perl6 14:01
yoleaux 06:29Z <sorear> jnthn: dylanwh is kind of putting me on the spot here with a question whose answer is intuitively obvious but not easy to express ;) ;) Why exactly cannot NativeCall functionality be used in CORE.setting? (beyond the obvious "it's a module")
14:02 gdey left 14:03 FROGGS joined
jnthn .tell sorear well, given it actually boils down to nqp:: ops and all NativeCall does is provide nice sugar, you could do native calls in CORE.setting through those already. But I'm pretty sure I don't want us to be doing that... 14:03
yoleaux jnthn: I'll pass your message to sorear.
14:05 atroxaper left 14:06 diederich joined, labster left, alester left
FROGGS morning 14:07
yoleaux 12:40Z <[Coke]> FROGGS: ll-exception gives me : gist.github.com/coke/5721191
timotimo how do i get out of "could not find perl6::BOOTSTRAP" again?
make install repeatedly until it works? ;)
oh, interesting
make install will install BOOTSTRAP into parrot/.../languase/nqp/lib/Perl6 14:08
and when trying to build rakudo::debugger it will look in parrot/.../languages/perl6/site/lib
jnthn wonders why the trouble building rakudo-debugger 14:09
Given I was hacking on it and demoing it yesterday...
And it build without trouble
14:09 census joined
moritz jnthn: because your (OS, Moon Phase, Panda Version, Rakudo Version) tuple was slightly different :-) 14:09
pmichaud good morning, #perl6 14:10
dalek kudo-debugger: bd3e80d | jonathan++ | bin/perl6-debug.nqp:
Ensure updated $/ in grammar debugging.
kudo-debugger: 66f7e15 | jonathan++ | lib/Debugger/UI/CommandLine.pm:
Make $<foo> and $0 dump nicer.
kudo-debugger: 342266a | jonathan++ | lib/Debugger/UI/CommandLine.pm:
Un-bit-rot tracepoint show.
kudo-debugger: 3176315 | jonathan++ | / (2 files):
Add breakpoint setting in code with "#?BREAK".
timotimo moon phase version :)
FROGGS .tell [Coke] I can imagine that this fails: github.com/rakudo/rakudo/blob/jvm-....nqp#L4924 14:11
yoleaux FROGGS: I'll pass your message to [Coke].
jnthn o/ pmichaud
FROGGS \o pmichaud
timotimo i just successfully installed it 14:12
all it took was to completely wipe my panda clean
FROGGS .tell [Coke] maybe you add a debug line right after that to check if $v has something in it
yoleaux FROGGS: I'll pass your message to [Coke].
jnthn use v6 fails ultimately 'cus it hits Str.match 14:13
Which in turn does gather/take
14:13 census left
FROGGS ahh 14:14
(I'm totally new to the JVM stuff)
jnthn ;)
FROGGS the bad thing is, I missed the breakfast time slot :/ 14:15
you are already in the LBJ/TCC?
14:16 colomon left
timotimo time for a swapfile... 14:16
jnthn FROGGS: No, not yet
FROGGS k, I hoe I get some coffee there
timotimo: yeah, I guess cleaning up pandas is some sort of challenge, given that it does quiet some magic 14:17
timotimo :) 14:18
14:19 sorear_ left
jnthn r: 'abc' ~~ /b/; eval 'say $/' 14:19
camelia rakudo b2072f: OUTPUT«(Any)␤»
jnthn Got a fix for ^ locally
timotimo oh! will that also make it work in the repl? 14:20
FROGGS to quote TimToady: "locally++"
jnthn timotimo: checking
timotimo hm, not sure if i'm referring to the error correctly
jnthn timotimo: yes
timotimo there's things like writing a try block and not having $! in the next line; does that work?
14:20 shinobicl joined 14:21 sorear joined
jnthn > try die "bratwurst" 14:22
Nil
> say $!
bratwurst
\o/
It also fixes looking at $! and $/ in the debugger :)
FROGGS I feel offended
jnthn
.oO( One patch to fix them all )
14:23
FROGGS hehe
timotimo cool!
jnthn FROGGS: Saussage is offensive? :)
timotimo i'd love to know what exact combination of nqp/parrot you're running, jnthn? master nqp and 5_2_0 parrot?
jnthn timotimo: exactly that. 14:24
FROGGS jnthn: no, the fact that it must die
[Coke] if we don't have java on feather, we'll also never get evalbot doing it. bother.
yoleaux 14:11Z <FROGGS> [Coke]: I can imagine that this fails: github.com/rakudo/rakudo/blob/jvm-....nqp#L4924
14:12Z <FROGGS> [Coke]: maybe you add a debug line right after that to check if $v has something in it
jnthn FROGGS: ah, I guess "die" is the wrong article ;)
FROGGS ahh
no, "die" would be right 14:25
jnthn :)
FROGGS just a matter of pronunciation
14:25 lizmat joined
[Coke] jnthn: where is the definition of 'use' in rakudo-java? 14:26
14:26 atroxaper joined
FROGGS [Coke]: <jnthn> use v6 fails ultimately 'cus it hits Str.match 14:26
<jnthn> Which in turn does gather/take
[Coke] (if no use's work right now, let's hack it so we can at least run the tests until we get to the tests on use)
FROGGS: yes, danke.
14:27 btyler left
jnthn [Coke]: Same place as it is in Rakudo Parrot: in statement_control:sym<use> in grammar, or the related method in actions. 14:27
FROGGS [Coke]: github.com/rakudo/rakudo/blob/jvm-....nqp#L1151
jnthn [Coke]: For most things, assuming "it's shared between backends" will be a good assumption, since 90% or so of the code is :) 14:28
FROGGS [Coke]: maybe you can try changing "| <version" to "| 'v6'"
jnthn Please #?if !jvm ...original thing... #?endif #?if jvm ...cheat... #?endif 14:29
So we don't regress on it on Rakudo Parrot.
oh, I wonder if those files go through the preprocessor...may have to look at that. 14:30
FROGGS what are "those files", the stuff in src/ ? 14:31
[Coke] jnthn: #?nIFty 14:32
jnthn: ok. any particular keyword you want in there to indicate that it's a bletcherous hack?
14:32 benabik joined, gdey joined
FROGGS would be cool if that works out 14:32
XXX
at least I have seen this at some places 14:33
diakopter jnthn: ping 14:34
jnthn: want a ride to the hackathon? it's invisible pouring outside, if you know what I mean 14:35
FROGGS: u want a ride too?
14:36 colomon joined, ajr_ left
colomon o/ from the hackathon 14:36
TimToady is there coffee there?
timotimo hey colomon. are you perl6-hackathoning? :)
diakopter well, of the silty dirt water kind
lizmat morning #perl6 14:37
colomon timotimo: absolutely!
14:37 guru joined
TimToady diakopter: as long as it has caffeine in it 14:37
14:37 guru is now known as Guest95867, Guest95867 is now known as ajr_
lizmat van leaving from doubletree for lot40 around 9:50 14:37
jnthn diakopter: pmichaud++ already offered me a ride, and I think I only need one ;-)
diakopter seems so
timotimo make: *** [NQPP5QRegex.pbc] Segmentation fault
aaw not again :( 14:38
diakopter TimToady: if you don't make Liz's 9:50 trip, I can head back there too
TimToady lemme see if I can rouse my progeny...
lizmat maybe rm-rf install dir
FROGGS diakopter: no, I can walk these 3 minutes :o)
but thanks
lizmat TimToady: we'll wait for you if you want the ride 14:39
TimToady he wants to stay here, but I can get there
I mean, I'd likea ride
FROGGS TimToady: I'll chec the coffee-state and report back (cause I really need one too)
TimToady I made some Wolfgang Puck on spec here 14:40
FROGGS the hackathon is still at the LBJ loppy?
diakopter no
tcc 2.122
FROGGS lobby*
diakopter and it's not as cool as it could be. calling the building.
FROGGS diakopter: thanks
atroxaper Does anybody know is there Perl6 BNF grammar?
14:43 salv0 left
diakopter atroxaper: afaik, the Perl 6 grammar has too many left & right recursions to be expressed as such 14:43
pmichaud my vehicle can probably depart around 10:05 14:44
arnsholt The Perl 6 grammar isn't expressed as a BNF grammar, but there is a grammar
TimToady actually, Lewis wants to come now
diakopter the Dr Pepper doesn't arrive until 10:30 14:45
pmichaud after jnthn++, I have two seats left
TimToady now that he is awake :)
atroxaper arnsholt: i heard that about P5. Are you sure about P6? 14:46
pmichaud or put another way, I'll be ready to depart in about 15-20 mins
atroxaper TimToady: is Perl6 expressed as a BNF ? :)
pmichaud Perl6 is expressed as a koan. 14:47
jnthn atroxaper: Perl 6 is expressed as github.com/perl6/std/blob/master/STD.pm6
14:47 FROGGS left
benabik Perl 6 is expressed as Perl 6 14:47
atroxaper What is koan?
diakopter masak: you're a koan 14:48
PerlJam atroxaper: why, perchance, do you want a BNF grammar for Perl 6?
atroxaper jnthn: yes, i khow.
pmichaud en.wikipedia.org/wiki/K%C5%8Dan
TimToady coming down (after I pack my computer :) 14:49
pmichaud TimToady: you have a ride already?
atroxaper PerlJam, i have a stupid idea to write Perl6 plugin for Intellij Idea. I have to have bnf grammar definition for that...
14:50 lizmat left
diakopter I need a photo of the Yapcna hackathon room - the Perl 6 folks are all on one side, and the Perl 5 folks on the other, separated by a deep chasm 14:50
pmichaud FROGGS++ can be in the middle :)
and the chasm isn't really that deep anymore
atroxaper pmichaud, nice )
PerlJam atroxaper: maybe you can get by with a partial grammar? 14:51
diakopter what is this, good cop bad cop? ;)
14:51 woolfy left 14:52 kaare__ joined
atroxaper PerlJam: what do you mean? I think it's not too easy to describe Perl6 in partial bnf for my now. 14:52
jnthn pmichaud: Will meet you down in the lobby at around 10:05 14:53
colomon atroxaper: well, for sure it's not easy to describe Perl6's grammar completely in BNF...
pmichaud jnthn: meet in lobby 10:05-ish?
colomon jnthn: gist.github.com/colomon/5722100 # debugger build failure on Linux
jnthn pmichaud: yes :) 14:54
14:54 labster joined
pmichaud see you there 14:54
jnthn colomon: It picks up whatever nqp is in the path...maybe it's picking the wrong one...
colomon perl 6: Now with exploding objects. # re arnsholt++'s current debugging run 14:55
It's getting /home/colomon/tools/rakudo/install/bin/nqp 14:56
which *sounds* rightish...
14:57 spider-mario joined, gdey left
colomon jnthn: unless there are two different nqps in the rakudo source tree? 14:57
timotimo i think this time i'll get a working rakudo and rakudo-debugger 14:58
jnthn Hm, not sure why it's getting it wrong then..
14:58 FROGGS joined
jnthn hackathon & 14:58
14:58 btyler joined
FROGGS \o/ 14:59
coffee and snacks are here
15:00 cognominal joined 15:02 kaleem left
PerlJam But no Dr Pepper :( 15:03
FROGGS true :o( 15:04
PerlJam untrue now! 15:05
FROGGS PerlJam: can you turn around?
diakopter and an introduction occurs 15:06
colomon \o/ 15:11
15:13 zby_home joined
RedditAnalytics are you guys in the same room on laptops? 15:13
15:13 btyler left, kaleem joined
colomon RedditAnalytics: yes 15:13
PerlJam Was that an unintentional pun? 15:14
RedditAnalytics you know technology has really done a lot for us. 50 years ago you would have had to walk over to the other side of the room to chat with your friend.
colomon RedditAnalytics: lizmat, FROGGS, diakopter, arnsholt, labster, PerlJam, and TimToady just showed up.
15:15 berekuk_ left
colomon RedditAnalytics: we're talking outloud, too. may make the channel traffic kind of confusing today. 15:15
15:16 lizmat joined
colomon RedditAnalytics: and pmichaud and jnthn just walked in the door. 15:16
jnthn: get_iter() not implemented in class 'Stash' is a bad thing, right? 15:17
Now that I've finally got module testing reporting something again, there are 25 more failures today than there were Saturday.
15:18 cognominal left
FROGGS what I dont like finding weird stuff in rakudo when trying to make things in v5 work, that just causes delay... (just mumbling) 15:18
15:19 cognominal joined, quietfanatic joined 15:21 kaleem left 15:22 kaleem joined
FROGGS TimToady: is this supposed to work? require Test; import Test; 15:22
15:22 Ulti left
FROGGS or only: require Test :ALL 15:23
I mean, "require Test; import Test;" can work basically, like this too: require Test:file("lib/Hurz.pm"); import Test; 15:24
moritz FROGGS: import is run at compile time 15:25
how would it know what to import?
colomon # b2072f578c04bbf7ac4d4b684ffdff42a2b2a29a 15:26
[Coke] woudl pay to have a few tuits shipped to albany, btw.
(the new ones with rakudo/moar)
FROGGS moritz: well, it knows the module name at least, and then could import symbols, which then can be accessed like ::('cmp_ok')
pmichaud [Coke]: I'll ship you some 15:27
how many would you like?
[Coke] eh. not a lot of sixers out this way. I could hand them out if we ever had another Albany.pm meeting... no more than a dozen, surely.
pmichaud dozen each?
colomon [Coke]: there are plenty here... ;) 15:28
[Coke] that would err on the side of damnthat'salot.
pmichaud six each?
15:28 JimmyZ left
[Coke] ... but in that case, sure! :) 15:28
pmichaud okay, dozen each :)
or if that's too much, ten of each
just the rakudo and moarvm ones?
[Coke] that works - you send 'em, I promise to have at least a beer meeting to distribute them.
those were the only ones I saw! 15:29
PerlJam [Coke]++
[Coke]: there are 7 tuit types (so far)
[Coke] O_o? . o O (I have only a single one from years ago with no type)
I will happily take a batch and redstribute them over beer as long as I get a rakudo one. :)
pmichaud "Rakudo" "MoarVM" "2013" "CPAN" "P5&P6" "Community" 15:30
FROGGS there is YAPC, rakudo, MoarVM, 2013, errhhhmmmm
right
pmichaud and "YAPC"
timotimo error:imcc:loadlib directive could not find library `perl6_group'
;_;
FROGGS hugs timotimo
timotimo but why was i able to do it on my vm? :/ 15:31
FROGGS moritz: maybe it can delay the importing till runtime if the module was require'd and not use'd
gah, I'm not sure 15:32
timotimo oh, i forgot to --gen-parrot=something 15:33
another hour of compilation >_<
FROGGS O.o
we need to use Moar cores 15:34
timotimo well, probably not an hour, but this is a laptop
15:35 gdey joined 15:37 stevan_ joined 15:38 raiph joined
jnthn arnsholt: heh, Moar's P6opaque had the bug too. 15:41
lizmat r: module Foo { module Bar { say $?PACKAGE } } # shouldn't this need to say (Foo::Bar) ???
camelia rakudo b2072f: OUTPUT«(Bar)␤»
lizmat n: module Foo { module Bar { say $?PACKAGE } } # shouldn't this need to say (Foo::Bar) ???
camelia niecza v24-66-gaee6525: OUTPUT«Unhandled exception: System.NullReferenceException: Object reference not set to an instance of an object␤ at Niecza.StashCursor.Core (System.String key, Boolean final, Niecza.StashCursor& sc, Niecza.Variable& v, Niecza.Variable bind_to) [0x00000] in <file…
FROGGS r: module Foo { module Bar { say $?PACKAGE } }; say Foo:: # I think so 15:42
arnsholt jnthn: Good thing we found it then!
camelia rakudo b2072f: OUTPUT«(Bar)␤("Bar" => Bar).hash␤»
arnsholt is recompiling stuff to test
FROGGS r: module Foo { module Bar { say $?PACKAGE } }; say Foo::Bar # I think so
camelia rakudo b2072f: OUTPUT«(Bar)␤(Bar)␤»
FROGGS hmmm
jnthn colomon: Yes, get_iter() error is a bad thing. Got a golf I can look at?
colomon jnthn: pondering that question right now 15:43
jnthn: so, not yet. :)
jnthn arnsholt++ # found it 15:44
colomon: OK. I suspect I can nail it fairly swiftly once it's golfed
dalek kudo/nom: c38c659 | jonathan++ | src/Perl6/ (2 files):
Fix $/ and $! visibility in eval/REPL.
15:45
lizmat wonders whether jnthn has a jonathan clone 15:46
pmichaud [Coke]: picasaweb.google.com/1005833335019...YapcNa2013 # tuits I've set aside for you
15:46 btyler joined
labster o/ masak 15:47
colomon arnsholt++
RedditAnalytics the worst kind of bugs are the ones that don't blow up your program
[Coke] pmichaud++
FROGGS arnsholt++ # The Bug got eliminated (no bug left)
pmichaud there's 10x of "Rakudo" and "MoarVM", and 1x of the rest
RedditAnalytics I'm going to start keeping a journal of the most interesting bugs i create
15:48 domidumont left
benabik I didn't have enough tuits to work on Perl stuff this year, so I didn't go to YAPC, so I don't get tuits? That doesn't seem right. ;-D 15:49
15:49 kaleem left
lizmat benabik: visiting any other Perl event this year? 15:51
benabik Maybe I should drive across the state and steal some of [Coke]'s.
15:51 stevan_ left
[Coke] benabik: just show up at an Albany.pm meeting! 15:51
PerlJam benabik: I bet if you ask woolfy she'll send you a bunch.
[Coke] ... last one, the guy who suggested a bar was the only one to show up. whoops.
pmichaud I'll be glad to be a U.S. distributor for tuits for a while. 15:53
15:53 jeffreykegler joined
pmichaud I'm going to be doing a lot of shipping-type-things over the next few months anyway 15:53
arnsholt timotimo: Did you get your NQP segfault fixed?
pmichaud it's no problem to add an extra parcel or or two
lizmat TimToady: do you have an opinion on "module Foo { module Bar { say $?PACKAGE } } # shouldn't this need to say (Foo::Bar) ???"
pmichaud s/or or/or/
TimToady lizmat: I'd think so 15:54
timotimo arnsholt: it seems so
lizmat submits rakudobug
jnthn Wait, stringifying a type object gives the short nam,e no?
lizmat ah, how to get the long name then? 15:55
arnsholt timotimo: Excellent. I got something that looked similar, so good that cleaning out and recompiling fixed it
jnthn .WHO iirc, though not sure that does the Right thing yet.
pmichaud $?PACKAGE.WHO stringifies to long name
jnthn Right 15:56
well, *should* :)
lizmat rn: module Foo { module Bar { say $?PACKAGE.WHO } }
camelia niecza v24-66-gaee6525: OUTPUT«Unhandled exception: System.NullReferenceException: Object reference not set to an instance of an object␤ at Niecza.StashCursor.Core (System.String key, Boolean final, Niecza.StashCursor& sc, Niecza.Variable& v, Niecza.Variable bind_to) [0x00000] in <file…
..rakudo b2072f: OUTPUT«().hash␤»
lizmat adapts rakudobug before sending off
timotimo i think i would do good to toss out my Coord2d class and use complex numbers instead? 15:57
they might be faster to create
arnsholt Could be 15:58
15:58 konundra left
arnsholt And complex numbers for plane geometry is a fairly normal thing to do 15:58
colomon timotimo: probably would be faster, I believe complex numbers are optimized a bit in Rakudo. 16:00
timotimo indeed. then i'll just turn my methods into subs 16:01
is there a cool mathematical construct that would give me the lowest real and lowest imaginary part for a set of numbers, or perhaps just two numbers?
16:01 gdey left
timotimo and for the highest? i'd like to get the bounding box of all my coordinates 16:02
dalek p: 61d80b9 | (Arne Skjærholt)++ | src/vm/parrot/6model/reprs/P6opaque.c:
Fix GC mark offset calculation.

Got the wrong answer in cases where complex inlines that are not box targets were made.
labster yay, nqp is fixed. Or rather, I rebuilt all the way down to parrot and it works now.
I'm guessing one of the libraries changed and parrot got confused. 16:03
colomon labster: it was the grand merge
jnthn: your requested ll-exception: gist.github.com/colomon/5722694 16:05
sorear good * #perl6 16:10
yoleaux 14:03Z <jnthn> sorear: well, given it actually boils down to nqp:: ops and all NativeCall does is provide nice sugar, you could do native calls in CORE.setting through those already. But I'm pretty sure I don't want us to be doing that...
16:10 konundra joined
colomon \o 16:13
16:13 gdey joined
timotimo arnsholt: does that fix nativecall? 16:14
16:14 domidumont joined
arnsholt Depends on what you mean by fix NativeCall =) 16:15
timotimo gggaaaaah
still unable to build Rakudo::Debugger
arnsholt But it fixes the segfault bug reported with callbacks
timotimo i am going insane
colomon timotimo: pretty sure Rakudo::Debugger is just broken at the moment. except it works for jnthn++
timotimo and except i just built it successfully on my other machine 16:16
colomon what's the difference between the machines? 16:17
16:17 ajr_ left
benabik colomon: The debugger works on one and not the other, obviously. 16:18
sorear o/ colomon
timotimo one's a mint VM, the other is a debian sid laptop 16:20
colomon timotimo: 32-bit versus 64-bit? maybe? 16:21
extra nqp causing trouble?
16:22 colomon left
timotimo no, both 64 16:23
16:25 salv0 joined
sorear nqp: say([]) 16:27
camelia nqp: OUTPUT«0␤»
sorear nqp-jvm: say([])
camelia nqp-jvm: OUTPUT«Cannot stringify this␤ in <anon>␤ in print␤ in say␤ in <anon>␤ in <anon>␤ in eval␤ in evalfiles␤ in command_eval␤ in command_line␤ in MAIN␤ in <anon>␤ in <anon>»
sorear jnthn: I have a want for a general way in nqp-jvm to get a string (probably just class+hashcode) for a general object 16:28
16:30 colomon joined 16:31 stevan_ joined
[Coke] ... if we have nqp-jvm, where is that java on feather? 16:32
sorear o/ stevan_
[Coke]: feather3?
stevan_ o/ sorear 16:35
16:35 stevan_ left
TimToady nr: package Foo { our proto foo($x) {*}; multi foo($x) { $x.say } }; Foo::foo(42); 16:35
camelia niecza v24-66-gaee6525: OUTPUT«Potential difficulties:␤ $x is declared but not used at /tmp/wPPV6nG8_u line 1:␤------> package Foo { our proto foo(⏏$x) {*}; multi foo($x) { $x.say } }; Foo␤ &foo is declared but not used at /tmp/wPPV6nG8_u line 1:␤------> packag…
..rakudo b2072f: OUTPUT«42␤»
16:36 fhelmberger left
TimToady nr: package Foo { our proto foo(|) {*}; multi foo($x) { $x.say } }; Foo::foo(42); 16:36
camelia niecza v24-66-gaee6525: OUTPUT«Potential difficulties:␤ &foo is declared but not used at /tmp/avzSn2CBoe line 1:␤------> package Foo { our proto foo⏏(|) {*}; multi foo($x) { $x.say } }; Foo␤␤Unhandled exception: System.NullReferenceException: Object reference not set…
..rakudo b2072f: OUTPUT«42␤»
TimToady I guess that shouldn't complain about 'our' declarations like that
but anyway, we can use our to put something proto-ish as a feed target into a package, and use the proto's multis to process events sent to that feed target 16:37
16:37 gdey left
sorear I HAVE FOUND IT 16:38
arnsholt Is it a pot of gold?
RedditAnalytics lol
sorear arnsholt: no, it's a bug I've been chasing since Sunday
arnsholt Goody! 16:39
RedditAnalytics that's a fast bug
TimToady if it's a butterfly, you have to be nice to it
sorear should've taken druthb's offer last night, this is a perfect opportunity for dancing... 16:40
labster lol, I blogged: blog.brentlaabs.com/2013/06/thankin...-yapc.html 16:41
and another checkbox down in my bingo card :)
16:41 konundra left
sorear labster: I don't see you in the room. :( 16:42
nwc10 *does* Rakudo-JVM use more memory than Rakudo on Parrot?
16:42 colomon left
nwc10 has anyone measured them? 16:42
labster waves 16:43
16:43 colomon joined
TimToady he's kitty-corner from me 16:43
flussence 152MB for java, 67MB for parrot just at the repl
but the java one does have -Xms100m
colomon sorear: he's sitting across from my and TimToady
*me 16:45
RedditAnalytics is there a live webfeed for these presentations? 16:46
nwc10 arguably the peak is more interesting - I guess compiling the setting is the RAM stress-test
flussence (oddly enough, changing that to -Xms16m *increases* the startup RAM usage...) 16:48
and 64m takes it down to 149MB. Maybe it's memory fragmentation or something.
16:51 dakkar left
sorear RedditAnalytics: no 16:53
RedditAnalytics: the free talks were all livestreamed, but that's over
RedditAnalytics: now it's just paid training sessions and hackathons 16:54
I don't think the hackathon is being livestreamed. it wouldn't be very interesting anyway
nwc10 you could take a silly picture every hour and upload it
but I suspect that that would get boring after the second. Or maybe even before the first 16:55
16:55 SamuraiJack left, nyuszika7h left
sorear just a bunch of people sitting at laptops and saying "rutabaga rutabaga rutabaga" 16:55
16:56 sqirrel joined
sorear o/ sqirrel 16:56
16:56 groky left, nyuszika7h joined
sqirrel o/ sorear 16:57
hi all
colomon rn: say "ß".uc
camelia rakudo b2072f, niecza v24-66-gaee6525: OUTPUT«ß␤»
16:58 rafl left, avar left, Gothmog_ left
sorear .u ß 16:58
yoleaux U+00DF LATIN SMALL LETTER SHARP S [Ll] (ß)
16:58 DrEeevil left, bonsaikitten joined
colomon on second thought, I think was just reading that traditionally it was uppercased as SS, and there is a new uppercase form for it, but also that some German rules specified using it in words that were otherwise uppercase. 16:59
17:00 rafl joined
colomon rn: say "lj".uc 17:00
camelia rakudo b2072f, niecza v24-66-gaee6525: OUTPUT«LJ␤»
17:00 bonsaikitten left
sorear traditionally uppercased to SS, there is an uppercase form for it, which is pretty much never user 17:01
s/user/used/
rafl!! o/
17:01 bonsaikitten joined 17:02 Gothmog_ joined, nyuszika7h left, jfried joined 17:04 kaleem joined 17:05 nyuszika7h joined 17:06 birdwindupbird left, groky joined
colomon rn: say "ljenčariti".tc 17:06
camelia rakudo b2072f: OUTPUT«No such method 'tc' for invocant of type 'Str'␤ in block at /tmp/V4DTR25DbN:1␤␤» 17:07
..niecza v24-66-gaee6525: OUTPUT«Ljenčariti␤»
colomon n: say 'ljenčariti'.tclc
camelia niecza v24-66-gaee6525: OUTPUT«Ljenčariti␤»
colomon n: say 'ljenčariti'.tcuc 17:08
camelia niecza v24-66-gaee6525: OUTPUT«LjENČARITI␤»
17:08 avar joined, avar left, avar joined 17:09 kevi joined, dmol joined
colomon TimToady: gist.github.com/patch/5717017 17:11
17:14 Ultali joined, Ultali is now known as Ulti 17:15 japhb_ joined
pmichaud r: my sub &*USAGE() { say "HELLO"; }; &*USAGE(); 17:16
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Missing block␤at /tmp/foA7qngGHV:1␤------> my sub ⏏&*USAGE() { say "HELLO"; }; &*USAGE();␤ expecting any of:␤ scoped declarator␤ new name to be defined␤»
pmichaud std: my sub &*USAGE() { say "HELLO"; }; &*USAGE();
camelia std ea24a29: OUTPUT«ok 00:00 47m␤»
pmichaud std: my sub &USAGE() { say "HELLO"; }; &USAGE(); 17:17
camelia std ea24a29: OUTPUT«ok 00:00 47m␤»
pmichaud r: my sub &USAGE() { say "HELLO"; }; &USAGE();
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Missing block␤at /tmp/VVQKeIVCMR:1␤------> my sub ⏏&USAGE() { say "HELLO"; }; &USAGE();␤ expecting any of:␤ scoped declarator␤ new name to be defined␤»
sorear implemented a patch for bug o' doom; testing now
pmichaud r: my &*USAGE = sub { say "Hello"; }; &*USAGE(); 17:18
camelia rakudo b2072f: OUTPUT«Hello␤»
17:18 btyler left
lizmat labster++ 17:19
pmichaud r: my &*USAGE = sub CORE_USAGE { say "Hello"; };
camelia rakudo b2072f: ( no output )
pmichaud CORE::USAGE 17:20
sorear t/nqp/24-module.t ............... ok
labster what did I do now?
sorear labster: blogd
labster oh, right. I also make origami butterflies, if anyone wants one. 17:21
colomon perl6 origami butterflies?
dalek ecs: d1f9930 | larry++ | S32-setting-library/Str.pod:
remove tcuc until someone demonstrates a use case
17:22
japhb_ Butterflies! I'm all in favor.
colomon TimToady++
dalek ast: 29a2813 | (Solomon Foster)++ | S32-str/tclc.t:
Update the tests for tclc a tad.
17:23
17:23 guru joined, guru is now known as Guest90360 17:24 Guest90360 is now known as ajr_
dalek ecza: b8f846e | (Solomon Foster)++ | lib/CORE.setting:
Remove tcuc, fix warning.

tcuc has been removed from the spec, so we remove it here as well.
17:24
17:26 stevan_ joined
colomon rn: say "hello".fc 17:27
camelia rakudo b2072f: OUTPUT«No such method 'fc' for invocant of type 'Str'␤ in block at /tmp/AHgBI6UW1N:1␤␤»
..niecza v24-66-gaee6525: OUTPUT«Unhandled exception: Unable to resolve method fc in type Str␤ at /tmp/ONXIVHs0iA line 1 (mainline @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 4502 (ANON @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 4503 (module-CORE @ 586) ␤ at /home/p6e…
colomon rn: say "hello world".worldcase 17:28
camelia niecza v24-66-gaee6525: OUTPUT«Unhandled exception: Unable to resolve method worldcase in type Str␤ at /tmp/GPdmX6lOqQ line 1 (mainline @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 4502 (ANON @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 4503 (module-CORE @ 586) ␤ at /h…
..rakudo b2072f: OUTPUT«No such method 'worldcase' for invocant of type 'Str'␤ in block at /tmp/k7rAcWrBOK:1␤␤»
colomon rn: say "hello world".wordcase
camelia rakudo b2072f, niecza v24-66-gaee6525: OUTPUT«Hello World␤»
timotimo worldcase :) 17:29
17:30 quietfanatic left
colomon n: say 'a' ∈ <a b c d> # should this work? 17:30
camelia niecza v24-66-gaee6525: OUTPUT«True␤»
TimToady I don't see a problem with it offhand 17:34
it doesn't seem to be a blockers to sets of sets
*er
17:35 kevi left
colomon (talking here, but for future reference) My question was basically whether it should do 17:36
'a' ∈ set <a b c d> 17:37
or 'a' ∈ <a b c d>.set
sorear jnthn: TEST_JOBS for nqp. Resonable?
colomon (since those two things are different now) 17:39
TimToady n: say set <a b c d>
camelia niecza v24-66-gaee6525: OUTPUT«set(a, b, c, d)␤»
TimToady n: say <a b c d>.set
camelia niecza v24-66-gaee6525: OUTPUT«set(a, b, c, d)␤»
TimToady are those really different?
sorear jnthn: The hell?
jnthn: it's not t/nqp/60-bigint.t that's broken.
17:39 _sri joined
sorear it's all the tests, 2% of the time 17:40
jnthn What?!
colomon TimToady: there's not different in current source. I have a branch where they are different.
TimToady: as I understand it, set <a b c d> should be Set.new(<a b c d>) should be a Set containing one element which is <a b c d> 17:41
TimToady <a b c d>.Set should be a single element 17:42
or am I misremembering
colomon TimToady: I don't believe that's spec'd at all.
unless you're thinking of that as distinct from [a, b, c, d].Set 17:43
jnthn sorear: What's the failure mode, ooc? 17:44
sorear jnthn: I get different failures every time I run make test
jnthn sorear: If you run the same thing over and over it fails? 17:45
(eventually)
17:45 zakalwe joined
sorear jnthn: I think so? I haven't gotten a random failure outside 'prove' 17:45
jnthn sorear: Maybe prove -v will give some info on exactly what the failure looks like... 17:46
At least we know your fix isn't to blame :) 17:47
sorear jnthn: yeah, just ran make test-loud | tee
jnthn presumes the sporadic failures happen even without the patch 17:48
sorear I haven't tested that. I suppose I ought to
jnthn has never seen these sporadic failures, fwiw 17:49
17:49 Chillance joined
sorear is running two attempts simultaneously to keep two cores busy 17:49
dalek kudo/nom: 57e740a | jonathan++ | src/Perl6/ModuleLoader.nqp:
Fix a module loader regression.

Merge fail when bringing in the jvm-support branch.
timotimo jnthn: cool, that should fix PSGI 17:50
jnthn Hope so
colomon: 57e740a should hopefully fix the issue you reported
colomon jnthn++
17:51 zakalwe left 17:52 zakalwe joined
colomon rn: say set [1, 2, 3] 17:55
camelia rakudo b2072f, niecza v24-66-gaee6525: OUTPUT«set(1, 2, 3)␤»
colomon rn: say set { a=>1, b=>2 } 17:56
camelia rakudo b2072f, niecza v24-66-gaee6525: OUTPUT«set(a, b)␤»
tadzik colomon: oh. THat's bad
colomon: I'll take a looks
colomon tadzik: very, very bad. :) 17:57
tadzik: but it's a simple-ish fix, I think.
TimToady yes, that's not right
colomon TimToady: right.
should be a set with two pairs
17:57 jeffreykegler left
TimToady simplify the constructor to not try to outsmart the slurpy 17:57
colomon TimToady: should be set(a=>1, b=>2), right? 17:58
TimToady uh, {} doesn't flatten, just as [] doesn't, so set({..})
tadzik oh, pull request already :) colomon++
TimToady a set with a single elem, which is a hash
colomon tadzik: that only fixing smoker, it doesn't fix the underlying panda issue
TimToady if the sig for set uses ** instead of *, it needs to change to * 17:59
dalek mentaler: 46eada4 | (Solomon Foster)++ | smoker:
Every exception is a failure, not just the carefully constructed Panda exceptions.
mentaler: fa1f1e4 | tadzik++ | smoker:
Merge pull request #1 from colomon/master

Do no allow any exceptions to slip by smoke test
[Coke] sorear: if it's on feather3 (not feather), then I suppose we could end up with an r-j: eval at some point. hokay.
colomon tadzik: basically, I changed smoker to signal failure no matter what exception it gets
tadzik colomon: yeah, that makes sense
sorear ok, *that* test-loud run finished with no spurious errors
colomon TimToady: right now it takes apart the hashes internally
pmichaud anyone interested in lunch? 18:00
jnthn pmichaud: yes
[Coke] rolls his eyes at people complaining about a sponsor on the mailing list. perhaps we need a rate for attendees that allows them to avoid seeing any advertisements anywhere.
colomon should Set.new([1, 2, 3]) also be set(1, 2, 3)
?
TimToady: ^ 18:01
pmichaud yes, I was thinking something similar. But I didn't want to stir the pot further anyway.
TimToady no
set is just short for Set.new
[Coke] ... and then duke goes over the top. *sigh*
TimToady both should make set( [1,2,3] ) with one elem
18:01 PacoAir left
jnthn pmichaud: Any thoughts on what/where? 18:02
sorear pmichaud: I would not mind lunch
pmichaud jnthn: I'm open for pretty much anything. Don't know what's nearby.
jnthn heh, me eihter ;)
TimToady would mind !lunch
RedditAnalytics I've made some changes and released my first alpha to prod -> www.redditanalytics.com/search/ it's running more Perl6 now
sorear oo 18:03
t/nqp/60-bigint.t ...............
Dubious, test returned 1 (wstat 256, 0x100)
No subtests run
18:03 PacoAir joined
jnthn Hm, so it doesn't even get to running 18:04
TimToady but !lunch is what happens by default :)
colomon TimToady: so far so good? gist.github.com/colomon/5723570
colomon wonders if any place delivers pizza to this building
TimToady implies pizza is !lunch 18:05
sorear Hehe
:D
TimToady yes, so far so good
pmichaud lunch eateries listed at www.yapcna.org/yn2013/wiki?node=LunchEateries 18:06
18:06 sivoais_ left 18:07 sivoais joined
colomon TimToady: two more: gist.github.com/colomon/5723570 18:07
and more now 18:08
TimToady my %h = { a=>1, b=>2 }; is non-sensical 18:10
18:10 domidumont left
colomon rn: my %h = a=>1, b=>2; say %h 18:10
camelia niecza v24-67-gb8f846e: OUTPUT«{"a" => 1, "b" => 2}␤»
..rakudo b2072f: OUTPUT«("a" => 1, "b" => 2).hash␤»
colomon TimToady: fixed
18:11 atroxaper left
TimToady set %h ==> set(a, b) is wrong, set %h.keys ==> set(a, b) 18:11
rn: my %h = a=>1, b=>2; print %h, "\n" 18:12
camelia rakudo b2072f, niecza v24-67-gb8f846e: OUTPUT«a 1b 2␤»
dalek ecs: 2f8ec12 | labster++ | S32-setting-library/Str.pod:
add IO and path methods to S32::Str
colomon TimToady: changed it: gist.github.com/colomon/5723570 18:13
18:14 kaleem left, konundra joined, PacoAir left 18:15 thou_ left, thou joined
jnthn sorear: d'oh...I may have found it 18:16
18:18 FROGGS left 18:20 PacoAir joined
colomon pmichaud, jnthn: I don't care where we go. 18:21
pmichaud I think we've settled on Ethiopian again 18:22
jnthn yays :)
pmichaud (walking distance, fast service, reasonable noms)
dalek p: 6293f05 | sorear++ | src/NQP/ (2 files):
Move $?PACKAGE and $?CLASS to the beginning of packages so that they are found correctly on JVM. Fixes t/nqp/24-module.t
18:23
18:26 lizmat left
FOAD Fried grasshoppers? 18:30
18:32 japhb_ left, domidumont joined 18:43 Kelder joined, Ayiko left 18:49 stevan_ left
dalek : 2b372cc | (Tobias Leich)++ | / (5 files):
Perl5::World is a class now rather a role

So Perl5::World doesnt get mixed in $*W, which wasnt working right in evals. Also added `import`-noop-token. I am not sure how to do require/import yet.
18:57
18:57 FROGGS joined
FROGGS I start to hate wifi 18:57
19:08 kaare_ joined 19:09 bluescreen10 joined 19:12 kaare__ left
moritz nqp-jvm segfaults while compiling src/stage1/gen/QAST.nqp :( 19:16
19:17 grondilu joined
benabik JVM segfault? Nice. 19:17
FROGGS has somebody recognized a slowdown when using nqp to compile stuff? 19:18
19:19 btyler joined 19:21 domidumont left 19:22 pecastro left 19:27 DreamingInCode joined, pecastro joined 19:30 DreamingInCode is now known as abnorman 19:31 xerox_ left
moritz (that's openjdk, iirc 19:31
yes, it is 19:32
labster lunch crowd is back 19:35
19:35 twigel joined
labster FROGGS: you're only now starting to hate the wifi here? 19:36
19:37 lizmat joined 19:38 japhb_ joined
FROGGS well, the first days I only had the re-enter the coupon once a day, now I have to do this every our or so, and need turn off/on my wifi 19:39
brb
19:39 FROGGS left 19:40 FROGGS joined 19:41 rindolf left, rindolf joined
sorear moritz: jnthn, yesterday: "You need 1.7.x. If it segfaults it means you need a newer patchlevel - we hit a bug in invokedynamic." 19:41
s/yesterday/yesterhackathon/
moritz and now my firefox segfaulted too. 19:43
FROGGS I get a segfault when building nqp/master/HEAD 19:44
parrot 5.2.0-devel is right?
sorear JOY
jnthn FROGGS: yes. Does it explode at the P5Regex stage at the end? If so, it's typically 'cus it finds a wrong installed ModuleLoader.pbc 19:45
sorear
.oO( no, we're using the JVM now )
Util++ # random acts of kindness
dalek p: 255a8bd | jonathan++ | src/vm/jvm/ (2 files):
Fix intermitent startup failure.

Boiled down to an accidental reliance on the order reflection gave back an array of methods.
19:46
jnthn sorear: Hopefully 255a8b helps
sorear jnthn: danke, testing
FROGGS src/stage1/gen/module_loader.pir 19:47
make: *** [src/stage1/nqpmo.pbc]
jnthn: so yes, module_loader stuff
jnthn FROGGS: um, no, that's not the failure I meant 19:48
FROGGS :/ 19:49
moritz installs an openjdk-7 from debian testing
19:50 bluescreen100 joined
colomon rn: say bag a=>1, b=>2; 19:50
camelia niecza v24-67-gb8f846e: OUTPUT«Unhandled exception: Excess arguments to bag, unused named a, b␤ at /home/p6eval/niecza/lib/CORE.setting line 0 (bag @ 1) ␤ at /tmp/o4Bcv12WmD line 1 (mainline @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 4499 (ANON @ 3) ␤ at /home/p6eval/niecza…
..rakudo b2072f: OUTPUT«2 unexpected named parameters passed (a, b)␤ in sub bag at src/gen/CORE.setting:13393␤ in block at /tmp/cMRYHyXvNR:1␤␤»
colomon rn: say bag (a=>1, b=>2);
camelia rakudo b2072f, niecza v24-67-gb8f846e: OUTPUT«bag("a" => 1, "b" => 2)␤»
colomon TimToady: again... gist.github.com/colomon/5723570 19:51
TimToady: except .Set instead of .set
(now corrected) 19:52
dalek : 8847b02 | (Tobias Leich)++ | rakudo.patch:
fudge patch for current rakudo
19:53
ecs: e85286e | pmichaud++ | S17-concurrency.pod:
[S17]: Add references to hyperops, feeds, and junctions.
19:53 bluescreen10 left 19:59 gudaht left
moritz labster++ # blog.brentlaabs.com/2013/06/thankin...-yapc.html 20:01
20:02 raiph left
labster Yeah, I wish you had been here, moritz 20:02
20:03 btyler left, btyler joined
labster r: X::NYI.new.say 20:05
camelia rakudo b2072f: OUTPUT«use of uninitialized value of type Any in string context in block at /tmp/xggYlAOqio:1␤␤ not yet implemented. Sorry. ␤␤Error while creating backtrace: No such method 'backtrace' for invocant of type 'Any'␤ in method new at src/gen/CORE.setting:8731␤ in method n…
20:06 kaare_ left
TimToady
.oO(A singleton object is only half as good as a doubleton object.)
20:06
labster .oO (When are we going to implement uninitialized in string context )
20:08 logie joined
moritz labster: I'm hopeful to make YAPC::EU 20:08
at least
20:09 btyler_ joined
labster I'm not hopeful for myself, but I hope you have a good time 20:09
moritz it's been way too long (Oslo Hackathon last year) since I last met some of #perl6 crowd in meat space
anyway, time for sleeping here 20:10
jnthn 'night, moritz
20:10 btyler left
colomon \o 20:10
japhb_ o/ moritz
sorear o/ 20:11
20:13 btyler_ left, btyler joined 20:21 quietfanatic joined 20:24 snearch joined
colomon TimToady: I've added some more: gist.github.com/colomon/5723570 20:24
TimToady: any thoughts on what bag %h should do? 20:25
labster r: any and all 20:26
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Undeclared routine:␤ and used at line 1. Did you mean '&rand', '&end', '&any'?␤␤»
labster 'and' not declared? that's a new one. 20:27
[Coke] r: any() and all() 20:32
camelia rakudo b2072f: ( no output )
20:33 stevan_ joined 20:39 tomyan joined 20:40 snearch left
FROGGS jnthn: I'm bisecting my build problem right now 20:41
jnthn ok
FROGGS must be somewhere within the last two weeks or so
20:42 sqirrel left 20:45 s1n1 left
lizmat r: class A is Hash {}; my $a=A.new; say $a.WHAT # shouldn't this have to say "(A)" 20:45
camelia rakudo b2072f: OUTPUT«(Hash)␤»
jnthn lizmat: Yes but we already know Hash.new is busted 20:46
colomon TimToady: tried to get some of it down... gist.github.com/colomon/5723570
lizmat jnthn: so what would be a quick fix / workaround ?
not something with BUILD I presume, buut rather with bless ? 20:47
jnthn Right
r: class A is Hash { method new() { self.bless(*) } }; my $a=A.new; say $a.WHAT
camelia rakudo b2072f: OUTPUT«(A)␤»
20:48 tomyan left
lizmat jnthn++! 20:48
colomon rn: say bag a=>1, b=>30 20:49
camelia rakudo b2072f: OUTPUT«2 unexpected named parameters passed (a, b)␤ in sub bag at src/gen/CORE.setting:13393␤ in block at /tmp/H9rEnf9ju6:1␤␤»
..niecza v24-67-gb8f846e: OUTPUT«Unhandled exception: Excess arguments to bag, unused named a, b␤ at /home/p6eval/niecza/lib/CORE.setting line 0 (bag @ 1) ␤ at /tmp/Ja41vqNiBa line 1 (mainline @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 4499 (ANON @ 3) ␤ at /home/p6eval/niecza…
sorear jnthn: hey. deletepos is indeed implemented only in src/vm/parrot; it is used only in nqp's own test suite, not used by rakudo at all 20:52
jnthn sorear: oh, hm
sorear could kill it I guess 20:53
jnthn I think Pm may know something of the history of this
FROGGS a history lesson \o/
jnthn Let's ask when he returns from $errand
sorear ok
seems the 59-nqpop fails are exclusively due to missing deletepos 20:55
20:57 zby_home left
tadzik heh, I discovered that too when I was implementing deletepos for nqp/jvm :) 20:57
20:58 _ilbot joined
sorear tadzik: there is no nqp::deletepos for jvm in my branch 20:58
have yu pushed?
don't want to dup here
tadzik sorear: I've probably opened a pull request
jnthn hm, does that mean there's a tadzik patch somewhere?
tadzik github.com/jnthn/nqp-jvm-prep/pull/14 :)
20:58 tgt joined
FROGGS jnthn: this one has borken it: [609c1f707cf1ef0bb708eedc8f735d255a3ef06e] Fix uninitialized memory access 20:58
tadzik old times 20:59
I don't think it's worth being there, if we got away without it for so long
21:00 s1n joined 21:02 _ilbot left 21:04 japhb_ left
jnthn FROGGS: Huh...that was just making sure some memory was zeroed 21:05
FROGGS jnthn: maybe is a bit over-reacting 21:06
it* is
you can have my laptop when you want and have time for it 21:07
right now I checked out one commit before that and can work now...
jnthn Hmm
That's really odd 21:08
Do you have a backtrace
?
FROGGS no
it is some sort of coredump
dunno the error message in english
pmichaud back again 21:09
21:09 konundra left 21:12 skids left
dalek : e09ea83 | (Tobias Leich)++ | Makefile:
pull in rakudos makefile changes
21:18
21:20 grondilu left
sorear jnthn: remove nqp::deletepos from src/vm/parrot for consistency? 21:21
jnthn If it's unused, +1
21:21 adu left
jnthn uh, which you told me it was earlier :) 21:21
tadzik I think we verifed that 4 months ago :P
21:22 konundra joined
sorear should I try to build nqp-parrot to see if I botched the removal? 21:25
colomon TimToady: should "a".Set be the same as set "a" ? 21:26
pmichaud should be the same as Set.new("a") 21:27
dalek p: f40d6de | sorear++ | / (3 files):
Remove nqp::deletepos

Per conversation at hackathon, we've decided that nqp::deletepos is not needed as an op. It has insufficient value-add compared to nqp::bindpos(nqp::null()).
colomon pmichaud: which is set "a". :)
pmichaud colomon: okay... it just depends on which you feel is more fundamental. :)
21:27 spider-mario left
pmichaud we're having trouble with hashes at the moment because Hash.new() is defined as "hash ..." instead of vice-versa. 21:28
colomon pmichaud: or how lazy the person typing is.
pmichaud: ah, ick.
pmichaud in general we consider the method forms to be fundamental and the sub forms to be built in terms of methods
21:29 mucker left
tadzik hm, I thought deletepos also shifted the elements. Whatever, it's history now :) 21:29
pmichaud it can shorten the array, yes. but it doesn't shift.
wait, let me rephrase :-) 21:30
tadzik shift as in "bit shift :)"
pmichaud Parrot's deletepos may have shifted elements.
tadzik it did
it was a loop assigning [i] = [i+1]
pmichaud if so, then that's something that can be correct for Parrot, but it's not useful to NQP or Rakudo.
tadzik or rather: it is
that's probably the reason why it was unused 21:31
pmichaud so NQP doesn't need an equivalent deletepos
21:31 rindolf left
sorear not sufficiently useful to optimize the n=1 case of splice eh 21:31
pmichaud not only that, but splice is likely more efficient :)(
:)
sorear :) 21:32
sorear now building for parrot
what 21:33
make: *** [src/stage1/nqpmo.pbc] Segmentation fault: 11 21:34
jnthn wtf 21:35
flussence ...have you tried taking your RAM out and blowing on it?
colomon rn: say "a" !~ Pair 21:36
camelia niecza v24-67-gb8f846e: OUTPUT«===SORRY!===␤␤Unsupported use of !~ to do negated pattern matching; in Perl 6 please use !~~ at /tmp/XPKcrpzLtr line 1:␤------> say "a" !~ ⏏Pair␤␤Parse failed␤␤»
..rakudo b2072f: OUTPUT«===SORRY!===␤Unsupported use of !~ to do negated pattern matching; in Perl 6 please use !~~␤at /tmp/9qU_jMRs9G:1␤------> say "a" !~ ⏏Pair␤»
sorear flussence: it's a macbook air, the RAM is soldered in place
colomon ah
colomon missed that error message in his backlog
21:37 dmol left
dalek p: 84d8240 | jonathan++ | examples/load_bytecode.nqp:
Toss useless example.
21:38
21:43 lizmat_ joined
dalek rl6-bench: 008f005 | (Geoffrey Broadwell)++ | / (2 files):
Add --verbose flag to timeall that prints extra info about current testing state
21:44
rl6-bench: 1e65b59 | (Geoffrey Broadwell)++ | timeall:
Catch requests to time compiler variants that don't have an entry in the compilers list
rl6-bench: 9874602 | (Geoffrey Broadwell)++ | timeall:
Fix a braino the swapped substitutions in two places; add support for magic DATA directory; show when a text is being skipped
rl6-bench: dc4cf46 | (Geoffrey Broadwell)++ | data/ (3 files):
Add data/ directory with a README, the Panda projects.json file, and an American English word list
rl6-bench: 564d244 | (Geoffrey Broadwell)++ | timeall:
Add more verbosity to verbose timeall output
21:45 japhb_ joined
diakopter arnsholt: where u? 21:46
21:46 lizmat left
lizmat_ r: my %h=a=>1,b=>2; say %h; say %h<a b>:p:!delete; say %h # ARGH, this works, the pb is in the test suite 21:48
yoleaux 28 May 2013 16:27Z <jnthn> lizmat_: I wonder if the Mu \key shouldn't be TKey \key
camelia rakudo b2072f: OUTPUT«("a" => 1, "b" => 2).hash␤"a" => 1 "b" => 2␤("a" => 1, "b" => 2).hash␤»
21:48 lizmat_ is now known as lizmat
colomon sorear: how did you do gather/take in Niecza? (general approach all that I'm curious about at the moment.) 21:49
21:50 tgt left
colomon TimToady: what stupid thing am I doing here? gist.github.com/colomon/5725313 21:53
TimToady: or at least, what stupid thing is making it not parse with an obscure message. 21:54
sorear colomon: by writing a stackless VM and switching the soft stacks
colomon sorear: thanks
sorear look at Kernel.Take and Kernel.CoTake, it's relatively straightforward 21:55
colomon sorear++
sorear much more straightforward than the approach jnthn++ invented for nqp-jvm
tadzik sorear: did you manage to fix the segfault?
I get it here too 21:56
sorear but has the disadvantage of slowing down all code and significantly complicating coding
tadzik my ram isn't soldered, but I very much doubt it's ram's fault
sorear tediating, really. writing CPS code is not hard, just seriously annoying
tadzik: no idea, but it crashes in the same way with a completely clean segfault so I'm sure it's not a hardware fault 21:57
a hardware fault would appear nondeterministic, because even if that physical byte is deterministically broken, the virtual memory mapping is random
tadzik completely clean segfault :) 21:58
yeah, cleaning doesn't help me either
well, I'll have to stick to rakudo/jvm then :) 22:01
22:02 lustlife1 left 22:04 lustlife left
dalek p: ede50c1 | jonathan++ | src/vm/parrot/6model/reprs/P6opaque.c:
Fix a thinko.
22:08
ast: b38771a | (Elizabeth Mattijsen)++ | S32-hash/delete-adverb.t:
It was the test for :p:!delete that was wrong. Duh
22:10 bluescreen100 left
arnsholt Derp. Sorry 'bout that >.< 22:10
22:12 stevan_ left
FROGGS nqp HEAD builds \o/ 22:12
dalek p: cd4197e | pmichaud++ | t/parrot/05-vtable.t:
Update plan for parrot/05-vtable.t .
22:15
kudo/nom: f737ff7 | labster++ | src/core/IO.pm:
move copy from IO::Handle to IO::Path, deprecate IO::Handle version
22:17
sorear how crazy would it be to add a mode to nqp which runs all the tests in the same JVM? 22:18
colomon TimToady, pmichaud: Do these look okay? gist.github.com/colomon/5725455
[Coke] I would be concerned about globals.
(how do you know that you have a clean state after a test completes, e.g.) 22:19
dalek ast: 060d569 | (Solomon Foster)++ | S02-types/set.t:
Some basic .Set tests.
22:23
sorear now beating on t/qast/01-qast.t 22:24
colomon [Coke]: if that was directed at me, how could I test for unclean state and globals?
dalek rl6-bench: 5bb409d | (Geoffrey Broadwell)++ | microbenchmarks.pl:
Convert some exit() calls to nqp::exit() for nqp microbenchmarks; remove nqp-moarvm skip for several microbenchmarks
22:25
pmichaud colomon: checking the roast commit 22:27
colomon: roast commit looks okay. I probably wouldn't have used showset(), but rather tested each element individually. :-) 22:31
colomon pmichaud: that would make set.t very long... :)
pmichaud: plus the count of elements? 22:32
(I mean, also testing the count.)
pmichaud is [ a=>3, b=>0, 'c', 'a' ].Set<a b c>, "True False True"; 22:33
is [ a=>3, b=>0, 'c', 'a' ].Set.<a b c>, "True False True";
(if that works at all... reading spec)
22:35 benabik left
colomon pmichaud: that does not test to make sure there are not other things in the set... but it's an interesting approach. 22:36
pmichaud is [ a=>3, b=>0, 'c', 'a'].Set.count, 2
22:36 sunnavy left, hypolin left, Util left, ruoso left, scottp left, awwaiid left, masak left, Nom- left, aghbas_ left, snarkyboojum_ left, estrabd_ left, LordV left, cooper left, arlinius left
pmichaud er, .elems 22:36
22:37 PacoAir left 22:38 LordV joined, Util joined, ruoso joined 22:39 masak joined, awwaiid joined 22:45 colomon left
dalek ast: b0f6ccb | (Solomon Foster)++ | S02-types/ (2 files):
Some basic .KeySet and .Bag tests.
22:45
22:45 japhb_ left 22:48 arlinius joined, sunnavy joined, hypolin joined, scottp joined, Nom- joined, snarkyboojum_ joined, aghbas_ joined, estrabd_ joined, cooper joined 22:50 btyler left 22:52 labster left 22:53 FROGGS left 22:54 tomyan joined 22:55 lizmat left 22:57 btyler joined 22:58 tgt joined, pecastro left 23:01 sunnavy left, hypolin left, scottp left, Nom- left, aghbas_ left, snarkyboojum_ left, estrabd_ left, cooper left, arlinius left 23:04 arlinius joined, sunnavy joined, hypolin joined, scottp joined, Nom- joined, snarkyboojum_ joined, aghbas_ joined, estrabd_ joined, cooper joined 23:09 raiph joined 23:20 estrabd joined 23:22 estrabd_ left 23:30 konundra left
cognominal I am watching TimToady's talk and apparently there is a COMPOSE phaser. What gives? 23:33
"there is only one way to do objects until you know better" :) 23:37
il parle des CoCs :) 23:39
lue S02:4347 should surely say "=begin END" and "=END" instead of "=begin finish" and "=finish", no? 23:40
23:41 ajr_ left, tgt left
cognominal oops wrong channel again for the last two posts :( 23:41
23:42 tomyan left
lue aw, it's actually the other way around. Apparently someone did s/=END/=finish/ at some point. 23:43
dalek rl6-roast-data: 7d7b8ee | coke++ | / (4 files):
today (automated commit)
lue prefers =END because it's shorter and the all-caps makes it feel more special, which its behavior is.
dalek kudo-star-daily: 8bd5a01 | coke++ | log/ (5 files):
today (automated commit)
23:46 btyler left, konundra joined
[Coke] r: COMPOSE * 23:47
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Undeclared name:␤ COMPOSE used at line 1␤␤»
dalek ecs: 26bc31d | lue++ | S02-bits.pod:
[S02] Updated table in "Double-underscore forms".

  =END is actually =finish now and =DATA is actually =data (this update
was already in the following paragraphs, but the table was neglected, which this fixes).
23:48
lue std: COMPOSE *
camelia std ea24a29: OUTPUT«ok 00:00 42m␤»
timotimo btw: random cool thing i found: 23:53
r: class Player { has $.color = <red green blue yellow pink potato>[(state $i)++]; }; say Player.new xx 6;
camelia rakudo b2072f: OUTPUT«Player.new(color => "red") Player.new(color => "green") Player.new(color => "blue") Player.new(color => "yellow") Player.new(color => "pink") Player.new(color => "potato")␤»
timotimo r: sub test($foo = <red green blue yellow pink potato>[(state $i)++]) { say $foo }; say test xx 3; say test("bar"); say test("baz"); say test("quux"); say test xx 2; 23:54
camelia rakudo b2072f: OUTPUT«===SORRY!===␤Undeclared routine:␤ xx used at line 1␤␤»
timotimo r: sub test($foo = <red green blue yellow pink potato>[(state $i)++]) { say $foo }; say test() xx 3; say test("bar"); say test("baz"); say test("quux"); say test() xx 2;
camelia rakudo b2072f: OUTPUT«red␤green␤blue␤True True True␤bar␤True␤baz␤True␤quux␤True␤yellow␤pink␤True True␤»
timotimo er, huh? 23:55
oh, of course