Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2011-09-14

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:08 replore_ joined #perl6
00:15 dukeleto joined #perl6
00:27 am0c joined #perl6
00:30 jevin joined #perl6
00:54 jevin joined #perl6
01:08 benabik joined #perl6
01:19 woosley joined #perl6
01:27 wolfman2000 joined #perl6
01:28 bluescreen10 joined #perl6
01:49 aindilis joined #perl6
01:59 woosley joined #perl6
02:01 frhodes joined #perl6
02:10 pernatiy joined #perl6
02:34 satyavvd joined #perl6
02:58 sorear good * #perl6
03:09 sorear jnthn: what is SAFE.setting?
03:17 TimToady nom: unlink 'perl6'
03:17 p6eval nom cca116: OUTPUT«unlink is disallowed in safe mode␤  in sub restricted at src/SAFE.setting:2␤  in sub unlink at src/SAFE.setting:6␤  in <anon> at /tmp/vt6IPO16ma:1␤  in <anon> at /tmp/vt6IPO16ma:1␤␤»
03:18 TimToady nom: CORE::unlink 'foo'
03:18 p6eval nom cca116: OUTPUT«Could not find symbol 'CORE::&unlink'␤  in <anon> at /tmp/xIhYMNPJZE:1␤  in <anon> at /tmp/xIhYMNPJZE:1␤␤»
03:18 benabik nom: CORE::say 'foo'
03:18 p6eval nom cca116: OUTPUT«Could not find symbol 'CORE::&say'␤  in <anon> at /tmp/1_a1LaEcFb:1␤  in <anon> at /tmp/1_a1LaEcFb:1␤␤»
03:19 TimToady nom: &CORE::say('foo')
03:19 p6eval nom cca116: OUTPUT«Method 'postcircumfix:<( )>' not found for invocant of class 'Any'␤  in <anon> at src/gen/Metamodel.pm:3023␤  in <anon> at /tmp/b04uTOjcgT:1␤  in <anon> at /tmp/b04uTOjcgT:1␤  in <anon> at /tmp/b04uTOjcgT:1␤␤»
03:19 TimToady nom: CORE::{'&say'}('foo')
03:19 p6eval nom cca116: OUTPUT«Could not find sub &CORE␤  in <anon> at /tmp/9KUhwoBz1X:1␤  in <anon> at /tmp/9KUhwoBz1X:1␤␤»
03:20 benabik CORE NYI
03:20 TimToady good thing, in this case ;)
03:22 TimToady niecza: only infix:<+>(Numeric $a, Numeric $b) { CORE::{'&infix:<+>'}($a,$b) }; say 1 + 1; say '1' + 1
03:22 p6eval niecza v9-23-g1a8efca: OUTPUT«2␤Unhandled exception: Nominal type check failed in binding Numeric $a in MAIN infix:<+>; got Str, needed Numeric␤  at /tmp/Sv5upAAz9h line 0 (MAIN infix:<+> @ 0) ␤  at /tmp/Sv5upAAz9h line 1 (MAIN mainline @ 3) ␤  at /home/p6eval/niecza/lib/CORE.setting li…
03:24 TimToady perl6: multi infix:<+>($a,$b) { die "Can't coerce" }; say 1 + 1; say '1' + 1
03:24 p6eval rakudo cca116: OUTPUT«2␤Ambiguous dispatch to multi 'infix:<+>'. Ambiguous candidates had signatures:␤:(Any $a, Any $b)␤:(Any $a, Any $b)␤␤  in sub infix:<+> at src/gen/CORE.setting:1865␤  in <anon> at /tmp/sOSYpoFYeO:1␤  in <anon> at /tmp/sOSYpoFYeO:1␤␤»
03:24 p6eval ..pugs: OUTPUT«2␤*** Can't coerce␤    at /tmp/f_rBZGmeWp line 1, column 63 - line 2, column 1␤       /tmp/f_rBZGmeWp line 1, column 26-45␤»
03:24 p6eval ..niecza v9-23-g1a8efca: OUTPUT«Potential difficulties:â�¤  $a is declared but not used at /tmp/BEEYkiR9fw line 1:â�¤------> [32mmulti infix:<+>([33mâ��[31m$a,$b) { die "Can't coerce" }; say 1 + 1[0mâ�¤  $b is declared but not used at /tmp/BEEYkiR9fw line 1:â�¤------> [32mmulti infix:<+>($a,[…
03:24 TimToady perl6: multi infix:<+>($,$) { die "Can't coerce" }; say 1 + 1; say '1' + 1
03:24 p6eval rakudo cca116: OUTPUT«2␤Ambiguous dispatch to multi 'infix:<+>'. Ambiguous candidates had signatures:␤:(Any $a, Any $b)␤:(Any, Any)␤␤  in sub infix:<+> at src/gen/CORE.setting:1865␤  in <anon> at /tmp/A0PYjkFNCI:1␤  in <anon> at /tmp/A0PYjkFNCI:1␤␤»
03:24 p6eval ..pugs: OUTPUT«*** ␤    Unexpected "$,$)"␤    expecting formal parameter or ")"␤    at /tmp/iJwJvS3NaG line 1, column 17␤»
03:24 p6eval ..niecza v9-23-g1a8efca: OUTPUT«Unhandled exception: Can't coerce␤  at /home/p6eval/niecza/lib/CORE.setting line 742 (CORE die @ 2) ␤  at /tmp/lndK_tGOaL line 1 (MAIN infix:<+> @ 2) ␤  at /tmp/lndK_tGOaL line 1 (MAIN mainline @ 1) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2046 (CORE…
03:31 TimToady looks like pugs does the best, as long as you name your params
03:32 envi_laptop joined #perl6
03:32 TimToady looks like niecza is calling a less specific inner multi in preference to a more specific outer multi
03:33 sorear TimToady: actually the issue is that niecza's CORE::infix:<+> is currently an only sub
03:33 TimToady ah
03:36 TimToady rakudo's behavior makes me want to add a tiebreaking rule where otherwise ambiguous dispatch prefers an inner lexical scope to an outer
03:40 TimToady then we don't have to calculate whether two sigs are identical to do multi shadowing
03:41 TimToady they only have to come out tied
03:41 TimToady jnthn: thoughts on ^^
03:41 diakopter math question for the channel: what's the length of the shortest string that contains all words of length Y from an alphabet of size X
03:44 sorear diakopter: for X=2, 2^Y-1 if you make the string circular
03:44 diakopter can't be circular
03:44 sorear and the solution takes the form of a maximal-length LFSR
03:45 sorear then I guess 2^Y+Y-1
03:45 diakopter X^Y+Y-1 ?
03:46 shinobicl_ joined #perl6
03:46 abercrombie joined #perl6
03:48 PerlJam Does that take into account overlapping words?
03:48 diakopter I don't know whether your guess is correct; 2nd question: how many unique strings satisfy those conditions?
03:48 sorear PerlJam: yes
03:48 sorear I have very little clue about the X != 2 case
03:49 sorear I'm using a result from coding theory, and those people don't consider non-binary much
03:50 diakopter 3rd task: write a program to generate those strings :)
03:50 diakopter this came about because of my garage door opener keypad, btw
03:51 PerlJam diakopter: are you trying to be the first person to cause #perl6 to say "go do your own homework!"  ;)
03:51 diakopter (it ignores extraneous presses before the correct code)
03:51 diakopter PerlJam: lol
03:51 diakopter no seriously, my garage door opener works this way
03:52 diakopter so I was trying to think of the minimum number of presses someone would have to press to guarantee opening the door
03:53 shinobicl_ hi... In 'nom', how do i define a constraint for a variable without using a subset?
03:54 cotto diakopter, what if you trigger the undocumented self-destruct function?
03:54 diakopter heh
03:55 cotto You laugh now.
03:55 diakopter oh, the secret Y^2 length code that always works
03:56 diakopter that's definitely a possibility
03:57 PerlJam all 9s, all 0s, etc.
03:57 packetknife joined #perl6
03:58 sorear diakopter: I can guararntee you that the minimum length string is *at least* X^Y+Y-1
03:58 sorear actually that's wrong
03:58 diakopter at least as small as, or at least as big as
03:58 sorear but not be enough to matter
03:58 sorear diakopter: at least as big as
03:59 sorear diakopter: a string of length X^Y+Y-1 has X^Y length-Y substrings
03:59 sorear diakopter: if it were any shorter, it would need to miss a word
03:59 diakopter I'll buy that for a dollar
04:01 diakopter maybe I'll submit the string generation part as a Rosetta Code entry
04:01 cotto Is this one of those nasty encrypted garage door openers whose manufacturers like using the DMCA to prevent interoperability?
04:02 cotto or is it just a matter of figuring out a fixed sequence
04:02 diakopter it's just like a normal push-button remote except it requires entering a preprogrammed code to send its signal to the opener
04:03 diakopter the opener has to be taught to trust it with the learning button
04:03 sorear My interpretation of diakopter's request is that he wants to know how vulnerable his house is to a brute-force attack.
04:03 diakopter just like any other remote you'd buy at Home Depot or Lowe's
04:03 sorear .oO( they could just drive through the garage door, if they're doing brute force hacking )
04:03 diakopter :D
04:04 cotto cars are an effective attack vector
04:04 cotto easy to detect though
04:04 cotto lots of side-effects
04:06 birdwindupbird joined #perl6
04:25 daniel-s joined #perl6
04:37 packetknife joined #perl6
04:41 dalek niecza: a279320 | sorear++ | lib/CORE.setting:
04:41 dalek niecza: Add mising at_pos/at_key methods to several classes with postcircumfix:*
04:41 dalek niecza: review: https://github.com/sorear/niecza/commit/a279320590
04:45 * sorear is trying to decide how much more nqp/nom-like niecza's bootstrap should become
05:07 _ilbot joined #perl6
05:07 Topic for #perl6 is now »ö« Welcome to Perl 6! | http://perl6.org/ | evalbot usage: 'perl6: say 3;' or rakudo:, niecza:, std:, or /msg p6eval perl6: ... | irclog: http://irc.perl6.org/ | UTF-8 is our friend!
05:07 sorear rakudo: say [].perl
05:07 p6eval rakudo cca116: OUTPUT«[]␤»
05:07 sorear rakudo: say Array.perl
05:08 p6eval rakudo cca116: OUTPUT«Array␤»
05:08 y3llow joined #perl6
05:08 risou_awy joined #perl6
05:08 Woodi joined #perl6
05:08 Lorn joined #perl6
05:11 diakopter joined #perl6
05:12 cotto tell jnthn Give win64-workaround another run whenever you get a chance.
05:12 cotto phenny tell jnthn Give win64-workaround another run whenever you get a chance.
05:12 cotto phenny, help
05:12 phenny cotto: Hi, I'm a bot. Say ".commands" to me in private for a list of my commands, or see http://inamidst.com/phenny/ for more general details. My owner is sbp.
05:12 sorear phenny: msg cotto This is how it's done
05:13 sorear phenny: tell cotto This is how it's done
05:13 phenny sorear: I'll pass that on when cotto is around.
05:13 cotto phenny: tell jnthn Give win64-workaround another run whenever you get a chance.
05:13 phenny cotto: I'll pass that on when jnthn is around.
05:13 phenny cotto: 05:13Z <sorear> tell cotto This is how it's done
05:13 packetknife joined #perl6
05:13 revdiablo joined #perl6
05:13 cotto stupid bots, only doing exactly what they're programmed to
05:13 revdiablo joined #perl6
05:14 sorear hehee.  we can't have our messaging bot gaining sentience now...
05:14 sorear rakudo: sub foo(@x?) { say @x.WHAT }; foo
05:14 p6eval rakudo cca116: OUTPUT«Array()␤»
05:14 cotto phenny: test the win64-workaround branch
05:15 sorear rakudo: sub foo(@x?) { say @x.defined }; foo
05:15 p6eval rakudo cca116: OUTPUT«Bool::True␤»
05:15 sorear rakudo: sub foo(@x?) { say @x.perl }; foo
05:15 p6eval rakudo cca116: OUTPUT«Array.new()␤»
05:15 sorear rakudo: sub foo(@x?) { say @x.perl; my @y; say @y.perl }; foo
05:15 p6eval rakudo cca116: OUTPUT«Array.new()␤Array.new()␤»
05:15 sorear hrmf
05:20 nsh- joined #perl6
05:21 SHODAN joined #perl6
05:28 tokuhirom joined #perl6
05:42 revdiablo joined #perl6
05:42 revdiablo joined #perl6
06:00 ingy joined #perl6
06:01 Util joined #perl6
06:02 envi_laptop joined #perl6
06:05 orafu joined #perl6
06:07 envi_laptop joined #perl6
06:12 sorear I have a BEGIN model that I think may be workable
06:16 mberends good. morning, sorear
06:19 mberends sorear: is there a part of RunMain (CLRBackend.cs) that I should concentrate on documenting, or just all of it?  I just arrived there, and it seems to reach out to so many other things.
06:21 donri joined #perl6
06:24 sorear mberends: the logical subdivisions of CLRBackend are 1. the constant pool generator 2. the JSON parser 3. the ClrOp layer (an AST for CIL) 4. the CpsOp layer (a CPS-aware AST) 5. the JSON-to-CpsOp converter in NamProcessor 6. the downcall handler
06:24 sorear mberends: but there's a significant amount of code that doesn't logically belong to any of the divisions
06:25 sorear mberends: RunMain is mostly part of #6
06:26 sorear mberends: will sleep& in 5min unless interrupted
06:26 mberends sorear: thanks, interesting, so much more to explore
06:26 mberends sorear: good night
06:29 * sorear out
06:33 mkramer joined #perl6
06:35 koban` joined #perl6
06:35 koban` left #perl6
06:40 wtw joined #perl6
06:40 REPLeffect joined #perl6
06:46 jevin joined #perl6
06:59 masak joined #perl6
06:59 masak goot mornink, #perl6.
06:59 masak phenny: tell diakopter http://www.hakank.org/comb/debruijn.cgi http://en.wikipedia.org/wiki/De_Bruijn_sequence http://www.theory.csc.uvic.ca/~cos/inf/neck/NecklaceInfo.html
06:59 phenny masak: I'll pass that on when diakopter is around.
07:01 masak before hakank provided me with all the theory, I had already gotten around to writing a brute-force solution, which seems to be doing fairly well: https://gist.github.com/1216010
07:02 eiro wesh comment ca va bien ?
07:02 eiro masak, i missed your line yesterday
07:02 masak eiro: :)
07:03 eiro i have to admit i'm not sure about the meaning of it
07:03 eiro what did you mean by "ont été la prise sur le channel"
07:05 masak "have taken over the channel".
07:06 masak this was the best that I and Google Translate could come up with together.
07:06 masak I veto'ed GT's translation "canale", because that sounded weird. but maybe that's actually correct.
07:07 masak the translation did roundtrip, which is a good sign, if not a perfect guarantee of correctness. :)
07:08 eiro haha: that's a word by word translation. "les polonais et finlandais (ont pris le controle du cannal|sont en surnombre)".
07:09 eiro canale ? is it the real spell ?
07:09 * eiro verify
07:09 eiro actually it's canal
07:09 masak ah; double fail :)
07:09 eiro yep :)
07:10 masak I can't even correctly remember the spelling I wrongly rejected :P
07:10 eiro :-)
07:12 masak well, it seems I meant "sont en surnombre" more than "ont pris le controle".
07:13 eiro it seem canal and channel have the same means (even the watercourse)
07:15 eiro vous avez "envahi le canal"
07:15 eiro je pense qu'un francais aurais dis ca.
07:16 masak "envahi"! beautiful!
07:17 eiro :)
07:17 masak Latin vādere, "to go, rush".
07:18 eiro btw: good for you: if you took over the chan, that's because you have lot of perl6 hackers :)
07:18 masak that's the idea :) hence the name.
07:18 eiro oh? where did you find ethimology of the world ?
07:19 eiro grmbl ... go back to code ... in perl5 :/
07:19 eiro some days i'm so borring by perl5 ...
07:20 cotto eiro, it could be php
07:23 eiro cotto, or python ...
07:23 eiro so yes: it could be way worse
07:24 cotto then again, you can write good and bad code in any language commonly used in production environments
07:25 eiro sure! that's not the pb.
07:25 cotto you with with what you've got
07:25 * moritz wouldn't know how to write good code in most languages
07:26 mj41 joined #perl6
07:26 moritz actually i'm not sure I know how to write good code in *any* language
07:26 cotto moritz, step 1 is to write bad code.  Step 2 is to come back to it 18 months later.
07:27 moritz I do have some code from 2007 which I recently modified
07:27 cotto I've heard of easier ways.
07:27 eiro the thing is i'm "fed up to unroll @_" (i don't know who gave this talk)
07:27 moritz and while it didn't look pretty, it was understandable and hackable
07:28 eiro :)
07:29 masak eiro: I mostly look on dictionary.com
07:31 eiro ok
07:32 masak moritz: may I recommend "The Practice of Programming" by Kernighan and Pike -- most of it may not be news to you, but it's the best no-nonsense capture of good programming principles that I've seen to date. I have it here on the desk right next to me. :)
07:33 lucs joined #perl6
07:33 cotto I like that book.
07:34 cotto I took it a little too dogmatically when I first read it, but it's great if you don't look at its recommendations as being hard laws.
07:34 jnthn mornin'
07:34 phenny jnthn: 05:13Z <cotto> tell jnthn Give win64-workaround another run whenever you get a chance.
07:35 cotto ohai.  I was just about to goto bed
07:36 jnthn cotto: At $dayjob with no Rakudo build env anyway...
07:36 jnthn Will check it this evening
07:36 eiro oh yes! Pike is one of my idols
07:36 jnthn moritz++ # applying mls++ patch
07:37 jnthn If supernovus++'s patches on socket didn't yet get merged, +1 from me on those too.
07:37 eiro (read "the good, the bad and the ulgy" about unix history and plan9: it's very interesting!)
07:37 jnthn Otherwise I do it this evening.
07:40 erez joined #perl6
07:45 jnthn TimToady: I was expecting to tie-break on lexical scope depth; I just hadn't gotten around to actually implementing it yet.
08:05 masak sorear: sometimes when some MyGrammar.parse fails in a certain way, I get "Unable to cast object of type 'Niecza.P6opaque' to type 'Cursor'." seems to me that should never happen. if it helps you, I could find out more about the exact nature of those circumstances.
08:08 daniel-s joined #perl6
08:21 * moritz tests supernovus++'s patch right now
08:22 moritz where by "test" I mean "see if it compiles" :-)
08:22 alvis joined #perl6
08:31 dalek rakudo/nom: f2857b7 | (Timothy Totten)++ | src/core/IO/Socket/INET.pm:
08:31 dalek rakudo/nom: IO::Socket::INET.accepts now returns the connected socket
08:31 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/f2857b7fc9
08:34 wamba joined #perl6
08:42 senobmeht joined #perl6
09:00 dakkar joined #perl6
09:00 wamba joined #perl6
09:09 dakkar joined #perl6
09:12 satyavvd joined #perl6
09:16 daxim_ joined #perl6
09:22 nebuchadnezzar joined #perl6
09:27 tadzik good morning morning
09:28 masak tadzik: good good, morning.
09:28 masak quick poll: in http://en.wikipedia.org/wiki/Newcomb%27s_paradox , would you go for both boxes, or just B?
09:31 tadzik B probably
09:31 skangas joined #perl6
09:32 moritz does 'running a fair random generator to determine my choice' count?
09:32 masak sure!
09:33 masak (though it feels a bit reckless)
09:34 satyavvd joined #perl6
09:34 moritz (note that this was before reading the full analysis
09:34 moritz )
09:36 dalek nqp: 8ba6001 | moritz++ | src/how/NQPClassHOW.pm:
09:36 dalek nqp: MRO calculation optimization by mls++
09:36 dalek nqp: review: https://github.com/perl6/nqp/commit/8ba60015ce
09:36 masak it's interesting to see what the precense of a Predictor does to one's willingness to think about one's choices rationally. :)
09:37 moritz thought just B sounds more sensible, once I think more about it
09:38 satyavvd_ joined #perl6
09:38 masak LessWrong has a fascinating analysis of this. he's a two-boxer.
09:38 masak his central argument should sound familiar to HPMoR readers: "Rational agents should WIN."
09:38 masak http://lesswrong.com/lw/nc/newcombs_problem_and_regret_of_rationality/
09:38 moritz but when he publishes his analysis, the predictor knows, no? :-)
09:39 masak most likely, yes.
09:40 satyavvd joined #perl6
09:41 tadzik the choice depends on a trust you put in the Predictor
09:41 masak let's assume that the Predictor is the real deal.
09:42 masak otherwise the problem sort of collapses into something much more mundane.
09:42 tadzik if he's never wrong, I'll take B
09:42 tadzik if he's a real deal, I assume his predictions are always exactly my choicnes
09:42 tadzik which makes the answer quite obvious, I suppose
09:43 tadzik picking both could feel a bit like cheating the system :)
09:44 moritz tadzik: since Omega knows you well enough to have understood that you'd arrive at that conclusion, you have now won 1Mio USD
09:44 tadzik I suppose the longish description of the Predictor in both texts is supposed to make you trust him a bit less, and to think more about the choice
09:44 tadzik moritz: true
09:44 moritz tadzik: spend them quickly before the exchange rate drops
09:44 tadzik and what is 1 M + 1 T compared to 1 M? It's almost nothing
09:45 masak that's irrelevant.
09:45 moritz it's a month of interest :-)
09:45 masak it's more money.
09:45 tadzik exactly
09:45 moritz well, less than a month, most likely
09:45 tadzik maybe there's a problem for the greedy :) I want moar, moar, moar :P
09:45 masak I want the 1,001,000 USD.
09:45 * moritz would probably just take box B, too
09:45 masak that's clearly the winning condition here ;)
09:45 tadzik I like good-enough choices
09:46 * moritz too, as long as they are as awesome as Perl 6
09:46 masak Perl 6 is all about both boxes :P
09:46 tadzik I like the saying "I don't have to be the best, I can be very good"
09:46 envi_laptop joined #perl6
09:46 tadzik masak: ng took both, nom tries to take only one :)
09:47 moritz without having read all of Elizer's analysis, I'd approach the problem like this:
09:47 moritz suppose the probabilty that you can outsmart Omega is p
09:47 moritz if you take both boxes, your expected value is 1001000 * p + 1000 (1-p)
09:48 moritz if you take box B, your expected value is 1mio
09:48 moritz now you set both sides equal, and see what p is
09:51 moritz so if you think you can outsmart Omega with a certainty of 99.9%, you take both boxes, and just B otherwise
09:51 moritz my hubris doesn't go that far.
09:53 tadzik moritz++ # rational analysis
10:04 masak I like the analysis :)
10:05 masak even though I'm not sure if outsmarting Omega comes down to a probability.
10:16 pmurias joined #perl6
11:01 tadzik make spectest  1265.52s user 92.35s system 99% cpu 22:48.75 total
11:01 tadzik nom: say (1398.19 - 1265.52) / 1398.19
11:01 p6eval nom f2857b: OUTPUT«0.094886960999578␤»
11:02 tadzik not bad
11:02 tadzik that is mostly mls++' patch I suppose
11:02 tadzik the previous result is from 2011-09-09 00:13:36
11:02 jnthn mls++
11:02 * tadzik afks
11:02 jnthn 10% spectest time win is nice
11:03 jnthn .oO( wonder if my box does it in 200s now... )
11:06 HarryS joined #perl6
11:08 odoacre joined #perl6
11:13 wamba joined #perl6
11:16 replore_ joined #perl6
11:20 mj41 joined #perl6
11:28 satyavvd joined #perl6
11:29 arthur-_ joined #perl6
11:30 djanatyn joined #perl6
11:35 Psyche^ joined #perl6
11:42 moritz spectest in 200s?
11:42 moritz how many cores?
11:43 jnthn moritz: I got it to 220 yesterday by running it with --jobs=8
11:43 jnthn moritz: 4 cores, but I guess then there's HT, and it can win something by being able to beat IO-bound things.
11:44 jnthn mls++ seems to have got us a ~10% reduction, so that could bring me to the 200ish mark :)
11:48 lichtkind joined #perl6
11:49 bbkr 200s? challenge accepted :)
11:51 jnthn Well, getting Rakudo improved to the point where it's more like 100s is a more interesting challenge :P
11:53 moritz hey, we could give mls_ a commit bit for nqp, iirc we don't require CLAs for that
11:53 jnthn +1
11:56 renormalist|work joined #perl6
11:59 moritz done
12:03 * daxim_ beckons jnthn
12:03 daxim_ http://conferences.yapceurope.org/tcpw2011/newtalk
12:06 HarryS joined #perl6
12:08 wamba joined #perl6
12:09 jnthn daxim_: Yes, I read the message on the ba.pm list this morning ;)
12:09 jnthn daxim_: Planujem prist. :)
12:10 Bzek joined #perl6
12:17 donri is the artistic lisence most similar to bsd or gpl?
12:19 daxim_ similarity under which aspect?
12:19 moritz in length of text, it's more similar to the GPL :-)
12:19 moritz it's also compatible to GPL
12:20 daxim_ copyleft: gpl yes, bsd no, artistic no    (afaik)
12:20 moritz but AL is more permissive than the GPL
12:20 donri aka is it permissive or copyleft
12:20 donri ok thanks
12:20 f00li5h joined #perl6
12:21 moritz donri: in case of modifications, it requires a rename OR publishing of the modifications (if redistributed)
12:22 lichtkind daxim_: your rt and the mail was broken, wrote domm yesterday about it
12:22 daxim_ damn, really?  we tested it when it was set up
12:22 lichtkind jnthn: you will come to vienna?
12:23 lichtkind daxim_: i got a No permission to create tickets in the queue 'TwinCity2011'
12:23 pernatiy joined #perl6
12:30 clairvy joined #perl6
12:38 abercrombie joined #perl6
12:39 jnthn lichtkind: Ano :)
12:40 lichtkind jnthn: than i pull back my p6 talk :)
12:40 benabik joined #perl6
12:40 jnthn lichtkind: I wasn't in Bratislava since I used to live there :)
12:40 jnthn lichtkind: And I really miss it. :)
12:40 lichtkind yes understand it
12:41 jnthn lichtkind: Hey, you can talk about Perl 6 too :P
12:41 jnthn lichtkind: What's your topic?
12:42 lichtkind jnthn: still improving this arrays and hashes talk
12:42 jnthn lichtkind: OK, that's probably quite different from what I'd talk about.
12:42 lichtkind you do signatures or the metamodel?
12:46 uasi joined #perl6
12:46 moritz Perl 6: all the cool features in just 5 minutes, explained in depth!
12:47 lichtkind stay away from moritz, he is a snake oils salesman
12:48 moritz I merely proposed a title for a talk :-)
12:53 lichtkind sounds like amn electronic noise experiment, to play a 20h talk in 5 min :)
12:54 benabik It's just a 5 minute talk delivered from depth... say 60' underwater?
12:54 lichtkind hahah
12:55 moritz good be a talk advertising a book or so :-)
12:56 jnthn lichtkind: Was pondering doing one on "Optimizing Rakudo" or "Optimizing Perl 6" or some such.
12:56 lichtkind yes
12:56 jnthn lichtkind: By November I shoulda done a bunch of work on optimizers, and have the BS stuff in place, or mostly in place. So should have plenty of things to say :)
12:58 moritz "Perlis of inlining - what can go wrong goes wrong"
12:58 moritz *Perils
12:58 jnthn ;)
12:58 jnthn Interesting typo :)
12:58 jnthn "Perils of Perl Inlining" :)
13:04 im2ee joined #perl6
13:05 Alias joined #perl6
13:05 donri joined #perl6
13:09 Holy_Cow joined #perl6
13:10 Holy_Cow left #perl6
13:11 mtk joined #perl6
13:17 mikemol TimToady: Nice answer. I'll be interested to watch what happens as Paddy gets more interested in P6.
13:23 bbkr has anybody experienced "my $i = now;" hanging forever? it happens on one of my machines on latest NOM, but I don't know what causes that.
13:24 moritz bbkr: it's Num.Rat on 32 bit systems
13:25 moritz bbkr: problem is that timestamp is a 10-digit number, and if you add a few decimals to that, it just doesn't fit into a Rat that's made of two 32 bit ints
13:27 bbkr moritz: thanks for explanation
13:30 bbkr that ends my attempt to crunch spectest under 200 seconds :)
13:31 replore_ joined #perl6
13:32 benabik_ joined #perl6
13:38 wamba joined #perl6
13:38 flussence (it'd be nice if prove had a thing to auto-kill a test after some length of time...)
13:38 moritz prove --exec='perl tools/perl6-limited.pl' ...
13:39 sorear masak: That error message would happen if you somehow produced a subclass of Cursor or Match using Mu.new
13:39 sorear masak: you *might* also be able to trip it by using undef in an unhandled way
13:39 flussence well that was simple enough :)
13:40 moritz sorear: what's this 'undef' you are talking about?
13:44 masak sorear: as far as I can see, all I did was grammar-match in a les-than-successful way.
13:55 im2ee Hello:)
13:56 masak im2ee: cz!
14:02 sorear moritz: 'Match' or 'Cursor'
14:03 sorear masak: maybe you're doing something with $/ after a failed match that I never expected anyone to do with the Match type object.  I can be very shortsighted sometimes.
14:04 sorear TimToady: +1 to tiebreaking on lexical scope depth
14:04 sorear I planned to do it in Niecza too
14:04 jnthn yay :)
14:04 masak sorear: on the line where it blows up, I'm doing $/.ast
14:04 sorear niecza: multi foo(Int $x) { "a" } ;  { multi foo(Int $x) { "b" }; say  foo(5) } # I can't remember if I implemented it yet
14:04 p6eval niecza v9-24-ga279320: OUTPUT«Potential difficulties:â�¤  $x is declared but not used at /tmp/rJfxOl78Bp line 1:â�¤------> [32mmulti foo(Int [33mâ��[31m$x) { "a" } ;  { multi foo(Int $x) { "b"[0mâ�¤  &foo is declared but not used at /tmp/rJfxOl78Bp line 1:â�¤------> [32mmulti foo[33mâ��[31m(…
14:05 sorear niecza: multi foo(Int $x) { "a" } ;  { multi foo(Int $x) { "b" }; say  foo(5) } #OK # I can't remember if I implemented it yet
14:05 p6eval niecza v9-24-ga279320: OUTPUT«Potential difficulties:â�¤  $x is declared but not used at /tmp/CtGPVSP1Kq line 1:â�¤------> [32mmulti foo(Int [33mâ��[31m$x) { "a" } ;  { multi foo(Int $x) { "b"[0mâ�¤  &foo is declared but not used at /tmp/CtGPVSP1Kq line 1:â�¤------> [32mmulti foo[33mâ��[31m(…
14:05 sorear niecza: multi foo(Int $x) { "a" } ;  { multi foo(Int $x) { "b" }; say  foo(5) } #OK
14:05 p6eval niecza v9-24-ga279320: OUTPUT«b␤»
14:07 sorear \o/
14:07 molaf joined #perl6
14:12 packetknife joined #perl6
14:12 mls_ good evening!
14:14 alvis joined #perl6
14:15 mls_ jnthn: NQPClassHOW.pm's "methods" method ignores the :$local flag. Is that intentional?
14:15 moritz \o
14:16 masak mls_! \o/
14:16 moritz mls_: I've given you commit access to nqp, in case you want to do more patches
14:16 moritz (nqp commit bit policy is rather liberal)
14:16 mls_ Yes, I got an emial notification. Thanks!
14:16 moritz niecza: 'a' ~~ /b/; say $<foo>
14:16 p6eval niecza v9-24-ga279320: OUTPUT«Any()␤»
14:16 moritz niecza: 'a' ~~ /b/; say $<foo>[0]
14:16 p6eval niecza v9-24-ga279320: OUTPUT«Any()␤»
14:17 moritz doesn't seem to be that simple
14:18 moritz niecza: 'a' ~~ /b/; say $/.to
14:18 p6eval niecza v9-24-ga279320: OUTPUT«Unhandled exception: System.InvalidCastException: Cannot cast from source type to destination type.␤  at CORE.C684_Match_2Eto (Niecza.Frame ) [0x00000] in <filename unknown>:0 ␤  at Niecza.Kernel.RunCore (Niecza.Frame& cur) [0x00000] in <filename unknown>:0…
14:18 moritz there you go
14:19 jnthn mls_: Is it not :$local! ?
14:19 mls_ yes.
14:19 jnthn mls_: So it's mandatory
14:19 moritz I'd expect that to throw an error like 'Cannot access attribute of a type object' or so
14:19 jnthn e.g. it doesn't know how to handle the non-local case.
14:20 mls_ It only handles the local case, I think
14:20 moritz that's what jnthn said, except that he used double negation
14:20 moritz :-)
14:21 mls_ yes, sorry ;)
14:21 moritz and not avoiding double negation makes it easy to non-misunderstand
14:21 moritz or something :-)
14:22 mls_ but shouldn't it also handle the non-local case?
14:23 jnthn mls_: It can be made to :)
14:23 mkramer left #perl6
14:23 jnthn mls_: There's not an NQP spec really, so much of what it's meta-objects do has come about on a "what do they need to do" basis.
14:24 jnthn mls_: Whereas S12 lays out what full-blown Perl 6 should do.
14:24 jnthn mls_: I've no objections if somebody wants to extend them a bit though.
14:25 mls_ ok
14:26 mls_ my goal was to speed up has_method in RoleToClassApplier a bit, which uses methods() to get all methods and then compares the name with each element of the array
14:27 mls_ so I wondered if I could use method_table() instead, but that only has the local methods
14:28 jnthn mls_: yeah
14:28 jnthn mls_: You could build a hash of names once though
14:28 jnthn as in, once in the applier
14:29 jnthn mls_: Or walk the mro and look in each method table.
14:29 uasi_ joined #perl6
14:30 mls_ yes
14:31 mls_ "can" seems to already do that
14:32 jnthn Not sure I'd fancy relying on can here though
14:32 jnthn Given we're still uncomposed.
14:33 mls_ so no @!meo?
14:33 mls_ @!mro?
14:33 jnthn well, that, but also .^can may want to delegate to the method cache or something...
14:33 tokuhirom joined #perl6
14:34 clairvy joined #perl6
14:34 jnthn Mostly it's just a case of trying to think ahead about what other people's meta-objects might do/look like.
14:36 mls_ btw, what does the 'methods' method return?
14:36 mls_ an array of code objects, right?
14:37 im2ee joined #perl6
14:38 mls_ so is it ok for has_method to compare with a name?
14:39 dalek roast: 698a5e1 | Coke++ | S03-smartmatch/any-array.t:
14:39 dalek roast: nom fudging
14:39 dalek roast: review: https://github.com/perl6/roast/commit/698a5e1e55
14:42 jnthn mls_: code objects, but it's the names that matter in role composition
14:42 kaare_ joined #perl6
14:43 jnthn Apart from non-collision, where sameness is interesting.
14:45 mls_ so code objects stringify to the name?
14:46 jnthn Yes
14:46 jnthn Stringification is safer than calling .name (more)
14:46 jnthn NQP doesn't (yet?) wrap up Parrot subs in code objects
14:46 mls_ ah, ok. makes sense.
14:47 jnthn Though it irks me more and more that it doesn't and the performance overhead of doing so is very minimal these days.
14:47 mls_ so you better don't all add_method with the wrong name ;)
14:47 jnthn ;)
14:48 mls_ (cause add_method takes both a name and a code object as arguments)
14:48 jnthn yeah
14:48 jnthn Looking at the method table's keys is "safer" in a sense.
14:49 jnthn Probably should move towards doing that.
14:51 mls_ hmm, perl6's RoleToClassApplier::has_method ignores $local and uses the method_table.
14:51 jnthn ah :)
14:51 jnthn Well, I guess I learned something between NQP and Perl 6 then :P
14:52 jnthn Generally, if Perl 6's meta-objects do things in a way that seems better than NQP's, then it probably makes sense to update the NQP ones accordingly.
14:52 mls_ ok, will do ;)
14:52 jnthn I need to re-visit the NQP roles implementation at some point to apply the lessons from the Perl 6 one too.
14:52 jnthn er, the parametricity bits of it.
14:54 mls_ perl6 calls has_method() with local=0 when composing the methods. Is this a bug?
14:54 mls_ (either that local is ignored or that it is called with local=0)
14:55 jnthn Depends where.
14:55 jnthn Potentially a thinko.
14:55 jnthn nom: class A { method m() { say 'a' } }; role R { method m { say 'r' } }; class C does R { }; C.m
14:55 p6eval nom f2857b: OUTPUT«r␤»
14:55 jnthn hm, it gets that right :)
14:56 mls_ cause it ignores the local=0 ?
14:56 jnthn nqp: class A { method m() { say('a') } }; role R { method m() { say('r') } }; class C does R { }; C.m
14:56 p6eval nqp: OUTPUT«r␤»
14:56 jnthn mls_: ah :)
14:56 jnthn mls_: Maybe. I need to decommute in a moment...can look more once I'm home
14:56 mls_ ok. till later...
14:57 jnthn But think is, Rakudo classes don't ignore the :local(0)
14:57 jnthn *thing
14:57 mls_ but Rakudo's RoleToClassApplier::has_method ignores it ;)
14:58 jnthn mls_: Oh.
14:58 jnthn mls_: ahh...thinking about it...
14:59 jnthn I've yet to implemented the "required method" thingy roles do
14:59 sorear masak: also, it's not clear what the correct behavior of #49 is
14:59 jnthn Which is the place I'd need :local(0)
14:59 jnthn mls_: but if the composition check passes 0...yeah, sounds like thinko.
15:01 thou joined #perl6
15:03 jnthn decommute &
15:04 saaki joined #perl6
15:05 masak sorear: huh? @a.elems should be 0, where's the unclarity?
15:05 moritz masak: did you see my example that triggered the System.InvalidCastException?
15:06 clairvy joined #perl6
15:06 sorear masak: what is the nature of @a?
15:06 moritz maybe that can be golfed further
15:07 moritz niecza: $/.to
15:07 p6eval niecza v9-24-ga279320: OUTPUT«Unhandled exception: Unable to resolve method to in class Any␤  at /tmp/qAZGUBmGMo line 1 (MAIN mainline @ 1) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2060 (CORE C968_ANON @ 2) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2061 (CORE module-CORE @ …
15:07 sorear masak: is it equivalent to [], creating a new array?
15:07 moritz niecza: Match.to
15:07 p6eval niecza v9-24-ga279320: OUTPUT«Unhandled exception: System.InvalidCastException: Cannot cast from source type to destination type.␤  at CORE.C684_Match_2Eto (Niecza.Frame ) [0x00000] in <filename unknown>:0 ␤  at Niecza.Kernel.RunCore (Niecza.Frame& cur) [0x00000] in <filename unknown>:0…
15:07 moritz niecza: say Num.sqrt
15:07 p6eval niecza v9-24-ga279320: OUTPUT«0+0i␤»
15:07 sorear masak: and why Array?  The implicit type constraint is Positional or maybe Positional[Any]
15:08 dalek niecza: bc5cf9d | sorear++ | lib/Builtins.cs:
15:08 dalek niecza: Zero sqrt does not need to promote
15:08 dalek niecza: review: https://github.com/sorear/niecza/commit/bc5cf9d5fa
15:09 masak sorear: for the same reason 'my @a' gets an empty Array, I guess.
15:10 masak Array is the "default" Positional.
15:10 sorear Hmm.  I guess that rationale works.
15:28 masak perl6: say Num.sqrt
15:28 p6eval niecza v9-24-ga279320: OUTPUT«Rebuild in progress␤»
15:28 p6eval ..rakudo f2857b: OUTPUT«No applicable candidates found to dispatch to for 'sqrt'. Available candidates are:␤:(Num, Mu %_)␤␤  in method sqrt at src/gen/CORE.setting:2364␤  in <anon> at /tmp/NA3sSYdnM3:1␤  in <anon> at /tmp/NA3sSYdnM3:1␤␤»
15:28 p6eval ..pugs: OUTPUT«0␤»
15:31 sorear #phasers in 119? :D
15:32 tadzik mebbe :)
15:32 moritz next week, I think
15:32 tadzik maybe I should do something to have a report
15:32 dalek nqp: f3b8f73 | (Michael Schroeder)++ | src/how/RoleToClassApplier.pm:
15:32 dalek nqp: copy has_method implementation over from perl6
15:32 dalek nqp:
15:32 dalek nqp: The perl6 implementation uses the method_table instead
15:32 dalek nqp: of iterating over all methods. It ignores $local for now,
15:32 dalek nqp: this may need to be fixed in the future.
15:32 dalek nqp: review: https://github.com/perl6/nqp/commit/f3b8f73e40
15:32 mls_ heh, my first commit ;)
15:32 moritz tadzik: #phasers-driven development
15:32 masak mls_++
15:32 tadzik nice one
15:33 tadzik moritz: or motivation-driven. Or even expectation-driven.
15:33 tadzik It's when someone gives you a commit bit, and you feel obliged to commit :P
15:33 masak mls_: is your last name "Schroeder" or "Schröder"?
15:33 masak just curious.
15:33 mls_ Schröder, actually
15:34 masak ooh, a Latin-1 client ;)
15:34 mls_ yes, parrot also doesn't like that. I always need to change LC_CTYPE to stop parrot from complaining ;)
15:34 * tadzik mumbles something about every package delivered to some So&#347;nierz guy
15:35 sorear can't the world just standardize on utf-8 already?
15:35 mls_ ;)
15:35 tadzik ups--
15:35 sorear maybe utf-16 for the East Asians
15:36 dalek niecza: 0173980 | sorear++ | / (2 files):
15:36 dalek niecza: Optional @/% defaults to empty Array/Hash (masak) (fixes #49)
15:36 dalek niecza: review: https://github.com/sorear/niecza/commit/0173980691
15:38 masak mls_: UTF-8 is your *frieeeeend*! :D
15:38 mls_ yes. It's just lazyness...
15:38 masak »ö«
15:38 masak oh wait, that's all Latin-1 too :P
15:39 sorear tadzik: have you seen? http://en.wikipedia.org/wiki/File:Letter_to_Russia_with_krokozyabry.jpg
15:40 * masak admires that post office
15:41 moritz sounds like they employ a geek :-)
15:41 tadzik oh wow
15:46 moritz a wrongly encoded address is just the riddle you want to solve if you have to repair horrible hand writing all day :-)
15:49 masak no Ch. 78 of HPMoR yet :/
15:51 moritz masak: but Ch. 20 of the podcast
15:52 masak I'm thinking of listening to it... but it'll take a year for it to catch up to where the text version is *now*...
15:54 moritz I find it very nice for the commute time
15:54 moritz reading on the bike is LTA :-)
15:54 * moritz -> commute :-)
16:01 * jnthn home
16:01 sayu joined #perl6
16:01 * masak decommutes
16:04 dalek roast: d09d566 | sorear++ | S06-signature/optional.t:
16:04 dalek roast: Add tests for Niecza #49
16:04 dalek roast: review: https://github.com/perl6/roast/commit/d09d566c46
16:04 SHODAN joined #perl6
16:04 sorear o/ jnthn
16:04 jnthn hi sorear
16:06 dukeleto joined #perl6
16:11 sorear oh wow, github++
16:11 sorear I went to close #49 but apparently it saw my commit and closed itself
16:12 * sorear hunts for documentation of this feature
16:12 uasi joined #perl6
16:24 sorear yay, found it
16:27 dalek specs: ac6eea7 | larry++ | S12-objects.pod:
16:27 dalek specs: add scoping in as multi dispatch tiebreaker
16:27 dalek specs: review: https://github.com/perl6/specs/commit/ac6eea7387
16:28 mberends joined #perl6
16:30 sorear o/ mberends
16:31 mberends o/ sorear.  I have about 12 hours of backlogging to do.
16:32 * moritz wonders how many days worth of IRC conversation one can backlog in 12 hours
16:34 sorear in early Wikipedia days there was a joke about classifying people by the number of hours they spent to study one hour of RecentCHanges
16:34 sorear or maybe that was TvTropes
16:37 dalek rakudo/nom: f8301da | Coke++ | t/spectest.data:
16:37 dalek rakudo/nom: track failures, run fudges.
16:37 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/f8301da74b
16:40 * sorear has found a new horrible hack which is less horrible than the old horrible hack, for ferrying data between compiler-space and run-space
16:41 TimToady sounds convergent, or at least less divergent :)
16:46 jnthn mls++ # my spectest run is down 193s! :)
16:47 sorear nice
16:47 sorear what's the mimimum memory these days?
16:48 jnthn sorear: for?
16:48 jnthn sorear: Build?
16:48 sorear jnthn: build/spectest
16:49 flussence my netbook can survive it with only 1GB (no swap)
16:49 jnthn Not sure, but I suspect higher than ng
16:49 flussence I think it's closer to 700MB now
16:49 jnthn flussence: 32-bit or 64-bit?
16:49 sorear so probably no hope just yet of a .25 build.  ok.
16:49 flussence 32, I have to manually kill that one test that hangs...
16:50 jnthn sorear: No, it probably still needs some reigning in :(
16:51 jnthn sorear: Though the amount of PAST nodes we build probably drops by half once the serialization stuff lands.
16:54 sorear pre-\o/
16:55 tadzik sorear: how come I hear of no such problems in niecza? Is it that CLR is superior in this sense, or you just got the hard stuff silently done?
16:57 sorear tadzik: you haven't heard of this because I haven't started tackling it
16:57 tadzik oh, I see
16:58 tadzik so you handle the compiletime-runtime boundary differently somehow?
16:58 sorear tadzik: also, AST nodes have always been real objects in Niecza
16:58 jnthn "real objects"?
16:58 sorear PCT nodes are Captures
16:58 jnthn Oh, as opposed to the capture thingy...right.
16:59 sorear I guess because PCT started in Parrot's infancy
16:59 jnthn aye
17:00 sorear tadzik: the compiletime-runtime boundary is handled very differently in niecza.  I actually think jnthn++ has done a very good job considering the redesign for nom, and while I think niecza's way is better than what b did, I'm going to steal a *lot* from nom...
17:08 jnthn cotto: Tried the branch. It doesn't give any warnings during Configure now. :)
17:10 localhost joined #perl6
17:11 jnthn cotto: The bad news is that it now blows up like:
17:11 jnthn .\miniparrot.exe -Iruntime/parrot/include -o frontend/parrot2/prt0.pbc frontend/parrot2/prt0.pir
17:11 jnthn PackFile_pack segment 'DIRECTORY' used size 2316 but reported 2596
17:12 mj41 joined #perl6
17:14 * jnthn git clean -xdf's just in case
17:15 jnthn no luck :(
17:21 cotto jnthn, dang
17:21 cotto I wish I had a 64-bit windows box so I could fiddle with it myself
17:23 packetkn_ joined #perl6
17:25 TimToady that looks like the kind of ratio you get when there's a CRLF vs LF foulup
17:26 TimToady if the lines average about 20 chars...
17:26 TimToady er, 10 chars
17:27 TimToady perl6: say (2596-2316)/2316
17:27 p6eval rakudo f8301d: OUTPUT«0.120898100172712␤»
17:27 p6eval ..pugs: OUTPUT«0.1208981001727115716753022452504317789292␤»
17:27 p6eval ..niecza v9-26-g0173980: OUTPUT«70/579␤»
17:28 masak joined #perl6
17:28 masak evening! \o/
17:28 frettled Indeed it is!
17:28 TimToady more like 8 chars
17:31 sorear o/ masak
17:32 sorear incidentally, I think pugs' output is a tad excessive.
17:32 [Coke] sorear: they're just showing off.
17:32 tadzik those haskellers
17:38 * moritz compiles and tests a fix for http://www.perlmonks.org/?node_id=925931
17:39 jnthn moritz++
17:39 Chillance joined #perl6
17:58 colomon did we decide not to have #phasers today?
17:59 tadzik I think so
17:59 benabik joined #perl6
18:03 masak yeah, we can't have status meetings every day. :)
18:03 masak at least not yet.
18:03 jnthn But I want to do scrum! :P
18:05 dalek rakudo/nom: 6c9731a | moritz++ | src/Perl6/Grammar.pm:
18:05 dalek rakudo/nom: die on use of match variable in declaration
18:05 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/6c9731a031
18:05 moritz .oO( a scrumateur )
18:10 dalek niecza: b85eeeb | sorear++ | / (5 files):
18:10 dalek niecza: Redesign to cross-domain communication system
18:10 dalek niecza:
18:10 dalek niecza: Instead of using a custom interface which is compiled once and shared
18:10 dalek niecza: between versions of Kernel.dll, a system which proved fragile and
18:10 dalek niecza: inflexible, we now coopt System.Collections.IDictionary.
18:10 dalek niecza:
18:10 dalek niecza: Also adds a C# code module to the compiler, finally allowing all
18:10 dalek niecza: aspects of the compiler/compilee interface to be controlled by current
18:10 dalek niecza: code.
18:10 dalek niecza: review: https://github.com/sorear/niecza/commit/b85eeeba94
18:14 im2ee joined #perl6
18:19 * sorear out
18:21 TimToady perl6: my $::("-O'Malley's") = 42; say $::("-O'Malley's");
18:21 p6eval rakudo f8301d: OUTPUT«===SORRY!===␤Variable variable names not allowed in declarations at line 1, near " = 42; say"␤»
18:21 p6eval ..pugs: OUTPUT«*** ␤    Unexpected "$::(\"-"␤    expecting "=", formal parameter, context, ":" or "("␤    at /tmp/YbA4nL6sIe line 1, column 4␤»
18:21 p6eval ..niecza v9-26-g0173980: OUTPUT«Any()␤»
18:21 TimToady I think that should be made to work eventually
18:22 jnthn ew :)
18:22 jnthn But yes.
18:22 TimToady given the string can be recognized as static
18:22 jnthn rajt
18:22 jnthn huh
18:22 jnthn *right
18:22 jnthn :)
18:22 TimToady rajt isn't right but it works :)
18:23 * TimToady isn't sure, actually, what niecza thinks it's doing with it
18:25 birdwindupbird joined #perl6
18:31 xinming joined #perl6
18:37 masak rakudo: role A { multi method foo { say "A" } }; role B { multi method foo { say "B" } }; class C does A does B {}; C.new.+foo
18:37 p6eval rakudo f8301d: OUTPUT«Ambiguous dispatch to multi 'foo'. Ambiguous candidates had signatures:␤:(C, Mu %_)␤:(C, Mu %_)␤␤  in method foo at src/gen/CORE.setting:246␤  in method dispatch:<.*> at src/gen/CORE.setting:681␤  in method dispatch:<.+> at src/gen/CORE.setting:669␤  in <anon> at /…
18:38 jnthn That's awkward.
18:38 jnthn I'm not quite sure how to fix that one yet.
18:38 jnthn It was easy in ng because we had a much flatter view of dispatch.
18:39 jnthn Now we have nested dispatchers and it's not obvious that the method dispatcher, which handles the .+/.*, should descend into a nested dispatcher too.
18:39 jnthn (And the method dispatcher's view is that there's one method to call. It just happens that one method is a proto.)
18:41 benabik rakudo: role A { method foo { say "A" } }; role B { method foo { say "B" } }; class C does A does B {}; my $_ = C.new; .foo; .+foo
18:41 p6eval rakudo f8301d: OUTPUT«===SORRY!===␤Method 'foo' must be resolved by class C because it exists in multiple roles (B, A)␤»
18:41 masak that one's correct.
18:42 benabik Not sure why that's different when multi is added.  It's still ambiguous.
18:42 TimToady I see an ambiguous dispatch there too
18:42 masak even with .+ ?
18:43 TimToady .+ applies to method dispatch, not multi dispatch
18:43 TimToady there's only one proto, and .+ finds it
18:43 masak ah.
18:43 benabik Doesn't it have to be resolved at class composition time?  A has a method foo, B has a method foo…  Does it matter if they're multis?
18:43 jnthn TimToady: oh, wow. I expected you to wave your hands and tell me to just make it work :)
18:43 jnthn benabik: It's undecidable in the general case whether a multi is ambiguous.
18:44 TimToady which is part of why we just added a scoping tiebreaker, which doesn't help in this case
18:44 poerschke joined #perl6
18:44 benabik Hm.  multi dispatch is probably turing complete…  :-
18:44 benabik :-/
18:44 jnthn benabik: Sure, as soon as you add where clauses.
18:44 masak with where clauses, sure.
18:44 jnthn benabik: Thankfully the common cases are not
18:44 TimToady however, if you mixed in role B after role A, then B would be more derived, and preferred
18:45 jnthn TimToady: Right, but this is composition, not mixins.
18:45 TimToady which is why it's ambiguous
18:45 jnthn Seems we're agreeing ;)
18:45 TimToady I'm trying to be as disagreeable as possible :)
18:45 benabik I would think that adding roles or classes with overlapping multis should be noted somewhere.  Seems ripe for error.
18:46 jnthn TimToady: :P
18:46 masak benabik: sure, but it'd be a compilte-time boon.
18:46 TimToady if it can be determined that a dispatch will be ambiguous, it's probably worth a warning at least
18:47 jnthn Well, if we know that .+ and .* won't descend into multi dispatch, then we can do a compile time error.
18:47 TimToady in this case, the signatures appear to be close enough for such a warning
18:47 jnthn Like we planned for subs.
18:47 szabgab joined #perl6
18:47 jnthn I thought we agreed the sub case was an error, anyway...
18:48 jnthn The only reason I argued against that for methods was because I'd forgotten that .+ and .* would work differently now.
18:49 wallberg joined #perl6
18:49 * jnthn remembers that he should eat something for dinner
18:49 benabik food++
18:49 masak the staticker, the better.
18:50 moritz TimToady: http://www.perlmonks.org/?node_id=925973 -- are you serious? why would anybody want that to work?
18:50 benabik I had that problem too.  Difficult to concentrate on class when my brain is going "when's lunch" every few minutes.
18:51 masak moritz: I guess one could argue language interop or sump'n.
18:53 jnthn hm, something to eat seems to require me to go and buy something to stick inside my sandwich...bbs :)
18:55 TimToady moritz: because there's no reason to prohibit it
18:56 moritz TimToady: I think it's dangerous, because it depends on the degree in which one can constant-fold, which is probably quite implementation dependent
18:57 moritz so I fear it's easy to accidentally create compiler dependent behavior
18:58 masak moritz: then don't use the feature :)
18:58 masak same way you wouldn't use Q:PIR in a compiler-agnostic setting.
18:58 moritz masak: except that Q:PIR is *obviously* compiler specific
18:59 masak sure, and $::("-O'Malley's") will be right there in perl6doc perl6port
19:02 benabik Constant folding isn't quite the right term…  There's no folding, just changing code gen based on the presence of a literal.  It's the difference between recognizing :16<deadbeef> as a literal and :16<$const> as a literal
19:02 alvis joined #perl6
19:03 moritz std: :16<$const>
19:03 p6eval std bb4f150: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Malformed radix number at /tmp/6f6fppCMaE line 1:â�¤------> [32m:16<[33mâ��[31m$const>[0mâ�¤    expecting number in radix notationâ�¤Parse failedâ�¤FAILED 00:01 117mâ�¤Â»
19:04 masak benabik: TimToady has a broad definition of constant folding, is all.
19:06 benabik masak: From an academic viewpoint, constant folding is an occasionally tricky optimization.  What TimToady's asking for is far simpler.  And since I'm sitting in a uni right now, I'm probably going to be a little more pedantic.  ;-)
19:06 masak benabik: I haven't learned about constant folding at uni. would you be so kind as to define it for me?
19:07 masak specifically, why's it tricky?
19:07 benabik masak: Constant folding is turning "1 + 1" into "2" at compile time.
19:07 masak that seems a very narrow definition. :P
19:07 benabik masak: Tricky part is recognizing what operations are and are not possible to do that with.
19:07 masak but given it I'm not surprised you disagree with TimToady's :P
19:07 benabik masak: That was an example.  :-D
19:07 masak oh, then I misunderstood the word "is" :P
19:08 jnthn .oO( Masak Clinton )
19:08 benabik masak: Constant folding is taking operations that operate on constants and replacing them with the result of the operation.
19:08 masak benabik: but that's exactly what TimToady's snippet would require!
19:08 benabik It's fair enough.  This is a result of me being somewhat pedantic.
19:09 jnthn The thing is that in Perl 6 we're starting to get to the point where we sometimes say that things working depends on constant folding, rather than it being an optimization (and thus optionalish)
19:09 masak nod.
19:09 PerlJam as long as constant-folding is specced  :)
19:10 benabik masak: The difference (to me) is that you can express the difference in a fairly simple bit of syntax.  $::(<literal string>) is fairly obviously different than $::(<string expression>)
19:11 benabik It's a type difference, not performing operations on constant.
19:12 Kivutarrr joined #perl6
19:13 jnthn I'd hope the type of a string operation is string :P
19:13 PerlJam jnthn: yeah, but mutable or immutable?  :)
19:14 benabik jnthn: literal string is a type of string expression, but the more specific type can be noticed very quickly in parsing.
19:15 TimToady well, 'my $::()' doesn't actually require constant folding, since it should just recognize that the name has to be evaluated at compile time anyway
19:15 benabik Which is, more or less, what I'm trying to say.
19:16 benabik I'm just being verbose about it.  :-/
19:17 TimToady but we did add explicit constant declarations for a Reason, and that reason is to make it clearer to the compiler when something that looks like a variable isn't
19:18 TimToady (and to force the initialization of said to constant to compile time, of course)
19:18 TimToady s/to //
19:21 masak "the 'constant' keyword just helps the constant folding along"
19:37 pernatiy joined #perl6
19:53 masak rakudo: my @a = a => 1, b => 2; for @a -> Pair $p ( $key, $value ) { say $key }
19:53 p6eval rakudo 6c9731: OUTPUT«Not enough positional parameters passed; got 0 but expected 2 in sub-signature of parameter $p␤  in block <anon> at /tmp/PoamncmwwV:1␤  in method reify at src/gen/CORE.setting:3649␤  in method reify at src/gen/CORE.setting:3554␤  in method reify at src/gen/CORE.set…
19:56 jnthn rakudo: my @a = a => 1, b => 2; for @a -> Pair $p ( :$key, :$value) { say $key }
19:56 p6eval rakudo 6c9731: OUTPUT«Mu()␤Mu()␤»
19:57 jnthn hmm
19:57 [Coke] (not have status meetings every day) I need you to take over this project I have at work. :P
19:57 jnthn rakudo: (a => 1).Capture.say
19:57 p6eval rakudo 6c9731: OUTPUT«␤»
19:57 jnthn rakudo: (a => 1).Capture.perl.say
19:57 p6eval rakudo 6c9731: OUTPUT«␤»
19:57 jnthn rakudo: (a => 1).^attributes>>.name>>.say
19:57 p6eval rakudo 6c9731: OUTPUT«$!key␤$!value␤»
19:57 jnthn rakudo: (a => 1).^attributes>>.has_accessor>>.say
19:57 p6eval rakudo 6c9731: OUTPUT«Bool::False␤Bool::False␤»
19:58 jnthn ah.
19:58 MayDaniel joined #perl6
19:58 masak [Coke]: :)
19:59 flussence joined #perl6
19:59 jnthn masak: Are you going to masakbug the above?
20:00 jnthn masak: I'm compiling a patch now but it deserves a test.
20:00 jnthn masak: And getting you to submit a bug report is the laziest way I can make sure it gets tested :P
20:01 jnthn > my @a = a => 1, b => 2; for @a -> Pair $p ( :$key, :$value) { say $key }
20:01 jnthn a
20:01 jnthn b
20:01 jnthn \o/
20:01 mikemol joined #perl6
20:02 benabik jnthn: Your code appears to print an extra "\o/".  You should figure out where that's coming from.  ;-)
20:03 * masak submits rakudobug, on request
20:03 masak jnthn: I'm sorry, I woulda done it sooner. I was sure that one had an RT ticket already.
20:03 mikemol joined #perl6
20:03 wallberg joined #perl6
20:03 birdwindupbird joined #perl6
20:03 localhost joined #perl6
20:03 mberends joined #perl6
20:03 dukeleto joined #perl6
20:03 SHODAN joined #perl6
20:03 Alias joined #perl6
20:03 djanatyn joined #perl6
20:03 arthur-_ joined #perl6
20:03 lucs joined #perl6
20:03 jevin joined #perl6
20:03 y3llow joined #perl6
20:03 jql joined #perl6
20:03 dju joined #perl6
20:03 frhodes joined #perl6
20:03 aloha joined #perl6
20:03 bbkr1 joined #perl6
20:03 kfo_ joined #perl6
20:03 simcop2387 joined #perl6
20:03 cognominal_ joined #perl6
20:03 pnu joined #perl6
20:03 pothos joined #perl6
20:03 geekosaur joined #perl6
20:03 Gothmog_ joined #perl6
20:03 jnthn joined #perl6
20:03 awwaiid joined #perl6
20:03 jjore joined #perl6
20:03 domidumont joined #perl6
20:03 sivoais joined #perl6
20:03 cotto_work joined #perl6
20:03 jasonmay joined #perl6
20:03 Exodist joined #perl6
20:03 Khisanth joined #perl6
20:03 prammer joined #perl6
20:03 slavik1 joined #perl6
20:03 jerome_ joined #perl6
20:03 jfried joined #perl6
20:03 baest joined #perl6
20:03 tadzik joined #perl6
20:03 TiMBuS joined #perl6
20:03 domm joined #perl6
20:03 silug joined #perl6
20:03 zamolxes joined #perl6
20:03 chitragupt joined #perl6
20:03 clkao joined #perl6
20:03 PerlJam joined #perl6
20:03 _sri joined #perl6
20:03 ribasushi joined #perl6
20:03 amkrankruleuen joined #perl6
20:03 avar joined #perl6
20:03 mathw joined #perl6
20:03 krakan joined #perl6
20:03 kcwu joined #perl6
20:03 arnsholt joined #perl6
20:03 sjohnson joined #perl6
20:03 spacebat joined #perl6
20:03 charsbar__ joined #perl6
20:03 jesk joined #perl6
20:04 masak I mean, how could it not? it's been known for ages.
20:04 donri joined #perl6
20:04 donri joined #perl6
20:04 * jnthn gets out the debugger... :P
20:05 y3llow joined #perl6
20:05 localhost joined #perl6
20:05 masak seems there isn't one, so here goes.
20:06 * jnthn wonders what me missed by being on the wrong side of the netsplit
20:06 tadzik yeah
20:07 pothos_ joined #perl6
20:07 dalek rakudo/nom: 16bb82c | jnthn++ | src/core/Enum.pm:
20:07 dalek rakudo/nom: key and value should declare themselves as public attributes in Enum.
20:07 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/16bb82c184
20:07 y3llow joined #perl6
20:07 jnthn There's the patch :)
20:09 masak jnthn: https://rt.perl.org/rt3/Ticket/Display.html?id=99134
20:13 masak jnthn: feel free to mark it as "take for tests".
20:15 \patch joined #perl6
20:15 nine joined #perl6
20:15 LoRe_ joined #perl6
20:15 ruoso joined #perl6
20:15 zostay joined #perl6
20:15 tewk joined #perl6
20:15 cosimo joined #perl6
20:15 rhr joined #perl6
20:15 athomason joined #perl6
20:15 phenny joined #perl6
20:15 ggoebel joined #perl6
20:18 pmichaud joined #perl6
20:18 shachaf joined #perl6
20:18 kthakore_ joined #perl6
20:18 frettled joined #perl6
20:18 broquaint joined #perl6
20:18 szbalint joined #perl6
20:18 zb joined #perl6
20:19 breinbaas joined #perl6
20:19 sunnavy joined #perl6
20:19 sorear joined #perl6
20:19 ascent_ joined #perl6
20:19 pochi joined #perl6
20:19 snarkyboojum joined #perl6
20:19 cxreg joined #perl6
20:19 renormalist joined #perl6
20:19 takesako joined #perl6
20:19 lumi_ joined #perl6
20:19 miso2217 joined #perl6
20:19 drbean joined #perl6
20:19 huf joined #perl6
20:19 kst joined #perl6
20:19 ashleydev joined #perl6
20:19 jamtech joined #perl6
20:19 felher joined #perl6
20:19 fhelmberger joined #perl6
20:19 jtpalmer joined #perl6
20:19 imarcusthis joined #perl6
20:19 nsh joined #perl6
20:19 s1n1 joined #perl6
20:19 ilogger2 joined #perl6
20:20 skangas joined #perl6
20:20 Grimnir_ joined #perl6
20:20 jlaire_ joined #perl6
20:22 daemon joined #perl6
20:23 eiro joined #perl6
20:23 betterworld joined #perl6
20:24 itz joined #perl6
20:24 im2ee joined #perl6
20:26 jnthn masak: tagged
20:28 masak jnthn++
20:29 jnthn OK, so I have one spectest that says that a nextsame should die if there's nothing to defer to
20:30 jnthn And another that says it shouldn't.
20:30 jnthn Which is right?
20:30 felipe joined #perl6
20:30 donri joined #perl6
20:31 masak I was wondering the same thing the other day.
20:31 masak feels like 'nextsame' falling off the edge of the world isn't such a big deal.
20:32 masak 'nextwith' maybe, but probably not even that.
20:32 masak if the candidates've run out, they've run out. doesn't feel like an error condition.
20:32 cotto_work you'd expect it to happen eventually
20:33 masak this is assuming some kind of "candidates are independent links in a chain" model of things.
20:34 jnthn Death does seem a little harsh.
20:35 tadzik jnthn: so, do you run windows 8 yet?
20:35 tadzik there is one on http://msdn.microsoft.com/en-us/windows/apps/br229516 makes me maybe even a bit curious :)
20:36 jlaire is the internal version number 6.2 this time? :)
20:36 tadzik if it's less annoying than 7, I could possibly try it on virtualbox
20:37 benabik masak: OTOH, doesn't "no candidates" mean "we can't do what you asked"?
20:37 benabik Depends on why you're calling next same, I guess.
20:37 benabik lion-- # no, really, I meant nextsame
20:38 masak benabik: when you go 'next' in the last iteration of a loop, what do you expect to happen?
20:38 benabik masak: Is function dispatch really like looping?
20:38 masak 'nextsame' and 'nextwith' used to be spelled 'next'. which was a fairly bad idea, so it got fixed. but the semantics is basically the same.
20:39 masak benabik: yes, this kind of function dispatch is very much like looping.
20:40 jnthn tadzik: Haven't tried 8 yet :)
20:40 benabik masak: When is next same supposed to be used?  I would expect it to be "Oh, I can't handle this, try the next one"
20:41 masak benabik: I don't put that much semantics into it. some routines simply want to defer to the next one in the chain for whatever reason.
20:42 masak rakudo: class A { method foo { say "I'm a proud parent!"; nextsame } }; class B is A { method foo { say "I'm a responsible child!"; nextsame } }; B.new.foo
20:42 p6eval rakudo 6c9731: OUTPUT«I'm a responsible child!␤I'm a proud parent!␤»
20:42 masak jnthn: no, 'nextsame' shouldn't die just because there are no more candidates.
20:42 bbkr1 joined #perl6
20:43 masak it shouldn't warn either.
20:45 masak niecza: die
20:45 p6eval niecza v9-27-gb85eeeb: OUTPUT«Unhandled exception: No value for parameter $msg in CORE die␤  at /home/p6eval/niecza/lib/CORE.setting line 0 (CORE die @ 0) ␤  at /tmp/_Xawqb5QBu line 1 (MAIN mainline @ 1) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2060 (CORE C968_ANON @ 2) ␤  at /ho…
20:45 * masak submits nieczabug
20:46 jevin joined #perl6
20:47 * jnthn tags #57788 as closable after tests arranged
20:49 masak sorear++ # niecza
20:51 jnthn #61500 is testneeded too :)
20:51 colomon masak: context?  not that I disagree....
20:51 im2ee Good night! :)
20:51 tadzik gnight!
20:52 masak colomon: I'm using it for the first time in a while, and while it's a different kind of experience that Rakudo and I feel a little out of place, I also recognize that it's very usable and powerful.
20:52 masak colomon: it's definitely underused considering what it can do.
20:53 masak and fast!
20:53 masak there's a perceptible compilation delay, but once it gets going...!
20:54 masak im2ee: dobranoc! (a bit too late)
20:58 pmurias joined #perl6
20:58 pmurias masak: did you succed with running the gtk example under niecza?
20:59 mkramer1 joined #perl6
21:01 araujo joined #perl6
21:01 araujo joined #perl6
21:02 masak pmurias: no, not yet. didn't follow up on mberends' last tip yet, though.
21:02 * masak does so
21:02 [Coke] (netsplit) man, do I hate IRC. ;)
21:04 masak the trick is to end up on the side of the split with all the nice people :P
21:04 jnthn If somebody fancies some test munging, I just tagged 5 RT tickets as "testneeded"
21:04 jnthn And that was just a cursory glance for the first couple of pages in RT.
21:04 masak mberends: hm, I already had libgtk2.0-cil installed, it seems. still doesn't work.
21:05 mkramer1 left #perl6
21:06 tadzik I may take a look at that, now having a proper amount of ram
21:07 tadzik I haven't wrote and p6 in a while
21:07 mberends masak: I'll search for more clues
21:08 masak thank you.
21:08 masak ++tadzik # today's test hero
21:09 tadzik (;
21:11 masak sorear: grr. that cast exception again. tomorrow I'm going to track down and golf it for you.
21:11 * jnthn rejects an invalid test
21:11 tadzik wow, use Test is _really_ slow
21:12 jnthn tadzik: Ticket for you: https://rt.perl.org/rt3/Ticket/Display.html?id=67662
21:12 tadzik like >6 times slower on startup
21:12 jnthn tadzik: Really?
21:12 tadzik really-really
21:12 tadzik perl6 -e 1  0.72s user 0.13s system 99% cpu 0.852 total
21:12 tadzik perl6 -e 'use Test'  5.95s user 0.17s system 99% cpu 6.123 total
21:12 jnthn tadzik: huh...is it somehow missing the pre-compiled version?
21:12 masak probably.
21:12 tadzik jnthn: could be
21:13 jnthn tadzik: Or is it hitting an installed Test.pm?
21:13 tadzik you having your blazingly fast machine don't even notice small people's problems! :P
21:13 jnthn Which of course has invalid PIR...
21:13 tadzik jnthn: don't think so, I'm in /tmp
21:13 jnthn tadzik: er, but there's no measurable difference here with those two :)
21:13 tadzik jnthn: I don't think the feature from this bug is even specced
21:14 jnthn tadzik: Feel free to reject ;)
21:14 tadzik I'll re-check
21:14 flussence hey, we were talking about this a few days ago in here... http://redd.it/kfr32
21:14 * jnthn really wants to cut this RT queue down to size
21:14 * masak will be there to build it up again, don't worry
21:15 masak 'night, channel
21:15 thou joined #perl6
21:16 sbp 'night masak
21:17 tadzik Ticket 67662: Permission Denied
21:17 tadzik damnit RT, why u so stubborn
21:18 tadzik could someone reject https://rt.perl.org/rt3/Ticket/Display.html?id=67662 for me?
21:19 jnthn tadzik: done
21:19 tadzik thanks-o
21:19 tadzik nom: eval "sub bar {}"; say $!.perl;
21:19 p6eval nom 16bb82: OUTPUT«Malformed block at line 1, near ""␤»
21:22 jnthn tadzik: ''
21:22 jnthn tadzik: the {} is an interpolated block
21:22 tadzik ooh, right
21:23 jnthn http://rakudo.org/testneeded links to an empty list :(
21:24 jnthn pmichaud: ^^ if it's you who maintains that.
21:26 jnthn (Looks like double-encoding of ' or some such)
21:30 flussence wfm
21:31 tadzik https://rt.perl.org/rt3/Ticket/Display.html?id=61126 not really fixed
21:31 tadzik nom: sub foo { map { return 1 }, 1; return 2 }; say foo;
21:31 p6eval nom 16bb82: OUTPUT«2␤»
21:31 tadzik eek, wait
21:32 tadzik nom: sub bar($code) { $code() }; sub foo { bar { return 1 }; return 2; }; say foo;
21:32 p6eval nom 16bb82: OUTPUT«1␤»
21:32 tadzik this one
21:32 tadzik oh, right. So the first one is broken :)
21:39 jnthn tadzik: Not returns' fault, though...
21:39 dalek roast: 78945cc | tadzik++ | S12-attributes/inheritance.t:
21:39 dalek roast: Tests for RT #61500
21:39 dalek roast: review: https://github.com/perl6/roast/commit/78945cc460
21:39 dalek roast: 1138a24 | tadzik++ | S06-advanced_subroutine_features/return.t:
21:39 dalek roast: Tests for RT #61126
21:39 jnthn nom: sub foo { eager map { return 1 }, 1; return 2 }; say foo;
21:39 dalek roast: review: https://github.com/perl6/roast/commit/1138a24f6f
21:39 p6eval nom 16bb82: OUTPUT«1␤»
21:40 tadzik oh, so testfail. Fixing
21:41 dalek roast: 2d28aa2 | tadzik++ | S06-advanced_subroutine_features/return.t:
21:41 dalek roast: Fix tests for RT #61126, jnthn++
21:41 dalek roast: review: https://github.com/perl6/roast/commit/2d28aa2de6
21:41 tadzik https://rt.perl.org/rt3/Ticket/Display.html?id=61500 closable
21:42 tadzik https://rt.perl.org/rt3/Ticket/Display.html?id=61126 closable
21:43 jnthn tadzik: both done
21:43 jnthn tadzik++
21:44 jnthn That one masak++ filed earlier today could get tests easily and be closed too :)
21:44 diakopter .
21:44 phenny diakopter: 06:59Z <masak> tell diakopter http://www.hakank.org/comb/debruijn.cgi http://en.wikipedia.org/wiki/De_Bruijn_sequence http://www.theory.csc.uvic.ca/~cos/inf/neck/NecklaceInfo.html
21:44 diakopter phenny: tell masak cool
21:44 phenny diakopter: I'll pass that on when masak is around.
21:47 dalek Heuristic branch merge: pushed 85 commits to rakudo/optimizer by jnthn
21:48 tadzik ho ho
21:49 jnthn Just bringing it up to date with master :)
21:49 jnthn er
21:49 jnthn with nom
21:49 jnthn :)
21:50 dalek roast: e42db46 | tadzik++ | S02-names_and_variables/contextual.t:
21:50 dalek roast: Test for RT #63226
21:50 dalek roast: review: https://github.com/perl6/roast/commit/e42db46331
21:50 tadzik https://rt.perl.org/rt3/Ticket/Display.html?id=63226 closable
21:51 jnthn done
21:52 tadzik huh, this throws an exception but does not die
21:52 tadzik nom: eval 'sub foo($n) { if $n { callsame() } }; foo(1)'; say $!.WHAT
21:52 p6eval nom 16bb82: OUTPUT«Exception()␤»
21:53 tadzik nom: use Test; eval_dies_ok 'sub foo($n) { if $n { callsame() } }; foo(1)'
21:53 p6eval nom 16bb82: OUTPUT«(timeout)»
21:53 tadzik oh great
21:53 tadzik not ok 1 - locally
21:53 tadzik nom: use Test; eval_dies_ok 'sub foo($n) { if $n { callsame() } }; foo(1)'
21:53 p6eval nom 16bb82: OUTPUT«(timeout)»
21:53 tadzik teehee
21:53 jnthn tadzik: I thought the discussion earlier settled on that living, not dying?
21:54 tadzik oh, my bad
21:54 tadzik anyway, this has a test it seems
21:54 tadzik https://rt.perl.org/rt3/Ticket/Display.html?id=69314#txn-705770
21:55 tadzik but it fails, it seems
21:55 tadzik because the test's broken :) Ok, I see it now
21:55 jnthn :)
21:57 tadzik and the test is not ran anyway
21:59 dalek roast: 328f005 | tadzik++ | S06-advanced_subroutine_features/callsame.t:
21:59 dalek roast: Fix a test for RT 69314, fudge S06-advanced_subroutine_features/callsame.t for Rakudo
21:59 dalek roast: review: https://github.com/perl6/roast/commit/328f005ffe
22:00 dalek Heuristic branch merge: pushed 40 commits to rakudo/nom-buf by tadzik
22:00 dalek rakudo/nom: ed8f4ad | tadzik++ | t/spectest.data:
22:00 dalek rakudo/nom: Run S06-advanced_subroutine_features/callsame.t
22:00 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/ed8f4ad131
22:00 tadzik https://rt.perl.org/rt3/Ticket/Display.html?id=69314 closable now
22:01 meraxes joined #perl6
22:04 jnthn meh, I put in undeclared routine detection and now the setting fails to compile...
22:04 tadzik :P
22:05 tadzik meh, I added a test to a file but nom does not run it yet
22:05 jnthn nom: warn('!')
22:05 p6eval nom 16bb82: OUTPUT«Could not find sub &warn␤  in <anon> at /tmp/z6i90_Tqzc:1␤  in <anon> at /tmp/z6i90_Tqzc:1␤␤»
22:05 jnthn oh, well, it's not entirely wrong ;)
22:05 jnthn tadzik: That callsame one?
22:06 tadzik jnthn: nope, protos
22:06 jnthn ah, yeah
22:06 tadzik t/spec/S06-multi/proto.t
22:06 dalek roast: b8d66dd | tadzik++ | S06-multi/proto.t:
22:06 dalek roast: Tests for RT #68242
22:06 dalek roast: review: https://github.com/perl6/roast/commit/b8d66dd49e
22:06 tadzik ...whatever :)
22:06 jnthn Still some work to do to make that one run but it may also be out of date.
22:06 sorear good * #perl6
22:06 jnthn tadzik: Anyway, best to leave the ticket open for now.
22:07 tadzik okay
22:07 tadzik sorear: hi
22:08 diakopter sorear: I'm curious to see an example of accessing the compiler from p6 code
22:08 sorear diakopter: eval("2 + 2")
22:09 diakopter I must've misunderstood your commit message
22:09 sorear the compiler and the user code are both p6, but they are compartmentalized and cannot directly share data
22:11 dalek roast: bb87d97 | tadzik++ | S04-statement-modifiers/for.t:
22:11 dalek roast: Tests for RT #66606
22:11 dalek roast: review: https://github.com/perl6/roast/commit/bb87d9718f
22:11 tadzik https://rt.perl.org/rt3/Ticket/Display.html?id=66606 closable
22:13 tadzik https://rt.perl.org/rt3/Ticket/Display.html?id=65482 not-really-testneeded
22:14 dalek joined #perl6
22:14 jnthn 66606...the ticket of the beast?!...closed
22:15 jnthn tadzik: 65482 can be left until we run the test file
22:16 tadzik I think so
22:18 dalek roast: f695da8 | tadzik++ | S (2 files):
22:18 dalek roast: Tests for #57788
22:18 dalek roast: review: https://github.com/perl6/roast/commit/f695da8e07
22:19 tadzik https://rt.perl.org/rt3/Ticket/Display.html?id=57788 testsnotran
22:20 tadzik rakudo: (a => 1).^attributes>>.has_accessor>>.say
22:20 p6eval rakudo 16bb82: OUTPUT«Bool::True␤Bool::True␤»
22:21 tadzik rakudo: my @a = a => 1, b => 2; for @a -> Pair $p ( :$key, :$value) { say $key }
22:21 p6eval rakudo 16bb82: OUTPUT«a␤b␤»
22:22 jnthn tadzik: Think test for that one wants to go in the unpacking tests
22:22 jnthn Somewhere in S06
22:22 jnthn Check that Pair is unpackable
22:22 tadzik I'll look at it tomorrow :]
22:22 jnthn ok :)
22:22 tadzik I'll get some sleep for now
22:22 jnthn good plan ;)
22:23 tadzik plan 'good'
22:23 * jnthn will be doing Perl 6 stuff tomorrow
22:23 tadzik good plan :)
22:23 jnthn :)
22:23 jnthn dobranoc o/
22:23 dalek niecza: df880f1 | sorear++ | lib/CORE.setting:
22:23 dalek niecza: Add a default message to &die, closes #51
22:23 dalek niecza: review: https://github.com/sorear/niecza/commit/df880f1174
22:23 tadzik 'branoc :)
22:25 lichtkind my slides are now on http://www.slideshare.net/lichtkind/perl-6-data-structures
22:26 replore_ joined #perl6
22:26 jnthn nom: 1 %% 60 or say 'hi'
22:26 p6eval nom 16bb82: OUTPUT«hi␤»
22:26 jnthn nom: 0 %% 60 or say 'hi'
22:26 p6eval nom 16bb82:  ( no output )
22:30 diakopter time for some std fuzzing
22:32 diakopter niecza: &Int**^&Int
22:32 p6eval niecza v9-28-gdf880f1: OUTPUT«Unhandled exception: System.Exception: Unable to find lexical &Int in mainline␤  at Niecza.CLRBackend.NamProcessor.ResolveLex (System.String name, Boolean upf, System.Int32& uplevel, Boolean core) [0x00000] in <filename unknown>:0 ␤  at Niecza.CLRBackend.Na…
22:33 sorear diakopter: std fuzzing?
22:33 diakopter er, niecza fuzzing
22:34 sorear diakopter: I think niecza is right to fail that, although it should fail earlier
22:35 diakopter niecza: say &1**&1
22:35 p6eval niecza v9-28-gdf880f1: OUTPUT«1␤»
22:35 sorear &1**&1 means $/[1] ** $/[1]
22:35 sorear Any ** Any is interpreted as 0 ** 0
22:36 diakopter niecza: say 2^^3 # should this fail earlier too?
22:36 p6eval niecza v9-28-gdf880f1: OUTPUT«Unhandled exception: System.Exception: Unable to find lexical &infix:<^^> in mainline␤  at Niecza.CLRBackend.NamProcessor.ResolveLex (System.String name, Boolean upf, System.Int32& uplevel, Boolean core) [0x00000] in <filename unknown>:0 ␤  at Niecza.CLRBac…
22:39 diakopter std: say 2^^^3 # should this fail?
22:39 p6eval std bb4f150: OUTPUT«ok 00:01 120m␤»
22:41 dalek rakudo/nom: 867ebd6 | jnthn++ | src/core/control.pm:
22:41 dalek rakudo/nom: Add a cheating &warn to avoid meta-errors.
22:41 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/867ebd6473
22:41 dalek rakudo/nom: 12b860d | jnthn++ | src/core/Temporal.pm:
22:41 dalek rakudo/nom: A doubled 'or' caught by work in the optimizer branch.
22:41 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/12b860df69
22:42 Tene jnthn: I'm curious why you're cheating like that.
22:42 diakopter niecza: say 2 ^ ^ 3
22:42 p6eval niecza v9-28-gdf880f1: OUTPUT«one(2, 0, 1, 2)␤»
22:43 jnthn Tene: No especially good reason.
22:43 Tene 'k
22:43 jnthn Tene: I didn't really mean to fix that.
22:43 diakopter sorear: should that junction have duplicates?
22:43 jnthn Tene: I'm working on optimizer branch, which also catches various "that could never work" things. It started catching a missing &warn in the setting and then I couldn't compile the setting in that branch any more :)
22:44 sorear wait, how is that even parsing
22:44 diakopter :) fuzzing is fun
22:44 sorear diakopter: for :36 - that's actually a NYI
22:45 dalek rakudo/optimizer: ed8f4ad | tadzik++ | t/spectest.data:
22:45 dalek rakudo/optimizer: Run S06-advanced_subroutine_features/callsame.t
22:45 dalek rakudo/optimizer: review: https://github.com/rakudo/rakudo/commit/ed8f4ad131
22:45 dalek rakudo/optimizer: 1088154 | jnthn++ | src/Perl6/Optimizer.pm:
22:45 dalek rakudo/optimizer: Stub in some check-time failure/warning infrastructure. First cut of detecting undeclared routines; fails on building CORE.setting.
22:45 dalek rakudo/optimizer: review: https://github.com/rakudo/rakudo/commit/108815461c
22:45 dalek rakudo/optimizer: e98e205 | jnthn++ | src/Perl6/Optimizer.pm:
22:45 dalek rakudo/optimizer: Don't complain about calling routines that we get passed as &foo style parameters.
22:45 dalek rakudo/optimizer: review: https://github.com/rakudo/rakudo/commit/e98e205735
22:45 dalek rakudo/optimizer: 867ebd6 | jnthn++ | src/core/control.pm:
22:45 dalek rakudo/optimizer: Add a cheating &warn to avoid meta-errors.
22:45 dalek rakudo/optimizer: review: https://github.com/rakudo/rakudo/commit/867ebd6473
22:45 dalek rakudo/optimizer: 12b860d | jnthn++ | src/core/Temporal.pm:
22:45 dalek rakudo/optimizer: A doubled 'or' caught by work in the optimizer branch.
22:45 dalek rakudo/optimizer: review: https://github.com/rakudo/rakudo/commit/12b860df69
22:45 dalek rakudo/optimizer: 6bb0357 | jnthn++ | / (3 files):
22:45 dalek rakudo/optimizer: Merge branch 'nom' into optimizer
22:45 dalek rakudo/optimizer: review: https://github.com/rakudo/rakudo/commit/6bb0357811
22:45 diakopter sorear: oh
22:45 sorear niecza: say ^ 3
22:45 p6eval niecza v9-28-gdf880f1: OUTPUT«0..^3␤»
22:45 jnthn Oops, that was a bit noisy...
22:45 sorear niecza: say [ ^ 3 ]
22:45 p6eval niecza v9-28-gdf880f1: OUTPUT«0 1 2␤»
22:45 sorear niecza: say 2 ^ ^ 3
22:46 p6eval niecza v9-28-gdf880f1: OUTPUT«one(2, 0, 1, 2)␤»
22:46 sorear I think you're actually looking at a display bug
22:46 sorear niecza: say (2 ^ ^ 3).WHAT
22:46 p6eval niecza v9-28-gdf880f1: OUTPUT«Junction()␤»
22:46 sorear niecza: say (2 ^ ^ 3).FALLBACK("perl")
22:46 p6eval niecza v9-28-gdf880f1: OUTPUT«one("2", "0..^3")␤»
22:46 sorear yeah
22:47 sorear Junction.perl is putting the context in list context when it shouldn't
22:47 sorear but as for your other question - diakopter - Niecza's Junctions deliberately do no duplicate elimination
22:50 diakopter oh
22:52 diakopter niecza: say %7238957928356.WHAT # eh
22:52 p6eval niecza v9-28-gdf880f1: OUTPUT«Any()␤»
22:53 whiteknight joined #perl6
22:56 dalek niecza: eefe939 | sorear++ | lib/CORE.setting:
22:56 dalek niecza: Fix Junction-containing-list stringification (diakopter)
22:56 dalek niecza: review: https://github.com/sorear/niecza/commit/eefe939dc2
23:00 diakopter niecza: say [][][0].WHAT # sry
23:00 p6eval niecza v9-28-gdf880f1: OUTPUT«Unhandled exception: No value for parameter $index in CORE postcircumfix:<[ ]>␤  at /home/p6eval/niecza/lib/CORE.setting line 0 (CORE postcircumfix:<[ ]> @ 0) ␤  at /tmp/UYKnnyfA7h line 1 (MAIN mainline @ 1) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2…
23:05 dorlamm joined #perl6
23:06 diakopter perl6: say [][]
23:06 p6eval niecza v9-28-gdf880f1: OUTPUT«Unhandled exception: No value for parameter $index in CORE postcircumfix:<[ ]>␤  at /home/p6eval/niecza/lib/CORE.setting line 0 (CORE postcircumfix:<[ ]> @ 0) ␤  at /tmp/pWDi1Vq7Mz line 1 (MAIN mainline @ 1) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2…
23:06 p6eval ..pugs, rakudo 12b860: OUTPUT«␤»
23:08 dalek rakudo/optimizer: bfcf87a | jnthn++ | src/Perl6/Compiler.nqp:
23:08 dalek rakudo/optimizer: While in the optimizer branch, turn optimizer on by default, for easier testing.
23:08 dalek rakudo/optimizer: review: https://github.com/rakudo/rakudo/commit/bfcf87a507
23:08 dalek rakudo/optimizer: 4cba402 | jnthn++ | src/Perl6/Optimizer.pm:
23:08 dalek rakudo/optimizer: Avoid some inlines when we might screw up the workings of control exception handlers (like with given/when). This fixes the spectest regressions that we got from the immediate block inlining optimization. (We now fail some tests, but it's because the tests have undeclared routines, which the optimizer complains about at CHECK time; these tests need fixing.)
23:08 dalek rakudo/optimizer: review: https://github.com/rakudo/rakudo/commit/4cba402000
23:08 sorear diakopter: zen slices nyi
23:09 thou joined #perl6
23:09 sorear jnthn: \o/  I want those routines fixed
23:09 sorear er, tests
23:09 diakopter niecza: say ()()
23:09 p6eval niecza v9-28-gdf880f1: OUTPUT«Unhandled exception: Unable to resolve method postcircumfix:<( )> in class Parcel␤  at /tmp/0j0OVfLy4B line 1 (MAIN mainline @ 1) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2060 (CORE C969_ANON @ 2) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2061 …
23:09 jnthn sorear: Hopefully somebody will take the bait... ;)
23:09 diakopter I suppose that one's expected
23:09 jnthn sorear: Otherwise I'll get to it when I want to merge the optimizer branch.
23:10 jnthn diakopter: That one looks reasonable to me.
23:14 plobsing joined #perl6
23:15 diakopter add Parcels to the list of things I don't understand: Parcels, Captures, containers, variables, values, binding/assignment :/ I'm so behind.
23:19 diakopter niecza: say 2*****3
23:19 p6eval niecza v9-28-gdf880f1: OUTPUT«Unhandled exception: System.Exception: Unable to find lexical &term:<**> in mainline␤  at Niecza.CLRBackend.NamProcessor.ResolveLex (System.String name, Boolean upf, System.Int32& uplevel, Boolean core) [0x00000] in <filename unknown>:0 ␤  at Niecza.CLRBack…
23:19 diakopter that confuses me
23:22 diakopter why would it think ** is a term
23:23 sorear HyperWhatever
23:23 diakopter oh
23:24 diakopter niecza: say %3.WHAT # what does %3 mean
23:24 p6eval niecza v9-28-gdf880f1: OUTPUT«Any()␤»
23:24 jnthn niecza: say % 3.WHAT
23:25 p6eval niecza v9-28-gdf880f1: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤Non-declarative sigil is missing its name at /tmp/4QDqKwVJK8 line 1:â�¤------> [32msay [33mâ��[31m% 3.WHAT[0mâ�¤â�¤Unhandled exception: Check failedâ�¤â�¤  at /home/p6eval/niecza/boot/lib/CORE.setting line 685 (CORE die @ 2) â�¤  at /h…
23:25 am0c joined #perl6
23:25 jnthn not that :)
23:25 diakopter nom: say %3.WHAT
23:25 jnthn Suspect it's aliased to $3
23:25 p6eval nom 12b860: OUTPUT«Nil␤»
23:25 jnthn nom: "abcde" /(.)(.)(.)(.)(.)/; say %3
23:25 p6eval nom 12b860: OUTPUT«===SORRY!===␤Confused at line 1, near "\"abcde\" /("␤»
23:26 jnthn nom: "abcde" ~~ /(.)(.)(.)(.)(.)/; say %3
23:26 p6eval nom 12b860: OUTPUT«d␤»
23:26 jnthn yeah.
23:26 jnthn niecza: "abcde" ~~ /(.)(.)(.)(.)(.)/; say %3
23:26 p6eval niecza v9-28-gdf880f1: OUTPUT«#<match from(3) to(4) text(d) pos([].list) named({}.hash)>␤»
23:26 diakopter eh. why is it d and not c; sorry I'm so ignorant
23:27 jnthn diakopter: 0-based in Perl 6.
23:27 jnthn niecza: "abcde" ~~ /(.)(.)(.)(.)(.)/; say %0
23:27 p6eval niecza v9-28-gdf880f1: OUTPUT«#<match from(0) to(1) text(a) pos([].list) named({}.hash)>␤»
23:27 diakopter oh
23:28 * jnthn wonders if anybody can remember the C function pointer syntax, or if everyone has to look it up everytime they want it... :)
23:28 sorear I can remember it
23:29 jnthn sorear: Somehow it doesn't stick in my brane. :(
23:29 jnthn Then every time I look it up I'm like "oh, yeah, of course, silly me..."
23:30 sorear the trick is it's rather irregular
23:30 * sorear had occassion to write a pointer to a function which takes two function pointer arguments earlier this year
23:30 sorear er, three
23:30 tokuhirom joined #perl6
23:31 jnthn I bet it looked...beautiful. :)
23:34 uasi joined #perl6
23:40 diakopter niecza: no CORE
23:40 p6eval niecza v9-29-geefe939: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤Action method statement_control:no not yet implemented at /tmp/utCE_cWWU4 line 1 (EOF):â�¤------> [32mno CORE[33mâ��[31m<EOL>[0mâ�¤â�¤Unhandled exception: Unable to resolve method statement_level in class Strâ�¤  at /home/p6eval/ni…
23:42 diakopter nom: require CORE # jnthn ?
23:42 p6eval nom 12b860: OUTPUT«Null PMC access in find_method('need')␤  in <anon> at /tmp/Xp5nfvbiEb:1␤  in <anon> at /tmp/Xp5nfvbiEb:1␤␤»
23:42 jnthn diakopter: ugh
23:42 jnthn diakopter: oh, nobody re-visited require in nom yet, probably
23:43 jnthn So probably it's relying on disappeared compiler guts or something.
23:43 TimToady trying to use a setting should probably fail
23:44 TimToady that isn't their interface
23:44 TimToady likewise no should fail
23:44 jnthn makes sense
23:44 jnthn nom: use CORE; # curious *how* it fails...
23:44 p6eval nom 12b860: OUTPUT«===SORRY!===␤Could not find CORE in any of: lib, /home/p6eval/.perl6/lib, /home/p6eval/nom-inst2/lib/parrot/3.7.0-devel/languages/perl6/lib, .␤»
23:44 jnthn oh
23:44 jnthn It doesn't look for things with the .setting extension :)
23:44 TimToady settings don't work by importation, hopefully
23:45 jnthn Problem averteD!
23:45 TimToady so you can't unimport 'em either
23:45 diakopter niecza: use CORE.setting
23:45 p6eval niecza v9-29-geefe939: OUTPUT«Unhandled exception: Cannot use a lexical psuedopackage as a compile time package reference␤  at /home/p6eval/niecza/boot/lib/CORE.setting line 685 (CORE die @ 2) ␤  at /home/p6eval/niecza/src/Metamodel.pm6 line 572 (Metamodel StaticSub._lexy_ref @ 2) ␤  at…
23:46 diakopter misspelled pseudo
23:46 TimToady ooh, a bug!
23:46 jnthn std: use CORE.setting
23:46 p6eval std bb4f150: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤Confused at /tmp/2Ob_1nmgJh line 1:â�¤------> [32muse CORE[33mâ��[31m.setting[0mâ�¤    expecting statement_controlâ�¤Parse failedâ�¤FAILED 00:01 122mâ�¤Â»
23:46 jnthn nom: use CORE.setting
23:46 p6eval nom 12b860: OUTPUT«===SORRY!===␤Could not find CORE in any of: lib, /home/p6eval/.perl6/lib, /home/p6eval/nom-inst2/lib/parrot/3.7.0-devel/languages/perl6/lib, .␤»
23:46 jnthn hm
23:47 diakopter TimToady: should %1 be aliased to $1 ?
23:47 TimToady mebbe, mebbe naught
23:47 TimToady niecza: use CORE is this even parsed?
23:47 p6eval niecza v9-29-geefe939: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤Undeclared routines:â�¤     'even' used at line 1â�¤  'is' used at line 1â�¤    'parsed' used at line 1â�¤        'this' used at line 1â�¤â�¤'use' with arguments NYI at /tmp/xrUn1gjDVw line 1:â�¤------> [32muse CORE is this even
23:47 p6eval ..parsed[33m�[31m?[0m��…
23:48 TimToady niecza: use CORE + 1;
23:48 p6eval niecza v9-29-geefe939: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤'use' with arguments NYI at /tmp/oq8SluBQuK line 1:â�¤------> [32muse CORE + 1[33mâ��[31m;[0mâ�¤â�¤Unhandled exception: Check failedâ�¤â�¤  at /home/p6eval/niecza/boot/lib/CORE.setting line 685 (CORE die @ 2) â�¤  at /home/p6eval/niecz…
23:48 TimToady hmm
23:48 TimToady niecza: use CORE++;
23:48 p6eval niecza v9-29-geefe939: OUTPUT«Unhandled exception: Cannot use a lexical psuedopackage as a compile time package reference␤  at /home/p6eval/niecza/boot/lib/CORE.setting line 685 (CORE die @ 2) ␤  at /home/p6eval/niecza/src/Metamodel.pm6 line 572 (Metamodel StaticSub._lexy_ref @ 2) ␤  at…
23:49 TimToady niecza: use ++CORE;
23:49 p6eval niecza v9-29-geefe939: OUTPUT«[31m===[0mSORRY![31m===[0mâ�¤â�¤Undeclared routine:â�¤      'use' used at line 1â�¤â�¤Unhandled exception: Check failedâ�¤â�¤  at /home/p6eval/niecza/boot/lib/CORE.setting line 685 (CORE die @ 2) â�¤  at /home/p6eval/niecza/src/STD.pm6 line 1136 (STD P6.comp_unit @ 36) â�¤  a…
23:49 TimToady ah well
23:50 ponbiki joined #perl6
23:52 dalek niecza: 2b3ba1c | diakopter++ | / (2 files):
23:52 dalek niecza: fix pseudospellos
23:52 dalek niecza: review: https://github.com/sorear/niecza/commit/2b3ba1c14c
23:52 thou joined #perl6
23:53 diakopter .oO( I added myself to pseudoers today )
23:55 diakopter nom: anon package foo { }
23:55 p6eval nom 12b860: OUTPUT«===SORRY!===␤Cannot use anon scope with package at line 1, near "{ }"␤»
23:55 diakopter niecza: anon package foo { }
23:55 p6eval niecza v9-29-geefe939:  ( no output )
23:56 TimToady niecza: (anon package foo { }).WHAT.say
23:56 p6eval niecza v9-29-geefe939: OUTPUT«Unhandled exception: Unable to resolve method say in class foo␤  at /tmp/2WePzoJSRH line 1 (MAIN mainline @ 2) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2060 (CORE C968_ANON @ 2) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2061 (CORE module-CORE @…
23:56 TimToady niecza: (anon package foo { }).WHAT.gist.say
23:56 p6eval niecza v9-29-geefe939: OUTPUT«Unhandled exception: Unable to resolve method gist in class foo␤  at /tmp/qhyD7TSq3Y line 1 (MAIN mainline @ 2) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2060 (CORE C968_ANON @ 2) ␤  at /home/p6eval/niecza/lib/CORE.setting line 2061 (CORE module-CORE …
23:56 TimToady *class* foo?
23:57 TimToady decommuting &
23:58 diakopter niecza: slang P6 { } # sorear NYI?
23:58 p6eval niecza v9-29-geefe939: OUTPUT«[31m===[0mSORRY![31m===[0m��Unable to resolve method xref in class Any at /tmp/xbLF74gq4Y line 1:�------> [32mslang P6 [33m�[31m{ } # sorear NYI?[0m��Nominal type check failed for scalar store; got Any, needed Str or subtype at /tmp/xbLF74gq4Y line …

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

Perl 6 | Reference Documentation | Rakudo