Perl 6 - the future is here, just unevenly distributed

IRC log for #perl6, 2016-06-19

Perl 6 | Reference Documentation | Rakudo

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

All times shown according to UTC.

Time Nick Message
00:00 b2gills Hotkeys: Zoffix came up with a shorter version using xx 9 rather than ^9.map
00:02 Zoffix I don't think it works. There's '9' in the output
00:02 m1ndgames joined #perl6
00:05 Hotkeys m: say sum await start {sleep 7;now -ENTER now} xx 9;say now -INIT now
00:05 camelia rakudo-moar f19241: OUTPUT«63.0100307252␤7.0132424␤»
00:05 Hotkeys it works
00:05 Zoffix oh
00:05 Zoffix :}
00:05 Hotkeys m: say sum await start {sleep 7;now -ENTER now}xx9;say now -INIT now
00:05 camelia rakudo-moar f19241: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Strange text after block (missing semicolon or comma?)␤at <tmp>:1␤------> 3sum await start {sleep 7;now -ENTER now}7⏏5xx9;say now -INIT now␤    expecting any of:␤        infix␤        infix s…»
00:05 Hotkeys m: say sum await start {sleep 7;now -ENTER now} xx9;say now -INIT now
00:05 camelia rakudo-moar f19241: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Strange text after block (missing semicolon or comma?)␤at <tmp>:1␤------> 3sum await start {sleep 7;now -ENTER now}7⏏5 xx9;say now -INIT now␤    expecting any of:␤        infix␤        infix …»
00:05 Hotkeys m: say sum await start {sleep 7;now -ENTER now} xx 9;say now -INIT now
00:05 Hotkeys weird
00:06 camelia rakudo-moar f19241: OUTPUT«63.014306␤7.01916107␤»
00:12 huggable joined #perl6
00:28 AlexDaniel joined #perl6
00:31 cpage_ joined #perl6
00:32 BenGoldberg joined #perl6
00:35 mohae joined #perl6
00:54 aries_liuxueyang joined #perl6
00:59 rgrinberg joined #perl6
01:08 m1ndgames left #perl6
01:18 aries_liuxueyang joined #perl6
01:21 molaf joined #perl6
01:21 jonadab joined #perl6
01:27 FROGGS_ joined #perl6
01:28 kurahaupo joined #perl6
01:47 ilbot3 joined #perl6
01:47 Topic for #perl6 is now »ö« Welcome to Perl 6! | https://perl6.org/ | evalbot usage: 'p6: say 3;' or rakudo:,  or /msg camelia p6: ... | irclog: http://irc.perl6.org or http://colabti.org/irclogger/irclogger_logs/perl6 | UTF-8 is our friend!
01:48 Zoffix masak, interesting results (re blog post).
01:48 timotimo next step is a physics simulation to run actual trains?
01:48 Xliff Thanks timotimo.
01:49 Xliff Also, is there a way to determine if an attribute exists at runtime?
01:49 timotimo sure, you can .^attributes on your object
01:50 Zoffix m: class Foo { has $!bar = 'attr'; method !bar { 'private' }; method bar { 'public' }; method Str { $!bar ~ self!bar ~ $.bar } }.new.put
01:50 camelia rakudo-moar f19241: OUTPUT«attrprivatepublic␤»
02:03 _webstrand joined #perl6
02:07 Zero_Dogg joined #perl6
02:11 Xliff There's definitely an endless loop issue relating to type checking in rakudo.
02:12 Xliff I have a reliable piece of code that generates it, too.
02:14 Zoffix whatisit?
02:15 Zoffix Now I remember where I left off with IRC::Client... I want a restartable Promise
02:15 Zoffix (as a module)
02:20 Xliff Well dammit... it won't do it now. *sigh*
02:24 noganex_ joined #perl6
02:25 Zoffix Perl 6 Mojolicious :) http://i.imgur.com/CzX5ZgW.png
02:26 Xliff Zoffix++
02:27 Xliff Good night!
02:34 BenGoldberg joined #perl6
02:36 mr-foobar joined #perl6
02:42 Ben_Goldberg joined #perl6
02:52 Xliff_zzzz m: role B { has $.a; method set($_a) { $!a = $_a}; method A { say $.a } }; class A does B; my $A = A.new; $A.set("Aaa"); $A.A
02:52 camelia rakudo-moar f19241: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Too late for unit-scoped class definition;␤Please use the block form.␤at <tmp>:1␤------> 3 method A { say $.a } }; class A does B;7⏏5 my $A = A.new; $A.set("Aaa"); $A.A␤»
02:52 Xliff_zzzz m: role B { has $.a; method set($_a) { $!a = $_a}; method A { say $.a } }; class A does B { }; my $A = A.new; $A.set("Aaa"); $A.A
02:52 camelia rakudo-moar f19241: OUTPUT«Aaa␤»
02:54 Sgeo joined #perl6
03:18 Zoffix m: class Bar { ... }; class Foo { trusts Bar; has $!meths = <foo bar ber>;  }; class Bar { method do ($z) { .say for $z!meths } }; Bar.new.do: Foo.new
03:18 camelia rakudo-moar f19241: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤No such private method 'meths' for invocant of type 'Bar'␤at <tmp>:1␤------> 3lass Bar { method do ($z) { .say for $z!7⏏5meths } }; Bar.new.do: Foo.new␤»
03:18 Zoffix Why is it looking for meths in Bar? :S
03:18 Zoffix m: class Bar { ... }; class Foo { trusts Bar; has $!meths = <foo bar ber>;  }; class Bar { method do ($z) { .say for $z.meths } }; Bar.new.do: Foo.new
03:18 camelia rakudo-moar f19241: OUTPUT«Method 'meths' not found for invocant of class 'Foo'␤  in method do at <tmp> line 1␤  in block <unit> at <tmp> line 1␤␤»
03:18 Zoffix Yet this looks in Foo
03:20 BenGoldberg joined #perl6
03:24 Zoffix m: class Bar { ... }; class Foo { trusts Bar; has $!meths = <foo bar ber>;  method !m { $!meths } }; class Bar { method do ($z) { .say for $z!Foo::m } }; Bar.new.do: Foo.new
03:24 camelia rakudo-moar f19241: OUTPUT«(foo bar ber)␤»
03:31 Zoffix Is there a way to avoid hardcoding 'Foo' in there?
03:31 Zoffix m: class Bar { ... }; class Foo { trusts Bar; has $!meths = self.^name.lc;  method !m { $!meths } }; class Bar { method do ($z) { .say for $z!Foo::m } }; class Meow is Foo { has $!meths = 'not-meow'; }; Bar.new.do: Foo.new; Bar.new.do: Meow.new;
03:31 camelia rakudo-moar f19241: OUTPUT«foo␤meow␤»
03:31 Zoffix I want the result here to be 'foo not-meow'
03:34 wibble joined #perl6
03:36 Zoffix bisect: m: class Bar { ... }; class Foo { trusts Bar; has $!meths = <foo bar ber>;  }; class Bar { method do ($z) { .say for $z!meths } }; Bar.new.do: Foo.new
03:36 bisectable Zoffix: on both starting points the exit code is 1 and the output is identical as well
03:36 Zoffix bisect: m: class Bar { ... }; class Foo { trusts Bar; has $!meths = <foo bar ber>;  method !meths { $!meths } }; class Bar { method do ($z) { .say for $z!meths } }; Bar.new.do: Foo.new
03:36 bisectable Zoffix: on both starting points the exit code is 1 and the output is identical as well
03:39 Zoffix m: class Bar { ... }; class Foo { trusts Bar; has $!meths = self.^name.lc;  method !m { $!meths } }; class Bar { method do ($z) { .say for $z!::($z.^name)::m } }; class Meow is Foo { has $!meths = 'not-meow'; }; Bar.new.do: Foo.new; Bar.new.do: Meow.new;
03:39 camelia rakudo-moar f19241: OUTPUT«===SORRY!===␤cannot stringify this␤»
03:41 Zoffix m: use MONKEY-SEE-NO-EVAL; class Bar { ... }; class Foo { trusts Bar; has $!meths = self.^name.lc;  method !m { $!meths } }; class Bar { method do ($z) { .say for EVAL "\$z!$z.^name()::m" } }; class Meow is Foo { trusts Bar; has $!meths = 'not-meow'; method !m { $!meths } }; Bar.new.do: Foo.new; Bar.new.do: Meow.new;
03:41 camelia rakudo-moar f19241: OUTPUT«foo␤not-meow␤»
03:41 Zoffix I can't use this. Is there a saner way to achieve the same result?
03:44 Zoffix m: class Foo { has $.gibblets = <foo bar ber>; }; my $f = Foo.new; say $f.gibblets; $f but role { method gibblets { die 'Tis private!' } }; say $f.gibblets
03:44 camelia rakudo-moar f19241: OUTPUT«(foo bar ber)␤(foo bar ber)␤»
03:45 Zoffix *sigh* there gotta be a way to have private methods for use by other classes in my distro that aren't available to the users of my distro :(
03:47 Zoffix Or remove/disable an existing method on a specific object :\
03:47 Zoffix m: class Foo { has $.gibblets = <foo bar ber>; }; my $f = Foo.new; say $f.gibblets; $f = $f but role { method gibblets { die 'Tis private!' } }; say $f.gibblets
03:47 camelia rakudo-moar f19241: OUTPUT«(foo bar ber)␤Tis private!␤  in method gibblets at <tmp> line 1␤  in block <unit> at <tmp> line 1␤␤»
03:47 Zoffix Ahhh
03:47 Zoffix Well, ain't perfect but at least something
03:47 Zoffix Digital duck tape
03:53 khw joined #perl6
04:10 Cabanossi joined #perl6
04:12 huggable joined #perl6
04:15 grondilu joined #perl6
04:17 wamba joined #perl6
04:28 kaare_ joined #perl6
05:54 gfldex Zoffix: pod5 got F<> for filenames
05:59 ChoHag joined #perl6
06:10 jjido joined #perl6
06:20 domidumont joined #perl6
06:22 CIAvash joined #perl6
06:25 domidumont joined #perl6
06:38 mr-foobar joined #perl6
07:06 Actualeyes joined #perl6
07:08 firstdayonthejob joined #perl6
07:18 jjido joined #perl6
07:22 rindolf joined #perl6
07:30 BenGoldberg joined #perl6
07:53 grassass joined #perl6
07:53 lizmat joined #perl6
08:05 mr-foobar joined #perl6
08:12 huggable joined #perl6
08:16 rurban_ joined #perl6
08:33 grondilu joined #perl6
08:36 darutoko joined #perl6
08:44 setty1 joined #perl6
08:52 kid51 joined #perl6
09:04 spider-mario joined #perl6
09:08 dalek doc: cc46aaa | (Steve Mynott)++ | doc/Language/operators.pod:
09:08 dalek doc: fix minor typo
09:08 dalek doc: review: https://github.com/perl6/doc/commit/cc46aaa8f5
09:25 RabidGravy joined #perl6
09:26 RabidGravy boom
09:44 cognominal joined #perl6
09:46 cognominal m, for 1..20  { print $_, Int.new($_) =:= Int.new($_) ?? '' !! '!', ' ' }
09:48 cognominal m: for 1..20  { print $_, Int.new($_) =:= Int.new($_) ?? '' !! '!', ' ' }
09:48 camelia rakudo-moar f19241: OUTPUT«1 2 3 4 5 6 7 8 9 10 11 12 13 14 15! 16! 17! 18! 19! 20! »
09:49 cognominal I suppose there is some form of caching going on here.
09:50 cognominal instanciating Ints below 15 gives an existing Int
09:53 cognominal j: for 1..20  { print $_, Int.new($_) =:= Int.new($_) ?? '' !! '!', ' ' }
09:53 camelia rakudo-jvm 40a953: OUTPUT«cannot connect to eval server: Connection refused␤»
09:55 araujo_ joined #perl6
09:57 araujo_ joined #perl6
10:00 sno joined #perl6
10:00 cognominal https://github.com/MoarVM/MoarVM​/blob/master/src/core/intcache.c  indeed thru a nqp::box_i called in the new method of Int class
10:02 araujo_ joined #perl6
10:06 timotimo yeah, that's right
10:06 timotimo those ints are incredibly common
10:06 timotimo we also do it for -1 i believe
10:11 cognominal the mystery is there is 4 type of ints according to the code.
10:12 cognominal m: for -3..20  { print $_, Int.new($_) =:= Int.new($_) ?? '' !! '!', ' ' }
10:12 camelia rakudo-moar f19241: OUTPUT«-3! -2! -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15! 16! 17! 18! 19! 20! »
10:19 cognominal at the moar level, there are P6int and P6Bigints. Probably under the hood, there are many sorts of bigints?
10:32 Util Zoffix: I have not read the thread past the long first post, but this might help:
10:32 Util http://www.nntp.perl.org/group/perl.​perl6.language/2010/03/msg33409.html
10:32 Util "You never have privacy from your children in Perl 6"
10:35 FROGGS o/
10:37 Zoffix Util, what is that about?
10:37 Zoffix gfldex, yeah, ok, but why are those filenames in that page?
10:38 Util Zoffix: in re: "there gotta be a way to have private methods for use by other classes in my distro that aren't available to the users of my distro"
10:38 Zoffix Ahh
10:38 Util Zoffix: Just woke up
10:39 ChoHag_ joined #perl6
10:39 * Zoffix too
10:39 Zoffix Actually, I'm going back in... 6:40AM
10:49 _4d47 joined #perl6
10:49 pmurias joined #perl6
10:50 timotimo cognominal: actually, p6int is the bigint
10:52 jnthn ...um, no? :) P6int is the REPR of native ints. P6bigint is the REPR of bigints :)
10:53 jnthn In reality you never see a naked object with REPR P6bigint in Perl 6, though; Int has it flattened into a P6opaque.
10:53 cognominal so why 4 kind of *Ints cahed
10:53 dalek snake: eebd456 | (Pawel Murias)++ | src/Snake/Actions.nqp:
10:53 dalek snake: Unbitrot.
10:53 dalek snake: review: https://github.com/arnshol​t/snake/commit/eebd456e84
10:54 jnthn cognominal: Cached where, exactly?
10:54 timotimo er, derp.
10:54 timotimo that's because there's BOOTInt and Int, for example
10:55 gfldex Zoffix: those ENV vars are an implementation detail, if you go that deep you might as well show where they come from. In my eyes they should be removed. They are quite long and ack can find them nicely.
10:58 iH2O joined #perl6
10:58 cognominal jnthn, https://github.com/MoarVM/MoarVM/b​lob/9c885bce8f65b969f52d82ef7fda7f​eb16b464f5/src/core/intcache.c#L7
10:59 cognominal there is 4 "types" of *ints in the int cache
10:59 jnthn Oh, that... It doesn't mean there are 4 types, it just means it can hold up to 4 types
11:00 cognominal ok
11:00 jnthn I don't know why 4 was picked...I think maybe timotimo put the cache in and may remember... :)
11:00 jnthn Certainly NQP's integer type and Perl 6's Int make 2
11:00 cognominal so, I can create my fancy ints and have -1..14 cached
11:01 jnthn I guess some of the allomprphs may end up in there...
11:02 cognominal anyway, probably that discussion probably belongs to #perl6-dev :)
11:02 timotimo anything that we box_i on; do we box_i into our allomorph classes?
11:02 jnthn timotimo: Not sure...we may :)
11:02 jnthn I mean, it'd be a valid way to implement it. :)
11:02 cognominal jnthn++ timotimo++ for the kind answers
11:03 timotimo P6bigint can't hold mixins, so that's not a concern for that cache
11:04 dalek snake: 2f89820 | (Pawel Murias)++ | Makefile:
11:04 dalek snake: Fix Makefile by adding missing dependencies.
11:04 dalek snake: review: https://github.com/arnshol​t/snake/commit/2f89820d87
11:07 masak ahoy, #perl6
11:07 masak ooh, a snake commit
11:08 masak arnsholt: I remember I promised-ish to contribute to snake -- what's the status of the project right now?
11:08 jnthn ahoj, masak
11:09 masak ahoĵ
11:09 jnthn .oO( Status: need to badger more people :P )
11:09 * masak .oO( there's no shroom for more puns )
11:09 timotimo also mushroom some more people
11:09 jnthn I'm sure there's mush room for inprovement :P
11:09 timotimo %)
11:09 jnthn hah :P
11:10 masak damaged minds think alike
11:13 iH2O obviously, all mind damages are alike
11:14 hahainternet what is this? #andyzaltzman?
11:15 iH2O is the trailing "?" really part of the channels name?
11:15 hahainternet i'd like to claim it's the name of the person, but even andy doesn't have such bombastic punctuation
11:16 iH2O no
11:17 masak for those who missed my blog posting late last night: http://strangelyconsistent.org/blog/train-tracks
11:18 iH2O ive never been good with knots
11:19 iH2O esp those uninvertible
11:19 hahainternet knots are really enjoyable
11:19 hahainternet especially the few you tie with a big inversion
11:19 hahainternet like the figure of eight loop
11:19 iH2O too much shrooming haha
11:19 hahainternet masak: i can't help but think this is scalextric track :)
11:20 masak "scalextric track" -- is that something I should know about? :)
11:20 hahainternet masak: https://2e7fd430838d304f1516-467f5d9f2ca7​b7b12f8a116e60ea9c1d.ssl.cf3.rackcdn.com/​catalog/product/cache/5/image/9df78eab335​25d08d6e5fb8d27136e95/c/1/c1319-scalextri​c-continental-sports-cars-set-track.jpg
11:21 masak oh, oh no. this is wooden trains, not cars.
11:21 hahainternet yeah i saw your real photo
11:21 hahainternet but you have done an impressively good job of reproducing a scalextric track in your diagrams too
11:21 masak there's no electricity involved. only small magnets keeping the train cars together.
11:21 hahainternet and they have almost exactly the same mechanism, with slightly less 'wiggle'
11:21 hahainternet you get almost exactly the same track layout in the box
11:21 hahainternet and you can produce approximately the same versions you showed
11:22 _4d47 left #perl6
11:22 hahainternet so, you didn't even realise it but you built a model for an entirely different system that mostly conforms :)
11:22 masak ah. wiggle, my nemesis.
11:23 hahainternet masak: scalextric tracks attach with two points at the joints, at the far edges, using a simple spring nub that fits into a hole
11:23 hahainternet masak: one thing you haven't accounted for that is possible, is track banking
11:24 masak doesn't come up with these wooden tracks
11:24 hahainternet only at the scale you're working on
11:24 masak but yeah. *shudder*
11:24 hahainternet it requires only torsion, so you can't avoid it :p
11:24 hahainternet i'm impressed by your vector space plot though, not quite sure how you did that, i'm no mathematician
11:26 masak if there's something I can do to help by clarifying, I'd be happy to do so
11:26 masak maybe it'd improve the blog post, too
11:27 hahainternet masak: well you don't label anything about it lol
11:28 hahainternet so, i'm not sure what the numbers represent, piece count?
11:28 hahainternet i'm not sure why there's an irregular grid
11:28 hahainternet i'm not sure why only a minority of marks are unlabelled
11:28 masak those unlabeled are just a single solution at that point
11:28 domidumont joined #perl6
11:29 masak each point (x, y) represents a particular displacement -- for example (0, 0) has the 10 exact solutions
11:29 hahainternet oh i see they're number of solutions at that position
11:29 masak aye
11:29 masak that's how you get a total of 380
11:29 hahainternet i'd note that using a 'luggage tag' shape may be a bad choice
11:29 hahainternet and adding a suffix 'x' is common
11:29 hahainternet 10x is many many times clearer than |10|
11:29 masak oh, that's a good suggestion
11:30 iH2O 10x or x10?
11:30 hahainternet are these top down X/Y offsets for the track end masak?
11:30 iH2O arent both confusing with hex?
11:30 hahainternet iH2O: either is fine, 'x' is read as 'times'
11:30 hahainternet so it's '10 times' or 'times 10', either is perfectly fine
11:30 hahainternet hex typically requires a leading 0 also
11:30 iH2O so its 10 in hex notation
11:30 hahainternet 0x10 is, x10 is not
11:30 iH2O which is 15?
11:31 iH2O good, im reassured
11:31 hahainternet brb 2 moments
11:32 masak re irregular grid: because of the 45-degree angles, the constant 1/sqrt(2) creeps into the coordinates
11:32 masak and so any given displacement coordinate is M + N/sqrt(2), M and N integers
11:32 masak all the thin lines in that diagram represent such M + N/sqrt(2) coordinates
11:32 masak the thicker lines are integral coordinates
11:33 iH2O why does perl6 implement sqrt(complex number), or any complex arithmetic. 99.99% of the folks  will never use that in their lifetime
11:33 iH2O ..
11:35 masak iH2O: that seems an odd complaint to me. are you upset that something is implemented?
11:35 iH2O folks, concentrate on the real stuff
11:35 AlexDaniel joined #perl6
11:35 iH2O implement the real stuff like parallel processing
11:35 iH2O i have 4 cores
11:35 * masak .oO( folks, concentrate on the real plus the imaginary stuff )
11:35 iH2O i need parallel processing, all of it
11:36 iH2O the "i" stuff is for apple
11:36 iH2O not for perl6
11:36 masak iH2O: oh, I get it. you're here to tell people what to spend their time and energy on.
11:36 iH2O just saying...
11:36 iH2O :)
11:37 timotimo you do realize that complex arithmetic is a billion times easier and faster to implement than any kind of multithreaded stuff?
11:37 iH2O im sure it is
11:37 gfldex m: await do start { say "your point is?" }, do start { say "works nicely for me" }
11:37 camelia rakudo-moar 53af02: OUTPUT«your point is?␤works nicely for me␤»
11:38 * hahainternet returns
11:38 hahainternet masak: it makes much more sense to me now
11:39 hahainternet masak: i feel that the asymmetry of the graph is now a problem though
11:40 hahainternet the centre being (0,0)
11:40 hahainternet i fail to comprehend how the graph should not be a reflection across x=0 or y=0
11:41 masak basically because of the symmetry breaking I'm doing on the tracks
11:42 masak let's say a track is a string like BBLLLLLSSRRRRR
11:42 hahainternet but if your initial track is oriented at a cardinal direction, and all tracks are equally 'wiggly', then the graph should exhibit symmetry regardless
11:42 hahainternet as all Ls and all Rs can simply be swapped
11:42 masak then that track's representative is the lexicographically first among these: the track itself, same track opposite direction, its mirror image, mirror image in the opposite direction
11:43 hahainternet oh, so you're culling potentials before you get to the graph, so selecting for those with x>0?
11:43 masak this symmetry breaking greatly favors tracks that run counter-clockwise. and so you have a lack of symmetry in the graph.
11:43 hahainternet masak: i think that's a problem tbqh
11:44 hahainternet without meaning to tell you how to do your blog
11:44 masak it gives me each solution only once, which I like
11:44 masak that's more important to me than symmetry in the graph :)
11:44 hahainternet it means that your solution count labels are inaccurate though
11:44 hahainternet it means the graph requires orientation you don't denote
11:44 masak but your point is well taken -- because it means the signs of the displacement are quite meaningless
11:44 hahainternet and it makes it generally harder to understand, as you could begin to talk about the symmetry
11:44 Actualeyes joined #perl6
11:44 hahainternet without the symmetry, the graph is considerably harder to understand
11:44 masak I could plot it all as just one quadrant instead
11:44 hahainternet (imho)
11:45 hahainternet masak: that, and perhaps some sort of tesselation, voronoi etc, would make a lot of sense to me
11:45 hahainternet a terrain map would be preferable
11:45 hahainternet where Z is joining force
11:45 hahainternet or 'difficulty' of joining
11:45 hahainternet but the 'islands of connectivity' to butcher a phrase would be interesting to see
11:45 masak I enjoy all your ideas
11:46 iH2O haha is genial
11:46 masak yeah, there's quite a fair bit to be done in order to fully understand all the non-exact solutions
11:46 hahainternet iH2O: i'm mostly useless, so when i can give somewhat valid criticism i'm super happy
11:52 iH2O you'll write the 1st program with real *artificial intelligence*, in perl6
11:53 hahainternet i think "artificial intelligence" is a term we should stop using now, because it's quite offensive, "machine intelligence" is the PC term :)
11:53 iH2O none exists at this time
11:54 iH2O they havent even solved the Turing test, theyre clueless
11:54 hahainternet iH2O: the logs of this channel aren't going away though
11:54 iH2O why? they have historical significance i think
11:54 hahainternet so when we create a viable machine intelligence, it will review history, and see us talking about it derisively, as only 'artificial'
11:54 domidumont joined #perl6
11:55 hahainternet but a machine intelligence is no more artificial than i am :)
11:55 iH2O they wont be any real machine intelligence or artificial intelligence before the year 2100
11:55 iH2O *there
11:55 hahainternet that's one hell of a claim iH2O!
11:56 iH2O why? theyre totally clueless about solving the Turing test
11:56 iH2O the hard problem of intelligence
11:56 iH2O theyve no clue
11:57 iH2O z....
11:57 dalek doc: 3f9f9aa | (Tom Browder)++ | htmlify.p6:
11:57 dalek doc: sort programs files by file name for desired order
11:57 dalek doc: review: https://github.com/perl6/doc/commit/3f9f9aac4e
11:57 dalek doc: 3c5ffe4 | (Tom Browder)++ | htmlify.p6:
11:57 dalek doc: Merge pull request #605 from tbrowder/progs-index
11:57 dalek doc:
11:57 dalek doc: sort programs files by file name for desired order
11:57 dalek doc: review: https://github.com/perl6/doc/commit/3c5ffe4c7c
11:57 hahainternet iH2O: i believe the topic is significantly more complex, i am not sure that free will exists, and i believe that building the mechanisms of the brain will cause intelligence to arise as emergent behaviour
11:57 gfldex you define artificial intelligence with "human like". That's quite narrow (in more then one sense).
11:58 masak also, there's a lot more to machine intelligence than the Turing test
11:58 iH2O haha, the human brain is a kludge loaded with cognitive biases/dissoances
11:58 iH2O you dont want to replicate that
11:58 brrt joined #perl6
11:58 iH2O what you want is clean, high level knowledge processing
11:59 hahainternet iH2O: it's not clear such a thing exists
11:59 iH2O AI papers do their best to figure it, i read them once in a while
11:59 brrt i'm fairly sure it doesn't exist
11:59 yoleaux 8 Jun 2016 09:53Z <Ven> brrt: interesting LuaJIT talk by one of the current DartVM engineers: http://mrale.ph/talks/vmss16
11:59 brrt .tell Ven thanks
11:59 yoleaux brrt: I'll pass your message to Ven.
11:59 hahainternet brrt: which ?:)
11:59 brrt 'clean high-level knowledge processing'
12:00 iH2O i glad you approve brrt
12:00 iH2O im
12:00 brrt well, let me phrase my complaint as accurately as i can
12:01 brrt what we call 'intelligence' is nothing more than a vague collective idea of 'what autonomous living organisms do to control the circumstances of their lives'
12:01 hahainternet one thing that amuses me, is that when scifi shows portray future computers, they're almost always sleek electronic / mechanical devices, but the most advanced computers we're aware of are all completely organic
12:01 iH2O i think there is also a transcendent side to it
12:02 iH2O besides what you mention
12:02 brrt we, as well as most of the kingdom of animalia, have hardwired processing facilities for several tasks
12:02 brrt iH20: I think there isn't, that's my point
12:02 iH2O O_O
12:02 iH2O we're not fancy automata
12:02 hahainternet i'm looking forward to my PC in 60 years, i'll feed it and myself with a glass of soylent each morning :)
12:02 iH2O we're transcendent
12:02 brrt if you are willing to give up that idea, then all of 'intelligence' etc becomes part of the regular world, the physical world
12:02 hahainternet iH2O: i disagree
12:03 hahainternet we're nothing but complex simultaneous modelling computers that very quickly notice differences with our models and contain a primitive high level reasoning facility to determine the cause of these differences
12:04 brrt if you're willing to give up transcendence, then conciousness becomes explicable
12:04 iH2O brrt: we can recognize ourselves in the mirror, which is already more than what you suggest
12:04 iH2O there is a "meta" part in us
12:04 hahainternet iH2O: that's just a level of cognizance, we're social animals and so it's practically required
12:04 hahainternet there's no evidence of a meta part in us
12:04 hahainternet in fact, the opposite is true
12:04 iH2O tsk tsk
12:04 hahainternet by studying your brain in a PET machine
12:04 hahainternet i can determine your response to yes/no questions
12:04 hahainternet approximately 2 seconds before you are aware of making the decision
12:05 brrt lots of animals can recognize themselves in a mirror
12:05 iH2O only a few
12:05 hahainternet very few
12:05 iH2O very few
12:05 hahainternet three or four total
12:05 iH2O yes
12:05 brrt lots of animals which can't are not biased on visual inuts
12:05 iH2O elephants, crows
12:05 brrt lots of birds can
12:05 hahainternet some apes
12:05 iH2O which indicates that this ability is "special", beyond ordinary knowledge processing
12:06 hahainternet iH2O: it just shows specialisation
12:06 hahainternet iH2O: babies can, for example, recognise facial differences between members of non human species
12:06 hahainternet but by the time you are 2, you lose this ability
12:06 brrt no, i disagree fully with the first premise
12:06 iH2O hmm
12:07 brrt recognising yourself in the mirror is a specific instance of recognising peers in the mirror
12:07 iH2O absolutely not
12:07 brrt this is only relevant for social animals which operate visually
12:07 iH2O its more
12:07 brrt yes, yes it is
12:07 hahainternet indeed, but it requires a certain level of abstract thought, but this is likely built off a physical mechanism, not a transcendental 'meta' mechanism
12:08 iH2O lol
12:08 hahainternet there's an organisation of cells in our brain that allows us to reference images in a mirror vs tagged sensory data on 'ourselves'
12:08 iH2O this is a high level ability. you suggest it is low level
12:08 brrt no
12:09 Actualeyes joined #perl6
12:09 iH2O s/no/yes
12:09 hahainternet iH2O: what i'm saying is, what we see as 'high level' is just emergent behaviour from a brain very very similar to all others
12:09 FROGGS :D
12:09 hahainternet the difference between humans and dogs for example, will be far far smaller than we'd be comfortable with
12:09 hahainternet imagine it in code, a language without $self and a language with it
12:10 hahainternet you can accomplish many of the same things regardless
12:10 brrt i reject the notion that there is such a thing as high and low level
12:10 hahainternet the mechanisms don't even change very much
12:10 hahainternet brrt: in the brain, or conceptually?
12:10 hahainternet because conceptually i don't think you've an argument to make
12:10 brrt in the brain
12:10 hahainternet brrt: well we have actual evidence of these high and low level processes
12:10 brrt i'm not able to continue the discussion at this point
12:10 brrt see you later :-)
12:10 hahainternet ciao
12:12 huggable joined #perl6
12:12 hahainternet anyhow, my favourite assertion i like to make is that of organic computers
12:12 hahainternet many people have a scifi, startrek view of the future
12:12 hahainternet where advanced medicine is tiny mechanical robots injected into your bloodstream
12:13 hahainternet but in reality we know how effective programmed biology is
12:13 hahainternet so when people see the future as VR goggles they strap on
12:13 hahainternet i see it as a squishy lump of meat you get injected
12:14 * masak .oO( that's all -- enjoy your lunch! )
12:14 hahainternet :D
12:14 hahainternet there's no good reason to think that the future of computers will be anything but biological though
12:14 hahainternet in fact, you can argue it might not even be possible to make a brain without biology
12:14 hahainternet it may be that fundamentally, non biological machine intelligences are not possible
12:15 iH2O intelligence has some *magic*
12:15 hahainternet [citation-needed] :p
12:15 iH2O we gotta to inject that into machines
12:17 masak I think describing it as "magic" is selling intelligence short
12:18 hahainternet i actually think it's doing the opposite, hyping it up as something notable
12:20 hahainternet my view is this, there's no evidence for any mechanism by which 'free will' can occur, without the concept of free will, intelligence is just rapid environmental modelling and historical correlation to find a best course
12:20 hahainternet and that is completely describable as a computable process right now, we just don't have the processing volume to test it
12:20 hahainternet and if that is the case, 'intelligence' is an illusion, our deluded perception of a continuous modelling process we don't even have access to
12:21 hahainternet our decisions bubble up from this mechanism and we, the deluded people living in our own brains, grasp on them as our own
12:22 cognominal this thread is interesting but does it belong to #perl6?
12:22 _mg_ joined #perl6
12:23 hahainternet cognominal: probably not, but it was quiet, feel free to get us on topic :)
12:23 AlexDaniel cognominal: well, if we're planning to change the world with perl 6, then… maybe? :D
12:23 iH2O cognominal: #AI is the quietest room on IRC, nothing ever happens there :(
12:23 iH2O i usually go there when i need a nap z...
12:24 masak I don't like the tendency to say "we figured out all the constituent parts, so X is an illusion"
12:24 telex joined #perl6
12:24 masak you can stop before the comma and things are still just as nice
12:24 masak and it leaves you with one more term that you can use and that people will understand
12:25 hahainternet masak: i don't have a problem with the term, i'm just pointing out how 'free will' is the part we find the biggest concern
12:25 hahainternet but there's literally no evidence for it
12:25 hahainternet and if we look at a person without assuming free will, the process of their brain is thoroughly understandable, and nothing but computing technology prohibits us recreating it
12:26 hahainternet when we do that, we will produce the first Machine Intelligence (is my contention)
12:27 masak I'm fine with leaving "free will" completely aside. I'm not sure what a conclusion in the free will debate would even look like.
12:27 masak "yes, we have free will!" -- "oh, great. I'll use mine to grab a cup of coffee."
12:27 masak "no, we don't have free will!" -- "that's too bad. I'll go get a cup of coffee."
12:28 hahainternet from my perspective, the conclusion would be linking of intelligence to some underlying physics mechanism, quantum indeterminacy being the obvious choice
12:28 pmurias hahainternet: "nothing but computing technology", doesn't that apply to a lot of the far future scifi things?
12:28 hahainternet my argument is that free will behaves as potentiation of various options
12:28 hahainternet pmurias: not at all, transporters are fundamentally impossible
12:28 hahainternet shields, phasers, all these sorts of things are likely impossible
12:28 hahainternet machine intelligences though? we'll be swimming in them before long
12:29 hahainternet oh, and another impossible thing i'm sick of people bringing up, holographic displays
12:29 hahainternet get over it hollywood, nobody but a designer would ever want one, and augmented reality is a better solution to that
12:29 cognominal Some strong deniers of free will have a strong will to prove their point
12:29 hahainternet i should really blog about light field displays in order that i can just link that
12:29 hahainternet cognominal: i enjoyed the irony in that :)
12:30 masak no irony. it's a strong non-free will, duh
12:30 hahainternet in fact
12:30 hahainternet i just realised
12:30 hahainternet a VR headset is perfect to reproduce a video of a light field display
12:30 hahainternet wonder if i can do the same with a phone's accel/gyro combo
12:31 FROGGS why not a more direct video "projection"? like, transmitting data directly into your head?
12:31 hahainternet FROGGS: nothing wrong with that, and that will come fairly soon, i just hate 'holographic displays' and they're everywhere at the moment
12:32 FROGGS I mean, you may also provide a solution for blind ppl
12:32 hahainternet endless people fawning over something few would ever want
12:32 cognominal So strong deniers of free will are just predestined to be,,, to get to another hot subject, predestination
12:32 FROGGS hahainternet: still, jarvis is awesome :o)
12:32 hahainternet what they actually would want is a 'light field display', but i've seen virtually no talk about them
12:32 hahainternet FROGGS: indeed the most impressive technology in those movies by a very long way
12:32 hahainternet cognominal: indeed
12:33 FROGGS I'd buy a voice-only jarvis system too, no doubt
12:34 pmurias hahainternet: the stuff in movies is supposed to look cool not actually be usefull/possible
12:34 hahainternet pmurias: i agree, but the meme has made its way into games and 'serious' discussion
12:34 FROGGS remember that joystick that riker used in star trek 10? >.<
12:35 hahainternet every futuristic setting now includes endless holographic displays
12:35 hahainternet drives me nuts! :D
12:35 masak cognominal: "predestined to be dree-will deniers" reminds me of the saying "how magnanimous of God to even create atheists"
12:36 pmurias masak: re blog post the rail tracks renderings look cool
12:36 hahainternet hmm i've realised i'm not very clear on this point, i assume everyone here gets what a 'light field display' is?
12:37 cognominal masak, yea
12:38 iH2O i think you can do away with predestination if you assume that each person has his own individual timeline
12:38 FROGGS hahainternet: probably not
12:38 cognominal and kind of mischevious creating believers of one true god, but not the same.
12:38 masak pmurias: thanks. countless hours wasted on drawing them :P
12:39 hahainternet FROGGS: a display with control over the angle of emission of photons as well as the wavelength, this allows it to display depth
12:39 masak pmurias: made slightly worse by the fact that little Noa helpfully rebooted my computer in such a way that I lost my first attempt.
12:39 hahainternet it would appear as a virtual world extending into your screen with 'real' depth
12:39 FROGGS hahainternet: inst that what the star trek holodecks are about?
12:40 hahainternet FROGGS: to an extent
12:40 hahainternet they also can magically cause those photons to be emitted from a 'forcefield' created in free air
12:40 hahainternet which is a bit over the top
12:40 hahainternet but simply imagine an FPS game, where you can move your head left and right to see left and right
12:40 hahainternet that is a plausible future of displays within the next decade
12:41 hahainternet holographic displays on the other hand would be translucent non physical nonsense
12:41 FROGGS ahh, I see
12:41 iH2O do you at least have a 3D LED monitor for now, ahah
12:41 hahainternet you can easily imagine a reverse lytro camera, if you understand how they work
12:42 hahainternet in reality (and ironically) the mechanism to create these displays will likely be 'holographic'
12:42 colomon joined #perl6
12:42 hahainternet as in, it will use diffraction gratings rather than micro lenslet arrays
12:42 iH2O progress never stops
12:42 moritz masak++ # blog
12:43 hahainternet yeah sorry for the big diversion, masak despite my criticisms, your blog is excellent :)
12:43 masak I wish I could blog more
12:43 moritz next up: ball paths :-)
12:43 masak moritz: aren't those just parabolas?
12:44 hahainternet masak: perhaps you answered it in the blog and i missed it, how did you model the joint flexibility?
12:44 pmurias hahainternet: I can imagine things that "look like" holographic displays from scifi/games being used in augumented reality so that things aren't real can be distinguished
12:44 masak hahainternet: I didn't -- I skipped that bit
12:45 masak hahainternet: instead of modeling wiggle, I just count the eventual displacement
12:45 hahainternet masak: right ok, i see now that the graph includes impossible ones
12:45 hahainternet my mistake
12:45 moritz masak: either my dictionay mistranslated, or the term is completely ambiguous
12:45 masak which term?
12:45 hahainternet pmurias: that'd be a funny one, but i am mostly just against the meme, i'd rather get people talking about light field displays, as they make so much more sense
12:46 moritz masak: ball path; I meant something like this: http://www.meinbaukasten.de/media/catalog/produc​t/cache/1/image/1000x1000/040ec09b1e35df13943388​7a97daa66f/k/a/kaden-kugelbahn-11-traversen.jpg
12:46 masak moritz: ooh!
12:46 masak er, I mean... "noooooooo!"
12:46 masak :P
12:46 hahainternet pmurias: imho we're likely to see glove type AR/VR interfaces within a decade or so, so that's a scenario where a 'hologram' would be useful, but it wouldn't be a hologram so, it's a bad term
12:46 masak I just just free of my last obsession!
12:46 hahainternet those actually seem very straightforward
12:47 hahainternet you can probably get away with a tiny bit of 'wiggle' and a 3d vector sum?
12:47 hahainternet i suppose you have the problem of calculating entry energy
12:47 hahainternet but that would be true of masak's trains, so you can just assume constant speed
12:49 masak moritz: those look lovely.
12:50 hahainternet masak: are you aware of 'marble run' toys? :D
12:50 hahainternet model this: https://www.youtube.com/watch?v=UxcJfaoK5xg
12:50 jast here you have to keep track of momentum. the train doesn't need to move on its own, but the marbles do, given an initial impulse
12:50 masak I'm not modeling the train at all. just focusing on the tracks and the possible ways to build them.
12:51 hahainternet jast: the problem masak was tackling wasn't 'ball paths' as such, but the potential arrangements in space
12:51 hahainternet ^^
12:51 masak basically a combinatorics problem.
12:51 jast sure, I read that.
12:51 jast my point is a train track in which trains don't have enough momentum is still useful, a marble slide not nearly as much
12:51 jast so the same simulation is quite a bit less interesting for the slide
12:51 hahainternet you can easily simplify the modelling though
12:52 hahainternet there's only one unrestricted curve
12:52 hahainternet so you can calculate the maximum velocity permissable
12:52 hahainternet and then model at that
12:52 hahainternet as there's no restrictions on entry energy, and as long as it's not a significant mach fraction it's good enough :)
12:53 finanalyst joined #perl6
12:53 hahainternet i suppose if you allow balls to fall off instead of using the z axis blocks that adds a significant challenge, but i'm just saying there are a lot of simplifying assumptions possible
12:53 jast sure
12:53 jast that marble machine is crazy btw
12:54 hahainternet it is, i've always wanted to build some big rube goldberg machine
12:54 jast so, a computer basically? ;)
12:55 hahainternet no no, just a ludicrously overcomplicated machine for a simple job
12:55 hahainternet like this marble machine could be replaced by a single straight pipe
12:55 jast isn't that what I just said? :P
12:55 hahainternet i suppose so :)
12:55 hahainternet being able to model that marble machine correctly would be such an unbelievably huge job people rarely comprehend it
12:55 jast not really something I'd be interested in, but it's definitely fun to see the result
12:56 hahainternet you'd require a datacentre of computers for a very very long time
12:56 jast depends on what level of accuracy you're after
12:56 hahainternet velocities are easily high enough that rigid body simulation would be woefully inadequate
12:56 hahainternet jast: nah not in any meaningful way
12:56 hahainternet it's into the range where damage is being done constantly and energy is being lost significantly in heating
12:56 hahainternet so you're basically screwed from step #1
12:57 jast well, there's always toy-level simulation
12:57 jast with cheating and such
12:57 iH2O haha: do you have a PhD in cybernetics?
12:57 hahainternet i am completely unqualified
12:57 hahainternet by US standards, and new UK standards, i didn't finish high school
12:58 hahainternet if i'm talking authoritatively, i apologise, everything i've said is my opinion
12:58 jast qualifications on paper = boring
12:58 iH2O = theoretical
12:58 masak until you're applying for a job
12:59 jast I have a BSc in $field and know just enough to know how much I don't know
12:59 jast (and to know, a bit later, that I didn't actually want to finish the Master's programme)
12:59 iH2O jast = just enough
12:59 hahainternet i wish i had gone to university because there's a lot of things i don't understand and would like to
13:00 hahainternet but i'm in my early 30s, so i still have time to learn things
13:00 jast there's always self-study, and talking to people who know stuff
13:00 jast most of the stuff I was "taught" at university I actually learned from the internet
13:00 iH2O from wikipedia, right
13:00 hahainternet one of my favourite quotes is "knowlege doesn't weigh anything" .. it's factually incorrect but i take its meaning to heart
13:00 jast the most valuable thing you get from university is a learning path of sorts
13:00 jast and peers
13:01 jast well, how would you measure the weight of knowledge? ;)
13:01 hahainternet carefully :)
13:01 jast good answer
13:01 hahainternet it's from Ray Mears, a survival guide, so what he's saying is "knowledge is free"
13:02 iH2O knowledge is made of holes in punch cards
13:02 hahainternet it has no downside
13:02 jast human functioning is actually my other big field of interest, and knowledge is an interesting topic in that context
13:02 iH2O holes arent heavy
13:02 jast knowledge does have downsides
13:03 hahainternet does it though?
13:03 iH2O holes have downsides
13:03 jast yes
13:03 jast aside from the whole 'ignorance is bliss' thing
13:03 iH2O lol
13:03 jast (and there's some truth to that, too)
13:04 hahainternet it's a difficult argument to make, because the ignorant don't realise their bliss
13:04 jast it's an easy argument to make for someone who thinks they know too much :}
13:04 jast anyway. knowledge, if you think of it as more than just being able to recite facts, involves a more or less intuitive understanding of at least some elements of it
13:04 jast for instance, if you don't know what differentation is all about, reasoning about it doesn't become easier
13:04 hahainternet yeah the term is quite loosely defined, but the general principle is one i think is valuable to apply
13:05 hahainternet 'you might as well learn, because it won't do any harm'
13:05 jast and intution takes time to build and, presumably, occupies a rough equivalent to space
13:05 hahainternet actually that's an interesting one
13:06 hahainternet because my personal view is that the brain is a series of constant modelling processes
13:06 jast I don't think equating human memory to disk space is a reasonable comparison, but still there are physical limits
13:06 hahainternet and it's arguable that a lot of 'memory' could be refinements of those models
13:06 hahainternet but it doesn't follow per se that a model refinement is an increase in complexity
13:06 hahainternet so there's an argument to be made that the brain of someone who has a vast corpus of knowledge and understanding may in fact be significantly simpler than the brain of someone without this
13:07 jast the more intuition you need (= more fields of knowledge), the more complexity there is in tuning your cognitive structure to fit them all together
13:07 jast simply put, sometimes understanding one thing makes it harder to understand some other thing
13:07 hahainternet it's not clear to me that that is inherently true
13:08 jast I don't want to put this up as a law, just as a tendency
13:08 hahainternet i'll give a brief example to try and illustrate my thoughts
13:08 masak sounds a bit like "the curse of knowledge", which is a thing
13:08 hahainternet take a physically impossible mechanism that is plausible on its face
13:08 jast I mean, often there is an equivalent payoff for getting a unified concept of two different fields
13:08 hahainternet i have this belief i find it hard to shake, and i know it's impossible, so if people don't mind the diversion i'll illustrate it
13:09 hahainternet imagine a 1 mile long flat rectangular beam
13:09 hahainternet atop that at one side, sit an electrical generator shaped as a cylinder, as it rolls down the beam, it spins and thus generates electricity
13:09 hahainternet now lift one end of the beam up by a microscopic amount, sufficient for the generator to begin rolling, a perfect cylinder will begin rolling immediately
13:09 lizmat joined #perl6
13:09 pmurias joined #perl6
13:09 jast (not taking into account friction)
13:09 hahainternet when it gets to the other end, drop the end you started with by 2x the distance
13:10 hahainternet allow it to roll back
13:10 hahainternet now, arbitrarily extend the length of the plate, in theory increasing the amount of electricity extracted
13:10 hahainternet now, this is clearly wrong, and i know it's wrong because i know enough to look at the sources of energy input, and i know that E=mgh
13:11 hahainternet but it's plausible on its face, and someone without the understanding of that equation may have to go through an extensive and very complex process to find out it is false
13:11 yanmc guys, I can't get p6doc to work
13:11 hahainternet ok back on topic :D
13:11 yanmc haha, don't mind me :)
13:11 jast what does it do instead of working?
13:11 moritz hahainternet: you missed a critical step: the claim (that is wrong)
13:12 yanmc apparently its path is empty?
13:12 yanmc Cannot locate Type::Str in any of the following paths:
13:12 hahainternet moritz: that you can generate greater electricity output than the cost of raising and lowering the track to cause the generator to roll
13:12 pmurias why is the hexint rule is HLL::Grammar so weird?
13:12 moritz hahainternet: ah
13:12 moritz yanmc: which version of rakudo are your using?
13:12 Emeric joined #perl6
13:13 yanmc rakudobrew version tells me moar-nom
13:13 jast hahainternet: okay, but that doesn't say anything about the complexity of the cognitive structure involved, only about the complexity of the overall process to determine the correct answer
13:13 Emeric is it possible to build a .exe from a perl6 script ?
13:13 yanmc perl6 says This is Rakudo version 2016.05-179-gf192415 built on MoarVM version 2016.05-37-ga126e0f
13:13 toddr joined #perl6
13:13 yanmc implementing Perl 6.c.
13:13 hahainternet jast: indeed, but it's simply an illustration of how more knowledge can lead to a simpler process
13:13 jast I don't disagree with that
13:14 hahainternet i don't pretend to have any facts about how the brain actually works :D
13:14 toddr FUI in Damian's Class right now. The link to download for OSX on http://rakudo.org/how-to-get-rakudo/ is bad.
13:14 jast nobody knows how the brain works, so that's good :)
13:15 moritz yanmc: and did you install p6doc?
13:15 yanmc moritz: I cloned the repo from github?
13:16 jast it's clear in this case that the knowledge helps. thing is, there was some mental effort involved in getting to a point where you can recognize that the formula applies to this, and what it means for the claim. the interesting question is whether that effort depends only on the complexity of the subject... clearly not. some people find some topics easier to grasp than others.
13:17 yanmc and added the executable to my path, but same thing happens regardless of how I call it
13:17 jast so, there are several forces in effect: more knowledge can have a holistic effect to speed up future learning, but we have some kinds of physical limitations and some concepts are contradictory. so where are the trade-offs? no idea :)
13:17 moritz yanmc: you might need to install it ("panda install .") for it find the doc files; but I'm kinda out of the loop wrt the docs
13:18 yanmc ok, will try
13:18 yanmc in the mean while, a related issue with trying to build the html version of the docs
13:19 yanmc it generates the graph images correctly, i.e. like type-graph-Int.png, but it looks for type-graph-int.png (lowercase) and errors out at that point
13:20 moritz huh, both int and Int exist
13:20 moritz yanmc: what platform are you on?
13:21 yanmc ubuntu linux
13:22 yanmc moritz: yeah, very strange, the file system should have no problem with that
13:22 pmurias joined #perl6
13:22 moritz also, the build for doc.perl6.org runs on a Debian
13:22 moritz so very similar
13:22 moritz though probably on an older rakudo version
13:23 moritz yes, on 2016.04
13:23 dha joined #perl6
13:23 yanmc moritz: first the folder was contained in my Dropbox and I got case conflicts on these 5 files
13:24 jast some unusual filesystems can mess up case in filenames
13:24 yanmc scratch that, 3, type-graph-Int.png type-graph-Int.svg Operators.X.html
13:24 moritz yanmc: oh, you might rerun htmlify.p6 with --type-graph to rerun them
13:25 moritz *need to
13:26 iH2O équit
13:26 * masak .oO( équit -- I'm outta hére )
13:28 yanmc oh boy, somehow I still had it installed in both locations, running it now, but don't expect that problem again
13:29 AlexDaniel .seen timeless
13:29 yoleaux I saw timeless 16 Jun 2016 03:33Z in #perl6: <timeless> this is what i was using to test a change like what you pushed https://www.irccloud.com/pastebin/TYiCATmm/
13:29 dha Hm. Damian is making Perl 6 look pretty cool. Also scary.
13:30 dha :-)
13:30 gfldex where is he doing that?
13:31 dha YAPC::NA
13:32 lizmat http://www.yapcna.org/yn2016/event/1858
13:33 BenGoldberg joined #perl6
13:36 dalek snake: 36e554f | (Pawel Murias)++ | / (2 files):
13:36 dalek snake: Use python rules for integer literals instead of NQP ones.
13:36 dalek snake: review: https://github.com/arnshol​t/snake/commit/36e554f533
13:36 * TimToady --> tpc hotel
13:37 yanmc moritz: panda install worked!
13:38 yanmc I don't know if this is correct and the path is hardcoded, but p6doc apparently looks for the pods in ~/.perl6/doc
13:38 yanmc maybe that information was missing from https://github.com/perl6/doc/blob/master/README.md
13:41 dalek snake: ec2bb60 | (Pawel Murias)++ | src/Snake/Grammar.nqp:
13:41 dalek snake: Remove TODO comment, it's python 3 not python 2 that we are implementing.
13:41 dalek snake: review: https://github.com/arnshol​t/snake/commit/ec2bb601fb
13:45 Woodi hahainternet: about holo displays :)  such strong emotion... but it's just augmented flowerpot vs your augmented glasses, both can be tools :)  also you looks you prefer short things that can be available in short term vs sci-fi... it's like don't liking poetry, it's for communicating idea but usually no immediate technical application
13:45 dalek doc: 114bd1a | (Zoffix Znet)++ | doc/Programs/00-running.pod:
13:45 dalek doc: Fix unsupported POD directives
13:45 dalek doc: review: https://github.com/perl6/doc/commit/114bd1abd4
13:45 yanmc where do panda installed packages go on the filesystem?
13:46 dalek doc: bf6b0ee | (Zoffix Znet)++ | htmlify.p6:
13:46 dalek doc: Move Programs to end of menu
13:46 dalek doc: review: https://github.com/perl6/doc/commit/bf6b0eee59
13:46 moritz yanmc: into $install/share/perl6/site I think
13:52 yanmc moritz: oh, but panda is really the only way to interact with that? (no renaming {,.disabled} which is not necessarily a bad thing)
13:54 pmurias joined #perl6
13:56 pmurias masak: re state of snake, looking at the commit log arnsholt hasn't worked on it recently
13:56 dha Damian: "There's really no difference between a variable and a subroutine" I bet the audience is loving that. :-)
13:56 Woodi dha: where you hear/saw that ?
13:57 dha Damian's giving his Perl 6 multi-paradigm course at YAPC::NA
13:57 dha As usual, he's occasionally trying to scare us. :-)
13:58 Woodi dha: Perl6 is scary by itself ;)
14:01 toddr joined #perl6
14:01 grondilu off topic:  Blue Origin's launch livestream:  https://www.youtube.com/watch?v=EI-tGVFg7PU
14:03 yanmc oh "earlier today", would be cool if it could be rolled out to launch pad and be launched in 30' time
14:03 * Zoffix looks
14:03 * Zoffix sees Russians dissing the rocket why others compare to a penis
14:04 AlexDaniel Zoffix: have you figured out the thing with \-- ?
14:05 Zoffix AlexDaniel, are you talking about Programs POD? I've not even attempted
14:05 AlexDaniel yea
14:05 toddr joined #perl6
14:05 AlexDaniel tbrowder: hi! By the way, why 00-running? Why not have a beautiful link instead?
14:06 AlexDaniel programs/running – looks fine!
14:08 kaare_ joined #perl6
14:09 sufrostico joined #perl6
14:09 pmurias (pythons tab/space mixing rule)-- # had to look at the interpreter source to understand how it works
14:12 mst pmurias: hrm? all the python code I see uses 4 space indent
14:14 jjido joined #perl6
14:16 lizmat joined #perl6
14:16 toddr joined #perl6
14:18 dha joined #perl6
14:30 kurahaupo joined #perl6
14:32 pmurias mst: that's the all way all the vaguely sane people indent Python code ;)
14:32 stevieb pep8 states 4-space tabs, and that's all I've ever seen: https://www.python.org/dev/​peps/pep-0008/#indentation
14:32 mst pmurias: which is why, obviously, I end up using 2 space
14:32 mst being confused for vaguely sane would not help my life
14:35 pmurias mst: the Python tokenizer both treats the tabs as indenting up to a multiple of 8 and then checks that things don't change when treating tabs as having the size of a single space
14:36 mst oh yow
14:37 pmurias had to look up the rule so that the Python on QAST (aka snake) has the same semantics
14:37 pmurias afk&
14:43 DarthGandalf joined #perl6
14:44 adu joined #perl6
14:44 zakharyas joined #perl6
14:50 khw joined #perl6
14:55 rgrinberg joined #perl6
14:56 rjbs Huh, I thought Damian was @thoughtstream on Twitter.  Seems not.
15:00 dha I'm not sure he's actually on twitter.
15:00 cognominal currently, is there any explicit types that does not fit in a register ? Meaning you are obliged to declare the boxed equivalent, say Buf instead of buf
15:04 BenGoldberg What are you calling a type?
15:05 BenGoldberg m: my rat $x = 1.2;
15:05 camelia rakudo-moar 53af02: OUTPUT«5===SORRY!5===␤Type 'rat' is not declared. Did you mean 'Rat'?␤at <tmp>:1␤------> 3my rat7⏏5 $x = 1.2;␤Malformed my␤at <tmp>:1␤------> 3my7⏏5 rat $x = 1.2;␤␤»
15:05 cognominal related to the previous question. Is there any $v for which $v.REPR is not P6Opaque
15:05 cognominal m: say rat.REPR
15:05 camelia rakudo-moar 53af02: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤    rat used at line 1␤␤»
15:05 cognominal oops
15:05 BenGoldberg m: say int.REPR
15:05 camelia rakudo-moar 53af02: OUTPUT«P6int␤»
15:05 BenGoldberg m: say num.REPR
15:05 camelia rakudo-moar 53af02: OUTPUT«P6num␤»
15:05 BenGoldberg m: say Num.REPR
15:05 camelia rakudo-moar 53af02: OUTPUT«P6opaque␤»
15:06 BenGoldberg m: say Mu.REPR
15:06 camelia rakudo-moar 53af02: OUTPUT«P6opaque␤»
15:06 cognominal ok
15:06 cognominal why not rat.REPR ?
15:07 BenGoldberg Because there is no native rat.
15:07 cognominal S02 says rat is native
15:07 lizmat it got implemented differently  :-)
15:07 BenGoldberg std: my rat $x = 2.3;
15:07 camelia std : OUTPUT«No such file or directory»
15:08 lizmat Rational is implemented as a role
15:08 camelia std : OUTPUT«(timeout)cat: /home/camelia/std/snap/revision: No such file or directory␤»
15:08 rjbs Trying to figure out what determines legal names for variables.
15:08 rjbs ...and whether that differs from $^vars
15:08 cognominal I think that native that don't fit in a register are in limbo
15:09 cognominal the specs talks about a lot of them but only their boxed counterpart exists
15:09 cognominal so my confusion
15:09 yanmc is there a way to make the REPL not quit on ^C?
15:10 toddr joined #perl6
15:11 rjbs how can I print a description of a subroutine signature?
15:12 geekosaur yanmc, not yet, it's a known bug
15:12 lizmat m: sub a($a,@b,%c) {}; dd &a.Signature
15:12 camelia rakudo-moar 53af02: OUTPUT«Method 'Signature' not found for invocant of class 'Sub'␤  in block <unit> at <tmp> line 1␤␤»
15:12 lizmat m: sub a($a,@b,%c) {}; dd &a.signature
15:12 camelia rakudo-moar 53af02: OUTPUT«:($a, @b, %c)␤»
15:12 lizmat rjbs: ^^
15:12 rjbs Got it, thanks.
15:13 dalek doc: cfd3ee3 | (Jan-Olof Hendig)++ | doc/Type/IO/Socket/Async.pod:
15:13 dalek doc: Fix typo
15:13 dalek doc: review: https://github.com/perl6/doc/commit/cfd3ee3318
15:13 rjbs So, it appears to be (to my great pleasure!!) that two $^x variables with canonical unicode equivalence are combined into one.
15:13 rjbs sub { $^Å + $^Å } is 1-ary, with the a-ring and angstrom sign unified
15:13 rjbs good job, perl6!
15:14 cognominal BenGoldberg, can you get a .REPR of a type non fitting in a register print anything than P6opaque without extensions or creating a new kind of OO using the metaobject systeem ?
15:14 lizmat rjbs: that's NFG for ya
15:14 moritz .U Å
15:14 yoleaux U+212B ANGSTROM SIGN [Lu] (Å)
15:14 moritz .u Å
15:14 yoleaux U+00C5 LATIN CAPITAL LETTER A WITH RING ABOVE [Lu] (Å)
15:15 finanalyst joined #perl6
15:15 rjbs how does NFG do that?  I knew it worked on grapheme clusters, but I didn't know it dealt with canonical equiv
15:17 lizmat rjbs: ah, good point
15:17 lizmat hmmm... not sure about that
15:17 rjbs Okay!  Who do I interrogate about that later?  Also, who do I high five? :)
15:17 moritz I think NFG is based off NFKC
15:18 CQ joined #perl6
15:18 rjbs moritz: but would I then lose precision round-tripping through NFG?
15:18 moritz and we have a synopsis on unicode... S19 maybe?
15:18 rjbs moritz: Great, I'll look into that!
15:18 rjbs S15
15:18 rjbs Um: http://design.perl6.org/S15.html
15:19 moritz synopsebot6: S15
15:19 AlexDaniel rjbs: There's a list of people to high five on the bottom: https://perl6advent.wordpress.co​m/2015/12/25/christmas-is-here/
15:19 rjbs AlexDaniel: Thanks, but I was looking specifically for the person(s) who addressed this one very specific issue.
15:19 AlexDaniel huggable: dunno
15:19 huggable AlexDaniel, ¯\_(ツ)_/¯
15:20 rjbs But I'm on that list, so I'll high five myself whilst I wait! \o/
15:20 AlexDaniel and I'm not :P
15:21 rjbs I'll high five you anyway.
15:21 rjbs moritz: synopsebot6 taking holiday? :)
15:27 rblackwe joined #perl6
15:27 moritz rjbs: I think it triggers only on more specific stuff
15:27 moritz S02:1234
15:27 synopsebot6 Link: http://design.perl6.org/S02.html#line_1234
15:28 AlexDaniel S02:
15:28 AlexDaniel S02:0
15:28 synopsebot6 Link: http://design.perl6.org/S02.html#line_0
15:28 dha joined #perl6
15:28 AlexDaniel S02:a
15:28 synopsebot6 Link: http://design.perl6.org/S02.html#a
15:30 Zoffix S02:⁣
15:31 webstrand joined #perl6
15:34 ocbtec joined #perl6
15:41 finanalyst joined #perl6
15:49 Zoffix webstrand, the bug was fixed
15:51 webstrand joined #perl6
15:53 BenGoldberg m: use native 'int'; # S02:Numeric_Types
15:53 synopsebot6 Link: http://design.perl6.org/S02.html#Numeric_Types
15:53 camelia rakudo-moar 53af02: OUTPUT«===SORRY!===␤Could not find native at line 1 in:␤    /home/camelia/.perl6␤    /home/camelia/rakudo-m-inst-2/share/perl6/site␤    /home/camelia/rakudo-m-inst-2/share/perl6/vendor␤    /home/camelia/rakudo-m-inst-2/share/perl6␤    CompUnit::Reposi…»
15:54 lizmat S15:1
15:54 synopsebot6 Link: http://design.perl6.org/S15.html#line_1
15:56 cosarara_ joined #perl6
15:59 lizmat lunch&
16:07 skids joined #perl6
16:12 huggable joined #perl6
16:31 finanalyst joined #perl6
16:34 zakharyas joined #perl6
16:37 lambd0x joined #perl6
16:38 Zero_Dogg joined #perl6
16:38 Zero_Dogg joined #perl6
16:39 lambd0x Hi everyone! Does Rakudo Perl compiler already has support receive from a program an int input?
16:39 trnh joined #perl6
16:42 * Zoffix doesn't understand what that means
16:44 ilmari perl6 porgrams can receive integers from other programs in many ways
16:47 lambd0x sorry typo. I want to receive an Integer from the Standard Input but am not getting it right, I think. the Prompt function for example can just return a string for as far as I saw.
16:47 lambd0x Is that right?
16:49 b2gills Do you mean a native integer, or do you mean a human readable integer?
16:49 lambd0x b2gills: I mean native integer type of Perl.
16:49 jonadab joined #perl6
16:50 Zoffix lambd0x, well, sure, there isn't a "I'm now typing in integers" mode :P Just coerce it with .Int
16:50 Zoffix m: my Str ( $a, $b ) = 2, 4; say $a + $b # it DWIMs anyway
16:50 camelia rakudo-moar 53af02: OUTPUT«Type check failed in assignment to $a; expected Str but got Int (2)␤  in block <unit> at <tmp> line 1␤␤»
16:50 Zoffix m: my Str ( $a, $b ) = '2', '4'; say $a + $b # it DWIMs anyway
16:50 camelia rakudo-moar 53af02: OUTPUT«6␤»
16:53 b2gills m: #`( here is a native integer ) $*OUT.write: Blob.new: 0x63
16:53 camelia rakudo-moar 53af02: OUTPUT«c»
16:53 b2gills m: #`( here is a human readable integer ) $*OUT.print:  0x63
16:53 camelia rakudo-moar 53af02: OUTPUT«99»
16:54 lambd0x ok, I'm just starting to learn Perl 6. Was playing around with some subs and a simple while menu that got input from the user to redirect it to some of the subs. Using variables in a static way.
16:56 lambd0x thanks :)
16:57 Ben_Goldberg joined #perl6
16:59 pmurias_ joined #perl6
17:02 lambd0x Here's what I actually got and what the compiler gives me so far: https://bpaste.net/show/2816d25cbf4e
17:03 iH2O joined #perl6
17:04 Zoffix lambd0x, just return prompt(); no need to assign it to a var
17:05 lambd0x really? that's nice. 'll try it.
17:05 iH2O left #perl6
17:07 finanalyst joined #perl6
17:08 Zoffix lambd0x, I'd write it like this: https://gist.github.com/zoffixznet​/8b115b496cdddd6469b26b8159a16980
17:09 Zoffix default { say ... } isn't even needed in this case. Just say ... is enough because `when` will exit the block if it matches.
17:09 Zoffix Though, may be good for clarity ¯\_(ツ)_/¯
17:10 lambd0x Zoffix: thanks I'm really enjoying learning Perl 6 so far.
17:10 Zoffix \o/
17:12 Deep_Thought joined #perl6
17:12 lizmat Zoffix: so I should / could merge now ?
17:12 lizmat Zoffix: just checking
17:14 TimToady rjbs, moritz: NFG is just based on NFC, not NKFC, but NFC itself canonicalizes a few weird characters such as greek semicolon, so this is probably similar; and yes, NFG does lose the original representation, just as NFC does for composing chars
17:15 AlexDaniel uhh! 「prompt '';」 without return but with semicolon /o\
17:16 ilmari m: say "\c[ANGSTROM SIGN]".uninames
17:16 camelia rakudo-moar 53af02: OUTPUT«(LATIN CAPITAL LETTER A WITH RING ABOVE)␤»
17:18 Zoffix lizmat, yup
17:21 dha joined #perl6
17:26 ptolemarch joined #perl6
17:28 rjbs TimToady: thanks!
17:29 abruanese joined #perl6
17:35 Some-body_ joined #perl6
17:39 dj_goku joined #perl6
17:41 woolfy joined #perl6
17:42 woolfy The Perl 6 course at YAPC::NA in Orlando by Damian Conway is well-visited, lots of "Perl 5 faces", and it's going well, and 40+ attendees.
17:44 Zoffix Awesome
17:53 masak nice!
17:54 BenGoldberg joined #perl6
17:56 tadzik sweet :)
17:56 _webstrand joined #perl6
17:58 lambd0x :'-)
18:04 woolfy (of course, also some well-know Perl 6 faces, like Larry, DrForr, lizmat, dha)(and Ricardo Signes and Chip Salzenberg!!!)
18:04 CIAvash joined #perl6
18:06 jdv79 is the material up to date?
18:06 * FROGGS .oO( well, Larry is perhaps a bit old :P )
18:06 lizmat jdv79: yes it is
18:08 jdv79 i meant things like glr and async
18:08 mst are the slides online now or do we have to wait til later for that?
18:09 woolfy mst: most of the time the courses by Damian are not made public, so I'm not sure.  We did get handouts, a centimeter thick booklet with loads of info.
18:09 mst ah
18:09 mst of course it's PJF and Jacinta who publish everything
18:10 woolfy I will ask Damian in the upcoming break and get back to you about it.
18:10 lizmat mst: yeah
18:10 mst yeah, 's fair enough, I was mooostly curious which things he chose to highlight
18:11 woolfy (well, some people, like Damian, have giving courses as their business, and when he gives away his materials, some people might be less inclined to attend his courses)
18:11 woolfy It's a whole day, and he covers more than I can handle...  but that does not say a lot.  At the moment he is doing OO, and for me it's too deep.
18:12 nine woolfy: Chip Salzenberg? How is he? :)
18:13 jdv79 will a video be avail?
18:13 mst it's a paid training course
18:14 mst I continue to be glad that perltraining.com.au has decided that it makes sense to them to release a lot of their course materials, but it's entirely understandable that not everybody does
18:16 woolfy Damian's tutorial is among the most affordable ones, only $75 per person ( http://www.yapcna.org/yn2016/tutorials.html ).
18:16 woolfy Chip is looking well, and and active attendee, asking questions too.
18:17 woolfy jdv79: no, there will be no video
18:19 jdv79 oh well.  id pay for materials or video or audio.  just didnt make it to yapc.
18:19 dha To be fair, it was implied that the course is only $75 because it's being subsidized.
18:20 dha Also, I did *tell* you to come to YAPC. ;-)
18:21 jdv79 cool.  i know.
18:26 _mg_ joined #perl6
18:27 ssotka joined #perl6
18:35 rjbs once again weeping because (obj.^methods.sort) doesn't dwiw; I should write a thiny to use in the reply to give me sorted method names
18:36 FROGGS m: say 42.^methods.sort: *.name
18:36 camelia rakudo-moar 1d93ad: OUTPUT«(ACCEPTS Bool Bridge Complex DUMP FatRat Int Num Numeric Range Rat Real Str WHICH abs acos acosec acosech acosh acotan acotanh asec asech asin asinh atan atan2 atanh base ceiling chr cis conj cos cosec cosech cosh cotan cotanh exp expmod floor gist is-prim…»
18:37 rjbs FROGGS: I end up using obj.^methods>>.name.sort
18:37 rjbs FROGGS: I'm just trained by Ruby. :)
18:37 finanalyst joined #perl6
18:37 FROGGS TIMTOWTDI :o)
18:37 rjbs I Wish My First Guess Always Was One Of The Ways To Do It
18:37 rjbs IWMFGAWOOTWTDI
18:38 dha WOOT
18:38 FROGGS ohh, that'd be the python way
18:38 kid51 joined #perl6
18:38 dha Guido once showed NY.pm TWO ways to do something.
18:38 FROGGS ohh wow
18:39 dha I wish I could remember what it was.
18:41 FROGGS rjbs: installing 5.22 on a production right now btw :o)
18:41 zhmylove_ joined #perl6
18:41 vendethiel- joined #perl6
18:42 Zoffix There's 5.24! With no-experimental postfix deref! :P
18:42 Zoffix *non
18:42 FROGGS hmmmm, worth a try I guess :o)
18:42 pRiVi_ joined #perl6
18:43 lizmat m: quietly say 42.^methods.sort   # rjbs: another way  :-)
18:43 camelia rakudo-moar 1d93ad: OUTPUT«(ACCEPTS Bool Bridge Complex DUMP FatRat Int Num Numeric Range Rat Real Str WHICH abs acos acosec acosech acosh acotan acotanh asec asech asin asinh atan atan2 atanh base ceiling chr cis conj cos cosec cosech cosh cotan cotanh exp expmod floor gist is-prim…»
18:44 obfusk joined #perl6
18:44 a3r0_ joined #perl6
18:44 ilmari_ joined #perl6
18:44 jdv79_ joined #perl6
18:44 kshannon joined #perl6
18:44 DrForr_ joined #perl6
18:45 jeek_ joined #perl6
18:45 rudi_s_ joined #perl6
18:45 Util_ joined #perl6
18:45 profan_ joined #perl6
18:45 johnjohn101 joined #perl6
18:45 mspo_ joined #perl6
18:45 Hor|zon_ joined #perl6
18:45 nowan joined #perl6
18:50 sftp joined #perl6
18:50 simcop2387 joined #perl6
18:56 samb1 joined #perl6
18:57 _webstrand joined #perl6
18:57 woolfy joined #perl6
18:58 autarch joined #perl6
19:03 rjbs m: say 0o10; say 2⁸
19:03 camelia rakudo-moar 1d93ad: OUTPUT«8␤256␤»
19:03 rjbs m: say 2⁰º¹⁰
19:03 camelia rakudo-moar 1d93ad: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Two terms in a row␤at <tmp>:1␤------> 3say 2⁰7⏏5º¹⁰␤    expecting any of:␤        infix␤        infix stopper␤        postfix␤        statement end␤        statement modifier␤  …»
19:03 rjbs too bad :-)
19:03 Zoffix heh
19:04 FROGGS dammit, was fighting 15 minutes to install 5.24 via perlbrew through a proxy after I recognized I have to specify 5.24.0 -.-
19:05 jdv79_ wah wah:)
19:06 Zoffix m: sub infix:<⁰º> ($b, $p) { say [$b, $p] }; say 2⁰º¹⁰
19:06 camelia rakudo-moar 1d93ad: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Two terms in a row␤at <tmp>:1␤------> 3x:<⁰º> ($b, $p) { say [$b, $p] }; say 2⁰7⏏5º¹⁰␤    expecting any of:␤        infix␤        infix stopper␤        postfix␤        statement …»
19:06 Zoffix too bad :)
19:07 Zoffix m: sub postfix:<⁰º¹⁰> { $^b ** 8 }; say 2⁰º¹⁰
19:07 camelia rakudo-moar 1d93ad: OUTPUT«256␤»
19:08 rjbs m: say 2⁰ᵇ¹¹
19:08 camelia rakudo-moar 1d93ad: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Two terms in a row␤at <tmp>:1␤------> 3say 2⁰7⏏5ᵇ¹¹␤    expecting any of:␤        infix␤        infix stopper␤        postfix␤        statement end␤        statement modifier␤  …»
19:08 rjbs Zoffix: Well done.
19:08 rjbs Really, I'm cheating.  There's no good legit superscript o or b.
19:08 Zoffix m: say '
19:08 camelia rakudo-moar 1d93ad: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Unable to parse expression in single quotes; couldn't find final "'" ␤at <tmp>:1␤------> 3say '7⏏5<EOL>␤    expecting any of:␤        argument list␤        single quotes␤        term␤»
19:08 Zoffix m: say 'ᵇ'.uninames
19:08 camelia rakudo-moar 1d93ad: OUTPUT«(MODIFIER LETTER SMALL B)␤»
19:09 pmichaud joined #perl6
19:09 pmichaud good afternoon, #perl6
19:09 Zoffix \o
19:09 dha hi pmichaud
19:09 jdv79 hello
19:10 lizmat pmichaud  o/
19:10 FROGGS pmichaud \o/
19:10 masak pmichaud! \o/
19:12 yqt joined #perl6
19:14 jdv79 pmichaud: at yapc?
19:14 jdv79 or going
19:17 pmichaud at yapc -- arrived about an hour ago
19:17 pmurias joined #perl6
19:18 stevieb m: enum test <a b>; b.value.say
19:18 camelia rakudo-moar 1d93ad: OUTPUT«1␤»
19:18 stevieb m: my @a=("a","b"); enum test @a; a.value.say
19:18 camelia rakudo-moar 1d93ad: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routines:␤    a used at line 1␤    enum used at line 1␤    test used at line 1␤␤»
19:19 stevieb how do I use an array to create an enum? I've tried several variants, none of which work
19:19 FROGGS m: my @a=("a","b"); enum test @a; say @a[0]
19:19 camelia rakudo-moar 1d93ad: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routines:␤    enum used at line 1␤    test used at line 1␤␤»
19:20 Zoffix m: my @a=("a","b"); enum test (@a); a.value.say
19:20 camelia rakudo-moar 1d93ad: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤    a used at line 1␤␤»
19:20 Zoffix m: my @a=("a","b"); enum test (@a); @a.value.say
19:20 camelia rakudo-moar 1d93ad: OUTPUT«Method 'value' not found for invocant of class 'Array'␤  in block <unit> at <tmp> line 1␤␤»
19:21 FROGGS the enum is called test
19:21 FROGGS m: my @a=("a","b"); enum test (@a); say test
19:21 camelia rakudo-moar 1d93ad: OUTPUT«(test)␤»
19:21 FROGGS m: my @a=("a","b"); enum test (@a); import test;
19:21 camelia rakudo-moar 1d93ad: ( no output )
19:21 FROGGS m: my @a=("a","b"); enum test (@a); import test; say a
19:21 camelia rakudo-moar 1d93ad: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤    a used at line 1␤␤»
19:21 FROGGS no, that probably wont work
19:21 stevieb m: <a b>.WHAT
19:21 camelia rakudo-moar 1d93ad: ( no output )
19:21 stevieb m: <a b>.WHAT.say
19:21 camelia rakudo-moar 1d93ad: OUTPUT«(List)␤»
19:22 stevieb m: my @a=<a b>; @a.WHAT.say
19:22 camelia rakudo-moar 1d93ad: OUTPUT«(Array)␤»
19:22 stevieb m: my @a=<a b>; @a.List.WHAT.say
19:22 camelia rakudo-moar 1d93ad: OUTPUT«(List)␤»
19:22 stevieb enum takes a list, so I even called List on it and still fail
19:23 stevieb m: my @a=<a b>; enum test @a.List; a.value.say
19:23 camelia rakudo-moar 1d93ad: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routines:␤    a used at line 1␤    enum used at line 1␤    test used at line 1␤␤»
19:23 stevieb m: my @x=<a b>; enum test @x.List; a.value.say
19:23 camelia rakudo-moar 1d93ad: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routines:␤    a used at line 1␤    enum used at line 1␤    test used at line 1␤␤»
19:23 stevieb ^^ just in case it was a name clash with @a, which (hopefully) could never happen
19:23 FROGGS m: BEGIN my @a=("a","b"); enum test (@a); say a
19:23 camelia rakudo-moar 1d93ad: OUTPUT«a␤»
19:24 Zoffix heh, damn. I tried with @a = BEGIN ... :)
19:24 stevieb m: my @a=("a","b"); enum test (@a); a.value.say
19:24 camelia rakudo-moar 1d93ad: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤    a used at line 1␤␤»
19:24 FROGGS stevieb: the enum declaration runs a compile time, and at that time the assignment has not happened yet
19:24 stevieb ahhhhh!
19:24 stevieb m: BEGIN {my @a=("a","b"); enum test (@a);}; a.value.say
19:24 camelia rakudo-moar 1d93ad: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Undeclared routine:␤    a used at line 1␤␤»
19:25 FROGGS so, all is well until you time travell
19:25 FROGGS now it is only visible within that block
19:25 stevieb m: BEGIN my @a=("a","b"); enum test (@a); say a.value.say
19:25 camelia rakudo-moar 1d93ad: OUTPUT«0␤True␤»
19:26 arnsholt masak: Snake isn't dead, but it's not super-alive either
19:26 stevieb FROGGS: I'll play around with BEGIN. thanks!
19:27 arnsholt My copious free time is extra copious these days, with a three days old and a 12 month over due thesis
19:27 stevieb in fact, enum won't work in the case I wanted it for, as I want to be able to change it. I'll use a hash instead :)
19:27 FROGGS arnsholt: \o/
19:27 arnsholt But it's a projecy that's still fermenting in my brain, and I think it'd be interesting to make more progress on it
19:27 FROGGS arnsholt: girl or boy?
19:28 arnsholt So I'm pondering stuff like how to get the Python type system right, where object's type object is type and type itself extends object
19:29 Zoffix m: my $xºbar = 'wat'; say $xºbar
19:29 arnsholt Sort of a type croissant?
19:29 camelia rakudo-moar 1d93ad: OUTPUT«wat␤»
19:29 arnsholt FROGGS: Boy =)
19:29 FROGGS arnsholt: awesome :D
19:29 dalek snake: 2c4204e | (Pawel Murias)++ | src/Snake/ (2 files):
19:29 dalek snake: Rename sports to spaces-or-tabs.
19:29 dalek snake: review: https://github.com/arnshol​t/snake/commit/2c4204e8de
19:29 pmurias arnsholt: hi
19:29 arnsholt pmurias: EHLO! =)
19:30 woolfy mst (and others who were interested): I just asked Damian, and he says "no", to "will the slides of the Perl 6 course be available online".  For exactly the reasons I stated.  It's his income.
19:31 pmurias arnsholt: renamed sports to spaces-or-tabs as I can't bend my mind not to parse it as a single word ;)
19:31 arnsholt pmurias: np =)
19:31 mst woolfy: well, yes, as I said, once I'd remembered it was only PJF+Jacinta who were outliers in that respect, I wasn't expecting any other response
19:32 * Juerd would gladly pay to watch recordings of such talks
19:32 Juerd That'd still be much cheaper than actually going to a YAPC :)
19:33 Zoffix m: my \ʳᵘⁿ_ᶠᵒʳ-ᵗʰᵉ_ʰⁱˡˡˢ = 'run for the hills';   say ʳᵘⁿ_ᶠᵒʳ-ᵗʰᵉ_ʰⁱˡˡˢ
19:33 camelia rakudo-moar 1d93ad: OUTPUT«run for the hills␤»
19:33 Zoffix That's just obscene :P
19:33 arnsholt No, it's Unicode ;p
19:35 japhb arnsholt: Not mutually exclusive statements.  :-)
19:35 arnsholt masak: Another thing that's annoyed me in Snake is the fact that the type() constructor is overloaded to also extract the type object of things
19:36 arnsholt Now *that's* obscene!
19:38 dalek doc: 6a252c5 | (Tom Browder)++ | htmlify.p6:
19:38 dalek doc: simplify for loop
19:38 dalek doc: review: https://github.com/perl6/doc/commit/6a252c5e5d
19:38 dalek doc: 276c613 | RabidGravy++ | htmlify.p6:
19:38 dalek doc: Merge pull request #606 from tbrowder/rm-flat
19:38 dalek doc:
19:38 dalek doc: simplify for loop
19:38 dalek doc: review: https://github.com/perl6/doc/commit/276c613cdb
19:40 dha joined #perl6
19:42 Supervisor joined #perl6
19:42 Supervisor left #perl6
19:42 arnsholt masak: Anyways, I've given you a commitbit to Snake. No need to feel pressured to hack on it (I'm not doing much to it anyways, after all), but if you have ideas on how to proceed I trust your judgement. Bug tickets discussing implementation issues are most welcome too, of course =)
19:43 jjido joined #perl6
19:48 woolfy Juerd: in my experience, when a video of something is made avaliable in a website where you have to pay for it, very soon after many more people have a copy of the video and most have not paid for it...
19:51 mst woolfy: I liked 37signals' trick with their e-books where they embedded the purchaser's name
19:55 woolfy mst: yeah, me too, have some of such e-books copied and uploaded shamelessly by other people...
19:58 cognominal joined #perl6
19:58 Zoffix m: say $_ %% Ⅴ*Ⅲ ?? 'Fizz Buzz' !! $_ %% Ⅲ ?? 'Fizz' !! $_ %% Ⅴ ?? 'Buzz' !! $_ for ^Ⅹ*Ⅹ
19:58 camelia rakudo-moar 1d93ad: OUTPUT«Fizz Buzz␤1␤2␤Fizz␤4␤Fizz Buzz␤Fizz␤7␤8␤Fizz␤Fizz Buzz␤11␤Fizz␤13␤14␤Fizz Buzz␤16␤17␤Fizz␤19␤Fizz Buzz␤Fizz␤22␤23␤Fizz␤Fizz Buzz␤26␤Fizz␤28␤29␤Fizz Buzz␤31␤32␤Fizz␤34␤Fizz Buzz␤Fizz␤…»
19:59 Zoffix :)
20:00 Juerd woolfy: That happens, yes.
20:01 girafe joined #perl6
20:11 jjido joined #perl6
20:13 erdic joined #perl6
20:14 telex joined #perl6
20:14 jack_rabbit joined #perl6
20:15 rindolf joined #perl6
20:22 pmurias_ joined #perl6
20:34 pmurias is passing an array of 2 integers cheaper in nqp than passing an object?
20:35 moritz the passing itself is just a reference, so should be the same performance wise
20:35 arnsholt Yeah, even a native array goes through the object paths
20:36 arnsholt The *lookups* might be faster, but I wouldn't expect the passing to be
20:36 arnsholt (lookups into the array vs. lookup of object attributes, that is)
20:36 moritz as for storage, I guess the array needs to store itse size somewhere, and the object needs to store its class
20:36 moritz so, donno
20:36 arnsholt Array is probably more compact
20:37 arnsholt As P6opaque also stores things like attribute name to offset tables and the like
20:37 pmurias arnsholt: that would be shared for all the object of the same class
20:37 moritz right, that's in the meta calss
20:38 moritz *class
20:38 moritz well, meta class instance
20:38 moritz and the array also has a meta class instance
20:40 arnsholt Oh, derp. That's true
20:41 moritz nqp-m: say([].^name)
20:41 camelia nqp-moarvm: OUTPUT«Confused at line 2, near "say([].^na"␤   at gen/moar/stage2/NQPHLL.nqp:706  (/home/camelia/rakudo-m-inst-1/sh​are/nqp/lib/NQPHLL.moarvm:panic)␤ from gen/moar/stage2/NQP.nqp:927  (/home/camelia/rakudo-m-inst-1/sha​re/nqp/lib/nqp.moarvm:comp_unit)␤ from gen/moar…»
20:42 moritz nqp-m: say([].HOW.name([]))
20:42 camelia nqp-moarvm: OUTPUT«NQPArray␤»
20:42 TEttinger joined #perl6
20:47 stevieb is there a way to constrain the size of an attribute? for instance, I want a read/write array, but I want to force it to contain exactly eight elements
20:48 moritz m: my int @array[8]; say @array.elems
20:48 camelia rakudo-moar 1d93ad: OUTPUT«8␤»
20:48 moritz stevieb: ^^ like that, but only for native arrays, afaict
20:49 moritz m: my @foo[8]; say @foo.elems
20:49 camelia rakudo-moar 1d93ad: OUTPUT«8␤»
20:49 moritz oh, seems to work rather generally
20:49 gfldex m: my int @array[8]; say @array.elems; .say for @array;
20:49 camelia rakudo-moar 1d93ad: OUTPUT«8␤0␤0␤0␤0␤0␤0␤0␤0␤»
20:49 gfldex m: my @array[8]; say @array.elems; .say for @array;
20:49 camelia rakudo-moar 1d93ad: OUTPUT«8␤(Any)␤(Any)␤(Any)␤(A​ny)␤(Any)␤(Any)␤(Any)␤(Any)␤»
20:49 stevieb moritz: thanks for that. I completely forgot I had read that... I should have known :)
20:54 mkende joined #perl6
20:55 gfldex m: my @a[8] is List; .say for @a;
20:55 camelia rakudo-moar 1d93ad: ( no output )
20:55 gfldex m: my @a[8] is List; @a[100] = 'oi!';
20:55 camelia rakudo-moar 1d93ad: OUTPUT«Cannot modify an immutable Nil␤  in block <unit> at <tmp> line 1␤␤»
20:56 gfldex m: my @a[8] is List; @a[5] = 'oi!';
20:56 camelia rakudo-moar 1d93ad: OUTPUT«Cannot modify an immutable Nil␤  in block <unit> at <tmp> line 1␤␤»
20:56 gfldex m: my @a[8] is List[Str]; @a[5] = 'oi!';
20:56 camelia rakudo-moar 1d93ad: OUTPUT«5===SORRY!5=== Error while compiling <tmp>␤Can't use unknown trait 'is ' in a variable declaration.␤at <tmp>:1␤------> 3my @a[8] is List[Str]7⏏5; @a[5] = 'oi!';␤    expecting any of:␤        TypeObject␤        default␤        dynami…»
20:58 mkende_ joined #perl6
20:58 gfldex stevieb: you didn't read it because we didn't wrote it
20:58 gfldex i will plug that hole tomorrow
21:00 stevieb perhaps something like it was in an intro or something I read somewhere...
21:03 mkende Hi all, I have an issue with Rakudo/moarVM compilation times: if I use Rakudo as provided by the Debian testing package or if I build and install it myself with a --prefix=/usr/local/bin, then just running "perl6 -e ''" takes more than 2 seconds (entirely spent in the "parse" stage). If I install the compiler with --prefix=/home/user/foobar then this issue disappear (compilation is fast).
21:03 mkende Is this a known issue?
21:03 rindolf joined #perl6
21:03 stevieb gfldex: the following doc describes "fixed size arrays", which is relatively similar to what I was asking, but not quite. That's where I remember it from http://perl6intro.com/
21:04 timotimo mkende: can you give us the exact perl6 --version output?
21:05 woolfy Perl 6 course by Damian just ended.  Several people will now go offline...  byebye <3
21:05 woolfy left #perl6
21:05 timotimo o/
21:05 mkende This is Rakudo version 2016.05-145-gac0dcdd built on MoarVM version 2016.05-34-gfbe9e24
21:05 mkende implementing Perl 6.c.
21:06 mkende that's the one that I built.
21:06 timotimo in that version we already have RAKUDO_MODULE_DEBUG, if you set that env var, you'll get a bunch of spam, but you could probably feel where more time is spent in the one version vs th eother
21:06 timotimo stage parse is, after all, also the place where modules get loaded and such
21:07 mkende The debian version is "version 2016.05 built on MoarVM version 2016.05
21:07 mkende implementing Perl 6.c.
21:07 mkende "
21:07 timotimo ah, different versions. that makes it a bit trickier
21:07 timotimo well, at least you're not running a parrot-based version from 2014 or something
21:08 mkende well, I can just ignore the debian version for the moment, as I see the issue with the other one, depending on where I install it.
21:08 timotimo ah, that's good!
21:09 timotimo there's also a flag for strace that might help locate what moar's doing around the time it slows down so much
21:09 timotimo i mean, you can make strace output the time since the start, so it'd be more visible
21:09 mkende I've already looked at the strace output and couldn't find anything suspicious
21:09 timotimo OK
21:09 timotimo do you have perf available on your machine?
21:10 mkende no, no "perf"
21:10 b2gills Could it be that one of them isn't precompiling corectly?
21:10 timotimo that's the c-level profiler i like using
21:10 timotimo might be
21:11 mkende That's my guess, that because the location where the compiler and libs are is not writtable, some precompilation is not happening. But the issue is the same if I call perl6 as root.
21:13 jjido joined #perl6
21:13 mkende with the debug flag I get this:
21:13 mkende Stage start      :   0.000
21:13 mkende Stage parse      :      1352 RMD: Loading settings CORE
21:13 mkende 1352 RMD: Setting up default paths: . blib
21:13 mkende 1352 RMD: going to load Perl6::BOOTSTRAP
21:13 mkende 1352 RMD: Settings CORE loaded
21:13 mkende 0.930
21:13 mkende Stage syntaxcheck:   0.000
21:14 _webstrand joined #perl6
21:14 timotimo right. on my machine i get only 0.156s stage parse for -e ''
21:17 lizmat joined #perl6
21:18 mkende Another data point is that when I built rakudo, running "make test" before "make install" resulted in any tests that had a "use test" to fail.
21:18 Zoffix try with the latest and greatest rakudo?
21:19 mkende they passed after make install. The failure was related to a directory (I did not copy the name unfortunately) that could not be created
21:20 timotimo build systems are hard, let's go have ice cream
21:21 mkende might have been /usr/local/share/perl6/precomp
21:21 cognominal joined #perl6
21:21 BooK mmm ice cream
21:22 BooK so, who do I ask for a code review of my first perl6 module?
21:23 Zoffix Anyone? :) Just give a github link
21:23 BooK I know it needs fixing, and I wanted to write some test for some of the internal functions (not methods)
21:23 BooK https://github.com/book/p6-Git-Vers​ion/blob/master/lib/Git/Version.pm
21:23 Zoffix My eyes!!
21:23 Zoffix 😜
21:23 BooK it's a port of Perl5  http://metacpan.org/pod/Git::Version
21:24 BooK Zoffix: sadly, my font does not have this unicode character
21:24 Zoffix m: '😜'.uninames.say
21:24 camelia rakudo-moar 1d93ad: OUTPUT«(FACE WITH STUCK-OUT TONGUE AND WINKING EYE)␤»
21:24 BooK nice
21:25 BooK emojis will be the undoing of unicode
21:25 Zoffix I'd write these three lines with `given` https://github.com/book/p6-Git-Versio​n/blob/master/lib/Git/Version.pm#L41  Like https://gist.github.com/zoffixznet​/f4bd43e24bb33e660eb756ee00f8d06d
21:26 cognominal Book, why don't you calculate %cached from a here-in to protect our collective eyes.
21:26 BooK here-in ?
21:27 BooK the cache is initialized with all the weird cases
21:27 BooK in practice caching that kind of stuff is useless, given the number of time one does git version comparisons (except in the test script)
21:28 cognominal BooK heredoc, sorry
21:28 BooK Zoffix: oh given, yes of course
21:28 BooK I don't like the push/append/push at the end either
21:29 Zoffix And I'd write this block as `@r = ( -1, @v.pop ) if @v[*-1] ~~ s/^rc//;`, but I'm guessing you wrote it that way for performance reasons https://github.com/book/p6-Git-Versio​n/blob/master/lib/Git/Version.pm#L63
21:29 AlexDaniel .seen as
21:29 yoleaux I haven't seen as around.
21:29 BooK I'm not at the stage of doing things for performance reasons :-)
21:30 Zoffix m: for ^1000 { my $foo = 'foobar'; s/^foo//; }; say now - INIT now
21:30 camelia rakudo-moar 1d93ad: OUTPUT«0.10660804␤»
21:30 Zoffix m: for ^1000 { my $foo = 'foobar'; $foo.subst: 0, 2 }; say now - INIT now
21:30 camelia rakudo-moar 1d93ad: OUTPUT«0.17093622␤»
21:30 Zoffix oh heh. Regex is faster
21:30 Zoffix Well, "faster" :) camelia isn't made for benchmarks
21:30 AlexDaniel ok, the book is talking, now I only have to see as…
21:31 masak 'night, #perl6
21:31 Zoffix night
21:31 webstrand So you can leave the { } off of phasers? that's really cool
21:32 AlexDaniel Zoffix: interestingly, you can probably use bisectable to find performance regressions. The regression has to be pretty bad though
21:32 webstrand Every new thing I find in perl 6 is a joy
21:32 Zoffix And this may or may not be the same thing as the three append/push/append lines you got: @v.append: 0 xx 4-@v, |@r, $c
21:32 AlexDaniel always wanted to try it
21:33 * Zoffix goes back to killing people
21:33 Zoffix (in FC4 :P)
21:35 timotimo what's that again?
21:35 dalek snake: fb46df5 | (Pawel Murias)++ | src/Snake/ (2 files):
21:35 dalek snake: Accept more ways of indenting that Python 3 accepts.
21:35 dalek snake: review: https://github.com/arnshol​t/snake/commit/fb46df5891
21:35 dalek snake: ff58c63 | (Pawel Murias)++ | src/Snake/Grammar.nqp:
21:35 dalek snake: Reject indentation as inconsistent if a source file mixes tabs and spaces in a way that makes the meaning dependent on the worth of a tab in spaces.
21:35 dalek snake: review: https://github.com/arnshol​t/snake/commit/ff58c63791
21:35 Zoffix Far Cry 4
21:35 timotimo ah
21:35 pmurias joined #perl6
21:36 TEttinger soon to be followed by Far Cry 4: Bloodier Dragonner
21:36 Zoffix Nah, it was followed by Far Cry: Primal
21:37 TEttinger I do like the idea of FC: Blood Dragon
21:37 TEttinger and the soundtrack is good
21:37 pmurias arnsholt: fixed the indentation handling in snake, had to check the python source to figure out how it works ;)
21:37 pomJ joined #perl6
21:37 Zoffix Really? I absolutely hated the soundtrack. It made otherwise fun game depressing
21:38 TEttinger I do admit I haven't seen how it's been paired in-game
21:38 TEttinger it does seem like it doesn't quite fit the titles
21:39 BooK Zoffix: thanks for the quick review
21:39 Zoffix no problem
21:39 BooK one last question: how can I call normalize() from outside ?
21:39 Zoffix Add `is export` on it
21:40 Zoffix It will be exported.
21:40 BooK that was just for test, I don't want to export it
21:40 Zoffix Oh
21:41 Zoffix m: module Foo { our sub bar { say 'hello' } }; Foo::bar
21:41 camelia rakudo-moar 1d93ad: OUTPUT«hello␤»
21:42 Zoffix I guess that may work (note the `our`), but that would let users use it too
21:42 mkende Zoffix: "latest and greated rakudo" does not compile :-p
21:42 mkende --vmlibs=dynext/libperl6_o​ps_moar.so=Rakudo_ops_init gen/moar/m-BOOTSTRAP.nqp
21:42 mkende Unknown constant 'CONTROL_RETURN'
21:43 Zoffix mkende, how did you try to compile it?
21:43 BooK Zoffix: ok
21:43 mkende the first line that was not copied above: /usr/local/bin/nqp-m --target=mbc --output=blib/Perl6/BOOTSTRAP.moarvm --encoding=utf8
21:43 TEttinger Zoffix: off topic, but I think maybe Karma To Burn might have been a better fit for the style of game than Power Glove https://www.youtube.com/watch?v=8bcCZ-tor8g
21:43 mkende git pull and then "make"
21:44 mkende I'm not using rakudo star if that is the question.
21:44 timotimo you'll have to check the exact contents of build/tools/*_REVISION
21:44 Zoffix mkende, but did you update nqp and moar?
21:44 timotimo that's how we ensure you get compatible versions
21:46 Zoffix CONTROL_RETURN ... that sounds like something to do with `return` stuff and that got overhaul on moarvm level a few days ago
21:46 mkende ah... well I will update them then :)
21:50 timotimo maybe you actually need a branch in nqp or rakudo for that to work. i dunno.
21:50 timotimo sometimes after a git pull and such you really should Configure.pl or at least make clean
21:52 _webstrand joined #perl6
21:58 rgrinberg Can grammars handle "resumable" parsing? By that I mean they can say something like "not enough input" and let you resume the parser with more input
21:59 timotimo we don't have support for that yet, sadly
21:59 timotimo not sure how much work it is to build such a solution yourself
22:00 rgrinberg timotimo: eh? were you answering me?
22:01 timotimo yes
22:01 rgrinberg timotimo: OK I see, thanks.
22:02 rgrinberg First time I'm looking at perl6 and grammars look very cool though. They seem like someone took ragel and embedded in the language.
22:02 rgrinberg Ragel's quite a powerful tool
22:04 timotimo haven't heard of it yet
22:05 jdv79 rgrinberg: you worked in nyc?
22:05 rgrinberg it's pretty old but definitely under used IMO - http://www.colm.net/open-source/ragel/
22:06 rgrinberg jdv79: briefly, why?
22:06 jdv79 boris?  i worked with a mr. grinberg once.
22:07 rgrinberg jdv79: Not Boris here :) It's a common name anyway, especially in NYC.
22:08 jdv79 yeah but the perl world is small:).  ok.
22:10 Emeric1 joined #perl6
22:19 lambd0x Later guys o/
22:28 mkende Well same things with the latest version of rakudo.
22:28 mkende Anyway thanks for the advice, I'll go back to using a version installed in my home dir for the moment.
22:29 Zoffix weird
23:03 bjz joined #perl6
23:04 skids joined #perl6
23:04 webstrand joined #perl6
23:08 sherwool joined #perl6
23:08 Score_Under left #perl6
23:32 mr-foobar joined #perl6
23:43 Xliff_zzzz When adding invocant declarations to signatures for role methods. Would you do Role::Name:D: ?
23:48 timotimo there's code examples for that in the core setting
23:48 timotimo i think you need ::?CLASS or something like that?
23:59 BenGoldberg joined #perl6

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

Perl 6 | Reference Documentation | Rakudo