Camelia, the Perl 6 bug

IRC log for #perl6, 2013-12-20

Perl 6 | Reference Documentation | Rakudo | Niecza | Specs

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

All times shown according to UTC.

Time Nick Message
00:00 lue r: my Array[Int] @a = 1,2,3; say @a.perl; @a[1] = Mu; say @a.perl
00:00 camelia rakudo-parrot 413650: OUTPUT«Type check failed in assignment to '@a'; expected 'Array[Int]' but got 'Int'␤  in method REIFY at gen/parrot/CORE.setting:8320␤  in method reify at gen/parrot/CORE.setting:7196␤  in method reify at gen/parrot/CORE.setting:7183␤  in method reify a…»
00:00 camelia ..rakudo-jvm 413650: OUTPUT«Type check failed in assignment to '@a'; expected 'Array[Int]' but got 'Int'␤  in method REIFY at gen/jvm/CORE.setting:8294␤  in block  at gen/jvm/CORE.setting:7177␤  in block  at gen/jvm/CORE.setting:7167␤  in block  at gen/jvm/CORE.setting:7267␤…»
00:00 lizmat r: my @a is default(42); say @a; say @a[0]:exists; @a[0] = Nil; say @a[0]:exists
00:00 camelia rakudo-parrot 413650, rakudo-jvm 413650: OUTPUT«␤False␤True␤»
00:01 lizmat even without the "is default":
00:01 lizmat my @a; say @a; say @a[0]:exists; @a[0] = Nil; say @a[0]:exists
00:01 lizmat r: my @a; say @a; say @a[0]:exists; @a[0] = Nil; say @a[0]:exists
00:01 camelia rakudo-parrot 413650, rakudo-jvm 413650: OUTPUT«␤False␤True␤»
00:02 lizmat lue: or are you agreeing with me  :-)
00:02 lue Would I be able to, say, assign a Mu or Any to an Array[Int] to do what Nil does now?
00:02 lizmat nope, not according to spec afaik
00:03 lue :/  I just like how Nil lets me set a spot as a type object / default value without me caring too much about it.
00:03 lue .oO(maybe a new Nada thing for old Nil behavior? ☺)
00:03 telex joined #perl6
00:04 lue lizmat: I don't object to the ":delete synonym" behavior of Nil by the way. I just happen to like Nil's current behavior too :)
00:04 lizmat the spec actually states: If the container's contents are deleted, the value is notionally set to the provided default value; this value may or may not be physically represented in memory, depending on the implementation of the container.  You should officially not care about that (much).
00:05 lizmat but it's not consistent with :delete (is my feeling)
00:05 lue lizmat: I suppose I'm thinking of times where my Array should really be a Buf, in terms of what I'm caring about :D
00:06 timotimo i don't want to do the code moving :(
00:06 lizmat anyways, busy day coming up and relatively early rise
00:07 lizmat so sleep&
00:08 lue ♞ lizmat o/
00:17 cognominal joined #perl6
00:25 finanalyst joined #perl6
00:35 colomon joined #perl6
01:13 Entonian joined #perl6
01:23 xenoterracide joined #perl6
01:41 dalek specs: 2387578 | larry++ | S07-lists.pod:
01:41 dalek specs: mention that we force sink context on some loops
01:41 dalek specs: review: https://github.com/perl6/specs/commit/2387578626
01:45 TimToady moritz: thinking about tap/map, I'd say map always returns a Supply that you can put further taps on, but a tap is an endpoint, and closing a tap is done from the user end rather than the supplier end
02:03 jnap joined #perl6
02:25 eternaleye joined #perl6
02:42 * timotimo has had hiccups for about 1 hour now
02:46 colomon :(
02:46 timotimo makes it very hard to fall asleep
02:48 colomon I can imagine.
02:48 timotimo drinking a big glass of water didn't help
02:49 colomon rj: say sprintf("%.1f", 1.969)
02:50 camelia rakudo-jvm 413650: OUTPUT«1.0␤»
02:50 colomon I am having problems recompiling nqp and the rakudo.
02:52 colomon I fixed that test locally, I think, but I am having one heck of time getting rakudo rebuilt with my nqp changes.
02:53 colomon I did it somehow, once.  But I don't know what I did.
03:02 jnap joined #perl6
03:06 timotimo i shall try lying down again :\
03:21 BenGoldberg joined #perl6
03:52 gfldex joined #perl6
04:19 gdey joined #perl6
04:21 kaleem joined #perl6
04:58 preflex_ joined #perl6
05:09 gdey joined #perl6
05:14 eternaleye joined #perl6
05:15 finanalyst joined #perl6
05:21 jnap joined #perl6
05:39 SamuraiJack joined #perl6
05:41 cognominal joined #perl6
05:42 rurban1 joined #perl6
06:10 xenoterracide joined #perl6
06:19 kaleem joined #perl6
06:21 moritz TimToady: ok, thanks for the explanation
06:22 [Sno] joined #perl6
06:34 dayangkun joined #perl6
06:39 krunen joined #perl6
06:47 gdey joined #perl6
06:54 denis_boyun joined #perl6
06:59 darutoko joined #perl6
06:59 xenoterracide joined #perl6
07:01 rurban1 joined #perl6
07:15 [Coke] [A
07:16 diakopter ]
07:18 [Coke] "ww"
07:44 sqirrel joined #perl6
07:53 pochi_ joined #perl6
07:56 rindolf joined #perl6
07:58 FROGGS joined #perl6
08:04 dmol joined #perl6
08:04 darutoko- joined #perl6
08:23 jnap joined #perl6
08:28 brrt joined #perl6
08:32 Tene joined #perl6
08:32 sqirrel joined #perl6
08:33 djanatyn joined #perl6
08:33 Piers joined #perl6
09:18 finanalyst joined #perl6
09:22 pochi joined #perl6
09:37 salv0 joined #perl6
09:54 fhelmberger joined #perl6
10:01 rurban1 joined #perl6
10:09 clkao_ win 25
10:24 jnap joined #perl6
10:36 tgt joined #perl6
10:50 timotimo o/
10:50 FROGGS \o
11:09 masak clkao_: you win 25 internets :)
11:09 masak oh o/, Perl6
11:09 masak #perl6*
11:11 timotimo hiccups ... nooooo :(
11:14 * masak .oO( #firstworldproblems )
11:28 lizmat_ joined #perl6
11:34 timotimo you may think it's funny, but it actually quite hurts :(
11:39 gfldex joined #perl6
11:46 atroxaper joined #perl6
12:02 GlitchMr joined #perl6
12:05 cognominal joined #perl6
12:07 daniel-s joined #perl6
12:29 logie joined #perl6
12:33 PZt joined #perl6
12:35 sqirrel joined #perl6
12:53 timotimo no perl6 tuits have been spent yet today?
12:53 daniel-s joined #perl6
12:54 tadzik because it's friday: http://www.belarr.com/bakercat/
12:54 tadzik (sound obligatory)
12:56 denis_boyun joined #perl6
13:05 timotimo te hiccups are back :\
13:13 lizmat joined #perl6
13:37 finanalyst joined #perl6
13:45 ajr joined #perl6
13:47 gtxcomm joined #perl6
13:53 gdey joined #perl6
13:53 kaare__ joined #perl6
13:57 PacoAir joined #perl6
14:04 jnap joined #perl6
14:05 tgt joined #perl6
14:06 araujo joined #perl6
14:25 lizmat_ joined #perl6
14:26 rurban1 joined #perl6
14:29 treehug88 joined #perl6
14:35 xenoterracide joined #perl6
14:36 jnap just catching up on advent, and I have to say, http://perl6advent.wordpress.​com/2013/12/16/day-16-slangs/  is very cool
14:37 jnap I do wonder though if we need to support everything in Perl5 for it to be very useful.  Perhaps the subset of Perl5 we consider modern is fine?
14:43 xinming joined #perl6
14:46 bluescreen10 joined #perl6
14:56 FROGGS jnap: my goal is that we can use Perl 5 modules, and these will tell us what v5 would need to support
14:56 colomon jnap: I don't think… oh, he can speak for himself.  :)
14:57 FROGGS :o)
14:58 colomon FROGGS++ # nice post
14:58 skids joined #perl6
14:58 mtk joined #perl6
14:58 FROGGS thanks
14:58 skids timotimo: you know the glass-of-water is just a device to make you hold your breath.  CO2 calms the hiccup nerve.
15:00 jnap FROGGS:  being able to pull in some important bits of my regular stack would really help.  DBIx::Class and Plack being the two most important things that I don't see coming to P6 anytime quickly.  But that involves DBI and so forth, so might be hard :(
15:01 jnap I did see a PSGI/Plack perl6 project somewhere but not sure how evolved it was.  given how much more Perl6 has out of the box it might be worth revisting the PSGI spec and porting it to something that makes sense for Perl6
15:04 FROGGS okay, then I target DBIx::Class (and its gazillion deps) when tackling the next things to do
15:04 moritz XS!
15:05 FROGGS O.o
15:05 FROGGS ó.ò
15:08 logie joined #perl6
15:10 thou joined #perl6
15:25 renormalist joined #perl6
15:45 ajr joined #perl6
15:45 rurban1 joined #perl6
15:50 hummeleB1 joined #perl6
16:00 jnap joined #perl6
16:06 fhelmberger joined #perl6
16:20 japhb DBIx::Class and similar beasts are probably much more suited to implementation in Perl 6 than they were to Perl 5.  Having a strong metamodel and type system is pretty damn cool when you're creating interfaces to foreign data models.  (Speaking from Protocol Buffers experience.)
16:20 FROGGS joined #perl6
16:21 japhb Talking of which ...
16:21 * japhb wonders whether to hack p6-pb, Net::IRC, or MAIN_HELPER
16:21 japhb .oO( HACK ALL THE THINGS )
16:22 FROGGS jnthn: are you still there?
16:22 japhb FROGGS: I believe he has already begun his travels.
16:22 FROGGS k
16:22 FROGGS maybe somebody else can help, so I'll tell:
16:23 FROGGS I try to access the symbols of an enum in sub EXPORT_SYMBOL in traits.pm, but I fail to do so... has somebody an idea?
16:24 japhb p: say Enum.^mro
16:24 camelia rakudo-parrot 413650: OUTPUT«(Enum) (Any) (Mu)␤»
16:24 japhb say Enum.^methods
16:24 japhb p: say Enum.^methods
16:24 camelia rakudo-parrot 413650: OUTPUT«new BUILD invert key kv value keys values pairs fmt at_key FLATTENABLE_LIST FLATTENABLE_HASH of ACCEPTS Str perl␤»
16:24 FROGGS japhb: I can't call these methods there...
16:25 * japhb goes to look
16:27 japhb Er, why can't you access those methods?
16:27 FROGGS maybe they get added later?
16:28 japhb Oh, hmmm.
16:28 FROGGS traits.pm comes right after the prolog
16:28 japhb prolog?
16:29 japhb Oh, you mean, before most of the setting.
16:30 FROGGS yeah
16:30 FROGGS /usr/bin/perl tools/build/gen-cat.pl parrot src/core/core_prologue.pm src/core/traits.pm src/core/Positional.pm ...
16:32 japhb p: enum Foo < bar baz >; say Foo.WHAT;
16:32 camelia rakudo-parrot 413650: OUTPUT«(Foo)␤»
16:32 japhb p: enum Foo < bar baz >; say Foo.^mro;
16:32 camelia rakudo-parrot 413650: OUTPUT«(Foo) (Int) (Cool) (Any) (Mu)␤»
16:33 japhb p: enum Foo < bar baz >; say bar.^mro;
16:33 camelia rakudo-parrot 413650: OUTPUT«(Foo) (Int) (Cool) (Any) (Mu)␤»
16:33 japhb p: enum Foo < bar baz >; say Foo.^roles;
16:33 camelia rakudo-parrot 413650: OUTPUT«No such method 'roles' for invocant of type 'Perl6::Metamodel::EnumHOW'␤  in block  at /tmp/uUxmuttE_H:1␤  in any  at /tmp/uUxmuttE_H:1␤  in any  at gen/parrot/stage2/NQPHLL.nqp:1146␤  in any eval at gen/parrot/stage2/NQPHLL.nqp:1133␤  in any e…»
16:34 japhb p: enum Foo < bar baz >; say Foo.HOW.WHAT
16:34 camelia rakudo-parrot 413650: OUTPUT«(Perl6::Metamodel::EnumHOW)␤»
16:34 FROGGS I think enum Foo results in a class Foo with bar and baz being nested classes... so I thought I can access it via .HOW
16:34 japhb p: enum Foo < bar baz >; say Foo.HOW.^methods
16:34 camelia rakudo-parrot 413650: OUTPUT«No such method 'gist' for invocant of type 'NQPRoutine'␤  in method gist at gen/parrot/CORE.setting:6807␤  in method gist at gen/parrot/CORE.setting:1014␤  in sub say at gen/parrot/CORE.setting:12960␤  in block  at /tmp/C2UVFjDoJ6:1␤  in any  a…»
16:35 japhb p: enum Foo < bar baz >; .Str.say for Foo.HOW.^methods
16:35 camelia rakudo-parrot 413650: OUTPUT«No such method 'Str' for invocant of type 'NQPRoutine'␤  in block  at /tmp/PEbTPvQXZw:1␤  in any  at /tmp/PEbTPvQXZw:1␤  in any  at gen/parrot/stage2/NQPHLL.nqp:1146␤  in any eval at gen/parrot/stage2/NQPHLL.nqp:1133␤  in any evalfiles at gen/p…»
16:35 FROGGS p: enum Foo < bar baz >; say Foo.HOW
16:35 camelia rakudo-parrot 413650: OUTPUT«Perl6::Metamodel::EnumHOW.new()␤»
16:35 japhb p: enum Foo < bar baz >; say Foo.HOW.^methods>>.name
16:35 camelia rakudo-parrot 413650: OUTPUT«archetypes new new_type add_enum_value enum_values enum_from_value enum_value_list compose set_composalizer composalize does_list set_name name set_stash_type add_stash add_attribute compose_attributes set_rw rw get_attribute_for_usage attributes add_met…»
16:35 FROGGS p: enum Foo < bar baz >; say nqp::how(Foo)
16:35 camelia rakudo-parrot 413650: OUTPUT«Perl6::Metamodel::EnumHOW.new()␤»
16:35 japhb Ah-hah!
16:36 japhb p: enum Foo < bar baz >; say Foo.^enum_values
16:36 camelia rakudo-parrot 413650: OUTPUT«("bar" => 0, "baz" => 1).hash␤»
16:37 FROGGS p: enum Foo < bar baz >; say Foo.HOW.enum_values(Foo) # I tried that
16:37 camelia rakudo-parrot 413650: OUTPUT«("bar" => 0, "baz" => 1).hash␤»
16:37 * japhb loves exploring around the metamodel
16:38 * TimToady wonders how many geeks are drawn unconsciously into this by tickling the "met a model" neurons..
16:38 japhb heh
16:40 japhb Timbus: Did you want to do the concurrent version of Net--IRC yourself, or accept a PR for it?
16:40 TimToady snowing here, which is yay in the "Californian who rarely sees snow" sense
16:40 japhb TimToady: Where is "here"?
16:40 TimToady Seattle
16:41 japhb Ah.  Given that it's going to be in the 60's for the next week here, I'm doubting we'll get any precipitation at all, let alone snow.
16:41 * TimToady realizes the Chicagoans are not particularly impresses by an inch or so of snow  :)
16:41 TimToady *pressed
16:42 TimToady well, yeah, SF is going to have the dryest calendar year on record
16:42 japhb Having grown up in a snowy place, neither am I.  Sure fun to watch the rubes freak out, though.  :-)
16:42 TimToady well, I also grew up here, so it's not like it's a brand new thing
16:43 TimToady but having lived in LA and MV for the last, oh, 33 years, I haven't had much snow since then
16:43 kaleem joined #perl6
16:44 TimToady though it did, in fact, snow in the San Fernando Valley, and stuck 0.25" for the first time in living memory, just before we moved to the bay area
16:44 japhb preflex: ask Timbus Did you want to do the concurrent version of Net--IRC yourself, or accept a PR for it?
16:44 preflex Consider it noted.
16:44 TimToady Heidi made a snowman that was about 3 inches tall
16:44 japhb heh
16:45 japhb It's portable -- just not to anywhere warm.  :-)
16:51 eternaleye joined #perl6
17:05 FROGGS japhb: `say sym.^enum_values;` results in an empty hash :o(
17:05 [Sno] joined #perl6
17:06 FROGGS wait
17:07 FROGGS in this example, the export might even run before the enum elems are known: module Foo { enum NN is export <Abe Bay> }; import Foo
17:07 FROGGS :/
17:12 denis_boyun_ joined #perl6
17:23 denisboyun joined #perl6
17:24 rurban joined #perl6
17:26 FROGGS damn it, that enum is created after exporting it
17:27 FROGGS so I might need to store something like a call to REQUIRE_IMPORT in the ast
17:27 japhb What does that do?
17:30 japhb Ah, implements runtime import.
17:33 REPLeffect joined #perl6
17:41 FROGGS but... no, that would not work
17:41 FROGGS I need the symbols at compile time
17:41 [Sno] joined #perl6
17:45 ajr joined #perl6
17:46 gdey joined #perl6
17:48 denisboyun joined #perl6
18:00 denis_boyun joined #perl6
18:01 treehug8_ joined #perl6
18:01 timotimo is jnthn already long-time-afk?
18:07 treehug88 joined #perl6
18:09 lizmat timotimo: yes, he was travelling today to his family
18:09 preflex lizmat: you have 2 new messages. '/msg preflex messages' to read them.
18:10 lizmat and am expecting he's spending some quality time afk
18:12 sqirrel joined #perl6
18:25 denisboyun joined #perl6
18:27 kivutar joined #perl6
18:28 Tedd1 joined #perl6
18:30 lizmat this just in on perl6advent: ".defined is a verb (method), even though it's being used grammatically as an adjective. Yet, :delete is an adverb, but grammatically a verb. Why is .defined a verb but :exists an adverb? Those choices are a bit confusing to me"
18:31 lizmat I'm not sure what to answer there, as I missed most of the discussions as to how and why
18:36 ssutch joined #perl6
18:43 Rotwang joined #perl6
18:45 timotimo i wonder how to get the protected static inner class in my superclass
18:45 timotimo i guess i'll just expose a method that does nothing but return a new instance of that class >_>
18:45 ajr joined #perl6
18:46 raiph joined #perl6
18:46 timotimo oh my. i have to rebuild all of rakudo to get changes in the RakudoJavaInterop to stick :(
18:46 timotimo this'll take a while :(
18:51 raiph lizmat: fwiw I also thought "can I test definedness of a slice? why :exists but .defined?"
18:58 xenoterracide are arrays in function parameters slurpy? can I do sub foo ( @bar, @baz ) { ... } ?
19:07 raiph p: sub f(@a,*@b) {say @b}; f [1], 1,2,3 # this?
19:07 camelia rakudo-parrot 413650: OUTPUT«1 2 3␤»
19:09 xenoterracide probably, interesting
19:10 FROGGS[mobile] joined #perl6
19:24 timotimo i can't get nqp::jvmrakudointerop to work :(
19:25 timotimo it says no such method 'jast' for invocant of type nqpmu
19:31 FROGGS[mobile] so there is something null-ish again in your ast
19:32 treehug88 joined #perl6
19:33 eternaleye joined #perl6
19:37 moritz when I run 'make m-test', all tests produce "No subtests run"
19:37 moritz but when I run perl t/harness --moar t/01-sanity (which is what make m-test calls), all tests succeed
19:39 FROGGS[mobile] moritz: can you change the merge=>1 to 0 and run m-test again?
19:39 FROGGS[mobile] you should then see what fails
19:39 moritz so "no subtests run"
19:40 FROGGS[mobile] hmmm
19:40 FROGGS[mobile] you are sure it passes --moar to t/harness?
19:41 moritz yes
19:41 FROGGS[mobile] weird
19:41 moritz curiosuly changing the verbisoty to 1 make the tests pass to
19:42 moritz except that t/01-sanity/20-defined.t fails an assetion in the end
19:42 moritz moar: 3rdparty/libuv/src/unix/loop.c:150: uv__loop_delete: Assertion `!((((*(&(loop)->active_reqs))[0]) == (&(loop)->active_reqs)) == 0)' failed.
19:42 FROGGS[mobile] both indicate a problem with stderr/stdout filehandles
19:45 FROGGS[mobile] for one that we might need the fix that jnthn pushed as windows-only
19:46 FROGGS[mobile] even when I think that a NYI thing for libuv on linux is unlikely
19:48 FROGGS[mobile] a test would be to pipe both stderr and stdout to somewhere and see how reliable that is
19:50 FROGGS[mobile] if libuv gave us the same handle in both cases we were in trouble (as we are now)
19:55 FROGGS moritz: that fail for you too? nqp-m -e 'nqp::say(42)' >>/dev/null 2>&1
19:58 timotimo with optimize=0 and optimize=3 i don't get past Test.pm :(
19:58 timotimo is that expected?
19:59 FROGGS no, I had thought optimize=0 would work
20:00 moritz FROGGS: it produces an exist status of 0
20:00 timotimo should it get past Test.pm?
20:01 FROGGS moritz: k, :o(
20:01 FROGGS timotimo: yes
20:01 timotimo :(
20:01 FROGGS timotimo: what about --optimize=1 ?=
20:02 timotimo i can try that
20:02 timotimo and --debug?
20:02 treehug88 joined #perl6
20:02 ivanshmakov joined #perl6
20:03 FROGGS doesn't hurt
20:08 FROGGS moritz: I can't reproduce it on my box
20:08 moritz FROGGS: :/
20:10 bluescreen10 joined #perl6
20:19 sqirrel joined #perl6
20:20 Entonian joined #perl6
20:27 FROGGS is Perl 5's `goto &name` like our `nextwith &name`?
20:28 timotimo should we get bindings to TkInter for perl6? so that we can has pretty UIs? :P
20:31 colomon joined #perl6
20:36 TimToady FROGGS: no, would be more like &name.callsame
20:37 TimToady well, .nextsame
20:37 FROGGS k, thank you
20:37 TimToady dunno if rakudo does the method form yet
20:45 treehug88 joined #perl6
20:49 Entonian joined #perl6
20:55 lue 🌲 hello world o/
20:55 dalek rakudo/nom: 2b027b4 | (Tobias Leich)++ | src/core/Backtrace.pm:
20:55 dalek rakudo/nom: addjust paths to new directory layout
20:55 dalek rakudo/nom: review: https://github.com/rakudo/rakudo/commit/2b027b4327
20:58 Timbus joined #perl6
20:59 autumn joined #perl6
20:59 sivoais_ joined #perl6
21:05 sivoais joined #perl6
21:11 lue TimToady: I just realized, we have a function that does "char" → Unicode properties, but no function to do properties → <list of chars> (to do things like  Foo(:Name(/:sPLAYING CARD/) for a deck of cards). I'm stuck on a name though :)
21:11 lue My only good thought so far is propsuni, though that could easily get confusing.
21:12 FROGGS propsuni does not sound very perlish
21:13 dalek rakudo-star-daily: 3f2f537 | coke++ | log/ (5 files):
21:13 dalek rakudo-star-daily: today (automated commit)
21:13 dalek rakudo-star-daily: review: https://github.com/coke/rakudo​-star-daily/commit/3f2f537571
21:13 TimToady do we need that function when we can just grep a list of chars?
21:13 * [Coke] ponders writing his article.
21:14 moritz or we can regex-search for uni props
21:15 lue TimToady: I don't imagine a gather/take setup with uniprops($_.chr) for ^0x11_0000 is very efficient :/
21:16 lue I just chose the :Name() property because it was the easiest one to write a valid example for. :)
21:17 TimToady how would your function be more efficient than scanning all the entries?
21:18 dayangkun joined #perl6
21:19 lue TimToady: I noticed that. There's still the "a lot of people will want to do this at some point" thing in any case :)
21:19 TimToady I guess the lists of chars are available for various properties, as long as it's a simple query
21:19 TimToady but scanning names isn't going to be much more efficient than a grep of all chars with a regex for the uni prop
21:20 lue TimToady: another idea is to add adverbs to &uniprops, to cut down on the search time.  uniprops($_, :Unicode_1_Name)  will only return Unicode_1_Name and na1 pairs in the Hash.
21:20 TimToady don't over-engineer it
21:20 TimToady but maybe that's okay
21:21 treehug88 joined #perl6
21:22 lue my @a; for ^0x11_0000 { @a.push(chr($_)) if uniprops(chr($_))<na1> ~~ /SIGN/ }   <-- I'm concerned over if this action will occur often enough for this snippet to be too much.
21:22 lue (though it's easily within the realm of modules...)
21:23 treehug88 joined #perl6
21:24 lue If we go with an adverb in uniprops, I don't know if we'd want the adverb to return a Hash with Pairs for all the aliases for the property name, or if it's just the name you used.
21:25 lue e.g. does uniprops($a, :na1) return a Hash of  Unicode_1_Name => "*", na1 => "*"  or just  na1 = "*"  ?
21:25 TimToady my @a = do for ^0x11_0000 { my $c = .chr; $c if $c ~~ /:name(/SIGN>)/ }
21:26 TimToady er, /<:name(/SIGN/)>/
21:26 TimToady or sump'n like that
21:27 lue there's that, I suppose :)
21:27 lue It's just that I realized "hey waitaminute, how do I do that‽" . I see now that there are ways, and a custom-made function isn't guaranteed to fair better than the existing solutions.
21:27 lue s/fair/fare/
21:29 lue (Of course, limiting the area of your search helps there, e.g. for 0x10_0000..^0x11_0000 to not bother with the BMP.)
21:32 lue And I'll refrain from adding adverbs to the uniprops function unless/until it's shown that uniprops("a" :Name) is much faster than uniprops("a")<Name>
21:32 FROGGS TimToady: we have subscripts for arrays and hashes, can we have something like that for accessing object attributes too?
21:33 TimToady how slow do you want it to run?
21:35 diakopter all the slows
21:36 FROGGS I basically want: class Foo { has $.a; has $.b }; my @a-and-b = Foo.new( :a(42), b(42) ).<<a b>> or so
21:37 educated_poo joined #perl6
21:37 FROGGS (a more beautiful version please :o)
21:37 educated_poo Did I miss the December Rakudo release?
21:39 lue FROGSS: Foo⁅bar⁆ ? :) [yeah, we'll probably have to play with Unicode to get something]
21:39 FROGGS ohh, cleaning the screen doesnt help O.o
21:40 FROGGS nah, would be cool if we could use < > in combination with ( )
21:40 lue Even better: since you want a hash-y access for attributes (which you normally get with Foo.bar), I found a better bracket pair: Foo⦑bar⦒
21:40 FROGGS .<(a b)> ? I guess that is too confusing
21:40 raydiak Foo<{([| a b |])}> :)
21:40 FROGGS hehe
21:41 FROGGS raydiak: at least this one isnt takes yet *g*
21:41 FROGGS taken*
21:41 TimToady educated_poo: no, still in the works
21:41 lue r: say "Foo\x[2993]bar\x[2994]"
21:41 camelia rakudo-parrot 2b027b, rakudo-jvm 2b027b: OUTPUT«Foo⦓bar⦔␤»
21:41 lue FROGGS: there's your < + (  ☺
21:41 educated_poo Okay, thanks.
21:42 educated_poo left #perl6
21:42 FROGGS lue: my font does not like it :o(
21:43 lue The stuff I've presented so far could have these Texas versions: [-bar-], <.bar.>, and <(bar)>
21:43 * FROGGS .oO(  Foo:o(<a b>)  )
21:43 FROGGS I like <(bar)>, but I have the urge to quote the bar there
21:44 lue p: say <"bar">.perl; say «"bar"».perl
21:44 camelia rakudo-parrot 2b027b: OUTPUT«"\"bar\""␤"bar"␤»
21:45 FROGGS p: class Foo { has $.a; has $.b }; say Foo.^attributes
21:45 camelia rakudo-parrot 2b027b: OUTPUT«Mu $!a Mu $!b␤»
21:46 ajr_ Overdoing the weird characters runs the risk of emulating APL and enjoying the same popularity.
21:46 FROGGS p: class Foo { has $.a; has $.b }; say Foo.new( :a(42) ).^attributes(<$!a>) # I'd be happy if that does it
21:46 camelia rakudo-parrot 2b027b: OUTPUT«Mu $!a Mu $!b␤»
21:46 lue ajr_: we're too late. We have been since we decided to allow Unicode operators :D
21:47 lue ajr_: we also have "Texas" versions of all the core Unicode-based operators.
21:47 FROGGS lue: nah, important things are still ascii-ish
21:47 lue ∪ is considered important in some circles :)
21:48 lue (don't listen to me; I rather enjoy the look (if not feel) of APL)
21:49 xenoterracide joined #perl6
21:50 jnap joined #perl6
21:51 raydiak I remember one of my former employers mocking Perl 6, with something like this in mind: http://upload.wikimedia.org/wikip​edia/commons/9/9f/APL-keybd2.svg
21:56 raydiak anyway, I did have something on topic to ask: can the R* MSI install the JVM backend?
21:57 rurban1 joined #perl6
21:57 FROGGS raydiak: I have not seen patches which would make me think that
21:58 TimToady R* does not yet support jvm, maybe next one
22:00 * lizmat is tired and goes to bed
22:00 raydiak thanks FROGGS++ and TimToady++ , all I needed to know for now...back to my spheres and cubes :)
22:00 FROGGS :o)
22:01 FROGGS if someone does the jvm+R*, I could support him/her
22:03 lue raydiak: yeah, I don't like that keyboard. Not nearly as featureful as the Space Cadet.
22:03 * lue &
22:06 djanatyn joined #perl6
22:06 gfldex joined #perl6
22:07 Tene joined #perl6
22:10 raydiak lue: I've never used such a thing, but think I saw something similar in a nightmare once :)
22:11 raydiak FROGGS: what does that entail, generally speaking?
22:12 FROGGS build system hacking
22:12 FROGGS and we would need to able to build the modules of course
22:19 raydiak I've been mulling over which ways I could best help, and it doesn't escape my notice that MSI releases are sporadic, and I have a windows box; where can I read about how the MSI is produced?
22:22 FROGGS raydiak: here https://github.com/rakudo/star
22:22 FROGGS I did the star release in august and september or so
22:22 treehug88 Hey TimToady : just wanted to personally thank you for perl and for making my software development life so, so much better. Thank you.
22:23 treehug88 ... and I know it's not JUST timtoady - so thanks to all the other players as well
22:32 jeffreykegler joined #perl6
22:49 xinming joined #perl6
22:52 raydiak FROGGS: I'll read about it and hassle you if/when I pursue it; life is busy atm, but I would love to contribute something concrete sooner or later
22:53 FROGGS raydiak: I'll be here :o)
22:59 raydiak on a mostly unrelated note, there's another Pray image up, CSG is almost working, and releasing this year is still quite likely
22:59 FROGGS \o/
22:59 FROGGS raydiak++
22:59 raydiak I hope spouting off about my own project here isn't too obnoxious...it's mostly b/c nobody around me IRL knows what any of that means :)
23:00 FROGGS link?
23:00 raydiak http://cyberuniverses.com/pray/#pray-image-8
23:00 [Coke] editing in the wordpress editor is very annoying. I have a preformatted block. I'm trying to change the next paragraph to "paragraph" style. it changes the preformatted block also. ARGH.
23:01 FROGGS raydiak: you need an environment map :o)
23:01 [Coke] any pointers on how to edit stuff in wordpress ?
23:04 xenoterracide joined #perl6
23:04 raydiak FROGGS: what I need is much more powerful hardware...my $5/mo VPS isn't going to cut it much longer for running Pray :)
23:04 FROGGS [Coke]: no idea
23:05 FROGGS raydiak: how long does it take to render the last image?
23:05 [Coke] and it loses leading spaces on preformatted code... wtf?
23:06 FROGGS I remember the time when I rendered stuff using 3dsmax on a dual eleron 466MHz... most stuff took a day for a single image
23:06 raydiak [Coke]: don't use the WYSIWYG editor?
23:06 [Coke] raydiak: then I have to format it when I paste it in later. Just trying to save a step.
23:06 [Coke] I cannot be the first advent poster to have this issue.
23:06 raydiak FROGGS: it depends on a lot on the scene and there's a lot of polishing to be done, but the last image took somewhere in the 5-10 hour range, while I slept
23:07 FROGGS yeah, and then you spot a glitch and you don't intent to sleep again for another 10 hours :/
23:08 raydiak [Coke]: I had several issues with it, as I have with pretty much every web-based WYSIWYG HTML editor I've used in my lifetime
23:08 FROGGS raydiak: do you know the brazil renderer plugin of 3dsmax?
23:08 raydiak FROGGS: that's the game right now...last night's CSG render didn't come out like I was hoping either
23:09 raydiak FROGGS: nope, I don't have much experience w/3dsmax
23:09 FROGGS it rendered to independent buckets, and several threads an run in parallel where each renders a single bucket
23:10 FROGGS so if you can render a, say, 32 by 32 pixel area in a channel on the jvm.... that might speed it up a "bit"
23:10 raydiak hmmm...the p5 version of pray would SCP itself over the network and then call itself over ssh and wait for the results from each such node
23:11 raydiak iow, multithreading is definitely in the works, and maybe poor man's clustering like I just described, too
23:11 FROGGS ahh, so you can even run it on several machines for one image?
23:11 FROGGS that is nice :o)
23:13 raydiak it was out of necessity, in a way...the unofficial subtitle of the p5 version was "Pray (that you're still living by the time it's done rendering)"
23:14 raydiak but that was like a decade ago, too, so hardware was slower
23:22 FROGGS gnight
23:22 raydiak \o
23:25 lue [Coke]: I personally type up my post in my good ol' text editor (just plain text for me, others differ), and then paste it into the non-WYSIWYG tab of editor, and do the sensitive handling of code blocks and stuff before I go back to WYSIWYG mode.
23:25 lue (i.e. I keep Wp from applying destructive formatting changes until after I've protected what would be destroyed.)
23:27 sftp_ joined #perl6
23:32 xenoterracide joined #perl6
23:39 BenGoldberg joined #perl6
23:40 skids joined #perl6
23:51 raiph joined #perl6
23:55 Entonian joined #perl6
23:57 japhb_ p: say 512 * 512 / (5 * 60 * 60)
23:57 camelia rakudo-parrot 2b027b: OUTPUT«14.563556␤»
23:58 japhb_ So raydiak's latest image rendered at ~ 10 pixels per second.
23:58 xinming joined #perl6
23:58 japhb_ It would be fun to get an order of magnitude or two improvement there.
23:58 BenGoldberg joined #perl6

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

Perl 6 | Reference Documentation | Rakudo | Niecza | Specs