IRC log for #perl6, 2012-08-09

Time	Nick	Message
00:02	sirrobert	r: class A {has $.a = 1}; my %cfg = {a => 2}; my $a = A.new(%cfg); say $a.a;
00:02	p6eval	rakudo 038a3c: OUTPUT«Default constructor only takes named arguments␤ in method new at src/gen/CORE.setting:609␤ in block at /tmp/aObT_XotEW:1␤␤»
00:02	sirrobert	how do you expand a hash to use this way?
00:02	diakopter	r: class A {has $.a = 1}; my %cfg = {a => 2}; my $a = A.new(\|%cfg); say $a.a;
00:03	p6eval	rakudo 038a3c: OUTPUT«2␤»
00:03	sirrobert	ahhh
00:03	sirrobert	I think I saw that before somewhere
00:03	sirrobert	what is the \| ... operator?
00:03	diakopter	"flattening"
00:03	sirrobert	makes sense =)
00:03	sirrobert	thanks
00:09	supernovus	Well, I'm out for now, sirrobert, if you like WWW::App, take a look at WWW::App::Easy -- https://github.com/supernovus/perl6-www-app-easy --- it's still a work in progress, but it'll be ready soon.
00:19	sirrobert	will do
00:19	sirrobert	thanks =)
00:19	sirrobert	I'm out too, all.
00:39		benabik_ joined #perl6
00:40	dalek	std: ff6eac2 \| larry++ \| Actions.pm:
00:40	dalek	std: better(?) warning suppression
00:40	dalek	std: review: https://github.com/perl6/std/commit/ff6eac2279
00:40	dalek	std: df68299 \| larry++ \| STD.pm6:
00:40	dalek	std: 2nd whack at speccish sigspace parsing
00:40	dalek	std: review: https://github.com/perl6/std/commit/df682992bd
00:40	dalek	std: 820200c \| larry++ \| STD.pm6:
00:40	dalek	std: repeat requires while or until
00:40	dalek	std: review: https://github.com/perl6/std/commit/820200c05f
00:40	dalek	std: 840c157 \| larry++ \| CORE.setting:
00:40	dalek	std: fix casemap fossils
00:40	dalek	std: review: https://github.com/perl6/std/commit/840c157aa4
00:40		clkao joined #perl6
00:40	TimToady	evalbot rebuild std
00:40	p6eval	OK (started asynchronously)
00:55	diakopter	... as if it could report it started synchronously
00:56	dalek	std: adada5f \| larry++ \| STD.pm6:
00:56	dalek	std: modernize ** to +%
00:56	dalek	std: review: https://github.com/perl6/std/commit/adada5f23f
01:00	TimToady	I've taken to just writing it "foo +% bar" everywhere instead of "foo+ % bar", since the space is harmless under token, and what you usually want under rule anyway
01:00	TimToady	seems to be healthier psychologically to keep the +% together most of the time
01:00	TimToady	std: repeat { ... };
01:00	p6eval	std f43a358: OUTPUTÂ«===[0mSORRY!===[0mâ�¤Undeclared routine:â�¤ 'repeat' used at line 1â�¤Check failedâ�¤FAILED 00:00 41mâ�¤Â»
01:02	* TimToady	suspects std isn't really rebuilding
01:02	benabik	f43a358 is from jul 30
01:18	sorear	I also doubt that rebuild works
01:18	sorear	for various painful reasons p6eval's rebuilder no longer runs on the same machine as p6eval and dalek
01:45	diakopter	std: repeat { } while do do { };
01:45	p6eval	std f43a358: OUTPUT«ok 00:00 41m␤»
01:45	* TimToady	needs to manage a 'make reboot' for the ** to +% transtion
01:48	TimToady	and a 'make reboot' isn't actually successful currently...
01:56		supernovus joined #perl6
01:58	dalek	ecosystem: b080b79 \| (Timothy Totten)++ \| META.list:
01:58	dalek	ecosystem: Added WWW::App::Easy, an MVC framework for Perl 6 Web Applications.
01:58	dalek	ecosystem: review: https://github.com/perl6/ecosy[…]commit/b080b79608
02:04	diakopter	n: say -0
02:05	p6eval	niecza v19-15-g051783d: OUTPUT«0␤»
02:06	Timbus	oh wow. supernovus, this module is pretty great
02:06		Chillance joined #perl6
02:08	dalek	ecosystem: f689bf9 \| (Timothy Totten)++ \| META.list:
02:08	dalek	ecosystem: Removed Esquel, it is currently very broken, rewrite planned for later.
02:08	dalek	ecosystem: review: https://github.com/perl6/ecosy[…]commit/f689bf9d64
02:08	Timbus	im assuming content-type is text/html by default and you dont hactually have to set it per handler
02:11	diakopter	n: say &1&&&11&&&1&
02:11	supernovus	Timbus: Glad you like it. Actually, the current WWW::App (the library WWW::App::Easy is based upon) doesn't set a default Content-Type header if none was found. I'll add that to the TODO list.
02:11	p6eval	niecza v19-15-g051783d: OUTPUT«{ ... }␤»
02:12	diakopter	TimToady: what would &&**& mean above
02:12	Timbus	oh, well yeah it definitely should
02:13	TimToady	** is HyperWhatever
02:14	sorear	n: **
02:14	p6eval	niecza v19-15-g051783d: OUTPUT«Unhandled exception: Unable to resolve method postcircumfix:<( )> in type Any␤ at /tmp/pfKJUi9mXx line 1 (mainline @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3929 (ANON @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3930 (module-CORE @ 564…
02:14	sorear	n: {**}
02:14	p6eval	niecza v19-15-g051783d: OUTPUT«Unhandled exception: Unable to resolve method postcircumfix:<( )> in type Any␤ at /tmp/hV_tZIkAIG line 1 (mainline @ 6) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3929 (ANON @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3930 (module-CORE @ 564…
02:14	sorear	n: -> { ** }
02:14	p6eval	niecza v19-15-g051783d: ( no output )
02:14	sorear	whatever it is, it only dies at runtime
02:17	diakopter	n: say (&1&&&1**1&&&1&*)({},{})
02:17	p6eval	niecza v19-15-g051783d: OUTPUT«all({}, 1)␤»
02:18	diakopter	n: say (&1&&&1**1&&&1&*)({2},{3})
02:18	p6eval	niecza v19-15-g051783d: OUTPUT«Unhandled exception: Unable to resolve method postcircumfix:<( )> in type Any␤ at /tmp/g4UDJ3GsSD line 1 (ANON @ 3) ␤ at /tmp/g4UDJ3GsSD line 1 (mainline @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3929 (ANON @ 3) ␤ at /home/p6eval/niecza/lib/C…
02:18	diakopter	n: say (&1&&&1**1&&&1&*)({2},{})
02:18	p6eval	niecza v19-15-g051783d: OUTPUT«Unhandled exception: Unable to resolve method postcircumfix:<( )> in type Any␤ at /tmp/1S5DK0H61P line 1 (ANON @ 3) ␤ at /tmp/1S5DK0H61P line 1 (mainline @ 8) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3929 (ANON @ 3) ␤ at /home/p6eval/niecza/lib/C…
02:19	diakopter	n: say (&1&&&1**1&&&1&*)({},{3})
02:19	benabik	Has someone been using GA to generate P6?
02:19	p6eval	niecza v19-15-g051783d: OUTPUT«all({}, 1)␤»
02:19	diakopter	n: say (&1&&&1**1&&&1&*)({})
02:19	p6eval	niecza v19-15-g051783d: OUTPUT«Unhandled exception: No value for parameter 'anon_2' in 'ANON'␤ at /tmp/axPOJjuD1F line 0 (ANON @ 1) ␤ at /tmp/axPOJjuD1F line 1 (mainline @ 8) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3929 (ANON @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting lin…
02:19	diakopter	n: say (&1&&&1**1&&&1&*)()
02:19	p6eval	niecza v19-15-g051783d: OUTPUT«Unhandled exception: No value for parameter 'anon_0' in 'ANON'␤ at /tmp/90ic21pylj line 0 (ANON @ 1) ␤ at /tmp/90ic21pylj line 1 (mainline @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3929 (ANON @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting lin…
02:19	diakopter	sorear: it counts from anon_0 to anon_2 ?
02:20	sorear	those are gensyms
02:20	sorear	they increment in the order structures are parsedr
02:21	sorear	n (* + ); ( + *)()
02:21	sorear	n: (* + ); ( + *)()
02:21	p6eval	niecza v19-15-g051783d: OUTPUT«Unhandled exception: No value for parameter 'anon_3' in 'ANON'␤ at /tmp/4iSfBtSTyU line 0 (ANON @ 1) ␤ at /tmp/4iSfBtSTyU line 1 (mainline @ 4) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3929 (ANON @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting lin…
02:21	sorear	and they don't reset except with new files
02:21	diakopter	oh
02:21	diakopter	why would it take {} but not {3}
02:21	diakopter	for the first argument
02:22	sorear	in that last sample I think anon_2 is the * + * closure; niecza does not support truly anonymous closures
02:22	sorear	{} is a Hash, {3} is a Block
02:22	diakopter	why would it take a Hash but not a Block
02:26	supernovus	Timbus: Okay, if no Content-Type or Location header is found prior to sending the PSGI response, it adds a Content-Type of text/html.
02:27	Timbus	rad
02:28	diakopter	n: do { } & while 1 { } # <blink>
02:28	p6eval	niecza v19-15-g051783d: OUTPUT«Unhandled exception: Unable to resolve method from in type P6+{term:sym<∅>},{infix:sym<∈>},{infix:sym<(elem)>},{infix:sym<∉>},{infix:sym<∋>},{infix:sym<(cont)>},{infix:sym<∌>},{infix:sym<∪>},{infix:sym<(\|)>},{infix:sym<∩>},{infix:sym<(&)>},{infix:sym<(-)>}…
02:28	Timbus	now make it automatically pick if the url has .txt or .json at the end and you have yourself a feature
02:29	supernovus	Well, that's a bit much for WWW::App itself, but for WWW::App::Easy, I'll make some magic dispatch rules that can easily be included in a web app to do just that.
02:31	supernovus	Anyway, enough for tonight. Hope someone finds the library useful. I need to expand the documentation, as currently you need to read the WWW::App documentation in addition to the WWW::App::Easy to figure out all of the different features.
02:32	diakopter	n: do { } , while 1
02:32	diakopter	std: do { } , while 1
02:32	p6eval	niecza v19-15-g051783d: OUTPUT«(timeout)»
02:32	p6eval	std f43a358: OUTPUT«ok 00:00 43m␤»
02:33	diakopter	TimToady: what does the above comma do?
02:34	diakopter	(or anyone, of course)
02:34	TimToady	treats do{} as a list of 1 element
02:38		Psyche^ joined #perl6
02:43	diakopter	n: say <-> { } >0
02:43	p6eval	niecza v19-15-g051783d: OUTPUT«Unhandled exception: Cannot use value like Block as a number␤ at <unknown> line 0 (ExitRunloop @ 0) ␤ at /home/p6eval/niecza/lib/CORE.setting line 295 (Any.Numeric @ 6) ␤ at <unknown> line 0 (ExitRunloop @ 0) ␤ at /tmp/bGx7tNqS1k line 1 (mainline @ 3) …
02:43	diakopter	n: say <-> { }>0
02:43	p6eval	niecza v19-15-g051783d: OUTPUT«{ ... }␤»
02:43	diakopter	@.@
02:43	diakopter	what would that mean?
02:45	diakopter	(why would it not like the space in the first example but accept the 2nd example?)
02:49		orafu joined #perl6
02:52	TimToady	it has something to do with the endargs memo, so it determines whether say's args end there
02:54	diakopter	:S
02:55	diakopter	well, it parsed the first one, just interpreted the > as greater than I think.
02:56	* diakopter	tries less hard to understand
02:59	diakopter	n: say { }<->{ }<->{ }
02:59	p6eval	niecza v19-15-g051783d: OUTPUT«Any()␤»
03:00	diakopter	o_O
03:00	diakopter	why wouldn't it be a boolean
03:00	diakopter	(how else could it parse?)
03:02	diakopter	oh :)
03:02		fgomez joined #perl6
03:03		sivoais joined #perl6
03:09	benabik	What is <->?
03:10	diakopter	rn: say { '-' => 4 }<->
03:10	p6eval	rakudo 038a3c, niecza v19-15-g051783d: OUTPUT«4␤»
03:10	benabik	Ah. either a list or a subscript. right.
03:15		benabik joined #perl6
03:16		localhost joined #perl6
03:50		benabik joined #perl6
03:52		gongyiliao joined #perl6
04:03	dalek	std: e20e129 \| larry++ \| STD.pm6:
04:03	dalek	std: fix endargs wrt whitespace, diakopter++
04:03	dalek	std: review: https://github.com/perl6/std/commit/e20e12932d
04:12		crab2313 joined #perl6
04:18		fgomez joined #perl6
04:37		huf joined #perl6
04:40		alvis joined #perl6
04:49		Entonian joined #perl6
05:09		Telgalizer joined #perl6
05:37		birdwindupbird joined #perl6
05:44		adu joined #perl6
05:50		kaleem joined #perl6
06:07	moritz	\o
06:08	adu	o/
06:08	sorear	o/
06:14		wtw joined #perl6
06:24	tadzik	'/
06:26		DarthGandalf joined #perl6
06:28		zhutingting joined #perl6
06:36	diakopter	moritz: I was considering replying to the "kill the spokesbug with fire" thread with something like this:
06:38	* TimToady	thinks replying with silence is one of the good options :)
06:39	sorear	replying with silence only works if we can enforce it
06:41	moritz	"Camelia is fire-proof. Killing it is fire is just as effective a killing a fish with water"
06:42	moritz	TimToady: well, if somebody offers to contribute a different design, it's not nice to let him do all the work, and not merge it in the end. Which is why I broke my silence after a mockup was made
06:45	diakopter	"...makes language with 10 years of development look less mature than..." - Perl 6 does not want to contrive an image of maturity. Instead, it intends to be cutesy and whimsical for the foreseeable future, and let a reputation of maturity grow organically in the long term. <Please correct me if/where my understanding is wrong>
06:45	cosimo	all this discussion makes me want to create a new design for perl6.org
06:46	cosimo	:-)
06:46	diakopter	I too am not a fan of the current design, and I'm not afraid to admit it because I don't believe that only those willing/able to help out should offer critiques.
06:47	diakopter	that said, I am not able to design a better one because I suck at design.
06:47	cosimo	me too, but would be nice to try anyway
06:50	diakopter	the multicolored spokesbug would make for a nice splash of color on a page with far less color contrasts/busyness than the current one
06:51	diakopter	making the color palette of the site the same as the logo makes the logo not stand out, and is way overkill
06:54	diakopter	(especially because the logo has so many colors)
06:57	diakopter	I recommend a palette for the rest of the page with some grays and perhaps one of the logo's colors, less neon and even more pastel than the current divs on the site
07:01	diakopter	dark pastel.
07:03		GlitchMr joined #perl6
07:07		brrt joined #perl6
07:15		Saravanan joined #perl6
07:15	Saravanan	Hi Everybody
07:16	Saravanan	i want to be a perl6 contributor
07:16	Saravanan	can anyone guide me how to start
07:16	diakopter	Saravanan: welcome! do you know how you would like to contribute?
07:17	TimToady	what are your interests?
07:17	diakopter	(implementation, tests, documentation, evangelism, or others I'm sure I'm forgetting)
07:17	Saravanan	i dono
07:17	Saravanan	i am blank
07:18	Saravanan	@diakopter .. can u please guide
07:18	Saravanan	where to start
07:18	moritz	Saravanan: do you know any programming languages already? if yes, which?
07:18	Saravanan	perl
07:18	moritz	Saravanan: then maybe pick your favorite Perl 5 module, and implement it in Perl 6
07:19		SamuraiJack joined #perl6
07:19	TimToady	don't start with DBI :)
07:19	moritz	right :-)
07:19	moritz	there's a (very loose) port of that already, called DBIish
07:19	sorear	Saravanan: how long have you been using p5? programming in general?
07:20	moritz	http://modules.perl6.org/ has the existing Perl 6 modules
07:23	diakopter	Saravanan: let's start by helping you obtain a working Perl 6 install
07:23	Saravanan	4 years
07:24	Saravanan	ok..
07:24	diakopter	Saravanan: do you use Linux? Windows? Mac?
07:25	Saravanan	fedora and mac
07:25	Saravanan	but i prefer fedora
07:26	diakopter	well, there are several implementations. Let's start with rakudo. Do you have experience building softare from source?
07:26	Saravanan	no..
07:27	diakopter	okay; we can help you with that
07:27	Saravanan	this is my first time
07:28	diakopter	you might need a few things installed in your fedora. in a terminal type sudo yum groupinstall "Development Tools"
07:29	brrt	for the record, and not to be annoying, i do believe fedora comes with rakudo star
07:29	GlitchMr	Fedora?
07:30	GlitchMr	I think they have Rakudo Star 2011.07
07:30	GlitchMr	yum install rakudo-star
07:30	Saravanan	ok..
07:30	GlitchMr	If you want newer release, I guess you would have to compile
07:30	TimToady	that's pretty old though
07:30	tadzik	yeah
07:35		imarcusthis joined #perl6
07:37	diakopter	Saravanan: did you decide whether you want the year-old version of rakudo or the most recent? Install git, gcc, and make to build the latest
07:37	Saravanan	ok..
07:37	Saravanan	which will be better
07:37	Saravanan	old
07:37	Saravanan	or new
07:38	diakopter	let's try building the new one
07:38	diakopter	there have been many changes
07:38	Saravanan	ok..
07:38	Saravanan	i will back after lunch
07:45	jnthn	mornin'
07:45	tadzik	hello jnthn
07:45	diakopter	ahoy
07:46		fhelmberger joined #perl6
07:46	sorear	yo jnthn
07:59		zhutingting joined #perl6
07:59	arnsholt	o/
08:01		fhelmberger joined #perl6
08:04	adu	Saravanan: or you can do what I do
08:07	GlitchMr	std: do { } for 1;
08:08	p6eval	std f43a358: OUTPUTÂ«===[0mSORRY!===[0mâ�¤Unsupported use of do...for; in Perl 6 please use repeat...for at /tmp/ocDd4DEbqD line 1:â�¤------> do { } for 1â��;â�¤Parse failedâ�¤FAILED 00:01 41mâ�¤Â»
08:08	GlitchMr	std: repeat { } for 1;
08:08	p6eval	std f43a358: OUTPUTÂ«===[0mSORRY!===[0mâ�¤Undeclared routine:â�¤ 'repeat' used at line 1â�¤Check failedâ�¤FAILED 00:00 41mâ�¤Â»
08:19	TimToady	the first message is bogus, and the second one is fixed (but not recompiled for the evalbot)
08:20	* jnthn	will pull that second one into Rakudo later today, if he's not beaten to it
08:21	moritz	. ..--.ö..ö-8,, ,m.,m ,mlm mm
08:22	sorear	guten Morgen Ronja
08:23	moritz	danke :-)
08:29		awoodland joined #perl6
08:44	* jnthn	-> meeting
08:46	* moritz	-> 6h commute to $vacation
09:00	* sorear	-> sleep
09:00	masak	guten ugt morning, #perl6
09:00	hoelzro	morgen
09:00	tadzik	masak! \o/
09:13	Saravanan	@diakopter as of now i will use mac
09:13	Saravanan	my fedora box is troubling
09:14	Saravanan	how to install rakudo in mac
09:15	hoelzro	Saravanan: I used to do with homebrew, but the latest formula doesn't work for me
09:16	Saravanan	oh..
09:26	brrt	parrot works on a mac, that i know for sure
09:26		mucker joined #perl6
09:26	brrt	Saravanan: whats troubling?
09:27	Saravanan	that's hardware problem
09:27	brrt	oh, then irc can't help you much :-)
09:27	Saravanan	:-)
09:27	brrt	you can certainly install parrot and rakudo on a mac, although i would not now know how
09:27	Saravanan	ok..
09:28	Saravanan	i will try to add vmware in mac
09:28	Saravanan	or virtualbox
09:28		bbkr joined #perl6
09:29	brrt	thats not necessary
09:32	bonsaikitten	prefix portage would be my lazy guess
09:33		daxim joined #perl6
09:48	masak	Saravanan: welcome.
09:48	masak	sorry to hear you're having trouble building Rakudo.
09:48	masak	we'll try to help.
09:49	Saravanan	ok.
09:54	masak	also, let us know if you need any more Perl 6 resources: documentation, tutorials, tips...
09:54	masak	there's plenty at perl6.org, but sometimes it's still easier to ask here ;)
09:59	Saravanan	:-)
10:14	daxim	can someone correct the spelling on <http://rakudo.org/how-to-get-rakudo/>? s/OpenSuSE/openSUSE/
10:16	Saravanan	Successfully installed rakudo :-)
10:17	Saravanan	diakopter: i had installed rakudo
10:17	Saravanan	next step?
10:21	masak	next step: awesomeness.
10:21	masak	r: say "OH HAI $_" for 1..5
10:21	p6eval	rakudo 038a3c: OUTPUT«OH HAI 1␤OH HAI 2␤OH HAI 3␤OH HAI 4␤OH HAI 5␤»
10:28	Saravanan	its working
10:28	Saravanan	:-)
10:28	Saravanan	great
10:30	masak	\o/
10:31	masak	Saravanan: pro tip: the fastest way to Perl 6 mastery is to be curious and ask a lot of questions about things you're learning. we won't bite, and we're very helpful and responsive.
10:31	Saravanan	ok..
10:31	Saravanan	how to install modules in perl6
10:31	Saravanan	is that same as perl5?
10:31	daxim	similar
10:31	Saravanan	ok..
10:32	Saravanan	in perl6 OO is inbuild?
10:32	daxim	yes!
10:32	Saravanan	i mean pure OO
10:32	daxim	it's unpure OO, perl6 is like perl5 a multi-paradigm language
10:33	Saravanan	ok..
10:33	huf	what's a pure oo?
10:33	Saravanan	as in java or c++
10:33	daxim	when you cannot do procedural programming
10:34	Saravanan	hmm
10:34	tadzik	so not like C++
10:34	huf	hm. both java and c++ let you do procedural programming :)
10:34	huf	yeah
10:35	Saravanan	so diff btw perl5 and perl6
10:35	huf	perl6's builtin OO is like perl5+Moose except so much more
10:35	huf	i guess...
10:35	daxim	too many to name here, there's a list in the online documentation somewhere
10:35	tadzik	Saravanan: you may like the "Perl 5 to 6" series of articles
10:35	tadzik	it's linked on perl6.org/documentation I believe
10:36	Saravanan	s
10:38	daxim	I just updated the page <http://en.opensuse.org/Rakudo> (uuurgh, mediawiki), do you see the newest version with a table at the bottom or the previous one with section headings?
10:39	flussence	looks out of date here
10:39	tadzik	no table here
10:39	daxim	then I really have to wait for someone to release
10:40	Saravanan	ok
10:42	masak	Saravanan: Perl 6's OO is "pure" in the sense that even if you don't care about objects, things tend to be objects.
10:42	masak	it's also "pure" in the sense that the OO system is implemented using the OO system itself.
10:43	masak	but really, maybe there isn't really any "pure" OO out there. even Smalltalk, the most obvious candidate, is opinionated in various ways.
10:44	masak	everything I've ever read purporting to explain what OO "really" is has turned out to be yet another opinion ;)
10:44	Saravanan	:-
10:44	Saravanan	:-)
10:44	huf	:)
10:45	masak	let me show you a few cool things about Perl 6's OO, though:
10:45	daxim	one can certainly form a majority report on the opinions
10:45	GlitchMr	OOP is just syntactic sugar. Tell me difference between $something.do_something() and do_something($something)
10:45	masak	daxim: yeah, and one would end up with... Java.
10:46	masak	GlitchMr: those used to be the same in Perl 6, but they're not any more.
10:46	masak	GlitchMr: you do have a point, though. even if you're not 100% right.
10:46	huf	whoa, foo($bar) used to do method lookup? that'd be surprising....
10:46	masak	r: class A { has $.x; has $.y; }; say A.^attrs>>.name
10:46	p6eval	rakudo 038a3c: OUTPUT«No such method 'attrs' for invocant of type 'Perl6::Metamodel::ClassHOW'␤ in block at /tmp/eJjESD4QOV:1␤␤»
10:46	masak	r: class A { has $.x; has $.y; }; say A.^attributes>>.name
10:46	p6eval	rakudo 038a3c: OUTPUT«$!x $!y␤»
10:46	masak	\o/
10:46	GlitchMr	Commonly in object oriented languages object is first.
10:46	GlitchMr	In procedural, method is first.
10:47	masak	huf: no, more like methods marked with 'our' (or a star, I don't remember) got auto-exported as subroutines.
10:47	GlitchMr	It's just different order (and somewhat different syntax)
10:47	masak	GlitchMr: no, there's a real distinction between non-method procedures and methods, IMO.
10:47	masak	GlitchMr: but you can simulate a method call by pretending that the first argument is an invocant.
10:48	lumi_	GlitchMr: $something.do_something() is like invoke(get_method(get_type($something),"do_something"),$something)
10:48	masak	and it's correct that many languages/VMs actually implement it that way.
10:48	masak	lumi_++ # nice
10:48	masak	yes, that captures the mro thing, too.
10:49	GlitchMr	print $_%2 for (0 ... *)[^100]
10:49	GlitchMr	...
10:49	GlitchMr	oops
10:49	GlitchMr	wrong paste
10:49	GlitchMr	String.prototype.toUpperCase.call('hello, world')
10:49	* masak	was just going to mention JavaScript :)
10:51	masak	yes, JavaScript is another language that emulates method invocation by folding the invocant as a first argument of some underlying invocation form (.call in this case)
10:51		Saran joined #perl6
10:51	masak	which means that you can take a method of object A and call it on an unrelated object B.
10:52	masak	just like you can in Perl 5.
10:52	masak	in Perl 6, that's not possible unless B is a kind of A.
10:52	masak	r: class A { our method foo { say "OH HAI" } }; class B {}; &A::foo(B.new)
10:52	p6eval	rakudo 038a3c: OUTPUT«Nominal type check failed for parameter ''; expected A but got B instead␤ in method foo at /tmp/i7nzxuZqya:1␤ in block at /tmp/i7nzxuZqya:1␤␤»
10:52	masak	r: class A { our method foo { say "OH HAI" } }; class B is A {}; &A::foo(B.new)
10:52	p6eval	rakudo 038a3c: OUTPUT«OH HAI␤»
10:52	masak	:)
10:52	GlitchMr	I know. I use this trick to emulate super in JavaScript
10:53	masak	if you emulate super in JavaScript, let me suggest that you're not going by the grain of the language ;)
10:53	masak	it's a prototypal language, not a super language.
10:53	masak	the inheritance is differential, not classical.
10:54	GlitchMr	B = new A(); B.prototype.method = function () { A.prototype.method.apply(this, arguments); alert('But this is B!'); }
10:55	masak	yeah, I've seen that pattern.
10:55	Saran	masak.. r u talking abt the difference between perl 5 and perl 6
10:55	masak	Saran: yes.
10:57	masak	and we also brought JavaScript into the discussion.
10:57	masak	JavaScript has the same looseness in pairing methods with objects.
10:58	masak	Perl 6 is a little more strict, as you can see in the above p6eval conversation.
10:58		whiteknight joined #perl6
10:59	GlitchMr	r: module Object { sub new { bless {}, 'Object' }; sub test($self) { $self.perl.print } }; Object.new.test
10:59	p6eval	rakudo 038a3c: OUTPUT«===SORRY!===␤CHECK FAILED:␤Undefined routine '&bless' called (line 1)␤»
10:59	GlitchMr	I guess I shouldn't attempt Perl 5 code in Perl 6
11:00	Saran	saravanan This link will guide with the differences and how to learn perl6 with the help of perl5 http://perlcabal.org/syn/Differences.html
11:00	hoelzro	GlitchMr: I believe in Perl 6 it's self.bless(*)
11:00	GlitchMr	r: module Object { sub new { self.bless(*) }; sub test($self) { $self.perl.print } }; Object.new.test
11:00	p6eval	rakudo 038a3c: OUTPUT«===SORRY!===␤'self' used where no object is available␤at /tmp/Y0BG9Zq3gd:1␤»
11:00	Saravanan	whether we have bless in perl6
11:01	Saravanan	if it has inbuilt OO then why we need bless in perl6?
11:02	tadzik	bless is a part of this OO we have
11:02	masak	GlitchMr: you can only use 'self' inside methods.
11:02	GlitchMr	r: module Object { sub method { self.bless(*) }; method test { self.perl.print } }; Object.new.test
11:02	p6eval	rakudo 038a3c: OUTPUT«===SORRY!===␤'self' used where no object is available␤at /tmp/BDe8BcnJ5X:1␤»
11:03	GlitchMr	r: module Object { method new { self.bless(*) }; method test { self.perl.print } }; Object.new.test
11:03	p6eval	rakudo 038a3c: OUTPUT«Useless declaration of a has-scoped method in module␤Useless declaration of a has-scoped method in module␤You cannot create an instance of this type␤ in method bless at src/gen/CORE.setting:618␤ in method new at src/gen/CORE.setting:606␤ in block at /tmp/m4kNnL…
11:03	masak	GlitchMr: and calling a sub 'method' doesn't make it a method :P
11:03	masak	r: my $m = method { self.bless(*) }; class A {}; say A."$m"()
11:03	p6eval	rakudo 038a3c: OUTPUT«No such method '<anon>' for invocant of type 'A'␤ in block at /tmp/mizHWATMNb:1␤␤»
11:03	masak	hm :)
11:03	masak	r: my $m = method { self.bless(*) }; class A {}; say $m(A)
11:03	p6eval	rakudo 038a3c: OUTPUT«A.new()␤»
11:03	masak	\o/
11:04	masak	r: my $m = method { self.bless(*, :x<42>) }; class A { has $.x }; say $m(A)
11:04	p6eval	rakudo 038a3c: OUTPUT«A.new(x => "42")␤»
11:04	GlitchMr	http://rosettacode.org/wiki/No[…]es/Convert#Perl_6
11:04	GlitchMr	Is it possible without eval (which he by accident called "val")
11:05	masak	no, it's "val".
11:05	masak	it's an eval for values.
11:05	GlitchMr	I'm confused
11:05	masak	and no, it's not possible to parameterize on the base. that's why "val" is used.
11:06	masak	n: say val ":2<1110>"
11:06	p6eval	niecza v19-15-g051783d: OUTPUT«:2<1110>␤»
11:06	masak	hrm.
11:06	GlitchMr	But, can I do :16('FF') or something like this?
11:06	masak	n: say val "0b1010"
11:06	p6eval	niecza v19-15-g051783d: OUTPUT«0b1010␤»
11:06	GlitchMr	It's good enough for me then
11:06	masak	n: say val "0x55"
11:06	p6eval	niecza v19-15-g051783d: OUTPUT«0x55␤»
11:06	Saran	saravanan ..it is advised to go with few basics of perl6..then it will be easy to understand
11:06	masak	sorear: I think all those should be parsed as ints.
11:09	Saravanan	ok..
11:11		MayDaniel joined #perl6
11:16	masak	rn: say "ß" ~~ m:i/ß/; say "ß" ~~ m:i/SS/; say "SS" ~~ m:i/ß/;
11:16	p6eval	rakudo 038a3c: OUTPUT«q[ß]␤␤#<failed match>␤#<failed match>␤»
11:16	p6eval	..niecza v19-15-g051783d: OUTPUT«#<match from(0) to(1) text(ß) pos([].list) named({}.hash)>␤False␤False␤»
11:17	masak	what's the rule for case insensitiveness in regexes?
11:17	masak	rn: say "ß".uc
11:17	p6eval	rakudo 038a3c, niecza v19-15-g051783d: OUTPUT«ß␤»
11:18	masak	hm.
11:19		JimmyZ joined #perl6
11:25	arnsholt	Pretty sure that's wrong
11:25	masak	aye.
11:33	GlitchMr	http://codegolf.stackexchange.com/a/6936/3103
11:33	GlitchMr	I'm almost sure I could golf this better :)
11:43	Timbus	say [*] [xx] .words for get
11:44	masak	say [*] [xx] get>>.words
11:44	Timbus	i was trying to do that
11:44	masak	one shorter if you use » :)
11:46	Timbus	»:)
11:46	Timbus	klingon smile
11:49	GlitchMr	The problem is that both Timbus and masak solutions end after first line :P
11:49	GlitchMr	But he hasn't specified that
11:49	GlitchMr	So it could work too
11:49	Timbus	nput given as: x y (separated by a space)
11:50	Timbus	looks perfect to me
11:50	GlitchMr	I'm already confused
11:52	GlitchMr	Perl 6 doesn't win with GolfScript, but it has chance in CodeGolf :)
11:52	GlitchMr	...
11:52	GlitchMr	Except I did it wrongly
11:55	Timbus	you did!
12:02	Timbus	say [**] [xx] get>>.words is the obvious one but incurs a penalty
12:06	Timbus	say [*] [xx] @ARGS
12:07		zhutingting left #perl6
12:08		MayDaniel joined #perl6
12:09	Timbus	30 pts including penalty. not bad
12:16	flussence	I'm getting nowhere but this looks interesting -
12:16	flussence	given '2 4'.words { say [*] [xx] .[0], ^(.[1])»+»1 }
12:16	flussence	r: given '2 4'.words { say [*] [xx] .[0], ^(.[1])»+»1 }
12:16	p6eval	rakudo 038a3c: OUTPUT«16777216␤»
12:16	flussence	only 8 bits off :)
12:19	Timbus	lol. what's it meant to be
12:19	Timbus	r: say [**] [xx] (2,4)
12:19	p6eval	rakudo 038a3c: OUTPUT«65536␤»
12:19	Timbus	close!
12:21		xinming joined #perl6
12:30	masak	r: say [**] [xx] 2,4
12:30	p6eval	rakudo 038a3c: OUTPUT«65536␤»
12:30	masak	r: say [**][xx]2,4
12:30	p6eval	rakudo 038a3c: OUTPUT«===SORRY!===␤Confused␤at /tmp/3U4zQfKKYR:1␤»
12:31	masak	r: say [**] [xx]2,4
12:31	p6eval	rakudo 038a3c: OUTPUT«===SORRY!===␤Confused␤at /tmp/g0ag4TFkC7:1␤»
12:31	masak	:)
12:32	GlitchMr	Why reduce operators care so much about spaces?
12:32	JimmyZ	array
12:33	JimmyZ	r: say $a[xx][*]
12:33	p6eval	rakudo 038a3c: OUTPUT«===SORRY!===␤Variable $a is not declared␤at /tmp/g_I4Uhls57:1␤»
12:34	JimmyZ	r: say $a [xx] [*]
12:34	p6eval	rakudo 038a3c: OUTPUT«===SORRY!===␤Variable $a is not declared␤at /tmp/392OiiL3_2:1␤»
12:41		sirrobert joined #perl6
12:41	sirrobert	hi p6
12:42	tadzik	hi
12:46	sirrobert	what is the : in this syntax doing: $foo.somemethod: $somesub;
12:47	tadzik	it's like $foo.method($sub)
12:47	sirrobert	ok
12:47	sirrobert	does it have a particular name?
12:48	tadzik	I don't know
12:50	sirrobert	huh, it's only for methods, not subs
12:50	JimmyZ	well, for subs is: sub args
12:51	sirrobert	yeah
12:51	sirrobert	was just understanding the ... domain of the operator
12:52	masak	I tend to think of it as the "rest of the statement is method args" colon :)
12:52	sirrobert	heh
13:03		jeffreykegler joined #perl6
13:04		SamuraiJack joined #perl6
13:06		SamuraiJack joined #perl6
13:08		Saravanan1 joined #perl6
13:08	PerlJam	r: sub foo ($alpha: $beta) { } # just checkign
13:08	p6eval	rakudo 038a3c: ( no output )
13:08	PerlJam	std: sub foo ($alpha: $beta) { }
13:08	p6eval	std f43a358: OUTPUTÂ«Potential difficulties:â�¤ $beta is declared but not used at /tmp/sPeLWXdUQR line 1:â�¤------> sub foo ($alpha: â��$beta) { }â�¤ $alpha is declared but not used at /tmp/sPeLWXdUQR line 1:â�¤------> sub foo (â��$alpha: $beta) { }â�¤ok 00:00 4â€¦
13:12	Saravanan1	hi
13:13	tadzik	hello Saravanan1
13:14		JimmyZ joined #perl6
13:14	masak	Saravanan1: how's everything going?
13:15		atrodo joined #perl6
13:16	masak	PerlJam: I believe we have an RT ticket for invocant colons in sub declarations.
13:16	masak	hm, seems not.
13:17	* masak	submits rakudobug
13:17	PerlJam	std: sub foo ($alpha: $beta) { say "$alpha $beta" }
13:17	p6eval	std f43a358: OUTPUT«ok 00:00 43m␤»
13:17	PerlJam	It would be a stdbug too then
13:18		Saravanan joined #perl6
13:19	masak	rn: sub foo($alpha: $beta) { say "$alpha: $beta" }; foo 42, "OH HAI"
13:19	p6eval	niecza v19-15-g051783d: OUTPUT«42: OH HAI␤»
13:19	p6eval	..rakudo 038a3c: OUTPUT«Lexical 'self' not found␤ in sub foo at /tmp/PCatPFMUMH:1␤ in block at /tmp/PCatPFMUMH:1␤␤»
13:19	masak	actually, I'm not so sure supplying an invocant this way to a sub is an error.
13:19	PerlJam	yeah, I was thinking that too :)
13:19	masak	but Rakudo's behavior above surely is.
13:19	* masak	submits rakudobug
13:19	PerlJam	masak++
13:21		bluescreen10 joined #perl6
13:21	sirrobert	What should that do?
13:21	sirrobert	sub foo ($alpha: $beta)
13:22	masak	sirrobert: we don't know.
13:22	PerlJam	sirrobert: separate the args that participate in multi-dispatch from those that don't (maybe)
13:22	sirrobert	heh ok
13:22	masak	either error or it should be fine.
13:23	sirrobert	fair enough =)
13:23	sirrobert	hey, btw
13:23	masak	PerlJam: no, both the invocant and the rest of the arguments before ;; participate in MMD to exactly the same amount.
13:23	masak	s/amount/extent/
13:23	sirrobert	as I've been using (learning) p6, I'm trying to be helpful by blogging about it. What kind of stuff is helpful to write about?
13:24	PerlJam	It's too early for me to think clearly and I haven't had my caffiene yet.
13:24	masak	sittobert: to a first approximation, anything you feel is interesting.
13:24	sirrobert	I'll try to find my "voice" about it =)
13:24	PerlJam	sirrobert: Have you looked as "Using Perl 6"? Anything that fills the gaps from that book would be good to blog about.
13:24	masak	r: class A {}; class B is A {}; class C { multi method foo(B) { say "C" } }; class D is C { multi method foo(A) { say "D" } }; D.new.foo(B.new)
13:24	p6eval	rakudo 038a3c: OUTPUT«Ambiguous call to 'foo'; these signatures all match:␤:(C , B , Mu %_)␤:(D , A , Mu %_)␤␤ in method foo at src/gen/CORE.setting:334␤ in block at /tmp/i2yosE1L2F:1␤␤»
13:24	sirrobert	PerlJam: good idea =)
13:25	masak	PerlJam: the above conflict demonstrates this.
13:27	masak	here's a random suggestion for how .sort should sort by default:
13:28	masak	(a) if the values compared are equal by ===, sort them as equal
13:29	masak	(b) if the values are of the same type, sort them by cmp
13:29	masak	(c) if the values are of different types, sort the type name by leg
13:29	masak	or maybe this behavior should be folded into cmp itself.
13:30		skids joined #perl6
13:35	[Coke]	sort on type name seems odd, if you have a type hierarchy where you can still compare things (like in the numbers)
13:36	masak	yes, but (a) will catch that.
13:36	masak	rn: say 4 === 4.0
13:36	p6eval	rakudo 038a3c, niecza v19-15-g051783d: OUTPUT«False␤»
13:36	masak	hm :/
13:36	masak	am I thinking of some other equivalence, like eqv, perhaps?
13:37	masak	rn: say 4 eqv 4.0
13:37	p6eval	rakudo 038a3c: OUTPUT«False␤»
13:37	p6eval	..niecza v19-15-g051783d: OUTPUT«True␤»
13:37	masak	heh!
13:37	* masak	wants to submit a bug, but doesn't know on what
13:37	* masak	reads spec
13:38	masak	infix:<eqv>: S03:1303: "Compares two objects for canonical equivalence. For value types compares the values."
13:38	masak	well, it's the same value, no?
13:38	* masak	submits rakudobug
13:51		tokuhiro_ joined #perl6
13:54		thou joined #perl6
13:55	masak	rn: say 4e1 eqv 4; say 4e1 eqv 4.0
13:55	p6eval	rakudo 038a3c, niecza v19-15-g051783d: OUTPUT«False␤False␤»
13:55	masak	sorear: why does Niecza go True for 4 eqv 4.0 but not for those two?
13:56	masak	at least Rakudo is being consistent.
13:58		Chillance joined #perl6
14:06		crab2313 joined #perl6
14:11		atrodo joined #perl6
14:14		SamuraiJack joined #perl6
14:20	TimToady	it is a nieczabug, not a rakudobug, since === and eqv can only match if the types are the same
14:21	* masak	rejects the rakudobug
14:21	* masak	submits nieczue
14:22	masak	TimToady: any thoughts on subs with an invocant parameter?
14:22	TimToady	not before coffee
14:23	masak	;)
14:23	TimToady	esp since I only got 5 hours sleep
14:23	masak	:/
14:23	tadzik	one does not cast spells without mana potions
14:23	TimToady	because I was hacking on STD :)
14:23	masak	heh :)
14:23		kaare_ joined #perl6
14:24	masak	TimToady: when I saw your "I can't commit this thing before I've committed these things", I wanted to tell you about git-stash
14:24	* TimToady	is close (cross fingers) to having the sigspace stuff working again
14:24		hoelzro joined #perl6
14:24		telex joined #perl6
14:25	sirrobert	git stash is super nice =)
14:25	sirrobert	love that it's a stack
14:26	TimToady	yes, that looks nice
14:32	masak	'git stash [save]' and 'git stash apply' are complementary actions, and the two I use the most.
14:33	masak	I love how it's using commits to save away my changes.
14:33	masak	(without hooking those commits onto any branch)
14:33	PerlJam	how long do your stashes stay around?
14:34	PerlJam	(I use stash quite a bit too but I find that sometimes I forget about them)
14:34	masak	that's a really good quesion.
14:34	sirrobert	I think they stay around indefinitely (until you clear them)
14:34	PerlJam	whereas if I put my changes in a branch, they are less likely to be forgotten
14:34	masak	somewhere between 10 seconds and 10 kiloseconds, most of the time.
14:34	hoelzro	PerlJam: that's my issue
14:35	masak	yeah, branches are nice, too. they're the "next step" after stashes.
14:35	hoelzro	I often forget about them =/
14:35	sirrobert	I use stash as something like a clipboard
14:35	sirrobert	I only put stuff in there I'm ok to lose, or that I want to copy from one branch to another.
14:35	masak	and I keep forgetting and re-remembering that branches are very throwaway, too, and can be created en masse for various experiments.
14:36	masak	yeah, stashing is like an anonymous mini-branch of sorts.
14:36	PerlJam	sirrobert: If you're copying from one branch to another, then surely you don't need the stash?
14:36	sirrobert	well, I don't mean literally copying, more like "sticking them in there to see if there are applicable bits"
14:37	sirrobert	I often have a stash that's several layers deep and pop to the one I want
14:37	PerlJam	ah
14:37	sirrobert	kinda like an undo/redo =)
14:38	lumi_	r: say 4e1
14:38	p6eval	rakudo 038a3c: OUTPUT«40␤»
14:38	lumi_	masak: That's why the first one fails
14:39	PerlJam	lumi++ (math skills ;)
14:39	sirrobert	heh
14:40	masak	I fail math today.
14:40	* masak	rises from the ashes and moves on ;)
14:40	masak	rn: say 4e0 eqv 4; say 4e0 eqv 4.0
14:40	p6eval	niecza v19-15-g051783d: OUTPUT«True␤True␤»
14:40	p6eval	..rakudo 038a3c: OUTPUT«False␤False␤»
14:41	masak	yeah, niecza is consistently wrong on those. lumi_++
14:41	TimToady	now if only git stash would save and restore all the vim sessions I have up in that directory, it might be useful to me
14:44	timotimo	surely you could write a perl one-liner that does this ... :)
14:44	TimToady	debugging STD/viv when you have no valid CORE setting and compiling CORE.setting fails in the first whitespace is...not trivial
14:44	timotimo	hm. there's python and ruby for vim, but is there also perl for vim? i wonder ...
14:45	hoelzro	timotimo: sure there is
14:45	hoelzro	if by "perl for vim", you mean that Vim can be built with an embedded Perl
14:45	TimToady	I went to bed last night when I managed to get CORE to parse up to line 8, so at least it's parsing some sigspace now
14:55	dalek	std: 66387b5 \| larry++ \| STD.pm6:
14:55	dalek	std: restrict do {} obs message to while/until
14:55	dalek	std: review: https://github.com/perl6/std/commit/66387b5246
14:57	TimToady	GlitchMr++ for that, btw
15:01	* jnthn	back
15:02	JimmyZ	welcome back
15:03	masak	good ugt morning, jnthn!
15:03	jnthn	TimToady: Heh, sounds like the fun I had getting CORE.setting parsing again when putting QRegex in place...
15:03	jnthn	TimToady: I'd say "have fun", but it was more "argh why" :)
15:03	masak	that's because you forgot to have fun
15:05	jnthn	masak: It's kind of a yak-shavey task :)
15:05	masak	yeah. I was kidding ;)
15:05		Chillance joined #perl6
15:05	jnthn	re 114438 I suspect invocants should be forbidden on subs, but I didn't backlog to see if any other discussion happened with regard to it
15:06	masak	mad props to both jnthn and TimToady, of course.
15:06	jnthn	ah, I didn't sleep well enough to do humor :P
15:06	jnthn	Well, I can probably pun still... :)
15:06	masak	jnthn: I started out supposing that it should forbid it too. it's certainly the conservative thing to do.
15:06	masak	if it works, then it should work like in Niecza.
15:07	jnthn	I know STD has some invocant_ok thingy. I think I didn't steal it into Rakudo's grammar.
15:07	jnthn	It may be related to taht.
15:07	masak	n: sub foo($alpha: $beta) { say "$alpha: $beta" }; foo 42, "OH HAI"
15:07	p6eval	niecza v19-15-g051783d: OUTPUT«42: OH HAI␤»
15:07	masak	jnthn: well, std allows it too!
15:07	masak	std: sub foo($alpha: $beta) { say "$alpha: $beta" }; foo 42, "OH HAI"
15:07	p6eval	std f43a358: OUTPUT«ok 00:00 44m␤»
15:08	jnthn	bah :)
15:14	tadzik	"We have an implementation of Perl 6 that's gonna start to be worth using in the next year or so" -- Damian on http://radar.oreilly.com/2012/[…]-conway-perl.html
15:14	tadzik	nice interview it is
15:15		obra joined #perl6
15:15	masak	tadzik: Damian Conway, is that the guy who said "the next year or so" back in 2001? :P
15:15		clkao joined #perl6
15:15	masak	s/said/wrote/
15:17	PerlJam	I'm not sure what to make of Damian anymore ... It's like he's disconnected from the community, but every once in a while pops up to say "I'm still here, look at this awesome thing"
15:17	PerlJam	(at least I don't quite see how he's connected like he used to be)
15:18	masak	I've had that feeling too. but that seems to be his MO.
15:18	masak	at least when he's involved, he is involved.
15:19	masak	I would prefer if he would pop up on the channel occasionally, but he doesn't seem to be a pop-up-on-the-channel kind of guy.
15:19	GlitchMr	back
15:19	GlitchMr	TimToady: I was just testing if https://github.com/perl6/std/issues/1 was fixed
15:20	GlitchMr	I will do moritz++ instead
15:23	TimToady	s/instead/in addition/ :)
15:23	TimToady	this isn't a zero-sum game...
15:24	diakopter	o_O TimToady before 9 a.m.?
15:25	* diakopter	apparently needs coffee too
15:26	masak	diakopter: we're all on UGT now :P
15:26	masak	I like http://blog.stackoverflow.com/[…]ess-on-a-qa-site/ not just because of the excellent geek troll picture. it makes good points.
15:26	TimToady	* can mean UGT too
15:28		MayDaniel joined #perl6
15:28		MayDaniel joined #perl6
15:30		alester joined #perl6
15:35		thelazydeveloper joined #perl6
15:42	diakopter	oh noes. no one show TimToady codegolf.stackexchange.com
15:42	TimToady	n: say +val '0x55'
15:42	p6eval	niecza v19-15-g051783d: OUTPUT«85␤»
15:42	TimToady	masak: when you use the string version of val, you just get the string back
15:42	diakopter	it's like rosettacode except crazy
15:44	TimToady	n: say <0x55>
15:44	p6eval	niecza v19-15-g051783d: OUTPUT«85␤»
15:44	TimToady	note that <> is specced to force numeric when there's just that one value
15:44	TimToady	but that's part of <>, not val
15:44	masak	ah.
15:45	TimToady	n: say +'0x55'
15:45	p6eval	niecza v19-15-g051783d: OUTPUT«85␤»
15:45	TimToady	and you don't need the val if you're forcing numeric anyway
15:47	TimToady	so the RC entry is wrongish, I guess
15:47	masak	aye.
15:48	masak	val
15:48	diakopter	I forget - will P6 have unsanitary macros?
15:48	masak	val, lightweight as it is, is too heavyweight there.
15:48	masak	diakopter: Pugs has them.
15:48	masak	pugs: macro foo { 'say $a' }; my $a = 42; foo
15:48	p6eval	pugs: OUTPUT«42␤»
15:49	diakopter	is that spec'd?
15:49	TimToady	and there's no reason to limit the RC entry to integers
15:49	masak	diakopter: yes.
15:49	TimToady	n: say 3.14159625.base(2)
15:49	p6eval	niecza v19-15-g051783d: OUTPUT«Unhandled exception: Unable to resolve method base in type Rat␤ at /tmp/UQAKN6XJRW line 1 (mainline @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3929 (ANON @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3930 (module-CORE @ 564) ␤ at /home/p…
15:49	TimToady	r: say 3.14159625.base(2)
15:49	p6eval	rakudo 038a3c: OUTPUT«11.001001000011111110101␤»
15:49		jeffreykegler joined #perl6
15:49	TimToady	nieczabug
15:49	* masak	submits
15:52	diakopter	that oreilly link - "the fastest dynamic language out there"?
15:53	PerlJam	diakopter: yeah, I caught that too. I would have agreed with him a few years ago, but not today.
15:56	masak	which language? Dart? :P
15:56	diakopter	Gabor Szabo needs an editor for his perl6maven posts; all of them have English errors
15:56	masak	well volunteered.
15:56	diakopter	:D
15:58	TimToady	http://rosettacode.org/wiki/No[…]es/Convert#Perl_6 fixed, GlitchMr++ masak++
15:58	diakopter	rn: my $a =~ /a/; say $a
15:58	p6eval	niecza v19-15-g051783d: OUTPUT«Regex()<instance>␤»
15:58	p6eval	..rakudo 038a3c: OUTPUT«␤»
15:59	diakopter	std: my $a =~ /a/; say $a
15:59	p6eval	std f43a358: OUTPUT«Can't bless non-reference value at CursorBase.pm line 1163.␤FAILED 00:00 42m␤»
15:59	diakopter	I don't know which one's righter
15:59	diakopter	niecza I guess, but isn't it supposed to warn on using =~ ?
16:00	masak	'FAILED'?
16:00	diakopter	crashed I guess
16:00	TimToady	std: $_ =~ /a/
16:00	p6eval	std f43a358: OUTPUTÂ«===[0mSORRY!===[0mâ�¤Unsupported use of =~ to do pattern matching; in Perl 6 please use ~~ at /tmp/xbebQGchbK line 1:â�¤------> $_ =~â�� /a/â�¤Parse failedâ�¤FAILED 00:00 41mâ�¤Â»
16:01	TimToady	diakopter: the my is messing you up there
16:01	geekosaur	interesting parser fail in std with the my version
16:01	TimToady	parsed as an initializer
16:01	TimToady	my $a = (~ /a/)
16:01	diakopter	std: my $a = ~ /a/; say $a
16:01	p6eval	std f43a358: OUTPUT«Can't bless non-reference value at CursorBase.pm line 1163.␤FAILED 00:00 42m␤»
16:01	TimToady	but =~ on a my is non-sensical in any case, so I'm not concerned
16:02	diakopter	did someone break p6eval's std?
16:02	TimToady	it's a month old
16:02	TimToady	and the current version certainly won't bootstrap
16:03	TimToady	gimme another few hours (cross fingers) to fix that
16:03	diakopter	git wishlist: make a commit of only a portion of the diff of a file
16:04	masak	diakopter: git-add -i
16:06	diakopter	oh!
16:06	diakopter	seems TortoiseGit doesn't have that feature
16:07		spider-mario joined #perl6
16:08	diakopter	oh wait it does
16:08	TimToady	maybe it has git stash save --patch
16:12	jnthn	diakopter: Also "git gui" brings up a GUI tool that can do that
16:16	[Coke]	masak: I had to walk away earlier, but how would you compare 4 and 4.1 then? as "Integer" vs "Number" ?
16:17	masak	yeah, that won't fly.
16:17	masak	never mind.
16:20		adu joined #perl6
16:38		[hds] joined #perl6
16:41	TimToady	I think subs with invocant parameters are benign, and : just means the same as comma there
16:41		am0c joined #perl6
16:42	TimToady	in some sense, exported methods are the same thing
16:43	masak	let me play devil's advocate, then.
16:43	masak	if we disallow them now but realize they're useful later, we will have an easier time than if we allow them now but realize they're harmful later.
16:44	masak	I see how they are "the same thing" as exported methods, but what are they chances they are not an indication of programmer confusion?
16:44	masak	i.e. what are the chances a programmer would be annoyed rather than helped by us flagging it?
16:45	masak	it might as well be a typo for 'method' as a typo for ','
16:45	TimToady	what are the chances that someone is attempting to communicate that this might end up as a method somehow as well?
16:45	jnthn	Should it have a slef?
16:45	jnthn	*self?
16:45	masak	definitely not.
16:45	masak	well, unless the sub is inside of a method :P
16:46	jnthn	:P
16:48	TimToady	"sub with invocant better written as 'my method'" maybe
16:52		sftp joined #perl6
16:53	* masak	rather likes that
16:53	masak	or 'our method'.
16:53	TimToady	nr: my method foo ($foo: $bar) { say "$foo $bar" }; foo(1,2)
16:54	p6eval	rakudo 038a3c: OUTPUT«1 2␤»
16:54	p6eval	..niecza v19-15-g051783d: OUTPUTÂ«===[0mSORRY!===[0mâ�¤â�¤Methods must be used in some kind of package at /tmp/yi8quc9FUC line 1:â�¤------> my method foo â��($foo: $bar) { say "$foo $bar" }; foo(1,â�¤â�¤Unhandled exception: Check failedâ�¤â�¤ at /home/p6eval/niecza/boot/libâ€¦
16:54	TimToady	nieczanyi I guess
16:54	masak	yeah.
16:54	masak	someone else is welcome to submit it. :)
16:58		kaleem joined #perl6
16:59	jnthn	r: class A { }; say A.new
16:59	p6eval	rakudo 038a3c: OUTPUT«A.new()␤»
16:59	jnthn	r: class A { }; say ~A.new
16:59	p6eval	rakudo 038a3c: OUTPUT«A<853331884>␤»
17:00	TimToady	ô.ó
17:01	jnthn	Matches spec, afaik
17:01	jnthn	I do now have the slight issue of what to do in NQP now that we have .gist and .Str doing different things in a bunch of cases.
17:05		cognominal joined #perl6
17:07	diakopter	nr: my method foo ($foo: $bar) { say self~" $bar" }; foo(1,2)
17:07	p6eval	niecza v19-15-g051783d: OUTPUTÂ«===[0mSORRY!===[0mâ�¤â�¤Methods must be used in some kind of package at /tmp/Iksn61Nzeo line 1:â�¤------> my method foo â��($foo: $bar) { say self~" $bar" }; foo(1â�¤â�¤Potential difficulties:â�¤ $foo is declared but not used at /tmp/Iksnâ€¦
17:07	p6eval	..rakudo 038a3c: OUTPUT«1 2␤»
17:08	diakopter	r: my method foo ($foo:) { say self === $foo }; foo("b")
17:08	p6eval	rakudo 038a3c: OUTPUT«True␤»
17:09	diakopter	r: class F { my method foo ($foo:) { say self === $foo }; }; F.foo()
17:09	p6eval	rakudo 038a3c: OUTPUT«True␤»
17:10		raiph joined #perl6
17:11	dalek	nqp/toqast: 319bb8d \| jnthn++ \| src/QAST/Compiler.nqp:
17:11	dalek	nqp/toqast: Implement default parameter handling, winning the optional and named argument tests back.
17:11	dalek	nqp/toqast: review: https://github.com/perl6/nqp/commit/319bb8d173
17:11	dalek	nqp/toqast: 4b85a2e \| jnthn++ \| src/NQPQ/ (2 files):
17:11	dalek	nqp/toqast: Get + and ~ prefixes workingish again.
17:11	dalek	nqp/toqast: review: https://github.com/perl6/nqp/commit/4b85a2e2b5
17:11	dalek	nqp/toqast: 67a605c \| jnthn++ \| src/NQPQ/ (2 files):
17:11	dalek	nqp/toqast: Compile =:= again.
17:11	dalek	nqp/toqast: review: https://github.com/perl6/nqp/commit/67a605ca2a
17:17	sorear	good * #perl6
17:19	masak	sorear! \o/
17:19	masak	sorear: I discovered some bugs in Niecza's eqv in the backlog.
17:19	masak	also filed some other things and didn't file some other things.
17:26	* [Coke]	files a bug on masak's bug filer.
17:26	[Coke]	"seems spotty"
17:27	masak	I wouldn't mind if someone took over all the nieczue filing, to be honest.
17:27	masak	I still feel like my primary responsibility is to keep RT's perl6 queue up-to-date.
17:31	diakopter	I think we can have a bot that takes this command: bugfiler: niecza 8 Title of Bug Report where 8 means number of prior irc lines to paste into the bug report
17:31	diakopter	and it would only accept commands on #perl6 so people couldn't spam it in privmsg
17:32	diakopter	seems the nick isn't taken
17:32	masak	well, there's usually some editing to filing a bug report.
17:35		benabik joined #perl6
17:36		quietfanatic joined #perl6
17:37	diakopter	quietfanatic: howdy
17:40	quietfanatic	hey diakopter
17:42		benabik_ joined #perl6
17:45	dalek	nqp/toqast: 00c649e \| jnthn++ \| src/NQP/Actions.pm:
17:45	dalek	nqp/toqast: Fix bug in v-table method installation.
17:45	dalek	nqp/toqast: review: https://github.com/perl6/nqp/commit/00c649e8d9
17:45	dalek	nqp/toqast: 507d55c \| jnthn++ \| src/NQPQ/Actions.pm:
17:45	dalek	nqp/toqast: Port previous fix to NQPQ.
17:45	dalek	nqp/toqast: review: https://github.com/perl6/nqp/commit/507d55cc5c
17:45	dalek	nqp/toqast: 93a716f \| jnthn++ \| src/core/NQPMu.pm:
17:45	dalek	nqp/toqast: Add some missing methods to NQPMu.
17:45	dalek	nqp/toqast: review: https://github.com/perl6/nqp/commit/93a716fd59
17:45	dalek	nqp/toqast: ce81241 \| jnthn++ \| src/QAST/Operations.nqp:
17:45	dalek	nqp/toqast: Add a QAST op for throwing control exceptions.
17:45	dalek	nqp/toqast: review: https://github.com/perl6/nqp/commit/ce81241a2a
17:45	dalek	nqp/toqast: 26b5b5b \| jnthn++ \| src/NQPQ/Actions.pm:
17:45	dalek	nqp/toqast: Get next/last/redo working again.
17:45	dalek	nqp/toqast: review: https://github.com/perl6/nqp/commit/26b5b5b6d9
17:49	jnthn	50 out of 60 tests in t/nqp now passing in nqp/toqast :)
17:51		Targen joined #perl6
17:58	quietfanatic	am I the only one who thinks arbitrary radix notation looks way too much like pair notation?
17:59	PerlJam	"too much"?
17:59	quietfanatic	:16<abcd> ought to make 16 => 'abcd'
17:59	* Tene	also
18:00	masak	quietfanatic: it is pair notation.
18:00	masak	it's just been hijacked for radix conversion.
18:01		bluescreen10 joined #perl6
18:01	quietfanatic	right
18:02		jeffreykegler joined #perl6
18:02	PerlJam	quietfanatic: so you think it should look different?
18:02	quietfanatic	I do think it should look different.
18:03	quietfanatic	mind you, I don't have any immediate suggestions s to how it should look :)
18:04	PerlJam	If there were a more procedural interface, would that assuage your concerns?
18:04	quietfanatic	probably
18:04	TimToady	normal colon pairs can only have identifiers on the left, so the numeric was completely unused
18:05	TimToady	and there's little use case for 16 => 'abcd'
18:05	quietfanatic	Arbitrary bases are not something people use enough that they need a syntactic shortcut for it.
18:05	TimToady	sez who? :)
18:05	PerlJam	quietfanatic: ah, so you think they are incorrectly huffmanized?
18:05	quietfanatic	Especially when that syntactic shortcut means they need to do an eval-like thing to parameterize the base
18:06	quietfanatic	sez me :)
18:06	quietfanatic	PerlJam: It's more that different things should look different
18:07	TimToady	parameterizing the base is something that is used even less, so +":$base\<$str>" is fine
18:10	TimToady	there's also something to be said for not inventing additional syntax
18:10	PerlJam	All I know is that bases 2, 8, 10, and 16 come up quite a bit in computer programs :)
18:10	TimToady	which is why those have special cases
18:11	GlitchMr	Perl 6 is already more popular than certain languages at GitHub :).
18:11	GlitchMr	https://github.com/languages/AppleScript
18:11		SamuraiJack_ joined #perl6
18:11	quietfanatic	reusing already existing syntax for something entirely unrelated is just as bad in my opinion as inventing new syntax
18:11	TimToady	using colon notation also gives you the () and [] variants for "free"
18:12	GlitchMr	In PHP there is base_convert() function. The name is long, but you rarely want to do it anyways.
18:12	quietfanatic	Something like that but with a better name would be best, I think.
18:12	PerlJam	quietfanatic: we do it all the time with contextual clues to help. Notice how many ways {} can be used.
18:13	PerlJam	quietfanatic: :NUMBER seems like a really good contextual clue to me.
18:13	TimToady	I have no trouble telling :foo apart from :16
18:13	GlitchMr	I wouldn't have problem with something like base-convert function to do it
18:14	quietfanatic	At a close look, you have no trouble, of course
18:14	PerlJam	GlitchMr: That's kind of where I was going initially. If there were some subroutine to do base conversion (in addition to the current notation), maybe that would help
18:14	quietfanatic	suppose you're skimming over code. You'll see what looks like a pair, but you'd have to examine the individul characters to see that it is in fact not a pair
18:14	TimToady	you always have to look at individual characters
18:15	quietfanatic	But : and < > are immediately distinctive.
18:15	GlitchMr	I mean, $num.base-convert($base => 10) looks better than +":$base\<$num>"
18:15	quietfanatic	whereas 16 and IG are not immediately distinguishable.
18:15	TimToady	so are 2 and x
18:15	PerlJam	quietfanatic: sounds like you need a better font :)
18:16	quietfanatic	.oO(The tyrrany of selective examples)
18:16	quietfanatic	Oh, I can tell them apart if I actually look at them
18:16	* PerlJam	boggles
18:17	GlitchMr	If I would see +":$base\<$num>" I would think it's... hacky method of converting bases
18:17	quietfanatic	My thesis here is that you shouldn't have to read every single character to get a gist of the structure of a piece of code.
18:17	quietfanatic	And yes, { } used for both hashes and code bothers me that way too.
18:17	TimToady	this seems to come down to almost the same argument as "I want to be able to read it without learning the language"
18:18	GlitchMr	What's wrong with { } being used for both functions and hashes?
18:18	TimToady	and I'd submit that someone who doesn't know Perl 6 would immediately figure out what :16<deadbeef> means
18:18	quietfanatic	That's because deadbeef is a culturally unambiguous hexadecimal number.
18:18	GlitchMr	:16<deadbeef> could be written as 0xdeadbeef
18:19	quietfanatic	GlitchMr: You have to use contextual cues to figure out which is which, just like the compiler
18:19	TimToady	the other thing is that if you're going to be using a weird base, you're likely to use it many times in teh same program, so the learning is amortized
18:19	GlitchMr	Well, .map will definitely won't take hash as function :P.
18:19	quietfanatic	TimToady: Oh that's a valid point
18:19	PerlJam	quietfanatic: "contextual cues" is almost the basis of all things Perl.
18:20	[Coke]	"GRAMMAR"
18:20	benabik	r: say <1 2 3>.map { a => 12 }
18:20	p6eval	rakudo 038a3c: OUTPUT«===SORRY!===␤Confused␤at /tmp/KGZaxPsRn6:1␤»
18:20	GlitchMr	We have binary, octal, hexadecimal, decimal... what other weird base you would want to use?
18:20	TimToady	GlitchMr: I certainly would not be opposed to a named function in place of the hack
18:20	benabik	Base64?
18:20	PerlJam	GlitchMr: base 64 seems to come up a lot
18:20	quietfanatic	The thing is, I skim code almost as much as I carefully read it. Actually, more.
18:21	quietfanatic	and so I like easily-skimmable code quite a bit.
18:21	GlitchMr	:64<1> doesn't work and shouldn't
18:21	PerlJam	GlitchMr: base 36 is occasionally useful too
18:21	TimToady	nr: say :64[1,1]
18:21	p6eval	niecza v19-15-g051783d: OUTPUT«Unhandled exception: :16[...] syntax NYI␤ at /home/p6eval/niecza/lib/CORE.setting line 1414 (die @ 5) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3550 (unbase @ 8) ␤ at /tmp/3VeZbgRWB8 line 1 (mainline @ 3) ␤ at /home/p6eval/niecza/lib/CORE.setting …
18:21	p6eval	..rakudo 038a3c: OUTPUT«===SORRY!===␤Malformed radix number␤at /tmp/5s5MwphmF6:1␤»
18:21	benabik	Although there is a difference between Base64 and base 64. :-/
18:21	GlitchMr	I would like to see use for base 36... but I also would like to see use for octal numbers :)
18:21	TimToady	octals make sense on a PDP11 :)
18:22	GlitchMr	Oh, I think I have one use for octals - chmod
18:22	PerlJam	GlitchMr: when talking to other weird systems, it helps if you speak their language :)
18:22		fhelmberger joined #perl6
18:22		quietfanatic_ joined #perl6
18:22	TimToady	base 4 makes some sense in biology :)
18:22	GlitchMr	But, I can see base 4 being useful
18:22	GlitchMr	TimToady said it before me :P
18:23	cosimo	so...
18:23	cosimo	if I had to propose a redesign of perl6.org, what concepts we want the site/design to convey?
18:23	GlitchMr	I like current Perl 6 site :)
18:23	PerlJam	base 6 might make lots of sense if you're dealing with molecules that have lots of carbon :)
18:24	quietfanatic_	Non-conformity.
18:24	quietfanatic_	(cosimo:)
18:24	GlitchMr	It reminds me WHATWG website for some reason, but it's nice nevertheless
18:24	TimToady	friendly to seven-year-olds
18:24	TimToady	fun
18:24	PerlJam	cosimo: friendly, fun, capable, serious, powerful, awesome, cool, nice, happy, ...
18:25	benabik	Wikipedia says base 15 is useful for telephony routing over IP
18:25	TimToady	btw, the .org doesn't have to do everything; there are plenty of other tlds
18:25	GlitchMr	Could you modify grammar to add 0p if you need it?
18:25	TimToady	so the .org should be fun
18:25	GlitchMr	p being for pentadecimal
18:26	quietfanatic_	I bet you could
18:26	benabik	postfix:<p> ?
18:26	TimToady	GlitchMr: it's possible, but we don't go out of our way to make it easy
18:26	benabik	-Ofun
18:26	benabik	And there do seem to be a good number of uses for base 36.
18:26	quietfanatic_	TimToady: I thought the Perl 6 grammar was supposed to be extensible at the drop of a hat
18:26	GlitchMr	postfix:<p> wouldn't work when your numbers include characters between A and E
18:27	quietfanatic_	TimToady: Or have we decided to go the Apple route? :)
18:27	GlitchMr	0t1201202201
18:27	PerlJam	base 60 has lots of uses too
18:28	GlitchMr	https://en.wikipedia.org/wiki/Ternary_computer
18:29	Tene	ternary would also be useful for malbolge interop
18:29	TimToady	r: say :60[12,34,56]
18:29	p6eval	rakudo 038a3c: OUTPUT«===SORRY!===␤Malformed radix number␤at /tmp/WdOzyP90D3:1␤»
18:29	TimToady	nyi I guess
18:29	PerlJam	rakudo doesn't grok bases above 36 for some reason
18:29	TimToady	I thought it did
18:30	benabik	36 is the highest unambiguous base. 0-9a-z.
18:30		jeffreykegler joined #perl6
18:30	TimToady	see S02:3247
18:31	GlitchMr	Bases above 36 aren't useful anyways until :X[] notation will be implemented
18:31	benabik	r: say :10[1,2,3,4]
18:31	p6eval	rakudo 038a3c: OUTPUT«===SORRY!===␤Malformed radix number␤at /tmp/vSiY9iJBSR:1␤»
18:31	TimToady	b: :60[12,34,56]
18:31	p6eval	b 922500: OUTPUT«===SORRY!===␤Malformed radix number at line 22, near "[12,34,56]"␤»
18:32	TimToady	ng: :60[12,34,56]
18:32	TimToady	alpha: :60[12,34,56]
18:32	* TimToady	forgets the names
18:32	benabik	p6eval: help
18:32	p6eval	benabik: Usage: <(star\|pugs\|nqp\|std\|niecza\|rakudo\|b\|toqast\|nom\|npr\|n\|r\|perl6\|prn\|rn\|p\|rnp\|nrp\|pnr\|rpn\|t\|p6\|nr)(?^::\s) $perl6_program>
18:32	TimToady	pugs: say :60[12,34,56]
18:32	p6eval	pugs: OUTPUT«45296␤»
18:32	PerlJam	pugs++
18:32	TimToady	maybe that's what I'm remembering
18:32	benabik	perlcabal's spec appears to be out of date.
18:33	GlitchMr	It's always out of date ;)
18:34	benabik	Well, that's what the irc logs point to, so it would be nice if it wasn't.
18:34	GlitchMr	Even when you think it isn't, something is out of date.
18:36	cosimo	is it possible to run stuff server side for perl6.org, or it's a completely static site?
18:36	GlitchMr	It's static site
18:37	PerlJam	std: say :60[12,34,56]
18:37	p6eval	std f43a358: OUTPUT«ok 00:00 42m␤»
18:39	diakopter	how does one regex match for graphs that could have negative codepoints?
18:40	diakopter	TimToady: ^^ re NFG
18:41		haidarian joined #perl6
18:41	quietfanatic	You can't know the codepoints of precomposed graphemes, can you?
18:41	diakopter	not under the proposed implementation
18:42	[Coke]	r: my $a = "This is a string"; say $a.chars
18:42	p6eval	rakudo 038a3c: OUTPUT«16␤»
18:42	diakopter	(and not under any implementation I can imagine)
18:42	[Coke]	wow. I counted that wrong at least 3 times. ;)
18:42	quietfanatic	I'd imagine that in the regex you have to assemble the equivalent grapheme out of its individual codepoints.
18:43	diakopter	quietfanatic: yeah, but is there a way to say "this base codepoint, with optionally whatever combining codepoints"
18:44	quietfanatic	[ :ii $base ], perhaps?
18:44	diakopter	(or optionally particular combining codepoints)
18:44	quietfanatic	well,that's probably more difficult
18:45	quietfanatic	If you are matching on the grapheme level it's probably difficult to selectively pull the graphemes apart into codepoints when you please.
18:46	diakopter	well, I'm just imagining it will be impossible to match against NFG strings that contain negative codepoints (except for . obviously), because the codepoints don't fall into any cclass and can't be matched directly
18:47	benabik	negative codepoints?
18:47	quietfanatic	The negative codepoints stand for a sequence of positive codepoints though.
18:47	diakopter	are you suggesting the regex engine should decompose them while matching?
18:47	quietfanatic	If it has to.
18:48	quietfanatic	Negative codepoints is just an implementation detail. An optimization.
18:48	quietfanatic	as far as I understand it, at least.
18:48	diakopter	eh; I'm fairly certain TimToady says it's an essential element of NFG
18:49	quietfanatic	Well, that's how you'd get a semantically variable-width encoding into an actually fixed-width encoding.
18:49	diakopter	right, but also how to store NFG strings at runtime
18:49	quietfanatic	right.
18:49	diakopter	(all strings)
18:50	quietfanatic	Somewhere hidden in the program is a table which maps negative indexes to strings representing one grapheme.
18:50	quietfanatic	Like the symbol table in Lisp dialects.
18:50		sftp joined #perl6
18:50	diakopter	right...
18:51	jnthn	I would rather expect that the negative codepoints have a bunch of unicode properties calculated for them.
18:51	quietfanatic	jnthn: Oh, that's possible too
18:51		PerlJam joined #perl6
18:52	diakopter	but what would determine whether they have those properties?
18:52	quietfanatic	whether the positive codepoints they stand for have those properties
18:52	jnthn	Giving it the properties of the base character is probably a good start.
18:52	diakopter	whether all the codepoints in the graph have the property or just one of them?
18:52	diakopter	ok
18:53	quietfanatic	Or is unicode not set up to assign codepoints to composed graphemes? I don't know actually
18:53	quietfanatic	ah
18:53	jnthn	The answer may also vary with property. I suspect there must be a sane way to do this.
18:53	quietfanatic	Excuse me
18:53	quietfanatic	to assign properties to composed grapheme
18:53	quietfanatic	s
18:53	* jnthn	guessed that's what was meant
18:53	* quietfanatic	made a wordo
18:54	diakopter	I don't think that's a word
18:54	quietfanatic	It's like a typo
18:54	diakopter	it was a joke..
18:55	masak	I don't think that's a joke :P
18:55	diakopter	:D
18:57	TimToady	all NFG codepoints (both negative and precomposed NFC) should have easily accessible NFD for any kind of partial grapheme matching or property
18:58	TimToady	so looking up property of the base char should generally not be difficult
18:58	diakopter	ok, that answers the question of whether regex matches should match against just the base character
18:59	jnthn	Well, in the context of properties anyway, I guess.
18:59	diakopter	including cclass?
18:59	jnthn	yes
18:59	jnthn	Those boil down to unicode props
18:59	diakopter	ok
19:01	TimToady	we will need to invent ways of doing partial matches that involve marks
19:01	TimToady	but I think that will become more obvious once we get NFG in place
19:01	diakopter	ah, that was the other question, which I had forgotten I asked
19:04	dalek	nqp/toqast: 476a500 \| jnthn++ \| src/guts/multi_dispatch.c:
19:04	dalek	nqp/toqast: Don't look into freed memory.
19:04	dalek	nqp/toqast: review: https://github.com/perl6/nqp/commit/476a500ac2
19:04	dalek	nqp/toqast: 9341160 \| jnthn++ \| src/NQPQ/ (3 files):
19:04	dalek	nqp/toqast: Partially fix multi-method dispatch.
19:04	dalek	nqp/toqast: review: https://github.com/perl6/nqp/commit/9341160d03
19:04	dalek	nqp/toqast: 4e3159b \| jnthn++ \| src/QAST/Operations.nqp:
19:04	dalek	nqp/toqast: Fix code-gen bug.
19:04	dalek	nqp/toqast: review: https://github.com/perl6/nqp/commit/4e3159b19e
19:05	TimToady	or course, one can always call out to a method in a grammar, but it's not clear how that helps normal regexes, unless we do it with <:&mymatch(...)> or some such
19:09		jeffreykegler joined #perl6
19:25	japhb	r: say +':60[12,34,56]'
19:25	p6eval	rakudo 038a3c: OUTPUT«45296␤»
19:25	dalek	nqp/toqast: 9ccb91c \| jnthn++ \| src/QAST/Operations.nqp:
19:25	dalek	nqp/toqast: Fix lexotic compilation bug.
19:25	dalek	nqp/toqast: review: https://github.com/perl6/nqp/commit/9ccb91c616
19:25	dalek	nqp/toqast: 4780b8e \| jnthn++ \| src/NQPQ/ (2 files):
19:25	dalek	nqp/toqast: Fix return; this also brings it inline with the mechanism we use to do it in Rakudo.
19:25	dalek	nqp/toqast: review: https://github.com/perl6/nqp/commit/4780b8e99c
19:25	japhb	TimToady, Str.Numeric handles :60[,,] format. The main compiler does not.
19:26	masak	feels like that should be a single code path :)
19:26	japhb	masak, Str.Numeric would ideally be implemented by just using that piece of the full grammar. Unfortunately:
19:27	japhb	A) When I started working on Str.Numeric, I don't think that was actually workable. Dunno if it is now.
19:27	japhb	B) The general grammar is (was?) way slower than the hand-coded Str.Numeric parser.
19:27	japhb	Hence, not one code path yet. :-)
19:29	japhb	Str.Numeric can handle some pretty strange stuff these days. Unfortunately, not Complex numbers yet. I should probably just do that when I continue work on val().
19:32	masak	++japhb
19:35		bluescreen10 joined #perl6
19:47	* [Coke]	wonders if nieczue rhymes with "miscue"
19:49		brrt joined #perl6
19:55		[particle] joined #perl6
19:59		jeffreykegler joined #perl6
20:01		fhelmberger joined #perl6
20:06		birdwindupbird joined #perl6
20:06	sirrobert	proving tests on JSON::Tiny seems to randomly skip/fail tests...
20:07	sirrobert	just ran "prove -re perl6 t/01-parse.t" three times in a row
20:08	sirrobert	twice they all passed, the third time "No subtests run"
20:08	sirrobert	sometimes a portion pass and a portion fail
20:08	sirrobert	is that a known phenomenon others have encountered?
20:10		fhelmberger joined #perl6
20:11	masak	sirrobert: no, never heard anything like that before.
20:11	sirrobert	hm
20:11	masak	[Coke]: no, it rhymes with "betcha" :)
20:12	sorear	sirrobert: is it still random if you disable address space layout randomization?
20:12	sirrobert	tell me how to do that and I'll check =)
20:12	masak	[Coke]: or, hm, maybe not... :)
20:13	sorear	sirrobert: are you on linux?
20:13	sirrobert	yep
20:14	sorear	echo 0 \| sudo tee /proc/sys/kernel/randomize_va_space
20:16	sirrobert	sorear: running prove several times now, will tell you in a sec
20:17	sirrobert	six consecutive runs of prove with no issue.
20:17	sirrobert	(previously I got at most 2 in a row)
20:18	sirrobert	you want me to report that somewhere?
20:18	sirrobert	woops, it happened again on the 11th run
20:19	sirrobert	but that's a lot MORE rare, at least =)
20:20		fgomez joined #perl6
20:24	sirrobert	my vm is being silly-- rebooting it
20:26	benabik	karma silly
20:26	aloha	silly has karma of 0.
20:30	masak	karma everybody
20:30	aloha	everybody has karma of 1.
20:36		jeffreykegler joined #perl6
20:50		francium joined #perl6
20:58		sirrobert joined #perl6
20:58	sirrobert	back
21:00	Tene	karma karma
21:00	aloha	karma has karma of 2.
21:01	sirrobert	What does the syntax "Any:U", "Any:D", "Hash:D" etc. mean?
21:02	masak	sirrobert: definedness of an object.
21:02	masak	r: sub foo(Int:U) { say "type object" }; sub foo(Int:D) { say "instance"
21:02	p6eval	rakudo 038a3c: OUTPUT«===SORRY!===␤Unable to parse blockoid, couldn't find final '}' at line 2, near ""␤»
21:02	masak	r: sub foo(Int:U) { say "type object" }; sub foo(Int:D) { say "instance" }; foo Int; foo 42
21:02	p6eval	rakudo 038a3c: OUTPUT«===SORRY!===␤Redeclaration of routine foo␤at /tmp/0CD1zHN2vD:1␤»
21:02	sirrobert	so if a param is (Hash:D $d) it means "a defined hash"
21:02	sirrobert	instead of say, Hash itself
21:02	sirrobert	?
21:02	masak	r: multi foo(Int:U) { say "type object" }; multi foo(Int:D) { say "instance" }; foo Int; foo 42
21:02	p6eval	rakudo 038a3c: OUTPUT«type object␤instance␤»
21:02	masak	sirrobert: right.
21:03	sirrobert	oh, or instead of a "my Hash %foo;"
21:03	sirrobert	ok, thanks
21:03	masak	sirrobert: 'my Hash %foo' is a hash of hash objects.
21:03	sirrobert	ohhh right
21:03	masak	sirrobert: both the type annotation and the sigil add something there.
21:03	sirrobert	righto
21:03	sirrobert	are there others besides :U and :D?
21:04	masak	I believe so.
21:04	* masak	checks
21:04	masak	yes.
21:04	masak	:_ and :T
21:04	sirrobert	where did you look? =)
21:05	masak	S12.
21:05	sirrobert	ok
21:05	masak	"Abstract vs Concrete types"
21:05	sirrobert	ah
21:05	masak	:_ allows anything, it seems.
21:05	masak	:T allows only type objects, not just any undefined thing.
21:05	masak	:U allows undefined things.
21:05		Chillance joined #perl6
21:06	sirrobert	ok, makes sense
21:06	sirrobert	mostly ;)
21:08	masak	r: sub foo(Int:U) { say "undefined" }; sub foo(Int:D) { say "defined" }; foo 42; foo 42 but role { method defined { False } }
21:08	p6eval	rakudo 038a3c: OUTPUT«===SORRY!===␤Redeclaration of routine foo␤at /tmp/AT1bT0Rj52:1␤»
21:08	masak	argh.
21:08	masak	r: multi foo(Int:U) { say "undefined" }; multi foo(Int:D) { say "defined" }; foo 42; foo 42 but role { method defined { False } }
21:08	p6eval	rakudo 038a3c: OUTPUT«defined␤defined␤»
21:09	masak	apparently that wasn't undefined enough :)
21:09	sirrobert	heh
21:09	quietfanatic	Rakudo might be cheating.
21:09	Tene	r: multi foo(Int:U) { say "undefined" }; multi foo(Int:D) { say "defined" }; foo 42; foo Int
21:09	sirrobert	the "but role" applied a "mixin" that returned False for the method "defined" ?
21:09	p6eval	rakudo 038a3c: OUTPUT«defined␤undefined␤»
21:09	Tene	sirrobert: yes
21:09	sirrobert	cool
21:10	masak	yes :)
21:10	jnthn	:U is currently taken to mean "type object"
21:10	masak	huh.
21:10	jnthn	Just after I implemented that lot, the spec got changed. :/
21:11	* masak	submits rakudobug
21:11	sirrobert	is there a way to apply a mixin temporarily? (or to remove it after applying it?)
21:11	jnthn	I'm hoping if I ignore the spec change enough it may change back :P
21:11	masak	sirrobert: just keep the original object around ;)
21:11	jnthn	If you use "but" and keep the original one around... :)
21:11	sirrobert	does applying a mixin copy the object?
21:12	jnthn	sirrobert: Yes if you use "but". No if you use "does"
21:12	quietfanatic	'does' is a
21:12	quietfanatic	^ that
21:12	sirrobert	meowf! that's great
21:12	sirrobert	I'd been trying to figure that out for a while (off and on)
21:12	sirrobert	everyone++ =)
21:14	sirrobert	and now it's grammatically obvious. heh
21:19	sirrobert	man that makes a couple of things a lot easier =)
21:22	sirrobert	ok, gotta go-- thanks all =)
21:22	masak	karma go
21:22	aloha	go has karma of 0.
21:24	jnthn	using hyphens-- a curious way to play with the karma bot.
21:27	masak	I took it to mean he was dissatisfied with having to go.
21:42	[Coke]	if :_ allows anything, why do we need to specify it?
21:45	sorear	because it overrides a 'use parameters :D' pragma in scope
21:47		brrt joined #perl6
21:50	[Coke]	hokay
22:19	japhb	rn: '<1/2>'.Numeric.perl.say;
22:19	p6eval	rakudo 038a3c: OUTPUT«Failure.new(exception => X::Str::Numeric.new(source => "<1/2>", pos => 0, reason => "base-10 number must begin with valid digits or '.'"))␤»
22:19	p6eval	..niecza v19-15-g051783d: OUTPUT«Unhandled exception: Cannot parse number: <1/2>␤ at /home/p6eval/niecza/lib/CORE.setting line 1414 (die @ 5) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3492 (ANON @ 11) ␤ at /home/p6eval/niecza/lib/CORE.setting line 3494 (NumSyntax.str2num @ 5) ␤ a…
22:19	japhb	TimToady, was it your intent that the above DWIM? Or for .Numeric, does the user need to leave off the < >?
22:20	japhb	er, Str.Numeric, I mean.
22:25	masak	'night, #perl6
22:26	japhb	o/
22:33		PacoAir joined #perl6
22:37		sftp joined #perl6
22:43		PacoAir joined #perl6
22:52		skids joined #perl6
23:06		cognominal_ joined #perl6
23:18		Chat3630 joined #perl6
23:18	Chat3630	Hey:)
23:18	sorear	Hello Chat3630.
23:18	sorear	Are you here for Perl 6?
23:23	sjn	tadzik: http://news.ycombinator.com/item?id=4363887
23:32		am0c joined #perl6
23:33	sjohnson	sorear: :)
23:33	sjohnson	if not, he may get perl6 fever
23:33		gongyiliao joined #perl6
23:50		jeffreykegler joined #perl6

Perl 6 | Reference Documentation | Rakudo | Niecza | Specs