IRC log for #perl6, 2009-04-21

Time	Nick	Message
00:03	skids	std: my @a; @a = do for 1..3 { 1;}; @a.perl.say
00:03	p6eval	std 26342: OUTPUT«ok 00:02 35m␤»
00:03	skids	rakudo: my @a; @a = do for 1..3 { 1;}; @a.perl.say
00:03	p6eval	rakudo cea34f: OUTPUT«[Perl6Iterator.new()]␤»
00:04	skids	rakudo: my $a; $a = do for 1..3 { 1;}; $a.perl.say; $a.get.say;
00:04	p6eval	rakudo cea34f: OUTPUT«Perl6Iterator.new()␤Could not locate a method 'get' to invoke on class 'Perl6Iterator'.␤current instr.: 'die' pc 17453 (src/builtins/control.pir:225)␤»
00:05		FurnaceBoy joined #perl6
00:06	skids	rakudo: my @a; @a = do for 1..3 { 1;}; @a.perl.say; @a[0].say
00:06	p6eval	rakudo cea34f: OUTPUT«[Perl6Iterator.new()]␤-1␤»
00:06	skids	(which comes from a stopped iterator, I know because in my own tree I have $a.get implemented)
00:07	skids	So if the Iterator stopped, where'd the data it produced go to?
00:21		bacek_ joined #perl6
00:24	* skids	ponders .iterator versus .Iterator
00:34		hudnix joined #perl6
00:36	hudnix	perl6: if(1) {say "hello";}
00:37	p6eval	rakudo cea34f: OUTPUT«Statement not terminated properly at line 1, near "{say \"hell"␤␤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:85)␤»
00:37	p6eval	..pugs: OUTPUT«hello␤»
00:37	p6eval	..elf 26342: OUTPUT«Parse error in: /tmp/BpPKUqa7Bi␤panic at line 1 column 0 (pos 0): Can't understand next input--giving up␤WHERE: if(1) {say "hello";}␤WHERE:/\<-- HERE␤ STD_red/prelude.rb:99:in `panic'␤ STD_red/std.rb:76:in `scan_unitstopper'␤ STD_red/std.rb:224:in `comp_unit'␤
00:37	p6eval	..STD_red/std.rb:...
00:38	hudnix	rakudo: if(1) {say "hello";}
00:39	p6eval	rakudo cea34f: OUTPUT«Statement not terminated properly at line 1, near "{say \"hell"␤␤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:85)␤»
00:39	hudnix	rakudo doesn't even have "if" yet? or has the syntax changed?
00:40	mikehh	hudnix: yes
00:40	hudnix	yes to which? :)
00:40	skids	ditch the parens
00:41	hudnix	rakudo: if 1 {say "hello"}
00:41	p6eval	rakudo cea34f: OUTPUT«hello␤»
00:41	hudnix	ah ok
00:42		orafu joined #perl6
00:43	skids	(actually you can use the parens, but you'd be asking for the truth value of a "list 1" not of "1" -- it's the lack of space after "if" that really killed your first try.
00:51		frew\|wor1 joined #perl6
00:56		scook0 joined #perl6
01:13		kimtaro joined #perl6
01:14		jlaire left #perl6
01:19		awarefish joined #perl6
01:26		wayland joined #perl6
01:26		meppl joined #perl6
01:30	s1n	hudnix: if(1) is a method call, you'll either need a space before the () or remove them altogether
01:32		awarefish joined #perl6
01:33		ispy_1 joined #perl6
01:35		orafu joined #perl6
01:39		ispy_ joined #perl6
01:45		ispy_1 joined #perl6
01:50		LimbicRegion joined #perl6
01:54		agentzh joined #perl6
01:54		kimtaro joined #perl6
02:01		cspencer joined #perl6
02:01	cspencer	good evening all
02:05	skids	oh hai
02:14		Kisu joined #perl6
02:23		nbrown joined #perl6
02:27		lumi joined #perl6
02:27		cspencer_ joined #perl6
02:43		coppro joined #perl6
02:46	coppro	I'm sure you've heard this a lot, and I apologise for being a broken record, but is there a link to a development roadmap or the like somewhere?
02:48	cspencer_	there should be one distributed in the docs/ directory
02:48	cspencer_	though i'm not sure of how often it's updated
02:48	cspencer_	http://github.com/rakudo/rakud[…]f436/docs/ROADMAP
02:50		alanhaggai joined #perl6
02:51	wayland	Not very often. It says it was in August last year, but I think they actually updated it in November
02:52	wayland	But most of the stuff in section 1 hasn't been completed (with the exception that we've got the beginning of the Setting, which they called the Prelude back then)
02:52	coppro	Ah
02:53	coppro	What about Parrot?
02:53	wayland	Parrot's a different kettle of fish :)
02:53	wayland	They're already at version 1.0
02:53	wayland	And while work continues they have a recent ROADMAP somewhere
02:56	wayland	Don't get the impression that there's no progress on Rakudo, though
02:56	coppro	yeah, I don't
02:56	wayland	They've been getting a fair bit done, but not following the ROADMAP
02:56	coppro	Although I'm a bit confused how the eventual deployment of Perl 6 is going to go
02:56		eternaleye joined #perl6
02:56	coppro	Pugs => initial Perl 6 => runs on Parrot => Rakudo => self-boostraps?
02:57	wayland	Pugs is mothballed
02:57	coppro	oh
02:57	wayland	Rakudo = Perl6 on Parrot
02:57	coppro	yeah
02:57	cspencer_	coppro: there're multiple implementations of perl6 in various states of completion
02:57	wayland	But Perl6 is a spec, and Rakudo is an implementation of that spec
02:57	coppro	right
02:57	wayland	SMOP is another one
02:58	coppro	What's Rakudo written in?
02:58	skids	PIR
02:58	cspencer_	PIR and some Perl 6
02:58	wayland	SMOP is in C, and is intended to integrate well with Perl5
02:58	coppro	ah, I get it now, you're writing directly in Parrot
02:58	coppro	okay
02:58	wayland	But also Perl6 now
02:59	wayland	They've got to the point where parts of Perl6 can be implemented in Perl6
02:59	cspencer_	and writing perl6 in perl6 is so much nicer than PIR
02:59	wayland	That's called the Setting
02:59	cspencer_	:)
02:59	skids	Also as Parrot adds stuff the PIR becomes smaller, because there are more features to use.
02:59	wayland	Parrot Roadmap: https://trac.parrot.org/parrot/wiki/ParrotRoadmap
02:59	coppro	neat
02:59	coppro	anyways, I have to go, but thanks!
02:59	wayland	ok :)
02:59	wayland	bye
03:00	skids	Eventually, rakudo will also integrate STD.pm, which is the self-parsing grammar, and is both a standard and code.
03:01	coppro	my head is starting to spin
03:01	* skids	remonds self not to scare newbies with acronyms
03:01		silug joined #perl6
03:01		meppuru joined #perl6
03:27		japhb joined #perl6
03:28		alanhaggai joined #perl6
03:30	wayland	skids: remond yourself all you want, but don't redmond yourself :)
03:36		perlooooooooo joined #perl6
03:36	perlooooooooo	helllo
03:37	perlooooooooo	i am trying to develop a email extractor and i get an odd error
03:37	perlooooooooo	can anyone help me?
03:43		orafu joined #perl6
03:43	skids	in perl6?
03:44	skids	wayland: I never redmond anything :-)
03:57	wayland	skids: Me either, if I can help it.
03:59		MrKrueger joined #perl6
04:03	perlooooooooo	thank you guys
04:04	perlooooooooo	do you want to see my code?
04:06	wayland	perlooooooooo: Yes, but in a pastebin.
04:06	wayland	http://sial.org/pbot/perl6
04:06	perlooooooooo	ok
04:06	wayland	Got to that website, and paste your code in. A message should automatically appear in this channel saying that it's there
04:08	perlooooooooo	http://pastebin.com/m6e6b4bb1
04:08	buubot	perlooooooooo: The paste m6e6b4bb1 has been copied to http://erxz.com/pb/17149
04:08	perlooooooooo	so
04:08	perlooooooooo	i have a function that should extcrat all the first email of a url from craiglist
04:08	c9s	oooooooooo
04:09	perlooooooooo	when i dont use it as a function it works perfectly
04:09	perlooooooooo	i dont undestand where is the problem
04:17	perlooooooooo	Missing base argument at /usr/local/share/perl/5.8.8/HTTP/Response.pm line 91
04:17	perlooooooooo	i have this error
04:17	perlooooooooo	Missing base argument at /usr/local/share/perl/5.8.8/HTTP/Response.pm line 91
04:18	perlooooooooo	what is it?
04:18	perlooooooooo	pleaaaaaaaaase
04:26	skids	I haven't worked with HTTP:: in quite some time.
04:26	skids	there would be more people who work with it on the main perl channel.
04:26	wayland	perlooooooooo: Note that this is a Perl 5 problem. If you join #perl, I'll try to help you there.
04:27	perlooooooooo	ok
04:27	wayland	This is a channel for Perl 6, not perl 5 :)
04:27	perlooooooooo	thanks
04:27		eternaleye_ joined #perl6
04:41	wayland	Turns out perloooooo was probably a spammer. No more help for him.
04:56		agentzh joined #perl6
05:12		payload joined #perl6
05:13		ticos joined #perl6
05:17		PZt joined #perl6
05:18		ezeki3l left #perl6
05:24		meppl joined #perl6
05:29	wayland76	moritz_++ # ARGH e-mail :)
05:30		justatheory joined #perl6
05:45		andreas12345647 joined #perl6
05:46		andreas12345647 left #perl6
05:46		aff joined #perl6
06:20		DemoFreak joined #perl6
06:34	moritz_	;)
06:34	pugs_svn	r26343 \| moritz++ \| [t/spec] fix some svn props
06:44		amoc joined #perl6
06:50		ejs joined #perl6
06:53		ejs joined #perl6
06:53		iblechbot joined #perl6
06:57		ejs joined #perl6
07:07		cj joined #perl6
07:21	Matt-W	Morning
07:30		payload joined #perl6
07:32	wayland76	Not here it's not :)
07:33	wayland76	rakudo: sub judge { rand > 0.5 ? "guilty" : "not guilty"; }; say judge, "have you seen my toothbrush";
07:33	p6eval	rakudo cea34f: OUTPUT«Statement not terminated properly at line 1, near "? \"guilty\""␤␤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:85)␤»
07:34	wayland76	rakudo: sub judge { rand > 0.5 ?? "guilty" :: "not guilty"; }; say judge, "have you seen my toothbrush";
07:34	p6eval	rakudo cea34f: OUTPUT«ResizablePMCArray: Can't pop from an empty array!␤current instr.: 'parrot;PGE;OPTable;parse' pc 1809 (compilers/pge/PGE/OPTable.pir:504)␤»
07:34	wayland76	That looks like a bug. Is it a known bug?
07:35	moritz_	I think so
07:35	moritz_	it should complain about :: vs. !!
07:46		charsbar joined #perl6
07:48	wayland76	rakudo: sub judge { rand > 0.5 ?? "guilty" !! "not guilty"; }; say judge, "have you seen my toothbrush";
07:48	p6eval	rakudo cea34f: OUTPUT«guiltyhave you seen my toothbrush␤»
07:49	wayland76	rakudo: sub judge { rand > 0.5 ?? "guilty" !! "not guilty"; }; say judge, "have you seen my toothbrush";
07:49	p6eval	rakudo cea34f: OUTPUT«guiltyhave you seen my toothbrush␤»
07:49		masak joined #perl6
07:49	wayland76	rakudo: sub judge { rand > 0.5 ?? "guilty" !! "not guilty"; }; say judge, "have you seen my toothbrush";
07:49	p6eval	rakudo cea34f: OUTPUT«guiltyhave you seen my toothbrush␤»
07:49	wayland76	rakudo: say rand
07:49	p6eval	rakudo cea34f: OUTPUT«0.291138380141096␤»
07:49	wayland76	rakudo: sub judge { rand > 0.5 ?? "guilty" !! "not guilty"; }; say judge, "have you seen my toothbrush";
07:49	p6eval	rakudo cea34f: OUTPUT«not guiltyhave you seen my toothbrush␤»
07:49	wayland76	Good :)
07:49	wayland76	std: sub judge { rand > 0.5 ?? "guilty" :: "not guilty"; }; say judge, "have you seen my toothbrush";
07:50	p6eval	std 26343: OUTPUT«##### PARSE FAILED #####␤Malformed routine␤Found ?? but no !!; possible precedence problem␤Please use !! rather than :: at /tmp/I5HYeKHWiG line 1:␤------> sub judge { rand > 0.5 ?? "guilty" :: "not guilty"; }; say judge, "have you␤ expecting any of:␤ infix or
07:50	p6eval	..met...
07:50	wayland76	Well, not a bug worth submitting, then, I think :)
07:52	masak	wayland76: there's one about ?? :: already in RT.
07:52	lambdabot	masak: You have 2 new messages. '/msg lambdabot @messages' to read them.
07:52	masak	@massage
07:52	lambdabot	moritz_ asked 15h 9m 39s ago: do you know if there were any tests for RT #61128?
07:52	lambdabot	moritz_ said 14h 32m 37s ago: never mind, I just added one
07:52	masak	moritz_++
07:52	* masak	backlogs
07:56		riffraff joined #perl6
08:04		cosimo joined #perl6
08:06	Matt-W	yay more tests
08:07	masak	Matt-W: morning, sir.
08:07	Matt-W	Good morning!
08:07	Matt-W	If you see me this evening, please nag
08:07	Matt-W	I must do some more Form.pm
08:08	Matt-W	got to figure out why it's crashing Rakudo and file a bug
08:08	masak	Matt-W: was just going to ask about that.
08:08	masak	Matt-W: oh, it's crashing Rakudo?
08:08	Matt-W	Yeah
08:08	* masak	suddenly gets curious
08:08	* masak	does './proto update form'
08:08	Matt-W	I wrote a fairly primitive full justification method, and the first time I called it it crashed Rakudo
08:08	Matt-W	it's not in github yet
08:08	Matt-W	I was feeling fairly ill at the time so I didn't follow it up straight away
08:08	masak	Matt-W: please upload a branch or something.
08:09	Matt-W	I can do it later, can't get at the code from my office
08:09	masak	ah, ok.
08:09	Matt-W	IT's on the laptop at home, which is asleep
08:09	masak	I see.
08:09	Matt-W	I used quite a few things in that method that I've not used before, so I currently have numerous candidates for the cause of the crash
08:10	Matt-W	I used [+] and map :)
08:10	Matt-W	and Z
08:10	Matt-W	or did I use Z in the end
08:10	pmichaud	grrrr...
08:10	Matt-W	can't remember now
08:11	pmichaud	who removed prefix:<=> from Rakudo?
08:11	Matt-W	I think I did
08:11	Matt-W	pmichaud: not me!
08:11	pmichaud	I'd prefer not to break existing scripts and slides so abruptly.
08:11	moritz_	pmichaud: it was me
08:11	moritz_	pmichaud: I can add it back in if you want
08:12	pmichaud	Until people have time to update their modules, tutorial slides, etc., I'd prefer to keep things working.
08:12	pmichaud	I.e., given that more people are using rakudo now, we need to start thinking in terms of deprecation cycles.
08:13	masak	phenny: ah, new bot. welcome.
08:13	masak	.u smiley
08:13	phenny	masak: Sorry, no results for 'smiley'.
08:13	pmichaud	or, if we don't honor prefix:<=>, we need to at least provide a useful error message that tells someone why the information on the slide they just read isn't working properly, instead of making them guess.
08:14	masak	.u white smiling
08:14	phenny	U+263A WHITE SMILING FACE (☺)
08:15	moritz_	what's the currently recommend method of dying from PIR in rakudo?
08:18	pmichaud	'die'
08:18		kidd` joined #perl6
08:18	pmichaud	'die'('message')
08:19	moritz_	ok, I have a patch, let me test that...
08:23	moritz_	pmichaud: http://nopaste.snit.ch/16355 feel free to push that, I can't push from $here :/
08:24	pmichaud	okay
08:25	masak	literal: congrats on being accepted for gsoc. :)
08:25	frettled	pmichaud: BTW, how far did you guys get on the roadmap?
08:25	masak	literal: how long do I have to implement P2T1A before you dig into your stuff?
08:25	pmichaud	frettled: you mean on creating the updated one?
08:26	frettled	pmichaud: yup
08:27		[particle] joined #perl6
08:28	pmichaud	we got the basic pieces in place. I expect to be updating the ROADMAP on the airplane.
08:28	dalek	rakudo: 0e73267 \| pmichaud++ \| docs/ChangeLog:
08:28	dalek	rakudo: Update ChangeLog.
08:28	dalek	rakudo: review: http://github.com/rakudo/rakud[…]15cde2a857c09c55a
08:28	dalek	rakudo: 69b3180 \| pmichaud++ \| src/ (2 files):
08:28	dalek	rakudo: Restore prefix:<=> to provide a useful error message (moritz++)
08:28	dalek	rakudo: review: http://github.com/rakudo/rakud[…]500faccc7cd5fb8cb
08:28	pmichaud	(i.e., turning it into something more readable/useful)
08:29	moritz_	pmichaud: if you apply such patches with 'git-am' you automatically preserve meta data
08:29	pmichaud	moritz_: oh yeah.
08:29	frettled	Ah, that's excellent. I was worrying that you might have been out of time yesterday. :)
08:30	moritz_	(I have no problem with this kind of attribution, just think it's nicer to our contributors)
08:30	moritz_	and git-am -s adds a signed-off-by lines
08:30	pmichaud	yes, but git's "patch" isn't quite as robust as the standard "patch"
08:30	moritz_	yes, ti seems to have problems with cr/lf issues
08:31	pmichaud	also it doesn't seem to work well whenever there's an offset
08:35	jnthn	morning folks
08:35	pmichaud	good morning, jnthn
08:35	pmichaud	wireless working at the hotel again?
08:35	jnthn	yeah
08:35	pmichaud	or are you someplace else now?
08:35	jnthn	though I gotta check out in not so long
08:35	jnthn	Plan to walk around Oslo for a bit.
08:35	jnthn	Then gotta head for airport at 4pm-ish.
08:36	pmichaud	svn.pugscode.org is down?
08:36	aff	Thank you guys for coming to Oslo and for making NPW an interesting event!
08:36	pmichaud	I can't seem to update t/spec here.
08:36	pmichaud	aff: Oslo and NPW have been great.
08:37	jnthn	I can reach it...
08:37	moritz_	seems to work here
08:37	jnthn	Yes, NPW was excellent.
08:37	pmichaud	svn: PROPFIND of '/pugs/t/spec': Could not read status line: connection was closed by server (http://svn.pugscode.org)
08:37	pmichaud	maybe it's some weird airport network issue.
08:37	aff	pmichaud and jnthn: I enjoyed your talks very much.
08:38	pmichaud	I can reach svn.pugcode.org via http
08:38	frettled	ditto, I learned lots from you guys.
08:39	aff	I am porting my first application to p6 :)
08:39	pmichaud	aff: excellent! That's exactly what helps us most.
08:39	wayland76	Matt-W needs Wake-on-LAN on his laptop :)
08:40	Matt-W	it's on wifi
08:40	Matt-W	I don't think you can do it on wifi
08:40	Matt-W	what I should do is have the code mirrored on my LAN server
08:40	Matt-W	and set that up for remote login
08:40	moritz_	wake-on-air? :-)
08:41	pmichaud	my laptop has "wake on power button".
08:41	pmichaud	ahhhh, svn up works now.
08:41	pmichaud	Good.
08:41	wayland76	Yay ROADMAP updates :)
08:41	moritz_	usually wake != boot
08:42	wayland76	Someone was asking about the Perl6 ROADMAP earlier today, and We pointed him at the one that claims not to have been updated since August last year :)
08:42	pmichaud	Would like to have up-to-date spectests before getting on plane
08:42	Matt-W	mine does wake on open lid :)
08:42	pmichaud	the file hasn't been updated much since August of last year.
08:42	Matt-W	I'll have to build a robot which I can get to open the lid over the internet
08:42	wayland76	Just don't control the robot from your laptop :)
08:43	frettled	heh
08:43	moritz_	that somehow reminds me of a nice little story... http://thedailywtf.com/Comments/Open-Sesame.aspx
08:43		kane__ joined #perl6
08:44	moritz_	probably the ssh solutions was much safer than any card reader (unless they used a broken random number generator :)
08:45	frettled	heh
08:49		kane__ joined #perl6
08:50		agentzh joined #perl6
08:55	masak	rakudo: multi method dispatch(@chunks, %param?) { say defined %param }; dispatch([1,2,3]); dispatch([1,2,3], { "OH" => "HAI" })
08:55	p6eval	rakudo 69b318: OUTPUT«No applicable candidates found to dispatch to for 'dispatch'␤current instr.: '_block14' pc 106 (EVAL_17:52)␤»
08:55	masak	bug in Rakudo or in masak?
08:55	moritz_	s/method/sub/?
08:55	masak	moritz_: ah, thank you.
08:55	frettled	and where do we post masak bugs, anyway?
08:56	moritz_	frettled: rakudobugperl.org
08:56	moritz_	oh wait
08:56	masak	frettled: masakbugmasak.org :P
08:56	frettled	:)
08:56	frettled	masak: \o/
08:56	masak	frettled: you can't submit them through the web interface, only mail.
08:56		yahooooo joined #perl6
08:57	frettled	masak: and I suppose patches are submitted via BK instead of svn or git ;)
08:58	moritz_	BK?
08:58	masak	Burger King?
08:58	frettled	BitKeeper
08:58	jnthn	I CAN HAZ CHEEZBURGER!
08:59	frettled	ohnoes
08:59	masak	rakudo: class A { multi method dispatch(@chunks, %param?) { say %param.perl, defined %param } }; A.new.dispatch([1,2,3]); A.new.dispatch([1,2,3], { "OH" => "HAI" })
08:59	jnthn	ohnoes now I can't
08:59	p6eval	rakudo 69b318: OUTPUT«{}1␤{"OH" => "HAI"}1␤»
08:59	masak	I'm slightly surprised that the former is defined.
08:59	masak	but maybe it has to be that way.
08:59	moritz_	masak: that's because %param is an instance
08:59	masak	an instance?
09:00	* Matt-W	writes some Perl 5 and finds it very strange
09:00	moritz_	objects
09:00	jnthn	An instance of Hash
09:00	moritz_	objects of normal classes are always defined
09:00	jnthn	ohnoes I have to check out
09:00	masak	ok, you two don't seem to think this is an error, so I guess it isn't :)
09:00	jnthn	catch y'all later :-)
09:00	masak	jnthn: o/
09:01	masak	Matt-W: I find the more frequently you switch, the easier it becomes.
09:01	wayland76	bye :)
09:05	masak	rakudo: my %h = { foo => "1" }; %h<foo bar baz> //= 2,3,4; say %h.perl
09:05	p6eval	rakudo 69b318: OUTPUT«{"foo" => "1", "bar" => undef, "baz" => undef}␤»
09:05	frettled	masak: I think I grokked the code in your previous example now, and yes, I think it makes sense that it works the way it did.
09:06	masak	frettled: with the instantiated hash?
09:06	moritz_	but the last one is... weird
09:06	frettled	masak: yep
09:06	masak	frettled: well, I just want a simple way to test if the parameter was passed.
09:06	masak	moritz_: submit a rakudo/
09:06	masak	...bug?
09:07	masak	I was thinking maybe some interaction between list assignment and defined-or disallows that.
09:08		payload joined #perl6
09:08		rblasch joined #perl6
09:09	moritz_	masak: I think it's a manifestation of a different bug
09:09	masak	which one?
09:09	moritz_	it's ok that the //= does nothing
09:09	moritz_	but mentioing %foo<bar> autovivifies it
09:10	frettled	argh, I forgot that Firefox is _very_ unhappy about searching through any of perlcabal.org/syn/S*.html.
09:10	moritz_	rakudo: my %h; %h<foo>; say %h.perl
09:10	p6eval	rakudo 69b318: OUTPUT«{"foo" => undef}␤»
09:10	moritz_	masak: that's long known and ugly
09:11	masak	moritz_: still, assigning with //= should overwrite the undef value.
09:11	frettled	masak: Anyway, S06/Optional parameters/ - "Missing optional arguments default to their default values, or to an undefined value if they have no default."
09:11	masak	moritz_: or do the autovivification and the assignment happen in the wrong order?
09:11	frettled	masak: so it appears that there is a bug, then, since you haven't specified a default for %param
09:11	moritz_	masak: does //= dwim over lists values?
09:11	moritz_	masak: if so, where's that specced?
09:11	masak	frettled: or maybe {} is the default for Hash...
09:12	masak	moritz_: that's part of what I'm wondering.
09:12	masak	moritz_: and if not, how does one write what I meant?
09:12	moritz_	masak: maybe you need a hyper
09:12	frettled	masak: the way I read the spec, the default values are the ones you specify in the sub declaration.
09:13	Matt-W	masak: I think you might be right - it works for musical instruments, anyway.
09:13	moritz_	for <a b c> Z 1, 2, 3 -> $k, $v { %h{$k} //= $v }
09:14	* masak	's head explodes from three conversations at once
09:14	wayland76	masak: Let's talk about having multiple conversations too :)
09:14	masak	:)
09:14	frettled	:)
09:14	moritz_	frettled: if %h were undef, you couldn't assign values to it
09:15	masak	ack. it's more dwimmy that it's {}.
09:15	wayland76	rakudo: for <a b c> Z 1, 2, 3 -> $k, $v { %h{$k} //= $v }
09:15	frettled	moritz_: hrm, that's breaking the principle of least surprise.
09:15	p6eval	rakudo 69b318: OUTPUT«Scope not found for PAST::Var '%h' in ␤current instr.: 'parrot;PCT;HLLCompiler;panic' pc 146 (src/PCT/HLLCompiler.pir:104)␤»
09:15	masak	but... I just want a nice way to see if the hash was passed or not.
09:15	moritz_	frettled: maybe tye spect needs to be updated to only refer to scalar optional parameters
09:15	moritz_	s/tye/the/
09:16	moritz_	rakudo: my %h = a => 5; for <a b c> Z 1, 2, 3 -> $k, $v { %h{$k} //= $v }; say %h.perl
09:16	frettled	moritz_: or to allow assignment to undefined non-scalars ;)
09:16	p6eval	rakudo 69b318: OUTPUT«{"a" => 5, "b" => 2, "c" => 3}␤»
09:16	moritz_	frettled: that would be much weirder, overall
09:16	frettled	moritz_: would it?
09:17	moritz_	frettled: yes.
09:17	frettled	moritz_: I don't see how, though.
09:17	moritz_	frettled: if you have an undef value, it's usually not a "normal" object, but a proto object
09:18	moritz_	frettled: so if a default value is such a proto object, a %h<a> = 3 would call the postcircumfix:<< < > >> method as a class method
09:18	moritz_	frettled: which doesn't have access to attributes - so it's going to barf
09:18	masak	moritz_: what does 'my %h;' do?
09:19	moritz_	masak: it calls Hash.new under the hood, I assume
09:19	masak	then it can be argued that '%h?' should, too.
09:19	frettled	moritz_: so essentially, it's been designed into a corner where you can't get out of without getting hacks all over your feet ;)
09:19	moritz_	frettled: right
09:19	frettled	moritz_: $expletive
09:19	moritz_	frettled: of course there are other options...
09:20	frettled	moritz_: yes, _is rw_, for instance, seems like it would help.
09:20	* frettled	is a Perl 6 spec n00b, though.
09:20	moritz_	frettled: like having %h? defaulting to a Hash.new(defined => 0) or so
09:20	frettled	moritz_: no, that's the wrong solution, becaus it still means you can't say whether you got passed an empty hash or not.
09:21	frettled	s/caus /cause/
09:21	moritz_	frettled: you could, because an empty hash defaults to being defined
09:21	frettled	ah, I misunderstood the magic
09:21	* frettled	<- stupid.
09:22	moritz_	but there's a drawback... every modifying call to a hash would have to set its defined flag
09:22	frettled	mm
09:22	frettled	I think this is a design weakness.
09:23	moritz_	the problem is that we're doing in-band communication
09:23	moritz_	instead of asking the hash parameter if it was defined, we should ask the capture if that hash was given
09:24	masak	aye.
09:24	moritz_	especially since objects can respond with False to .defined at will anyway
09:24	frettled	the way you put it, it seems like there is a straight-forward way of doing so, and that would be good.
09:24	frettled	Conceptually, that's much cleaner than asking the variable.
09:25	moritz_	the simplest solution that works with the current spec is to split it into two multis
09:25	moritz_	but that's not always convenient
09:26	frettled	I was considering to suggest that, but I thought it would be neater to use the optional argument as, well, optional.
09:26		agentzh joined #perl6
09:26	frettled	In more complex cases it would make more sense to create multiple methods.
09:27	frettled	(going Java)
09:29	frettled	There doesn't appear to be a test for this in S06, though. Unless I'm misreading the tests or missing a test, only scalars are tested. But my test-fu is weak.
09:29		pmurias joined #perl6
09:30	moritz_	frettled: it might well be missing... but let's try to get clarification on the spec first, and then write tests :-)
09:30	frettled	moritz_: yep.
09:30	frettled	moritz_: though the spec is pretty clear about it, really, I just had to read the right part of it.
09:31	frettled	It's right there in the intro of S06/Parameters and arguments/
09:31	frettled	It speaks of "container argument" as opposed to a scalar.
09:32		payload joined #perl6
09:37	frettled	so if anyone has the test-fu to write a test for testing whether optional hash and array parameters are 1) immutable, 2) defined with their correct default value, and 3) handles correctly with is rw, is copy, etc.
09:37	frettled	...that would be nice. Then I can leech on and see how that's tested. :D
09:41	frettled	Lunch in T-19 minutes
09:43	masak	frettled: that's mixing of two constructs. :) it's either just "T minus 19" or "lunch in 19 minutes"...
09:44	Matt-W	my lunch today will include arguments
09:44	antiphase	t(lunch) = NOW() + 19;
09:46		mhsparks joined #perl6
09:52		rindolf joined #perl6
09:57	frettled	masak: Ah, but I didn't tell you what T was. waves arms desperately
09:57	masak	:P
09:57	masak	you'd probably need to redefine infix:<-> as well.
10:03		patmat joined #perl6
10:04	wayland76	class T { return -NOW() };
10:04	pugs_svn	r26344 \| pmurias++ \| [re-smop] LOST compiles (the MESSAGE is still empty)
10:05	masak	wayland76: a slight nit: you can't return from a class body.
10:06	wayland76	That's why I didn't send it to rakudo :)
10:06	wayland76	Ah, I know...
10:06	masak	rakudo: class A { method closure() { return { say $!greeting } } }; class B { has $!greeting; method run-code(Callable $c) { $c() } }; B.new().run-code( A.new().closure() )
10:06	p6eval	rakudo 69b318: OUTPUT«No such attribute '$!greeting'␤current instr.: 'parrot;A;_block29' pc 347 (EVAL_22:129)␤»
10:06	wayland76	class T { new { return -NOW() } };
10:06	masak	that works.
10:06	masak	'method new', though.
10:07	wayland76	masak is a fully-implemented perl6 compiler :)
10:07	wayland76	rakudo: class T { method new { return -NOW() } };
10:07	p6eval	rakudo 69b318: ( no output )
10:08	masak	rakudo: class A { has $!greeting = "I'm guessing this is actually the specced behaviour"; method closure() { return { say $!greeting } } }; class B { has $!greeting; method run-code(Callable $c) { $c() } }; B.new().run-code( A.new().closure() )
10:08	p6eval	rakudo 69b318: OUTPUT«I'm guessing this is actually the specced behaviour␤»
10:08	masak	Rakudo++
10:09		agentzh joined #perl6
10:22		orafu joined #perl6
10:26	aff	Is hash:delete not implemented yet?
10:27	masak	aff: correct. only %h.delete($key)
10:27	aff	masak: Thanks
10:29	pugs_svn	r26345 \| pmurias++ \| [re-smop] SMOP__NAGC__RI__create takes a WEAKREF argument
10:32		orafu joined #perl6
10:35		ascent__ joined #perl6
10:36		pjcj joined #perl6
10:48		Maghnus joined #perl6
10:51		patmat left #perl6
11:02		orafu joined #perl6
11:08		azawawi joined #perl6
11:14		smtms joined #perl6
11:18		orafu joined #perl6
11:38		iblechbot joined #perl6
11:44		fridim_ joined #perl6
11:49	moritz_	masak: yes, it's specced like that... $!greeting is not a lexical, so a block won't close over it
11:50	masak	but note that the block did in effect close over it.
11:50	moritz_	how so?
11:51	moritz_	it's a method, so it has access to the attributes
11:51	moritz_	I don't see any closing.
11:56	masak	the closure uses the 'self' defined in the closure() method, even though it's called in the run-code($c) method.
11:58	pugs_svn	r26346 \| pmurias++ \| [re-smop] LOST passes tests
11:59		nsh joined #perl6
12:03		abra joined #perl6
12:04		Psyche^ joined #perl6
12:13		cosimo left #perl6
12:20	masak	class A { has Callable $.c; method foo() { say defined $.c } }; A.new.foo
12:20	masak	I must say this is highly counterintuitive.
12:20	masak	rakudo: class A { has Callable $.c; method foo() { say defined $.c } }; A.new.foo
12:20	p6eval	rakudo 69b318: OUTPUT«1␤»
12:21	masak	if that one is defined, what in the world is it defined to be?
12:21		ihrd joined #perl6
12:23	masak	I feel both this case and the case with the default parameters undermines the whole concept of definedness.
12:24	masak	to me, 'defined' should be able to show whether the variable in question has been assigned to, initialized etc.
12:27	masak	moreover...
12:28	masak	rakudo: class A { has $.c; method foo() { say defined $.c } }; A.new.foo
12:28	p6eval	rakudo 69b318: OUTPUT«0␤»
12:28	masak	now it works as we all expect it to!
12:28	masak	so the current behaviour is actually a kind of punishment for adding types!
12:28	frettled	muahaha
12:29	* masak	mutters inaudibly
12:29	frettled	masak: it's a shame that you never got to hear sjn and I practice our evil laughter.
12:30	masak	frettled: looking forward to hearing it next time we meet.
12:30	frettled	masak: that's a deal!
12:31	baest	masak: I think it's only Roles
12:31	frettled	yes, we play them as best we can.
12:32	baest	rakudo: class B { }; class A { has B $.c; method foo() { say defined $.c;} }; A.new.foo'
12:32	LylePerl	pmichaud: The issue I was having on Vista turned out to be an issue with Rakudo's realclean and not the code itself.
12:32	p6eval	rakudo 69b318: OUTPUT«Statement not terminated properly at line 1, near "'"␤␤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:85)␤»
12:32	baest	rakudo: class B { }; class A { has B $.c; method foo() { say defined $.c;} }; A.new.foo';
12:32	LylePerl	pmichaud: Maybe that's the same issue on Darwin?
12:32	p6eval	rakudo 69b318: OUTPUT«Statement not terminated properly at line 1, near "';"␤␤current instr.: 'parrot;PGE;Util;die' pc 129 (runtime/parrot/library/PGE/Util.pir:85)␤»
12:32	baest	rakudo: class B { }; class A { has B $.c; method foo() { say defined $.c;} }; A.new.foo;
12:32	p6eval	rakudo 69b318: OUTPUT«0␤»
12:32	baest	sorry for the noise
12:32	masak	baest: I think you're right.
12:32	masak	but I don't think roles should behave like this either.
12:32	baest	indeed
12:33	baest	would be _very_ weird
12:34	frettled	And the guys who can answer conclusively are either having a good walk in the sunshine or on their way to/from an airport.
12:34	masak	moritz_ seemed to argue earlier that the optional-params case was correct.
12:37	frettled	Yes, at least correct according to spec, and as far as I could tell, it was.
12:38	masak	right.
12:38	masak	that's why I'm not submitting it as a rakudobug, because it's really a spec issue.
12:39	masak	by the way, what's a way to write an empty code block in Perl 6?
12:39	frettled	But Someone� should probably write new tests. I don't know how yet. :)
12:39	masak	i.e. 'my Callable $c = ...'
12:40	frettled	oh, an empty anonymous/implicit sub?
12:40	masak	rakudo: my Callable $c = -> {}
12:40	p6eval	rakudo 69b318: ( no output )
12:40	masak	ah, that's the way.
12:40	frettled	That was almost too easy.
12:41	frettled	no black candles or goats necessary. But then again, it's not SCSI.
12:42	masak	frettled: the important point being that the 'obvious' solution is wrong here.
12:42	masak	rakudo: my Callable $c = {}
12:42	p6eval	rakudo 69b318: OUTPUT«Type mismatch in assignment.␤current instr.: 'die' pc 17466 (src/builtins/control.pir:225)␤»
12:42	masak	because {} is an empty Hash.
12:42	masak	rakudo: say Hash ~~ Callable
12:42	p6eval	rakudo 69b318: OUTPUT«0␤»
12:44	frettled	masak: Yes, and I don't think I quite got an answer to how to check whether an argument was used in any other way than doing multiple dispatch to eliminate the alternatives.
12:44	masak	frettled: indeed.
12:45	masak	this discussion is not-yet-resolved.
12:45	frettled	Since defined() obviously isn't the correct way to do it, an alternative method is necessary, e.g. another property for each parameter that's specified in the sub/method definition.
12:46	frettled	That would be quite useful, since it allows us to separate between the following cases:
12:46	frettled	1) The parameter was not used.
12:46	frettled	2) The parameter was used, but was undefined.
12:47	frettled	I'm not eager to fight for this, since I'm not sure where the snowball will stop rolling.
12:47	masak	whoa, "obviously isn't the correct way"? please explain -- I didn't know it was obvious that it isn't the correct way, let alone that it isn't the correct way...
12:48	masak	frettled: for a typed variable, the case 2) will never occur.
12:48	masak	s/variable/parameter/
12:48	frettled	masak: ah, right, that was mentioned earlier as well.
12:49	frettled	I used "obviously" loosely; it's not quite obvious to me that this is the way it _ought_ to be, but it's obvious that it's the way it _is_.
12:49	masak	ah. :)
12:50	masak	I think 'defined' should be used to test whether a variable, parameter or attribute has been initialized.
12:50	masak	I think that typing shouldn't affect this maxim.
12:53	frettled	Yup.
12:53	frettled	As I said before, it breaks with the principle of least surprise, and it also breaks with the principle of not having special exceptions for common cases.
12:54	masak	mm.
12:54	masak	I think I'm ready to submit a rakudobug after all.
12:54	masak	but it'll be one helluva rakudobug... :)
12:54	* masak	submits
12:59	aff	In a function call, must the argument list follow a function name without whitespace in between?
12:59	aff	E.g. ok(sum (1,2,3) == 6);
13:00	aff	use Test; plan 2; sub sum (*@numbers) { return [+] @numbers; }; ok(sum(1,2,3) == 6); ok(sum (1,2,3) == 6);
13:00	aff	I get not ok 2
13:01	masak	rakudo: say +(sum (1,2,3))
13:01	p6eval	rakudo 69b318: OUTPUT«Could not find non-existent sub sum␤current instr.: '_block14' pc 85 (EVAL_16:49)␤»
13:01	masak	rakudo: sub sum (*@numbers) { return [+] @numbers; }; say +(sum (1,2,3))
13:01	p6eval	rakudo 69b318: OUTPUT«6␤»
13:01	masak	rakudo: sub sum (*@numbers) { return [+] @numbers; }; say +(sum (1,2,3) == 6)
13:01	p6eval	rakudo 69b318: OUTPUT«0␤»
13:07	masak	aff: the expression (1,2,3) == 6 is treated as one value and sent as a param to sum.
13:08	aff	Not a bug then?
13:08	masak	don't think so.
13:08	aff	Ok thanks
13:08	masak	use parentheses. :)
13:10	frettled	hmm.
13:10	frettled	rakudo: sub sum (*@numbers) { return [+] @numbers; }; say +(sum <1 2 3> == 6)
13:10	p6eval	rakudo 69b318: OUTPUT«0␤»
13:10	frettled	just curious :)
13:13	masak	I guess the answer can be formulated like this, too: infix:<==> has higher precedence than listop function calls.
13:15	frettled	mm
13:16	frettled	Just use the periodic table of operators (GIYF). Caveat: there are two syntax changes since it was updated in February.
13:18	aff	Ok thanks guys
13:19		patmat joined #perl6
13:20		cosimo joined #perl6
13:20	frettled	masak: Emacs? Jolly good!
13:21	masak	frettled: I'm sorry? :)
13:21		aff left #perl6
13:22		ihrd left #perl6
13:22	frettled	masak: I just asked for your IRC client's version, it responded with ERC 5.2, which runs under Emacs. So, does that mean you have syntax hilighting for Perl 6 in Emacs? :D
13:23	masak	frettled: I don't use Emacs much to write Perl 6 code.
13:23	frettled	I was wondering why people us e.g. <==> to quote == instead of using bold, underline or inverse.
13:23	masak	I prefer vim for coding.
13:25	patmat	is there an actual, really good ebook/tutorial for Perl6 in the net?
13:25	patmat	i mean for programming newbies
13:26	* frettled	guesses "no".
13:26	masak	patmat: unfortunately not, only bits and pieces.
13:26	patmat	:(
13:26	masak	patmat: on the other hand, the current implementations are equally fractured and incomplete, so you'd need a high quotient of patience and research anyway.
13:27	patmat	ok so i let it be :P
13:27	patmat	at the moment
13:27	masak	patmat: I have a bunch of recommended reading for you though, if you plan to stay long enough to let me remember it.
13:27	masak	especially moritz_++'s perl5-to-perl6 blog posts.
13:27	frettled	I suspect that programming newbies might be better off learning another language, as the situation is right now.
13:27	masak	(assuming you have prior Perl 5 experience)
13:28	frettled	I'd even go so far as to suggest that programming newbies don't look at Perl 5 either.
13:28	masak	frettled: well, it's theoretically possible to learn programming by learning Perl 6.
13:28	masak	frettled: but it'll be a tough ride. :)
13:28	frettled	masak: it's also _practically_ possible to learn programming without learning a programming language. :)
13:28	masak	frettled: I'd tend to agree.
13:28	frettled	(though it's not very practical, ha ha)
13:29	frettled	I appreciated learning Simula and then LPC (Lars Pensj�'s C, object-oriented C for MUD programming), but YMMV.
13:32	wayland76	patmat: The advice I tend to give people is that Perl is a good last language to learn.
13:33	wayland76	After you've learned one or two others that can only do some things, then learn Perl, because it can do all of them :)
13:33	wayland76	The disadvantage to perl is that, because it can do everything, it tends to be big
13:33	patmat	i wouldnt look to Perl 5 :)
13:33	patmat	better Python/Perl
13:35	wayland76	patmat: You say "Programming Newbie", but you seem to have some background. Do you mind if we ask what languages you've already tried?
13:35	masak	patmat: here's why Perl 6 might be interesting to learn. let's say you have a file with a number on every line, and you want to sum those numbers. now, in most languages that would be a lot of trickery with opening files and for loops.
13:35	frettled	Python is also pretty far down on the list of languages for newbies, since it tends to be confusing for the next language.
13:36	masak	patmat: in Perl 6, it's 'say [+] lines("myfile.txt")'
13:36	frettled	In awk, it's: awk '{ n+=$1 } END { print n }' myfile.txt
13:37	frettled	>:->
13:38	masak	what I like about the Perl 6 version is that it aligns very well with the problem specification. it contains the primitives "say" "addition" "lines" and the file.
13:38	masak	that's much the same as was used to describe the problem.
13:38	baest	frettled: shorter in p5: perl -n -e ' $n+=$_; END { print $n }' myfile.txt
13:39	frettled	baest: perl -ne >:->
13:39	patmat	wayland76: i learned a bit of Ruby, and Python
13:39	patmat	and in eaaarly days, QBasic :P
13:39	frettled	yow
13:39	frettled	well, there are worse things to learn than QBasic.
13:40	masak	indeed.
13:40	wayland76	I did GWBASIC, followed by QBasic, followed by QuickPascal, all before I got to Uni
13:40	wayland76	Oh, and Assembly, and had a look at ForTran :)
13:41	wayland76	Anyway, patmat, it sounds like you're not quite a newbie
13:41	masak	I did GWBASIC, Turbo Basic, QBasic, Turbo Pascal, 386 Assembly, Visual Basic and a smattering of C before I got to Uni.
13:42	wayland76	I'd say stick with Ruby until Perl 6 is readier -- check back in about 6 months :)
13:42	* frettled	did assembly, MS BASIC(A), Logo, then had a few years' worth of recess before going to the University and discovering that I was supposed to do informatics.
13:42	* frettled	will not recommend assembly as the starting point.
13:43	wayland76	I had BASICA too, and a little Logo at school
13:43	wayland76	oh, and apple basic (as in Apple ][ )
13:43	masak	patmat: I'd add to wayland76's suggestion that looking at Perl 6 at the current stage is quite awesome too, but at times frustrating.
13:43	frettled	As opposed to masak, I don't recall _which_ assembly, because I think I was only 8 or 9 at the time and only wanted to use the beeper to create music with loops and conditionals.
13:44	masak	ah, the beeper.
13:44	frettled	or "speaker", as it would be called in the IBM PC.
13:44	wayland76	Who remembers what the GW in GWBASIC stood for? (I do, I'm just testing... :) )
13:44	masak	aye.
13:44	* moritz_	does not
13:44	wayland76	Well, it was a fine speaker, but had lame circuitry leading to it
13:45	* masak	doesn't either
13:45	frettled	but it wasn't a PC, it was a loose board with 1 or 2 KiB memory and a 5- or 6-digit calculator-type display.
13:45	frettled	I didn't know, but it was apparently "Graphics and Windows".
13:46	* frettled	cuddles his uni's local acronym/initial database.
13:46	wayland76	frettled: I was told "Gee-Whiz" :)
13:46	frettled	hehe
13:46	masak	seems the name question is disputed: http://en.wikipedia.org/wiki/GW-BASIC#Name
13:47	wayland76	masak: I was just looking at that to check if I was right :)
13:47	masak	hm. today is not a productive day, $WORK-wise.
13:48		yahooooo6 joined #perl6
13:49	moritz_	same here
13:49	wayland76	Lets make that anonyous!
13:50	wayland76	(anonymous = clown version of unanimous)
13:52	* frettled	mumbles something about $expletive script kiddies and $expletive customers and @expletive.
13:53	wayland76	Is it just me, or are we mostly sysadmins here?
13:53	Matt-W	I'm not a sysadmin
13:54	Matt-W	I run my home network sloppily, but that's just so I can get stuff done
13:54	moritz_	wayland76: I'm one, though only part time
13:54	frettled	I fled consulting to become a sysadmin. I'm not sure that's a wise choice, but at least I'm not earning loads of money coding Java, which of course would be awful.
13:54	masak	I'm decidedly not a sysadmin.
13:54	masak	I'm a regular code monkey.
13:54	frettled	ooh, a code monkey!
13:54	Matt-W	I'm a code monkey too
13:54	wayland76	ok :)
13:54	frettled	I need one or two of those!
13:54	* frettled	holds on to masak tightly.
13:55	Matt-W	C++ mostly, bits of Perl and some Java when I can't avoid it
13:55	* frettled	holds on to Matt-W tightly.
13:55	masak	frettled: let go of my tail!
13:55	Matt-W	Plus a lot of Perl 6 evangelism
13:55		ispy_ joined #perl6
13:56	wayland76	ok, only about half sysadmins then
13:56	frettled	I like solving problems, but when I've done so, I don't like implementing the solution. That's the boring part.
13:56	Matt-W	I like making things
13:56	wayland76	frettled: tell me about it :)
13:56	Matt-W	Preferably flying by the seat of my pants, experimental stuff
13:56	wayland76	I think we're all exhibiting the virtue of laziness :)
13:57	frettled	This is outrageous, by the way: I printed the periodic table of elements and showed it to one of my co-workers, and he called me "very nerdy".
13:58	Matt-W	Try the table of perl 6 operators
13:58	frettled	That's theone.
13:58	frettled	s/eo/e o/
13:58	Matt-W	it's beautiful
13:58	Matt-W	tempted to print it out at work
13:58	frettled	Yes, especially after it got antlers, signatures and hand-written corrections.
13:58	moritz_	where is it btw? we should link to it from perl6-projects.org
13:58	Matt-W	we can do A3 on the colour laser
13:59	frettled	moritz_: http://www.ozonehouse.com/mark/periodic/
14:00	moritz_	frettled: thanks
14:03	frettled	np
14:03		pmurias joined #perl6
14:03	pugs_svn	r26347 \| pmurias++ \| [re-smop] started adding .m0ld tests to s1p
14:03	pugs_svn	r26347 \| pmurias++ \| even this simple one uncovers a memory leak
14:09	pugs_svn	r26348 \| moritz++ \| [perl6-projects.org] add link to periodic table
14:12	patmat	masak ok i stick to it, that it could be awesome.. but i'd need a good documentation or better say ebook/tutorial for it
14:12	literal	moritz_: the "Perl 6 highlighting for vim" link can be changed to http://github.com/hinrik/vim-perl
14:12	masak	patmat: did you check out moritz_'s perl5-to-perl6 blog?
14:13	patmat	whats the page again? sorry im in work
14:13	patmat	i hate it to be a Supporter... even for Windows stuff
14:13	masak	patmat: http://perlgeek.de/blog-en/perl-5-to-6/
14:13	patmat	its my first job, i want to be a coder once
14:13	patmat	thanks
14:13	masak	should be enough to get your feet wet.
14:14	patmat	so is there an easy way to install perl6 on a windows machine? cause for work...
14:14	patmat	private i like to use linux/bsd
14:14	literal	moritz_: never mind, I'll update :)
14:15	pugs_svn	r26349 \| hinrik++ \| [perl6-projects.org] update perl6.vim link
14:15	moritz_	literal++ # JFDI :-)
14:15	masak	moritz_: the order of the index in your blog linked above doesn't correspond to the order of the posts. just thought you'd like to know.
14:17	moritz_	masak: that's intional. One is alphabetic, the other is by date
14:17	patmat	so is there a way to install it on Windows?
14:17	masak	moritz_: oh, ok.
14:17		cosimo left #perl6
14:18	masak	patmat: http://rakudo.org/how-to-get-rakudo
14:18	pugs_svn	r26350 \| hinrik++ \| [perl6-projects.org] link to Perl 6 Programming on Wikibooks
14:19	frettled	Now all we need is that syntax hilighter for Emacs.
14:19	masak	patmat: I'd recommend the tarball approach mentioned in the "If you don't have git installed..." paragraph.
14:19	masak	frettled: I've found that cperl-mode does a decent job.
14:20	masak	s/\./ highlighting Perl 6./
14:20	literal	there is an old modified cperl-mode.el in the pugs repo
14:21	literal	don't know how much it supports
14:21		rewt joined #perl6
14:22	masak	literal: do you know when you'll be starting your GSoC work? I'd like to have P2T1A implemented by then.
14:22	literal	P2T1A ?
14:23	masak	yes, a throwaway prototype.
14:23	masak	actually, not even that.
14:23	masak	it's enough documentation to be able to say something sensible about the documentation data format.
14:23	moritz_	May 23:
14:23	moritz_	* Students begin coding for their GSoC projects
14:23	masak	oh, good.
14:23	moritz_	http://socghop.appspot.com/doc[…]gsoc2009/timeline
14:24	literal	yeah, may 23rd officially
14:24	moritz_	of course nobody can prevent literal from starting earlier :)
14:24	literal	http://nix.is/gsoc/gsoc-u4x.html
14:24	literal	there's a schedule there
14:25	masak	seems my deadline can still be May 23, given that schedule.
14:25	frettled	masak: cperl-mode is okay, except that it does different parsing mistakes than perl-mode. :)
14:25	masak	the 'start prototyping grok' will ideally be helped by my musings.
14:25	frettled	I suspect that Perl 6 may be easier to hilight.
14:26	masak	frettled: I've always found cperl-mode much, much more stable than perl-mode. don't know why.
14:26		cosimo joined #perl6
14:26	masak	I think it might have been implemented with failure robustness in mind, whatever that means for Perl code parsing.
14:26	moritz_	frettled: I don't think so - the ambiguty between terms and operators is more present in Perl 6
14:26	literal	frettled: Perl 6 has more orderly syntax yeah, but it also has much more of it...
14:27	literal	and yeah, the predictive parsing thing
14:27	pugs_svn	r26351 \| pmurias++ \| [re-smop] removed surplus REFERENCE from capture
14:28	literal	frettled: though emacs has much better support for external processes so one could probably use STD.pm to highlight it
14:28	frettled	yup, that was suggested this weekend.
14:29	moritz_	that said, the grammar for Perl 6 is much shorter than perly.{c,y} and toke.c
14:29		cosimo left #perl6
14:29	masak	moritz_: that's because it's written in a much more high-level language. :)
14:29	literal	moritz_: any reason your perl-5-to-6 isn't mentioned on perl6-projects.org ?
14:31	moritz_	literal: the blog is assembled on planetsix, which is already there... but we could add it anyway
14:31	literal	ok
14:32	moritz_	just don't add the normal perl 6 part of my blog, that would be a duplicate
14:32	literal	yeah
14:36	pugs_svn	r26352 \| hinrik++ \| [perl6-projects.org] add link to Perl 5 to 6 lessons
14:37		skids joined #perl6
14:40		nihiliad joined #perl6
14:48		nsh joined #perl6
14:48	* skids	backlogging -- wifi wake on lan is called WOWLAN, and theres a smattering of support on various adaptors.
14:49	* moritz_	knows WOW without the LAN part :-)
15:01	patmat	i think i choose PUGS
15:02	patmat	for the work here
15:02		mofino joined #perl6
15:04	literal	what work?
15:04	wayland76	patmat: That's kind of dangerous -- most of the active developers are working on Rakudo
15:04	* skids	backlogging: not a sysadmin. A net admin. As in actual network hardware, not PCs/servers, e.g. those sysadmins that call themselves netadmins.
15:08	skids	however it would not hurt to keep pugs somewhat in sync with the test suite as the standard changes.
15:08	skids	If not the big stuff, at least the simple method renames and such.
15:09	moritz_	skids: feel free
15:09	skids	:-)
15:09	skids	Was more encouragement to patmat if he really likes haskel
15:12	moritz_	that said there is a Dec 2008 pugs release here: http://hackage.haskell.org/cgi[…]ipts/package/Pugs
15:12	skids	So last night I was trying to suss whether we need both .iterator and .Iterator, or whether .iterator is now cruft.
15:13	PerlJam	skids: what's .Iterator or .iterator?
15:13	PerlJam	skids: aren't they all ['Iterator'] now?
15:13	PerlJam	or am I thinking of something completely different?
15:14	moritz_	PerlJam is talking abouth Rakudo, skids about the spec
15:14	skids	.Iterator would have to follow the "methods named after a class are for casting/coercing" whereas .iterator might offer more flexibility in semantics.
15:15	skids	But I can't think of a situation where the difference would matter.
15:15	PerlJam	ah.
15:16	skids	Of course as I've often said esoteric CS minutia elude me.
15:18	skids	If it is the case that the spec should now exclusively say .Iterator I wouldn't mind doing that, but don't want to if there's some reason for .iterator.
15:18		yahooooo joined #perl6
15:19	[particle]	lowercase means it's a base type
15:20	[particle]	er, that's too many loaded words.
15:20	moritz_	.true usually returns a Bool
15:20	PerlJam	yeah, I was about to ask what a "base type" is :)
15:20	moritz_	not a native type
15:20	moritz_	if you meant s/base/native/
15:25	skids	Yeah rakudo-wise I get that some of the .iterator calls are looking for Parrot 'iterator', but is there a reason for .iterator on the spec level?
15:31		lumi joined #perl6
15:32		aindilis` joined #perl6
15:34		edenc joined #perl6
15:35	patmat	wayland76: ok pugs doesnt work anyway.. so i need Rakudo for Windows :)
15:36	PerlJam	patmat: unless you want to make pugs work.
15:36	patmat	literal: i'm at a company, in the IT Support.. sucks a bit, but its my first job.. so.. its ok for the beginning
15:36	patmat	PerlJam: nah, the reason that most people work on rakudo is enough :)
15:36	literal	why do you need pugs there?
15:37	PerlJam	literal: he needs perl6 to do Great and Wonderous Things.
15:37	wayland76	literal: He needs something to learn Perl 6 with
15:38	wayland76	That's why I told him that it's not recommended
15:38	wayland76	(pugs not recommended; perl 6 is :) )
15:38		cognominal joined #perl6
15:38	patmat	right
15:38	literal	yeah, Rakudo would be a better choice
15:38	patmat	i dont need Perl or any other language for work :P but i need it, because i want to learn something!!
15:38	patmat	and to be a programmer once :)
15:39	patmat	because Support simply sucks.
15:39	moritz_	aye
15:40	literal	if you haven't programmed at all, then Perl 5 would probably be a better starting point, I think
15:40	wayland76	patmat: I had a job that involved support once. You're right that programming is better :)
15:40	patmat	no, not Perl 5..
15:41	wayland76	literal: He's done some python and ruby and QBasic, and didn't like Perl 5 for some reason :)
15:41	PerlJam	literal: don't discourage him! How often do you think someone will be able to say "Perl 6 was my first programming language"?
15:41	literal	hehe
15:41	baest	hehe, that would be cool
15:41	literal	PerlJam: too late, if what wayland76 says is true
15:41	PerlJam	yeah
15:42		Psyche^ joined #perl6
15:42	PerlJam	After using python and ruby web frameworks and then using Catalyst, I can see how they win. Syntax matters and they have less extra stuff you have to type to get work done.
15:43	PerlJam	speaking of web frameworks ... how is Web.pm coming along?
15:54		nsh- joined #perl6
15:59		justatheory joined #perl6
16:02		exodist joined #perl6
16:03	patmat	argh, cant get rakudo running on that windows machine, hrm
16:04	patmat	even git port is closed
16:05	[particle]	try git clone http://github.com/rakudo/rakudo/tree/master
16:05	arnsholt	What does the make function/operator/thing do in regexes?
16:06	[particle]	'make' generates the ast
16:07	[particle]	which can then be extracted with Match.ast
16:07	patmat	soo initializing.. wait wait
16:07	arnsholt	[particle]: Aha. Thanks. And I pass anything I want it to have as children as arguments?
16:07	patmat	download error- failed connect to github.com:80; No error
16:08		Eimi left #perl6
16:08	[particle]	arnsholt: you pass what you want to generate the ast of
16:09	[particle]	/ (\d+) { make $0 ) / # create the ast for the number matched
16:09	[particle]	/ (\d+) { make $0.sqrt ) / # create the ast for the square root of the number matched
16:10	arnsholt	Righties
16:10		cspencer joined #perl6
16:11	arnsholt	Frustratingly enough, { make $0 } causes a syntax error
16:12	[particle]	inside a regex, in rakudo?
16:12	arnsholt	Yah.
16:12	[particle]	yeah, doesn't work yet.
16:12	[particle]	you can't embed perl 6 closures in rakudo regexes yet iirc
16:12	arnsholt	Damn. That'd be a real relief
16:14		sproingie left #perl6
16:43		nsh joined #perl6
16:54		FurnaceBoy joined #perl6
16:57		cdarroch joined #perl6
17:01		revdiablo joined #perl6
17:03	revdiablo	I just noticed that unary = was killed in favor of .get() and .lines(). Any suggestions on where I can read up on the discussion that lead up to that?
17:04	revdiablo	I think it's a nice change, but it seems to have changed suddenly after sitting in the spec for a long time
17:04		azawawi joined #perl6
17:04	azawawi	hi
17:04		pmurias_ joined #perl6
17:05	literal	azawawi: perl6.vim has moved to https://github.com/hinrik/vim-perl
17:09		sri_kraih_ joined #perl6
17:13		azawawi joined #perl6
17:21	cspencer	is there any way of getting at the unicode accent info on a character? (ie. identifying that a character contains an umlaut)
17:21	cspencer	(in either PIR or P6)
17:22		nsh- joined #perl6
17:22	[particle]	revdiablo: it's in the irclog over the past few days
17:24	revdiablo	[particle]: I've been looking through there, I see discussions about the consequences of the decision... but I still wonder how the decision came in the first place
17:25	revdiablo	[particle]: It looks like it's out of the blue, but that's probably because I haven't been following well enough
17:26	[particle]	revdiablo: could have been in person at npw, i suppose
17:27	[particle]	i didn't see the discussion either, but i don't read scrollback
17:27	revdiablo	Would be interesting to know how it came up... though really I'm glad, I like lines() and get() a lot better.
17:30	[particle]	i'll miss the standard fish operator =<>
17:31	[particle]	but, that's okay.
17:31	skids	revdiablo: it's my understanding that it happened at the oslo conference.
17:32	PerlJam	=<> went away?
17:32	* skids	kinda liked the mnemonic of hotting "=" on a calculator to repeat the last operation.
17:32	[particle]	prefix:<=> is dead dead dead
17:32	skids	But yeah it was scary in some places, e.g. = =$it
17:32	* Tene	also mourns the loss of the fish operator.
17:32	PerlJam	oh! so what takes prefix:<=>'s place?
17:32	skids	.get
17:33	Tene	PerlJam: .get() and .lines()
17:33	Tene	for a single line and all lines
17:34	* skids	could have sworn he saw .end mentioned in an svn log but still its .eof and .eoi in spec.
17:34	skids	Mustabeen a hllucination
17:35	PerlJam	how does one do the while(<>) { ... } idiom in perl 6 then?
17:35	PerlJam	I don't know what the iterator is called.
17:35	* skids	suggests .eoi also be specced to return Inf for "it is known this will never stop, so don't bother asking anymore"
17:35	revdiablo	PerlJam: while (lines()) {}
17:36	PerlJam	and lines() knows to work on files from @ARGS or $IN ?
17:36	revdiablo	PerlJam: From what I understand, yeah. Let me find the spec patch, it includes that example.
17:36	[particle]	for <>.lines {}
17:37	[particle]	while is a perl5ism
17:37	PerlJam	ah, the "null iterator" still exists though? It's just prefix:<=> that's died?
17:37	[particle]	prefix:<=> is dead dead dead
17:37	PerlJam	for ''.lines { ... } seems weird :)
17:39	azawawi	interesting
17:40	PerlJam	I liked prefix:<=>, why did it die?
17:40	revdiablo	PerlJam: That's what I was trying to find out too
17:41	* [particle]	doesn't know, but hopes to raise the question in the design meeting tomorrow
17:42	revdiablo	PerlJam: btw, http://www.nntp.perl.org/group[…]/04/msg31458.html
17:42	revdiablo	PerlJam: The patch has lots of before-and-after code examples
17:43	PerlJam	revdiablo: thanks. I must've just deleted that message from my inbox without reading.
17:43	revdiablo	I started searching because I saw this, http://twitter.com/rakudoperl/status/1564922056
17:44	PerlJam	Hmm. for @$handle {...} makes me think that I could do $thingy = $$handle; too
17:45	arnsholt	Am I right in thinking that the parsing of regexes is defined only in PIR, in quote_expression.pir?
17:46	PerlJam	ah, $*ARGFILES
17:46	PerlJam	I hadn't seen that before.
17:47		LadyLunacy joined #perl6
17:47	revdiablo	Ah, $*ARGFILES is the replacement for magical <>?
17:48	revdiablo	Also I was a tad curious about the rakudo commit that pushed it past 10k passing tests
17:49	PerlJam	That's where pmichaud fixed all sort of unicode stuff.
17:50	revdiablo	All these mysterious goings-on, pleasant surprises coming out of the blue =)
17:51		LadyLunacy left #perl6
18:02		M_o_C joined #perl6
18:08		barney joined #perl6
18:09	skids	arnsholt: I think so, but pmichaud could forklift STD.pm in without much warning there, so don't expend too much effort.
18:12	* skids	adds stuff to http://www.perlfoundation.org/[…]x.cgi?perl6_tasks just to keep it full after oslo results resolve a bunch of it :-)
18:17		smtms joined #perl6
18:18		fridim_ joined #perl6
18:21		masak joined #perl6
18:22		lumi joined #perl6
18:24		fridim_ joined #perl6
18:31		hercynium joined #perl6
18:36	moritz_	can you guys see the favicon on http://perl6-projects.org/ ?
18:36	moritz_	it shows as a red box on firefox 3, but works with other browser
18:36	PerlJam	is it supposed to be camelia?
18:36	Tene	red box for me
18:36	cspencer	moritz_: i see a red box
18:36	moritz_	PerlJam: it is
18:37	skids	red box here.
18:37	literal	red box here
18:37	[particle]	ff3 red box
18:37	literal	firefox 3
18:37	PerlJam	red box here too (FF3 on linux)
18:37	* moritz_	lost
18:37	moritz_	ah
18:37	moritz_	opening it in gimp reveals three layers
18:38	moritz_	I didn't know that .icos could have it
18:38	[particle]	ie7 camelia
18:39		PhatEddy joined #perl6
18:39	moritz_	epiphany browser: camelia
18:40	PhatEddy	rakudo: my @x; print "ok\n" if defined (@x)
18:40	p6eval	rakudo 69b318: OUTPUT«ok␤»
18:40	Tene	moritz_: maybe they can't (hence the issue)
18:40	skids	flattened in gimp and it displays in ff3 now as local file.
18:40	moritz_	skids: care to check that in?
18:40	skids	Where's it at?
18:40	PhatEddy	Just noticed RT 64928 on definedness and thought my smaller example might be relevant ...
18:40	moritz_	I removed the other locally here, but then I'll loose the background
18:41	PhatEddy	You don't see the OK with p5
18:41	moritz_	skids: docs/feather/perl6-projects.org/favicon.ico
18:43		rindolf joined #perl6
18:43	pugs_svn	r26353 \| skids++ \| Flatten favicon.ico for FF3
18:43	moritz_	skids++
18:43		xinming joined #perl6
18:44	PerlJam	so ... does perl6-projects.org update automatically?
18:45	PerlJam	(i.e., If I reload, will I see the camelia favicon?)
18:45	moritz_	yes, every 10 or 15 minutes or so
18:45		rblasch joined #perl6
18:45	moritz_	but most browsers cache the favicon much longer than normal images
18:46	moritz_	so even if it's updated you might not see it
18:48	moritz_	ok, updated now
18:49	PerlJam	cool
18:50		payload joined #perl6
18:52	PhatEddy	I am vaguely wondering whether my example is the same bug or a related one requiring a new ticket referencing the existing 64928. Any thoughts?
18:52	moritz_	(everyone who helped with perl6-projects.org)++ # feedback also counts
18:54	M_o_C	also works with opera, btw
18:54	frettled	BTW, is Camelia licensed in the public domain, creative commons $foo, or something else?
18:55	M_o_C	nevermind my last message, my scrollbar ended up a few lines above...
18:55	moritz_	frettled: you have to ask TimToady, but most stuff in the pugs repo is Artistic 2, so I'd guess that's it
18:56	frettled	moritz_: ok.
18:56	* skids	hopes TimToady/perlfoundation trademark-lefts "use of the image of a butterfly in conjunction with Perl6"
18:56	alester	Is that Official?
18:57	alester	That the butterfly is the P6 animal?
18:57	moritz_	(usually I don't use images before knowning the licence before, but I couldn't imagine TimToady complaining about spreading the logo ;-)
18:57		szabgab joined #perl6
18:57	literal	probably unless/until someone comes up with a better suggestion :)
18:57	moritz_	alester: it's an official candidate, and will be the one unless somebody suggests something that Larry likes better ;-)
18:58	literal	alester: btw, I forked your vim-perl repo
18:58	frettled	moritz_: Oh, I wasn't wondering whether perl6-projects.org had permission, I thought that was given.
18:58		rindolf joined #perl6
18:58	alester	literal: Good, if you have stuff to merge back, od let me know.
18:59	alester	I just started it, but have done precious little.
18:59	moritz_	frettled: it wasn't, I just hoped I had :)
18:59	frettled	moritz_: it's covered by TimToady's policy of �better to ask forgiveness than permission� ;)�
18:59	moritz_	frettled: yes
19:01	meppl	good night
19:02		azawawi left #perl6
19:02	cspencer	moritz_: i added the "samecase" method to Any.pm last week, and I was wondering if you had any insight as to how I might do "sameaccent"
19:03	moritz_	cspencer: I implemented in perl 5, wait a sec...
19:03	cspencer	in particular, i dont know how i might get at the accent data for characters
19:03	cspencer	thanks :)
19:03	moritz_	Perl6::Str
19:04	cspencer	does that do sameaccent?
19:04	frettled	moritz_: http://howcaniexplainthis.blogspot.com/ - I shrunk it to a size suitable for various sidebars and stuff
19:04	moritz_	frettled: there's a svg version in misc/ in the pugs repo
19:05	frettled	moritz_: yeah, if I click it, I get the svg version, but svg doesn't work well in all browsers always; PNG is more web friendly.
19:05	pmichaud	hello, all
19:06	moritz_	cspencer: http://cpansearch.perl.org/src[…]/lib/Perl6/Str.pm - it relies on unicode normalization
19:06	cspencer	moritz_: found it, it wasn't in the perldocs so i looked at the source :)
19:06	pmichaud	the scariest instance of prefix:<=> that we found was reading from an embedded pod section.
19:06	pmichaud	my $line = =$=pod;
19:07	PhatEddy	rakudo: my @x; say 'ok' if @x.defined
19:07	p6eval	rakudo 69b318: OUTPUT«ok␤»
19:07	frettled	pmichaud: welcome back, looks like you had a safe trip back home.
19:07	PhatEddy	golf
19:07	pmichaud	I'm not home yet :-)
19:07	pmichaud	my flight in Newark is delayed.
19:08	frettled	ah!
19:08	pmichaud	The discussion about prefix:<=> that led to its demise took place at the Oslo hackathon, yes.
19:08	pmichaud	It was a very interesting discussion. But when it was over, TimToady++ was extremely eager to get it out of the spec. And the tests. :-)
19:08	pmichaud	Personally, I'm really glad to see it go. It was causing implementation issues.
19:08	pmichaud	(which is why it came up for discussion in the first place.)
19:08	masak	the discussion started as a discussion about laziness.
19:08	frettled	I think I walked in just when the decision had been made.
19:09	pmichaud	if people are really interested for the details, I could probably summarize the discussions in a blog post.
19:09	frettled	That was one of several times I walked by, saw something new and interesting on the whiteboard, and got to learn some more.
19:09	literal	do all iterators then have a .get now?
19:09	skids	literal: if moritz_ commits my patch :-)
19:10	frettled	pmichaud: please do, I think it will help in answering the questions from people who weren't there.
19:10	pmichaud	literal: that's not certain yet. AFAIK iterators still respond more to "shift" than to "get".
19:10	pmichaud	Perhaps they're synonymous.
19:10	moritz_	skids: or if somebody else does - I claim insufficient knowledge in that area to decide
19:10	literal	oh, that'd be cool
19:11	skids	pmichaud: is it still .eoi .eof for end testing?
19:11	pmichaud	skids: we didn't discuss those.
19:11	pmichaud	(which means no change yet. :-)
19:11	PhatEddy	The small example looks a bit like a few other tickets but seems different. Any objection to my opening a new one?
19:12	moritz_	PhatEddy: go right ahead, we can still merge them
19:12	skids	TimToady replaced prefix:<=> with .get in S07 (I think it was him) so if that can be used as an indication...
19:14	skids	(I only picked off one straggler there, the rest had all been done before I opened the file)
19:14		orafu joined #perl6
19:15	dalek	rakudo: e31a6f8 \| pmichaud++ \| src/ (2 files):
19:15	dalek	rakudo: Remove obsolete fish operator ("=<>"). It didn't work anyway.
19:15	dalek	rakudo: review: http://github.com/rakudo/rakud[…]b39238727a2dc969f
19:15	dalek	rakudo: 6ea0aa1 \| pmichaud++ \| src/classes/ (2 files):
19:15	dalek	rakudo: More fixes to Match objects.
19:15	dalek	rakudo: review: http://github.com/rakudo/rakud[…]4bc06215efb3823c1
19:16	moritz_	pmichaud: what do you mean by "it didn't work anyway"?
19:16	pmichaud	the implementation was wrong.
19:16	pmichaud	It just iterated over @*ARGS
19:17	moritz_	oh
19:17	pmichaud	so, if anyone was using it, they were using it wrongly. :-)
19:18	* literal	looks over the symbols on his keyboard. Looks like backticks and the backslash are the only ones not used by Perl 6, or am I missing something?
19:18	moritz_	literal: backslash is used
19:18	literal	what for?
19:18	moritz_	literal: backticks are left out intentionally
19:18	moritz_	literal: for captures
19:18	literal	oh
19:19	literal	not using backticks is good though, they can be hard to see
19:19	pmichaud	backslash is also "unspace"
19:19	moritz_	right
19:19	moritz_	and used iin regexes all over ;-)
19:19	literal	oh right, forgt about unspace
19:19		fridim_ joined #perl6
19:19	skids	http://www.perlfoundation.org/[…]ex.cgi?witch_hack
19:19	moritz_	pmichaud: when we bump PARROT_REVISION we can re-enable autothreading.t
19:20	moritz_	pmichaud: should I do that?
19:20		PhatEddy left #perl6
19:21	pmichaud	PARROT_REVISION will bump to the Parrot 1.1.0 release when it's done.
19:21	pmichaud	Then it has to remain there until the Rakudo release.
19:21	moritz_	ok
19:21	pmichaud	if the tests pass with a current PARROT_REVISIOn bump, it's okay with me.
19:22	pmichaud	since the parrot release is expected within the new few hours anyway (barring any problems encountered), it might just be worth waiting.
19:22	pmichaud	up to you. :-)
19:22	moritz_	I can wait for the release
19:22	pmichaud	btw, 6ea0aa1 fixes #63616 .
19:22	pmichaud	(and quite a few others :-)
19:23	moritz_	half of them by me, probably ;-)
19:24	skids	pmichaud: any thoughts on .iterator versus .Iterator on a spec level?
19:25	pmichaud	skids: not at the moment.
19:25	pmichaud	(more)
19:26	literal	.iter :P
19:26		orafu joined #perl6
19:26	moritz_	pmichaud: RT #63616 will be tested indirectly as soon as I unfudge S05-capture/caps.t - is that sufficient for you? (Match.caps and match.chunks both use the fixes)
19:26	pmichaud	moritz_: sure.
19:26	pmichaud	that's what I was hoping for.
19:27	moritz_	it seems that I have other bugs in .caps that I didn't even reach before, but the basic thing works now
19:27	pmichaud	yes, there are a few bugs in .caps
19:28	pmichaud	skids: On one hand, we have .Str for getting a Str and .Num for getting a Num
19:28	skids	And that typewise behavior is specced generally, I believe.
19:28	pmichaud	skids: On the other hand, we have .list and .hash for getting a List or Hash
19:28	moritz_	and .true for Bool
19:29	pmichaud	and .match for Match :-)
19:29	skids	well, true isn't the downcased name of a type
19:29	moritz_	I haven't seen that
19:29	pmichaud	moritz_: much of the smartmatching table is defined in terms of .match :-)
19:29	* skids	tries to swallow .match as a coerce.
19:30	moritz_	pmichaud: ah, but that's not coercing to a Match, isn't it?
19:30	pmichaud	moritz_: it would be in the same sense that .iterator is coercing to an Iterator :-)
19:31	pmichaud	I don't think of getting an Iterator as a coercion.
19:31	moritz_	you've got a point there
19:31	pmichaud	beyond that, for any given .iterator we're not guaranteed to get exactly an Iterator. We get something that "does Iterator".
19:31	pmichaud	For example, iterating a Range results in a RangeIterator
19:32	pmichaud	so, I totally agree with the notion that we should be consistent... but it's a little difficult to know exactly what that consistency is/should be.
19:33	pmichaud	Perhaps TimToady has a rule of thumb he uses to decide these things; if so, I don't know what it is.
19:34	skids	OK, well it's on the things to haunt TimToady with list now :-)
19:37	pmichaud	oh? where's that list kept? ;-)
19:38	moritz_	I started a small list in t/TASKS, but that's not the one skids is refering to, I think
19:38		orafu joined #perl6
19:39	pmichaud	well, I kept a list on the Oslo hackathon notes :-)
19:39	skids	pmichaud: grep { not is_answer_obvious } @perl6_tasks_wiki_page
19:40	pmichaud	I'm wandering off to see if I can grab a snack.
19:41	pmichaud	bbiab
19:41	pugs_svn	r26354 \| moritz++ \| [t/spec] fix and unfudge caps.t (I wrote half of that with Perl 5 in mind, it seems :/ )
19:42		ejs joined #perl6
19:44	dalek	rakudo: a76abb3 \| (Moritz Lenz)++ \| src/setting/Match.pm:
19:44	dalek	rakudo: fix .value confusion in Match.caps
19:44	dalek	rakudo: review: http://github.com/rakudo/rakud[…]9218f2c17cb370b45
19:51	skids	stronger words than "coerce": extort, pry, gouge :-)
19:52		orafu joined #perl6
19:53	cspencer	is there a .Int for getting an integer value or is that just .int?
19:53	masak	is there something like this in Perl 5 or 6? http://docs.python.org/library/time.html#time.strftime
19:53	[particle]	isn't it Int(...) ?
19:54	masak	I usually do int(...) :)
19:54	moritz_	masak: POSIX::strftime
19:54	moritz_	masak: in Perl 5
19:54	masak	moritz_: thanks.
20:03		orafu joined #perl6
20:05		lumi_ joined #perl6
20:05	cspencer	i suppose I'm more asking when a class can define an Int method to determine how it int-ifies
20:05	cspencer	s/when/whether/
20:05	moritz_	I think it can
20:06	moritz_	rakudo: class A { method Int { 3 } }; say A.new.int()
20:06	p6eval	rakudo a76abb: OUTPUT«Method 'Num' not found for invocant of class 'A'␤current instr.: 'parrot;Perl6Object;' pc 2119 (src/classes/Object.pir:719)␤»
20:06	moritz_	lol
20:06	cspencer	moritz_: it's probably a similar patch to the get_number one you'd applied to Object.pir the other day
20:06	cspencer	i can do it later on this evening
20:07	[particle]	rakudo: class A { method int { 3 } }; say A.new.int()
20:07	p6eval	rakudo a76abb: OUTPUT«3␤»
20:07	cspencer	hmmm
20:07	[particle]	Int vs int
20:07	cspencer	or there's that method too :)
20:07	Tene	rakudo: class A { method int { 3 } }; my A $a .= new(); say +$a;
20:07	p6eval	rakudo a76abb: OUTPUT«Method 'Num' not found for invocant of class 'A'␤current instr.: 'parrot;Perl6Object;' pc 2119 (src/classes/Object.pir:719)␤»
20:07	Tene	rakudo: class A { method Num { 3 } }; my A $a .= new(); say +$a;
20:07	p6eval	rakudo a76abb: OUTPUT«3␤»
20:08	moritz_	maybe we'll just leave it as-is
20:08	cspencer	alright
20:08	cspencer	i believe i had a patch to Range.pm that needed something along that lines, though I'll look at that again later too
20:09	Tene	pmichaud: can you confirm that the $() syntax for getting the AST is deprecated, and languages should update to .ast() ?
20:10	skids	In the case of int here really is both an Int and an int type.
20:11	skids	what would put the kibosh on any "lets make lowercase ones have different semantics" ideas.
20:12	masak	rakudo: /<[\-] + alpha>/
20:12	p6eval	rakudo a76abb: ( no output )
20:12	masak	rakudo: /<alpha + [\-]>/
20:12	p6eval	rakudo a76abb: OUTPUT«perl6regex parse error: Quantifier follows nothing in regex at offset 9, found ' '␤current instr.: 'parrot;PGE;Perl6Regex;parse_error' pc 10792 (compilers/pge/PGE/Perl6Regex.pir:1367)␤»
20:12	moritz_	uhm
20:12	* masak	feels he has not yet reached the bottom of the richness of S05...
20:12	moritz_	is that allowed?
20:12	masak	apparently not :P
20:12	moritz_	std: /<alpha + [\-]>/
20:12	p6eval	std 26354: OUTPUT«##### PARSE FAILED #####␤quantifier quantifies nothing at /tmp/IPdcuHMWZz line 1:␤------> /<alpha + [\-]>/␤ expecting quantmod␤FAILED 00:03 35m␤»
20:12	masak	oh.
20:13	masak	I stand corrected.
20:13	moritz_	masak: I guess whitespaces are only allowed if the beginning identifies it as a character class
20:13	moritz_	std: /<+alpha + [\-]>/
20:13	p6eval	std 26354: OUTPUT«ok 00:02 35m␤»
20:13	moritz_	std: /<-alpha + [\-]>/
20:13	p6eval	std 26354: OUTPUT«ok 00:02 35m␤»
20:13	masak	ah. there you go.
20:14	moritz_	otherwise it's parsed as <rule $args>
20:14	pmichaud	Tene: yes, confirm that $(...) becomes .ast()
20:15	pmichaud	but .ast also works now (parens not required)
20:15		orafu joined #perl6
20:16	pmichaud	it actually ends up being parsed as <rule pattern>
20:16	Tene	pmichaud: so should I update rakudo's actions.pm to .ast()?
20:16	pmichaud	Tene: if you want to do that, yes. But it might also be worthwhile to wait until after April release.
20:17	Tene	howso worthwhile?
20:17	pmichaud	I dunno -- just don't want little things to delay me :-)
20:17	pmichaud	also we'll probably prefer .ast to .ast()
20:17	moritz_	put it into a branch
20:17	Tene	pmichaud: will it delay you if I update $() to .ast ?
20:18	pmichaud	Tene: if it doesn't work or causes an issue, it might :-)
20:18	pmichaud	Tene: anyway, as long as spectests keep passing, I'm okay with it :-)
20:18	pmichaud	if it were me doing it, I'd be waiting until after the release (only because I have other things to focus on :-)
20:18	pmichaud	argggh, flight delayed even further.
20:18	masak	rakudo: "foo" ~~ /<nonesuch>/ # does RT know about this one?
20:19	p6eval	rakudo a76abb: OUTPUT«Unable to find regex 'nonesuch'␤Null PMC access in invoke()␤current instr.: 'parrot;PGE;Grammar;' pc 259 (EVAL_15:115)␤»
20:19	moritz_	masak: yes.
20:19	masak	moritz_: guessed so. thanks.
20:19	pmichaud	what did you expect it to do (differently)?
20:20	moritz_	not have a Null PMC access?
20:20	masak	aye, that.
20:20	pmichaud	but dying would be okay, yes?
20:20	masak	sure.
20:20	moritz_	sure
20:20	masak	Perl 6 is all about dying.
20:20	pmichaud	...and living to tell the tale.
20:20	Tene	okay, looks like I'm done. Testing a bit before commit.
20:21	pmichaud	..."a bit"?
20:21	Tene	;)
20:21	masak	testing if it's 1 or 0...
20:22	moritz_	btw we might have lost some test numbers on the Str.reverse -> flip renaming, and on the .get/.lines thing (because they made stuff easier, hence less to test)
20:23	pmichaud	that's okay.
20:23	moritz_	aye
20:23	moritz_	just wanted to explain some possible humps in the plot :)
20:23	pmichaud	speaking of which, I haven't updated today's plot.
20:23	* pmichaud	starts that.
20:26	pmichaud	looks like the Parrot 1.1.0 revision will be 38250 -- running spectests on that now. If that passes, we can bump PARROT_REVISION to that (and hold it there until tomorrow night).
20:27	moritz_	pmichaud: I'm also running that spectest now ;)
20:27	* masak	wants XPath for Match objects
20:27	moritz_	... with S03-junctions/autothreading.t re-enabled
20:27		jedai joined #perl6
20:27	pmichaud	moritz++
20:28		payload joined #perl6
20:28	frettled	pmichaud: BTW, do you recall that small discussion regarding representation and presentation of complex number that we had after your talk on hacking Rakudo? I said I was going to look at that during the hackathon, but I lead myself astray by actually wanting to get an overview first. I'm not quite there yet, but am I right in thinking that it should be in S02?
20:28	pmichaud	frettled: good question.
20:28	* moritz_	thinks that much that's in S02 now should be elsewhere
20:29	masak	buubot: spack complex
20:29	buubot	masak: S02-bits.pod:5 S09-data.pod:6 S22-package-format.pod:1 S29-functions.pod:8
20:29	pmichaud	Since that's where Int/Num/Str/others are listed, it would seem like the immediate place.
20:29	pmichaud	OTOH, it might go better in S32
20:29	pmichaud	or S03 (operators)
20:30	pmichaud	certainly the sqrt() stuff likely belongs in S32
20:30	frettled	Yep.
20:30	frettled	From my preliminary look-see, I thought I would have to stuff it in two, three, or possibly four different places.
20:31	pmichaud	I'd suggest trying to consolidate it into fewer places, not more.
20:31	frettled	And I think it would also be beneficial to write something general about data representation, in a separate paragraph, since that seems to be all over the place.
20:31	frettled	Yes.
20:49	* masak	frowns at the new, improved inability to introspect $/ in Rakudo
20:50	moritz_	masak: hum, what's wrong now?
20:50	frettled	masak: you liked the old, worse inability to introspect $/ better?
20:50	masak	frettled: I'm just trying to find a way through the current inability.
20:50	moritz_	Match.perl now works in more cases than before, I think
20:50	pmichaud	are there cases where it's not working?
20:51	masak	hm. maybe I'm running an old Rakudo.
20:51	pmichaud	(I can believe that there are... just curious about them)
20:51	moritz_	masak: in which case (except Perl6::Grammar) does it fail?
20:51	moritz_	same here
20:51	masak	I'll update and continue complaing afterwards. :)
20:51	pmichaud	okay. I may be boarding a plane by then, though. ;-\|
20:51	masak	the specific problems I had were with $/.hash and %($_).keys
20:51	moritz_	I still have some segfaults during make() calls
20:52	moritz_	masak: did you get Null PMC accesses?
20:52	pmichaud	%/.hash was previously broken, no question about it. So was %($_).keys
20:52	masak	moritz_: aye, on get_bool() for the former and can() for the latter.
20:52	masak	pmichaud: but in bleeding it works?
20:52	pmichaud	it works more than it used to, certainly.
20:53	masak	Rakudo is all about bleeding.
20:53	Matt-W	Evening
20:53	masak	Matt-W: o/
20:53	frettled	masak: and remember: down, not across
20:53	pmichaud	I have a significant update for Regexes in one of my local checkouts
20:53	masak	Matt-W: I was told to remind you...
20:53	Matt-W	to do some work on Form?
20:53	pmichaud	it allows us to distinguish Regex from other types of blocks
20:53	masak	Matt-W: I think so, yes.
20:53	pmichaud	but it depends on a parrot update, which means it'll have to wait until after the release.
20:54	masak	Matt-W: at least push your interesting failures so I can see them :)
20:54	Matt-W	[matthewasymptote rakudo]$ git pull
20:54	Matt-W	Updating 7a22e0d..a76abb3
20:54	Matt-W	error: Entry 'build/gen_parrot.pl' not uptodate. Cannot merge.
20:54	Matt-W	any ideas?
20:54	frettled	pmichaud: so we essentially have to bully someone into fixifying parrot soonish, then.
20:54	masak	Matt-W: dirty working copy.
20:54	Matt-W	I've done make realclean
20:54	masak	nonono.
20:54	pmichaud	frettled: no, I have the parrot fix in queue also.
20:54	masak	Matt-W: git status
20:54	Matt-W	oooh
20:54	Matt-W	I remember
20:54	Matt-W	I played with it
20:55	pmichaud	But I didn't want to push my fix to parrot just an hour before its release.
20:55	masak	thought so.
20:55	* Matt-W	feels stupid now
20:55	frettled	pmichaud: ah!
20:55	pmichaud	And I don't want to make the April Rakudo release dependent on anything but a Parrot release
20:55	masak	Matt-W: that's why it's called "git". :)
20:55	Matt-W	I should've remembered
20:55	skids	lol
20:55	Matt-W	I just assume I never edit the rakudo code, but I did...
20:56	Matt-W	now I can't remember how to unmodify it
20:56	masak	Matt-W: git stash
20:56	skids	Will loops returning Lists be waiting until after a lazy overhaul?
20:56	pmichaud	...anyone have suggestions about dealing with spectests in releases? should I bundle a copy of the current spectests into the tarball, or should I adjust the makefile to pull the correct revision of spectests?
20:56	Matt-W	that puts it off to one side somehow doesn't it
20:56	masak	Matt-W: or, if you don't want to keep the change, git checkout
20:57	pmichaud	skids: likely as part of lazy overhaul
20:57	Matt-W	mmm git stash apply
20:57	Matt-W	cunning
20:57	pmichaud	skids: or perhaps even more likely as part of implementing NEXT/LAST/REDO blocks
20:57	moritz_	pmichaud: pull a specific version
20:57	masak	Matt-W: I see you went with the former method.
20:57	frettled	pmichaud: I'd say "both please".
20:58	pmichaud	I think I'll go with "pull a specific version"
20:58	Matt-W	masak: well the patch is a hack that builds parrot in parallel so I thought I might keep it
20:58	masak	ah, nice.
20:58	frettled	I like having them in the tarball, because that means they're available offline as well.
20:59	frettled	(okay, so you could always unroll the tarball, run the appropriate make command, and reroll it if you plan on being without a connection)
21:00	masak	both my problems went away after updating Rakudo. now I have a new one. I get a StopIteration when doing $/.perl
21:00	* masak	looks at source
21:00	pmichaud	the .perl method has a big bug, yes.
21:00	* source	looks innocent
21:00	pmichaud	I saw that while looking at Match in general
21:00	pmichaud	It doesn't know how to handle quantified captures.
21:00	pmichaud	i.e., it assumes that every named/positional entry will be a Match. But sometimes they're lists of matches.
21:01	masak	right.
21:01	moritz_	I'll look into it
21:01	masak	I have such a $/.
21:01	masak	moritz_++
21:02	masak	yup, that's it.
21:02	masak	rakudo: regex f { f }; regex o { o }; "foo" ~~ /<f> <o>/; say $/.perl
21:02	p6eval	rakudo a76abb: OUTPUT«Match.new(␤ # WARNING: this is not working perl code␤ # and for debugging purposes only␤ ast => "fo",␤ text => "fo",␤ from => 0,␤ to => 2,␤ named => {␤ 'f' => Match.new(␤ ast => "f",␤ text => "f",␤ from => 0,␤ to => 1,␤ ),␤ 'o' => Match.new(␤ ast =>
21:02	p6eval	.."o",...
21:02	masak	rakudo: regex f { f }; regex o { o }; "foo" ~~ /<f> <o>+ /; say $/.perl
21:02	p6eval	rakudo a76abb: OUTPUT«StopIteration␤current instr.: 'parrot;Match;_block2802' pc 56671 (src/gen_setting.pir:10137)␤»
21:02	* masak	submits rakudobug
21:03		wknight8111 joined #perl6
21:03	pmichaud	oops, looks like my plane will board soon.
21:04	[particle]	safe flight, ttyl
21:04	masak	pmichaud: have a safe flight. watch out for snakes!
21:04	frettled	I'll sign that one.
21:04	Matt-W	remember, the windows on a plane don't open
21:04	Matt-W	so no smashing
21:04	pmichaud	moritz: if spectests pass for you using r38250, feel free to bump PARROT_REVISION
21:05	Matt-W	right, time to see if I get the same crash with latest rakudo
21:05	pmichaud	see you all tomorrow
21:05	pmichaud	(unlikely I'll be on tonight -- it's a looooong day today. Feels like it's about 11p :-)
21:05	dalek	rakudo: 7c8346d \| pmichaud++ \| docs/spectest-progress.csv:
21:05	dalek	rakudo: spectest-progress.csv update: 373 files, 10389 passing, 0 failing
21:05	dalek	rakudo: review: http://github.com/rakudo/rakud[…]e81cdc0d6621c3bcb
21:06		dolmen joined #perl6
21:07	Matt-W	woohoo, splat
21:09	frettled	pmichaud: strange thing, that ;)
21:10	Matt-W	hmm, no splat
21:11		fridim_ joined #perl6
21:11	masak	does 'given' have any guarantees about only eval'ing its topic once?
21:11	masak	say, 'given rand { ... }' or something.
21:11	Matt-W	it would be quite nice if it did
21:12	masak	Matt-W: sometimes necessary.
21:12	moritz_	masak: what do you mean by "once" - once per call?
21:12	masak	moritz_: once per given.
21:13	masak	just like 4 < $a < 10 promises only to look at $a once.
21:13	dalek	rakudo: ad73895 \| (Moritz Lenz)++ \| (2 files):
21:13	dalek	rakudo: bump PARROT_REVISION to parrot-1.1 release
21:13	dalek	rakudo: Also re-enable a test which was broken before by a parrot change
21:13	dalek	rakudo: review: http://github.com/rakudo/rakud[…]cebc24b621835ef33
21:13	literal	what software is dalek running?
21:14	moritz_	masak: so for 1..100 { given rand { ... } } should evaluate rand only once?
21:14	Matt-W	masak: form isn't crashing anymore, somoene must have already fixed it
21:14	Matt-W	masak: but it doesn't actually work right
21:14	masak	Matt-W: that's software in a nutshell, sir.
21:15	masak	moritz_: no, that's not what I meant.
21:15	Matt-W	masak: aye
21:15	masak	once per dynamic given, not once per static given.
21:16	* masak	grasps for words
21:16	moritz_	masak: for me 'given' is just a normal sub call that sets $_ for the block's caller (but it has a grammar rule)
21:16		w0lf7 joined #perl6
21:16		w0lf7 left #perl6
21:17	masak	moritz_: right, but in the case of 'given rand', it sets $_ to the value of the 'rand' call, and then leaves it alone, right?
21:17	Matt-W	oh man now I feel even stupider
21:17	Matt-W	the tests were wrong, not the algorithm
21:17	masak	that happens sometimes.
21:17	masak	that's why we have metatests. :P
21:18	moritz_	masak: sure
21:18	masak	good. I'm happy with that.
21:18	* Matt-W	pushes the changes
21:18	Matt-W	I need to decide on a consistent naming policy
21:18	Matt-W	I've got some methods_like_this and some methods-like-this
21:19	masak	tell me about it...
21:19	Matt-W	the disadvantage of using - is that vim doesn't see them as words
21:19	Matt-W	which can be handy when editing
21:19	[particle]	-++ _--
21:19	Matt-W	but _ is not as nice to read or type on an english keyboard
21:20	masak	screw vim. use -
21:21	Matt-W	it's a minor inconvenience :)
21:21	[particle]	vim can be bent to recognize - as a word char for perl 6 files
21:23	Matt-W	good point
21:23	kidd	:set isk+=-
21:23	kidd	(IIRC)
21:24	moritz_	it's not so easy
21:24	literal	that will mess with the highlighting though :P
21:24	literal	kidd: that changes \k
21:24	moritz_	- is only part of an identifier if it's between two letters
21:25	kidd	oh, so it's not so simple.. :/
21:25	[particle]	check the perl6.vim in pugs repo
21:26	moritz_	no, the one on github ;-)
21:26	literal	latest version is here now: https://github.com/hinrik/vim-perl
21:26	[particle]	ah
21:31	literal	Matt-W: how is it a disadvantage though? do you often need to highlight all identifiers or something?
21:31	Matt-W	no I just tend to use cw to replace the name of a method or something
21:31	masak	my, that's a new one: "Object must be created by a class." from PGE...
21:31	masak	what did I do? :)
21:32	Matt-W	oh if it's PGE it must be pmichaud's fault...
21:32	masak	yep.
21:32	literal	Matt-W: hm, but how is a wildcard useful in that case, when you know the name of the method?
21:32	moritz_	literal: another good use is * or # which search the current word forward/backward - it would be nice if that search for the whole sub/method name
21:32	moritz_	I use that a lot
21:32	literal	ah
21:33		hercynium joined #perl6
21:33	Matt-W	because if I typed the name of the method, I might as well just hit x that many times and switch into insert mode
21:33	literal	arguably the method name consists of more than one word if it's got a dash in it :)
21:33	Matt-W	yes but what I'm really using as 'word' in programme code is generally 'identifier'
21:34	literal	are there no quick movement commands for \i like \w is for words?
21:35	PerlJam	literal: <ident> :)
21:35	literal	though \i is sort of useless because vim discourages changing it, it might break stuff
21:35	literal	isident, that is
21:35	moritz_	literal: of course you could re-define * and # in a Perl 6 ft plugin... but I fear I lack the vim knowledge
21:37	masak	rakudo: regex o { o }; "foo" ~~ /f<o>+/; say $<o>.WHAT; say ~$<o>
21:38	p6eval	rakudo ad7389: OUTPUT«List␤2␤»
21:38	* masak	was expecting 'oo' there
21:38	* moritz_	too
21:38	* masak	submits rakudobug
21:38	Matt-W	goodnight!
21:38	masak	Matt-W: 'night
21:40	frettled	http://www.shadowcat.co.uk/blo[…]s-trout/iron-man/ - JFDI
21:44		FurnaceBoy joined #perl6
21:44	masak	I really like that.
21:45	moritz_	I've been preparing another blog post about how Perl 6 / Rakudo is great today
21:45	moritz_	but got distracted by Match.perl
21:45	moritz_	which is much more important, arguably
21:52		kate21de joined #perl6
21:52	frettled	hehe
21:52	frettled	Well, five blog lines don't cost much time.
21:53	frettled	masak: I really enjoy mst's talks/rants.
21:53	masak	I already blogged on Sunday, so I guess I'm Paper Man now.
21:53	frettled	h5!
21:53	masak	o/
21:53	frettled	\o
21:54	frettled	wait, wasn't I going to bed?
21:54	masak	whoa! midnight!
21:54	masak	that wasn't supposed to happen...
21:55	moritz_	speaking of which... I haven't managed to close in on the Match.perl bug so far :/
21:57	masak	rakudo: say %("foo" ~~ /foo/).exists("foo")
21:57	p6eval	rakudo ad7389: OUTPUT«Method 'exists' not found for invocant of class 'Hash'␤current instr.: 'parrot;P6metaclass;dispatch' pc 662 (src/classes/ClassHOW.pir:171)␤»
21:57	masak	that one's funny.
21:57	* masak	submits rakudobug
21:59	moritz_	debugging a recursive function where I have no idea on which level it fails, and why, and without proper debugging tools is... hard
21:59	frettled	g'night!
22:00	moritz_	any idea where a StopIteration comes from?
22:00	masak	moritz_: aye.
22:00	moritz_	masak: please share.
22:00	masak	just as pmichaud++ said, it occurs when a List is encountered.
22:00	masak	the algo expects a Match.
22:00	moritz_	I meant more low-level
22:01	masak	oh.
22:01	masak	no, no idea. I haven't looked at the algorithm.
22:01	moritz_	like "whenever you try to iterate $something that's not iteratable" or something
22:01	masak	well, StopIterations usually occur when elements don't add up in a for loop, so keep a watch out for that.
22:02	masak	are there any 'for ... -> $a, $b, ... {' in the code?
22:02	moritz_	no
22:02	moritz_	oh
22:02	masak	hm.
22:02	moritz_	wait
22:02	moritz_	sure there is
22:02	masak	:)
22:02	moritz_	for %(self).kv
22:02	masak	there you go.
22:03	moritz_	if .kv returns and odd sized list it's broken
22:04	masak	aye.
22:06	* moritz_	wonders if strings should be represented as trees internally
22:06	* masak	has to sleep now
22:06	masak	see y'all tomorrow!
22:06	moritz_	so that a string concatenation is a mere pointer operation
22:06	moritz_	good night
22:10	PerlJam	moritz_: you're familiar with judy-strings?
22:11	moritz_	PerlJam: no
22:11	[particle]	judy++
22:13	PerlJam	I don't have a link handy, but google for it. audrey was incorporating them into pugs at one point.
22:25		majensen joined #perl6
22:28		majensen left #perl6
22:30		ilogger2 joined #perl6
22:30	* moritz_	is close to tackling the Match.perl problem
22:31		ispy_ joined #perl6
22:38		FurnaceBoy joined #perl6
22:42		lumi joined #perl6
22:42		Limbic_Region joined #perl6
22:48	s1n	frew\|work: ping
22:50		skids joined #perl6
22:52	frew\|work	s1n: hey
22:56	moritz_	.say for @('ab' ~~ m/(.)+/)
22:57	moritz_	rakudo: .say for @('ab' ~~ m/(.)+/)
22:57	p6eval	rakudo ad7389: OUTPUT«a␤b␤»
22:57	moritz_	rakudo: 'ab' ~~ m/(.)+/; say $0.WHAT
22:57	p6eval	rakudo ad7389: OUTPUT«List␤»
22:57	moritz_	rakudo: 'ab' ~~ m/(.)+/; say $0.WHAT; say $1.WHAT
22:57	p6eval	rakudo ad7389: OUTPUT«List␤Failure␤»
23:12	dalek	rakudo: 50f6111 \| (Moritz Lenz)++ \| src/setting/Match.pm:
23:12	dalek	rakudo: handle quantified captures in Match.perl
23:12	dalek	rakudo: Actually this works only decently for named (and not positional) captures
23:12	dalek	rakudo: because of another rakudobug (RT #64952).
23:12	dalek	rakudo: review: http://github.com/rakudo/rakud[…]ed0bd04f4dec936bf
23:22	dalek	rakudo: 581f573 \| (Stephen Weeks)++ \| src/parser/actions.pm:
23:22	dalek	rakudo: Migrate actions.pm to use .ast instead of $()
23:22	dalek	rakudo: review: http://github.com/rakudo/rakud[…]29c7efbb26677bb55
23:28		[particle]1 joined #perl6
23:30		kate21de1 joined #perl6
23:39		DemoFreak joined #perl6
23:52	s1n	is there a perl6 equivalent to require?
23:55		orafu joined #perl6
23:57	skids	s1n: require is specced, S11
23:58	dalek	rakudo: 2665575 \| (Stephen Weeks)++ \| src/parser/actions.pm:
23:58	dalek	rakudo: Fix actions.pm by replacing .ast() with .ast
23:58	dalek	rakudo: review: http://github.com/rakudo/rakud[…]b0a08543a140141c2

Perl 6 | Reference Documentation | Rakudo | Niecza | Specs