<< 01-01-2015 >>

---

00:09:18	dom96	Happy new year everybody in the UK :)
00:11:50	ldlework	BlaXpirit: there are non-zero benefits to putting all the types in a single file
00:12:04	BlaXpirit	...
00:12:06	ldlework	BlaXpirit: nothing substantial to make me like it
00:12:15	ldlework	but it is kind of nice to see all the types at once..
00:12:42	ldlework	BlaXpirit: I mean other than defeating the recursive types thing :P
00:16:21	BlaXpirit	dom96, do you think a list of valid licenses for nimble could be useful?
00:16:42	dom96	BlaXpirit: yeah
00:16:48	BlaXpirit	I don't see any library licensed under zlib/libpng and I'm not entirely sure what to write in license: ""
00:17:15	BlaXpirit	dom96, i'm not suggesting that I make it :p but i could look into it
00:18:04	BlaXpirit	this is cool: http://spdx.org/licenses/
00:18:58	*	Varriount|Remote quit (Ping timeout: 246 seconds)
00:19:19	*	z1y joined #nim
00:19:30	BlaXpirit	but some packages already are different from it
00:20:06	BlaXpirit	I'm mentioning that site because Rust apparently limits licenses to this list in its package/whatever manager
00:20:47	BlaXpirit	what use is that thing in .json if you can't categorize by license etc
00:25:30	ldlework	putting all the types in one file means everything is public too :(
00:26:05	BlaXpirit	ldlework, who cares
00:26:15	ldlework	...
00:36:00	*	JStoker quit (*.net *.split)
00:36:28	*	JStoker joined #nim
00:42:49	*	Demon_Fox joined #nim
00:43:53	BlaXpirit	I don't understand how to import packages if they are not in the current directory. I've tried literally adding to PATH but it didn't help
00:44:16	Araq	--path ?
00:44:35	BlaXpirit	aaand it worked immediately. thank you.
00:44:55	jsudlow	\0/
00:45:05	BlaXpirit	it's so hard for me to find info without stackoverflow first result instant answer :p
00:46:59	def-	BlaXpirit: post all your questions to stackoverflow to help future learners of Nim?
00:47:44	BlaXpirit	i dunno
00:49:46	*	skyfex quit (Quit: (null))
00:50:40	*	gokr joined #nim
00:52:19	*	BlaXpirit quit (Quit: Quit Konversation)
00:54:20	*	gokr_ quit (Ping timeout: 250 seconds)
00:56:22	*	z1y quit (Ping timeout: 255 seconds)
01:00:53	*	gsingh93 quit (Quit: Connection closed for inactivity)
01:02:40	*	skyfex joined #nim
01:06:57	*	Jesin quit (Quit: Leaving)
01:16:00	*	BitPuffin quit (Ping timeout: 250 seconds)
01:18:29	ldlework	It would be neat if you could store int slices as consts
01:19:13	Araq	planned feature
01:19:18	ldlework	neat
01:19:23	Araq	not even hard to do
01:19:45	*	Jesin joined #nim
01:22:12	*	Varriount|Remote joined #nim
01:31:24	Varriount|Remote	There, force builds are finally fixed...
01:35:47	Demon_Fox	Just wondering, because it is annoying
01:35:54	Demon_Fox	How come nim does not use automake tools?
01:36:08	*	gokr quit (Remote host closed the connection)
01:36:16	Demon_Fox	autobuild, my bad
01:36:21	*	gokr joined #nim
01:40:49	Varriount|Remote	Demon_Fox: Uh... what?
01:42:01	Varriount|Remote	Demon_Fox: Do you mean continuous integration? Or something that rebuilds files on modification? Or do you mean http://wiki.secondlife.com/wiki/Autobuild ?
01:44:03	Demon_Fox	I mean like a makefile
01:44:27	Varriount|Remote	Demon_Fox: We don't need one?
01:45:46	Varriount|Remote	Demon_Fox: If you want to build Nim from source, you just checkout the source code repo, then checkout the csources repo, and then run one of the script files to bootstrap Nim.
01:46:03	Varriount|Remote	Then you compile koch, and bootstrap again.
01:49:25	Varriount|Remote	Demon_Fox: Nim doesn't have any dependencies, other than a working C/C++ compiler
01:50:54	Demon_Fox	I get it now
01:51:07	Demon_Fox	It was to keep the dependencies low
01:53:57	ldlework	Is there a way to echo the type of a name?
01:54:30	def-	ldlework: yes
01:55:14	def-	.eval import typetraits; name(type(12))
01:55:17	Mimbus	def-: eval.nim(4, 5) Error: invalid indentation
01:55:22	Demon_Fox	By the way
01:55:22	def-	.eval import typetraits;name(type(12))
01:55:25	Mimbus	def-: eval.nim(4, 0) Error: value of type 'string' has to be discarded
01:55:31	def-	.eval import typetraits;echo name(type(12))
01:55:35	Mimbus	def-: 0.004321813583374023sec: int
01:55:41	Demon_Fox	How many more mile stones do we need to cross to get to the next version?
01:55:47	ldlework	bleh this sucks
01:55:56	ldlework	I have a type
01:56:00	ldlework	with a field of type A
01:56:07	ldlework	which I set to a subclass
01:56:09	ldlework	type B
01:56:17	ldlework	and my methods are getting sent an A
01:56:20	ldlework	which I have to cast to B
01:56:34	ldlework	I don't like this, but this is probably inherent to statically typed languages?
01:56:36	def-	you're using method instead of proc?
01:56:40	ldlework	yes
01:56:45	ldlework	oh wait
01:57:04	ldlework	yeah it still get's send the base type
01:57:13	def-	I'd like to see the code
01:57:16	Varriount|Remote	ldlework: Care to give us an example?
01:57:25	ldlework	let me explain better first
01:57:35	ldlework	I have a type Foo, which has a field bar of type Bar
01:57:44	ldlework	I create a foo, but give it a SubBar
01:57:52	ldlework	When I call Foo.doSomething
01:58:06	ldlework	and it does self.bar
01:58:09	ldlework	it gets a Bar
01:58:20	Varriount|Remote	ldlework: Is 'Bar' a reference type, or a value type?
01:58:36	ldlework	well, all the types are value types, but all fields are ref types
01:58:46	EXetoC	the type of a name or the name of a type? :p
01:59:06	ldlework	all types are declared as '= object'
01:59:09	ldlework	all fields are ref T
01:59:47	ldlework	err
01:59:51	ldlework	'= object of RootObj'
01:59:54	ldlework	sorry
02:00:16	ldlework	let me try to come up with a minimal example
02:00:17	Varriount|Remote	ldlework: Hm, I don't suppose you could post a code sample?
02:01:07	Varriount|Remote	ldlework: You just need to be careful that, somewhere along the line, your SubBar isn't being implicitly downcast such that the extra data is being cut off.
02:01:55	ldlework	I mean, the field is defined as the base type, so that's probably likely isn't it?
02:09:30	ldlework	Varriount|Remote: https://gist.github.com/8d4fd16e63c00dce744e
02:09:37	ldlework	if you run this, this is the issue I'm running into
02:09:49	ldlework	If you try to cast, by commenting the current echos, and uncommenting the others
02:09:54	ldlework	you get a different error
02:10:13	ldlework	sorry
02:10:16	ldlework	I typed proc instead of method
02:10:21	ldlework	https://gist.github.com/c712c5228e214fe6c6b1
02:10:23	ldlework	same error though
02:12:05	def-	that looks expected to me
02:12:28	def-	why don't you just do "bar: ref Bar" and "baz: ref Baz"?
02:12:35	ldlework	should it know that
02:12:39	ldlework	f.bar is a Bar and not a Child?
02:12:58	def-	I don't think the compiler can assume that, you could just set f.bar to any other Child subtype
02:13:06	ldlework	I see
02:13:13	ldlework	fuck
02:13:23	ldlework	its so hard to divorce myself from duck type assumptions
02:13:29	ldlework	like, it is legitimately hard
02:13:39	def-	but you could have a printer method for Child, Bar and Baz
02:13:56	def-	then you can call the printer and since it's a method the correct one gets called automatically
02:14:00	ldlework	the problem is that since it interprets it as a child
02:14:08	ldlework	it Bar has no clue about Baz, ec
02:14:24	ldlework	it sees Baz as a child
02:14:30	ldlework	and therefore would print the wrong thing
02:14:42	def-	no, should work, let me try
02:14:51	Varriount|Remote	ldlework: This is what I thought you were trying to explain: https://gist.github.com/Varriount/3d473d742e88b78009b0
02:15:17	def-	Yeah, I would have done it exactly as Varriount|Remote did
02:15:26	*	gokr quit (Remote host closed the connection)
02:15:34	Varriount|Remote	And that works.
02:15:48	ldlework	they are not doing the same thing def-
02:16:05	Varriount|Remote	Actually, I was surprised that the pointer dereference didn't mess method dispatch up.
02:16:19	ldlework	they're not the same thing tho
02:16:57	ldlework	in mine, the problem is that when calling a method on the child, all parent fields are interpreted as the base type as shown in the definition
02:17:04	ldlework	but I understand why it does that now, thanks
02:17:33	ldlework	oh
02:17:41	ldlework	my problem is even more annoying
02:18:34	Varriount|Remote	ldlework: Your code confuses me. >_<
02:18:39	def-	ldlework: here's how your code works: https://gist.github.com/def-/e9fb834da751f1eac9e8
02:19:03	flaviu	sorry about the Mimbus number nonsense, it's a temporary change to show how long compilation takes.
02:19:13	ldlework	def-: I understand, it can only call the lowest common interface
02:19:15	def-	flaviu: np, first time I tried mimbus out
02:20:03	def-	flaviu: an online compiler would be nice too, is anyone working on that?
02:20:20	Varriount|Remote	ldlework, def-: You don't even have to have the methods take references, as long as you dereference the field inside the method call.
02:20:22	def-	people love to try out a language interactively without installing anything
02:20:38	def-	Varriount|Remote: yeah, true
02:20:53	flaviu	def-: Looks like ideone supports it, although their interface isn't too glamorous.
02:20:55	Varriount|Remote	def-: The hard thing about that would be abuse... It's easy to curtail bad behavior on an irc channel
02:21:04	Varriount|Remote	Much less so on a website.
02:21:20	flaviu	Varriount|Remote: Nah, it's a solved problem
02:21:28	Varriount|Remote	flaviu: Oh?
02:21:30	def-	I'm more thinking about something that you can use to make fun Nim tutorials
02:21:39	flaviu	linux containers for sandboxing and a bit of rate limiting.
02:21:51	def-	flaviu: what about infinite loops in the compiler?
02:21:58	Varriount|Remote	def-: I know, lets compile the compiler to javascript.
02:22:14	flaviu	def-: Like rust's rust-by-example?
02:22:15	def-	Varriount|Remote: yes, that would be interesting
02:22:18	def-	flaviu: exactly
02:22:25	flaviu	def-: Look at this:
02:22:26	flaviu	.eval while true; discard
02:22:28	Mimbus	flaviu: eval.nim(4, 8) Error: ':' expected
02:22:39	Varriount|Remote	def-: I don't actually know if that would work though...
02:22:45	flaviu	.eval while true:; discard
02:23:01	Varriount|Remote	?
02:23:20	Varriount|Remote	.eval echo("I'm still here")
02:23:26	flaviu	It'll time out in a minute, I couldn't easily make it shorter.
02:23:42	def-	yeah, stuff like this, coming from many ip addresses at the same time
02:23:48	Mimbus	flaviu: <no output>
02:23:51	Mimbus	Varriount|Remote: 0.004374027252197266sec: I'm still here
02:23:59	Varriount|Remote	flaviu: What about remotely stopping a job?
02:24:09	def-	Maybe I'm just used to this stuff from the gaming community, but I expect a lot of abuse
02:24:15	flaviu	remotely stopping a job?
02:24:44	Varriount|Remote	flaviu: Stopping the currently executing eval
02:24:56	flaviu	Varriount|Remote: reading from stdout is blocking
02:25:20	Varriount|Remote	flaviu: Not with threads...?
02:25:26	flaviu	def-: I keep abuse in mind at all times. There are solutions available for it all though.
02:25:29	Varriount|Remote	Or if you use non-blocking IO
02:25:49	ldlework	Okay this is my actual type relationships: https://gist.github.com/e674661ee967548d0c5c
02:26:00	ldlework	You can see how the base classes might be implemented in a library
02:26:07	ldlework	and you're implementing all the concrete stuff
02:26:08	flaviu	Varriount|Remote: Here is the code: def-: I keep abuse in mind at all times. There are solutions available for it all though.
02:26:12	flaviu	oops
02:26:13	ldlework	you're going to have to cast all over the place
02:26:17	flaviu	Varriount|Remote: https://gist.github.com/8251221aa23dde7385ba
02:27:51	flaviu	def-: see http://www.ucw.cz/moe/isolate.1.html
02:28:12	Varriount|Remote	ldlework: That looks like a problem with method dispatch..
02:28:46	ldlework	Varriount|Remote: I think it is because parent is defined as BaseParent
02:28:52	Varriount|Remote	ldlework: You might consider using dispatch tables as well, although that also requires casts...
02:29:20	ldlework	and the compiler can't upgrade any reference to .parent to a Parent
02:29:35	ldlework	so any use of .parent gives you a BaseParent that has no a or b fields
02:29:38	ldlework	this sucks
02:29:46	ldlework	for me
02:29:50	ldlework	because its how I model stuff
02:29:51	Varriount|Remote	Oh wait. Isn't it not doing what you expect because echo() doesn't look at the actual type info?
02:30:03	ldlework	no
02:30:13	ldlework	its because the parent field is declared as BaseParent
02:30:13	Varriount|Remote	echo does static dispatch, not dynamic dispatch.
02:30:48	ldlework	that isn't the reason
02:30:58	ldlework	https://gist.github.com/70f748b19eed5bafa32c
02:31:00	ldlework	same error
02:31:20	ldlework	its because the compiler does not upgrade any use of .parent
02:31:26	ldlework	to a Parent type from a BaseParent type
02:31:27	fowlmouth	ldlework, because it cant
02:31:41	ldlework	right I think I understand that
02:32:22	Varriount|Remote	Oh, yes, I see now.
02:32:24	ldlework	fowlmouth: in this situation, BaseParent is my Scene class, Parent is an actual Scene subclass in my game, and A and B are two different entities in the scene
02:33:02	ldlework	It seems you just cannot use the pattern where you organize a few things that talk to each other under a parent type, or at least in the ways I've tried
02:33:20	ldlework	when using inheritance
02:33:21	Varriount|Remote	Hm... Would C++ be able to do any better?
02:33:26	ldlework	I have no idea
02:33:46	ldlework	I don't want to change Nim, but in Python this would work perfectly
02:33:59	ldlework	So I have to evolve my own design sensibilities
02:34:01	Varriount|Remote	ldlework: You could always interface with python...
02:34:02	ldlework	but that is so painful
02:35:23	ldlework	What I wish
02:35:33	ldlework	is that I could redefine parent on the sub class maybe?
02:35:45	ldlework	so instead of casting everywhere
02:36:06	ldlework	I just inform the compiler, "so yeah, all .parent fields on this subclass are actually Parent not BaseParent"
02:37:12	fowlmouth	ldlework, parent:Parent doesnt work?
02:37:20	ldlework	huh?
02:37:36	ldlework	let val = self.parent.b.b_val
02:37:39	ldlework	is the line
02:38:53	ldlework	fowlmouth: in the subclass? no
02:38:56	ldlework	it says redefinition
02:39:20	ldlework	let val = (ref Parent)(self.parent).b.b_val
02:39:23	ldlework	fixes it
02:39:24	ldlework	but ew
02:40:15	ldlework	I wonder if I defined my BaseChild.parent to be a typeclass..?
02:40:33	ldlework	hmm no
02:41:27	*	brson_ quit (Quit: leaving)
02:45:31	ldlework	I guess this is what event systems are for
02:45:46	ldlework	so instead of calling methods directly on other stuff I send it a message
02:46:00	*	ldlework cries a little and goes off to drink
02:46:04	ldlework	happy new year everyone
02:46:38	*	z1y joined #nim
02:51:35	fowlmouth	ldlework, i got it working
02:51:43	ldlework	whatwhat
02:52:41	fowlmouth	https://gist.github.com/fowlmouth/74f52167f0e71ae81a42
02:53:31	ldlework	heh
02:53:37	ldlework	I don't exactly see what you changed
02:54:13	Varriount|Remote	ldlework: Hm, I don't remember nim having (type)(value) casting syntax... or is that conversion syntax?
02:54:34	ldlework	no idea what the difference between casting and conversion is
02:54:47	fowlmouth	Varriount|Remote, thats conversion
02:54:57	def-	conversion is type safe, casting is just interpreting the bits as a new type (totally unsafe)
02:55:00	fowlmouth	ldlework, conversion does a runtime type check
02:55:08	fowlmouth	in your case
02:55:15	ldlework	fowlmouth: it looks like you solved it for the old version
02:55:18	fowlmouth	if its int(0.1) obv there is no runtime type check
02:55:26	ldlework	the problem lies in Foo having a base class
02:55:26	Varriount|Remote	fowlmouth: What did you change?
02:55:34	ldlework	Varriount|Remote: he solved the old version
02:55:37	ldlework	with no BaseParent
02:55:45	ldlework	dunno if the solution applies to that as well
02:56:37	ldlework	fowlmouth: https://gist.github.com/anonymous/e674661ee967548d0c5c
02:57:15	ldlework	in this case the .parent field points/is declared to be a reference to the BaseParent
02:57:21	ldlework	it is Parent that gets the a/b fields
02:57:42	fowlmouth	you understand why that doesnt work right
02:58:00	ldlework	fowlmouth: yeah I just explained it
02:58:07	ldlework	the parent field is defined as BaseParent
02:58:18	ldlework	so when you are in a ChildA or ChildB and you do .parent
02:58:25	ldlework	that is a BaseParent and has no .a or .b
02:58:38	ldlework	so from ChildA you can never reach ChildB, unless you convert .parent to a Parent
02:58:50	jsudlow	happy new year germany!
02:59:36	ldlework	so in our game subsystems
02:59:45	ldlework	we're going to need to figure out a different way for them to communicate
02:59:56	fowlmouth	ldlework, ok that example is easy to fix
03:00:07	ldlework	fowlmouth: with a conversion?
03:00:18	ldlework	(ref Parent)(self.parent).b ?
03:00:19	fowlmouth	ldlework, you change line 8 to parent: ref Parent like i suggested 30 minutes ago
03:00:40	ldlework	fowlmouth: uh
03:00:49	ldlework	yes in this trivial example they are all defined together
03:01:01	ldlework	I said earlier imagine the base classes are in a library
03:01:05	ldlework	and you are implementing the subclasses
03:01:10	fowlmouth	ldlework, oh christ
03:01:26	fowlmouth	you're still on about the module limits
03:01:40	ldlework	are you serious?
03:01:45	ldlework	how is this about module limits?
03:01:52	fowlmouth	since that is the test case why dont you take the .a/.b fields out
03:02:02	ldlework	dude
03:02:13	ldlework	is this language supposed to have no libraries I don't get your point at all
03:02:36	fowlmouth	ldlework, well your design needs work if you want to do it this way
03:02:41	ldlework	is your answer to vendorize libs?
03:02:46	ldlework	fowlmouth: I've already admitted that
03:02:54	jpoirier	is there any difference between cast[bool](100) and bool(100)?
03:03:00	ldlework	I mentioned in python this would have worked fine
03:03:06	ldlework	and that I have to change the way I think about things
03:03:21	*	darkf_ joined #nim
03:04:31	Varriount|Remote	jpoirier: In the end, no.
03:05:10	Varriount|Remote	jpoirier: However, bool(100) will check that the type can actually be sensibly converted to a bool, whereas cast[bool](100) won't.
03:05:35	Varriount|Remote	Actually, wait, yes, the two will differ. The cast will work, the conversion will error.
03:05:37	*	skyfex quit (Quit: (null))
03:06:45	Varriount|Remote	ldlework: Have you looked at entity and component architectures?
03:06:53	ldlework	Varriount|Remote: very much so
03:07:09	ldlework	I am sort of implementing that without a generalized database implementation
03:07:17	jpoirier	Varriount|Remote: k, thanks
03:07:48	ldlework	where multiple game systems use game structures directly instead of having logic wrapped up in game structurs and having them talk to each other
03:07:58	*	darkf_ is now known as darkf
03:08:06	ldlework	I think I just need to externalize the state out of the systems themselves
03:11:54	Varriount|Remote	ldlework: If it helps, think about this as a challenge.
03:12:23	Varriount|Remote	And yes, I do understand the frustration that a duck-shaped hole leaves. :<
03:13:19	Varriount|Remote	You just have to... let it go. http://www.youtube.com/watch?v=moSFlvxnbgk
03:13:52	jsudlow	beat it ldlework https://www.youtube.com/watch?v=Ym0hZG-zNOk
03:14:10	EXetoC	something wrong with duck shapes?
03:27:16	flaviu	.eval bool(100)
03:27:19	Mimbus	flaviu: eval.nim(3, 4) Error: value of type 'bool' has to be discarded
03:27:26	flaviu	.eval echo bool(100)
03:27:29	Mimbus	flaviu: eval.nim(3, 9) Error: conversion from int literal(100) to bool is invalid
03:27:35	flaviu	.eval cast[bool](100)
03:27:38	Mimbus	flaviu: eval.nim(3, 0) Error: value of type 'bool' has to be discarded
03:27:38	flaviu	.eval echo cast[bool](100)
03:27:41	Mimbus	flaviu: 0.004118919372558594sec: true
03:30:47	Varriount|Remote	.eval echo(cast[string](cast[int]("Hello world")))
03:30:50	Mimbus	Varriount|Remote: 0.004240989685058594sec: Hello world
03:31:03	Varriount|Remote	Interesting.
03:31:16	Varriount|Remote	.eval echo(cast[string](cast[int]("Hello world"[2])))
03:31:19	Mimbus	Varriount|Remote: 0.006471157073974609sec: SIGSEGV: Illegal storage access. (Attempt to read from nil?)
03:31:41	Varriount|Remote	.eval echo(cast[char](cast[int]("Hello world"[2])))
03:31:45	Mimbus	Varriount|Remote: 0.004175901412963867sec: l
03:33:11	Varriount|Remote	Hm. I should probably get started on installer generation...
03:33:42	Varriount|Remote	You know, someday we should really abandon inno and nsis, and just make our own installer.
03:34:17	Varriount|Remote	It might be NIH, but dependancies really suck, and nsis scripting is... weird.
03:34:21	EXetoC	why so little precision?
03:35:04	Varriount|Remote	EXetoC: ?
03:35:33	EXetoC	jk. cya
03:36:15	flaviu	EXetoC: Because it's just temporary
03:36:20	flaviu	so I don't really care about polish
03:41:45	ldlework	jsudlow: updated and pushed
03:41:52	ldlework	I'm off to a party
03:41:54	ldlework	o/
03:42:10	jsudlow	ldlework: aweomse I"ll review it enjoy the party happy new year
03:43:16	Triplefox	cube drawing performance tests...still using glBegin and displaylists https://vine.co/v/OwFDpVb1itp
03:43:56	*	vbtt joined #nim
03:47:42	Triplefox	going to move on to the buffers and shaders approach now so that i can drive up the counts for dynamic cubes...
03:49:13	flaviu	Triplefox: The encoding on that video is pretty bad
03:50:05	Triplefox	it's a cell phone camera held up to the display...the vine aesthetic
03:50:19	flaviu	oh.
03:50:25	*	flaviu left #nim (#nim)
03:50:54	*	flaviu joined #nim
03:50:56	flaviu	oops
03:50:59	Triplefox	on some material it works well, this time it was kind of blah
03:52:43	Triplefox	i'm happy with how nim interacts with a c-style api, anyway...it's not perfectly seamless but i don't feel like i'm losing anything
03:54:13	*	Varriount|Remote quit (Ping timeout: 246 seconds)
03:57:51	*	Demos joined #nim
03:58:06	*	saml_ joined #nim
04:03:58	*	BitPuffin joined #nim
04:04:28	*	vbtt quit (Ping timeout: 255 seconds)
04:12:52	*	brson joined #nim
04:39:46	def-	Hm, this doesn't look updated: http://nim-lang.org/trmacros.html
04:45:14	*	kapil__ joined #nim
04:47:30	*	MyMind joined #nim
04:52:04	*	Sembei quit (Ping timeout: 264 seconds)
04:52:37	*	ARCADIVS joined #nim
05:05:27	*	BitPuffin quit (Ping timeout: 244 seconds)
05:16:59	*	vbtt joined #nim
05:35:43	*	z1y quit (Quit: Leaving.)
05:51:18	*	saml_ quit (Remote host closed the connection)
06:04:05	*	dts|pokeball joined #nim
06:13:17	*	vbtt quit (Ping timeout: 240 seconds)
06:20:08	*	BillsPC joined #nim
06:27:53	*	BillsPC quit (Remote host closed the connection)
06:29:28	*	BillsPC joined #nim
06:39:26	*	vbtt joined #nim
06:59:58	*	vbtt quit (Ping timeout: 255 seconds)
07:14:35	*	gour joined #nim
07:56:47	*	Demos quit (Ping timeout: 244 seconds)
08:02:04	*	BitPuffin joined #nim
08:06:55	*	BitPuffin quit (Ping timeout: 256 seconds)
08:07:25	*	vbtt joined #nim
08:22:05	*	nullmove joined #nim
08:24:24	*	darkf_ joined #nim
08:24:27	*	darkf_ quit (Changing host)
08:24:28	*	darkf_ joined #nim
08:25:43	*	darkf quit (Ping timeout: 244 seconds)
08:31:19	*	vbtt quit (Ping timeout: 255 seconds)
08:33:52	*	Demon_Fox quit (Quit: Leaving)
08:36:53	*	darkf_ is now known as darkf
08:43:45	*	brson quit (Quit: leaving)
08:55:37	*	nullmove quit (Ping timeout: 255 seconds)
08:56:26	*	nullmove joined #nim
09:15:06	*	BlaXpirit joined #nim
09:23:12	*	gour quit (Remote host closed the connection)
09:53:42	*	tgkokk joined #nim
10:00:01	BlaXpirit	this is not really important, but i wonder what is the best way to make an alias for a function
10:00:21	BlaXpirit	e.g. setCharacterSize `characterSize=`
10:01:03	dts|pokeball	let foo = someFunc # BlaXpirit ?
10:01:13	BlaXpirit	dts|pokeball, i don't think that's how it works
10:01:30	dts|pokeball	well considering ive done that before...
10:02:03	BlaXpirit	dts|pokeball, first of all, there can be multiple versions/methods
10:02:20	dts|pokeball	.eval proc foo()=echo 1;let bar = foo;bar()
10:02:24	Mimbus	dts|pokeball: 0.007239818572998047sec: 1
10:02:43	BlaXpirit	but it seems to work somehow wtf
10:02:54	dts|pokeball	because as i tried to tell you, that is how it works...
10:03:01	dts|pokeball	functions are first class like in python
10:03:28	BlaXpirit	proc foo()=echo 1;proc foo(a:int)=echo a;let bar=foo;bar()
10:03:33	BlaXpirit	.eval proc foo()=echo 1;proc foo(a:int)=echo a;let bar=foo;bar()
10:03:36	Mimbus	BlaXpirit: eval.nim(5, 4) Error: invalid type: 'None'
10:03:40	BlaXpirit	okay
10:03:56	BlaXpirit	.eval proc foo()=echo 1;proc foo(a:int)=echo a;let bar=foo;bar(5)
10:03:59	Mimbus	BlaXpirit: eval.nim(5, 4) Error: invalid type: 'None'
10:04:05	BlaXpirit	so there, dts|pokeball
10:04:29	dts|pokeball	hmmmm
10:04:30	BlaXpirit	.eval proc foo(a:int)=echo a;let bar=foo;bar(5)
10:04:33	Mimbus	BlaXpirit: 0.004065036773681641sec: 5
10:04:47	dts|pokeball	does nim have bind?
10:04:59	BlaXpirit	considering "bind" is a keyword, probably :p
10:05:14	dts|pokeball	probably not the same bind im thinking of
10:05:16	BlaXpirit	but i think it's a completely different thing
10:05:22	dts|pokeball	im thinking of c++'s bind
10:05:42	dts|pokeball	hold on, whats the type of proc?
10:05:46	BlaXpirit	oh man, livescript is so amazing in that regard
10:06:02	BlaXpirit	func _, 1, _ # bind 2 argument
10:13:01	novist	why echo $cpuTime() results in "coro.nim(66, 5) Error: type mismatch: got (proc (varargs[expr]){.gcsafe, locks: 0.}, float)"? I would expect it to print float value converted to string. No clue whats going on here..
10:13:35	BlaXpirit	novist, which module?
10:13:58	*	gour joined #nim
10:14:04	novist	cpuTime() comes from times module if you mean that
10:14:10	BlaXpirit	gawd it's impossiblt to google such things
10:14:56	novist	i noticed quite some compiler errors are rather cryptic. its not immediately clear what error means
10:15:20	BlaXpirit	novist, ok, something is wrong for sure
10:15:28	dv-	why the $? just do echo cpuTime()
10:15:48	dv-	or echo($cpuTime())
10:16:08	novist	hmm yeah w/o $ it works. second one doesnt work still, tried
10:16:15	novist	but with $ it should still work no?
10:16:21	BlaXpirit	yes
10:16:24	BlaXpirit	what's up with $
10:16:25	dv-	the second works for me
10:17:00	*	tgkokk quit (Ping timeout: 244 seconds)
10:17:02	novist	err actually indeed it does. guess i got confused by all the random things i try to get it working hehe
10:17:22	BlaXpirit	.eval var t0: float = cpuTime();echo $t0
10:17:25	Mimbus	BlaXpirit: eval.nim(3, 16) Error: undeclared identifier: 'cpuTime'
10:17:34	novist	btw same error happens if i do var t = cpuTime(); echo $t
10:17:36	BlaXpirit	.eval import times;var t0: float = cpuTime();echo $t0
10:17:36	ekarlso-	dom96: around ?
10:17:39	Mimbus	BlaXpirit: eval.nim(5, 5) Error: type mismatch: got (proc (varargs[expr]){.locks: 0.}, float)
10:17:42	novist	Mimbus import times
10:18:13	dv-	when you do echo $cpuTime() it seems to think you're passing the cpuTime proc to echo instead of cpuTime()'s return value?
10:18:34	novist	thats how it looks yes
10:18:46	dv-	it's weird like that sometimes
10:19:21	BlaXpirit	guys
10:19:28	BlaXpirit	var t0: float
10:19:41	BlaXpirit	i'm doing echo $t0
10:19:49	BlaXpirit	it's definitely something horribly bad
10:20:21	dts|pokeball	because i think its trying to convert the function itself to a string, not the return value
10:20:23	dts|pokeball	Mimbus doesnt have anything imported
10:20:25	dts|pokeball	.eval proc foo: int = result = 42;echo $foo()
10:20:28	Mimbus	dts|pokeball: eval.nim(4, 5) Error: type mismatch: got (proc (varargs[expr]){.locks: 0.}, int)
10:20:29	novist	so didnt i just score in another compiler bug..? two bugs in two days. ill get smug if i get one more hoho
10:20:46	novist	weird thing is that t0 is actually float
10:20:57	novist	and echo t0 would work just fine
10:21:03	BlaXpirit	.eval var t: float;echo $t0
10:21:06	Mimbus	BlaXpirit: eval.nim(4, 6) Error: undeclared identifier: 't0'
10:21:07	novist	but $ makes it misbehave, but echo() fixes it
10:21:08	BlaXpirit	.eval var t: float;echo $t
10:21:11	Mimbus	BlaXpirit: eval.nim(4, 5) Error: type mismatch: got (proc (varargs[expr]){.locks: 0.}, float)
10:21:17	BlaXpirit	there, minimal example
10:21:18	dts|pokeball	see, as i said, $ has higher precedence, so $some_proc() shouldnt work
10:21:24	BlaXpirit	but it's probably just lack of parentheses
10:21:35	BlaXpirit	dts|pokeball, do you not see what i wrote
10:21:36	nullmove	yeah echo($t0) works
10:21:57	*	tgkokk` joined #nim
10:22:13	novist	err.. are you guys sayng in echo $t operator $ acts on echo?
10:22:13	*	tgkokk` is now known as tgkokk
10:22:20	dts|pokeball	no
10:22:30	BlaXpirit	ok the error is obvious now
10:22:33	BlaXpirit	echo $ t
10:22:34	novist	then whats happening? i dont get it
10:22:37	BlaXpirit	no binary operator
10:22:38	BlaXpirit	yay
10:23:31	dts|pokeball	.eval var t: float = 1.0;echo $t
10:23:33	Mimbus	dts|pokeball: eval.nim(4, 5) Error: type mismatch: got (proc (varargs[expr]){.locks: 0.}, float)
10:23:36	novist	proc (varargs[expr]){.gcsafe, locks: 0.} looks much like definition of echo bw
10:23:43	BlaXpirit	dts|pokeball, that's what i wrote a few minutes ago -_-
10:23:52	dts|pokeball	BlaXpirit, no its not
10:24:04	novist	wait a moment, Mimbus is fucking bot? and i was talking to him. omg..
10:24:07	BlaXpirit	1.0 instead of 0.0 value
10:24:20	dts|pokeball	you never wrote 0.0
10:24:28	BlaXpirit	but it's implied
10:24:39	dts|pokeball	implied isnt always neccesarry a good thing
10:24:46	dts|pokeball	it can cause bugs
10:25:03	dts|pokeball	and you arent neccesary always guaranteed it to be implied
10:25:17	BlaXpirit	yes, you are
10:25:23	dts|pokeball	no
10:25:36	BlaXpirit	{.push noInit.} or what
10:25:41	dts|pokeball	im pretty sure if its decided to be taken out, it will be taken out
10:25:50	BlaXpirit	it will not be taken out
10:25:56	dts|pokeball	says who?
10:26:03	dts|pokeball	its a terrible feature
10:26:21	BlaXpirit	no
10:26:25	dts|pokeball	yes
10:26:53	BlaXpirit	in any case, if someone missed, this whole thing was not a bug
10:27:02	BlaXpirit	[:22:34] <BlaXpirit> echo $ t
10:27:19	novist	could you explain what actually happens here? why $ ruins things?
10:27:32	BlaXpirit	no binary $ operator defined
10:28:05	*	gour quit (Remote host closed the connection)
10:28:14	novist	there is operator $ that works on float. i dont get it why it breaks right here but works with ()
10:28:32	BlaXpirit	`$`(echo, t)
10:28:34	novist	unless its works much like * would in a * b
10:28:35	BlaXpirit	don't you get it
10:28:40	BlaXpirit	yes, exactly
10:29:03	novist	right.. thats totally not what one would expect. i see now
10:29:23	BlaXpirit	there are languages where this no-parentheses things works much better
10:29:39	BlaXpirit	main problem here is arbitrary operators
10:29:59	BlaXpirit	significant whitespace could help though
10:30:51	*	nullmove quit (Ping timeout: 256 seconds)
10:31:00	BlaXpirit	welp, even #! strongSpaces doesn't help in this case
10:31:02	BlaXpirit	too bad
10:31:09	novist	echo is simple proc too?
10:31:38	novist	in other langs stuff with no parentheses usually is keyword
10:31:58	BlaXpirit	you can call any procedure like tha
10:32:05	BlaXpirit	(limitations apply)
10:33:21	novist	that kind of freedom is pretty cool. bites in the ass sometimes though
10:33:43	novist	i read on wiki of 2014 GSOC. did anyone take wrapping Qt?
10:33:58	BlaXpirit	someone wrapped QML
10:33:59	BlaXpirit	...
10:34:11	BlaXpirit	i couldn't care less about it
10:34:18	novist	im not sure what that means. saw on forum and thought to myself "wtf is this shit?" :D
10:34:29	BlaXpirit	but looks like i will have to, because large parts of Qt are basically abandoned
10:34:30	dts|pokeball	hmmm good catch BlaXpirit
10:34:53	novist	maybe it could be automated by reusing some other projects
10:35:12	BlaXpirit	novist, i've gone quite far in autogenerating source code
10:35:20	BlaXpirit	but that's not what baffles me
10:35:39	novist	what is it then?
10:35:47	BlaXpirit	the whole proprocessing/macro thing
10:35:56	BlaXpirit	extensive usage of advanced c++ features
10:36:00	*	irrequietus joined #nim
10:36:14	BlaXpirit	qt is one of the hardest libraries to wrap
10:36:16	BlaXpirit	for sure
10:36:17	novist	Q_OBJECT/signals/slots?
10:36:21	BlaXpirit	yeah
10:36:24	BlaXpirit	it's all a big hack
10:36:58	novist	yeah.. that may warrant a custom nim compiler additions just to emit these
10:37:13	dts|pokeball	flaviu, tomorrow when i push v1 i will have your .give thing, except modified
10:38:16	novist	anyhow once there is anything to test consider you have a tester. i would gladly attempt to migrate my one app to nim. its in python now. and not exactly tiny so should be good testcase
10:38:43	*	jefus__ joined #nim
10:42:54	*	jefus_ quit (Ping timeout: 265 seconds)
10:43:00	novist	dts|pokeball btw now testing my coroutines thing with onlySafeCode = false - it still is not correct. printed k value gets confused and after a bit app crashes. there may be more to it than shallowCopy() problem
10:43:46	novist	https://paste2box.com/6/#/FZnWGQ/WoNIrbUckq1IrhXr3EtIN_37c244H8rgHHW3Vu2ORNE/SgugPXVV.py with output https://paste2box.com/6/#/FJnWGQ/VXyNs93fPr5i8qX-Lj7tL_MMHgVg6ZZMu_Gy3EIVMDE/STqlcjYq.txt
10:44:14	novist	notice "a1 7" printed lots, out of place
10:47:43	*	gour joined #nim
11:04:24	*	Matthias247 joined #nim
11:09:40	*	Varriount|Mobile joined #nim
11:13:42	Varriount|Mobile	.eval proc foo (s: string {.byCopy.}): string = return repr(s);var s = "Hello";echo(repr(s), " ", foo(s))
11:13:44	Mimbus	Varriount|Mobile: eval.nim(3, 22) Error: ')' expected
11:14:42	Varriount|Mobile	.eval proc foo(s: string {.byCopy.}): string = return repr(s);var s = "Hello";echo(repr(s), " ", foo(s))
11:14:44	Mimbus	Varriount|Mobile: eval.nim(3, 21) Error: ')' expected
11:15:41	BlaXpirit	well this is weird
11:15:59	Varriount|Mobile	.eval proc foo (s: {.byCopy.} string): string = return repr(s);var s = "Hello";echo(repr(s), " ", foo(s))
11:16:02	Mimbus	Varriount|Mobile: eval.nim(3, 15) Error: expression expected, but found '{.'
11:16:43	BlaXpirit	check this out though
11:16:47	BlaXpirit	.eval proc test(a: var int) = echo a;proc test(a: int) = echo a;test 5
11:16:50	Mimbus	BlaXpirit: eval.nim(5, 0) Error: ambiguous call; both eval.test(a: var int) and eval.test(a: int) match for: (int literal(5))
11:16:52	BlaXpirit	.eval proc test(a: var int) = echo a;test 5
11:16:55	Mimbus	BlaXpirit: eval.nim(4, 5) Error: for a 'var' type a variable needs to be passed
11:17:26	BlaXpirit	looks quite nasty to me
11:17:35	BlaXpirit	prevents me from doing awesome things
11:19:15	BlaXpirit	no, this is seriously a huge problem for me
11:19:24	BlaXpirit	i just realized :(
11:19:40	Varriount|Mobile	BlaXpirit: You could try fixing it in the compiler.
11:19:47	dv-	var n = 5; test n?
11:20:04	BlaXpirit	dv-, i get it but i want both a var-version and a non-var version
11:20:14	Varriount|Mobile	It's probably a 3 line fix. The challenge is finding the right 3 lines
11:20:28	BlaXpirit	i'm definitely not going for it
11:20:36	Varriount|Mobile	:<
11:20:44	BlaXpirit	if i wanna release a library, i'm not gonna go telling people "edit your compiler to run it"
11:21:01	BlaXpirit	in any case, i wrote an issue on github
11:21:04	*	fractalb joined #nim
11:21:24	Varriount|Mobile	Anyway, I could have sworn there was a way to add pragmas to arguments
11:23:52	Varriount|Mobile	.eval type valString {.byCopy.} = string;proc foo(s: valString): string = return repr(s);var s = "Hello";echo(repr(s), " ", foo(s))
11:23:55	Mimbus	Varriount|Mobile: 0.004104852676391602sec: 0x7fcbb136e050"Hello" 0x7fcbb136e050"Hello"
11:24:48	BlaXpirit	hey, by the way, maybe you guys have any ideas to make that one a template or something
11:25:09	Varriount|Mobile	(Varriount|Mobile) .eval type valString {.byCopy.} = string;proc foo(a: valString): string = return repr(a);var s = "Hello";echo(repr(s), " ", foo(s))
11:25:12	BlaXpirit	maybe then non-var would work or somethin
11:25:22	Varriount|Mobile	BlaXpirit: Huh?
11:25:41	Varriount|Mobile	.eval type valString {.byCopy.} = string;proc foo(a: valString): string = return repr(a);var s = "Hello";echo(repr(s), " ", foo(s))
11:25:44	Mimbus	Varriount|Mobile: 0.003957986831665039sec: 0x7f97c9c3b050"Hello" 0x7f97c9c3b050"Hello"
11:26:01	BlaXpirit	Varriount|Mobile, I'm looking for any workarounds to have 2 versions of functions
11:26:07	ekarlso-	hmmm
11:26:10	BlaXpirit	1 with var, 1 with non-var argument
11:26:35	dom96	ekarlso-: sup?
11:26:59	Varriount|Mobile	.eval type valString {.byCopy.} = object of string;proc foo(a: valString): string = return repr(a);var s = "Hello";echo(repr(s), " ", foo(s))
11:27:02	Mimbus	Varriount|Mobile: eval.nim(3, 35) Error: inheritance only works with non-final objects
11:27:27	*	vbtt joined #nim
11:27:33	BlaXpirit	or maybe... how can I guarantee that an argument gets passed to a function by reference
11:27:37	ekarlso-	your json example the other day
11:28:06	*	fractalb quit (Quit: Page closed)
11:28:07	ekarlso-	%{"method": %"git"} < how can I use smth like that in building up the json from a seq of stuff ?
11:28:09	BlaXpirit	(like var, but without using var, because i don't need to modify the argument)
11:28:10	ekarlso-	lemme post the code..
11:28:41	ekarlso-	http://pastebin.com/XP9pxik3 < dom96
11:28:49	ekarlso-	I wanna make a response out of the pkgs
11:31:05	Varriount|Mobile	(Varriount|Mobile) .eval type valString {.byCopy.} = object;proc foo(a: valString): string = return addr(a).toString;var s: object;echo(addr(s), " ", foo(s))
11:31:23	Varriount|Mobile	.eval type valString {.byCopy.} = object;proc foo(a: valString): string = return addr(a).toString;var s: object;echo(addr(s), " ", foo(s))
11:31:26	Mimbus	Varriount|Mobile: eval.nim(4, 47) Error: undeclared identifier: 'toString'
11:31:37	*	vbtt quit (Ping timeout: 240 seconds)
11:32:18	Varriount|Mobile	.eval type valString {.byCopy.} = object;proc foo(a: valString): pointer = return addr(a);var s: object;echo(addr(s), " ", foo(s))
11:32:21	Mimbus	Varriount|Mobile: eval.nim(4, 41) Error: expression has no address
11:33:48	Varriount|Mobile	(Varriount|Mobile) .eval type valString {.byCopy.} = object;proc foo(a: valString): pointer = return repr(a);var s: object;echo(addr(s), " ", foo(s))
11:34:02	Varriount|Mobile	.eval type valString {.byCopy.} = object;proc foo(a: valString): pointer = return repr(a);var s: object;echo(addr(s), " ", foo(s))
11:34:05	Mimbus	Varriount|Mobile: eval.nim(4, 45) Error: type mismatch: got (string) but expected 'pointer'
11:34:23	dom96	ekarlso-: var jArr = newJArray(); jArr.add(%{"method": %"git", ...})
11:34:46	ekarlso-	ah
11:35:35	Varriount|Mobile	.eval type valString {.byCopy.} = object;proc foo(a: valString): string = return repr(a);var s: object;echo(addr(s), " ", foo(s))
11:35:38	Mimbus	Varriount|Mobile: Error: internal error: GetUniqueType
11:36:00	ekarlso-	dom96: -,,-
11:36:13	Varriount|Mobile	.eval type valString {.byCopy.} = object; a: int;proc foo(a: valString): string = return repr(a);var s: object;echo(addr(s), " ", foo(s))
11:36:16	Mimbus	Varriount|Mobile: Error: internal error: GetUniqueType
11:36:27	Varriount|Mobile	I give up
11:37:18	ekarlso-	dom96: what does the %{} mean ?
11:37:45	dom96	ekarlso-: It constructs JSON
11:38:03	dom96	it creates a JObject node out of that
11:38:36	*	Varriount|Mobile quit (Quit: AndroIRC - Android IRC Client ( http://www.androirc.com ))
11:38:42	*	jux quit (Ping timeout: 250 seconds)
11:50:23	*	tgkokk quit (Read error: Connection reset by peer)
11:59:23	ekarlso-	dom96: how u do the "resp" method then ?
11:59:39	ekarlso-	resp jsArr, "application/json" ?
11:59:43	dom96	resp $jsArr
12:03:06	ekarlso-	resp jsArr, "application/json" ?
12:03:09	ekarlso-	ehm, wrong place
12:03:15	ekarlso-	resp $pkg_array, "application/json"
12:03:23	ekarlso-	I did that and now wrong nr of args on that line :
12:05:50	Araq	resp($jsArr) ?
12:06:22	Araq	Varriount_: there are no pragmas for parameters (yet)
12:06:34	ekarlso-	ah k
12:07:37	ekarlso-	hmm, %{"name": pkg.name} not possible ?
12:07:50	def-	%pkg.name maybe
12:09:05	ekarlso-	ya def- :P
12:09:16	Araq	ekarlso-: dunno what happened to the JSON macro. it's definitely possible in theory
12:09:36	ekarlso-	but what the, now it's saying my resp line is on wrong indentation
12:20:27	novist	ekarlso- wrong indentation error can happen if you mess up line before
12:25:10	ekarlso-	http://pastebin.com/bNMSTZaM < novist
12:25:14	ekarlso-	I dont see the wrongfulness there
12:25:40	BlaXpirit	Araq, tell me about that workaround please
12:27:27	Araq	BlaXpirit: proc foo(x: (var int){lvalue})
12:27:32	Araq	proc foo(x: int)
12:27:40	novist	ekarlso- i see some empty lines contain 5 spaces while indentation is 4 spaces. maybe thats the culprit?
12:28:00	novist	or maybe its just copying from pastebin
12:28:25	BlaXpirit	thank you
12:29:18	*	flaviu left #nim (#nim)
12:29:29	*	flaviu joined #nim
12:29:43	BlaXpirit	Araq, doesn't seem to compile though :o (0.10.1)
12:29:51	ekarlso-	I dont get this thing...
12:30:08	ekarlso-	doesn't have a meaning to it :|
12:30:13	BlaXpirit	Araq, sorry, it works
12:31:07	BlaXpirit	.eval proc foo (s: (string){byCopy}): string = return repr(s);var s = "Hello";echo(repr(s), " ", foo(s))
12:31:10	Mimbus	BlaXpirit: eval.nim(3, 22) Error: illformed AST: byCopy
12:32:02	BlaXpirit	Varriount_, is this what you were looking for?
12:32:07	*	Ven joined #nim
12:32:19	ekarlso-	Araq: u get it why ? :(
12:32:58	novist	ekarlso- maybe you forgot return statement on ln 41?
12:33:16	ekarlso-	novist: ehhh, should u need a return ?
12:33:25	ekarlso-	vs resp
12:33:50	novist	well you resp is obviously a call, but what is then string next to it? must put it somewhere i guess
12:33:56	novist	idk how web lib works, im very new to nim too
12:34:08	novist	but that statement does not quite make sense to me
12:34:33	novist	oh resp($pkg_array, "application/json") worked
12:35:21	ekarlso-	dang, u right!
12:35:25	*	Ven quit (Client Quit)
12:37:24	BlaXpirit	wow
12:37:37	BlaXpirit	so i've just been fighting the weirdest SIGSEGV i've ever seen
12:37:48	BlaXpirit	removing one part of the code fixed it
12:37:59	BlaXpirit	removing other unrelated part of code fixed it
12:38:06	BlaXpirit	removing import unicode fixed it
12:38:08	BlaXpirit	T_T
12:39:24	*	ARCADIVS quit (Quit: ARCADIVS)
12:40:05	def-	BlaXpirit: what was the result?
12:40:13	BlaXpirit	def-, what do you mean?
12:40:26	def-	did you find out why it segfaulted like that?
12:40:28	BlaXpirit	no
12:40:37	BlaXpirit	import unicode still consistently crashes it
12:40:56	BlaXpirit	well it's probably something i'm doing wrong, i have a bunch of ptrs
12:41:11	BlaXpirit	i bet it's just a coincidence. not importing unicode aligned the memory differently
12:43:03	*	gokr_ joined #nim
12:44:33	Araq	BlaXpirit: yup, if you don't have memory safety, spooky action from a distance is the result
12:47:25	*	gokr_ quit (Ping timeout: 256 seconds)
13:03:33	*	BitPuffin joined #nim
13:05:14	def-	https://www.reddit.com/r/programming/comments/2r06ej/what_is_special_about_nim/
13:05:17	def-	https://news.ycombinator.com/item?id=8822816
13:10:42	EXetoC	BlaXpirit: wanna try to use aur/nim-git?
13:10:48	BlaXpirit	EXetoC, sure
13:11:39	dom96	def-: Upvoted.
13:12:00	dom96	For everybody else who wants to upvote go to https://news.ycombinator.com/newest and upvote from there.
13:12:17	dom96	HN won't like it if lots of people upvote by going directly to the link
13:18:49	*	nullmove joined #nim
13:18:58	BlaXpirit	cool cool
13:19:32	*	vbtt joined #nim
13:21:18	flaviu	dom96: Asking people to vote might not be looked kindly upon by hacker news and reddit.
13:22:14	EXetoC	maybe just say "here's a new article" then
13:22:56	flaviu	yep, it's best not to mention votes at all.
13:25:10	*	vbtt quit (Ping timeout: 255 seconds)
13:27:17	*	yeye123 joined #nim
13:35:37	*	Ven joined #nim
13:43:41	*	Trustable joined #nim
13:45:03	novist	what kind of grammar is specified there? http://nim-lang.org/manual.html#grammar
13:45:34	novist	im trying to put simple intellij plugin and it requires bnf grammar. im totally clueless about grammars and parsing though :|
13:48:48	def-	Woo, top post on HN!
13:50:27	ekarlso-	dom96: what does it mean when jester goes 502 on a POST route ?
13:50:44	dom96	ekarlso-: no 'resp'?
13:51:31	ekarlso-	dom96: http://pastebin.com/55TnFmdk
13:51:34	ekarlso-	there's a resp there :/
13:52:06	ekarlso-	<html xmlns="