<< 05-01-2015 >>

---

00:01:31	*	brson joined #nim
00:08:56	*	irrequietus quit ()
00:11:58	MyMind	cool nim 0.10.2 is on homebrew finally
00:17:18	MyMind	yay https://github.com/infininight/nim.tmbundle
00:20:34	BlaXpirit_	Can someone try out or review/nitpick my CSFML bindings library? https://github.com/blaxpirit/nim-csfml
00:20:39	BlaXpirit_	Varriount_?
00:21:05	*	Epic|gmpreussner is now known as gmpreussner
00:22:03	BlaXpirit_	maybe def-
00:26:40	ekarlso-	https://bpaste.net/show/0121ed5d868b < doesn't that look valid ? it gives me a 400 though I dont see a reason why
00:29:10	MyMind	spaces instead of tabs are mandatory right?
00:29:29	renesac_	yes, for indentation
00:30:21	reactormonk	How can I interweave C code dependencies? It would depend on nim code, and other nim could would depend on the first one and the C code
00:33:04	*	wan quit (Quit: WeeChat 1.0.1)
00:35:16	ekarlso-	dom96: u around ?
00:35:30	def-	I think it's pretty cool that you can easily make a macro to make JSON work directly in Nim: https://github.com/Araq/Nim/pull/1869
00:36:45	ekarlso-	is postContent broken somehow ? the asynchttpserver is going 400 on my ass saying no content-length was given but python requests works fine
00:37:23	def-	ekarlso-: that's interesting, can you give me your code? I have an open pull request with postContent, can just add it there probably
00:37:24	*	yglukhov__ joined #nim
00:37:45	def-	(add the fix for content-length there)
00:38:04	def-	BlaXpirit_: i only have linux though
00:38:14	ekarlso-	def-: take a look at the link I pasted :)
00:38:26	ekarlso-	https://bpaste.net/show/0121ed5d868b
00:38:28	BlaXpirit_	def-, that's good, because i'm pretty sure it will only work on linux, without some minor changes
00:38:44	BlaXpirit_	i'm gonna start up that windoze virtual machine one day too
00:41:07	BlaXpirit_	also, I didn't consider that I'm going to bed soon, so I won't be able to hear your comments right away
00:41:33	*	yglukhov__ quit (Ping timeout: 244 seconds)
00:42:15	def-	ekarlso-: you probably need to add an \c\L at the and of headers
00:42:22	def-	maybe we should do that automatically
00:50:54	MyMind	adam_s: exist any yasnippet for nim?
00:53:06	*	hakand quit (Quit: Be back later ...)
00:57:52	*	Matthias247 quit (Read error: Connection reset by peer)
00:58:47	adam_s	MyMind: none that I'm aware of, but I haven't looked
00:58:58	adam_s	it would be pretty easy to write some if you wanted, though
00:59:07	adam_s	I don't really use yas
00:59:47	renesac_	http://nim-by-example.github.io/oop/ <-- this page calls multi-methods "methods" and implies they are needed for OOP, but it is only needed if you want inheritance, and in most cases it is the wrong tool
01:00:02	renesac_	flaviu: ?
01:01:32	flaviu	renesac_: It appears my ignorance is showing.
01:02:30	flaviu	I suppose I can change that, but I the words "OOP" imply inheritance for many people.
01:02:52	renesac_	actually, I'm not sure if multi-methods are needed for inheritance
01:03:08	renesac_	I never used neither in nimrod
01:03:42	renesac_	http://nim-lang.org/tut2.html#object-oriented-programming
01:03:44	Triplefox	oop is basically a "whatever i think it is" term at this point
01:04:16	BlaXpirit_	true dat
01:07:21	*	superfunc[mobile quit (Quit: Connection closed for inactivity)
01:08:22	def-	BlaXpirit_: i can't even get csfml installed^^
01:08:32	BlaXpirit_	def-, distro?
01:08:39	def-	i'm building from source
01:08:55	BlaXpirit_	uhhhhhhhhhhhhhhh okay
01:09:03	BlaXpirit_	what seems to be the problem?
01:09:10	def-	only have csfml 1.6 in gentoo strangely
01:09:15	BlaXpirit_	D:
01:10:10	def-	BlaXpirit_: https://gist.github.com/def-/561538f0a59243836fcb
01:10:49	BlaXpirit_	well i dont know gentoo at all
01:11:13	def-	that has nothing to do with gentoo, I'm just building from source and cmake is doing weird stuff
01:12:14	BlaXpirit_	lol i don't remember the last time i had to execute "make" myself
01:13:05	phira	kids these days
01:13:21	phira	in my day it was all configure;make;make install
01:13:35	def-	phira: i wish it was for sfml, don't understand this cmake error
01:13:39	BlaXpirit_	yaourt -S name here
01:13:40	phira	now it's # curl http://insecure.site.com/installer.sh | bash
01:13:46	BlaXpirit_	def-, do u even have SFML?
01:13:49	BlaXpirit_	phira, yuck
01:13:55	def-	BlaXpirit_: yeah
01:13:57	phira	BlaXpirit_: I know :/
01:14:07	BlaXpirit_	def-, I'm not sure if CSFML 2.1 will work with SFML 2.2
01:15:25	BlaXpirit_	whatever, this is not urgent :|
01:15:36	def-	BlaXpirit_: on archlinux sfml 2.2 gets installed with csfml 2.1 at least
01:15:46	BlaXpirit_	def-, yeah, and csfml doesn't work :D :D :D
01:15:51	BlaXpirit_	i had to downgrade
01:15:55	BlaXpirit_	sfml
01:16:03	renesac_	I don't know what SFML is, but I know what FML is, and it seems related
01:16:04	renesac_	:P
01:16:10	BlaXpirit_	no, not related
01:16:17	BlaXpirit_	this has gotten really offtopic, sorry
01:16:24	BlaXpirit_	and i definitely should go now.
01:16:29	renesac_	I ws being sarcastic
01:16:30	renesac_	XD
01:16:33	renesac_	*was
01:17:06	BlaXpirit_	it's not even jokingly related, because there is virtually nothing negative about it
01:17:27	renesac_	the compilation/instalation
01:17:43	renesac_	bah, forget it, it was a dumb joke
01:21:25	reactormonk	Araq, How can I interweave C code dependencies? It would depend on nim code, and other nim could would depend on the first one and the C code
01:21:51	Araq	reactormonk: .compile pragma?
01:28:20	*	z1y joined #nim
01:28:52	*	BlaXpirit_ quit (Quit: Quit Konversation)
01:32:22	reactormonk	Araq, I only see a compileTime pargma, not a compile pragma in the manual
01:32:27	*	z1y quit (Remote host closed the connection)
01:33:19	Araq	http://nim-lang.org/nimc.html#compile-pragma
01:33:25	*	z1y joined #nim
01:36:18	reactormonk	Araq, where do I put the code that the .compile pragma depends on? Above?
01:36:44	Araq	it doesn't matter, but it should be in the same module
01:39:29	*	JohnnyLee quit (Quit: Page closed)
01:42:12	*	z1y quit (Ping timeout: 264 seconds)
01:43:09	reactormonk	ok
01:43:26	reactormonk	Will try and report.
01:49:30	*	gmpreussner_ joined #nim
01:49:57	*	gmpreussner quit (Ping timeout: 244 seconds)
01:58:06	MyMind	Araq: in few hours will be official for Textmate https://github.com/infininight/nim.tmbundle
01:58:38	def-	MyMind: nice
01:59:16	MyMind	It's very basic support but will be shouted on the official twitter
01:59:26	*	VinceAddons quit (Read error: Connection reset by peer)
02:10:24	*	darkf joined #nim
02:14:18	*	z1y joined #nim
02:26:14	*	yglukhov__ joined #nim
02:27:35	*	z1y quit (Remote host closed the connection)
02:28:12	*	z1y joined #nim
02:30:23	*	yglukhov__ quit (Ping timeout: 240 seconds)
02:45:19	*	kapil__ joined #nim
02:50:10	*	yglukhov__ joined #nim
02:54:48	*	yglukhov__ quit (Ping timeout: 264 seconds)
03:04:24	*	willwillson quit (Ping timeout: 245 seconds)
03:07:43	*	z1y quit (Remote host closed the connection)
03:08:05	*	z1y joined #nim
03:18:20	*	Varriount joined #nim
03:19:56	*	Varriount quit (Client Quit)
03:20:48	*	Varriount_ is now known as Varriount
03:23:51	*	superfunc[mobile joined #nim
03:27:45	renesac_	nim still don't have any interface to write bits (rather than bytes) to a file?
03:29:11	renesac_	and when I should use the stream module rather than simply writing to a file using the procs in system?
03:32:18	EXetoC	there are bit-granularity file systems?
03:33:29	renesac_	no, you need at least one bit of buffer
03:33:47	renesac_	but there are bit-granularity data that a proc can generate
03:34:09	renesac_	*one byte of buffer
03:38:34	Varriount	renesac_: The stream module allows for multiple stream interfaces, such that you can (theoretically) write to a socket as easily as to a file.
03:39:02	EXetoC	but what should happen if a single bit is written? since you need to then store a byte
03:39:17	renesac_	EXetoC: it is written to the buffer
03:39:45	renesac_	when you close the file, it is written, followed by some padding (zeroes?)
03:39:53	EXetoC	ok
03:40:33	Varriount	renesac_: So make the char equivalent of a bit padded by some zeroes?
03:41:06	renesac_	Varriount: well, what is the difference of defining the same procs for the socket as for a file?
03:42:12	renesac_	w/o using that stream object?
03:44:29	*	brson quit (Quit: leaving)
03:47:33	*	z1y quit (Remote host closed the connection)
03:48:05	*	Boscop quit (Read error: Connection reset by peer)
03:48:11	*	z1y joined #nim
03:48:30	*	Boscop joined #nim
03:49:46	*	saltylicorice joined #nim
03:53:38	renesac_	Varriount:: I will probably use a uint8, as data with bit granularity is binary data, and I will use lots of bitwise ops to manipulate it
03:54:30	renesac_	if there is no such IO/stream library, I will write it, but I'm not sure what model I should use
03:54:49	renesac_	I don't see what value the "stream" object full of callbacks is adding
03:55:24	renesac_	instead of the regular static dispatching by the compiler
03:55:37	*	z1y quit (Ping timeout: 240 seconds)
03:58:48	Varriount	renesac_: Ask araq
04:10:51	renesac_	ok, it indeed makes easier to write extra functions that works on any type of stream
04:11:25	renesac_	but zahari's user defined type classes also should make the same thing possible
04:12:37	renesac_	just define a stream type testing if the required operations are defined for that type, and avoid that runtime overhead of calbacks
04:12:53	renesac_	I'm missing something?
04:13:33	renesac_	Varriount:: the manual says that user defined type classes are still in development, how stable are they?
04:14:43	EXetoC	they often break
04:15:35	*	z1y joined #nim
04:17:56	*	ARCADIVS joined #nim
04:19:52	renesac_	:P
04:24:55	*	dts|pokeball joined #nim
04:26:58	renesac_	oh, but I think with pure generics it may be possible to make this work, but then the functions accepting Stream will have a "T" instead of the much more descriptive "Stream" as type
04:28:32	renesac_	and the tests will happen in the function bodies, instead of the requeriments being documented on a centralized place
04:28:54	renesac_	not very pretty
04:29:17	renesac_	is there work being done to fix user defined type classes, or is it low priority?
04:31:25	EXetoC	it's not of priority for 1.0, and zahary is very busy
04:31:35	renesac_	ok
04:33:20	renesac_	but whole libraries types (like the stream one) could be rewritten taking advantage of this, aparently
04:34:43	*	boydgreenfield quit (Quit: boydgreenfield)
04:38:39	*	jefus__ joined #nim
04:40:10	renesac_	and I find strange that the buffer and buffer lenght should be given as parameters to every relevant stream operation, as oposed to be stored in the stream object as state
04:40:34	renesac_	like it is in C I think
04:40:48	renesac_	that one is for araq to answer though
04:41:10	*	renesac_ is now known as renesac|away
04:41:57	*	jefus_ quit (Ping timeout: 240 seconds)
04:45:29	renesac|away	oh, nevermind the last question, it is the lower level interface
05:04:01	*	saltylicorice quit (Ping timeout: 244 seconds)
05:05:41	*	x4nt joined #nim
05:06:01	x4nt	Does echo always end in a newline?
05:10:14	EXetoC	yes. see 'write' (stdout.write)
05:18:40	*	Sergio965 joined #nim
05:47:21	*	superfunc[mobile quit (Quit: Connection closed for inactivity)
05:49:53	*	yenic left #nim (#nim)
06:17:54	*	jpoirier quit (Ping timeout: 244 seconds)
06:23:21	reactormonk	What's the current branch? devel or master?
06:26:23	dts|pokeball	reactormonk, i think devel just merged into master
06:43:14	*	dyu joined #nim
06:51:14	*	dts|pokeball quit (Changing host)
06:51:14	*	dts|pokeball joined #nim
06:52:51	*	SplinterOfChaos quit (Ping timeout: 256 seconds)
07:16:11	*	gmpreussner__ joined #nim
07:16:57	*	gmpreussner_ quit (Ping timeout: 240 seconds)
07:39:35	*	BlaXpirit joined #nim
07:46:39	*	yglukhov__ joined #nim
08:00:31	*	gour joined #nim
08:08:38	*	Sergio965 quit (Quit: Bye! :))
08:08:59	*	Sphax joined #nim
08:27:19	*	adam_s quit (Quit: Leaving)
08:29:11	*	dts|pokeball is now known as blackmist
08:29:37	*	blackmist is now known as dts|pokeball
08:29:57	*	hakand joined #nim
08:49:02	*	dyu quit (Ping timeout: 245 seconds)
08:51:53	*	dyu joined #nim
09:00:18	*	z1y quit (Remote host closed the connection)
09:00:51	*	z1y joined #nim
09:05:03	*	Sphax quit (Quit: ZZZZZzzzzz)
09:11:53	ekarlso-	yet what is it that will replace parseopt ?
09:14:51	*	goobles joined #nim
09:16:02	*	Sphax joined #nim
09:22:07	*	lyro joined #nim
09:36:01	ekarlso-	can a exception be used in a case statement ?
09:36:22	*	zahary1 quit (Quit: Leaving.)
09:36:42	dts|pokeball	try it and see?
09:36:47	ekarlso-	packages.nim(247, 12) Error: selector must be of an ordinal type, float or string
09:36:50	ekarlso-	meh -,,-
09:36:54	ekarlso-	if statement then I guess
09:42:11	*	yglukhov___ joined #nim
09:42:11	*	yglukhov__ quit (Read error: Connection reset by peer)
09:48:57	*	hakand quit (Quit: Be back later ...)
09:51:05	*	repax joined #nim
09:51:48	*	hakand joined #nim
09:53:16	*	superfunc quit (Ping timeout: 246 seconds)
10:00:13	*	z1y quit (Remote host closed the connection)
10:01:24	*	z1y joined #nim
10:24:35	*	ekarlso- quit (Changing host)
10:24:35	*	ekarlso- joined #nim
10:33:03	*	gokr quit (Quit: Leaving.)
10:40:45	*	gokr joined #nim
10:40:57	*	Matthias247 joined #nim
10:41:17	*	gokr_ quit (Ping timeout: 240 seconds)
10:42:17	*	Sphax quit (Quit: ZZZZZzzzzz)
10:42:49	*	VinceAddons joined #nim
10:42:50	*	Sphax joined #nim
10:55:23	*	z1y quit (Ping timeout: 256 seconds)
11:00:16	*	gokr1 joined #nim
11:00:19	*	gokr1 quit (Read error: Connection reset by peer)
11:01:33	*	kapil__ quit (Quit: Connection closed for inactivity)
11:01:44	*	gokr1 joined #nim
11:05:12	*	willwillson joined #nim
11:05:38	*	bjz joined #nim
11:08:40	*	milosn quit (Remote host closed the connection)
11:09:34	*	milosn joined #nim
11:20:39	*	Sphax quit (Quit: ZZZZZzzzzz)
11:32:06	*	Epic|gmpreussner joined #nim
11:32:22	*	gmpreussner__ quit (Ping timeout: 245 seconds)
11:36:36	*	VinceAddons quit (*.net *.split)
11:36:36	*	x4nt quit (*.net *.split)
11:36:36	*	MightyJoe quit (*.net *.split)
11:38:20	*	MightyJoe joined #nim
11:38:29	*	xant1 joined #nim
11:39:52	*	Forty-3 quit (Ping timeout: 259 seconds)
11:40:49	*	Matthias247 quit (Read error: Connection reset by peer)
11:40:55	*	bjz quit (Read error: Connection reset by peer)
11:46:22	*	hakand quit (Quit: Be back later ...)
11:47:21	*	bjz joined #nim
11:47:27	*	soggybread quit (Quit: leaving)
11:48:00	*	hakand joined #nim
11:56:46	*	gokr1 quit (Quit: Leaving.)
12:08:30	*	z1y joined #nim
12:09:58	*	bjz quit (Read error: Connection reset by peer)
12:10:06	*	bjz_ joined #nim
12:11:33	*	repax_ joined #nim
12:14:27	*	z1y quit (Quit: Leaving.)
12:14:49	*	repax quit (Ping timeout: 245 seconds)
12:20:03	*	Sphax joined #nim
12:20:20	*	repax_ quit (Quit: repax_)
12:26:47	*	z1y joined #nim
12:33:17	*	z1y quit (Remote host closed the connection)
12:42:20	*	bjz joined #nim
12:42:43	*	bjz_ quit (Read error: Connection reset by peer)
12:46:42	*	Matthias247 joined #nim
12:50:30	*	bjz quit (Ping timeout: 250 seconds)
12:51:13	*	bjz joined #nim
12:59:16	*	irrequietus joined #nim
12:59:29	*	irrequietus quit (Changing host)
12:59:29	*	irrequietus joined #nim
13:05:42	*	z1y joined #nim
13:09:23	*	bjz quit (Ping timeout: 240 seconds)
13:14:02	*	jefus__ is now known as jefus
13:16:30	*	VinceAddons joined #nim
13:16:54	*	gokr quit (Ping timeout: 245 seconds)
13:18:01	*	bjz joined #nim
13:22:14	*	bjz quit (Read error: Connection reset by peer)
13:24:06	joebo	It looks like asynchttpserver is single threaded. Are there any multi-threaded versions or other ways of enabling it to use all the cores on a machine, aside from having 8 processes listening on different ports and having a reverse proxy round-robin it
13:26:32	*	Trustable joined #nim
13:26:48	acidx	joebo: not sure about technicalities regarding num, but if your operating system supports it, you can enable SO_REUSEPORT on the main socket, so that various processes can bind on the same port. the kernel will then schedule the connections.
13:27:41	joebo	true
13:28:18	Araq	joebo: we still haven't tested async with spawn :-)
13:28:21	joebo	I was hoping I there was an existing solution that used spawn
13:28:43	joebo	Araq: Thanks, I figured... it's still quite new it seems
13:28:56	Araq	both features are bleeding edge
13:29:17	Araq	spawn works with classical sockets though
13:30:39	*	bjz joined #nim
13:32:01	joebo	ok... the async/await pattern looks really nice
13:32:09	joebo	very clean code
13:32:26	Araq	that said, async has still performance problems, I think and so is slower than the boring old sockets, so spawn + old sockets might be the winner
13:32:46	joebo	I may try spawning from the asynchttpserver callback and see what happens
13:33:28	joebo	Araq: are you aware of any server examples using boring old sockets ? I can go check rosettacode
13:35:56	joebo	http://goran.krampe.se/2014/10/25/nim-socketserver/
13:37:24	Araq	yeah that one uses a non-optimized deepCopy of the socket's buffer
13:37:32	Araq	but apparently performance is acceptable
13:39:21	joebo	hmm, it still only used one core on windows (100% cpu on one core) even with --threads:on
13:39:49	*	bjz quit (Ping timeout: 245 seconds)
13:40:56	joebo	but maybe the response callback is so quick that it still keeps using these same thread/core
13:55:55	*	kapil__ joined #nim
13:56:07	joebo	that seemed to be it. I added a more CPU intensive workload to the spawned callback and it started using more cores
13:56:32	*	bjz joined #nim
13:56:48	Araq	good
13:57:25	joebo	nim is really impressive. thank you
14:00:57	*	ramnes joined #nim
14:01:11	*	Matthias247 quit (Read error: Connection reset by peer)
14:06:20	BlaXpirit	so destructors is a new thing? :o
14:07:22	BlaXpirit	or has it been experimental for long?
14:08:32	*	dirkk0 joined #nim
14:09:16	ramnes	hi
14:09:34	*	gokr joined #nim
14:09:50	ramnes	why didn't I heard about nim before and why so much hype now?
14:10:10	*	bjz quit (Ping timeout: 255 seconds)
14:10:52	gokr	ramnes: Because it was recently renamed to Nim - and because of the hype around Rust I suspect (where Nim pops up as a similar lang)
14:11:27	ramnes	yeah I know it just has changed it's name, but I never heard about nimrod neither
14:11:38	gokr	Me neither, until like a month or two ago
14:11:48	ramnes	and Rust is hype for monthes now
14:12:00	gokr	I have been cranking out articles btw, http://goran.krampe.se/category/nim
14:12:06	gokr	Might be of help if you are learning.
14:12:10	*	hakand quit (Quit: Be back later ...)
14:12:33	ramnes	nice, thanks gokr
14:13:09	ramnes	any serious project on Github to see some nice code?
14:13:11	*	bjz joined #nim
14:14:23	joebo	gokr: your socket server example was very help. thanks
14:14:31	gokr	Good!
14:15:12	gokr	I am all for slick async stuff - but a "regular" forking server is also very useful since then you can rely on blocking code if you need to.
14:15:34	gokr	For an async server all code handling the requests will need to be async.
14:16:04	*	hakand joined #nim
14:16:47	gokr	Its also worth noting that the "spawn" uses the thread pool that Andreas has written - so it doesn't actually create a new thread for each request.
14:17:34	gokr	ramnes: I think the forum code, Aporia etc are some of the public examples.
14:17:37	*	z1y quit (Ping timeout: 264 seconds)
14:18:12	gokr	I wrote a little utility called blimp that makes it easy to handle big binary files in git. So it may be an interesting example of a command line utility.
14:19:17	*	dirkk0 quit (Quit: Leaving)
14:19:34	dv-	nimble is nice too
14:20:12	*	z1y joined #nim
14:21:07	*	Epic|gmpreussner quit (Ping timeout: 245 seconds)
14:22:04	*	z1y quit (Remote host closed the connection)
14:22:19	*	z1y joined #nim
14:23:01	*	Epic|gmpreussner joined #nim
14:23:33	*	bjz quit (Read error: Connection reset by peer)
14:23:52	*	Epic|gmpreussner is now known as gmpreussner
14:24:58	*	gmpreussner quit (Client Quit)
14:25:43	*	bjz joined #nim
14:31:48	ramnes	why mixedCase for functions? :(
14:32:31	gokr	Why not?
14:33:27	ramnes	because snake_case is easier to read
14:34:18	BlaXpirit	ramnes, at least you don't have to write mixedCase in nim
14:34:48	*	hakand quit (Quit: Be back later ...)
14:35:05	ramnes	yeah but it looks like it's already the defacto standard
14:35:11	flaviu	.give ramnes let fooBar = "123"; echo foo_bar
14:35:16	Mimbus	ramnes: 123
14:35:34	ramnes	wat
14:36:49	gokr	ramnes: "is easier to read" - says who? Its quite odd to argue like that given the amount of people thinking otherwise.
14:37:09	*	hakand joined #nim
14:38:03	BlaXpirit	mixedCase is about as easy to read as mixed_c_ase
14:38:26	ramnes	gokr: see http://programmers.stackexchange.com/questions/27264/naming-conventions-camelcase-versus-underscore-case-what-are-your-thoughts-ab
14:38:32	ramnes	aRatherLongSymbolName vs a_rather_long_symbol_name
14:38:35	ramnes	seems fair to me
14:39:02	flaviu	arguing about identifier style is unlikely to be productive
14:39:11	ramnes	I know
14:39:49	*	bjz quit (Ping timeout: 245 seconds)
14:39:50	ramnes	but this is a big part of a language standards design
14:40:09	ramnes	so I guess there is some reasons about mixedCase being used
14:40:28	ramnes	hence my question
14:40:59	gokr	ramnes: Reading a bit in that link you posted, its definitely not a black/white question.
14:41:01	flaviu	The language is designed so that you can use whatever style you like.
14:41:17	gokr	I am a Smalltalker at heart, camelCase since 1980s :)
14:42:05	ramnes	flaviu: you know that "do what you want" never makes readable code :)
14:42:32	flaviu	ramnes: Of course, your personal style should stay consistent.
14:42:39	gokr	Also, IMHO you can't argue that a single identifier is easier/harder to read - its the code as a whole that is interesting. And then, I (humble opinion) tend to think that camelCase holds the identifiers "together" better.
14:43:15	gokr	But either way, I just reacted to your view that there was some sort of "fact" that one was superior than another :)
14:43:42	flaviu	ramnes: But the idea is that if some library author decides to do stuff like `underscores_everywhere` and you prefer camelCase, you can keep your code consistant
14:44:42	BlaXpirit	flaviu, and that author will be looked down upon -_-
14:45:01	ramnes	I really like the PEP about naming styles
14:45:33	ramnes	and in my question, I was actually wondering if there is such a document for nim
14:45:43	ramnes	not really saying that snake_case is superior
14:46:18	ramnes	like I said, I'm just asking a question, don't really want to go in an infinite debate :)
14:46:55	flaviu	https://github.com/Araq/Nim/wiki/NEP-1-%3A-Style-Guide-for-Nimrod-Code
14:47:11	*	repax joined #nim
14:47:30	flaviu	It's intended for the compiler and standard library, but there's no reason that it can't be used elsewere.
14:47:46	ramnes	nice
14:48:28	*	jpoirier joined #nim
14:49:10	*	jpoirier quit (Client Quit)
14:52:22	ramnes	it's funny how it nim feels heavily inspired by Python, like NEP instead of PEP
14:52:51	ramnes	but still change things without explaining, like that naming convention
14:53:27	ramnes	(and I don't mean to be harsh or to troll, seriously)
14:53:50	ramnes	s/it//
14:54:08	BlaXpirit	it's harder to find a language it was not inspired by
14:54:52	ramnes	well, this might be biased since I'm a Python developer :p
14:54:52	repax	There have been much discussion on naming
14:56:00	repax	The steps away from T and P prefixes were very welcome
14:57:05	BlaXpirit	and instead we get even longer suffixes -_-
14:57:21	*	willwillson quit (Ping timeout: 244 seconds)
14:57:28	flaviu	BlaXpirit: IMO, you shouldn't have to use a suffix
14:57:34	ramnes	:D
14:57:47	flaviu	Decide if you want your type to be a ref or a value, and stick with it.
14:58:01	BlaXpirit	i have been able to avoid it quite successfully lately, but i'm afraid it won't always be the case
14:59:35	*	bjz joined #nim
14:59:36	ramnes	for what kind of project would you absolutely use nim instead of an other language?
14:59:39	flaviu	BlaXpirit: Well, many languages get along fine treating everything as a ref.
15:00:08	BlaXpirit	(and i ended up looking for an alternative to such a language...)
15:01:10	repax	Why not simply use the ptr and ref keywords?
15:01:53	repax	As well as `var`
15:02:35	repax	The need for specific ref prefixes are very uncommon
15:03:08	flaviu	repax: Because `new` is annoying to use, and you can't use the type construction syntax with prefixes.
15:03:20	repax	ramnes, lately, wherever I can
15:03:38	EXetoC	flaviu: you mean, without them? well you can actually
15:03:47	flaviu	That's why I wrote https://gist.github.com/flaviut/1ae1d7deac4c57d0a84c
15:03:51	EXetoC	with another parenthesis pair
15:04:24	flaviu	.give EXetoC type Typ = object; val: int; discard (ref Typ)(val : 1)
15:04:27	Mimbus	EXetoC: <no output>
15:04:33	flaviu	wait, what?
15:04:43	EXetoC	:-p
15:04:45	flaviu	huh, that must have been implemented recently.
15:05:13	flaviu	I still like box better though
15:08:03	EXetoC	BlaXpirit: there are so many other aspects to a language though - major ones that matter
15:08:35	BlaXpirit	it's not the prefixes themselves, it's the confusion that they cause
15:08:58	flaviu	.eval "1.2.3.4.5.6".split('.').map(strip)
15:09:01	Mimbus	flaviu: eval.nim(3, 24) Error: type mismatch: got (seq[string], proc (string, bool, bool): string{.noSideEffect.})
15:09:11	flaviu	but the last two bools are optional.
15:09:40	BlaXpirit	bugs are everywhere
15:10:02	*	bjz quit (Ping timeout: 250 seconds)
15:11:21	*	saml joined #nim
15:11:27	EXetoC	BlaXpirit: there are many other things a newcomer has to learn
15:11:44	EXetoC	what do you think about Obj/ObjRef/ObjPtr?
15:12:04	BlaXpirit	EXetoC, yeah, there are many undocumented things a newcomer has to learn
15:12:04	EXetoC	*Foo
15:12:15	*	vqrs joined #nim
15:12:24	EXetoC	and then FooObj
15:12:43	BlaXpirit	what do you mean, what I think about them
15:12:45	EXetoC	that's a temporary issue, but ok good point
15:13:17	EXetoC	BlaXpirit: yes, that's the replacement. the primary type would just be Foo, and then you would introduce some of the others if necessary
15:13:42	BlaXpirit	how do I know if it's necessary?
15:14:00	BlaXpirit	what if i make a library and think it is not necessary to make the object type public
15:14:36	BlaXpirit	(which is exactly what i did)
15:15:38	flaviu	BlaXpirit: Then I suppose your users will have to stop worrying about that small inefficiency.
15:16:08	EXetoC	yes, you can either add them later, or the user can, or the user can just do (ref Foo)
15:16:15	BlaXpirit	nope
15:16:36	BlaXpirit	i mean, the first part - yes, the rest - nope
15:17:19	flaviu	Has anyone written a dynamic typing library?
15:17:33	EXetoC	you'd like to avoid ref Foo at all costs?
15:17:49	repax	Is this really a problem with nim? No matter what language you use you have to consider the boundary between API and implementation
15:17:51	BlaXpirit	EXetoC, you misunderstood... i wasn't clear enough
15:18:09	BlaXpirit	I'd like to avoid having these different types at all costs
15:18:10	EXetoC	repax: right
15:18:33	vqrs	Is anyone using nim for JavaScript stuff?
15:18:35	BlaXpirit	which is why i enforce object-type-only for some and ref-type-only for others
15:18:37	EXetoC	BlaXpirit: then do so if you choose. I'm just saying that it's new convention
15:19:08	flaviu	vqrs: only project I know of is the package list on the nim website.
15:19:19	BlaXpirit	but I have a scarier problem at hand
15:19:22	BlaXpirit	"usage of a type with a destructor in a non destructible context"
15:19:43	EXetoC	BlaXpirit: I don't know how that approach is any different
15:19:50	vqrs	flaviu
15:19:56	vqrs	flaviu How do you mean?
15:20:20	BlaXpirit	nim-lang/packages
15:20:32	BlaXpirit	no, that's not it
15:20:46	flaviu	vqrs: http://nim-lang.org/nimblepkglist.js
15:21:10	flaviu	it runs on this page: http://nim-lang.org/lib.html
15:21:12	*	bjz joined #nim
15:21:13	BlaXpirit	don't scare people like that
15:21:14	EXetoC	BlaXpirit: you mean, because the basic type would then sometimes not be exposed? then maybe it would be a good idea to always define it
15:21:32	vqrs	I see
15:21:34	BlaXpirit	EXetoC, so it's just a ton of boilerplate
15:22:02	*	tane joined #nim
15:22:42	EXetoC	maybe. I don't think many people have adhered to this convention yet
15:23:32	BlaXpirit	I think a JS minifier is a must-have in any nim-js pipeline
15:24:32	repax	Sounds useful
15:25:28	BlaXpirit	http://marijnhaverbeke.nl/uglifyjs http://nim-lang.org/nimblepkglist.js result is 53.6% of original
15:25:50	BlaXpirit	nobody's gonna look at it anyway
15:26:12	BlaXpirit	i mean actively read it
15:26:27	flaviu	It could be even smaller, if a more advanced compiler was used
15:26:49	BlaXpirit	lol
15:26:51	flaviu	there's lots of stuff in the global scope that doesn't need to have long identifiers.
15:26:53	BlaXpirit	it just pollutes namespace
15:26:54	onionhammer	i would use closure compiler
15:26:56	onionhammer	too
15:26:58	BlaXpirit	that's compiler's fault
15:26:58	onionhammer	see which gives smaller results
15:27:04	onionhammer	https://developers.google.com/closure/compiler/
15:27:05	BlaXpirit	it should be wrapped in anonymous function
15:27:10	flaviu	I can't get closure compiler to work.
15:27:14	onionhammer	http://closure-compiler.appspot.com/home
15:27:17	flaviu	Compilation did not complete successfully. See warnings pane for details.
15:27:37	vqrs	I was looking for a powerful language for a single page application, and Nim looks nice, but http://nim-lang.org/backends.html#the-javascript-target makes it sound like it's not a very good idea
15:27:49	onionhammer	for advanced it didnt, hm
15:28:33	onionhammer	doesnt look like valid js to me..
15:29:03	onionhammer	whoa.. did not know JS had labels
15:29:05	onionhammer	thats crazy ;)
15:29:12	BlaXpirit	no goto though
15:29:24	onionhammer	yeah
15:29:33	onionhammer	thats where the closure compiler is hanging up
15:30:09	EXetoC	vqrs: what do you think is missing?
15:30:18	BlaXpirit	i don't understand that code
15:30:25	BlaXpirit	there is a bunch of function definition
15:30:28	BlaXpirit	but where is "main"?
15:30:51	vqrs	EXetoC:It sounds like I can't make use of virtually the entire standard library.
15:32:04	BlaXpirit	the main question is, can js libraries be used
15:32:23	flaviu	BlaXpirit: All the code that is not in a function is main
15:32:38	vqrs	That, too. The page only shows a rudimentary funciton call to JS, not however whether the JS object model can be accessed from within Nim
15:32:41	BlaXpirit	flaviu, it's hard to tell because of no indentation xD
15:32:47	*	z1y quit (Ping timeout: 265 seconds)
15:33:28	BlaXpirit	vqrs, I can't find even that
15:33:35	BlaXpirit	pls link to "rudimentary funciton call to JS"
15:33:45	vqrs	BlaXpirit: http://nim-lang.org/backends.html#javascript-invocation-example
15:33:52	kokozedman	hey guys... how do I do a C : return *((uint32_t*)(addr)) in nim? addr is an arbitrary pointer
15:33:53	BlaXpirit	thank you
15:34:08	BlaXpirit	cast[ptr uint32] i think
15:34:24	BlaXpirit	cast[ptr uint32](addr)[] i think
15:34:25	gokr	vqrs: Amber smalltalk is quite cool (one page app)
15:34:36	kokozedman	BlaXpirit: ok, thanks
15:35:16	BlaXpirit	ah, vqrs
15:35:21	EXetoC	it's just that addr is a keyword
15:35:27	vqrs	gokr: Hmm the overview sounds nice
15:35:31	BlaXpirit	so you have to predefine all functions and it has to be static
15:35:36	BlaXpirit	nope nope, not usable for js, sadlt
15:35:39	BlaXpirit	sadly
15:35:48	gokr	vqrs: I know Nicolas Petton who created Amber. Its used for www.smalltalkhub.com
15:35:58	BlaXpirit	not that i would want to confine myself to a compiled language if it's all gonna perform even worse than default
15:36:15	gokr	Its a very clean Smalltalk and the cool part is that the IDE runs "live" in your browser
15:36:24	gokr	So you can code as you go.
15:37:14	flaviu	Well, I got rid of all the code it kept whining about, but it still fails
15:37:33	BlaXpirit	whatever
15:37:50	vqrs	Well, I'm going to write a compiler that runs in the browser, and visualizes its execution step by step, and for correctness' sake I wanted some more bells and whistles than JS has... I looked at TypeScript so far and that didn't look too shabby
15:38:01	BlaXpirit	the js target really, really, really should output everything wrapped in a function+call
15:38:03	*	bjz quit (Read error: Connection reset by peer)
15:38:29	BlaXpirit	guys, the best thing that tries to hide JS's horrors is LiveScript
15:38:33	BlaXpirit	please just trust me
15:38:33	gokr	vqrs: Do look at Amber. Its already a compiler that runs in a browser :)
15:39:00	EXetoC	some people seem to have nothing but praise for js
15:39:11	EXetoC	but that's true for everything I guess
15:39:13	gokr	vqrs: Check this: http://amber-lang.net/learn.html
15:39:17	gokr	Interactive tutorial.
15:39:29	vqrs	gokr: I will, the language is set to be Eiffel though, it's a Bachelor's Thesis
15:40:00	gokr	vqrs: Either case, the Amber compiler is a full multistage compiler with an IR etc, so ... it should be useful to study.
15:40:10	BlaXpirit	does the "Got 5 minutes?" part refer to the time it takes to load?
15:40:11	gokr	And its written in itself.
15:40:18	vqrs	BlaXpirit: Fair question
15:40:45	gokr	Note that I don't think the tutorial is stripped/minified etc.,
15:40:57	gokr	Any "one page app" will have a bit of loading time I guess.
15:41:11	gokr	Also, the latest Amber and IDE etc is newer than those pages.
15:41:12	BlaXpirit	no but it seriously loads for minutes
15:41:15	vqrs	I'm waiting a minute now...
15:41:15	gokr	What?
15:41:31	gokr	Oh, then it must be broken
15:41:46	BlaXpirit	it loads for a minute and then just stops download and is simply broken
15:41:51	BlaXpirit	yay
15:41:57	flaviu	Yeah, there's a message in the console
15:42:01	flaviu	it looks broken
15:42:17	vqrs	yep
15:42:17	gokr	a pity.
15:42:27	gokr	But you can check the old IDE by clicking that button on the main page
15:42:51	gokr	And in fact, on www.smalltalkhub.com - you can get into the IDE and browse all the code of that site.
15:43:31	flaviu	Neat. Is the code stored as an AST?
15:43:45	gokr	The code is stored compiled.
15:43:55	gokr	If you hit the console on www.smalltalkhub.com
15:43:58	gokr	And type in:
15:43:59	gokr	smalltalk.Browser._open()
15:44:09	gokr	And hit enter, you get the IDE up. Then you can browse the code.
15:44:29	gokr	(agian the older IDE, the new is called Helios)
15:44:55	gokr	Every time you save a method it will be compiled into js and "hooked in".
15:44:57	vqrs	smalltalk.Browser is undefined here
15:45:35	gokr	Chrome?
15:45:40	gokr	It works for me in Chrome.
15:45:41	vqrs	Yeah, that explains it I guess: http://puu.sh/e6EsZ/7d5ea5c6f2.png
15:46:28	gokr	interesting
15:47:26	gokr	Weird.
15:47:30	gokr	It works for me.
15:47:37	vqrs	Maybe you have it cached?
15:47:40	def-	works for me as well
15:47:48	gokr	Smalltalkhub is an Amber app - that uses REST to talk to a REST based server that uses MongoDB.
15:48:03	gokr	So the whole UI is in Amber.
15:48:21	gokr	vqrs: I may, but I doubt def- has
15:48:32	vqrs	hmm
15:48:56	vqrs	now it works...
15:49:23	gokr	Its a sortof github kinda site but for Smalltalkers.
15:49:35	gokr	Nicolas wrote it - same guy who created Amber.
15:52:17	gokr	Sorry for the offtopic flurry, but its a very nice example of a language on/in js (not just a trivial transformation).
15:52:38	BlaXpirit	it never works out
15:55:43	*	bjz joined #nim
15:56:48	gokr	BlaXpirit: What do you mean?
15:57:11	vqrs	gokr: Shame the tutorial is down...
15:57:23	BlaXpirit	gokr, "a language on/in js (not just a trivial transformation)." never works out
15:57:36	gokr	vqrs: I can alert them on the list
15:57:45	vqrs	gokr: Thanks
15:57:56	saml	why not?
15:58:06	gokr	vqrs: But it a pretty clean Smalltalk, so any tutorial on Smalltalk
15:58:59	gokr	BlaXpirit: Its kinda interesting to claim that "it never works out" for a language that actually runs a site with over 2000 repositories and almost 2000 register accounts.
15:59:41	vqrs	I've never used SmallTalk before, so I it's kinda hard to judge for me whether I should go down that road
16:00:25	BlaXpirit	no, it is not hard
16:01:25	vqrs	You like being cryptic, eh?
16:01:33	*	kapil__ quit (Quit: Connection closed for inactivity)
16:01:49	BlaXpirit	being direct would've been worse
16:02:41	*	darkf quit (Quit: Leaving)
16:03:06	gokr	vqrs: Its kinda like Ruby since Ruby is a Smalltalk "clone" in a different syntax
16:03:50	gokr	As usual, its all about what you want to do. Smalltalk has been successful in many different contexts. But its very different from Nim of course.
16:04:43	gokr	vqrs: Good starting place: www.world.st
16:05:21	vqrs	gokr: Currently doing this: http://amber-smalltalk.github.io/trysmalltalk/ Based on the name "prof stef" I guess that it's the same
16:05:39	gokr	Ah, great. Yep, same.
16:06:22	*	gour quit (Remote host closed the connection)
16:07:45	*	gour joined #nim
16:10:49	BlaXpirit	took me long enough to realize that my library has memory leaks virtually everywhere
16:11:15	BlaXpirit	who would've thought {.override.} is so relevant
16:11:31	BlaXpirit	and now that I added it, I'm getting "usage of a type with a destructor in a non destructible context" for no reason
16:11:34	*	bjz quit (Ping timeout: 250 seconds)
16:11:56	BlaXpirit	are destructors even a popular thing? are there alternatives maybe?
16:12:24	BlaXpirit	because i do need to destroy ptr object when it goes out of scope
16:15:25	*	bjz joined #nim
16:15:31	EXetoC	pointers are unmanaged. you could either do it manually or wrap the type
16:15:58	BlaXpirit	I'm gonna need a destructor either way.
16:16:38	EXetoC	destructors are experimental
16:16:45	BlaXpirit	what does that mean?
16:17:06	BlaXpirit	not experimental in general
16:17:24	BlaXpirit	I'm interested for the outlook of community on destructors, the progress of their implementation, etc
16:17:39	BlaXpirit	in any case... is this a bug? https://bpaste.net/show/bcf10f3a5640
16:18:05	BlaXpirit	oh, should've added the error message. it's "(8, 5) Error: usage of a type with a destructor in a non destructible context"
16:18:05	EXetoC	don't you need {.experimental.} in order to use destructors now? they simply might not work as expected in some cases
16:18:43	BlaXpirit	EXetoC, they seem to work nicely when I don't reassign variables
16:19:10	BlaXpirit	at least the rest of my code compiled and memory leaks were gone
16:19:40	*	hakand_ joined #nim
16:20:02	BlaXpirit	oooh I have an idea. I wish I could let users of my library choose whether to activate destructors or not
16:20:14	*	hakand quit (Ping timeout: 245 seconds)
16:20:16	EXetoC	it's yet another thing that might not have priority over other things
16:21:11	BlaXpirit	I'm pretty sure I know how to conditionally apply a pragma, but what about passing options to modules?
16:21:39	*	matkuki joined #nim
16:22:14	BlaXpirit	--symbol?
16:23:47	EXetoC	I don't know, but I've only used --define so far
16:26:04	BlaXpirit	sure would be nicer to define it in code
16:28:54	BlaXpirit	define is the thing. thanks.
16:31:25	*	dyu quit (Quit: Leaving)
16:32:27	matkuki	Happy new year guys!
16:33:12	dv-	that was two weeks ago matkuki
16:33:27	BlaXpirit	optional destructors, not too shabby https://bpaste.net/show/d62540a0a074
16:33:31	saml	matkuki, happy birthday
16:35:05	matkuki	Yes I know, I haven't been on this channel for some time :)
16:36:29	matkuki	Why does array access with an enum member require an .ord to be added? example:
16:37:01	matkuki	a[MyEnum.en.ord], instead of just a[MyEnum.en]
16:37:28	EXetoC	that works if it's an array that takes MyEnum indices
16:38:01	BlaXpirit	matkuki, enums members are not numbers
16:38:36	*	jefus_ joined #nim
16:39:11	matkuki	BlaXpirit: Can you explain in more detail? I'm new to Nim.
16:39:48	BlaXpirit	what is MyEnum.en?
16:40:24	BlaXpirit	it's a thing, a concept, just one of MyEnum's options. it is not a number
16:40:27	*	bjz quit (Ping timeout: 265 seconds)
16:40:48	BlaXpirit	so why are you surprised that an operation that wants a number does not accept something that is not a number?
16:41:07	EXetoC	some languages have more relaxed semantics
16:41:20	EXetoC	but I approve of the added type safety
16:41:41	matkuki	EXetoC: That's why I was a little confused.
16:42:01	matkuki	BlaXpirit: Got it, thanks.
16:42:11	*	jefus quit (Ping timeout: 256 seconds)
16:42:19	EXetoC	it's similar to having a "T = distinct int", where T is not an int, but if you want the underlying value then you have to do "val.int"
16:44:55	matkuki	EXetoC: When is this useful?
16:45:51	EXetoC	matkuki: you could prevent something like this for example: "10.kg + 5.miles"
16:46:20	matkuki	good example, thanks
16:46:44	EXetoC	had those not been distinct, but then it would be exactly the same as "10 + 5"
16:48:20	EXetoC	implicit conversions can be defined with converters, but that should be done sparingly, so as to not create needless confusion
16:51:42	BlaXpirit	why does nim have to converge from posix-defined command line arguments specification?
16:51:47	BlaXpirit	diverge*
16:52:25	ldlework	to keep you on your toes
16:52:31	ldlework	BlaXpirit: I don't think Araq is a big fan of posix
16:52:41	BlaXpirit	and even in parseopt2 error handling seems nonexistent
16:52:44	BlaXpirit	i don't understand
16:52:53	BlaXpirit	does it just accept any options and
16:53:08	BlaXpirit	leave almost everything to you?
16:53:21	BlaXpirit	geez
16:53:32	BlaXpirit	not even an automatic help message generation?
16:53:54	ldlework	BlaXpirit: settle down
16:54:00	*	gokr quit (Quit: Leaving.)
16:54:45	BlaXpirit	someone really should check out docopt
16:56:01	*	kniteli joined #nim
16:56:12	BlaXpirit	ldlework, how can I settle down, when the language with the most potential suffers from such silly things
16:56:27	ldlework	BlaXpirit: I mean, you've been here for a while
16:56:39	BlaXpirit	i still keep getting surprised
16:57:03	ldlework	It should come as no surprise Nim is full of these wierd edges that require careful handling
16:58:02	BlaXpirit	does nim compiler command line interface actually use this thing?
16:58:15	BlaXpirit	that explains why it's so wonky
16:59:30	flaviu	EXetoC: In that case, mightn't an implicit converter between kg and g be a good idea?
16:59:31	*	enquora joined #nim
16:59:50	BlaXpirit	flaviu, they should be the same type
16:59:53	flaviu	so you could add 100.g + 1.kg
17:00:03	*	brson joined #nim
17:00:14	BlaXpirit	proc kg(x) = G(x:x)
17:00:24	ldlework	flaviu: I think that's what they were saying so but to not go over board
17:00:24	BlaXpirit	i mean proc g(x) = G(x:x)
17:00:37	BlaXpirit	proc kg(x) = G(x:1000*x)
17:00:41	flaviu	Yeah, I think that BlaXpirit has the best solution here.
17:01:03	flaviu	best to keep the units consistent
17:01:26	BlaXpirit	G is like distinct float
17:01:26	EXetoC	flaviu: why not overload the operators?
17:01:35	BlaXpirit	and no idea why i wrote "x:" there
17:01:46	flaviu	EXetoC: Combinatorial explosion.
17:02:02	BlaXpirit	kg(x) = Grams(1000.0*x)
17:02:29	flaviu	BlaXpirit: Yes, I understand
17:02:49	BlaXpirit	i just wanted to write it out better
17:03:44	BlaXpirit	seriously though, should I look into porting docopt to Nim?
17:04:27	flaviu	BlaXpirit: Please do. parseopt's syntax is unconventional and the api is meh too.
17:04:41	BlaXpirit	lol syntax is the least of its problems
17:04:52	BlaXpirit	that module probably has like 20 lines of code because it just does almost nothing
17:05:19	*	Matthias247 joined #nim
17:05:38	EXetoC	https://github.com/acolley/docopt.nim.git
17:05:40	repax	weight is just *one* type. Use templates to implement others too (length, energy, etc)
17:05:48	BlaXpirit	EXetoC, oh.
17:06:07	def-	repax: I did something like that: https://github.com/def-/nim-units/blob/master/units.nim
17:06:30	BlaXpirit	crazy
17:06:34	BlaXpirit	i like it
17:06:43	repax	def-, looks very nice!
17:07:01	flaviu	Yeah, really impressive
17:07:22	flaviu	although it looks like you forgot to add a flag to turn off tests.
17:08:27	def-	flaviu: yeah, the whole thing was more of a test, should clean it up a bit
17:08:48	BlaXpirit	docopt.nim - 1 star on github
17:08:52	BlaXpirit	is the situation really that dire?
17:09:26	flaviu	docopt.nim doesn't seem to have tests
17:09:39	EXetoC	I've never heard anyone mention it
17:09:43	flaviu	but docopt provides a test suite: https://github.com/docopt/docopt/blob/master/testcases.docopt
17:09:46	EXetoC	it might not have been announced
17:10:21	BlaXpirit	it seems unfinished
17:10:29	BlaXpirit	tons of todo
17:11:10	*	hakand__ joined #nim
17:11:10	*	hakand_ quit (Read error: Connection reset by peer)
17:11:11	flaviu	yep: "# TODO: finish"
17:11:16	BlaXpirit	D:
17:11:38	flaviu	BlaXpirit: So get to work!
17:11:46	EXetoC	that TODO sure carries a lot of weight
17:11:52	BlaXpirit	i've been working 5 days and nobody wants to check out my library
17:11:57	BlaXpirit	:<
17:12:23	BlaXpirit	well i haven't really announced it, but yeah
17:12:24	EXetoC	link?
17:12:30	EXetoC	right.. :p
17:12:35	BlaXpirit	https://github.com/blaxpirit/nim-csfml
17:13:04	flaviu	I'm not interested in making games at the moment, but maybe a short example would be good?
17:13:05	EXetoC	I'm pretty sure fowl would've accepted PRs, but ok
17:13:09	EXetoC	or were there any other issues?
17:13:11	flaviu	oh, there are examples
17:13:18	def-	flaviu: pretty nice examples even
17:13:18	BlaXpirit	there sure are examples
17:13:28	BlaXpirit	1 doesnt work because i added destructors and there is a bug
17:13:31	BlaXpirit	gonna fix it soon
17:13:41	BlaXpirit	(bug in nim, seems likely)
17:13:57	BlaXpirit	EXetoC, there are many issues
17:14:15	BlaXpirit	the main ones I've written as issues on fowl's library
17:14:28	BlaXpirit	but it's just manually maintained or even not maintained in some places
17:14:39	BlaXpirit	human error is likely
17:15:19	flaviu	It's also more fun to start from scratch, right? :P
17:15:23	BlaXpirit	yes
17:15:39	BlaXpirit	well in this case it also was easier
17:16:01	BlaXpirit	i did learn a lot from the other library though
17:18:13	*	willwillson joined #nim
17:18:30	repax	Is there a principle behind the variance in naming convention: new_RenderWindow() vs newSoundBufferRecorder()
17:19:03	BlaXpirit	repax, I write examples in 2 different styles
17:19:19	BlaXpirit	the library itself adheres to official style guide as far as i'm aware
17:19:23	*	goobles quit (Quit: Page closed)
17:19:25	BlaXpirit	but some examples i use my own preferred style
17:19:37	repax	Ok, just wondered
17:19:43	BlaXpirit	repax, underscores are ignored in nim btw
17:19:49	repax	I know
17:19:53	flaviu	#fcf8f4 looks really great. Nice choice boydgreenfield.
17:20:07	BlaXpirit	wha
17:20:13	repax	But sometimes one might want to put emphasis on something
17:20:39	*	bjz joined #nim
17:21:57	BlaXpirit	flaviu, took me long enough to figure out that you're probably talking about documentation
17:22:03	flaviu	oh, yes.
17:22:13	flaviu	the background color
17:24:24	*	gokr joined #nim
17:24:49	*	gokr_ joined #nim
17:24:59	BlaXpirit	And if someone is still looking at my library, make sure to nitpick.
17:25:03	*	gokr1 joined #nim
17:25:26	ldlework	simple.nim(3, 32) Error: type mismatch: got (int literal(800), int literal(600))
17:25:43	BlaXpirit	okay, that's not good..?
17:25:58	gokr1	BlaXpirit: Help me evolve lapp then
17:26:10	BlaXpirit	gokr1, sorry?
17:26:34	gokr1	https://gitlab.3dicc.com/gokr/lapp
17:26:47	gokr1	It uses the synopsis (= help message) as the grammar
17:27:00	gokr1	Originall ported to Nim from Lua by Steve Donovan
17:27:01	BlaXpirit	gokr1, but that's what docopt does
17:27:11	gokr1	docopt?
17:27:14	BlaXpirit	why is your library good and why why not port docopt and why i
17:27:19	BlaXpirit	why is your library good and why not port docopt
17:27:37	gokr1	I didn't know about docopt - and lapp was already (as I said) ported by Steve
17:27:43	BlaXpirit	okay
17:27:49	gokr1	I simply fixed it from a bit of rot - and improved it a bit.
17:27:58	gokr1	I use it in blimp
17:28:51	*	gokr quit (Ping timeout: 264 seconds)
17:29:01	flaviu	I personally like docopt better
17:30:17	gokr1	Seems neat
17:30:21	gokr1	I hadn't heard of it
17:30:36	BlaXpirit	I updated nim-csfml so that snakes example should work now
17:30:42	BlaXpirit	it's my favorite
17:31:44	BlaXpirit	ooh it might be a cool idea to implement a compile-time option parser
17:32:11	BlaXpirit	docopt for C works by generating C code using Python :|
17:32:12	flaviu	Cool, yes, but it'll be a lot more work
17:32:22	BlaXpirit	how cool of a showcase would it be though
17:32:33	BlaXpirit	i have no idea if it's even possible
17:32:39	flaviu	of course it is
17:32:58	flaviu	if nothing else, you could use parseStmt
17:33:21	flaviu	then it's just standard string mangling
17:35:12	BlaXpirit	ldlework, so please tell me more about that error
17:35:12	*	yglukhov___ quit (Read error: Connection reset by peer)
17:35:17	*	yglukhov____ joined #nim
17:36:28	BlaXpirit	proc videoMode*(modeWidth, modeHeight: cint, bitsPerPixel: cint = 32)
17:36:36	BlaXpirit	can you not call it videoMode(800, 600) ?
17:37:08	flaviu	gokr1: your lapp library doesn't appear to be in pacakges.json
17:37:24	flaviu	You should submit it: https://github.com/nim-lang/packages/compare
17:38:47	*	bjz quit (Read error: Connection reset by peer)
17:39:13	ldlework	BlaXpirit: I'm just running your master branch
17:39:16	ldlework	and doing
17:39:20	ldlework	nim c simple.nim
17:39:31	BlaXpirit	well that's not good
17:39:42	BlaXpirit	you can do cint(800), cint(600)
17:39:52	BlaXpirit	but i thought you can pass int literal to cint arguments
17:40:04	BlaXpirit	it's pretty lame if you can't......
17:40:20	BlaXpirit	but it works on my system, and on def's system
17:40:36	ldlework	it wants three ints
17:40:39	ldlework	not two
17:40:46	ldlework	csfml.videoMode(width: cint, height: cint, bpp: cint): TVideoMode
17:41:15	BlaXpirit	ldlework, you're using fowl's library
17:41:21	BlaXpirit	D:
17:41:34	BlaXpirit	the T gives it away
17:41:51	ldlework	That sounds like a problem with packaging then
17:41:55	ldlework	Well I guess not
17:42:08	BlaXpirit	it's quite a problem that they're named the same :|
17:42:46	ldlework	how do you do a nimble install from source?
17:42:54	BlaXpirit	go into the folder
17:42:56	BlaXpirit	nimble install
17:43:01	BlaXpirit	i should mention that too, huh
17:43:31	BlaXpirit	ldlework, but the way I'm doing it is --path:../src
17:44:24	*	yglukhov____ quit (Ping timeout: 240 seconds)
17:55:50	*	bjz joined #nim
17:58:16	ldlework	BlaXpirit: I installed csfml 2.2 from source but I get
17:58:24	ldlework	could not load: libcsfml-window.so
17:58:28	BlaXpirit	ldlework, is there such thing as csfml 2.2?
17:58:35	ldlework	yeah its the latest version
17:58:35	*	hguux_ quit (Ping timeout: 272 seconds)
17:58:45	BlaXpirit	https://github.com/LaurentGomila/CSFML/releases
17:58:58	*	phira quit (Excess Flood)
17:59:03	ldlework	http://www.sfml-dev.org/download/sfml/2.2/
17:59:07	ldlework	I take it csfml is something else?
17:59:13	BlaXpirit	yes
17:59:20	*	phira joined #nim
17:59:28	*	Sphax quit (Quit: CYA!!)
17:59:41	BlaXpirit	ldlework, it looks like you had used fowl's library before
17:59:53	BlaXpirit	did you not have those libraries already installed?
17:59:59	ldlework	alright
18:00:32	BlaXpirit	i'm gonna try to help you in private messages cuz this is offtopic
18:00:37	*	ARCADIVS quit (Quit: ARCADIVS)
18:01:07	*	rpag joined #nim
18:01:07	*	hguux_ joined #nim
18:06:16	*	dts|pokeball quit (Quit: Leaving)
18:06:36	*	dts|pokeball joined #nim
18:07:37	*	dts|pokeball quit (Remote host closed the connection)
18:07:57	*	dts|pokeball joined #nim
18:08:50	*	dts|pokeball quit (Remote host closed the connection)
18:09:05	*	dts|pokeball joined #nim
18:10:19	*	skyfex quit (Read error: Network is unreachable)
18:10:56	*	skyfex joined #nim
18:11:48	*	iivvoo quit (Ping timeout: 265 seconds)
18:12:33	*	dts|pokeball quit (Client Quit)
18:14:13	*	EastByte_ quit (Ping timeout: 265 seconds)
18:15:03	*	CARAM__ quit (Ping timeout: 244 seconds)
18:16:11	*	Varriount|Busy joined #nim
18:16:26	*	EastByte_ joined #nim
18:17:05	Varriount|Busy	Meep.
18:17:14	BlaXpirit	hi
18:17:47	*	dtscode joined #nim
18:17:58	*	CARAM__ joined #nim
18:18:11	*	iivvoo joined #nim
18:18:29	ldlework	I wanted to share this all with you: http://labs.echonest.com/Uploader/index.html?trid=TRXKEZN13AF58CD0A8&thresh=9
18:19:00	ldlework	you're welcome
18:21:05	*	hakand__ quit (Quit: Be back later ...)
18:22:28	*	gour_ joined #nim
18:25:19	*	bjz quit (Ping timeout: 256 seconds)
18:25:22	*	Trixar_za quit (Ping timeout: 244 seconds)
18:25:42	*	gour quit (Ping timeout: 245 seconds)
18:27:58	*	Trixar_za joined #nim
18:34:02	*	superfunc joined #nim
18:39:52	*	yglukhov____ joined #nim
18:42:33	*	gour_ is now known as gour
18:43:04	*	bjz joined #nim
18:43:42	Varriount|Busy	Hrm. Decisions, decisions...
18:45:33	superfunc	just curious, on variant types, why can't the names clash across discriminators?
18:48:19	superfunc	as in: Term = ref object ... case kind : TermKind ... of Succ: t1 : Term .... of Pred: t1 : Term
18:49:12	Varriount|Busy	superfunc: I think it's an implementation detail, and an ambiguity issue
18:49:37	ekarlso-	flaviu: yo
18:49:43	flaviu	hi
18:50:00	superfunc	Varriount|Busy: Ok, I just wanted to make sure there wasn't some flaw in doing it that way I was missing
18:52:20	BlaXpirit	superfunc, Varriount|Busy, I just recently also ran into that
18:53:20	superfunc	I've been doing a small writeup on implementing simply typed LC in nim to add some documentation to the language.
18:53:30	Varriount|Busy	LC?
18:53:37	superfunc	Hopefully I'll get it done in the next few days, still a few more things I need to read
18:53:42	superfunc	lambda calculus
18:54:21	superfunc	so far my Nim implementation is much faster, and equally as clean as my old Ocaml one
18:54:30	*	tgkokk joined #nim
18:54:44	def-	superfunc: nice, anything online yet?
18:55:16	superfunc	I'll ping you in the next couple days I'm ready to show it
18:55:17	*	bjz quit (Ping timeout: 245 seconds)
18:55:45	*	ldlework is now known as superproc
18:56:16	*	superproc is now known as ldlework
18:57:02	superfunc	had I known I'd be so into nim... lol
18:57:15	ldlework	hehe
18:59:28	*	tgkokk quit (Ping timeout: 264 seconds)
19:04:00	*	tgkokk joined #nim
19:04:26	*	bjz joined #nim
19:04:56	*	SplinterOfChaos joined #nim
19:07:16	onionhammer	Varriount|Busy deciding on what to work on next? :)
19:07:23	onionhammer	Varriount|Busy cuz I have an idea
19:08:56	Varriount|Busy	onionhammer: What is it?
19:09:12	onionhammer	new NimLime release
19:09:17	Varriount|Busy	onionhammer: I was thinking about working on IDE Tools, since people on the forum seem to be clamoring for it.
19:09:31	onionhammer	that would be awesome
19:09:42	Varriount|Busy	onionhammer: Sure we can't just set the package control to just use the head revision all the time?
19:09:43	onionhammer	I could help you test that too
19:09:56	onionhammer	im not sure - i didn't submit it to package control
19:10:04	onionhammer	im not sure we want to do that either ;)
19:10:20	onionhammer	also I think some of your stuff might have gotten borked over the past several months
19:10:27	Varriount|Busy	onionhammer: ?
19:10:45	onionhammer	just thnk you should retest it, not sure specifically
19:10:51	onionhammer	if you havent already retested latest
19:11:02	onionhammer	think there was something with ##
19:11:06	onionhammer	iirc
19:11:06	Varriount|Busy	onionhammer: I accounted for the executable renaming, I think.
19:11:22	Varriount|Busy	When I get home I'll test it.
19:11:25	onionhammer	okay
19:12:08	Varriount|Busy	onionhammer: At the moment, in my spare time, I'm designing a filepath module.
19:12:15	onionhammer	interesting
19:12:35	Varriount|Busy	The problem is both storing things efficiently, *and* allowing extensions...
19:13:25	Varriount|Busy	For instance, I plan on storing the parts of a path in a single string, with each part prefixed by an integer. The only problem is, where do I store a drive (when on windows)?
19:15:28	def-	Varriount|Busy: what about ropes?: http://nim-lang.org/ropes.html
19:16:48	Varriount|Busy	def-: Ropes are trees. Part of my goal is to *minimize* references.
19:16:53	def-	alright
19:17:01	onionhammer	hm nim or nimlime isnt happy with spaces in path on windows
19:17:03	onionhammer	:)
19:17:10	onionhammer	during build
19:17:50	onionhammer	Varriount /c/Path/On/Windows ?
19:20:34	flaviu	Are ropes really effective at improving performance compared to giant char vectors?
19:21:02	Varriount|Busy	flaviu: Ropes are good for many large strings.
19:21:44	Varriount|Busy	flaviu: They can help save space. It's one of the tme vs space tradeoffs
19:21:47	Varriount|Busy	*time
19:23:22	*	bjz quit (Read error: Connection reset by peer)
19:24:43	*	tgkokk quit (Ping timeout: 255 seconds)
19:28:29	*	gsingh93 joined #nim
19:29:35	BlaXpirit	lol so i just installed nim on Windows and it can't compile anything
19:29:48	BlaXpirit	.eval echo "ok" # not even this
19:29:48	Varriount|Busy	BlaXpirit: How, and did you install a compiler?
19:29:53	Mimbus	BlaXpirit: ok
19:30:38	*	tgkokk joined #nim
19:30:57	BlaXpirit	I downloaded Windows installer 32bit and installed on Windows 7 64bit virtual machine. I had MinGW installed apparently but also checked to install MinGW in Nim installer
19:31:24	Varriount|Busy	BlaXpirit: And what happens when you try to compile something?
19:31:28	BlaXpirit	C:\Nim\lib/nimbase.h:385:13: error: size of array 'assert_numbits' is negative typedef int assert_numbits[sizeof(NI) == sizeof(void*) && NIM_INTBITS == sizeof (NI)*8 ? 1 : -1];
19:31:43	Varriount|Busy	BlaXpirit: That means it's still picking up your previous compiler.
19:32:06	*	bjz joined #nim
19:32:10	BlaXpirit	Varriount|Busy, I don't think there was anything wrong with that compiler
19:32:13	Varriount|Busy	The 'add to path' option *appends* to your path
19:32:37	Varriount|Busy	BlaXpirit: That error occurs when the bit-ness of the Nim executable and the bit-ness of the C compiler don't match.
19:32:45	BlaXpirit	but I think it's 32bit
19:33:02	Varriount|Busy	" I had MinGW installed apparently but also checked to install MinGW in Nim installer"
19:33:25	BlaXpirit	I only install 32 bit things
19:33:33	BlaXpirit	although who knows..
19:33:50	Varriount|Busy	BlaXpirit: Please check.
19:33:58	BlaXpirit	wish I knew how
19:34:21	BlaXpirit	I see mingw32 everywhere
19:36:01	Varriount|Busy	BlaXpirit: That doesn't necessarily mean anything. The 64 bit version of Mingw32 has a similar folder layout to the 32 bit version
19:36:38	Varriount|Busy	BlaXpirit: What does 'gcc --version' print out?
19:37:00	BlaXpirit	I did a SET PATH=C:\Nim\dist\mingw\bin;%PATH% and it works now. thanks.
19:37:39	BlaXpirit	hm interesting, Varriount|Busy
19:37:54	BlaXpirit	I didn't even have that old compiler in PATH, that special Nim shortcut seems to add it
19:38:00	*	tgkokk left #nim ("Textual IRC Client: www.textualapp.com")
19:38:10	Varriount|Busy	BlaXpirit: Huh? What old compiler?
19:38:20	BlaXpirit	the old mingw that i had installed
19:38:26	Varriount|Busy	Hm...
19:38:55	Varriount|Busy	BlaXpirit: The shortcut leads to a batch file. Could you gist the contents of it please?
19:38:57	BlaXpirit	i never added it to PATH but when i start Nim command line shortcut, it initializes the session by prepending that external mingw to the beginning of PATH and its own is near the end
19:39:57	BlaXpirit	https://bpaste.net/show/a162a51233a7
19:41:04	BlaXpirit	when i start my normal cmd, mingw32-gcc does not exist
19:41:25	BlaXpirit	when i use this batch file/shortcut, it is mingw32-gcc (GCC) 4.8.1
19:41:31	Varriount|Busy	BlaXpirit: Um, I don't see where it's prepending the external mingw?
19:41:40	BlaXpirit	Varriount|Busy, that surprised me too
19:42:57	BlaXpirit	no, I'm misinterpreting something
19:43:35	BlaXpirit	very sorry for the confusion
19:43:49	*	flaviu quit (Read error: Connection reset by peer)
19:43:50	BlaXpirit	I actually have gcc (rev5, Built by MinGW-W64 project) 4.8.1
19:44:01	BlaXpirit	and it is present in PATH
19:44:19	BlaXpirit	should've just checked everything
19:45:10	*	flaviu joined #nim
19:45:19	BlaXpirit	but that still doesn't explain why it doesn't prefer its own GCC
19:45:54	BlaXpirit	ah because I change directory
19:46:02	BlaXpirit	NIMRODPATH=. duh
19:46:11	BlaXpirit	that shouldn't be the way it is
19:46:18	Varriount|Busy	Hrm, your right.
19:46:38	*	Varriount|Busy makes another note of what to when he gets home
19:46:46	BlaXpirit	thx
19:50:33	BlaXpirit	cool stuff. nim-csfml works on Windows!
19:52:39	BlaXpirit	didnt even have to mess with compilers or anything, CSFML dll binary download works
19:52:49	BlaXpirit	not on Linux, though
19:52:59	Varriount|Busy	BlaXpirit: Make sure to test with a mingw64 compiler
19:53:17	BlaXpirit	Varriount|Busy, why?
19:53:33	Varriount|Busy	BlaXpirit: Platform differences
19:53:46	BlaXpirit	I will notbut i have 64 bit on linux anyway
19:55:32	BlaXpirit	yeah, i should've installed 64bit first, then uninstalled and left 32bit permanently
19:55:44	BlaXpirit	don't want to redo it twice though
19:58:21	Varriount|Busy	BlaXpirit: You know, you could fix the compiler script, and then send it to me.
19:58:40	Varriount|Busy	Since it's going to take me about half an hour to get home
19:59:58	*	BlaXpirit quit (Read error: Connection reset by peer)
20:00:17	*	Varriount|Busy quit (Quit: Page closed)
20:00:48	*	BlaXpirit joined #nim
20:01:31	*	matkuki quit (Quit: ChatZilla 0.9.91.1 [Firefox 34.0.5/20141126041045])
20:02:10	BlaXpirit	Varriount, I have no idea what to fix
20:04:19	BlaXpirit	I mean, `SET NIMRODPATH=D%` is an OK fix as long as you make sure current directory path is set correctly in the shortcut
20:04:27	BlaXpirit	I mean, `SET NIMRODPATH=%CD%` is an OK fix as long as you make sure
20:04:35	BlaXpirit	%C apparently is something
20:09:54	*	superfunc quit (Ping timeout: 250 seconds)
20:16:51	*	flaviu quit (Remote host closed the connection)
20:17:08	*	flaviu joined #nim
20:18:08	*	flaviu quit (Remote host closed the connection)
20:18:22	MyMind	https://twitter.com/macromates/status/552134978692472832
20:18:23	*	flaviu joined #nim
20:22:12	*	hasenj joined #nim
20:24:24	*	bjz quit (Ping timeout: 244 seconds)
20:41:16	*	flaviu quit (Remote host closed the connection)
20:41:30	*	flaviu joined #nim
20:41:59	*	flaviu quit (Remote host closed the connection)
20:42:18	*	flaviu joined #nim
20:49:12	*	hasenj quit (Quit: hasenj)
20:49:19	onionhammer	http://forum.nimrod-lang.org/t/721
20:51:29	*	nande joined #nim
20:51:45	*	Mat4 joined #nim
20:51:51	Mat4	hello
20:53:51	flaviu	Why does https://github.com/nim-lang/nits exist?
20:54:19	BlaXpirit	D:
20:54:25	flaviu	Also, a redirect for http://nim-lang.org/nimrodc.html would be good.
20:56:01	flaviu	https://github.com/nim-lang/nimble/blob/master/developers.markdown makes no mention of tests.
20:56:13	flaviu	What is the recommended directory structure for that?
21:09:44	*	gour quit (Remote host closed the connection)
21:10:22	*	jpoirier joined #nim
21:12:07	*	matkuki joined #nim
21:12:48	ekarlso-	flaviu: u wanna look at the packages thing again ? :p
21:13:07	ekarlso-	i'm gonna push a patch for the license thingie
21:13:07	flaviu	ekarlso-: Sure, let me pull
21:13:39	ekarlso-	flaviu: lemme push first :p
21:13:41	ekarlso-	will notify u
21:20:09	matkuki	How do you convert a C pointer to a static string to something that can be used with "echo"?
21:20:34	matkuki	I trying to pring the OpenGL version using glGetString(GL_VERSION)
21:20:45	matkuki	sorry, print
21:21:25	EXetoC	cast to cstring?
21:22:04	EXetoC	and then it will be converted to string by echo
21:22:13	EXetoC	which calls $ implicitly
21:22:50	matkuki	Gives me "SIGSEGV: Illegal storage access."?
21:23:20	matkuki	Also tried "cast[ptr cstring]"
21:23:25	BlaXpirit	is there some C parser for nim that gives AST of C?
21:23:53	repax	matkuki, are you sure that the pointer is non nil?
21:24:17	repax	and can you read its content as, say an int?
21:25:10	EXetoC	matkuki: does it return nil?
21:25:25	EXetoC	it might
21:25:27	matkuki	just a sec...
21:26:09	repax	BlaXpirit, have you looked at clang?
21:26:14	BlaXpirit	no
21:26:16	EXetoC	but the proc should then raise an GL error, unless something else is going on
21:26:20	repax	sorry, didn't see the nim part
21:26:36	BlaXpirit	what does c2nim use
21:26:37	matkuki	EXetoC: If I use "== nil" it returns false
21:27:24	BlaXpirit	rolls with its own..
21:27:32	BlaXpirit	matkuki, give code pls
21:27:58	repax	matkuki, echo cast[int](versionString[])
21:28:14	EXetoC	?
21:28:25	Mat4	is someone here interested in cross compilation for Parallax Propeller MCU's ?
21:28:37	matkuki	I'm using the nim-lang/opengl library.
21:28:42	matkuki	The code is var tmp :cstring; tmp = cast[cstring](glGetString(GL_VERSION)); echo(tmp)
21:30:35	def-	matkuki: https://www.opengl.org/wiki/FAQ#What_is_an_OpenGL_context_and_why_do_you_need_a_window_to_do_GL_rendering.3F
21:30:48	EXetoC	it didn't return nil though
21:31:57	*	Amrykid joined #nim
21:32:00	repax	possible reasons: 1) the address can't be read at all by the process. 2) it's not null-terminated
21:32:32	BlaXpirit	something is wrong again :(
21:32:33	EXetoC	you could print the first 20 chars or something, one by one
21:32:38	BlaXpirit	$ nimble install c2nim
21:32:52	BlaXpirit	nim not found
21:33:00	BlaXpirit	(Looking for nim (>= 0.10.2))
21:34:53	BlaXpirit	any ideas?
21:35:17	BlaXpirit	nim -v 0.10.3
21:35:43	BlaXpirit	ah damn, i probably have old nimble
21:36:56	BlaXpirit	now it says c2nim.nim(11, 33) Error: cannot open 'llstream'
21:37:37	matkuki	repax: also tried "glGetString(GL_SHADING_LANGUAGE_VERSION​)" and "glGetString(GL_VENDOR​)", same error.
21:37:49	BlaXpirit	matkuki, give code pls
21:38:00	BlaXpirit	why are you making it hard for everyone?
21:38:33	BlaXpirit	oh, you gave 1 line
21:38:36	EXetoC	you compared the output of glGetString to nil directly? though it shouldn't matter
21:38:43	matkuki	http://pastebin.com/3e3eASRa
21:38:46	BlaXpirit	but I want a complete example so i can try and tweak
21:38:52	EXetoC	BlaXpirit: that problem has remained for some time
21:38:58	BlaXpirit	:(
21:39:04	EXetoC	because the compiler is not being included in the installation
21:39:14	matkuki	no sorry wait...
21:39:29	EXetoC	it is indeed easier to just add the bin dir to the path
21:39:55	BlaXpirit	EXetoC, ?
21:39:59	BlaXpirit	which path
21:40:07	BlaXpirit	the normal one?
21:40:14	matkuki	http://pastebin.com/hw8t3ChR
21:40:17	matkuki	line 132
21:40:46	EXetoC	BlaXpirit: yes. that would be the simplest "installation"
21:40:48	EXetoC	for now
21:40:58	BlaXpirit	ah
21:41:07	BlaXpirit	i thought maybe i could fool nimble with PATH somehow
21:41:09	BlaXpirit	ok thx
21:41:10	EXetoC	but maybe you can copy the compiler dir to that lib dir
21:41:30	EXetoC	and I could update the build script if it then works
21:42:02	*	Amrykid quit (Changing host)
21:42:02	*	Amrykid joined #nim
21:42:12	EXetoC	I think it would be best if it was /usr/lib/nim/lib though, in that case
21:42:23	BlaXpirit	okay, matkuki, i can run it and get sigsegv. will look.
21:42:41	matkuki	BlaXpirit: Great, thanks!
21:44:22	BlaXpirit	glGetString(GL_SHADING_LANGUAGE_VERSION) returns 61
21:44:23	BlaXpirit	okay
21:44:42	BlaXpirit	wrapper is broken, matkuki
21:44:46	BlaXpirit	very very broken
21:44:55	BlaXpirit	opengl.nim
21:44:56	BlaXpirit	proc glGetString(name: GLenum): GLubyte
21:45:10	BlaXpirit	actually is ptr GLubyte
21:45:19	EXetoC	crap
21:45:32	BlaXpirit	EXetoC, who writes wrappers like that, srsly
21:45:38	EXetoC	me?
21:45:54	BlaXpirit	i thought my wrapper was not ready
21:46:08	BlaXpirit	now i think otherwise :p
21:46:14	matkuki	I noticed that. I'm not that good at Nim yet, I thought the conversion was implicit or something.
21:46:29	EXetoC	it would be very confusing
21:46:50	flaviu	BlaXpirit: Yes, your wrapper is excellent!
21:46:57	flaviu	Most are nothing more than c2nim output.
21:47:17	Mat4	what's the status of c2nim (and where can I found it) ?
21:47:58	BlaXpirit	Mat4, you can find it in google. 2nd search result https://github.com/nim-lang/c2nim
21:48:06	BlaXpirit	wait
21:48:09	BlaXpirit	you mean "status"?
21:48:42	BlaXpirit	well it's definitely not gonna translate C code, it's mostly just for headers, and not perfect even at that, last time i've tried
21:49:14	renesac|away	there is some improvment ideas on GSOC 2014 proposal
21:49:21	renesac|away	for c2nim
21:49:27	*	renesac|away is now known as renesac
21:50:56	BlaXpirit	flaviu, i wish someone would use it xD
21:51:13	Mat4	I need wrapping the standard libraries of the propGCC compiler for using Nim for cross-compilation
21:51:27	EXetoC	some things were fixed the last couple of months
21:51:57	matkuki	BlaXpirit, EXetoC: yup, with adding "ptr" to the glGetString return type it works.
21:52:02	EXetoC	the biggest issue I had was the unsupported ##, and multiline macros
21:52:32	Mat4	ok, I will how well the sources are converted
21:53:03	BlaXpirit	Mat4, it will not convert sources well
21:53:09	BlaXpirit	only C header (.h) files
21:53:49	BlaXpirit	EXetoC, do you think I could take advantage of clex, cparse somehow?
21:54:15	BlaXpirit	because i don't understand it at all :|
21:55:32	EXetoC	no idea
21:56:07	*	johnsoft quit (Ping timeout: 245 seconds)
21:56:44	ekarlso-	any people here good at webdesign ? :)
21:57:20	*	johnsoft joined #nim
22:01:40	EXetoC	I'm alright at developing functional websites
22:02:11	EXetoC	but I've never bothered to learn how to do fancy stuff with css since it's so quirky
22:02:27	matkuki	EXetoC: Should I open an issue on github for glGetString?
22:03:17	ldlework	EXetoC: amen
22:03:17	EXetoC	sure
22:03:21	EXetoC	my wrapper generator must be broken. I hope I can find it again
22:03:41	EXetoC	but I'll have to look at what else is broken
22:04:27	matkuki	I'm going to be playing with opengl for a while, will report if I find anything else.
22:05:55	flaviu	I can't say I'm good at CSS, but things typically turn out at least reasonable-looking.
22:05:55	BlaXpirit	i should try to pair some of this with sfml :3
22:06:46	BlaXpirit	I'm so tempted to just steal an OpenGL piece of code from somewhere, but the licenses, those damn licenses
22:07:08	BlaXpirit	there is a lot of code under MIT, but can I publish it in my repo under zlib license?
22:07:09	EXetoC	no one will notice :>
22:08:45	flaviu	Is there some way to disable `Info: instantiation from here`?
22:09:05	BlaXpirit	why
22:09:15	ldlework	EXetoC: :P
22:09:40	flaviu	BlaXpirit: Because it's useless and irritates me.
22:09:50	flaviu	I suppose grep will suffice :P
22:11:06	*	Mimbus quit (Read error: No route to host)
22:11:10	flaviu	As there looks like there's no way to turn off ProveInit either
22:12:09	*	flaviu left #nim (#nim)
22:12:12	*	flaviu joined #nim
22:12:40	*	Mimbus joined #nim
22:12:50	BlaXpirit	matkuki, where did you get that code that you pasted from?
22:13:41	BlaXpirit	also, any particular reason why you're using SDL? :p
22:13:49	ldlework	Its better.
22:14:01	ldlework	Packaged better, for one.
22:14:25	matkuki	BlaXpirit: a little modified from "nim-lang/sdl2/examples/sdl_opengl_example.nim"
22:14:35	BlaXpirit	thank you, matkuki
22:17:22	matkuki	I'm want to make a game. Nim looks and feels great to me. I already have a prototype is Python/SDL2. I kinda like SDL2.
22:18:09	dtscode	is nim good for writing games?
22:18:31	BlaXpirit	dtscode, probably
22:18:42	BlaXpirit	that's what I'm expecting to use it for
22:18:49	dtscode	interesting
22:19:21	ldlework	matkuki: I too am using Nim for games
22:19:27	BlaXpirit	matkuki, are you going for 3D?
22:19:42	*	enquora quit (Quit: enquora)
22:20:27	dtscode	ldlework, can i see an example?
22:20:27	matkuki	BlaXpirit: not yet, I'm just starting to play with OpenGL. The game will be primarily 2D. For now
22:21:01	BlaXpirit	"primarily" 2D ?
22:21:58	Araq	BlaXpirit, Mat4 c2nim does translate function bodies as well
22:22:14	repax	I have an sdl2 wrapper with nicer names for procs and types (i.e. no T/P prefixes etc): https://github.com/repax/sdl2
22:22:19	matkuki	Idlework: Good to know. How far have you come?
22:22:23	Araq	it's not perfect, but when you know its limitations you can convert thousands of lines easily
22:22:37	ldlework	matkuki: I've cloned Qix and I'm working on a Lights Out! clone with a friend
22:23:00	ldlework	matkuki: I'm working on my own little game library similiar to pygame
22:23:13	Araq	and I'm making it compile 'template' right now
22:23:38	BlaXpirit	ldlework, sfml..?
22:24:15	matkuki	Idlework: That's great. Keep it up! I'm currently working out the art direction for the game.
22:24:36	Araq	repax: put it on nimble
22:24:51	*	superfunc|busy joined #nim
22:25:02	repax	Araq, I've sent a pull req to dom
22:25:16	Araq	it's usually merged within a day
22:25:17	BlaXpirit	I do not see why "sdl2" is in standard library
22:25:38	Araq	BlaXpirit: graphics.nim uses sdl and is still in the stdlib
22:25:47	BlaXpirit	graphics, eh
22:26:05	repax	Hopefully, most libraries will move over to the new naming convention
22:26:20	ldlework	BlaXpirit: no, I'm not impressed with sfml, I use SDL
22:26:31	Araq	repax: that's already happening afaict
22:26:46	BlaXpirit	ldlework, what exactly are you not impressed with?
22:26:51	ldlework	BlaXpirit: the packaging for one
22:27:02	BlaXpirit	lol
22:27:08	BlaXpirit	what is wrong with packaging?
22:27:18	ldlework	there is none?
22:27:24	BlaXpirit	in my experience SDL has been much, much harder to get working
22:27:26	ldlework	you get some folders with the headers and libs in them
22:27:31	BlaXpirit	what do you mean "there is none"
22:27:34	ldlework	you have to install them by hand
22:27:35	matkuki	BlaXpirit: For now it's just 2D, maybe I'll add a depth of field element or something to it later.
22:27:40	BlaXpirit	ldlework, no..?
22:27:40	dtscode	BlaXpirit, i get sdl working pretty easily
22:27:51	ldlework	BlaXpirit: point to the debs I can install sfml 2.1
22:28:01	ldlework	(official debs)
22:28:24	BlaXpirit	ldlework, lol you think everyone is supposed to make official debs for everything?
22:28:43	ldlework	BlaXpirit: I'll stick with SDL then
22:28:48	BlaXpirit	facepalm
22:29:09	BlaXpirit	that's like saying "i'll stick with Windows then"
22:29:16	flaviu	I'm using fsanitize, and the GC is causing a "stack-buffer-overflow" at lib/system/gc.nim:888
22:29:22	ldlework	If you say so. Notice how I'm unmoved.
22:29:42	Araq	flaviu: how do you think conservative stack marking works?
22:29:46	*	dtscode is moved at ldlework's steadfast unmovingness
22:30:00	repax	You all move upwards on my screen
22:30:02	BlaXpirit	oh look, ldlework https://launchpad.net/~sonkun/+archive/ubuntu/sfml-stable
22:30:10	*	tane quit (Quit: Verlassend)
22:30:15	BlaXpirit	seems abandoned though -_-
22:30:30	ldlework	BlaXpirit: I don't think I'll be asking my users to install a PPA for a project that can't provide packages
22:30:48	BlaXpirit	ldlework, you can't be serious here
22:30:59	ldlework	BlaXpirit: you keep saying that, and yet I keep using SDL
22:31:06	Araq	ldlework: *we* don't provide packages
22:31:14	Mat4	BlaXpirit: There are similar projects like this out there
22:31:18	BlaXpirit	why do you think a developer is supposed to make packages for every single distro?
22:31:21	ldlework	Araq: sure, but we're not even 1.0
22:31:23	Mat4	(even for C++)
22:31:34	ldlework	Araq: and you've asked people to do so
22:31:38	ldlework	so we're ahead of the SFML project
22:31:40	Araq	ldlework: in opensource land, few things are
22:31:40	ldlework	\o/
22:31:52	BlaXpirit	ldlework, look, if your sh**ty distro doesn't have packages, that's nobody else's fault
22:31:54	ldlework	Araq: how does it feel to be ahead of the game?
22:32:04	ldlework	BlaXpirit: does your OS have packages for SFML 2.1?
22:32:10	BlaXpirit	yes
22:32:16	ldlework	neat which OS?
22:32:23	BlaXpirit	Arch Linux
22:32:25	ldlework	And who maintains them?
22:32:40	flaviu	ldlework:
22:32:41	Araq	ldlework: I don't know the opposite feeling.
22:32:42	flaviu	Sven-Hendrik Haase
22:32:43	dtscode	arch is really awesome about keeping its repos upto date
22:32:48	ldlework	Araq: ;P
22:32:50	BlaXpirit	https://www.archlinux.org/packages/community/x86_64/sfml/ Sven-Hendrik Haase
22:32:51	dtscode	Araq++
22:33:03	dtscode	now if arch could just be more stable :/
22:33:13	BlaXpirit	sorry for offtopic tho :(
22:33:19	flaviu	dtscode: You just aren't l33t enough ;)
22:33:28	ldlework	BlaXpirit: cool! I hope it works well for you.
22:33:47	dtscode	flaviu, probably :} i will admit to having lost entire arch installations on my hdd before
22:33:57	BlaXpirit	ldlework, you probably should go back to C++ though, because distros don't have packages for nim
22:33:59	dtscode	i did use manjaro though
22:34:07	ldlework	BlaXpirit: I'm not looking to distribute Nim
22:34:09	BlaXpirit	i don't know why i said "back"
22:34:19	dtscode	BlaXpirit, didnt ubuntu use to have a nimrod compiler package?
22:34:25	flaviu	BlaXpirit: Relax. He's not required to use your package if he prefers SDL.
22:34:31	ldlework	BlaXpirit: hey I have an idea, lets be a complete shitass
22:34:48	BlaXpirit	i'm just asking for any reason whatsoever why you prefer sdl
22:34:56	ldlework	BlaXpirit: and I gave you one
22:35:00	ldlework	did you ask so that you could combat me
22:35:02	ldlework	and put me down
22:35:10	*	saml_ joined #nim
22:35:18	ldlework	Or are you literally offended in CFML's place?
22:35:42	*	saml quit (Ping timeout: 265 seconds)
22:36:47	superfunc|busy	why you gotta keep ruffling feathers dude, just chill
22:36:52	*	adam_s joined #nim
22:37:59	ldlework	dtscode: this is a game I'm helping my friend write to help him learn how to use Nim: https://github.com/dustinlacewell/lights-out
22:39:06	dtscode	awesome. ty
22:39:14	ldlework	I don't remember where we left off, or if it is in a working state
22:43:51	flaviu	Is there a way to convert an exception to a string?
22:44:18	flaviu	Not getStackTrace, but to the same format that gets printed
22:45:05	*	vendethiel joined #nim
22:45:38	flaviu	auxWriteStackTrace isn't exposed anywhere :/
22:46:54	*	tinAndi joined #nim
22:48:45	*	superfunc|busy quit (Ping timeout: 265 seconds)
22:49:04	Araq	flaviu: when I expose all the good stuff that is internally used people complain about the shitty API design :P
22:49:27	flaviu	Well, I guess `auxWriteStackTrace` isn't a good thing to expose.
22:49:39	Araq	see newException. was some internal helper, people wanted it badly
22:49:53	Araq	got exported and now we have better means and confused users
22:50:04	flaviu	but something like `$`(ref Exception) that returns the exception name, message, and stacktrace as a single string.
22:50:22	BlaXpirit	the main concern is bloating system module
22:50:31	BlaXpirit	people won't complain about separate modules
22:51:00	Araq	BlaXpirit: ldlework does just that.
22:51:11	flaviu	BlaXpirit: Exceptions are already part of system.nim, and `$` is just omitted functionality IMO.
22:51:26	BlaXpirit	well exceptions should be part of system
22:51:35	BlaXpirit	that's for sure
22:51:59	BlaXpirit	but defining `$` for exceptions in any particular way is not good
22:52:23	ldlework	Well I don't complain that system.nim is too bloated
22:52:27	flaviu	BlaXpirit: The way that the compiler outputs it on the console is the correct way to define `$`
22:52:44	ldlework	I just think that we could mitigate those who do thing it is bloated by refactoring everything to depend on the preamble rather than system.nim
22:52:54	ldlework	we can then break up system.nim into more focused modules
22:53:08	BlaXpirit	flaviu, I'd rather just see exception message and optionally type
22:53:08	ldlework	and not have to import basic string and sequence mainipulations
22:53:56	Araq	ldlework: that's however even this way for scripting langs: import os, os.path, re, sys
22:54:18	Araq	is my typical first python line
22:54:24	ldlework	I don't think those things should be in this preample
22:54:28	ldlework	note you don't do
22:54:47	ldlework	"import basic_string_features" "import things_that_make_sequences_minimally_useful"
22:55:12	flaviu	ldlework: Try out numpy
22:55:21	flaviu	that's what you end up doing
22:55:24	EXetoC	matkuki: only glGetString and glGetStringi have similar signatures in gl.xml, so let's hope those are the only mistakes
22:55:25	ldlework	flaviu: numpy isn't a language
22:55:40	EXetoC	matkuki: you could edit it directly via github if you want to, otherwise I'll fix it
22:55:44	ldlework	flaviu: numpy is a library for a specific collection of domains
22:55:49	ldlework	odd analogy
22:55:59	ldlework	This is a threshold question
22:56:04	ldlework	We already have things in system.nim
22:56:09	ldlework	and we already have shit in stdlib
22:56:11	matkuki	EXetoC: I already opened an issue.
22:56:34	ldlework	so there's already a line, I think our line is just *so ever slightly* crossed into the inconvenient zone
22:56:57	ldlework	Especially compared with default functionality in things like Python, or even modern system languages
22:57:02	EXetoC	matkuki: I don't see how that makes any difference, but I'm looking into it
22:57:40	ldlework	There's only a handful of things I'd move from sequtils and strutils into core, but there is the bloated argument, so my response is lets refactor the core.
22:57:50	*	Matthias247 quit (Read error: Connection reset by peer)
22:58:22	BlaXpirit	Python probably would have more things imported by default, by an order of magnitude, if we actually considered how many methods every built in type has
22:58:33	BlaXpirit	but it's not a problem because documentation is logically separated
22:58:44	Mat4	I think splitting the system module is advantageous. Possible in different levels of system support
22:58:57	ldlework	https://docs.python.org/2/library/stdtypes.html#string-methods
22:59:17	ldlework	While Python's string type does have about 2 dozen methods, they are all hardcore fundamental operations you want to do on strings
22:59:24	ldlework	And Python has a 'strutils' module too
22:59:43	BlaXpirit	no..?
23:00:08	ldlework	There are a couple I'd take off of Python's string type if I was redesigning PYthon in order to make this argument
23:00:16	ldlework	but only like 3 or 4 that seem like anachronisms
23:00:33	ldlework	BlaXpirit: no..? what?
23:00:34	*	BillsPC quit (Remote host closed the connection)
23:00:35	*	dtscode quit (Quit: ZNC - http://znc.in)
23:01:02	*	dts|pokeball joined #nim
23:01:31	BlaXpirit	I don't know any such "strutils" module, and there is no other module that is like nim's strutils
23:01:50	ldlework	https://docs.python.org/2/library/string.html
23:01:58	ldlework	BlaXpirit: better disagree before you're sure
23:02:14	BlaXpirit	just look at that module and what it defines -_-
23:02:31	BlaXpirit	I wasn't wrong, in any case
23:02:31	Araq	Mat4: splitting system.nim has been tried at least twice now. it's harder than it looks and at the end of the day doesn't help much. but ymmv
23:02:34	ldlework	Stuff that doesn't belong on the type?
23:02:59	EXetoC	matkuki: done
23:02:59	BlaXpirit	ldlework, it just defines string constants!
23:03:07	ldlework	BlaXpirit: and... several other things
23:03:09	ldlework	god you are lazy
23:03:19	ldlework	please stop participating in public debate until you can have a better attitude
23:03:23	ldlework	seriously
23:03:39	BlaXpirit	what other things are there that are not available without importing it?
23:03:51	ldlework	capwords?
23:03:52	BlaXpirit	just Template, it seems
23:03:54	ldlework	formatting?
23:04:04	BlaXpirit	formatting is available as a method of str
23:04:09	ldlework	all of the deprecated functionality ives there
23:04:22	ldlework	BlaXpirit: string.Formatter
23:04:31	matkuki	EXetoC: Great, thanks.
23:04:37	ldlework	BlaXpirit: this is what happens when you take a fundamentalist position, just to argue
23:04:41	ldlework	what the fuck is your actual position?
23:05:58	matkuki	EXetoC: By the way, when you say edit directly via github, what do I need to do? I'm haven't used github that much.
23:06:27	ldlework	EXetoC: when looking at files on github, there is usually an edit button nearby
23:06:47	*	superfunc joined #nim
23:07:02	EXetoC	*matkuki
23:07:10	ldlework	right, sorry :)
23:09:11	matkuki	I don't see the button. I see only "Raw", "Blame", "History" and a few others.
23:09:30	ldlework	matkuki: are you a contributor to the repo?
23:09:50	ldlework	matkuki: its the little pencil icon
23:11:18	EXetoC	not everyone knows what a pencil is anymore
23:11:28	ldlework	hehe
23:11:58	matkuki	Idlework: Oh I see. Sorry, I'm new to this. Got it for next time!
23:13:30	*	saml_ quit (Read error: Connection reset by peer)
23:25:21	*	saltylicorice joined #nim
23:26:51	Araq	hi saltylicorice welcome
23:28:29	*	z1y joined #nim
23:28:59	*	repax quit (Quit: repax)
23:32:05	*	z1y quit (Client Quit)
23:36:57	*	dts|pokeball quit (Quit: Leaving)
23:37:50	MyMind	I tried to compile aporia on os x and got:
23:37:52	MyMind	CustomStatusBar.nim(41, 24) Error: undeclared identifier: 'False'
23:37:54	MyMind	FAILURE: Execution failed with exit code 256
23:38:11	EXetoC	matkuki: do you know about the automatic error checking?
23:38:18	*	dtscode joined #nim
23:38:21	EXetoC	for the opengl interface
23:38:23	Araq	MyMind: used aporia from github?
23:38:32	MyMind	nimble install aporia
23:38:35	*	dtscode quit (Remote host closed the connection)
23:38:38	Araq	should be fixed since a few weeks
23:38:48	matkuki	EXetoC: No, how does it work?
23:38:58	*	tinAndi quit (Quit: ChatZilla 0.9.91.1 [Firefox 34.0/20141125180439])
23:39:02	Araq	maybe nimble doesn't use the latest commit but some "official stable" commit?
23:39:14	*	dtscode joined #nim
23:39:15	*	saltylicorice quit (Ping timeout: 256 seconds)
23:39:24	flaviu	MyMind: nimble install aporia@#head
23:39:24	MyMind	maybe
23:41:18	MyMind	flaviu: that command doesn't work on zsh it does on bash
23:41:25	EXetoC	matkuki: you call a gl function and don't think more about it
23:41:33	def-	MyMind: "aporia@#head"
23:41:42	EXetoC	it's actually a massive wrapper, created with the help of a macro
23:41:43	flaviu	MyMind: ZSH does something special with # iirc, just escape it.
23:42:14	MyMind	aporia installed succesfully
23:43:12	MyMind	could not load: libpango-1.0.dylib
23:43:48	matkuki	EXetoC: the "wrapErrorChecking" macro?
23:43:53	EXetoC	matkuki: yes
23:44:11	matkuki	I'm just now looking at the code.
23:44:27	EXetoC	I could've just done it without a macro since I was generating from the spec, but where's the fun in that?
23:44:43	Mat4	have someone tried compiling Nim with Visual C ?
23:45:02	matkuki	EXetoC: :)
23:45:05	Araq	Mat4: iirc Demos at least does that regularly
23:46:45	Araq	and I'll do it in a few days too
23:47:35	flaviu	Is there a way to get the exception name other then casting?
23:48:27	Araq	flaviu: dunno, why is that so important? you cannot do much with the name
23:48:47	*	dtscode quit (Quit: ZNC - http://znc.in)
23:49:07	flaviu	Araq: unittest.nim's exception handling is inadequate. It's best to give as much information as possible to the user.
23:49:29	*	dtscode joined #nim
23:49:51	Araq	flaviu: unittest itself is inadequate
23:50:04	flaviu	Yeah, but at least it can give me a stack trace.
23:50:23	Araq	if brokenCode: writeStackTrace()
23:50:27	ldlework	:(
23:50:42	Araq	everything can give you a stacktrace, it's a simple call
23:50:59	flaviu	I still don't like it, but ok.
23:51:23	ldlework	Don't stack traces usually include what went wrong?
23:51:38	ldlework	Not just "someone called writeStackTrace from here --->"
23:51:57	Araq	ldlework: getStackTrace(someException) then
23:52:14	ldlework	Araq: ah okay
23:52:28	*	johnsoft quit (Ping timeout: 255 seconds)
23:53:14	*	johnsoft joined #nim
23:55:12	*	adam_s left #nim ("Leaving")
23:58:18	flaviu	Nice commit: https://github.com/flaviut/Nim/commit/7f02174ded35fc39f93657fbceb2a3dcd3885304 :P
23:58:41	flaviu	I have no idea what happened there