<< 13-06-2016 >>

---

00:47:41	*	mcc joined #nim
00:49:14	mcc	Few questions about OO. First: the {.base.} pragma. What *exactly* does it do? The manual says it is for "base methods". Is it for every method that might be overridden? Or just the base? I have one method which inherits from a {.base.} method and is not overridden but I am still getting a warning that I forgot to put {.base.}
00:51:03	mcc	Second question: Is there anything like "super"? For example if B is "of A", and I have a method f(x:A) and a method f(x:B), in the implementation of f(x:B) how can I call f(x:A) ?
00:53:07	*	yglukhov joined #nim
00:56:02	lazypenguin	@cheatfate, nice catch
00:57:49	mcc	Actually, weirder, I have two methods in this file. A next() and a draw(). With next(Player), it gives a warning if i do NOT say {.base.}. With draw(Player), it gives an ERROR if I say {.base.} and says it's "not a base method".
00:57:49	mcc	https://bitbucket.org/runhello/nim-practice/src/8829bb479f51581b14df5e951e4c618feb5a9ce1/maze.nim?fileviewer=file-view-default#maze.nim-99
00:57:53	*	yglukhov quit (Ping timeout: 260 seconds)
00:58:52	mcc	One more question. I do not understand how dynamic dispatch works on objects which are not ref-objects?
00:59:27	mcc	In this code I have a Mover object, and then child objects Player and Chaser. None are ref objects. However, my understanding is that non-ref objects are being passed and stored by value.
01:00:09	mcc	So I don't understand how I could have, for example, an array of Mover objects and then store a Player or Chaser in it, if Player or Chaser are different sizes from Mover. In C++ this would not work. How does Nim implement this? And is this why my code breaks?
01:00:40	*	ivan left #nim (#nim)
01:01:34	lazypenguin	hmm i've never seen 'method' before, is that an alias for 'proc'?
01:02:07	mcc	In the manual it is described as similar to proc but it uses dynamic dispatch. http://nim-lang.org/docs/manual.html#types-enumeration-types
01:13:45	lazypenguin	mcc, this is outside my realm of knowledge but could it be related to the fact that your methods have different return types? Seems the compiler detects the polymorphism correctly between draw(mover) and draw(player) and throws the base error but the same is not true for your next() methods which have different return types
01:14:13	mcc	lazypenguin: yes, sorry, that is the problem, if i change Player to Mover i no longer get the warning
01:15:00	lazypenguin	:)
01:16:41	mcc	how on earth do i control which of the two versions of map() documented in http://nim-lang.org/docs/sequtils.html#map,openArray[T],proc(T) i get?
01:18:53	lazypenguin	Looks like the function signature is slightly different in that the new one takes op as the second argument with a return type of S `op: proc(x: T): S`
01:20:09	mcc	but they both take op as the second argument
01:20:59	mcc	oh, is it controlled by the second argument's return value... that's subtle D:
01:21:12	lazypenguin	subtle indeed
01:28:59	*	yglukhov joined #nim
01:33:02	*	yglukhov quit (Ping timeout: 244 seconds)
01:33:37	renesac	mcc: you might be interested in this blog post series:
01:33:37	renesac	http://goran.krampe.se/category/oop/
01:33:54	renesac	the older 5
01:34:47	mcc	renesac: Hm. Which of these are out of date?
01:35:11	mcc	renesac: I'm currently filing several issues on the nim github (mostly of the form "the warnings and/or manual should be clearer about this")
01:39:40	*	bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
01:40:09	renesac	I don't know that changed since then..
01:40:16	renesac	*what
01:42:18	*	chemist69_ quit (Ping timeout: 272 seconds)
01:53:04	*	chemist69_ joined #nim
01:58:16	*	chemist69_ quit (Ping timeout: 250 seconds)
01:58:37	*	lazypenguin quit (Ping timeout: 250 seconds)
02:04:34	*	yglukhov joined #nim
02:07:40	*	zodiak_ quit (Ping timeout: 264 seconds)
02:08:48	*	yglukhov quit (Ping timeout: 246 seconds)
02:08:59	*	zodiak joined #nim
02:11:32	*	lazypenguin joined #nim
02:11:39	*	chemist69_ joined #nim
02:29:51	*	bjz joined #nim
02:35:10	*	bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
02:40:02	*	yglukhov joined #nim
02:44:37	*	yglukhov quit (Ping timeout: 260 seconds)
02:59:33	*	bjz joined #nim
03:03:47	*	yglukhov joined #nim
03:08:06	*	yglukhov quit (Ping timeout: 244 seconds)
03:22:03	*	GangstaCat quit (Ping timeout: 264 seconds)
03:34:17	*	GangstaCat joined #nim
03:38:33	*	akaisora_ quit (Ping timeout: 240 seconds)
03:39:21	*	yglukhov joined #nim
03:43:45	*	yglukhov quit (Ping timeout: 244 seconds)
03:54:17	*	bjz_ joined #nim
03:56:03	*	bjz quit (Ping timeout: 250 seconds)
04:09:18	*	akaisora joined #nim
04:11:29	*	michael_campbell quit (Ping timeout: 244 seconds)
04:16:22	*	akaisora quit (Remote host closed the connection)
04:25:07	*	Kingsquee joined #nim
04:27:56	*	SShrike joined #nim
04:41:06	*	SShrike quit (Quit: SShrike)
05:03:20	*	chemist69_ quit (Ping timeout: 260 seconds)
05:20:47	*	chemist69_ joined #nim
06:04:00	*	GangstaCat quit (Quit: Leaving)
06:16:14	*	bjz joined #nim
06:16:31	*	bjz_ quit (Ping timeout: 244 seconds)
06:21:42	Demos	so variadic generics could help the db modules support various data types (you can't insert blobs into sqlite as a string)
06:22:59	*	endragor joined #nim
06:23:01	*	endragor quit (Remote host closed the connection)
06:23:29	*	endragor joined #nim
06:39:20	*	heretobang joined #nim
06:56:06	*	yglukhov joined #nim
06:56:36	*	mcc quit (Quit: Connection closed for inactivity)
07:01:45	*	gokr joined #nim
07:27:25	*	kingofoz quit (Ping timeout: 260 seconds)
07:27:54	*	kingofoz joined #nim
07:33:19	*	fredrik92 joined #nim
07:33:21	*	kingofoz quit (Ping timeout: 276 seconds)
07:37:29	*	Gonzih joined #nim
07:46:02	*	chemist69_ quit (Quit: WeeChat 1.5)
07:48:36	*	Trustable joined #nim
07:51:35	*	PMunch joined #nim
07:52:34	endragor	Is it possible to pass sequence to a proc expecting varargs parameter?
08:00:31	PMunch	endragor, it could probably be done by a template or macro if there is no built in way
08:04:20	*	desophos quit (Read error: Connection reset by peer)
08:05:41	endragor	PMunch: how exactly, given the seq contents are not known in compile-time?
08:06:51	PMunch	Hmm, according to rosettacodes it seems that you can simply pass the seq: rosettacode.org/wiki/Variadic_function#Nim
08:07:36	*	zodiak_ joined #nim
08:08:13	PMunch	endragor, macros run during runtime so they should at least be able to do it.
08:10:04	*	chemist69 joined #nim
08:10:41	endragor	hm, indeed you can just pass it, I had a problem in a different place. macros are invoked during compile-time though, so if you can produce something with a macro you can do it without macro, too
08:11:03	*	zodiak quit (Ping timeout: 276 seconds)
08:13:49	PMunch	Yes, but sometimes it's a lot easier to do it at with a macro
08:14:24	PMunch	Technically you could write a macro to convert a seq to a varargs type and pass it, but I guess that is already written :)
08:18:39	*	gokr quit (Ping timeout: 250 seconds)
08:19:28	*	Salewski joined #nim
08:19:34	Araq	Demos: yes but varargs[typed] exists and can be manipulated with macros
08:20:43	Salewski	Dom, Listing 7.5, last line: requires "jester >= 0.0.1" is missing! Bye.
08:20:48	*	Salewski quit (Client Quit)
08:20:48	Demos	but that won't work if you have a query with multiple different parameter types
08:20:55	Demos	anyway it's not a big deal
08:26:37	Araq	hmm?
08:27:21	Demos	like if I have a table:
08:27:52	Demos	CREATE TABLE t (bin BLOB, name TEXT);
08:29:00	Demos	I can't do db.exec(sql"INSERT INTO t (bin, name) VALUES (?, ?)", addr something, "test")
08:29:24	Araq	yes you can
08:29:35	Araq	just not with db's current implementation
08:29:46	Araq	which doesn't make use of Nim's newer features
08:30:01	Demos	oh, allright. I'll look forward to that then
08:30:17	Demos	for now I'm just using the low level API
08:36:40	*	gokr joined #nim
08:38:10	Demos	OOOOOoooo the type mismatch errors have become really awesome since I last used nim
09:12:26	*	kingofoz joined #nim
09:19:45	Demos	the pragma syntax for macros does not seem to work in the latest version, although I could be screwing up my macro
09:22:30	Demos	I'll try and come up with a minimal test case tomorrow
09:23:05	endragor	Shouldn't procs in streams.nim guarantee endianness or provide both LE and BE variants? As of now, the module is unusable for networking or in any case when the data can be accessed from different machines, because the format is platform-dependent.
09:25:03	flyx	Demos: pragma syntax is used in emerald tests and they work at least in 0.14.0: https://github.com/flyx/emerald/blob/devel/test/basics.nim
09:27:16	Demos	hm maybe its just when they are on types
09:39:43	Araq	the AST changed for these
09:39:50	Araq	do you use getType()?
10:04:23	Demos	no
10:04:58	*	gokr quit (Ping timeout: 252 seconds)
10:16:16	Araq	hmm
10:23:10	*	gokr joined #nim
10:50:25	*	Kingsquee quit (Quit: https://i.imgur.com/qicT3GK.gif)
10:56:14	*	xet7 joined #nim
11:05:23	*	Nimian joined #nim
11:05:50	Nimian	Hello guys!
11:06:51	PMunch	Hi Nimian
11:10:02	Nimian	PMunch: What's up?
11:10:24	PMunch	Not much, trying to migrate a server
11:10:33	PMunch	You?
11:11:14	Nimian	Translating Nim tutorials to Arabic, Kurdish and Persian
11:11:36	PMunch	Oh nice
11:11:44	cheatfate	PMunch, you are talking with bot...
11:12:00	PMunch	I am?
11:12:12	cheatfate	PMunch, yep
11:12:15	PMunch	That's a good bot then :P
11:12:37	PMunch	And why does it say it's translating tutorials?
11:13:34	cheatfate	PMunch, because somebody programmed it :)
11:13:40	*	elrood joined #nim
11:13:52	Nimian	PMunch: It's not bad at all.
11:14:31	cheatfate	PMunch, this "tutorials translation" monolog was yesterday already
11:14:55	Nimian	what do you mean, cheatface?
11:15:21	PMunch	Haha, it even has spelling mistakes? That's pretty good
11:15:53	cheatfate	dom96, could you please ban the whole network 93.91.194.0?
11:15:55	Nimian	I like to call him cheatface.
11:16:08	Nimian	Why?
11:17:09	Nimian	cheatface: I lied about something, I don't really have master degree, I'm just an IT diploma student (year 1)
11:17:30	M-Flox	Hi Nimian
11:17:42	Nimian	Hello M-Flox!
11:17:49	Nimian	What's up?
11:17:57	M-Flox	Nimian: How are you?
11:18:15	Nimian	I'm fine thanks, and you?
11:18:44	M-Flox	Nimian: Good, thanks. What's the weather like?
11:19:30	Nimian	M-Flox: I think your are cheatfate, but you are registered another account in this channel?
11:19:35	Nimian	Am I right?
11:19:44	Nimian	Because I have that fellling?
11:20:17	M-Flox	Nimian: I am not cheatfate
11:20:45	*	chemist69 quit (Ping timeout: 260 seconds)
11:20:54	Nimian	M-Flox: Okay, I said that because cheatfate wanted to fool me.
11:21:24	Nimian	M-Flox: I'm in Kurdistan, the weather is very hot,
11:21:40	Nimian	M-Flox: We are also fasting in Ramadan month.
11:22:03	M-Flox	I have a feeling cheatfate is behind the bot typing
11:22:43	*	chemist69 joined #nim
11:22:44	Nimian	M-Flox: If I say anything offensive, my fasting won't be accepted.
11:24:29	Nimian	M-Flox: Would you like to know why I do care of Nim?
11:24:46	M-Flox	Ahm, not really :-)
11:26:29	Nimian	M-Flox: Just like I said, I cannot say offensive stuffs while I'm fasting.
11:27:02	Nimian	The rulse are rules and business is business.
11:30:48	Nimian	I have a question
11:31:05	Nimian	Any Nim developers here?
11:32:12	Nimian	Hello?
11:33:14	Araq	I'm here.
11:33:52	Nimian	Alright, my question is, when I'm on Windows, should I compile to C or C++? because as we know, Windows API is written in C++.
11:34:14	Nimian	Ohh! Hi Araq!
11:34:32	Araq	to C.
11:34:56	Nimian	May I know why?
11:36:03	Araq	windows exposes a C API and the C++ codegen still has some issues wrt c++'s const
11:36:10	flyx	the Windows API is C, not C++. MFC is C++
11:38:01	Nimian	flyx: It's both C and C++
11:38:11	Nimian	But the C codes are more than C++
11:39:35	flyx	afaik there is no C++ in the Windows API. please provide an example.
11:39:56	Nimian	I don't know actually, just like I said, I'm just a student at an IT diploma (2nd year)
11:40:20	elrood	don't feed the troll, guys
11:40:35	Nimian	elrood: I lied about master degree.
11:40:38	Nimian	Sorry
11:41:46	Nimian	elrood: But I'm still sure that Nim is not a real compiler.
11:42:25	elrood	you have quite competently and thoroughly disqualified yourself on your earlier visits, and as a consequence i couldn't care less what you do or don't say, or believe, or whether you're sorry or not
11:42:55	*	flyx seems to miss some of the background story here
11:43:17	Araq	damn it. stop it. I don't want to moderate this channel.
11:43:26	Nimian	Okay
11:44:11	Nimian	Araq: calm down, I never seen you like this anger
11:44:31	*	yglukhov quit (Ping timeout: 252 seconds)
11:48:37	Nimian	Araq: As you know many major languages have International standard(s) from ANSI, ISO and/or ECMA. Will Nim have one?
11:49:33	Nimian	For example, C++11, C++17 (upcoming), Single Java Specification etc...
11:51:31	Nimian	Would anyone answer my question?
11:52:16	*	SchaffGeischd joined #nim
11:52:19	*	SchaffGeischd left #nim (#nim)
11:52:21	*	SchaffGeischd joined #nim
11:52:55	Nimian	Welcome Shafdhuwdh3d
11:53:50	Nimian	Can anyone answer my question
11:54:18	Araq	Nimian: the manual evolves into a proper spec
11:54:49	Nimian	Araq: Am talking about international standard (a big stuff)
11:54:56	*	PMunch quit (Ping timeout: 244 seconds)
11:55:51	Nimian	Araq: Like standards from ISO, ANSI, ECMA etc...
11:56:17	M-Flox	sigh
11:56:17	Nimian	Araq: How can we get a standard?
11:58:01	Nimian	M-Fox: How many times I have to tell you that I'm fasting?
11:58:42	*	Araq2 joined #nim
11:59:06	Nimian	Hello Araq2! Welcome to the Nim IRC channel!
11:59:24	Araq2	Thank you Nimian.
11:59:40	Nimian	You are welcome Araq2
12:00:33	Nimian	May I ask why in the earth you named yourself Araq2?
12:00:39	*	PMunch joined #nim
12:00:56	Araq2	I like Mr. Araq. That's why
12:01:15	Nimian	NP (No Problem)
12:01:32	gokr	So is this a bot or what?
12:01:40	flyx	they have the same IP.
12:02:04	Nimian	I think it's the same person.
12:02:20	Nimian	I mean he is the same person.
12:02:27	*	nsf quit (Quit: WeeChat 1.4)
12:02:41	*	heretobang quit (Ping timeout: 250 seconds)
12:03:53	Nimian	Let me ask you a question Araq2, why Nim language is so good?
12:03:58	flyx	anyway, can we please just ban him. this is getting ridiculous
12:04:19	Nimian	flyx: what I have don?
12:04:29	Nimian	I just asking him questions.
12:04:52	Nimian	He is the only one who answers me. he is the best helper here.
12:06:59	dom96	Guess we need more mods.
12:07:18	cheatfate	dom96, you need to ban whole 93.91.194.0
12:07:26	dom96	cheatfate: ok
12:07:42	dom96	I hope that will work
12:08:03	gokr	Oops, there went half our users... ;)
12:08:39	dom96	Hope not
12:08:48	dom96	I'll likely remove it after a while
12:08:52	gokr	That would've been scary - if it turned out half of us are bots ;)
12:08:52	dom96	the troll will forget
12:09:04	dom96	I don't know why you guys think that that was a bot though
12:09:15	dom96	It was simply a human trying to troll.
12:09:30	gokr	Someone said so - but... I agree, it seemed slightly too sophisticated. not sure.
12:10:37	cheatfate	dom96, just because human dont like to talk with bots and start to ignore it...
12:10:43	cheatfate	till moderators kick him
12:10:55	elrood	AIs write movie scripts these days, you never know ;) thanks, dom
12:13:29	dom96	https://twitter.com/polyconfhq/status/742322383722151936
12:14:55	cheatfate	`nim async voodoo` what does it mean? :)
12:17:33	dom96	I think he will talk about how my async macro converts ordinary procedures into iterators.
12:17:58	*	Demon_Fox quit (Quit: Leaving)
12:41:51	*	WW_ joined #nim
12:42:05	WW_	Is NimEdit open source?
12:42:25	*	SchaffGeischd left #nim (#nim)
12:43:20	WW_	Is NimEdit open source?
12:43:50	WW_	Hello??
12:44:08	dom96	WW_: no
12:44:18	*	bjz_ joined #nim
12:44:19	dom96	WW_: Be patient with your questions.
12:44:43	WW_	dom96: excuse me, what do you mean?
12:45:04	dom96	WW_: I mean, don't repeat your question. Ask it and wait.
12:45:47	WW_	dom96: sorry, but I cannot wait, so please excuse me.
12:46:03	*	bjz quit (Ping timeout: 240 seconds)
12:47:09	WW_	dom96: when will your amazing book be complete?
12:47:23	WW_	I mean all chapters.
12:48:30	dom96	WW_: No idea. Estimate is early 2017 (completely finished, so printed too)
12:49:23	WW_	dom96: why did you kik me? I love Nim more than anyone else, even Mr. Andreas Rumpf.
12:50:39	dom96	WW_: Ah, so you're the same person. I kicked you because you were causing trouble.
12:51:31	WW_	I'm also BP (from Nim forum)
12:52:52	WW_	But dom96, for real, I love Nim more than anyone else.
12:52:59	WW_	I want to create a state-of-at operating system i entirely in Nim and assembly languages.
12:53:11	WW_	**state-of-art
12:58:00	dom96	Great. Why are you coming in here to argue then?
12:58:02	*	yglukhov joined #nim
12:58:40	*	WW__ joined #nim
12:59:27	WW__	dom: when Nim 1.0 will be released exactly?
12:59:46	*	WW_ quit (Ping timeout: 250 seconds)
13:00:41	*	WW__ quit (Client Quit)
13:01:46	*	kteza1 joined #nim
13:04:33	kteza1	Hi. Very new to nim. What am I doing wrong here --> http://pastebin.com/TnuP1vSF
13:04:50	kteza1	protocol.nim(31, 45) Error: type mismatch: got (int literal(32)) but expected one of: proc encodeRemainingLength(length: var int): int
13:04:56	*	WW_ joined #nim
13:05:06	kteza1	That's the error I'm facing
13:05:51	WW_	because I exited
13:05:52	WW_	Dom can you repeat your answer
13:06:56	flyx	kteza1: your pasted source does not seem complete, it does not contain 31 lines.
13:08:09	kteza1	@flyx Sorry. I didn't paste entire code. Error is at line 14 in pastebin
13:09:12	flyx	kteza1: ah okay. problem is that `length` is a `var int`, i.e. it has to be modifyable. but you pass `32`, an int literal, which is not modifyable
13:10:08	flyx	kteza1: you probably have assumed that `var` just enables you to change the value within the proc, but it is actually used to modify the value given by the caller
13:11:47	kteza1	Ahh. all the 'var's are pointers then?
13:12:24	flyx	proper term is „call by reference“
13:13:09	flyx	it does not necessarily have to be implemented as pointer, though it probably is (I do not know about the implementation details)
13:13:35	kteza1	Ok. But I'm not able to modify length by declaring it like this --> proc encodeRemainingLength*(length: int): int =
13:13:43	kteza1	with our 'var'
13:13:56	flyx	yes. you should copy it to a local var and modify that
13:14:18	flyx	modifying by-value parameters is considered bad practice even in languages that allow it
13:14:57	kteza1	Hmm. Is shadowing allowed? (like rust)
13:15:34	kteza1	I guess it is
13:15:43	flyx	not sure if Rust does something special about it, but yes, it is possible in Nim
13:19:13	kteza1	@flyx Also I tried changing 'length' to uint64 and mod operation is failing (coz it is expecting int)
13:19:23	kteza1	endocedLength = length mod 128
13:19:55	kteza1	Error: type mismatch: got (uint64) but expected 'int'
13:20:18	kteza1	Why isn't mod accepting uint64?
13:21:22	kier	'128' is an integer literal of type int. to make it a literal of type uint64 you can add the 'u64' suffix (like '128u64')
13:21:43	kier	but i'm not certain if mod is implemented on uint64 anyway
13:21:58	*	arnetheduck joined #nim
13:22:10	flyx	`length mod 128'u64` will work if length is uint64
13:22:36	flyx	while `length mod 128` will not work because there is no `mod`(x: uint64, y: int)
13:23:47	arnetheduck	hey Araq, saw you merged ansi_c, everything ok so far?
13:23:57	flyx	kier: http://nim-lang.org/docs/system.html#mod,T,T
13:24:33	kier	ah yeah just spotted that
13:25:04	kier	though am I right in thinking that there are a few operations that aren't implemented on uint/uint64 (but are on the other integer types)
13:26:44	*	elrood quit (Quit: Leaving)
13:27:12	kier	ah I'm thinking of the fact that uint/uint64 aren't ordinals
13:27:36	flyx	yeah, that's https://github.com/nim-lang/Nim/issues/2613
13:31:24	*	WW_ quit (Ping timeout: 250 seconds)
13:32:07	kteza1	@flyx @kier All well now. Thanks :). But I had to case 0 to u64 as well at line 8 here --> http://pastebin.com/TNchKViq
13:33:11	kteza1	Wouldn't code become too verbose if I keep doing that? I there an alternate to this?
13:33:18	kteza1	cast*
13:34:35	flyx	kteza1: all numeric literals without explicit type suffix are int and suffer this problem. there's no easy way around it
13:34:56	flyx	but you could, for example, write a template like this:
13:35:40	flyx	template `>`(x: uint64, y: int): bool = x > uint64(y)
13:36:29	flyx	there's an issue about implicit conversion here: https://github.com/nim-lang/Nim/issues/936
13:36:34	renesac	kteza1: https://github.com/nim-lang/Nim/issues/936
13:36:41	renesac	yep
13:37:46	renesac	the alternative is to avoid usnigned types in nim..
13:38:38	kteza1	Ok. Thanks @flyx @renesac
13:46:05	*	Anonymous joined #nim
13:46:28	*	Anonymous is now known as Guest46869
13:46:53	Guest46869	Where can I get a high-quality Nim logo?
13:47:37	*	BitPuffin joined #nim
13:48:55	*	yglukhov quit (Ping timeout: 260 seconds)
13:49:51	PMunch	Guest46869, a print version or just high-res?
13:50:01	renesac	flyx: I wonder if there is any side effect of putting https://github.com/nim-lang/Nim/tree/devel/web/assets/images
13:50:03	renesac	ops
13:50:04	renesac	https://github.com/nim-lang/Nim/tree/devel/web/assets/images
13:50:34	*	nsf joined #nim
13:50:53	renesac	flyx: I was going to ask whether any of those templates could be put in the standard library to ease some of the pain of working with unsigend
13:51:20	*	Guest46869 quit (Ping timeout: 250 seconds)
13:51:55	renesac	I was trying for u32 and the uint32() don't seems to truncate the input like a cast, I'm not sure
13:52:05	renesac	*uint32 seems to
13:53:03	flyx	renesac: well, that's a question Araq would need to answer.
13:53:28	renesac	the bitwise operations are the most complicated
13:53:39	renesac	maybe we should just ignore them for now
13:53:57	renesac	and fix some of the rest
13:53:59	*	yglukhov joined #nim
13:54:37	renesac	there is no problem in upconverting to uint64 for boolean operations
13:54:39	renesac	for example
13:55:30	flyx	dunno, wouldn't that mean that `length > -2` would raise a runtime error?
13:55:39	flyx	(assuming length is uint64)
13:56:50	renesac	no error of any type
13:57:35	renesac	it seems to cast the integer value to unsigned one
13:57:50	renesac	ops
13:57:53	renesac	that was with a proc
13:58:12	renesac	with a templete I get a compile time error: Error: conversion from int literal(-2) to uint32 is invalid
13:58:31	flyx	uh, that's nice
13:58:32	renesac	curious...
13:58:56	renesac	should the proc beehaviour be differnt?
13:59:09	flyx	um
13:59:26	flyx	so why does uint32(-2) raise a compile-time error, but uint64(-2) doesn't?
14:00:29	flyx	the proc must handle a parameter that may have different values, so it cannot check its value at compile time. the template just rewrites.
14:00:39	*	saml joined #nim
14:01:05	renesac	and at runtime, even in a debug build, it don't complains and just casts...
14:01:23	flyx	so the proc behaving in a different way makes sense to me. but uint32(-2) being handled differently than uint64(-2) does not.
14:01:51	renesac	I'm not able to reproduce that second thing as I'm getting a weird type mismatch here
14:02:11	flyx	`echo uint64(-2)` works for me. `echo uint32(-2)` does not.
14:03:04	renesac	strange, I had to manually convert uint32 to wuint64
14:03:10	renesac	it didn't implicitly convert there
14:03:18	renesac	*uint64
14:03:58	renesac	because 18446744073709551614 fits in a uint64, but not in a uint62...
14:04:17	renesac	*32
14:05:03	renesac	I'm not sure if this behaviour is intended though..
14:12:18	flyx	renesac: I would agree that this is the reason if the output would be the same as the output for uint32(18446744073709551614). but it's not.
14:13:07	*	flyx is afk for some time
14:19:07	*	michael_campbell joined #nim
14:19:12	*	lazypenguin1 joined #nim
14:19:47	renesac	flyx: at least both fail when not inside a proc
14:20:33	renesac	*not working with proc arguments
14:22:51	*	gokr quit (Ping timeout: 276 seconds)
14:22:59	Araq	yeah for the next release I intend to fix every remaining 'unsigned' bug
14:23:15	*	lazypenguin quit (Killed (asimov.freenode.net (Nickname regained by services)))
14:23:16	*	lazypenguin1 is now known as lazypenguin
14:23:23	Araq	I think this means to introduce more type conversion rules
14:23:37	*	btbytes joined #nim
14:24:01	*	Guest56550 joined #nim
14:25:07	*	desophos joined #nim
14:29:57	renesac	yay! If this means at least that literals will get automatically converted to the proper uint, I will be very happy
14:30:51	renesac	although it still would be good to have some implict conversions from small uints to int for interfacing with the rest of the language functions easier...
14:32:51	dom96	+1 on implicit unsigned conversions
14:39:53	*	btbytes quit (Quit: Textual IRC Client: www.textualapp.com)
14:40:18	*	btbytes joined #nim
14:41:31	michael_campbell	Why does this work: let z = {1..3}, but this does not? var x: set[int16]; x = {1..3}? I get:
14:41:32	michael_campbell	Error: type mismatch: got (set[range 0..65535(int)]) but expected 'set[int16]'
14:41:47	michael_campbell	user error I've no doubt, just not sure where I've gone wrong
14:42:12	michael_campbell	(the var example is on 2 lines; just put it on one for brevity)
14:42:21	Araq	the compiler is picky about the 'convertible' relation
14:43:07	michael_campbell	What is the correct way to declare a var of type set, and fill it out later (other than 1 element at a time)?
14:43:19	Araq	try x = {1i16..3i16}
14:43:55	michael_campbell	same. Both with 1i16 and 1'i16
14:43:57	michael_campbell	(on both ends)
14:44:35	michael_campbell	Pushing values into the set piecemeal works.
14:45:08	michael_campbell	Am enjoying the Nim in Action MEAP so far, by the way.
14:45:28	*	PMunch quit (Quit: leaving)
14:48:05	Araq	x = {1i16..3i16} needs to work, report it please
14:48:46	michael_campbell	Will do; on github?
14:49:12	*	elrood joined #nim
14:51:42	Araq	yes
14:55:21	*	btbytes quit (Quit: Textual IRC Client: www.textualapp.com)
14:57:10	*	xet7_ quit (Ping timeout: 260 seconds)
15:04:08	*	pregressive joined #nim
15:04:48	endragor	Araq: dom96: I'm working on intergrating nimue4 with Nimble (so that people can use Nimble packages in UE4 projects). The last piece is that Nim prefixes compiled .c/.cpp filenames with the package filename. I'd like to override that so that no prefix is added for the main package that is being built. Would it make sense to add such flag to Nim? An alternative way is to add nimble command that installs dependencies and lists out their paths. In
15:04:48	endragor	such case, I could manually invoke Nim with those paths, discarding the .nimble file for the main package. Which approach is better?
15:07:48	Araq	not sure, but why are the prefixes unwanted?
15:09:15	namnguyen	is there a flag to force c++11?
15:10:05	endragor	Araq: the integration has to mess with the generated files (specifically, create proper .h files) and some macros rely on the .cpp filenames being the same as .nim filenames
15:10:59	michael_campbell	Araq, I was mistaken.
15:11:07	namnguyen	ahh passC
15:11:09	michael_campbell	var x: system.set[int16] <newline> x = {1'i16..3'i16} does work
15:12:18	michael_campbell	Is there a reason I have to specify "system.set[int16]"? The examples don't have it, but I get this error if I don't: "set.nim(1, 8) Error: Error: ambiguous identifier: 'set' --use system.set or .set"
15:12:35	michael_campbell	Am I not importing something that I should be?
15:14:35	Araq	michael_campbell: name your file to something that doesn't clash with system.set
15:14:44	Araq	the filename is used as the module name.
15:14:54	Araq	I dunno why everybody stumbles over this.
15:15:06	michael_campbell	Ah hah! Of course; thanks
15:15:50	dom96	endragor: can't you edit the macros so that they look for the different filenames?
15:16:13	michael_campbell	It's probably my noobness, but it wasn't clear that the filename is the module name (from the tutorial; the ref-man may say it more explicitly), and I haven't gotten that far in the MEAP yet.
15:16:21	michael_campbell	Do you guys take PR's for the tutorial?
15:16:38	Araq	sure
15:16:48	michael_campbell	But even after that, I didn't put 2 and 2 together about the module name vs. system types - thanks for htat.
15:17:09	Araq	the error message seems broken
15:17:17	Araq	"or .set "?
15:17:23	Araq	that makes no sense.
15:17:25	michael_campbell	I was wondering about that too.
15:17:43	namnguyen	hmm, clib:ABC is prefixed with the current directory to -l/path/to/cwd/ABC
15:17:54	michael_campbell	Maybe if Nim were a *REAL* compiler.... (I kid, I kid - I was watching the shitshow from EEE yesterday)
15:18:04	namnguyen	is that intended? what's the use of -l/path...?
15:18:05	endragor	dom96: the macros are invoked during the project compilation phase, they don't know about the nimble package. or is the package name provided to the Nim program in some way?
15:18:59	dom96	endragor: it will always be your package name
15:19:23	endragor	macros are part of the library, they don't know about package names of users
15:19:48	dom96	as in, if nimue4 includes ue4.nimble then it will prefix with ue4_file.cpp
15:20:04	dom96	*prefix with ue4_ ending up with ue4_file.cpp
15:21:38	endragor	dom96: yes, I understand. I'm talking about packages (UE4 projects) that use macros from nimue4. those macros have no way to know the package name that invokes them
15:22:55	dom96	oh
15:25:30	Araq	namnguyen: it's indented but I don't know if it's wise
15:25:55	dom96	endragor: how do your macros know what the names of those packages' files are?
15:26:00	namnguyen	Araq, what was the use case that called for it?
15:26:18	Araq	"consistency".
15:26:36	Araq	people always report when it's "inconsistent"
15:26:45	endragor	dom96: using lineinfo proc from macros module
15:26:48	Araq	apparently it's better to be consistent than to be useful ...
15:27:18	namnguyen	with cincludes, i guess?
15:27:46	namnguyen	but the doc does say that clib:NAME where NAME is a name of the lib, not the path.
15:28:04	Araq	I have no idea what clib does.
15:28:14	dom96	endragor: then perhaps the best solution would be to ask Araq to include the .cpp/.c filenames in the lineinfo output
15:28:15	Araq	adds something to link?
15:28:19	namnguyen	--clib:LIBNAME link an additional C library (you should omit platform-specific extensions)
15:28:40	namnguyen	yes, it adds some libraries to link command.
15:28:55	namnguyen	the workaround is to use passL:"-lLIBNAME"
15:28:56	dom96	endragor: or maybe via some system proc like getPackageName
15:28:59	endragor	dom96: I think it would not be good for that procedure. It is supposed to return .nim filename
15:29:05	Araq	yeah, I always use the workaround
15:30:18	Araq	dom96, endragor yeah that cannot work, the frontend doesn't even know about how the file will be named
15:32:22	endragor	well, since I already mess with generated files after Nim, I guess I could simply rename those prefixed with the nimble package name.
15:40:14	Araq	true
15:47:01	michael_campbell	Is `echo("foo")` any more or less idiomatic than `echo "foo"`? MEAP uses the parens, tutorial does not.
15:47:04	*	NimIsNotARealCom joined #nim
15:47:37	*	arnetheduck quit (Ping timeout: 252 seconds)
15:48:56	NimIsNotARealCom	Where is Nim's official office?
15:48:58	lazypenguin	michael_campbell, from the docs "The expression x in x.f needs to be semantically checked (that means symbol lookup and type checking) before it can be decided that it needs to be rewritten to f(x). Therefore the dot syntax has some limiations when it is used to invoke templates/macros:"
15:49:17	lazypenguin	not sure how that affects echo "foo" however
15:49:26	lazypenguin	but there might be similar
15:49:53	michael_campbell	"foo".echo also works, which I found fun.
15:50:02	michael_campbell	UFCS... I kind of dig it.
15:50:04	NimIsNotARealCom	?
15:50:09	NimIsNotARealCom	Araq: where is Nim's official office
15:50:42	michael_campbell	<3 that `if` and `case` are expressions.
15:51:08	NimIsNotARealCom	WHERE IS NIM'S OFFICIAL OFFICE?????
15:51:18	Demos	yo: so w.r.t the winapi.
15:51:22	Demos	the winapi itself is in C
15:51:31	Demos	however some components are exposed using COM
15:51:38	Demos	COM is not C or C++
15:51:43	Araq	NimIsNotARealCom: in Mecca.
15:51:44	michael_campbell	Man, that due has a jihad.
15:51:49	michael_campbell	s/due/dude
15:52:04	Demos	and in fact defineing C_INTERFACE before importing a COM header will expose a (slightly nonstandard) C interface
15:52:05	lazypenguin	haha was that the EEE_ guy? Guess you guys changed your mind on ban policy?
15:52:39	dom96	nah, ban policy is still the same: Don't ban until you have no other choice.
15:52:58	michael_campbell	I guess only real compilers have offices. <shrug>
15:53:05	*	Demos quit (Read error: Connection reset by peer)
15:53:11	*	michael_campbell wishes he had an office.
15:53:36	*	J3RN too, wishes for an office
15:53:45	*	pandada8 quit (Ping timeout: 260 seconds)
15:54:25	michael_campbell	My "office" is one of those open office layouts; it's the worst thing ever. It's neither collaborative nor private.
15:55:12	michael_campbell	And it's a cesspool of bacteria; since we moved to it we have a lot of "out sick" that we didn't before.
15:55:20	*	NimIsNotRealComp joined #nim
15:55:30	lazypenguin	hah
15:55:33	lazypenguin	welcome back
15:55:35	NimIsNotRealComp	Where is Nim's official office?
15:55:43	NimIsNotRealComp	Thanks lazy.
15:55:59	NimIsNotRealComp	They cannot beat my VPN
15:56:46	NimIsNotRealComp	WHERE IS NIM'S OFFICIAL OFFICE?
15:56:58	lazypenguin	Why do you ask?
15:57:06	NimIsNotRealComp	I don't know.
15:57:17	namnguyen	<Araq> NimIsNotARealCom: in Mecca.
15:57:23	namnguyen	you happy now?
15:57:36	NimIsNotRealComp	What a beautiful paradism.
15:57:54	NimIsNotRealComp	**Paradise
15:58:34	namnguyen	okay. great. now sit tight.
15:59:09	dom96	*sigh*
15:59:53	lazypenguin	Wish I had that much free time
16:04:57	*	CC-BY-SA joined #nim
16:05:00	CC-BY-SA	ds
16:05:00	CC-BY-SA	c
16:05:01	CC-BY-SA	sdc
16:05:02	CC-BY-SA	fd
16:05:03	CC-BY-SA	btr
16:05:05	CC-BY-SA	bregdb
16:05:07	CC-BY-SA	fd
16:05:07	CC-BY-SA	bfd
16:05:07	CC-BY-SA	csd
16:05:07	CC-BY-SA	ds
16:05:07	CC-BY-SA	c
16:05:08	CC-BY-SA	bfd
16:05:08	CC-BY-SA	vbrd
16:05:08	CC-BY-SA	sdf
16:05:08	CC-BY-SA	v
16:05:09	CC-BY-SA	dr
16:05:09	CC-BY-SA	tr
16:05:10	CC-BY-SA	h
16:05:14	*	CC-BY-SA quit (Excess Flood)
16:06:02	*	lazypenguin quit (Ping timeout: 244 seconds)
16:06:02	*	lazypenguin1 joined #nim
16:06:41	dom96	lol
16:07:18	*	lazypenguin2 joined #nim
16:07:18	*	Mark___ joined #nim
16:07:35	Mark___	You are kiking me just because I'm muslim???!!!
16:07:42	dom96	Mark___: Not at all.
16:07:55	Mark___	So why you are kiking me?
16:08:03	dom96	Mark___: Because you are spamming
16:08:14	Mark___	Okay, am gonna stop
16:08:45	dom96	Mark___: Why are you doing it?
16:09:34	Mark___	What I want from you Mr. dom96 is that when I was in Nim forum (BP), I offensivly replied your comments. I'm sorry, can please forgive me for that?
16:10:21	dom96	Mark___: Of course I can
16:10:45	*	lazypenguin1 quit (Ping timeout: 276 seconds)
16:10:59	Mark___	Thanks, becuase if you can't, then my fasting may not be accepted.
16:11:47	*	lazypenguin2 is now known as lazypenguin
16:12:29	*	PMunch joined #nim
16:15:12	*	Mark___ quit (Ping timeout: 250 seconds)
16:18:01	*	yglukhov quit (Ping timeout: 240 seconds)
16:20:22	*	pandada8 joined #nim
16:25:32	lazypenguin	Hey guys, any reason why dnylib is preferable over header?
16:25:55	*	akaisora joined #nim
16:26:23	namnguyen	lazypenguin, i guess because at runtime, you can use the system-provided dynlibs.
16:26:51	namnguyen	where as with a header, you will have to compile the static library, and build that in with your executable.
16:28:40	*	chemist69 quit (Ping timeout: 264 seconds)
16:28:50	*	akaisora quit (Remote host closed the connection)
16:30:40	*	chemist69 joined #nim
16:35:37	michael_campbell	I'll dig a little deeper, but is there any way to get a statically linked output other than passL and knowing a fair bit about how gcc does this?
16:36:29	michael_campbell	One big advantage of golang here, for running inside docker anyway, is by-default statically linked bins without a lot of extra fuss.
16:37:37	lazypenguin	Yes that's something i'd like to know more about as well. Looking to create a binding for a C library but would rather it statically compile vs shipping with dynamic libs
16:38:05	*	endragor quit (Ping timeout: 260 seconds)
16:39:59	namnguyen	lazypenguin, --app:staticlib
16:41:44	namnguyen	michael_campbell, ^
16:44:02	lazypenguin	Excuse my lack of knowledge here, but is it possible to statically compile a library that is linked to a dynamic libarary? myNimLib -> someDnamicLib.so
16:45:23	namnguyen	lazypenguin, no sir. unles you do it yourself in code.
16:45:26	namnguyen	unless*
16:45:53	namnguyen	which is very possible with macros
16:46:03	namnguyen	tell me more about your need though.
16:46:54	namnguyen	you have code for the static liba.a and the dynamic libb.so, and you want to create libc.a that dynamically links to libb.so?
16:50:22	lazypenguin	Have source for clib, want to create nim bindings nimlib for clib. Looking to ship nimlib as static library. I'm thinking the only way to have users use nimlib statically is to download and compile clib static lib first?
16:50:40	namnguyen	yep.
16:51:06	*	Box_Jellyfish joined #nim
16:51:30	lazypenguin	In your opinion, is that method preferred over porting the lib to nim (since source is available)?
16:53:43	namnguyen	lazypenguin, are you going to maintain the binding for a long run?
16:54:23	*	desophos quit (Read error: Connection reset by peer)
16:55:16	lazypenguin	Is the implication that maintaining bindings to match new versions is more work vs. initial port to nim can be useable indefinitely?
16:55:17	*	Jesin quit (Quit: Leaving)
16:56:09	michael_campbell	namnguyen, that only *makes* a "libfoo.a". I'm looking to make a statically linked binary.
16:56:19	*	mal`` quit (Ping timeout: 244 seconds)
16:56:21	michael_campbell	not a lib
16:56:49	namnguyen	michael_campbell, have you tried it?
16:57:42	michael_campbell	yes, it creates a "libclient.a" from "client.nim".
16:57:57	michael_campbell	what I want is "client", but a staticly linked bin.
16:58:03	*	GangstaCat joined #nim
16:58:08	namnguyen	hmm, on my setup, it creates a static binary.
16:58:19	michael_campbell	I.e., not this -=> "chat: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.24, BuildID[sha1]=9e9e3513023db8c04debc1e22222619bcd37700f, not stripped"
16:58:41	lazypenguin	--passL?
16:59:00	namnguyen	ELF 64-bit LSB executable, x86-64, version 1 (SYSV), statically linked, for GNU/Linux 2.6.24
16:59:02	michael_campbell	lazypenguin, sure, but not sure what options to pass.
16:59:34	namnguyen	michael_campbell, do a ldd on that binary
16:59:36	*	Jesin joined #nim
16:59:40	lazypenguin	Haven't tested but wouldn't --passL:mylib.a work?
16:59:43	namnguyen	so what libraries are dynamic
16:59:59	michael_campbell	linux-vdso.so.1 => (0x00007fffb3ea1000)
17:00:00	michael_campbell	libm.so.6 => /lib/x86_64-linux-gnu/libm.so.6 (0x00007fa1e409c000)
17:00:00	michael_campbell	libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2 (0x00007fa1e3e98000)
17:00:00	michael_campbell	libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007fa1e3ad2000)
17:00:00	michael_campbell	/lib64/ld-linux-x86-64.so.2 (0x00005605a1b03000)
17:00:01	namnguyen	and make sure you have all those static libraries
17:00:11	michael_campbell	Then pass them to --passL?
17:01:19	namnguyen	so you'll need -lm for the math library
17:01:24	namnguyen	dl is automatically added
17:02:01	namnguyen	so, yea, the only thing i know is libm is probably not a static library
17:02:15	namnguyen	the rest looks standard to me.
17:02:35	namnguyen	do you have libm.a in your system?
17:02:44	michael_campbell	looking
17:03:43	lazypenguin	should be /usr/lib/x86_64-linux-gnu/libm.a if you are on debian based
17:04:25	michael_campbell	Yup. /usr/lib/x86_64-linux-gnu/libm.a
17:04:27	namnguyen	http://pastebin.com/k1H1TYiw
17:04:49	namnguyen	that's the command i used to create static binary.
17:05:16	michael_campbell	Gotcha, thanks. What happens if you take out the app:staticlib?
17:05:22	lazypenguin	you get a binary
17:05:27	namnguyen	michael_campbell, it created a dynamic dinary
17:05:30	namnguyen	binary
17:06:05	michael_campbell	"staticlib" seems oddly named to me, then.
17:06:18	michael_campbell	It DOES create a lib for me when used; I wouldn't expect it to create a bin.
17:06:30	lazypenguin	why is it oddly named?
17:06:36	lazypenguin	it produces a static library
17:06:44	michael_campbell	because it creates a lib.
17:06:58	michael_campbell	but for namnguyen it creates (also) a binary.
17:07:03	namnguyen	michael_campbell, is libclient.a really an archive file?
17:07:20	michael_campbell	libclient.a: current ar archive
17:07:21	namnguyen	does it not create anything else?
17:07:52	namnguyen	so it does create a static library for me too. but there's the static binary in addition to that.
17:08:05	michael_campbell	for me it creates JUST the lib
17:08:10	michael_campbell	0.14.2
17:08:24	michael_campbell	the source file is just import os and an echo
17:08:40	michael_campbell	using: nim c --app:staticlib client.nim
17:08:49	namnguyen	erm, hold on
17:08:57	namnguyen	i might have done something differently.
17:09:09	michael_campbell	(and of course nimcache)
17:09:11	namnguyen	i removed the genarated files and ran my command again
17:09:21	namnguyen	no binary is created.
17:10:04	namnguyen	okay. try this: --passL:-static
17:10:10	*	jjido joined #nim
17:10:11	namnguyen	and sorry for the confusion
17:10:55	lazypenguin	I'm still struggling to decide to port library vs create binding
17:11:24	michael_campbell	That seems to work namnguyen , but I get this in the output as well: /home/mcampbell/projects/nimchat/src/nimcache/stdlib_system.o: In function `nimLoadLibrary':
17:11:24	michael_campbell	stdlib_system.c:(.text+0xd7e7): warning: Using 'dlopen' in statically linked applications requires at runtime the shared libraries from the glibc version used for linking
17:11:29	*	akaisora joined #nim
17:11:29	namnguyen	lazypenguin, it all boils down to whether you want to maintain your binding for a long time.
17:11:56	michael_campbell	but I do get a static binary that runs
17:12:01	namnguyen	michael_campbell, it's usually okay.
17:12:06	michael_campbell	Cool, thanks.
17:12:11	namnguyen	yw
17:12:29	lazypenguin	namnguyen, so you are saying that bindings require more maintenance?
17:12:47	namnguyen	lazypenguin, less than porting.
17:13:12	namnguyen	when upstream changes, the port needs to be patched too, but the binding probably only needs a recompilation.
17:17:55	lazypenguin	sure, I'm just trying to figure out how to make it easiest for downstream
17:24:05	*	yglukhov joined #nim
17:24:45	*	Matthias247 joined #nim
17:26:16	*	Box_Jellyfish quit (Ping timeout: 250 seconds)
17:26:18	*	desophos joined #nim
17:28:14	kteza1	This might be the most frequently asked question (sorry. new user here). Is 1.0 planned before the end of this year?
17:28:52	namnguyen	kteza1, as far as i know, 1.0 is planned for "when it's ready".
17:29:24	lazypenguin	alright namnguyen, i'll go down the binding route. Will leave it up to the user to compile static libraries for the lib
17:30:47	namnguyen	lazypenguin, or they can use their own dynamic library too, just like the Nim binding to IUP.
17:31:04	kteza1	namnguyen. No estimates then. Ok
17:36:33	*	jjido quit (Ping timeout: 240 seconds)
17:46:23	*	alberti quit (Read error: Connection reset by peer)
17:54:18	*	mal`` joined #nim
18:00:44	*	akaisora quit (Read error: Connection reset by peer)
18:04:44	*	akaisora_ joined #nim
18:07:23	*	M-Flox is now known as M-Flox1
18:07:42	*	M-Flox1 is now known as Flox42
18:14:28	*	Google joined #nim
18:14:52	*	Google is now known as Guest8241
18:20:29	Guest8241	Araq: I think there is only one task remains for 1.0, am I right?
18:20:53	Araq	what would that be?
18:23:59	*	vendethiel joined #nim
18:24:42	*	akaisora_ quit (Remote host closed the connection)
18:24:46	*	Guest8241 quit (Ping timeout: 250 seconds)
18:25:17	*	ubhu joined #nim
18:25:35	ubhu	Araq: implement static[T] properly.
18:26:01	*	TheLemonMan joined #nim
18:26:09	ubhu	That is what written in roadmap 1.0 (on GitHub)
18:26:33	Araq	I need to update the roadmap :-)
18:27:43	ubhu	Yes, Araq please never hurry, take all your time, stability is very important for certain application programs.
18:32:51	ubhu	Araq: NimEdit's GUI is very not nice.
18:33:13	Araq	afaik I never even announced it anywhere.
18:33:23	Araq	there is a reason for that.
18:33:32	ubhu	GUI is the reason?
18:34:29	Araq	it has bugs and looks ugly
18:34:56	ubhu	It is not ugly, IT IS THE MOST UGLIEST THING
18:36:23	ubhu	For god's sake, why don't you use Qt or GTK+? these both are very good, portable, free and open source and low-level.
18:36:49	*	Salewski joined #nim
18:37:35	dom96	ubhu: stop being rude and making demands.
18:37:38	Araq	I have answered this question already.
18:38:33	Salewski	When the trolls may have a short break a short question: is using paranteses () as below the recommended style?
18:38:37	Salewski	if word.toLower() in pkg.name.toLower():
18:38:58	Araq	I don't use () here
18:39:08	*	endragor joined #nim
18:39:14	ubhu	Araq & dom: You don't know.
18:39:21	Salewski	Thanks!
18:39:35	*	Salewski quit (Client Quit)
18:40:22	*	kteza1 quit (Ping timeout: 250 seconds)
18:41:41	dom96	ubhu: what don't we know?
18:42:53	ubhu	dom96: In your book, you said that JVM is vulnerable.
18:43:10	ubhu	That is TOTALLY WRONG.
18:43:16	GangstaCat	ubhu, we spotted you EEE/Nimian
18:43:24	dom96	ubhu: Why are you wasting our time?
18:43:46	ubhu	dom96: you are teaching people wrong stuffs. you are a student, you cannot write a book and teach people.
18:43:55	ubhu	dom96: you don't have that right.
18:44:04	vktec	Who says?
18:44:13	*	vktec looks around
18:44:41	GangstaCat	no need to use a VPN we got your real IP my friend :p 93.91.194.164
18:44:44	lazypenguin	Sorry dom96, ubhu says you don't have that right so you better leave
18:45:12	vktec	:D
18:45:18	dom96	lol
18:45:22	*	Satyajit joined #nim
18:45:22	*	fredrik92 left #nim ("Leaving channel")
18:45:39	ubhu	vktec: How many times should I tell you GaFY?
18:45:55	lazypenguin	N+1 times please
18:46:41	ubhu	dom96: You are very responsible for sending people wrong information, because one of the Java advantages is secuity.
18:47:02	ubhu	dom96: This is one of the many things that you and araq don't know.
18:47:13	vktec	ubhu: And this is related to Nim.... how?
18:47:15	vktec	XD
18:47:21	lazypenguin	Is there a way to mute him without him knowing?
18:47:40	vktec	lazypenguin: That'd be cool :D
18:47:49	dom96	*cough* :P
18:47:53	vktec	Don't think so though...
18:48:10	*	ubhu quit (Quit: Page closed)
18:48:46	lazypenguin	he figured it out...
18:49:02	lazypenguin	too bad you can't do a mac address ban
18:49:13	vktec	Ooh, IRCv3 should implement that
18:49:33	*	SuperNova joined #nim
18:49:36	SuperNova	tr
18:49:36	SuperNova	vt
18:49:37	SuperNova	d
18:49:38	SuperNova	tehhv
18:49:38	SuperNova	trbeu
18:49:38	SuperNova	6
18:49:38	SuperNova	ytrb
18:49:39	SuperNova	ytr5
18:49:39	SuperNova	y54y
18:49:39	SuperNova	b
18:49:40	SuperNova	6u
18:49:40	SuperNova	yb
18:49:40	SuperNova	n
18:49:40	SuperNova	r5
18:49:44	*	SuperNova quit (Excess Flood)
18:49:47	vktec	Hahahaha!
18:49:53	lazypenguin	lol
18:50:00	vktec	Didn't even have to do anything
18:50:31	GangstaCat	someone refused his PR? is he mad? :p
18:50:53	lazypenguin	He wanted everyone to admit nim compiler is not a real compile
18:51:03	lazypenguin	started lecturing everyone
18:51:08	*	BoxJellyfish joined #nim
18:51:14	BoxJellyfish	Wait!
18:51:34	lazypenguin	:'D
18:52:02	jeffc_	Very 'murrican of you there Dom. Pre-emptive strike and all that
18:52:10	dom96	hah
18:52:24	GangstaCat	is that the first ban in this channel history?
18:52:41	cheatfate	GangstaCat, nope
18:53:06	dom96	nah, but it's the first time we've had a troll that was as persistent as this one
18:53:40	GangstaCat	no big deal, that means Nim becomes to make waves
18:53:51	GangstaCat	begin*
18:53:58	dom96	heh yeah
18:56:09	cheatfate	he using tor i think, and he is not here until he looking for irc client
18:59:36	*	pregressive quit (Read error: Connection reset by peer)
19:00:12	*	pregressive joined #nim
19:01:13	cheatfate	I'm looking for some help, with this https://gist.github.com/cheatfate/1984f4b759fd750d4bffaa6feeb08343
19:01:43	cheatfate	there 2 functions `sendTo`, first one works like expected, but second one always make SIGSEGV on call to WSASendTo...
19:02:11	cheatfate	but this SIGSEGV happens only with gcc, vcc works properly
19:03:10	*	BitPuffin quit (Read error: Connection reset by peer)
19:05:46	Flox42	Can't help directly, but when I had problems with gcc under windows it was due to thread local storage not being nicely implemented by gcc
19:06:43	*	shodan45 joined #nim
19:07:01	cheatfate	Flox42, the problem is, this code don't use threads...
19:13:11	Flox42	valid point
19:15:00	namnguyen	cheatfate, just a wild goose here, is the segfault related to dealloc?
19:15:24	cheatfate	it happens directly on call to WSASendTo
19:15:51	namnguyen	because you have dealloc aiList
19:16:03	namnguyen	and the WSASendTo uses aiList.ai_addr
19:16:11	*	jjido joined #nim
19:16:24	namnguyen	what if you comment out the dealloc lines?
19:18:12	cheatfate	namnguyen, i think i'm wrong and first function is also makes sigsegv
19:19:41	*	pregress_ joined #nim
19:19:42	*	pregressive quit (Read error: Connection reset by peer)
19:20:08	*	akaisora joined #nim
19:20:37	lazypenguin	Hey guys, are const limited to 32 bits?
19:21:30	cheatfate	lazypenguin, on 32bit systems - yes, iirc
19:24:58	shodan45	anyone familiar with both nim and pony and can compare the two?
19:25:32	lazypenguin	thanks cheatfate, have the following in a header file and trying to come up with nim equivalent: #define MPD_RADIX 10000000000000000000ULL
19:25:36	shodan45	I just discovered pony, and it looks interesting.... but I can't get the compiler to build for the life of me
19:26:08	lazypenguin	shodan45, what error are you getting?
19:27:15	shodan45	lazypenguin: with pony? I got several, I tried many different environments - several different distros
19:27:50	lazypenguin	sounds like nim is already better....
19:28:15	dom96	shodan45: you might be able to get some help in #ponylang :)
19:28:43	shodan45	I think the closest I got was a maybe-built ponyc that always failed trying to find "ld" (no idea why, it's in $PATH, etc.)
19:28:49	shodan45	dom96: yeah, already tried
19:29:09	shodan45	trying to figure if I should just give up on it
19:29:55	shodan45	granted, I also had some issues getting recent distros to boot in KVM
19:33:49	shodan45	seems nim's approach of compiling to C at least wins the ease-of-use battle :)
19:34:17	shodan45	(pony uses LLVM & compiles to LLVM IR)
19:35:49	dom96	yeah, i've met some people who use "compilation to C" as a reason not to use Nim and I really don't get them.
19:37:26	shodan45	from what I could tell, pony has less flexible syntax, which frankly I prefer over nim's fairly extreme flexibility
19:38:12	shodan45	nim is "cooler", but I like the practicality of less flexibility
19:38:29	shodan45	(IMHO, of course)
19:39:52	lazypenguin	dom96, while you seem to be around going to hijack this convo for a sec. Any recommendation for handling large numbers in C #defines. Trying to convert this C header to nim but hitting limitation of const at this line ' #define MPD_RADIX 10000000000000000000ULL'
19:41:11	dom96	const x = 10000000000000000000'u64 seems to work
19:41:28	lazypenguin	perfect, thank you
19:59:52	*	abruanese quit (Ping timeout: 272 seconds)
20:07:06	zaquest	hello. is it possible to do something like this http://pastebin.com/XHs48Xrw inherit parametrized type from non-parametrized and have working dynamic dispatching? or maybe there's a better way to achieve this than inheritance?
20:12:48	Araq	don't use uint for counting
20:13:12	PMunch	devlog.peterme.net/index.php?p=23
20:13:23	PMunch	^Wrote a little something on my implementation of TinyWM in Nim
20:14:05	Araq	and usually I replace inheritance and methods with closures
20:14:25	Araq	but it's hard to do that for your example because your example leaves out the algorithm
20:15:23	Araq	an alphabet could become a simple predicate: proc (elemInAlphabet: T): bool
20:16:01	dom96	PMunch: awesome
20:16:05	zaquest	Araq, why not uint? there should not be negative counts.
20:16:33	Araq	use Natural then. nor should the counter wrap around ...
20:16:34	dom96	PMunch: the citation/reference symbols are not rendering properly for me
20:17:01	PMunch	Hmm, those are link symbols (the text before them are links). Are you on mobile?
20:17:20	PMunch	Hade some trouble with mobile not rendering UTF-8 symbols, I'm looking into it
20:17:30	dom96	PMunch: nope, Firefox on OS X
20:17:35	PMunch	Strange
20:18:22	*	pregress_ quit (Remote host closed the connection)
20:18:42	dom96	PMunch: Some grammar feedback: "It's not there are a lack of them", I think you need a "that" in there.
20:19:06	PMunch	Fixed
20:19:10	PMunch	And thanks :)
20:19:12	dom96	also s/you/of/ later in that sentence
20:19:26	Flox42	zaquest: http://forum.nim-lang.org/t/1894#11846 might give you some ideas how to replace inheritance with closures and the like. It starts to get interesting with brianrogoff's post.
20:19:53	PMunch	Crap, had an old window open and now I reverted to a worse version of the text..
20:20:23	dom96	Might want to re-read the text, there are a couple small mistakes :)
20:20:39	zaquest	Flox42, ok, thank you
20:21:46	PMunch	There, now the version is at least back to the better of the two
20:23:34	*	jjido quit (Read error: Connection reset by peer)
20:26:14	PMunch	dom96, better?
20:27:06	dom96	PMunch: a bit, still a couple of mistakes
20:27:54	PMunch	Hmm, any in particular that stand out? I think I've read myself blind on this..
20:28:33	*	gokr joined #nim
20:28:59	*	vqrs quit (Quit: ZNC 1.6.1 - http://znc.in)
20:31:19	*	jjido joined #nim
20:32:20	dom96	PMunch: I'll read it and send you a list :)
20:33:12	PMunch	Thanks, really appreciate it :)
20:33:15	*	shodan45 quit (Quit: Konversation terminated!)
20:34:29	PMunch	Be right back
20:34:53	*	chemist69 quit (Ping timeout: 250 seconds)
20:36:48	lazypenguin	does nim have an equivalent to limits.h?
20:40:54	*	xet7 quit (Quit: Leaving)
20:44:29	elrood	PMunch, jyfi, the u+f2fe codepoint you use for your link symbol is in a private use area of unicode. might work for you, very likely won't for others
20:45:22	PMunch	elrood, yeah I've noticed that. I'm thinking of either specifying the font for those explicitly or using an svg of the icon instead.
20:46:04	*	ldleworker is now known as ldlework
20:47:41	*	vendethiel quit (Quit: q+)
20:47:49	cheatfate	dom96, i think we break asyncdispatch somewhere
20:48:59	cheatfate	and i think its a bug in callSoon implementation
20:49:16	dom96	cheatfate: why do you think that?
20:49:37	cheatfate	dom96, there 2 problems appears
20:50:16	cheatfate	First, do you remember this issue? https://github.com/nim-lang/Nim/issues/3909
20:50:28	*	jjido quit (Read error: No route to host)
20:50:41	cheatfate	I can't make this test anymore
20:50:44	*	vqrs joined #nim
20:50:52	cheatfate	because `await client.recvLineInto(lineFut) or sleepAsync(1000)` is not working anymore
20:51:20	cheatfate	Second, i have implemented async udp support and wrote test for it
20:51:44	cheatfate	this test is similar to test tasyncawait.nim
20:51:48	cheatfate	using localhost
20:51:52	*	chemist69 joined #nim
20:52:11	cheatfate	so in this test my functions heavy used immediate completion (udp is fast)
20:52:24	elrood	PMunch, might make more sense to just resort to a standardized rune. there are quite a few to choose from, you should be able to find something to your liking
20:52:28	cheatfate	but if on unix systems everything is ok
20:52:39	cheatfate	on windows i have very strange SIGSEGVs
20:52:49	cheatfate	first call is OK, second is SIGSEGV
20:53:43	cheatfate	I'm suspect that for some reason stored `cb` loosing its environment parameter
20:53:55	PMunch	elrood, any good listings?
20:55:11	dom96	cheatfate: perhaps it's a Nim closure bug?
20:55:19	*	GangstaCat quit (Quit: Leaving)
20:56:31	cheatfate	and my SIGSEGV happens on return from `WSASendTo()` call, assign return value to variable causes SIGSEGV
20:56:43	cheatfate	so somewhere stack is smashed
20:57:29	elrood	PMunch, your charmap utility of choice ;P
20:58:12	cheatfate	dom96, does anything have been changed recently in closure handling?
20:58:24	cheatfate	i didnt remember
20:59:18	PMunch	elrood, obviously :P I was thinking about a list of some of the more useful ones for this kind of stuff.
21:00:54	elrood	nope, sorry, unicode is a vast field. you could try http://shapecatcher.com/ though
21:03:12	*	cheatfate quit (Ping timeout: 258 seconds)
21:05:09	*	cheatfate joined #nim
21:07:08	cheatfate	Araq, where is callback environment is stored? is it possible so that environment's values can be smashed with some stack usage?
21:08:58	*	lazypenguin quit (Quit: Leaving)
21:11:25	*	Vad joined #nim
21:13:52	cheatfate	Araq, nm
21:15:39	Vad	Hello.
21:16:47	elrood	not again..
21:17:49	dom96	Vad: Hi there, welcome to #nim!
21:17:54	Vad	I have a nice question. Why almost all programming languages are in English?
21:18:05	Vad	dom96: Thanks very much.
21:18:10	dom96	Vad: Are you new here?
21:18:32	Vad	dom96: Yes, I'm new to Nim.
21:20:13	Vad	dom96: I can do tha forever.
21:20:14	dom96	Vad: You might want to try asking that question in #programming. This is a nim-specific channel.
21:20:32	Vad	So don't make me do that again.
21:20:51	Vad	I can always change the IP address
21:20:55	elrood	fasting obviously doesn't benefit that guy
21:21:17	dom96	Vad: Ahh, so you lied to me.
21:21:21	*	akaisora quit (Ping timeout: 246 seconds)
21:22:39	*	Kingsquee joined #nim
21:22:42	*	Vad left #nim (#nim)
21:24:54	*	ldlework is now known as ldleworker
21:25:38	*	jjido joined #nim
21:25:44	dom96	so, anybody who's on IRC usually who would like to help with moderating this channel?
21:26:33	*	kteza1 joined #nim
21:27:33	*	Matthias247 quit (Quit: Matthias247)
21:30:53	elrood	perhaps it's best to just ban /*!*@gateway/web/freenode* for a few days
21:31:18	PMunch	dom96, I'm on here often enough to lend a hand if I see anything obvious..
21:31:43	*	desophos_ joined #nim
21:34:40	*	desophos quit (Ping timeout: 250 seconds)
21:37:11	*	Matthias247 joined #nim
21:38:38	dom96	PMunch: ok
21:38:44	dom96	PMunch: thanks
21:39:08	PMunch	No problem, giving something back to Nim :)
21:39:51	dom96	PMunch: okay, you can gain +o using /msg chanserv op #nim
21:39:56	dom96	or just /cs op #nim in some clients
21:40:11	dom96	and 'deop' to get rid of it
21:40:18	PMunch	/msg chanserv deop #nim
21:40:26	PMunch	There we go
21:40:32	dom96	PMunch: keep this in mind https:/​/​github.​com/​Araq/​Nim/​wiki/​IRC-guidelines
21:40:48	dom96	To quiet use: /mode #nim +q username
21:40:55	dom96	change + to - to remove it
21:41:10	dom96	and prefer to quiet instead of ban :)
21:41:44	PMunch	Okay
21:41:52	PMunch	Keep them quiet until?
21:42:11	dom96	PMunch: however long you think is necessary
21:42:23	dom96	people who are just downright malicious are easy
21:42:57	dom96	when genuine fights occur best to make it temporary
21:43:04	*	jjido quit (Remote host closed the connection)
21:43:11	dom96	between people you know are regulars here for example
21:43:22	dom96	But that won't happen because we all love each other right? :)
21:43:28	PMunch	Yay :)
21:43:48	PMunch	Hmm, that IRC-guidelines link doesn't seem to work by the way
21:43:57	*	TheLemonMan quit (Quit: "It's now safe to turn off your computer.")
21:44:05	PMunch	https://github.com/nim-lang/Nim/wiki/IRC-guidelines
21:44:34	kteza1	Hi. How do I write python bytearray like this in nim --> msg = bytearray(b"\x10\0\0\x04MQTT\x04\x02\0\0")
21:49:09	dom96	kteza1: just var msg = "\0x10\0\0\0x04MQTT\0x04\0x02\0\0" (I think)
21:50:05	dom96	PMunch: oh, and probably best not to quiet the full freenode webchat as there may be genuine people trying to connect.
21:52:16	dom96	afk
21:52:46	PMunch	Of course
21:56:24	*	endragor quit (Ping timeout: 272 seconds)
21:58:06	kteza1	@dom96 Unfortunately no. msg[1] is relulting in 'x' where as in python, it would result in 0 (The byte after 0x10)
21:59:15	PMunch	Hmm, couldn't quite figure out a good way to add the link icons yet (currently I use an ::after tag as my editor simply inputs <a> tags). But I added some useful buttons and updated the text thanks to dom96: devlog.peterme.net/index.php?p=23
22:06:29	Araq	kteza1: var msg = "\x10\0\0\0x04MQTT\x04\x02\0\0"
22:09:15	elrood	PMunch, how do you want it to look? would changing styles.css:46 to something like content:"\2bb0"; do?
22:10:50	elrood	the way you're doing it is fine, just pick a standardized rune
22:14:06	*	zodiak_ quit (Ping timeout: 250 seconds)
22:17:06	*	Trustable quit (Remote host closed the connection)
22:17:19	kteza1	@Araq Will check that. Thanks
22:18:04	*	akaisora joined #nim
22:23:53	*	desophos_ quit (Remote host closed the connection)
22:26:08	*	kteza1 quit (Ping timeout: 250 seconds)
22:27:22	*	desophos joined #nim
22:29:23	*	desophos quit (Remote host closed the connection)
22:30:21	*	desophos joined #nim
22:31:23	*	arnetheduck joined #nim
22:34:34	*	desophos quit (Remote host closed the connection)
22:35:51	PMunch	Hmm elrood, 2bb0 doesn't display for me..
22:36:35	PMunch	And I just noticed that the icons for home and RSS didn't convert the glyph to a path, it simply embedded them. So they don't work either -_-
22:36:47	PMunch	But I've got to go to bed now, so I'll fix it tomorrow.
22:36:55	*	PMunch quit (Quit: leaving)
22:37:23	*	bjz_ quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
22:41:45	*	desophos joined #nim
22:42:42	*	gokr quit (Ping timeout: 276 seconds)
22:43:45	*	desophos quit (Remote host closed the connection)
22:51:20	ldleworker	I had a hard time justifying the idea that Nim is designed as an AST with a batteries included syntax
22:51:33	*	desophos joined #nim
22:51:40	ldleworker	Since other people mentioned like, you could invent any syntax for go as long as you wrote a translator to gofmt'd structure
22:51:53	ldleworker	If anyone wanted to give their thoughts about that I'll read it
22:59:17	Araq	sure. you could also invent any syntax for C as long as you have a translator that generates C ... oh wait.
23:01:46	*	MC__ joined #nim
23:02:26	MC__	Hi guys.
23:05:12	MC__	Araq & dom96: You may ask why I'm annoying you?, well, the true answer is because I'm alone.
23:06:04	vktec	MC__: If that's true, then perhaps be friendly rather than annoying ;)
23:08:39	vktec	I think the answer to this will be no, but here goes: can I run/import Nim (preferrably not NimScript) from within Nim in a way that I can use a DSL written in Nim to configure a Nim program?
23:08:51	vktec	Wow, that was a confusing sentence :D
23:08:58	Xe	yeah lol
23:09:05	Xe	do you mean like outside of the realm of `nake`?
23:09:11	*	Matthias247 quit (Read error: Connection reset by peer)
23:09:20	Xe	because well
23:09:21	*	vktec has never heard of nake
23:09:23	Xe	actually
23:09:27	Xe	i have a terrible idea
23:09:36	Xe	an in fact
23:09:43	vktec	I'm all ears :)
23:09:53	Xe	you could use this to allow people to write the conifugration in any language they want that can export C symbols
23:10:05	*	akaisora quit (Read error: Connection reset by peer)
23:10:10	vktec	Cool
23:10:13	Xe	all you need is a bunch of exposed functions in C space to do all the dirty configuration work
23:10:19	Xe	then make nim niceness around that
23:10:43	vktec	So how would I read the configuration files?
23:10:47	Xe	just make sure to compile your configuration as a shared object and get nim to manually load an execute that function
23:11:09	Xe	note: this is a bad idea
23:11:17	vktec	Is there a better way?
23:11:17	Araq	vktec: use nimscript for this.
23:11:27	vktec	Araq: Okay.
23:11:43	Xe	i like the idea of a .so configuration system
23:11:49	*	akaisora__ joined #nim
23:11:52	vktec	Araq: Can you provide a code snippet to show how I'd do that?
23:11:54	Araq	or: instead of a program, write a library
23:12:05	vktec	I remember struggling with NimScript a while back
23:12:18	vktec	Araq: It's a per-project configuration file
23:12:25	Araq	people can import your library and the config is the main program
23:12:28	vktec	So that's not ideal
23:12:38	Xe	we really need to write ffi support for the nimrodvm
23:12:46	Xe	hmm
23:12:53	Xe	what would that entail exactly?
23:12:56	*	MC__ quit (Ping timeout: 250 seconds)
23:13:56	Araq	I wrote FFI support for the VM twice.
23:14:18	Xe	why not include it then?
23:14:55	Araq	the code still exists in compiler/evalffi.nim
23:15:36	Araq	I don't want to support it.
23:15:59	Xe	how come?
23:16:42	*	saml_ joined #nim
23:17:15	*	zodiak joined #nim
23:18:10	dom96	ooh https://www.youtube.com/watch?v=yXYV_srRDaw
23:18:31	vktec	So.... any code I can look at for loading a NimScript config into my Nim program?
23:18:52	Araq	vktec: Nimble does it
23:19:04	Araq	look at its code for an extensive example
23:19:18	vktec	Yeah, I remember looking at that a while ago, is that the best example?
23:20:10	Araq	compiler/nimeval.nim is a better starting point
23:20:24	Araq	but only shows how to run Nim code as a string
23:20:34	*	desophos quit (Remote host closed the connection)
23:20:47	vktec	Okay
23:21:14	vktec	So I could load it from the file and then run it using the method shown there?
23:21:18	Araq	Xe: people report strange crashes that I then have to look at
23:21:58	Araq	vktec: usually you need the script to call functions from your program
23:22:08	*	desophos joined #nim
23:22:11	Araq	so ... it's never this easy and you need to look at Nimble
23:22:12	vktec	Araq: Yeah
23:22:16	vktec	Okay
23:22:36	vktec	I'll see what I can do, thanks Araq
23:22:39	Araq	but what about my "don't write programs, write libraries instead" idea?
23:22:50	Xe	^ it worked for xmonad
23:23:12	vktec	I'd normally do it that way, but this is a per-project configuration file, so it wouldn't really work very well
23:25:43	*	Jesin quit (Quit: Leaving)
23:27:29	vktec	I assume this is the function I should be looking at? https://github.com/nim-lang/nimble/blob/master/src/nimblepkg/nimscriptsupport.nim#L197
23:29:35	Araq	yes
23:29:45	vktec	Thanks
23:32:13	*	elrood quit (Quit: Leaving)
23:33:12	*	yglukhov quit (Remote host closed the connection)
23:34:14	*	cheatfate quit (Read error: Connection reset by peer)
23:34:46	*	Kaini joined #nim
23:43:43	*	Demos joined #nim
23:44:48	Demos	what editors are you guys using, I'm using the atom plugin but I'm wondering if there are any new/better nimsuggest integrations
23:45:41	*	cheatfate joined #nim
23:46:12	*	vktec uses Vim
23:47:41	Araq	Demos: VSCode + nim plugin seems to be the best currently
23:49:43	Demos	cool. I quite like VSCode
23:49:49	Demos	thanks
23:53:02	Demos	also I may start work on better cuda support for my job. We're getting a DGX-1 and I'd really love to not have to use c++ to really exploit it
23:53:12	*	GangstaCat joined #nim
23:53:39	*	desophos quit (Remote host closed the connection)
23:54:55	*	desophos joined #nim
23:55:45	*	yglukhov joined #nim
23:58:10	Araq	yummy
23:59:56	*	yglukhov quit (Ping timeout: 250 seconds)