<< 13-03-2017 >>

---

00:00:29	*	chemist69 quit (Ping timeout: 258 seconds)
00:00:52	Xe	yes but you should be able to support this usecase even if it is weird
00:01:27	ftsf_	Xe, sure, but someone's gotta do the work to implement it
00:02:40	ldlework	ftsf_: I love nim. I'm lifelong appreciator of the project from a conceptual point of view.
00:02:55	ldlework	I'm merely demonstrating something.
00:02:58	*	chemist69 joined #nim
00:03:10	ldlework	I migh even be wrong, never tried to install mono in a docker container
00:03:12	Xe	if it means modifying the `Addr -> FileDescriptor` function then allow it, etc
00:03:16	ldlework	(that part was super easy for nim)
00:03:39	ftsf_	ldlework, i love nim too, but gotta understand it's young and you'll get frustrated from time to time and probably have to implement features yourself and give back to the project =)
00:04:48	*	bjz joined #nim
00:04:54	*	def-pri-pub quit (Quit: leaving)
00:06:21	ldlework	ftsf_: I'm almost 30, been writing code since I was 8 or so. I'm not without context.
00:06:36	ldlework	I'm merely demonstrating. Its objective data points. It doesn't admonish anyone. It orients.
00:07:03	ftsf_	ack
00:07:32	ldlework	ok c-sharp compiler working in docker
00:07:36	ldlework	now to find docker client
00:07:38	ldlework	well
00:07:43	ldlework	lets try it just using an http client
00:09:43	ldlework	shit I forgot how to write a C# console app
00:09:56	ldlework	good thing the msdn is amazing
00:37:26	ldlework	this would have went wayyyy faster
00:37:36	ldlework	if I didn't systematically typo mcs for msc
00:44:48	*	girvo_ joined #nim
00:53:29	*	girvo_ quit (Ping timeout: 260 seconds)
01:17:28	ldlework	Welllllll
01:17:42	ldlework	apparently the C# world has never heard of a fucking domain socket either
01:17:52	*	ldlework wonders if he's dreaming.
01:29:57	*	bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
01:41:53	*	girvo quit (Ping timeout: 246 seconds)
01:46:19	demi-	ftsf_: tbh, nim has been a great deal less frustrating than many languages i've used; been greatly enjoying that about it :)
01:46:59	Xe	ldlework: this kind of frustration is why i've ended up doing things in go :/
01:49:42	ldlework	Xe: yeah its kinda ridiculous
01:50:01	ldlework	Xe it seems C# has the same problem
01:50:12	ldlework	it has no way to use its httpclient with domain sockets
01:50:40	ldlework	I suppose its a bit more shameful for Nim since it actually is a linux systems programming language
01:50:49	ldlework	And Windows doesn't even have domain sockets afaik
01:50:51	ldlework	lol
01:53:43	*	yeeve quit (Ping timeout: 256 seconds)
01:58:30	*	girvo joined #nim
02:02:26	ldlework	Xe: I take this problem simply does not exist for Golang?
02:02:56	Xe	ldlework: there's several packages for doing it and if you need to you can override that part of the HTTP client with your own Dialer
02:03:14	ldlework	haha god damn it
02:03:21	*	girvo quit (Ping timeout: 256 seconds)
02:03:26	ldlework	Xe: you busy today?
02:03:42	ldlework	wanna sprint this thing in Golang?
02:03:49	ldlework	I'm supposed to be learning Golang for work anyway
02:04:56	*	yeeve joined #nim
02:07:13	*	chemist69 quit (Ping timeout: 260 seconds)
02:13:40	*	vlad1777d quit (Remote host closed the connection)
02:19:23	*	bjz joined #nim
02:22:32	*	girvo joined #nim
02:26:26	ldlework	You know what would be a neat tool
02:26:41	ldlework	take some JSON and generate the nim types for deep unmarshaling
02:26:55	*	girvo quit (Ping timeout: 240 seconds)
02:27:24	Xe	ldlework: marshal in stdlib?
02:27:29	ftsf_	ldlework, at compile time?
02:27:32	*	girvo joined #nim
02:28:10	Xe	ldlework: if this works still it will get you most of the way there: https://github.com/Xe/dev
02:31:43	ldlework	ftsf_: yeah
02:31:52	ldlework	take some known json, generate a schemea for it
02:31:57	*	girvo quit (Ping timeout: 260 seconds)
02:32:04	ldlework	so you can unmarshal arbitrary json trees into real types in nim
02:32:07	Xe	ldlework: kinda like json-to-go?
02:32:10	ldlework	so you don't have to use the json objects
02:32:12	ldlework	is that a thing?
02:32:16	Xe	yeah
02:32:28	ldlework	Oh golang
02:32:31	ldlework	stahp
02:32:37	ldlework	Xe: I'm gonna have to ignore you soon
02:32:38	Xe	go is kinda horrible, but the tools people have made make it the least painful thing to use
02:32:41	ldlework	And I really really like you
02:32:44	ldlework	lol
02:32:59	Xe	ldlework: ?
02:33:31	ldlework	haha I'm just kidding because you keep pointing to golang solutions of my problems and its grating (not really I'm kidding!)
02:34:10	Xe	oh no
02:34:24	Xe	i was more referring to the tools as comparators so i can fit things into my head
02:34:36	Xe	i usually have similar tools in other languages assicated with the one in go
02:35:00	Xe	it's so i can look for "how do i take <input> and get <output> with <constraints>?"
02:35:05	ldlework	Xe: docker outputs all sorts of ridiciulous arbitrary json you know
02:36:16	Xe	ldlework: life throws all kind of ridiculous arbitrary shit at you and the only viable solution is to take things a dat at a time and bit by bit or you'll blow your head off because of stress
02:37:59	ldlework	Xe: no I just mean, it'd be cool to have a json-to-nim so we'd get nice type safe objects when we parse its json
02:38:39	Xe	yeah
02:39:11	ldlework	Xe: I bet there's some crazy Nim thing that would make it like 20 line long lol
02:42:25	ldlework	Xe: I mean, couldn't we just load a jsonfile during compilation and use a macro to generate a type?
02:42:29	ldlework	I've done crazier shit
02:42:41	Xe	well
02:42:43	Xe	that's an idea
02:42:48	ldlework	hehe
02:43:04	Xe	compile time expansion of swagger json into full fledged api clients
02:43:12	ldlework	hah
02:43:25	ldlework	I wonder if the docker api has a swagger definition
02:43:29	ldlework	that'd be nuts !
02:44:26	ldlework	whaaa https://github.com/docker/docker/blob/master/api/swagger.yaml
02:47:07	ldlework	I can't do a swagger thingie right now
02:47:11	ldlework	but neat to know the option is there
02:47:16	ldlework	I bet nim magic might make it easy
02:53:11	*	chemist69 joined #nim
03:00:51	*	girvo joined #nim
03:05:33	*	girvo quit (Ping timeout: 260 seconds)
03:06:33	*	lenstr quit (Ping timeout: 260 seconds)
03:10:40	*	girvo joined #nim
03:16:11	*	Nobabs27 joined #nim
03:26:38	*	cheatfate quit (Read error: Connection reset by peer)
03:26:43	*	cheatfate_ joined #nim
03:27:01	*	cheatfate_ is now known as cheatfate
03:34:26	*	def-pri-pub joined #nim
03:35:47	*	kulelu88 quit (Quit: Leaving)
03:43:58	*	bozaloshtsh quit (Ping timeout: 240 seconds)
03:44:27	*	bozaloshtsh joined #nim
04:15:33	*	bozaloshtsh quit (Ping timeout: 260 seconds)
04:30:24	*	def-pri-pub quit (Quit: Lost terminal)
04:31:13	*	bozaloshtsh joined #nim
04:37:29	*	girvo quit (Ping timeout: 260 seconds)
04:41:25	*	bozaloshtsh quit (Ping timeout: 240 seconds)
04:47:57	ldlework	Does unmarshalling simply not work if there is extra data in the serialized data or something?
04:48:59	ldlework	Anyone?
04:49:04	*	Nobabs27 quit (Quit: Leaving)
04:49:11	ldlework	Its been a whole minute.
04:49:19	ldlework	HELLO!?
04:49:22	ldlework	;)
04:53:00	ftsf_	ldlework, correct
04:53:12	ftsf_	it's annoying, i patched it so it ignored it =)
04:56:56	ldlework	ftsf_: can you share?
04:56:58	*	yeeve quit (Ping timeout: 240 seconds)
04:57:06	ftsf_	not currently
04:57:11	ldlework	...?
04:57:16	ftsf_	i'm at work
04:57:38	ldlework	https://gist.github.com/
04:57:46	ldlework	https://gist.github.com/dustinlacewell/ec6951a21b3cf08da74751d0d8f5ffa9
04:57:51	ldlework	how do I get this out as an int
04:57:56	ldlework	wtf is BiggestInt
04:58:10	ldlework	Nim is freaking crazy town
04:58:27	ldlework	I seriously just wanna work with some json
04:58:53	*	girvo joined #nim
04:58:59	ftsf_	getNum() doesn't work?
04:59:29	ldlework	It returns BiggestInteger
04:59:33	ldlework	whatever the hell that is
04:59:41	ftsf_	a type of int
04:59:46	ldlework	I just want int
04:59:51	ldlework	cuz that's what my type is
04:59:55	ftsf_	getNum().int
05:00:11	ldlework	got (BiggestInt) but expected 'int'
05:00:20	ldlework	the compiler can't figure that out?
05:00:33	ftsf_	seems not
05:00:41	ftsf_	you need to be explicit
05:01:01	ftsf_	potential for data loss i guess?
05:01:03	ldlework	Nim is the only language I use where I have to care about int types
05:01:21	ftsf_	try C =)
05:01:40	ftsf_	C doesn't force you to think about it until stuff breaks
05:01:51	ftsf_	nim forces you to think about it before it breaks
05:02:06	ldlework	I dunno why you think comparing Nim to one of the first languages that has no features is a good idea
05:02:11	*	Snircle quit (Quit: Textual IRC Client: www.textualapp.com)
05:02:19	zachcarter	because Nim compiles to it?
05:02:42	zachcarter	I don’t think having to think about int types at the time of writing code is a band thing
05:02:45	zachcarter	bad thing*
05:03:16	ldlework	Nim compiling to C isn't a justification for anything that Nim does.
05:03:27	zachcarter	what?
05:03:34	ftsf_	nim and C are both statically typed languages that run on baremetal
05:03:35	ldlework	Its clear english.
05:03:42	zachcarter	it’s not logical English
05:03:55	ldlework	You can disagree; it is surely coherent language.
05:04:10	zachcarter	that’s what I”m doing then :P
05:04:21	*	tax quit (Quit: Leaving)
05:04:28	ldlework	Saying, "Nim must work this way because it compiles to C"
05:04:45	ldlework	is instantly refuted by showing "Nim can do this despite C not being able to do this."
05:05:06	ldlework	Having C as a transpiling intermediary has nothing to do with the behavior of Nim as designed.
05:05:07	zachcarter	You can build abstractions
05:05:14	ldlework	Ah glad you agree.
05:05:21	*	yeeve joined #nim
05:05:34	ldlework	Nim is that abstraction, and so I'm just pointing out the reason you gave for why nim works this way is bad.
05:05:51	zachcarter	I think you’re over-generalizing but okay
05:06:03	ldlework	Its actually the most specific point I can make
05:06:34	ldlework	Anyway!
05:06:58	zachcarter	yeah I have better things to fight than you at the moment :P
05:07:06	zachcarter	like opengl code
05:07:16	*	user0_ joined #nim
05:07:42	user0_	I'm very fascinated with Nim.
05:07:51	user0_	How close are we to v1.0?
05:08:00	zachcarter	there’s a roadmap on github user0_
05:08:15	zachcarter	that’s the short answer, I don’t know what the long answer is
05:08:55	ldlework	ftsf_: heh I wish I had that patch :(
05:09:30	ftsf_	ldlework, for clarity, i'm using the marshal module not the json module
05:09:48	ftsf_	i vendored it and then modified it to ignore the errors
05:18:18	ldlework	ftsf_: yeah that's what I want
05:18:21	ldlework	partial unmarshaling
05:28:22	ldlework	I don't understand marshal.nim
05:28:25	*	ldlework sighs
05:28:54	ftsf_	just use it and when you see exceptions raised, put them in a try =)
05:29:35	ldlework	I don't know what you ean just use it
05:29:38	ldlework	I've written a type
05:29:44	user0_	zachcarter: I just read the roadmap. I'm still fascinated. I learned python as my first programming language. But after I started to learn other languages I realized python is pretty slow. *To me*, Nim is the perfect combination of comfort and speed.
05:29:49	ldlework	and I'm using to[TheType](the_data)
05:29:51	ldlework	no errors
05:30:00	ftsf_	ldlework, if no errors then great
05:30:05	ldlework	and yet the return value's fields are all empty
05:30:15	zachcarter	user0_ it’s a great language in my opinion, I’m working on a 2d game framework at the moment using Nim
05:30:27	zachcarter	very few complaints about the language thus far for my use case
05:30:34	ldlework	ftsf_: https://gist.github.com/dustinlacewell/e4b18274d3aa2624a299a01b5c5cb673
05:30:39	ftsf_	ldlework, try load
05:30:43	ldlework	I did
05:31:21	ldlework	this language drives me crazy
05:31:27	user0_	zachcarter: Nice. I want to start my personal project. It's a text editor written in D. Am I too crazy to implement a plugin system (or plugins) written in Nim as opposed to python? I know python has larger libraries but I want performance not latency.
05:31:35	ftsf_	var foo: Foo; load("foo.json", foo) ?
05:31:56	ldlework	ftsf_: there is no file
05:32:02	ldlework	I'm querying a service
05:32:08	ldlework	you want me to write the json to a file first?
05:32:27	ftsf_	the first argument is a stream
05:32:39	zachcarter	I don’t think so user0_ I believe demi- was talking earlier today about writing a plugin system
05:32:44	ldlework	is a string a stream?
05:33:16	ftsf_	no, but i think there's a StringStream type
05:33:26	ftsf_	https://nim-lang.org/docs/streams.html
05:33:46	ftsf_	var ss = newStringStream(mystring)
05:35:00	ldlework	ftsf_: no error but still no data in the actual object
05:35:11	ldlework	and if I print the json, its full of data
05:35:21	ldlework	so I think marshal is silently failing
05:35:27	user0_	zachcarter: Fantastic. I'll stick around.
05:35:42	ftsf_	mysterious
05:36:03	ldlework	ftsf_: do you have docker?
05:36:07	ftsf_	ldlework, nope
05:36:43	ftsf_	user0_, you can compile libraries (.so/.dll) and load them
05:37:38	ftsf_	user0_, https://nim-lang.org/docs/dynlib.html
05:37:53	ldlework	ftsf_: I could embed the json into theh file
05:38:13	ftsf_	https://forum.nim-lang.org/t/1400
05:38:39	ftsf_	which file?
05:38:55	zachcarter	user0_ sweet :)
05:39:07	user0_	Awesome, thanks ftsf_
05:39:25	ldlework	https://gist.github.com/dustinlacewell/8ea556eebe6d3eb689cfa672521094d7
05:39:40	ldlework	ftsf_: this hardcodes the json
05:39:54	ldlework	just run `./dmt ls`
05:39:57	ldlework	after compiling
05:40:54	ldlework	oops I left in an echo statement
05:41:10	*	bozaloshtsh joined #nim
05:41:15	ldlework	updated
05:41:24	ldlework	if you run ./dmt ls you wont get any output
05:41:32	ldlework	even though that causes it to unmarshal that json
05:43:29	ftsf_	Error: unhandled exception: invalid field name: ID [ValueError]
05:44:15	ldlework	for what
05:44:43	ftsf_	when running ./docker ls
05:45:15	ldlework	sure, but what's the error
05:45:18	ldlework	location
05:45:21	ldlework	the nature of it
05:45:29	*	bjz quit (Ping timeout: 240 seconds)
05:45:43	ftsf_	DMed
05:45:58	*	bjz_ joined #nim
05:46:50	ldlework	ugh I badly wanna give up on this
05:46:58	ldlework	how hard could it be to read some json and deserialize it
05:47:00	ldlework	ffs
05:47:43	ftsf_	you need to use object not tuple
05:47:50	ftsf_	tuple loses the field names
05:48:32	ldlework	didn't fix it
05:48:56	ftsf_	gives another similar error for ContainersStopped
05:49:43	ftsf_	typo in your DockerInfo
05:49:45	ftsf_	ConstinersStopped
05:50:09	ldlework	how come I don't get these errors?
05:50:14	ftsf_	i don't know...
05:50:20	ftsf_	what happens when you run it?
05:50:31	ldlework	I get no output
05:50:54	ftsf_	i get it echoing the json input and then the error
05:51:03	ldlework	this version https://gist.github.com/dustinlacewell/7367904089b39215161237cd48ad019b
05:51:09	ldlework	use this
05:51:14	ldlework	it wont try to create a home directory its
05:53:56	ftsf_	Error: unhandled exception: invalid field name: DriverStatus [ValueError]
05:54:10	ldlework	how are you getting these errors?
05:54:16	ldlework	I dont understand this whatsoever
05:54:23	ldlework	Hint: operation successful (44318 lines compiled; 1.050 sec total; 95.059MiB; Debug Build) [SuccessX]
05:54:25	ldlework	~/s/n/dmt $ ./dmt ls
05:54:27	ldlework	nothing
05:54:45	ldlework	how are you compiling it and running it
05:55:02	ftsf_	nim c -r docker ls
05:55:27	ldlework	me too
05:55:30	ldlework	what version of nim?
05:55:39	ftsf_	0.16.0
05:55:49	ldlework	how can this be explained?
05:56:22	ftsf_	what is dmt?
05:57:20	ldlework	its just the name of the app I'm writing
05:57:36	ldlework	you have the full source code
05:57:48	ftsf_	ok
06:00:36	ldlework	dunno, I think I give up on Nim for another couple years
06:01:06	ftsf_	also maybe try back later when the americans and europeans are online
06:01:16	ftsf_	rather than the 2-3 active people here at this time
06:01:52	*	rosshadden quit (Quit: WeeChat 1.7)
06:01:54	ldlework	Dunno, Nim has a strange attitude that, if someone figured it out, and it worked once, then that's literally all we can ask for
06:02:48	*	user0_ quit (Quit: user0_)
06:02:56	*	user0_ joined #nim
06:03:13	ldlework	it doesn't even ship its own package manager
06:04:10	ldlework	I love to hate on golang, I just feel utter sorrow to hate on Nim
06:05:15	ldlework	Oh look, Nim does bundle Nimble now but you have to go to the Nimble repo to find that out
06:05:31	ldlework	its just like they have no awareness of user experience
06:06:18	tstm	ldlework: I'm pretty sure feedback on that is greatly appreciated.
06:06:37	ftsf_	feedback and also work fixing it =)
06:07:00	tstm	May of the users of nim are quite savvy and thus don't necessarily notice the hurdles of starting with nim.
06:08:21	ldlework	I've been writing software for 15 years
06:09:15	ldlework	If they spent more than 0% effort/time, as in, it was literally part of the required process of managing the project, things would be a lot nicer.
06:10:01	ldlework	If I can't download your language, and download a json file from a URL and parse it into a type-safe structure in <20 minutes, your shit is fucked.
06:10:30	ldlework	There may be many reasons why one's shit is fucked, but none of thoses reasons magically un fuck the shit, so to speak.
06:11:45	tstm	For sure.
06:11:52	ftsf_	there's a reason it's alpha
06:13:05	ftsf_	there's a lot of room for improvement, but if you're keen then i'm sure your help would be appreciated. if not, come back in a while once it's stable
06:13:54	ftsf_	i've been using it happily for my own projects for some time with the occasional hurdle, which i can usually work around or get help with here.
06:14:36	ldlework	I've written a lot of code in Nim myself
06:14:57	ldlework	But its always through blood, sweat, and literally blind-folded traversal of rooms I've never seen before.
06:15:05	ldlework	The compiler messages are utterly mysterious
06:15:22	ldlework	I know that when the westerners wake up tomorrow they will explain why things are the way they are
06:15:40	ldlework	Or even some small simple thing I missed, but I dunno how long it'll take before they realize that isn't the point
06:17:31	ldlework	I'll just use JsonNode to get on with life
06:17:42	ldlework	(literally interface{} all over again!)
06:19:39	tstm	ldlework: Is your input json serialized with marshal or just some json you've written?
06:20:58	ldlework	tstm: Its json from a server
06:21:01	ldlework	service
06:21:15	ldlework	I expect it would say "Woah, your nim type doesn't define X field."
06:21:19	tstm	(dislaimer: I know next to nothing about the subject)
06:21:21	ldlework	or something, its just does nothihng
06:21:47	ftsf_	ldlework, which is what I got when I ran your code
06:21:52	ftsf_	not sure why it's not doing it for you
06:21:56	tstm	Yeah. I'd expect marshal to be pretty dependant on its own structure, whatever it uses for serialization.
06:21:59	ldlework	ftsf_: yes I am boggled.
06:24:31	FromGitter	<Varriount> ldlework: Pull request?
06:25:21	ldlework	Varriount for what?
06:29:35	ldlework	ftsf_: on interesting thing is that if I put some echo's after that deserialization line they don't get output
06:29:43	ldlework	I wonder if the error is taking place and I'm just not seeing it
06:29:59	ftsf_	running on linux?
06:34:13	ldlework	yeah
06:39:49	ldlework	yyyyyeap fuck this
06:46:56	*	nsf joined #nim
07:15:10	zachcarter	Cool truetype font loading via freetype and rendering is a thing now - http://imgur.com/a/6YZ28
07:24:38	*	user0_ quit (Read error: Connection reset by peer)
07:24:47	*	user0_ joined #nim
07:28:11	*	ftsf_ quit (Quit: :q!)
07:31:45	*	Vladar joined #nim
07:34:15	*	couven92 joined #nim
07:34:24	Araq	ldlework: what's the problem? missing json.toNim ?
07:34:51	*	girvo quit (Ping timeout: 256 seconds)
07:35:36	tstm	Araq: I think the problem was that marshal didn't give any sensible error messages for whatever reason.
07:35:45	tstm	it did for the other guy
07:36:20	tstm	But, marshal was used because of lack of json.toNim
07:38:25	Araq	yeah, maybe I can implement that later
07:40:54	stisa	There's https://github.com/cjxgm/jsonob that should do json->nim
07:44:01	*	tankfeeder joined #nim
07:47:10	Araq	"Dunno, Nim has a strange attitude that, if someone figured it out, and it worked once, then that's literally all we can ask for". I have no idea what that's supposed to mean.
07:51:17	*	chemist69 quit (Ping timeout: 256 seconds)
07:53:06	*	gokr joined #nim
07:56:28	*	chemist69 joined #nim
07:56:48	ldlework	Araq: There's no knowledge transfer
07:56:59	ldlework	The docs just report what is
07:57:25	ldlework	I gave up, and having lots of success in Python
07:58:15	Araq	sure but that doesn't mean I don't want to improve the experience.
07:59:39	Araq	so there is no knowledge transfer and the docs just report what is. well what should they report instead?
08:01:23	ldlework	Araq: I'm just saying, has no one ever tried to download nim, download a some json from a web service and deserialize into types?
08:01:55	ldlework	Either I'm stupid, my machine is strangely configured, people have tried and failed and were unheard, or literally no one has tried this before.
08:02:17	ldlework	I dunno which it is, but everytime I try Nim its a test of my loyalty to the concept of the project.
08:03:21	Araq	I think it's "people noticed it's missing and wrote their own json->object proc"
08:03:41	ldlework	I just hope one day I will try Nim randomly, and miraculously get a bunch of work done.
08:04:12	Araq	well actually I *know* people created better de/serialization and do not share it ;-)
08:09:20	*	Arrrr joined #nim
08:09:20	*	Arrrr quit (Changing host)
08:09:20	*	Arrrr joined #nim
08:09:55	*	xet7 quit (Quit: Leaving)
08:09:59	Araq	so ... where should we document that Nim ships with nimble? tutorials?
08:10:24	Araq	search index?
08:11:13	*	xet7 joined #nim
08:12:57	Araq	hmm github lets me resolve merge conflicts on their site now. impressive job.
08:22:35	*	zachcarter quit (Quit: zachcarter)
08:24:29	*	bozaloshtsh quit (Ping timeout: 240 seconds)
08:28:00	*	Andris_zbx joined #nim
08:33:48	ftsf	Araq, did it work out for you? i tried that in the past and it didn't work... ended up having to resolve it locally anyway
08:34:16	Araq	seems so, yes, the CI is still running though.
08:37:58	cheatfate	Araq, looks like zahary wants to break my `asyncpg` again :)
08:38:33	cheatfate	so it looks like macros is the most unstable part of Nim
08:38:40	Araq	hey #nim, stop talking in riddles please.
08:39:01	Araq	cheatfate: why? do you check for nkBracket in your macros?
08:39:08	cheatfate	yep
08:39:57	Araq	I can patch the code for you. ``n.kind in {nkBracket, nkArgList}``
08:40:25	Araq	<-- compatible with upcoming Nim version and older Nim versions
08:40:37	cheatfate	ok, then
08:41:49	*	rokups joined #nim
08:43:03	cheatfate	Araq, is it possible to have this workaround in macros docs?
08:46:31	Araq	you keep asking me for making the stdlib explain the basics of the programming language to you, but fine.
08:47:49	Araq	^ this 'you' here refers to #nim
08:48:01	Araq	in general.
08:48:29	ftsf	more examples in stdlib docs would probably do wonders
08:49:02	ftsf	bonus points if they're also tests =)
08:49:20	FromGitter	<dom96> It really pains me to see ldlework trying to read JSON using the marshal module.
08:49:28	FromGitter	<dom96> That's not what the marshal module is for!
08:49:35	ftsf	dom96, sorry, that was my suggestion as it worked for me
08:50:02	FromGitter	<dom96> I see.
08:50:15	FromGitter	<dom96> We need a large warning in marshal against this I guess.
08:50:55	FromGitter	<dom96> And a way to actually read/write JSON this way
08:51:00	*	xet7 quit (Quit: Leaving)
08:57:04	Araq	yup.
08:59:15	FromGitter	<dom96> Examples would help indeed. The JSON examples are a mess IMO https://nim-lang.org/docs/json.html
08:59:38	FromGitter	<dom96> Compare to httpclient docs https://nim-lang.org/docs/httpclient.html
09:00:20	FromGitter	<dom96> It's not just about examples, it's also about how you present them.
09:02:00	FromGitter	<dom96> @ldlework this is not how you should be deserialising JSON in Nim.
09:02:58	FromGitter	<dom96> How would you do it in Python?
09:03:21	*	bjz_ quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
09:04:05	FromGitter	<dom96> In fact, you shouldn't be deseralising at all.
09:06:37	*	Salewski joined #nim
09:07:15	Salewski	Araq, I am a bit confused about latest nimsuggest error text format...
09:07:22	Salewski	Strings I get from reading socket look like this when echoed:
09:07:31	Salewski	"\'s\' is declared but not used [XDeclaredButNotUsed]"
09:07:35	Salewski	literal(0))\x0Abut expected
09:07:45	Salewski	So I do now 3 replace operations before displaying erros in tooltip:
09:07:53	Salewski	if doc[0] == '"' : doc = doc[1 .. ^2]
09:08:01	Salewski	doc = doc.replace("\\'", "'")
09:08:11	Salewski	doc = doc.replace("\\x0A", "\n")
09:08:16	ldlework	dom96 what kind of advice is that
09:08:17	Salewski	But ok, when that is intended I am fine.
09:08:45	ldlework	I'm not "trying to read json using the marshal module"
09:09:00	ldlework	I'm trying to get type-safe json, for which has a rigid and stable schema.
09:09:07	ldlework	Your pitty is kind of patronizing.
09:09:51	ldlework	I wouldn't do this in Python because python has no type safety. If I really wanted it, I'd invent some sort of JSON schema non-sense.
09:10:00	Araq	Salewski: it's always in quotes, I think?
09:10:23	ldlework	But I figure, I can write some simply Nim types, and deserialize Docker's standard REST responses to actual types so I can use dot-notations and so on
09:10:25	Salewski	Yes, always i ""
09:10:31	ldlework	And to implement methods against them.
09:10:38	ldlework	dom96: still feel the same?
09:10:43	Araq	ldlework: "dom96, sorry, that was my suggestion as it worked for me"
09:10:57	Araq	I think he also referred to this statement
09:11:28	stisa	Idlework: https://gist.github.com/stisa/fa48fb24ba0929b42c7b39b2a0505029 do you mean something like this for json->object ?
09:12:18	ldlework	stisa: yeah, its called the Marshall Module.
09:12:55	ldlework	I assumed that since I did not totally define every single field, and every nested possible json object with a corresponding Nim type I would get errors
09:13:09	ldlework	But I just got silence and we couldn't figure out why I wasn't getting errors so I moved on
09:13:12	ldlework	I'll visit it again
09:13:27	ldlework	But "oh my god that's not what marshalling is for!" isn't the right thing.
09:14:02	*	bjz joined #nim
09:14:26	stisa	oh sorry, I must have missed part of the conversation
09:14:38	ldlework	If I figure it out, I'll make sure to come back and report what I come up with
09:15:25	*	user0_ quit (Quit: user0_)
09:19:55	FromGitter	<dom96> Sorry, I didn't mean to be patronizing. I just got a bit annoyed.
09:20:54	FromGitter	<dom96> The marshal module is essentially a pickle module.
09:21:04	FromGitter	<dom96> Perhaps we should rename it?
09:21:37	FromGitter	<dom96> That said, the name "marshal" doesn't imply that it handles JSON.
09:23:33	FromGitter	<dom96> But sure, we need something like that for JSON in the stdlib.
09:25:55	*	Salewski left #nim (#nim)
09:32:51	Arrrr	If the problem is that in the future it may change its internal representation, you might allow several backends.
09:58:15	*	bozaloshtsh joined #nim
10:00:00	*	vlad1777d joined #nim
10:00:38	*	chemist69 quit (Ping timeout: 246 seconds)
10:03:56	*	chemist69 joined #nim
10:37:41	*	arnetheduck joined #nim
11:02:16	*	Snircle joined #nim
11:09:28	*	couven92 quit (Quit: Client disconnecting)
11:16:00	*	couven92 joined #nim
11:16:53	*	fredrik92 joined #nim
11:20:49	*	couven92 quit (Ping timeout: 240 seconds)
11:42:33	*	fredrik92 is now known as couven92
11:44:16	*	couven92 quit (Quit: Disconnecting)
11:44:42	*	couven92 joined #nim
12:00:45	*	zachcarter joined #nim
12:04:58	*	zachcarter quit (Ping timeout: 240 seconds)
12:11:48	*	couven92 quit (Quit: Disconnecting)
12:19:17	*	zachcarter joined #nim
12:40:29	*	Arrrr quit (Ping timeout: 260 seconds)
12:44:10	*	def-pri-pub joined #nim
12:50:18	def-pri-pub	Hey, morning guys. I got around to doing another Nim article. This one is a tutorial instead: https://16bpp.net/page/hot-loading-code-in-nim
12:57:25	*	elrood joined #nim
12:57:57	Araq	def-pri-pub: I like it but the compiler can easily make this easier
12:58:07	Araq	want to patch the compiler?
12:58:34	Araq	I can guide you but I don't want to do it myself
13:00:38	FromGitter	<Varriount> dom96: Did you just change the documentation on the marshal module?
13:01:51	FromGitter	<Varriount> Or has the warning "The serialization format could change in future!" always been there?
13:05:53	FromGitter	<Varriount> ldlework: If you don't mind using yaml (which is a superset of json) NimYaml does what you want.
13:05:57	FromGitter	<Varriount> https://nimyaml.org
13:09:06	def-pri-pub	Araq: I bet it could. I'm far too busy right now to hack up the compiler to do that
13:10:12	def-pri-pub	I'd rather make a system that operates a little better first,
13:10:32	def-pri-pub	Or some sort of "easy hot loading," package for nimble
13:11:56	flyx	ldlework, I can confirm the NimYAML will do exactly what you expect for JSON deserializing, including emitting an error for missing / superfluous fields
13:13:00	flyx	if it hadn't the YAML data stream as intermediate representation, it would be a general serialization library which happens to have a YAML backend ^^
13:16:29	FromGitter	<Varriount> flyx: I've probably said this before, but NimYAML is an awesome module.
13:19:06	flyx	heh, thanks.
13:22:37	*	zachcarter quit (Quit: zachcarter)
13:46:18	def-pri-pub	Araq: actually, how could the compiler make this easier? I mentioned that I would like a system that could detect changes in the code, instead of file write times. Can you elaborate on this a little?
13:46:51	Araq	def-pri-pub: system.reloadDlL(name)
13:47:02	Araq	requires codegen support though.
13:47:22	Araq	but it should be a simple patch, we already use dlsym under the hood for .dynlib
13:47:57	Araq	the loading code needs to be moved into a generated C function that system.reloadDll calls again
13:48:24	def-pri-pub	Where is reloadDll located in the Nim source? A gitHub code search didn't reveal any results.
13:50:51	Araq	well it doesn't exist.
13:50:56	Araq	I'm asking you to implement it.
13:51:00	def-pri-pub	ah
13:53:02	def-pri-pub	Would this proc then also reset all of the function pointer symbols it grabbed from the first load? That's one of the issues that I see.
13:53:27	rokups	Araq: did you see https://github.com/nim-lang/Nim/issues/5507 ?
13:54:49	Araq	def-pri-pub: yes, it would. that's its whole point
13:57:13	def-pri-pub	Okay, I'll think about it, but don't except too much from me (busy workin'). Do you think this would be done in pure Nim or C?
13:58:59	Araq	don't be silly. we don't write C here.
13:59:17	Araq	it would be a compiler patch, the compiler is written in Nim.
14:00:02	*	IrcDroidClient joined #nim
14:00:50	*	IrcDroidClient quit (Remote host closed the connection)
14:06:17	def-pri-pub	Araq: can I do it in Haskell instead :)
14:23:21	Vladar	Any Linux users wanna test out my attempt at packing AppImage game? https://github.com/Vladar4/ng2planetoids/blob/master/ng2planetoids-i386.run
14:23:44	Vladar	waisted 3 days trying to figure it out
14:24:44	*	zachcarter joined #nim
14:25:05	*	elrood quit (Quit: Leaving)
14:39:00	def-pri-pub	Vladar:
14:39:06	def-pri-pub	I dl'd it
14:39:13	def-pri-pub	chmod u+x'd it
14:39:18	def-pri-pub	and tried to run it.
14:39:20	def-pri-pub	Got this:
14:39:24	def-pri-pub	bash: ./ng2planetoids-i386.run: No such file or directory
14:39:53	def-pri-pub	¯\_(ツ)_/¯
14:41:53	Vladar	that's… strange. what ls tells?
14:42:29	Vladar	even ls -la
14:47:41	Vladar	def-pri-pub: ?
14:48:28	def-pri-pub	It says that it's marked as executable
14:49:02	Vladar	could you give me the output of ls -la?
14:49:59	def-pri-pub	-rwxrw-r-- 1 ben ben 1541392 Mar 13 10:38 ng2planetoids-i386.run
14:51:17	Vladar	hm, let me check something…
14:53:03	Vladar	ok, I think I know why
14:53:28	flyx	well it is a 32bit executable, it naturally does not work in 64bit environments
14:55:05	Vladar	actually, it should
14:55:23	Vladar	here's the correct link https://github.com/Vladar4/ng2planetoids/blob/master/ng2planetoids-i386.run?raw=true
14:56:04	Vladar	def-pri-pub: what about now?
14:57:31	ftsf	dlopen(): error loading libfuse.so.2
14:57:36	ftsf	AppImages require FUSE to run.
14:57:49	Vladar	yes, you should have fuse installed
14:58:11	*	nsf quit (Quit: WeeChat 1.7)
14:58:27	Vladar	32bit one, probably
14:58:31	ftsf	i do, but no lib32-fuse available on arch it seems
14:58:41	Vladar	it is on aur
14:59:12	ftsf	"Linux apps that run anywhere" hehe
14:59:23	ftsf	sadly still not that easy
14:59:33	Vladar	better than nothing, I guess )
14:59:47	federico3	Vladar: have you looked into flatpak?
14:59:49	ftsf	Error: unhandled exception: No such file or directory [OSError]
14:59:50	ftsf	runs now
14:59:55	ftsf	but gives that, echos -1
15:00:24	ftsf	best i've found is a shell script that sets the library path and executes the correct binary for the platform
15:00:43	ftsf	and just distribute a tar.gz
15:00:48	Vladar	ftsf: flatpak requires flatpak installed, IIRC
15:00:52	demi-	is there any IPC mechanism in nim's standard library yet?
15:01:14	demi-	or would i have to fall back on whatever is available on my specific platform
15:01:54	federico3	Vladar: it does - however it looks like distributions are shipping it
15:03:30	Vladar	ftsf, what are you running, btw? arch?
15:04:08	ftsf	Vladar, yep
15:04:28	Vladar	it's strange, I'm running it too
15:04:49	Vladar	though no errors happening
15:06:53	Vladar	actually, I have looked into flatpak but didn't like it much
15:07:23	Vladar	AppImage seems like a more simple approach
15:07:59	ftsf	tar is nicer and simpler =)
15:09:07	Vladar	def-pri-pub: have you tried new link?
15:11:49	ftsf	o/ night!
15:13:21	*	tankfeeder quit (Quit: Leaving)
15:16:46	def-pri-pub	Vladar: can you resend it?
15:16:55	Vladar	https://github.com/Vladar4/ng2planetoids/blob/master/ng2planetoids-i386.run?raw=true
15:17:50	*	rosshadden joined #nim
15:18:32	def-pri-pub	Same thing happened
15:18:58	Vladar	weird…
15:20:58	Vladar	def-pri-pub: let's try this command then:
15:20:59	Vladar	wget -O ng2planetoids-i386.run https://github.com/Vladar4/ng2planetoids/blob/master/ng2planetoids-i386.run?raw=true && chmod +x ng2planetoids-i386.run && ./ng2planetoids-i386.run
15:21:38	Vladar	should download and run it
15:22:25	def-pri-pub	Yeah, that's not working either
15:22:34	Vladar	what distro are you running?
15:24:21	def-pri-pub	Ubuntu 16.04.1
15:25:25	Vladar	32 or 64?
15:25:51	Vladar	and it still says No such file or directory ?
15:26:34	def-pri-pub	Im' running 64 bit
15:27:32	*	sz0 joined #nim
15:27:42	Vladar	gonna try to run livecd myself to find out what the hell is going on
15:31:00	rosshadden	Vladar: I'll try running it on Arch. Do you have a 64bit file, or do you expect that to work anyway?
15:31:39	Vladar	rosshadden: it should work if you have lib32-fuse installed
15:31:46	rosshadden	Okay I'll install it.
15:31:57	*	rauss quit (Quit: WeeChat 1.7)
15:32:02	*	rosshadden is now known as rauss
15:32:14	Vladar	I'll try to build 64 bit version later, but I need to setup 64-bit Ubuntu somewhere
15:32:20	*	rosshadden joined #nim
15:33:10	Vladar	and I have exactly one 64-bit machine here — my working one >_<
15:33:29	rauss	Ehh I don't want to replace gcc with gcc-multilib
15:33:40	rauss	I'll wait for the 64 bit version later
15:33:46	Vladar	sure
15:36:41	Vladar	I'm wondering, could I just pack 32-bit app with a 64-bit AppImage?
15:36:57	Vladar	worth to try…
15:39:28	*	bozaloshtsh quit (Changing host)
15:39:28	*	bozaloshtsh joined #nim
15:41:54	*	niv quit (Remote host closed the connection)
15:42:07	Vladar	rauss: could you try this version? http://dropcanvas.com/68a4o
15:42:50	Vladar	packed it with 64bit appimage, so, theoretically, it should work with 64bit fuse
15:47:31	rauss	Sure
15:48:25	rauss	Vladar: I get the "No such file or directory" error, which is a good thing I think for you
15:48:29	rauss	more people to recreate it :)
15:48:37	rauss	Plus I'm on Arch, so it rules out Ubuntu as the issue.
15:50:03	Vladar	I'm starting to lose my mind. Post "ls -la" output, pls
15:51:27	Vladar	It might be related to 32/64bit, I think. Need to build a true 64-bit app to prove it, though
15:52:38	FromGitter	<rosshadden> @Vladar ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=58c6c0447ceae5376ab11c5c]
15:53:07	Vladar	thanks
15:53:38	Vladar	I'll try to set up 64bit ubuntu 12 platform and then contact you
15:54:21	rauss	Vladar: You said you use arch. Don't you use a 64bit arch?
15:54:35	rauss	Vladar: They officially deprecated 32bit arch a few weeks ago, even
15:55:09	rauss	Maybe it wasn't you that said you used arch. I skimmed the chat log but might have misread.
15:55:34	Vladar	sure, and I'm sure if I build it on arch it will run for you perfectly. The thing is, it should be build on oldest possible system to run everywhere
15:56:35	*	Snircle quit (Ping timeout: 246 seconds)
15:57:44	Vladar	however, there's no harm to try it out too, give me a couple of minutes…
16:02:50	*	Snircle joined #nim
16:03:42	Vladar	rauss: here's arch-built one http://dropcanvas.com/p1c8v
16:07:34	*	Nobabs27 joined #nim
16:11:29	*	chemist69 quit (Ping timeout: 240 seconds)
16:11:52	*	chemist69 joined #nim
16:26:51	*	Arrrr joined #nim
16:26:57	*	Nobabs27 quit (Quit: Leaving)
16:33:13	zachcarter	TrueType font rendering via FreeType :D
16:33:15	*	Trustable joined #nim
16:33:17	zachcarter	http://imgur.com/a/gnmYu
16:43:57	*	xet7 joined #nim
16:46:24	*	gokr quit (Ping timeout: 268 seconds)
16:49:27	*	Andris_zbx quit (Remote host closed the connection)
16:56:51	dom96	Hello guys
16:58:00	dom96	demi-: I don't think IPC is a must-have for a standard library.
16:59:02	dom96	I created this a while back https://github.com/dom96/nim-jsonrpc
16:59:43	*	ofelas quit (Quit: Leaving)
17:04:25	*	gokr joined #nim
17:04:26	*	vlad1777d quit (Remote host closed the connection)
17:05:05	demi-	interesting, though doesn't provide the same sort of transport security that an authorized IPC/XPC process would. I might just wrap the XPC headers on OS X and stick them up as a package
17:09:38	*	zaquest quit (Quit: Leaving)
17:19:16	*	zaquest joined #nim
17:27:14	rauss	Vladar: I still get dir not found
17:27:38	rauss	the first thing that prints is `-1` btw.
17:28:11	rauss	I agree with dom96 that IPC doesn't really need to be in a std lib
17:28:35	*	devted joined #nim
17:40:47	def-pri-pub	dom96: while you're here, I've been meaning to ask you about how I can go about fixing up the .nimble file for my stb_image wrapper
17:41:26	def-pri-pub	When I do an install, I get that warning about namespaces telling me to put my files in a subdirectory that has the same package name
17:42:08	def-pri-pub	The only issue is that I have to separate files that are meant to be imported, each with different names
17:42:20	def-pri-pub	`stb_image.nim` and `stb_image_write.nim`
17:42:54	def-pri-pub	Is there a way I can stuff that _write.nim one in the subdir, but still have it where my users can type `import stb_image_write as stbiw`?
17:54:15	*	Ven joined #nim
17:56:00	*	gk-- joined #nim
17:56:47	*	gk-- quit (K-Lined)
17:58:16	*	gk-- joined #nim
17:58:47	*	gk-- quit (K-Lined)
18:00:28	*	Ven quit (Ping timeout: 240 seconds)
18:01:00	*	vlad1777d joined #nim
18:01:46	*	Jesin quit (Quit: Leaving)
18:02:08	stisa	def-pri-pub : make a folder named src and put .nim files there
18:02:24	*	Jesin joined #nim
18:02:30	stisa	Then in .nimble add srcDir = "src"
18:03:19	*	elrood joined #nim
18:03:25	*	niv joined #nim
18:04:11	*	gokr quit (Quit: Leaving.)
18:04:14	stisa	Oh no wait I misread,sorry, that would still give warnings.
18:04:29	*	zachcarter quit (Quit: zachcarter)
18:10:16	dom96	def-pri-pub: make a directory called stb_image/
18:10:25	dom96	and create a 'write' module in there
18:10:43	dom96	(better to rename it to 'writer' probably)
18:11:25	def-pri-pub	I'm trying to keep it as similar to that of the original stb libs, e.g. `stb_image_write.h`. Guess I'll have to break that...
18:11:39	dom96	def-pri-pub: btw ideally Nimble's hints should explain this. Do you think there is a way to make them easier to understand?
18:12:24	dom96	Afraid so. The only other way would be to create a 'stb_image_write', 'stb_image_read', etc. packages
18:12:38	def-pri-pub	Nah, I think the hints are good, I just was wondering if there was a way to get the `import`s to work the way that I want with the newer nimble
18:12:56	def-pri-pub	two separate packags? Absolutely disgusting.
18:13:43	*	Ven joined #nim
18:15:23	*	chemist69 quit (Ping timeout: 264 seconds)
18:17:54	FromGitter	<Varriount> Rauss: Wow, deprecation an entire architecture? The Arch team makes done pretty bold moves
18:19:48	*	chemist69 joined #nim
18:20:16	rauss	Varriount: The rest of the OS's could learn something from that
18:20:16	*	LeNsTR joined #nim
18:21:19	dom96	While i'm sure it takes a load of work off their hands, I did enjoy running Arch on my old Pentium 3 IBM PC.
18:22:11	FromGitter	<Varriount> I'm still miffed that the Arch team stopped including an installer with their live CD iso
18:24:04	*	Vladar quit (Quit: Leaving)
18:26:24	ehmry[m]	is it possible to catch C++ exceptions from Nim? I've been putting exception handling in external code but then I have to deal with compiling an external object.
18:27:49	FromGitter	<Varriount> ehmry: I don't know, but if you can't do it in pure Nim, you can probably manage something with macros and the emit pragma
18:28:21	ehmry[m]	yea, I was consideing emit but it seemed distasteful
18:30:30	FromGitter	<Varriount> @Araq I'm curious - why is the `this` pragma module-wide?
18:32:17	*	Ven quit (Ping timeout: 260 seconds)
18:33:26	*	vlad1777d quit (Remote host closed the connection)
18:33:49	*	nsf joined #nim
18:34:32	*	Ven joined #nim
18:36:28	*	def-pri-pub quit (Quit: leaving)
18:46:17	*	Ven quit (Ping timeout: 260 seconds)
18:50:22	*	Trustable quit (Remote host closed the connection)
18:53:29	*	Ven joined #nim
19:02:24	*	Ven quit (Ping timeout: 240 seconds)
19:07:31	*	gokr joined #nim
19:13:25	*	Ven joined #nim
19:14:31	*	Ven quit (Read error: Connection reset by peer)
19:17:35	*	brson joined #nim
19:22:52	*	PMunch joined #nim
19:25:06	*	Matthias247 joined #nim
19:33:24	*	Ven joined #nim
19:34:49	Araq	varriount: what else should we attach it too?
19:37:46	Araq	dom96: can we please make nimble less rigid about the directory structure somehow?
19:38:09	Araq	not even the packages under nim-lang adhere to it
19:38:23	Araq	nor do I like excessive subdirectories, but you know that.
19:38:40	Araq	we can change how --nimblePath works if that's what it takes.
19:39:16	dom96	Propose something
19:39:48	dom96	It's a small price to pay for the feature that we get
19:40:07	Araq	compiler sees package/module.nim and considers package-$version/module.nim if that's in the nimble path
19:41:04	dom96	Do you understand why Nimble needs to be so anal about the directory structure?
19:46:22	*	Ven quit (Ping timeout: 260 seconds)
19:46:56	PMunch	I'm curious dom96, why?
19:47:03	Araq	because we like to enforce package/module in the import statements?
19:47:50	*	user0_ joined #nim
19:48:11	dom96	Because a package named 'http' shouldn't encroach on another package's namespace.
19:48:48	dom96	A package named 'http' shouldn't be able to install a module that can be imported via `import ftp/client`
19:49:08	dom96	Because this module belongs to an 'ftp' package.
19:50:15	dom96	Does that make sense?
19:51:04	PMunch	Ah
19:51:22	*	kulelu88 joined #nim
19:51:22	PMunch	So http would need to have an import like "import http/ftp/client"
19:51:33	dom96	yeah
19:51:41	PMunch	(Not that you should do that)
19:51:47	PMunch	That makes sense
19:51:54	dom96	Indeed. You could have a package named 'network_utils'
19:52:00	dom96	Where that would make sense
19:52:06	PMunch	Yeah
19:52:41	Araq	yeah, so what I said.
19:52:56	Araq	--nimblePath should work differently
19:53:14	*	Ven joined #nim
19:54:55	*	bjz quit (Ping timeout: 256 seconds)
19:54:59	dom96	Araq: I don't really understand what you propose
19:55:52	demi-	work differently how?
19:55:55	*	bjz_ joined #nim
19:56:34	*	BitPuffin|osx joined #nim
19:56:55	Araq	nimblePath is too greedy. it adds .nimble/pkgs/p-1.0 to the search path
19:57:14	Araq	and then in p-1.0 I need yet another 'p' dir to get the "namespacing"
19:57:37	Araq	instead it should only add .nimble/pkgs
19:57:48	dom96	I see. And then treat p-1.0/ as p/
19:57:54	Araq	exactly.
19:57:56	PMunch	Hopefully :P
19:58:09	dom96	This breaks the "Jester case"
19:58:21	PMunch	Which is?
19:58:29	dom96	I want to be able to write `import jester`
19:58:33	dom96	not `import jester/jester`
19:58:52	PMunch	Maybe when you ignore -1.0 you could also ignore p/p?
19:59:47	*	nsf quit (Quit: WeeChat 1.7)
20:00:37	PMunch	Or maybe even add an option to the config file for a package
20:00:55	PMunch	So that you would be able to do "p/p as p" or something similar
20:01:32	PMunch	To allow the package creator more flexibility with folder structure
20:02:59	*	Ven quit (Ping timeout: 246 seconds)
20:04:34	Araq	we can have some sugar like 'import jester/_' for this
20:04:54	PMunch	That's still a bit ugly though..
20:05:16	PMunch	I mean if I run nimble install jester I assume the import will be import jseter
20:05:24	PMunch	s/jseter/jester
20:06:45	dom96	While the way it works currently may prove annoying, it does work and does ensure there are no unintended files installed.
20:06:59	*	kulelu88 quit (Ping timeout: 264 seconds)
20:07:37	dom96	Changing the way --nimblePath works won't help def-pri-pub
20:07:58	PMunch	def-pri-pub?
20:08:09	dom96	Read the IRC logs
20:08:22	PMunch	Sorry :Y
20:11:23	PMunch	Hmm, he just wants to stick to the C syntax though
20:11:51	PMunch	In that case I think the "Nimian" way (whatever that turns out to be) would be better for a Nim package
20:11:54	*	LeNsTR quit (Ping timeout: 240 seconds)
20:12:09	PMunch	So you have "import stb_image/write" or something like that
20:12:14	*	rokups quit (Quit: Connection closed for inactivity)
20:12:21	Araq	well I consider --nimblePath a mistake, I use --noNimblePath more and more often and explicit --path commands
20:12:54	dom96	Araq: Why?
20:13:18	*	Ven joined #nim
20:13:47	Araq	because I tend to have outdated stuff in my pkgs/ that interferes with my compilations
20:14:37	Araq	that's what 'nimble build' itself does. it uses --noNimblePath plus explicit --paths, right?
20:14:37	*	Ven quit (Read error: Connection reset by peer)
20:15:19	dom96	yes
20:15:53	Araq	so nimble doesn't want this feature and I don't want it either. ;-)
20:16:08	dom96	Well, I do.
20:16:25	Araq	maybe not. consider this:
20:17:02	Araq	I install 'jester'. then my program has a dependency to 'jester'.
20:17:14	Araq	so I add jester to my own .nimble file as dependency.
20:17:30	Araq	and that's enough for 'import jester' to work then, no --nimblePath required
20:18:02	*	couven92 joined #nim
20:18:10	Araq	well that *could* be enough. maybe the compiler should read .nimble files.
20:19:08	*	kulelu88 joined #nim
20:19:12	dom96	This is not how you use this
20:19:29	dom96	"Hrm, I wanna try this new fangled jester package"
20:19:33	dom96	nimble install jester
20:19:37	*	dexterk_ joined #nim
20:19:47	dom96	echo "import jester; blah" > test.nim
20:19:49	dom96	nim c test.nim
20:20:06	dom96	Do you seriously want me to create a .nimble file every time?
20:20:11	*	dexterk quit (Ping timeout: 264 seconds)
20:20:51	PMunch	Yeah that would be a pain
20:21:23	Araq	<cwd> nimble install jester # nimble creates the cwd.nimble file if not existing
20:21:53	*	chemist69 quit (Ping timeout: 246 seconds)
20:21:56	Araq	but yes. I think "creating a .nimble file every time" is the lesser evil
20:22:10	PMunch	But I don't install jester for every program that uses it :S
20:22:25	PMunch	And I tend to open a new shell when installing things
20:22:30	Araq	so add jester to your --path in your $nim.cfg
20:22:49	Araq	we already have dozens of solutions for this in the config system
20:22:57	*	Arrrr quit (Read error: Connection reset by peer)
20:23:00	PMunch	But that's annoying..
20:23:17	PMunch	I feel like nimble install jester, import jester should work
20:23:36	PMunch	Otherwise I'd say something was wrong with the package manager (or another part of the system)
20:23:52	dom96	Araq: The real problem is that you have outdated Nimble packages in your $nimbleDir
20:24:21	Araq	dom96: yes and no. I am also a control freak.
20:24:30	dom96	Araq: This is already an issue and will be fixed
20:24:38	dom96	Why not?
20:25:08	dom96	But hey, if you hate it so much feel free to remove `nimblePath` from your compiler's config file
20:25:24	*	chemist69 joined #nim
20:25:42	Araq	problem with this is my config file is the config file everybody uses
20:26:00	Araq	since it's in Nim's repo.
20:26:15	dom96	Then this problem should be fixed
20:27:06	dom96	yay, aporia-git in arch was updated. Thank you to whoever did that :)
20:27:40	Araq	well IMO --nimblePath causes a fragile compilation model and enforces a rigid dir structure for Nimble packages. and Nimble itself doesn't use it.
20:28:05	*	libman joined #nim
20:28:32	Araq	lots of other solutions are thinkable.
20:28:48	PMunch	Nim is still stuck on 0.15.0 in Arch though..
20:28:50	Araq	for example, the existance of foo.nimble disables --nimblepath.
20:29:01	*	bozaloshtsh quit (Ping timeout: 268 seconds)
20:29:04	Araq	then you can have your easy 'import jester' to play with
20:29:15	Araq	but as soon as you have explicit deps, these are used instead.
20:30:23	*	shashlick quit (Ping timeout: 264 seconds)
20:31:07	dom96	Araq: The rigid dir structure is still necessary for packages even without --nimblepath
20:31:42	dom96	What does your proposed solution solve?
20:31:48	dom96	It only makes things more complicated
20:32:39	*	shashlick joined #nim
20:33:03	*	Ven joined #nim
20:33:47	Araq	fine, whatever
20:38:08	dom96	PMunch: btw, sorry about giving you the IRC equivalent of RTFM earlier :)
20:38:21	PMunch	Haha, no problem :P
20:38:34	PMunch	I should've read the log before asking silly questions
20:40:41	*	kulelu88 quit (Ping timeout: 260 seconds)
20:46:27	*	Ven quit (Ping timeout: 260 seconds)
20:51:59	*	user0_ quit (Read error: Connection reset by peer)
20:53:12	*	Ven joined #nim
20:53:36	*	kulelu88 joined #nim
20:55:54	*	Vladar joined #nim
20:56:03	*	devted quit (Quit: IRC Client has quit.)
20:56:26	*	user0_ joined #nim
21:00:37	Vladar	ok, I think I finally figured out that but with appimages
21:01:14	Vladar	It cost me running ubuntu live usb for a couple of hours so it better be working now…
21:02:11	*	Ven quit (Ping timeout: 264 seconds)
21:05:47	*	Ven joined #nim
21:12:16	Vladar	def-pri-pub, rauss, are you still here?
21:19:09	*	bozaloshtsh joined #nim
21:20:42	*	Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
21:20:50	Vladar	anyway, here's working (I hope!) appimages for both architectures:
21:21:09	Vladar	https://github.com/Vladar4/ng2planetoids/blob/master/ng2planetoids-i386.run
21:21:16	Vladar	https://github.com/Vladar4/ng2planetoids/blob/master/ng2planetoids-x86_64.run
21:21:55	Vladar	If any Linux users have free time to test-run it, go ahead =)
21:22:08	Araq	Vladar: what was the problem?
21:22:25	Araq	(maybe at some point I feel like producing binaries for linux...)
21:23:13	PMunch	Vladar, just run the .run file?
21:23:46	Vladar	Araq, interesting bug, triggering only when run inside an appimage, caused by a const call to os module: https://github.com/Vladar4/ng2planetoids/commit/b18a43b13ea0d9fbc68b2098deef0e2ac7f26bff#diff-8886a6b85db9f93f54efa34d256d81b5
21:23:54	Vladar	PMunch: that's the idea
21:24:50	PMunch	Seems to work fine
21:24:56	Vladar	Araq: I know, stupid mistake, but worked when not in the appimage just fine for some reason
21:25:12	Araq	yeah well. I think you know why that couldn't work :-)
21:25:15	PMunch	Manjaro i3
21:25:27	Vladar	PMunch: thanks, now I could sleep well )
21:25:27	Araq	interesting we actually support this at compiletime
21:25:45	Vladar	PMunch: 32 or 64 one?
21:25:49	PMunch	64
21:25:53	Vladar	nice
21:26:03	PMunch	I don't have my IBM-6050 at hand :P
21:26:20	PMunch	(Pretty sure that's not actually a machine, and if it was it would probably be 8-bit or something :P)
21:26:55	Vladar	It was such a pain to build entire SDL2*, Nim and the project on a live usb… man…
21:27:27	Vladar	I should setup persistent ubuntu usb tomorrow
21:27:42	PMunch	I can recommend Manjaro if you need Linux
21:28:14	Vladar	I need ubuntu 12.04 LTS as a lowest common denominator for my appimage builds, I run Arch Linux myself
21:29:50	Vladar	btw, the engine show pretty good 60FPS performance on Pentium 4, and even works on my IBM T30 nobebook _decent_ enough
21:30:14	stisa	are there packages for the js target that use enums a lot? I think I fixed a bug with enums with holes, but I want to check I didn't break something else
21:33:46	*	bjz_ quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
21:39:12	FromGitter	<Varriount> Are we going to be using appimages to distribute Nim now?
21:39:31	dyce[m]	how would you your html/css/js with jester so that everything is contained in the executable?
21:39:44	FromGitter	<Varriount> dyce: Use staticRead
21:39:44	dyce[m]	would you package*
21:40:22	FromGitter	<Varriount> dyce[m]: https://nim-lang.org/docs/system.html#staticRead,string
21:40:43	dyce[m]	Varriount: thanks
21:40:54	PMunch	Vladar, oh that's a good point
21:40:58	PMunch	And nice that it runs so well
21:41:27	Vladar	PMunch: Just tried it on T30 and got 60FPS most of the time!
21:41:38	PMunch	That's really good!
21:41:51	PMunch	I mean, the game is basic enough but it's nice to see such good performance
21:41:58	dyce[m]	is it just me or is nodejs/bower/webpack making frontend development become over engineered like java
21:41:59	PMunch	Using SDL2?
21:42:28	Vladar	PMunch: yes, there's a lot of collision detection going on under the hood too
21:42:41	PMunch	Oh right, that's a good point
21:43:27	Vladar	and now I'm ready enough for Linux game jam this weekend
21:43:31	PMunch	I've been playing with creating a game library (not an engine) for SDL2 in Nim myself
21:43:37	PMunch	Oh cool
21:43:44	PMunch	Wish there were more of those where I live
21:43:55	PMunch	Game jams in general, never seen a Linux one here :P
21:43:57	Calinou	Vladar: use a chroot for building your AppImages, or a virtual machine
21:43:57	Calinou	:P
21:44:08	Calinou	or systemd-nspawn if your target is a modern enough distro
21:44:20	Vladar	PMunch: https://forum.nim-lang.org/t/2841
21:44:25	Calinou	dyce[m]: Bower is pretty much dead, be happy
21:44:40	Calinou	dyce[m]: and unfortunately, modern web workloads tend to drive towards Webpack since gulp is too limited
21:44:46	Calinou	Webpack can do really nice things, but it's also really hard to master
21:44:50	PMunch	Oh, it's a global thing :P
21:44:51	Vladar	Calinou, any tutorials to setup ubuntu 12 this way?
21:45:00	PMunch	Guess I'm just not paying attention
21:45:10	Calinou	Vladar: note that Ubuntu 12.04 does not use systemd, so you can't use systemd-nspawn, but you can use a chroot
21:45:20	Calinou	https://help.ubuntu.com/community/BasicChroot
21:45:28	Vladar	thanks, I'll look into it
21:45:29	Calinou	this guide is valid for any host distro (or almost)
21:45:42	Calinou	*perhaps* systemd-nspawn can work with Ubuntu 12.04, but with caveats
21:46:05	PMunch	https://github.com/PMunch/SDLGamelib/tree/master/gamelib
21:46:11	PMunch	That's my game library by the way
21:46:25	dyce[m]	Calinou: i started learning angular 1 and it was dead simple easy. no backend tools required, now everything requires webpack and a developer environment
21:46:32	PMunch	Currently it's pretty basic, but it supports textureatlases as created by TexturePacker
21:46:40	Calinou	dyce[m]: Vue.js can be used without a build system (for basic usage)
21:47:43	dyce[m]	Calinou: yea i started using Vue, but it also recommends webpack for routing
21:47:48	Vladar	PMunch: interesting approach
21:47:50	PMunch	http://backbonejs.org/ is pretty cool. But it requires a REST API that outputs JSON
21:48:21	PMunch	Vladar, yeah. I was reading this: https://hookrace.net/blog/writing-a-2d-platform-game-in-nim-with-sdl2/
21:48:40	PMunch	And I liked the simplicity and transparency that the style of programming offered for game dev
21:48:54	PMunch	So many times have I been annoyed by magic in the underlying engine
21:49:22	Vladar	may be you could use some of my engine's code there, like collisions, etc. https://github.com/Vladar4/nimgame2/blob/master/nimgame2/collider.nim
21:49:33	PMunch	So with that library I wanted to focus on letting people pick the features they needed and use that
21:50:06	PMunch	Yeah, the collision stuff is the last I added. It was more to have at least something
21:50:13	PMunch	And there are many other things that could be added
21:51:59	dyce[m]	riotjs seems pretty dead simple
21:51:59	*	rauss quit (Quit: WeeChat 1.7)
21:52:44	Vladar	PMunch: I struggle with any geometry stuff, but this link helped me a lot: http://paulbourke.net/geometry/
21:52:52	Vladar	tons of ready to use algorithms
21:53:06	PMunch	Huh, that's cool
21:53:46	PMunch	I've got a pretty good grasp of it since I've been doing gamedev as a hobby and partially as a job for the past 14 years or so :P
21:54:05	PMunch	But those algorithms are probably more efficient than what I would write, so it's a nice resource for sure
21:54:47	Vladar	whoa, 14 years of professional gamedev?
21:55:49	*	bozaloshtsh quit (Ping timeout: 260 seconds)
21:56:12	PMunch	No
21:56:20	PMunch	I started tinkering with it 14 years ago
21:56:47	PMunch	But at the time I was 10, so I didn't start working with it professionally until I was 17-18
21:57:15	PMunch	Now I'm soon through my masters degree in computer science and I'm thinking of getting into gamedev full-time
21:58:15	Vladar	Nice. Interesting enough, I've got about 14 years of amateur/enthusiast gamedev myself, 0y professional yet, sadly )
21:59:00	PMunch	That's a shame. Tried getting a job and failed or just busy with other stuff?
21:59:07	PMunch	Gamedev is a lot of fun
21:59:47	Vladar	worked an engineer job previously, and doing some freelancing stuff now
22:00:08	PMunch	Cool
22:00:34	PMunch	I've done some freelance besides my studies but I think it would stress me too much to live off-of it
22:03:00	Vladar	it depends on the work mostly )
22:03:49	*	gokr quit (Ping timeout: 240 seconds)
22:04:40	PMunch	Yeah, but getting it :P
22:05:26	Vladar	yeah for sure )
22:06:46	Vladar	OK, going to sleep now. See you around.
22:07:05	PMunch	Yeah, see you
22:07:52	*	Vladar quit (Quit: Leaving)
22:28:15	*	Jesin quit (Quit: Leaving)
22:39:17	*	_yeeve joined #nim
22:39:32	*	Jesin joined #nim
22:41:49	*	yeeve quit (Ping timeout: 240 seconds)
22:54:21	*	bozaloshtsh joined #nim
22:56:11	*	Jesin quit (Quit: Leaving)
22:56:13	*	nsf joined #nim
22:58:31	*	rauss joined #nim
23:00:30	*	bozaloshtsh quit (Quit: WeeChat 1.7)
23:00:37	*	kulelu88 quit (Ping timeout: 260 seconds)
23:02:39	*	user0_ quit (Read error: Connection reset by peer)
23:03:42	*	user0_ joined #nim
23:06:54	*	def-pri-pub joined #nim
23:07:10	*	Jesin joined #nim
23:07:51	*	nsf quit (Quit: WeeChat 1.7)
23:12:39	*	kulelu88 joined #nim
23:17:01	*	couven92 quit (Quit: Switching to VPN)
23:17:33	*	couven92 joined #nim
23:23:16	*	user0_ quit (Read error: Connection reset by peer)
23:23:32	*	user0_ joined #nim
23:46:28	*	arnetheduck quit (Ping timeout: 240 seconds)
23:46:40	*	ftsf_ joined #nim
23:49:24	*	kulelu88 quit (Ping timeout: 240 seconds)
23:52:18	*	elrood quit (Quit: Leaving)
23:53:55	*	user0__ joined #nim
23:55:45	*	user0_ quit (Ping timeout: 256 seconds)
23:57:28	*	user0__ quit (Read error: Connection reset by peer)
23:58:08	*	user0__ joined #nim