<< 10-05-2020 >>

---

00:00:01	*	junland quit (Quit: %ZNC Disconnected%)
00:01:51	*	dadada joined #nim
00:02:14	*	dadada is now known as Guest62079
00:03:41	*	junland joined #nim
00:05:38	nisstyre	oh the null byte is always there in string, ok
00:18:50	*	Guest76944 quit (Ping timeout: 260 seconds)
00:25:50	*	Guest62079 quit (Ping timeout: 260 seconds)
00:31:49	*	dadada__ joined #nim
00:39:30	*	exelotl joined #nim
00:51:24	skrylar[m]	pushed the docs for the xxhash update
00:54:59	*	Hideki joined #nim
00:55:22	*	Hideki is now known as Guest38843
00:56:02	*	dadada__ quit (Ping timeout: 272 seconds)
01:00:40	*	Guest38843 quit (Remote host closed the connection)
01:01:47	*	dadada joined #nim
01:02:10	*	dadada is now known as Guest56553
01:03:04	*	konvertex quit (Ping timeout: 256 seconds)
01:15:45	skrylar[m]	i don't know that its worth it, but i figured it might be better to just store hashes of parameter blocks and then when changing 'brushes' it just checks if the gl brush is the same as what is already used... maybe cuts a pile of glget tests
01:16:12	skrylar[m]	altho dunno yet if glget 'counts' as a gpu call
01:25:57	*	Guest56553 quit (Ping timeout: 260 seconds)
01:31:43	*	dadada__ joined #nim
01:39:53	FromDiscord	<Yardanico> So guys can you share your thoughts in https://github.com/nim-lang/Nim/issues/14272 ?
01:39:55	disbot	➥ symbols defined in nim.cfg file aren't visible in user code? ; snippet at 12https://play.nim-lang.org/#ix=2leZ
01:41:22	*	chemist69 quit (Ping timeout: 256 seconds)
01:43:36	*	chemist69 joined #nim
01:48:46	skrylar[m]	@Yardanico it should be fixed but at least its not hard to work around
01:49:50	FromDiscord	<Yardanico> Yeah I know
01:50:12	*	NimBot joined #nim
01:50:15	FromDiscord	<Yardanico> It's just that it's not documented anywhere except the NimScript manual where it's just a note
01:50:31	*	Hideki is now known as Guest39366
01:50:34	FromDiscord	<Yardanico> And it's not really related to nim.cfg
01:52:30	FromDiscord	<Yardanico> (I mean that note in NimScript)
01:52:31	skrylar[m]	i don't suppose theres an undefine flag you could shove in the cfg
01:54:56	*	Guest39366 quit (Ping timeout: 272 seconds)
01:55:50	*	dadada__ quit (Ping timeout: 256 seconds)
01:56:22	FromDiscord	<Yardanico> It's not about that really
01:56:33	FromDiscord	<Yardanico> It's about the order of evaluation of configs
01:57:22	FromDiscord	<Yardanico> User config gets evaluated the last, so the switches under @if release in main nim config don't get enabled
01:57:46	FromDiscord	<Yardanico> And nim compiler switches (passed in command line) are always there before any configs
01:58:13	skrylar[m]	imho there isn't anything wrong with that, but with using the general define syntax to denote a release build
01:59:03	skrylar[m]	it should be fine for system configs to include stuff like -d:corporateoverlordflag
02:01:47	*	dadada joined #nim
02:02:12	*	dadada is now known as Guest35672
02:06:45	skrylar[m]	re-reading the ticket. i dunno. i don't think -d:release should be special. if its going to be special it should be --release, things shouldn't use generic syntax like `-d:` if they are not going to behave generically
02:07:51	skrylar[m]	other than that the order of execution seems okay, esp. if there is an undefine statement somewhere
02:11:17	*	monok joined #nim
02:13:58	*	mono quit (Ping timeout: 256 seconds)
02:14:03	skrylar[m]	i think the part about game engines i didn't grok until now is you were never supposed to rely on libs as external things but to tear them apart and graft them on to the engine. people don't worry about the deps going stale because they just fuse the deps to the engine and never update them anyway
02:17:26	*	muffindrake quit (Ping timeout: 244 seconds)
02:19:31	*	muffindrake joined #nim
02:25:58	*	Guest35672 quit (Ping timeout: 272 seconds)
02:31:50	*	dadada__ joined #nim
02:32:51	skrylar[m]	hrmm. i kind of expected this to work `with foo: a = 1; b = 2` but it seems you are intended not to use with as a block, and also it does not work with field assignments
02:47:53	skrylar[m]	dumb question; what are we using for weak references these days
02:55:22	*	dadada__ quit (Ping timeout: 246 seconds)
02:55:23	disruptek	cursor
02:55:26	disruptek	!memory
02:55:27	disbot	total: 360 Megabytes; free: 95 Megabytes; owned: 250 Megabytes; max: 360 Megabytes
02:56:20	disruptek	skrylar[m]: that was to you, btw. {.cursor.}
03:01:50	*	dadada joined #nim
03:02:13	*	dadada is now known as Guest870
03:05:07	skrylar[m]	i don't see a cursor in the language manual or the index o.o
03:07:42	disruptek	arc weak refs.
03:08:00	disruptek	var foo {.cursor.} = someRef
03:09:35	disruptek	i guess they aren't in doc/destructors.rst for some reason.
03:11:22	disruptek	there are no cursors for closures, though.
03:14:20	*	Senketsu joined #nim
03:14:39	skrylar[m]	yea they aren't referenced anywhere that i can find
03:24:34	disruptek	~cursor is ```let foo {.cursor.} = anyRef # with --gc:arc```
03:24:35	disbot	cursor: 11```let foo {.cursor.} = anyRef # with --gc:arc```
03:24:53	disruptek	~cursor is let foo {.cursor.} = anyRef # with --gc:arc
03:24:53	disbot	cursor: 11let foo {.cursor.} = anyRef # with --gc:arc
03:24:58	disruptek	better.
03:25:53	disruptek	~bentley lyybb
03:25:54	disbot	no matching footnotes for `bentley` with regexps `lyybb`. 🙁
03:26:01	*	Guest870 quit (Ping timeout: 264 seconds)
03:31:58	*	dadada__ joined #nim
03:55:20	*	dadada__ quit (Ping timeout: 256 seconds)
04:01:49	*	dadada joined #nim
04:02:12	*	dadada is now known as Guest21835
04:06:02	*	supakeen quit (Quit: WeeChat 1.9.1)
04:06:44	*	supakeen joined #nim
04:26:01	*	Guest21835 quit (Ping timeout: 264 seconds)
04:31:53	*	dadada__ joined #nim
04:55:24	*	dadada__ quit (Ping timeout: 256 seconds)
05:00:43	*	dadada joined #nim
05:01:13	*	dadada is now known as Guest63444
05:02:38	skrylar[m]	contemplating a design pattern in nim
05:03:23	*	Hideki joined #nim
05:03:47	*	Hideki is now known as Guest29324
05:09:18	skrylar[m]	pfft you're actually allowed to use typeof when defining parameter types
05:11:46	skrylar[m]	anyway the idea is for a type that has to have an up to date hash/dirty field if instead of saying "make sure to run 'update' when you are done" if you have a macro that just sets write mode and so it like stuffs all the mutation in a sub-block with a defer to do it. so it reads `change sometexture: sometexture.magfilter = ...`
05:18:27	skrylar[m]	https://gist.github.com/Skrylar/8c42061559e03894fb009210d0d25857 hm
05:20:47	*	Guest29324 quit (Remote host closed the connection)
05:31:05	*	solitudesf joined #nim
05:46:10	*	Guest63444 quit (Ping timeout: 246 seconds)
05:48:37	*	exelotl quit (Ping timeout: 246 seconds)
05:54:57	*	Hideki joined #nim
05:55:21	*	Hideki is now known as Guest85298
06:08:54	*	Guest85298 quit (Ping timeout: 272 seconds)
06:25:34	*	Senketsu quit (Quit: WeeChat 2.8)
06:53:08	*	rockcavera quit (Remote host closed the connection)
07:38:09	*	dadada joined #nim
07:38:34	*	dadada is now known as Guest65412
07:43:31	FromDiscord	<Yardanico> Good morning I guess
07:45:20	Prestige	o/
07:45:20	FromGitter	<bung87> good afternoon
07:45:25	Prestige	Sup @Yardanico
07:51:59	Prestige	If I assign a new variable to someOption.get(), it won't copy the variable right?
07:53:34	FromDiscord	<Yardanico> Not sure
07:54:04	FromDiscord	<Yardanico> !eval import options; let a = some("hi"); echo a.get(), a.get()
07:54:07	NimBot	hihi
07:55:26	Prestige	!eval import options; let a = some("hi"); let b = a.get(); echo a.get(), b
07:55:28	NimBot	hihi
07:55:55	Prestige	Wonder if b is a copy of the value in a, like if we were using objects
07:56:33	FromDiscord	<Yardanico> !eval import options; var a = some("hi"); let b= a.get(); a = some("lo"); echo a.get(), b.get()
07:56:36	NimBot	Compile failed: /usercode/in.nim(1, 85) Error: type mismatch: got <string>
07:57:36	FromDiscord	<Yardanico> I guess with arc if you annotate get proc as "lent" it won't make a copy in most cases but I don't know if that's correct
07:58:52	Prestige	Hm okay, interesting. I'd hope it would just return the same object with Options
07:59:08	Yardanico	well let's see
07:59:44	skrylar[m]	@Yardanico mornin'
08:00:24	Yardanico	Prestige:
08:00:27	Yardanico	!eval import options; var a = some("abc");var b = a.get();a = some("def");echo a.get(), b
08:00:31	NimBot	defabc
08:00:42	Yardanico	a.get() changed but not b
08:00:53	Prestige	Wouldn't that only change a?
08:01:16	Yardanico	well if it didn't copy b would change too
08:01:31	Yardanico	ah actually .get() returns var T hmm
08:03:33	Yardanico	with ref objects it'll be the same as expected
08:03:54	Prestige	I was attempting an example: https://play.nim-lang.org/#ix=2lvg
08:03:58	Yardanico	!import options; var da = new(string); da[] = "abc"; var a = some(da);var b = a.get(); da[]="def"; a = some(da);echo (a.get()[], repr b)
08:04:00	Prestige	but it looks like I cannot assign to y.s here
08:04:10	Yardanico	because you did "let y"
08:04:13	Yardanico	and it's an "object"
08:04:13	Prestige	woops
08:04:14	Yardanico	not "ref object"
08:04:25	Yardanico	"object" is a value type
08:04:49	Yardanico	"An object is a value type, which means that when an object is assigned to a new variable all its components are copied as well."
08:04:52	Yardanico	https://nim-lang.org/docs/tut1.html#advanced-types-objects
08:05:58	Prestige	Yeah, just not entirely clear on how ref objects work I suppose
08:06:09	Yardanico	ref objects are managed pointers to objects
08:06:17	Yardanico	"ref" is a managed (by the GC or other MM) pointer
08:06:32	Yardanico	so you copy the pointer
08:06:36	Yardanico	but it still points to the same place
08:06:39	skrylar[m]	yup. and ptr is unmanaged, and pointer is untyped
08:06:43	skrylar[m]	(and unmanaged)
08:06:48	Yardanico	pointer is void* in C yeah
08:06:49	Prestige	ah so like https://play.nim-lang.org/#ix=2lvj
08:06:59	Yardanico	Prestige: yep
08:07:03	Prestige	Perhaps that's what I want for my project
08:07:07	Prestige	Thanks Yardanico
08:07:32	Yardanico	to copy ref objects recursively we have deepCopy
08:07:37	Yardanico	although it doesn't work with arc (yet)
08:10:09	*	konvertex joined #nim
08:13:56	*	chemist69 quit (Ping timeout: 244 seconds)
08:14:20	*	chemist69 joined #nim
08:15:13	*	dddddd joined #nim
08:21:15	*	exelotl joined #nim
08:30:18	Yardanico	nice seems like https://github.com/def-/nimes/pull/6 will be merged :P
08:30:19	disbot	➥ Use streams-like interfaces instead of methods
08:30:49	Yardanico	so we have 1 more easy example to show people how to use streams-like interfaces :P
08:32:03	Yardanico	now isObj is only used once in the entire C source code and only for checking exception type
08:32:11	Yardanico	if (isObj(nimBorrowCurrentException()->Sup.m_type, "|ValueError|CatchableError|Exception|RootObj|")) {
08:32:31	Yardanico	(with arc)
08:38:37	Yardanico	yay it got mergefd
08:48:17	Yardanico	lol seems like changing from methods to interfaces made it use ~5% less cpu (from 25% to 20%)
08:48:23	Yardanico	and with LTO + PGO it uses ~15%
08:48:40	Yardanico	I mean one core of course
08:48:43	Yardanico	one cpu thread*
08:54:12	*	ertp07 quit (Ping timeout: 272 seconds)
08:55:04	*	ertp07 joined #nim
08:57:04	*	ertp07 quit (Read error: Connection reset by peer)
08:57:22	*	ertp07 joined #nim
09:01:28	*	letto quit (Quit: Konversation terminated!)
09:02:13	*	letto joined #nim
09:06:25	*	ftsf joined #nim
09:07:29	*	ertp07 quit (Read error: Connection reset by peer)
09:10:37	*	ertp07 joined #nim
09:11:25	Yardanico	I managed to compile it with emscripten (well it was already done by def- but I did it with arc and latest nim)
09:11:37	def-	Yardanico: cool
09:11:42	Yardanico	def-: oh you're here :P
09:11:42	skrylar[m]	we have interfaces now?
09:11:53	Yardanico	skrylar[m]: object with proc pointers
09:11:59	Yardanico	https://forum.my-toolbox.xyz/out/hello.html
09:12:04	skrylar[m]	oh.
09:12:07	Yardanico	ah wait forgot to modify html file
09:12:31	Yardanico	https://forum.my-toolbox.xyz/out/hello.html works now
09:12:57	skrylar[m]	methods are uh, binary dispatch trees on a type code if i recall. used to have a springer paper that talked about how they're pretty nice
09:13:08	Yardanico	skrylar[m]: well with arc methods use string comparisons :P
09:13:20	skrylar[m]	absolute trash
09:13:20	Yardanico	methods can be implemented differently, you know
09:13:56	skrylar[m]	the bindispatch technique is nice if you don't want a full fat pointer
09:14:04	skrylar[m]	guess i wouldn't be surprised fat pointers were faster tho
09:15:17	*	ertp07 quit (Ping timeout: 258 seconds)
09:16:02	*	ertp07 joined #nim
09:17:11	skrylar[m]	don't closures allocate heap stuff so an object ful of closures is actually a heap alloc per method
09:17:26	Yardanico	well you can always do {.cdecl.} as krux recommended me to do
09:17:31	Yardanico	so these function pointers won't be closures
09:24:56	*	Senketsu joined #nim
09:25:21	*	Senketsu quit (Client Quit)
09:30:06	Yardanico	hah, I managed to use gmail smtp for nimforum, nice, now you can register on https://forum.my-toolbox.xyz/ :P
09:30:09	Yardanico	and email activation works
09:30:35	Yardanico	literally 1-line change (added await client.startTls() because google smtp works with TLS only)
09:32:08	Yardanico	now I need to learn how to remove moderated status lol
09:32:47	*	krux02 joined #nim
09:33:04	Yardanico	ah I found out
09:33:14	Yardanico	"You can modify anyone's rank. Remember: with great power comes great responsibility."
09:33:33	Yardanico	nimforum backend is surprisingly lightweight
09:35:53	planetis[m]	Yardanico: you didn't have to change the `[]` procs to idx, just the proc fields and then keep the base "methods" which just call the fields
09:36:55	planetis[m]	oh there templates now cool
09:37:38	planetis[m]	i have a macro for that btw: https://github.com/b3liever/protocoled
09:37:54	planetis[m]	but hadn't got any usage yet
09:37:56	Yardanico	yeah I figured out there'd be a macro for that, niec :)
09:38:19	Yardanico	really cool
09:38:38	Yardanico	i'll add it to https://github.com/nim-lang/Nim/wiki/Curated-Packages so it doesn't get lost
09:38:57	planetis[m]	thanks!
09:39:39	Yardanico	added under Language -> Metaprogramming
09:39:44	Yardanico	sorry under Language -> Object orientation
09:40:24	planetis[m]	you are fast lol
09:40:33	planetis[m]	didn't expect to see it there
09:41:50	planetis[m]	there is also https://nimble.directory/pkg/interfaced and another one similar
09:42:09	planetis[m]	https://github.com/bluenote10/oop_utils
09:50:11	*	krux02 quit (Remote host closed the connection)
09:52:03	*	Guest65412 quit (Ping timeout: 260 seconds)
09:58:39	*	Vladar joined #nim
10:06:20	*	Senketsu joined #nim
10:25:26	federico3	https://github.com/mukul-rathi/bolt <- more fine-grained concurrency than Rust
10:28:45	*	Trustable joined #nim
10:29:15	*	krux02 joined #nim
10:29:16	*	PMunch joined #nim
10:32:30	*	liblq-dev joined #nim
10:32:42	FromDiscord	<wiremoons> I have not been using Nim very long - but have written a few blog posts about some of the things I have learnt so far: https://www.wiremoons.com/
10:33:20	FromDiscord	<wiremoons> Probably of limit use to more experienced users 🙂
10:34:56	Yardanico	@wiremoons about https://www.wiremoons.com/development/debugging/nim/2020/04/18/Nim-Debugging-Approaches.html
10:35:14	Yardanico	there's a very useful proc in the system module https://nim-lang.org/docs/system.html#instantiationInfo
10:35:23	dom96	wiremoons: nice! Thanks for writing these up! :)
10:35:26	Yardanico	with it you can show on which lines of code you printed your debug statements
10:35:30	Yardanico	but yeah, nice posts :)
10:35:36	FromDiscord	<Recruit_main707> @wiremoons ^
10:36:43	PMunch	@wiremoons yeah these are actually pretty nice
10:36:46	FromDiscord	<wiremoons> Thanks for the link @Yardanico - I will have a look
10:37:52	federico3	https://github.com/mukul-rathi/bolt-dissertation/blob/master/dissertation.pdf
10:38:57	FromDiscord	<wiremoons> @Recruit_main707 🙂
10:39:07	Yardanico	@wiremoons something like https://play.nim-lang.org/#ix=2lww
10:39:36	Yardanico	you can also get full paths with instantiationInfo(true)
10:39:42	Yardanico	ah sorry
10:39:49	Yardanico	instantiationInfo(fullPaths = true)
10:40:12	Yardanico	then it'll show absolute paths to your code so if your editor supports that you can directly use Ctrl+Click to go to that line of code
10:40:16	Yardanico	(vscode supports that for example)
10:41:43	PMunch	By the way @wiremoons, looking at the latest post: {.strdefine.} is a way to tell Nim to extract a parameter given by the compiler: https://nim-lang.org/docs/manual.html#implementation-specific-pragmas-compile-time-define-pragmas what is new in that Nimble version is that it passes it as a flag to the Nim compiler (as you can see if you do `nimble build --debug`, looks something like this: ` -d:NimblePkgVersion=0.1.0`)
10:41:51	FromDiscord	<wiremoons> @Yardanico - I have a module I use for adding debugging now, so I can add your suggested extra bit if code to it!
10:42:25	Yardanico	yeah, and in more complex programs you can wire up all sorts of cool things like sending these debug logs to your server or something
10:42:43	Yardanico	also astToStr (shown in instantiationInfo) is cool
10:44:24	FromDiscord	<wiremoons> Thanks @PMunch - I hadn't got around to looking for that info yet, so thanks for the link
10:45:22	PMunch	federico3, baller page "ii" in that dissertation :P
10:45:27	Yardanico	something like https://play.nim-lang.org/#ix=2lwB
10:45:37	PMunch	"Special Difficulties - None."
10:46:01	PMunch	That's my problem with writing things on my site
10:46:23	PMunch	I tend to delve deep into things and try to understand every nook and cranny of it so I can explain it properly
10:46:28	PMunch	Takes forever to write stuff
10:47:07	PMunch	And more than once have I ended up with an article on hold while I'm waiting for a PR to merge to fix a bug I encountered :P
10:47:28	PMunch	Still waiting on this for example for my C interop article: https://github.com/nim-lang/Nim/pull/14258
10:47:29	disbot	➥ Allow let to not have value when using importc
10:50:24	*	ertp07 left #nim ("Leaving")
11:01:46	Yardanico	lol
11:01:47	Yardanico	https://github.com/nim-lang/Nim/issues/12382#issuecomment-626308745
11:01:50	disbot	➥ OSError: Bad file descriptor in asyncnet server ; snippet at 12https://play.nim-lang.org/#ix=2lwQ
11:02:00	Yardanico	replacing code with same code did the trick :P
11:02:36	FromDiscord	<Recruit_main707> classic
11:02:54	Yardanico	and here I am, know nothing about web dev and trying to add ability to remove categories in nim forum lol
11:03:18	Yardanico	well at least it's karax so it's all in nim code
11:09:41	*	fredrikhr joined #nim
11:10:43	*	ertp07 joined #nim
11:14:54	*	ertp07 quit (Ping timeout: 240 seconds)
11:21:52	*	ertp07 joined #nim
11:24:42	FromDiscord	<KrispPurg> thats life hacks
11:28:28	PMunch	Is there a way to define a macro that has an optional untyped argument with the default just being something like `discard`?
11:29:54	FromDiscord	<Rika> overload it?
11:32:21	krux02	PMunch, do default values for untyped not work?
11:33:40	krux02	you can also do ``args: varargs[untyped]`` and check the length of `args`
11:33:48	PMunch	krux02, I can't just do `default: untyped = discard`
11:34:11	PMunch	That's quite a misleading signature though..
11:35:22	*	Trustable quit (Remote host closed the connection)
11:35:23	krux02	I don't think so.
11:35:44	krux02	I think it would fit the default argument model pretty well.
11:36:01	PMunch	With `default: varargs[untyped]`?
11:36:05	krux02	but it doesn't work, so you have to work around it. And I guess overloads are the solution to go here.
11:36:20	PMunch	Yeah, that's what I'm thinking as well
11:36:29	krux02	yes default varargs[untyped] is a misleading signature
11:36:34	krux02	but it would work
11:36:44	PMunch	Tried to do it with `default: untyped = nnkDiscard.newTree(newEmptyNode())` but that didn't work either
11:36:50	PMunch	Produced a rather odd error message
11:38:21	*	Hideki_ joined #nim
11:42:03	PMunch	Is there no way to return an anonymous procedure from a macro?
11:42:16	PMunch	So you do something like `let myProc = generateProcMacro()`
11:42:27	PMunch	And can then call `myProc(someargument)`
11:43:20	Yardanico	hmmmm
11:43:24	FromDiscord	<mratsim> yes you can
11:43:42	FromDiscord	<mratsim> one of the macro in sugar does this, pretty sure it was the list comprehension macro
11:43:58	FromDiscord	<mratsim> it used nnkLambda.newTree
11:44:08	PMunch	Aah right
11:45:07	Yardanico	yeah https://play.nim-lang.org/#ix=2lx0
11:45:14	liblq-dev	does nim have a shell expansion thingy in the stdlib? eg. "(a|b|c)def" would get expanded to @["adef", "bdef", "cdef"]
11:45:27	Yardanico	well that syntax is used for .so files
11:45:31	Yardanico	hmm
11:45:39	liblq-dev	I know there's one in dynlib.nim, it's called libCandidates
11:45:48	liblq-dev	but using it feels... off
11:45:58	Yardanico	apart from it I don't know any
11:46:21	Yardanico	and I've never heard of shell expansion like that
11:48:12	liblq-dev	well bash and fish have {}
11:48:20	liblq-dev	eg. {a,b,c}dawd
11:48:29	liblq-dev	and they can nest
11:48:44	liblq-dev	libCandidates doesn't support nesting
11:48:53	Yardanico	do it yourself then :P
11:49:08	liblq-dev	too much work
11:49:22	liblq-dev	I'll just list all possibilities manually, I guess
11:52:15	Yardanico	planetis[m]: btw your protocoled package is invalid
11:52:21	Yardanico	you specify "srcDir = src" but you don't have that
11:52:26	Yardanico	so you can't use the package if you install it from nimble
11:55:08	Yardanico	btw it's the first time I see 3-space indent used in the wild :P
11:59:28	*	Senketsu quit (Quit: WeeChat 2.8)
12:02:47	*	krux02 quit (Remote host closed the connection)
12:04:09	*	krux02 joined #nim
12:06:01	*	supakeen quit (Quit: WeeChat 1.9.1)
12:06:43	*	supakeen joined #nim
12:07:58	*	Hideki_ quit (Remote host closed the connection)
12:08:49	*	Hideki_ joined #nim
12:09:49	*	ertp07 quit (Ping timeout: 264 seconds)
12:12:44	*	ertp07 joined #nim
12:13:26	*	Hideki_ quit (Ping timeout: 256 seconds)
12:26:14	*	ertp07 quit (Ping timeout: 256 seconds)
12:27:04	*	Hideki_ joined #nim
12:27:10	PMunch	Hmm, is there a way to inject a block label in a macro?
12:27:51	*	ertp07 joined #nim
12:28:22	PMunch	Like I have a macro that generates an infinite loop (event handler) but I want the user to be able to catch an event and break the loop (e.g. for exiting the program). I was thinking creating a `block eventLoop` around the whole thing would be a good way of doing it, and then have the user call `break eventLoop`
12:31:56	Yardanico	PMunch: https://nim-lang.org/docs/macros.html#statements-block-statement ?
12:32:42	FromDiscord	<mratsim> block statement first param is the block name
12:33:12	Yardanico	https://play.nim-lang.org/#ix=2lxl
12:34:38	PMunch	Huh
12:34:49	PMunch	That's pretty much exactly what I tried..
12:34:55	PMunch	I mean I did it like this: https://play.nim-lang.org/#ix=2lxm
12:35:00	PMunch	But that also works..
12:35:06	Yardanico	heh
12:35:14	PMunch	Aah
12:35:17	*	ertp07 quit (Read error: Connection reset by peer)
12:35:39	PMunch	Hmm, no that wasn't it..
12:35:55	PMunch	Oh wait..
12:36:06	PMunch	Hmm
12:36:49	Yardanico	protocoled is quite elegant ngl, although a bit of boilerplate but still cool :)
12:36:55	Yardanico	https://play.nim-lang.org/#ix=2lxn
12:37:12	Yardanico	stripped and adapted mathexpr to do parsing instead of evaluation :P
12:37:43	*	Hideki_ quit (Ping timeout: 260 seconds)
12:39:22	Yardanico	of course that will be slower than my mathexpr (because i don't construct any AST and immediately evaluate everything) but still cool
12:39:41	Yardanico	around 2x slower for 2+2*2-4^2+-1
12:40:07	Yardanico	6m vs 2.5m iterations/sec :P
12:40:49	Yardanico	funny fact: seems like clang optimizes mathexpr more than this protocol version
12:40:55	PMunch	Hmm, yeah my issue was that I pass the break statement into a macro that creates a pragma out of the body..
12:41:06	Yardanico	so mathexpr is faster with clang, but protocol is faster with gcc (I mean relatively to clang)
12:41:18	Yardanico	maybe gcc recognizes that pattern and does some more smart optimizations
12:41:25	PMunch	s/pragma/lambda
12:42:26	PMunch	Which is a bit weird as I create the lambda inside the block..
12:43:08	PMunch	Yeah, this fails: https://play.nim-lang.org/#ix=2lxo
12:45:09	PMunch	Ah, so block labels aren't captured in lambdas at all: https://play.nim-lang.org/#ix=2lxp
12:45:21	PMunch	Is that intended?
12:50:38	Yardanico	PMunch: I think that's not possible in Nim at all
12:50:45	Yardanico	you might have to use a global variable for that or something
12:50:58	PMunch	Found a workaround that works for my application
12:50:58	Yardanico	https://play.nim-lang.org/#ix=2lxu
12:53:29	Yardanico	btw it's the first time I saw the "Error: invalid control flow: test" error
12:53:52	FromDiscord	<clyybber> I think you get that when the VM can't find the break label
12:53:59	Yardanico	and see https://play.nim-lang.org/#ix=2lxu
12:54:05	FromDiscord	<clyybber> .s/VM/compiler
12:54:08	Yardanico	yeah
12:54:26	*	sirn joined #nim
12:55:17	Yardanico	also seems like my mathexpr is actually faster than https://github.com/codeplea/tinyexpr (well because i don't do AST at all)
12:55:26	Yardanico	nim-mathexpr initially was a wrapper to tinyexpr :P
12:55:44	Yardanico	https://github.com/Yardanico/nim-mathexpr/commit/039f5f07714b675ca51a1ce5d0bf9928c8b9b77d
12:55:48	Yardanico	https://github.com/Yardanico/nim-mathexpr/commit/64348b534edecb390c0035416217199358a3397f
13:01:02	*	zacharycarter joined #nim
13:03:56	*	solitudesf quit (Ping timeout: 256 seconds)
13:06:22	*	Hideki_ joined #nim
13:06:50	shashlick	Nimterop 0.5.0 is almost ready to be tagged, if anyone wants their wrapper tested before release, please get in touch
13:09:16	zacharycarter	shashlick: I'm wrapping cimgui with nimterop and I think I have an issue - not sure if it will be addressed with 0.5.0 or not
13:09:37	zacharycarter	but essentially there are serveral types that are predeclared and then declared again later
13:10:08	zacharycarter	nimterop skips the initial type defs since they're just empty structs but then doesn't define them when they're actually fully defined
13:11:09	krux02	zacharycarter, and it is not a forward def?
13:11:27	zacharycarter	here's an example
13:11:43	zacharycarter	https://github.com/cimgui/cimgui/blob/master/cimgui.h#L53
13:12:10	zacharycarter	it's fully defined later in the header but Nimterop just skips this all together
13:12:20	Yardanico	#ifdef CIMGUI_DEFINE_ENUMS_AND_STRUCTS
13:12:22	Yardanico	is that defined?
13:12:24	zacharycarter	yes
13:12:49	*	dddddd quit (Ping timeout: 264 seconds)
13:13:07	krux02	so tha alias is declared first, before the struct is declared.
13:13:17	zacharycarter	yes
13:13:18	krux02	That is interesting, I didn't know that this is valid C code.
13:13:35	krux02	I thought you had to have the struct definition first.
13:14:32	krux02	But the ``typedef struct Typename Typename;`` is a very common pattern. No reason not to support it in nimterop.
13:16:10	zacharycarter	well it just forces me to manually override all of these for now and then using nimterop in the first place becomes questionable
13:16:40	krux02	what if you comment them out?
13:16:53	krux02	is it feasible to invert the declaration order)
13:16:56	krux02	?
13:17:55	krux02	oh no it's not really, it really is a lot of API
13:18:46	krux02	zacharycarter: Did you see this: This file is automatically generated by generator.lua from https://github.com/cimgui/cimgui
13:19:58	*	leorize joined #nim
13:20:16	krux02	maybe it is feasible to extend generate.lua so that it generates Nim bindings.
13:20:23	zacharycarter	maybe
13:20:43	zacharycarter	I could probably also just use c2nim
13:20:47	zacharycarter	and automate that
13:21:49	*	Hideki_ quit (Remote host closed the connection)
13:22:07	*	Hideki_ joined #nim
13:22:41	PMunch	https://github.com/PMunch/xevloop
13:23:11	*	sz0 quit (Quit: Connection closed for inactivity)
13:28:02	PMunch	https://github.com/nim-lang/packages/pull/1466
13:28:02	disbot	➥ Add xevloop package
13:31:28	Yardanico	btw for nim forum frontend (the JS file) google closure compiler makes the file 3x smaller (well it's minimization but also advanced optimization)
13:31:34	Yardanico	and says "0 error(s), 266 warning(s), 66.1% typed" which is pretty good I guess
13:31:59	FromDiscord	<Rika> thats a lot of warnings
13:32:04	Yardanico	that JS code is beyond obfuscation though, it's completely undreadable
13:32:38	Yardanico	@Rika redeclared variables/unreachable code/condition always false/some other stuff
13:33:10	FromDiscord	<Rika> o i see
13:33:13	Yardanico	https://i.imgur.com/YOdCOXB.png
13:33:13	*	filcuc joined #nim
13:33:22	FromDiscord	<Rika> that bg tho 😮
13:33:46	Yardanico	found it in https://www.youtube.com/watch?v=6WQjgHuUDWU
13:35:05	Yardanico	I wonder what difference does -d:release vs -d:danger make
13:35:08	Yardanico	for nim forum frontend
13:35:27	Yardanico	with -d:release - 696kb, with -d:danger - 660kb
13:36:00	Yardanico	also 70% typed now
13:36:19	Yardanico	and output from closure compiler is 204kb
13:36:47	Yardanico	and that code is really interpreter and JIT friendly
13:36:59	Yardanico	because interpreters are happier then you don't have any whitespace ;)
13:37:04	Yardanico	if the language isn't whitespace-significant
13:37:10	Yardanico	*when
13:38:02	Yardanico	https://i.imgur.com/MQak4rF.png
13:38:25	FromDiscord	<Rika> interpreted languages really benefit from non-whitespace-significant languages
13:40:43	*	zacharycarter quit (Ping timeout: 260 seconds)
13:41:58	Yardanico	tfw js "compressors" make the size bigger
13:42:52	FromDiscord	<Rika> lmao
13:49:33	*	Vladar quit (Quit: Leaving)
13:52:11	*	tane joined #nim
13:58:25	FromDiscord	<Doongjohn> ```nim
13:58:26	FromDiscord	<Doongjohn> block:
13:58:26	FromDiscord	<Doongjohn> let textPath = "./sometext.txt"
13:58:26	FromDiscord	<Doongjohn> let textFile = open(textPath, FileMode.fmReadWrite)
13:58:26	FromDiscord	<Doongjohn> defer: close textFile
13:58:27	FromDiscord	<Doongjohn>
13:58:27	FromDiscord	<Doongjohn> write textFile:
13:58:28	FromDiscord	<Doongjohn> """Lorem ipsum dolor sit amet,
13:58:30	FromDiscord	<Doongjohn> consectetur adipiscing elit.
13:58:32	FromDiscord	<Doongjohn> Morbi lorem eros, sagittis sed laoreet vel,
13:58:33	FromDiscord	<Doongjohn> bibendum vitae dui."""
13:58:35	FromDiscord	<Doongjohn> textFile.flushFile()
13:58:36	FromDiscord	<Doongjohn>
13:58:38	FromDiscord	<Doongjohn> echo textPath.readFile() # prints properly.
13:58:40	FromDiscord	<Doongjohn> echo textFile.readAll() # prints nothing? What am I doing wrong?
13:58:41	FromDiscord	<Doongjohn> ```
13:58:42	FromDiscord	<Rika> oof
13:58:44	FromDiscord	<Rika> rest in peace irc
13:58:55	FromDiscord	<Rika> @Doongjohn you arent seeking back to the end of the file
13:58:58	PMunch	Doongjohn, please don't paste code into Discord..
13:59:10	*	lritter joined #nim
13:59:14	FromDiscord	<Rika> before the echo, do `textFile.setPosition 0`
13:59:25	FromDiscord	<Doongjohn> oh ok thanks
13:59:40	PMunch	Yardanico, bot, when?
13:59:42	FromDiscord	<Rika> or was it setFilePos...
13:59:56	FromDiscord	<Rika> i dont remember, check the `io` module if you must
14:02:11	FromDiscord	<Doongjohn> should I use forum instead for pasting the code?
14:02:39	FromDiscord	<Rika> no, use a paste service
14:02:46	FromDiscord	<Rika> pastebin if you want
14:03:03	FromDiscord	<Doongjohn> ok sry
14:03:04	leorize	or the playground
14:07:06	PMunch	Or ix.io
14:07:29	FromDiscord	<Yardanico> @PMunch I can do it today just to see how things will go xdd
14:07:39	PMunch	Hooray!
14:07:52	FromDiscord	<Yardanico> The bot has two days uptime which is pretty good
14:07:53	PMunch	Finally the days of telling people to stop pasting code might come to an end!
14:08:04	PMunch	Yeah it seems solid in offtopic
14:08:28	Yardanico	yeah and I can always swap the older version easily
14:08:47	PMunch	As long as you're around :P
14:08:55	PMunch	Don't swap the bots and then go on holiday or something
14:08:57	Yardanico	well dom has access to that VPS as well :)
14:09:07	*	ertp07 joined #nim
14:09:07	Yardanico	mattebridge runs on the same droplet as NimBot
14:09:12	PMunch	(Not that I know where you would go on holiday..)
14:09:16	Yardanico	xd
14:09:46	Yardanico	well let's try to do it
14:10:48	Yardanico	it'll bridge 5 channels at the same time :P
14:10:55	Yardanico	ah sorry 4
14:11:16	shashlick	@zacharycarter - nimterop supports forward declarations
14:11:35	shashlick	But you need to use ast2
14:13:51	PMunch	Is there a limit to how many arguments a program can take?
14:14:18	Yardanico	yeah
14:14:34	Yardanico	https://serverfault.com/questions/163371/linux-command-line-character-limit
14:15:00	shashlick	@zacharycarter https://github.com/nimterop/nimterop/issues/148
14:15:01	disbot	➥ Doesn't properly handle forward declarations
14:15:54	leorize	on windows the limit is < 32800 ansi or unicode chars
14:15:56	*	Hideki_ quit (Ping timeout: 272 seconds)
14:16:15	PMunch	leorize, this is for a window manager, so Windows isn't an issue
14:16:26	PMunch	Or, well, Windows is an issue, but not for this
14:16:48	PMunch	Okay so that might be enough..
14:16:50	Yardanico	so first I'll try to start the bridge for all channels except #nim on the nim vps
14:16:56	Yardanico	and then I uncomment #nim and restart :P
14:17:22	PMunch	Good idea
14:17:23	Yardanico	nim vps is really pretty old though
14:17:50	*	Hideki_ joined #nim
14:17:53	Yardanico	ubuntu 16.04
14:19:35	PMunch	Hmm, good idea or bad idea: Have a program called the "configurator" for my WM. It can be written in anything (reference implementation in Nim) and it reads a configuration file and converts it to calling programs with command line arguments. This way you can write programs that interface with the WM, but which doesn't care about the configuration language
14:20:09	PMunch	I guess it could also pass them along by environment variables..
14:21:13	*	filcuc quit (Ping timeout: 264 seconds)
14:22:11	PMunch	Problem though is for the program that is meant to handle keyboard shortcut stuff. Currently it can only send a custom XEvent so it just takes an atom name and a shortcut, something like `--toggleExpand "Mod4+e"`
14:22:55	PMunch	But what happens if I want to allow e.g. scripting I can't have an argument `--"exec something" "Mod4+e"`
14:23:25	PMunch	Maybe multiple `--bind "Mod4+e:exec something" or something like that..
14:24:04	*	dadada joined #nim
14:24:07	*	Hideki__ joined #nim
14:24:27	*	dadada is now known as Guest39393
14:24:50	*	PMunch quit (Quit: leaving)
14:27:45	*	Hideki_ quit (Ping timeout: 256 seconds)
14:39:12	*	filcuc joined #nim
14:42:20	Yardanico	~restarting the bridge~
14:42:21	disbot	no footnotes for `restarting`. 🙁
14:42:27	*	FromDiscord quit (Remote host closed the connection)
14:43:47	*	FromDiscord joined #nim
14:44:02	*	Guest39393 quit (Ping timeout: 260 seconds)
14:44:06	FromDiscord	<Yardanico> 123
14:44:08	Yardanico	123
14:44:27	FromDiscord	<Yardanico> ```nim↵echo "Hello world"↵echo "yes finally"↵```
14:44:38	FromDiscord	<Yardanico> asdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasf↵asdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasf↵asdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasf↵asdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdasfasdasdas
14:44:51	FromDiscord	<Yardanico> sent a long message, see http://ix.io/2gcc
14:45:27	FromDiscord	<Rika> interesting
14:45:33	FromDiscord	<Rika> testing in production 😛
14:45:39	*	Hideki_ joined #nim
14:45:55	Yardanico	wait I'll restart it to enable screen logging
14:45:58	*	FromDiscord quit (Remote host closed the connection)
14:46:24	Yardanico	hmm
14:46:25	*	FromDiscord joined #nim
14:47:03	Yardanico	ah sorry need to do it again
14:47:09	*	FromDiscord quit (Remote host closed the connection)
14:47:21	*	rockcavera joined #nim
14:47:29	*	FromDiscord joined #nim
14:47:55	Yardanico	well let's see how it works out
14:48:23	Yardanico	if something's bad I can always get the old bridges back up :P
14:48:42	*	Hideki__ quit (Ping timeout: 256 seconds)
14:48:45	Yardanico	!status
14:48:45	FromDiscord	Uptime - 1 minute, 28 seconds, 815 milliseconds, 148 microseconds, and 119 nanoseconds
14:49:45	Yardanico	@KrispPurg btw I used the latest devel is it stable? :P
14:49:47	Yardanico	and yes pinging works
14:49:57	Yardanico	or it doesn't hmm
14:49:59	Yardanico	@Rika 123
14:50:01	Yardanico	hmm
14:50:16	Yardanico	@KrispPurg ^
14:50:17	Yardanico	wat
14:50:45	FromDiscord	<exelotl> holy shit I've been waiting so long for this moment
14:50:50	Yardanico	@exelotl lmao
14:51:34	Yardanico	ah I see why
14:51:38	Yardanico	lemme do a hotfix (TM)
14:52:39	*	FromDiscord quit (Remote host closed the connection)
14:52:54	*	FromDiscord joined #nim
14:52:57	Yardanico	@KrispPurg ping
14:53:01	Yardanico	ok im stupid
14:55:09	FromDiscord	<Rika> LMAO
14:55:54	*	FromDiscord quit (Remote host closed the connection)
14:56:09	*	FromDiscord joined #nim
14:57:13	*	dadada__ joined #nim
14:57:47	*	FromDiscord quit (Remote host closed the connection)
14:58:02	*	FromDiscord joined #nim
14:58:07	*	FromDiscord quit (Remote host closed the connection)
14:58:23	*	FromDiscord joined #nim
14:58:24	*	FromDiscord quit (Remote host closed the connection)
14:59:28	*	FromDiscord joined #nim
15:00:51	*	FromDiscord quit (Remote host closed the connection)
15:01:07	*	FromDiscord joined #nim
15:01:17	FromGitter	<sealmove> Guys how do nimble package tags work? Can one choose from a certain set of tags? Or he can put any string as tag? When I try publihs using arbitary tags it fails.
15:01:46	*	exelotl quit (Ping timeout: 260 seconds)
15:02:36	leorize	ask dom96 lol
15:02:59	FromGitter	<sealmove> @dom96 ping :P
15:03:08	Yardanico	idk why that didn't work honestly
15:04:20	*	FromDiscord quit (Remote host closed the connection)
15:04:33	*	FromDiscord joined #nim
15:05:01	*	FromDiscord quit (Remote host closed the connection)
15:05:16	*	FromDiscord joined #nim
15:09:13	*	FromDiscord quit (Remote host closed the connection)
15:09:28	*	FromDiscord joined #nim
15:17:01	*	FromDiscord quit (Remote host closed the connection)
15:17:18	*	FromDiscord joined #nim
15:19:44	*	FromDiscord quit (Remote host closed the connection)
15:20:01	*	FromDiscord joined #nim
15:21:39	*	FromDiscord quit (Remote host closed the connection)
15:21:54	*	FromDiscord joined #nim
15:22:37	*	poohman_1 joined #nim
15:25:11	*	FromDiscord quit (Remote host closed the connection)
15:25:28	*	FromDiscord joined #nim
15:28:40	*	FromDiscord quit (Remote host closed the connection)
15:28:56	*	FromDiscord joined #nim
15:29:57	*	FromDiscord quit (Remote host closed the connection)
15:30:16	*	FromDiscord joined #nim
15:31:12	*	ftsf quit (Ping timeout: 256 seconds)
15:31:36	*	FromDiscord quit (Remote host closed the connection)
15:33:13	*	Romanson joined #nim
15:34:01	*	FromDiscord joined #nim
15:34:29	*	FromDiscord quit (Remote host closed the connection)
15:35:59	*	FromDiscord joined #nim
15:37:11	*	FromDiscord quit (Remote host closed the connection)
15:37:19	*	FromDiscord joined #nim
15:38:20	*	FromDiscord quit (Remote host closed the connection)
15:38:28	*	FromDiscord joined #nim
15:39:45	*	Hideki_ quit (Remote host closed the connection)
15:41:32	FromDiscord	<Recruit_main707> ircord in general??
15:42:52	*	filcuc quit (Ping timeout: 260 seconds)
15:43:06	FromDiscord	<Rika> yes
15:43:12	Yardanico	ill revert for now
15:43:16	Yardanico	coz the stupid thing doesn't get member list
15:43:17	*	FromDiscord quit (Remote host closed the connection)
15:43:34	*	FromDiscord joined #nim
15:43:34	FromDiscord	<Rika> try re-enabling the intent lol
15:43:34	FromDiscord	<Yardanico> I TRIED
15:43:34	FromDiscord	<Rika> in the discord dev settings
15:43:35	FromDiscord	<Rika> really
15:43:37	FromDiscord	<Yardanico> yes
15:43:37	FromDiscord	<Yardanico> really
15:43:38	FromDiscord	<Rika> mmmmmmmmmm
15:43:44	FromDiscord	<Rika> weird
15:43:46	FromDiscord	<Yardanico> I even created a new bot and enabled same intent
15:43:48	FromDiscord	<Rika> discord dumb
15:44:04	FromDiscord	<Rika> i declare discord dumn
15:44:06	FromDiscord	<Rika> dumb
15:44:08	Yardanico	i'll test on my pc for now to find why it doesn't work
15:44:10	Yardanico	#nim-test
15:44:11	Yardanico	#test
15:44:34	*	exelotl joined #nim
15:47:49	*	dadada__ quit (Ping timeout: 246 seconds)
15:52:48	FromDiscord	<exelotl> i cri
15:54:22	*	Hideki_ joined #nim
15:54:24	*	Hideki_ quit (Remote host closed the connection)
15:55:01	*	Hideki_ joined #nim
16:01:28	*	Hideki_ quit (Ping timeout: 246 seconds)
16:03:49	*	filcuc joined #nim
16:04:34	*	dddddd joined #nim
16:08:54	FromDiscord	<Rika> does nim have a "sanitize filename" proc that strips invalid chars from filenames
16:09:49	leorize	I don't think so
16:10:01	leorize	invalid chars differs from os to os anyway
16:10:23	FromDiscord	<Rika> same thing with dir.sep.s, yet we have `os`
16:10:44	FromDiscord	<Rika> nonetheless, that's fine
16:10:58	FromDiscord	<Rika> was just wondering if i could replace a handmade proc with a stdlib one
16:11:09	leorize	and invalid chars is context-dependent
16:11:18	leorize	we do have one to verify for invalid names iirc
16:11:30	leorize	none to strip them though, since that sounds, uh, terrible?
16:12:09	shashlick	Seeing different makefile behavior when shell = bash vs sh - any insights?
16:12:18	shashlick	Paths don't work in bash mode
16:12:25	FromDiscord	<Rika> ooh, if we have a verify one, that works for me i guess
16:13:02	leorize	https://nim-lang.org/docs/os.html#isValidFilename%2Cstring
16:24:13	*	ertp07 quit (Ping timeout: 246 seconds)
16:26:14	FromGitter	<Willyboar> !stream
16:28:34	*	ertp07 joined #nim
16:29:08	rockcavera	Look at this example: https://play.nim-lang.org/#ix=2lyR Is there any way to know which type `a.edata` is storing?
16:29:26	leorize	Willyboar: it's ~stream
16:29:52	FromGitter	<Willyboar> ~stream
16:29:53	disbot	stream: 11https://twitch.tv/disruptek (live video/audio) and mumble://uberalles.mumbl.io/ (live voice chat)
16:30:19	FromGitter	<Willyboar> thank you @leorize
16:30:24	leorize	rockcavera: use the `of` operator
16:30:38	leorize	ie. `a.edata` of `EDataString`
16:31:06	rockcavera	leorize case?
16:31:30	leorize	`if a.edata of EDataString` <- it's an operator
16:31:33	*	Hideki_ joined #nim
16:31:54	rockcavera	hummm
16:32:22	rockcavera	leorize Thanks, I didn't know that operator.
16:32:24	rockcavera	:D
16:33:15	rockcavera	the best way to access is by casting?
16:33:17	leorize	yea, people don't frequently use it, so it doesn't have too much exposure
16:33:24	leorize	use conversion
16:33:33	leorize	`EDataString(a.edata)`
16:33:48	rockcavera	thanks
16:33:49	rockcavera	;)
16:33:51	leorize	^ that's a typed conversion, if the type doesn't match, it will error out
16:34:18	Yardanico	~bridge down for a bit, trying to fix that issue so I can start ircord~
16:34:19	disbot	no footnotes for `bridge`. 🙁
16:34:22	*	FromDiscord quit (Remote host closed the connection)
16:34:24	rockcavera	you saved me. I was racking my brain about this.
16:34:38	*	FromDiscord joined #nim
16:34:46	*	Vladar joined #nim
16:34:55	Yardanico	@yardanico 123123123
16:35:02	Yardanico	@KrispPurg 123123213
16:35:03	*	ptdel joined #nim
16:35:06	Yardanico	ok good enuf
16:36:07	*	dadada joined #nim
16:36:31	*	dadada is now known as Guest62821
16:37:06	FromDiscord	<Yardanico> https://media.discordapp.net/attachments/371759389889003532/709081590784983180/2020-05-10-16-37-48_grim.png
16:37:12	Yardanico	ok that works as well
16:37:12	*	couven92 joined #nim
16:39:30	FromGitter	<bung87> which software using mumble:// protocol?
16:39:36	Yardanico	Mumble
16:39:53	Yardanico	It's an open source VOIP app
16:39:58	Yardanico	for all major platforms
16:40:06	*	FromDiscord quit (Remote host closed the connection)
16:40:10	Yardanico	wat
16:40:17	FromGitter	<bung87> ah like skype?
16:40:20	*	FromDiscord joined #nim
16:40:22	Yardanico	not really
16:40:26	leorize	like teamspeak
16:40:58	*	fredrikhr quit (Ping timeout: 272 seconds)
16:41:03	*	FromDiscord quit (Remote host closed the connection)
16:41:05	FromGitter	<bung87> thanks , I will give a try :)
16:41:19	*	FromDiscord joined #nim
16:42:25	FromDiscord	<Rika> this is so funky
16:42:33	FromDiscord	<Rika> now all your messages arent condensed as fuck
16:42:52	Yardanico	don't worry I started the old bridge again since this one is a bit unstable because of some code changes
16:43:14	*	filcuc quit (Ping timeout: 256 seconds)
16:43:37	leorize	how come your bridge broke the moment it goes in production? :P
16:43:53	Yardanico	leorize: because apparently discord decided to stop sending full member list to it
16:43:54	Yardanico	I mean it works
16:44:02	Yardanico	but the mention functionality wouldn't really work
16:49:58	leorize	gg discord
16:50:06	Yardanico	well it can be solved
16:50:09	Yardanico	not that hard
16:50:15	Yardanico	btw arc is magic
16:50:53	*	couven92 is now known as fredrikhr
16:51:18	Yardanico	I'm not sure if I'm doing this right but https://play.nim-lang.org/#ix=2lz0
16:51:28	Yardanico	idk how it manages to access textArea from another thread btw :D
16:51:53	Yardanico	run with arc/orc and then go to localhost:8080
16:52:05	Yardanico	that's a really good way to make GUIs for async apps and stuff
16:52:21	Yardanico	no need to limit your async loop in any way
16:52:45	Yardanico	and idk if the global dispatcher part is needed, but I added it just to be safe
16:52:53	Yardanico	to create the global dispatcher in the thread where async runs
16:53:41	leorize	yea, the default dispatcher can do threads but not well tested iirc
16:54:22	leorize	I guess it can access textArea because the underlying object is usually thread-safe (inherited from the native ui toolkit)
16:54:32	Yardanico	hmm I guess
16:54:54	*	poohman_1 quit (Ping timeout: 272 seconds)
16:54:59	Yardanico	otherwise we can just have two threads and use channels
16:55:05	Yardanico	to send stuff
16:55:13	Yardanico	ah right it wouldn't be that simple
16:55:18	dom96	anyone here used random.sample with the CDF set?
16:56:47	leorize	I thought the point of arc is to let us safely share memory between threads?
16:57:00	dom96	oh, I think I get it. You create a list of weights and then you call cumsum on them
16:57:01	leorize	and now it seems like that you gotta move the graph to another thread
16:57:08	Yardanico	well that too but if it's memory from a native UI toolkit which is not threadsafe all funny things would happen leorize
16:57:31	Yardanico	leorize: btw GTK doesn't seem to be threadsafe
16:57:31	Yardanico	lol
16:57:38	leorize	I was just ranting about how arc doesn't do much :/
16:57:42	Yardanico	ah no it is
16:57:44	Yardanico	"GLib is completely thread safe (all global data is automatically locked), but individual data structure instances are not automatically locked for performance reasons. So e.g. you must coordinate accesses to the same GHashTable from multiple threads."
16:57:47	Yardanico	leorize: it does
16:57:50	Yardanico	much better interop
16:58:00	Yardanico	ohh
16:58:05	Yardanico	"GTK+, however, is not thread safe. You should only use GTK+ and GDK from the thread gtk_init() and gtk_main() were called on. This is usually referred to as the “main thread”."
16:58:26	leorize	the heap is thread-safe
16:58:33	leorize	the objects are not, obviously
16:58:42	*	Guest62821 quit (Ping timeout: 256 seconds)
16:58:48	leorize	since you're controlling gtk from one thread only, you're fine
16:58:58	*	waleee-cl joined #nim
16:59:46	Yardanico	well nigui is a good lib :)
16:59:53	Yardanico	without it I wouldn't have found a big bug in arc
17:00:23	Yardanico	need to do some arc bug hunting again
17:00:30	leorize	I'm not sure how better does arc help with our multi threading story
17:00:48	leorize	at least now we can move memory around, but we still can't safely share them?
17:03:58	livcd	leorize you are alwasy so pessimistic :D
17:04:44	*	poohman_1 joined #nim
17:05:29	leorize[m]	:P
17:06:05	leorize[m]	anyhow, for anyone interested, here's the full coverage map of Nim: https://codecov.io/gh/alaviss/Nim/tree/8f63aa0ef01e14f795d5afd8d368f4999addf103
17:06:19	leorize[m]	this is probably not going to become a part of the CI anytime soon
17:06:28	Yardanico	leorize: I saw your branch, but I didn't figoure out how to do it myself without a lot of commits
17:06:37	Yardanico	can you explain it with bash commands/nim compiler arguments? :D
17:06:50	Yardanico	or if I want to do it for other nim programs
17:07:06	leorize[m]	look at `config/nim.cfg`
17:07:12	Yardanico	I did
17:07:16	leorize[m]	the `@if coverage` section
17:07:21	Yardanico	but for me lcov failed and generated empty nim.info
17:07:47	Yardanico	lemme try for my lib
17:07:52	leorize[m]	well you also need to control where your objects are gonna be generated too
17:08:27	leorize[m]	use `--nimcache` to relocate the nimcache to somewhere deterministic
17:09:57	Yardanico	ok what do I do after that to generate cool html thing like you did? :P
17:10:07	leorize[m]	the coverage system is usable, but I'd say that it's not that good yet
17:10:22	leorize[m]	the cool html thingy? you need a codecov account lol
17:10:30	leorize[m]	for now use lcov genhtml
17:10:33	leorize[m]	it's not pretty but it works
17:11:56	leorize[m]	s
17:11:57	leorize[m]	so after you ran your program against a bunch of inputs to get the coverage data, run:
17:13:00	leorize[m]	`lcov -c --exclude '*generated_not_to_break_here' --exclude '\?\?\?' --rc 'lcov_branch_coverage=1' -b /where/you/run/nim/c --no-external --no-compat-libtool -d /the/nimcache/location -o project.info`
17:13:29	leorize[m]	`lcov -z -d /nimcache/location` can be used to wipe out all prior data
17:13:41	leorize[m]	use that when you recompile your program
17:14:23	leorize[m]	for best coverage data, you'd need to generate two `info` files: one for the raw source-only data, one for the data generated at runtime
17:14:46	FromDiscord	<dom96> This is pretty cool actually: https://gist.github.com/dom96/370e3f82a4b818cbd20c183220028748
17:15:03	Yardanico	>cumsummed :P
17:15:10	Yardanico	cumulative summed I assume
17:15:40	*	Jesin quit (Ping timeout: 246 seconds)
17:15:56	leorize[m]	`lcov -c --initial --exclude '*generated_not_to_break_here' --exclude '\?\?\?' -b /where/you/run/nim/c --no-external -d /the/nimcache/location -o project-initial.info`
17:16:44	Yardanico	leorize: and I again get an empty .info file :(
17:16:51	Yardanico	with tons of
17:16:52	Yardanico	geninfo: WARNING: cannot find an entry for indexerrors.nim##10b59710a9ebf1c8521c41778e6f416a.gcov in .gcno file, skipping file!
17:17:15	Yardanico	with nimcache folder, excessiveStackTrace, linedir, and --coverage options for gcc (basically same as in nim.cfg in your branch)
17:17:38	leorize[m]	is your program available somewhere? I give it a go
17:17:42	leorize[m]	I can*
17:17:55	Yardanico	https://github.com/Yardanico/nim-mathexpr/ just run tests/tests.nim :P
17:18:45	leorize	why is your LICENSE and README.md executable btw :P
17:18:46	Yardanico	maybe I'm doing something wrong, idk
17:18:53	Yardanico	leorize: lol idk :D
17:18:54	leorize	even your .gitignore is executable
17:19:09	Yardanico	i don't know when I changed that
17:27:16	FromDiscord	<codic> Some places I've noticed large numbers writetn as `10_100_100` for example, is that for readability? Can I till usse `10100100`/
17:27:26	Yardanico	yes
17:27:28	Yardanico	for readability
17:27:37	Yardanico	leorize: ahhhhh
17:27:38	Yardanico	https://github.com/linux-test-project/lcov/issues/58
17:27:41	disbot	➥ lcov incompatible with GCC 9.x coverage data ; snippet at 12https://play.nim-lang.org/#ix=2lzc
17:27:42	Yardanico	might be due to this
17:27:44	Yardanico	I have GCC 9.3
17:27:52	leorize	I have gcc 10
17:27:55	Yardanico	lol
17:27:58	leorize	and I just made it work lol
17:28:09	Yardanico	what was the issue?
17:28:18	FromDiscord	<codic> ah
17:28:23	FromDiscord	<codic> also, how do I create a pragma?
17:28:33	Yardanico	depends on what you mean by that
17:28:38	FromDiscord	<codic> for example if I had
17:28:55	FromDiscord	<codic> proc test(){.xyz.}=echo "O"
17:29:03	Yardanico	https://nim-lang.org/docs/manual.html#userminusdefined-pragmas
17:29:11	FromDiscord	<codic> it'd do something from xyz before the execution
17:29:12	FromDiscord	<codic> ah
17:29:13	FromDiscord	<codic> thanks
17:29:17	FromDiscord	<Rika> userrminusdefined lmao
17:29:37	Yardanico	well it's just replacing - with minus
17:29:41	FromDiscord	<codic> wait what `They cannot be imported from a module.`
17:29:45	FromDiscord	<codic> welp
17:29:47	FromDiscord	<Rika> yeah, its just funny
17:29:51	Yardanico	@codic the pragma pragmas can't
17:30:04	Yardanico	custom pragmas (with macro or template) can
17:30:04	FromDiscord	<clyybber> Will be "fixed"
17:30:05	FromDiscord	<codic> oof :\
17:30:19	Yardanico	you can create a macro pragma which does your pragma pragma
17:30:20	Yardanico	big brain
17:30:21	FromDiscord	<codic> with a `template thing() {.pragma.}`?
17:30:34	FromDiscord	<Rika> "pragmas made with the pragma pragma" is such a confusing sentence
17:30:39	FromDiscord	<Rika> no need to add pragma there
17:30:44	Yardanico	not really
17:30:47	*	ertp07 quit (Ping timeout: 260 seconds)
17:32:00	FromDiscord	<codic> argh I just want it so I can do from another file `import library; proc thing(){.x.}=echo "a"` and something from the pragma `x` can be executed before and after thing()
17:32:23	*	krux02 quit (Remote host closed the connection)
17:32:24	Yardanico	not hard
17:32:40	FromDiscord	<codic> how would it work though/
17:32:45	Yardanico	just okay
17:32:57	FromDiscord	<codic> considering that you can't import pragmas
17:33:01	leorize[m]	Yardanico: here's your mathexpr.info: http://ix.io/2lze
17:33:02	Yardanico	???????????????????
17:33:06	FromDiscord	<Rika> it'd rewrite the function to be (before stuff); function stuff; (after stuff)
17:33:07	Yardanico	@codic you can
17:33:09	Yardanico	just not pragma pragmas
17:33:50	FromDiscord	<Rika> `{.pragma ---.}` pragmas are not importable; `macro pragmaname()` are
17:34:00	FromDiscord	<codic> Ohhhh
17:34:04	FromDiscord	<codic> hold on, testing something
17:34:05	Yardanico	try to import it from another file https://play.nim-lang.org/#ix=2lzg
17:34:12	Yardanico	leorize: so what was the problem?
17:34:17	leorize[m]	Yardanico: and here's the diff that do the magic: http://ix.io/2lzf
17:34:25	leorize[m]	nothing really :P
17:34:27	leorize[m]	it just works
17:34:44	*	solitudesf joined #nim
17:34:51	FromDiscord	<Rika> i'm thinking of rewriting memo into something more readable (maybe) for no reason
17:35:27	Yardanico	still doesn't work for me XDD leorize[m]
17:35:32	Yardanico	something wrong with my compiler or lcov
17:35:43	Yardanico	lcov: LCOV version 1.0
17:35:45	leorize	lol what version of gcov are you using?
17:35:49	Yardanico	ah gcov?
17:35:52	*	Hideki_ quit (Ping timeout: 256 seconds)
17:36:00	leorize	I'm on lcov 1.14
17:36:07	Yardanico	lol
17:36:08	leorize	what oddly outdated distro are you on?
17:36:13	FromDiscord	<codic> Wait what, nim play lang says `/usercode/in.nim(2, 13) Error: undeclared identifier: 'stmt'`
17:36:22	Yardanico	because that's from like 2015
17:36:30	FromDiscord	<clyybber> use typed and untyped
17:36:30	Yardanico	stmt/expr were before typed/untyped
17:36:31	FromDiscord	<clyybber> instead
17:36:39	FromDiscord	<codic> alright
17:36:40	Yardanico	leorize: not sure either
17:36:56	FromDiscord	<clyybber> Yardanico: I think we could get rid of the pragma pragma completely
17:36:57	FromDiscord	<codic> but if I have a t of type untyped, how can I call it from within a macro
17:37:07	Yardanico	@codic ??????
17:37:10	Yardanico	macros are not templates
17:37:14	Yardanico	macros operate on AST and output AST
17:37:27	FromDiscord	<codic> Oh lol
17:37:37	FromDiscord	<codic> Ohhhhhhhhhhhhh
17:37:40	FromDiscord	<codic> That makes much more sense
17:38:02	FromDiscord	<codic> or not
17:38:06	FromDiscord	<codic> https://play.nim-lang.org/#ix=2lzh
17:38:20	FromDiscord	<codic> that throws an error for some reason
17:38:23	Yardanico	return untyped
17:38:35	Yardanico	ah wait
17:38:37	Yardanico	I see
17:38:37	FromDiscord	<clyybber> No
17:38:42	FromDiscord	<clyybber> Simply don't call t
17:38:44	FromDiscord	<clyybber> t is code
17:38:45	Yardanico	@codic you're trying to call t
17:38:53	FromDiscord	<codic> but then where will it be called if I don't call it?
17:38:59	Yardanico	you don't "call" code
17:39:00	FromDiscord	<clyybber> yeah
17:39:01	FromDiscord	<Rika> t == `echo("hi")`
17:39:04	Yardanico	code just get inserted
17:39:05	FromDiscord	<clyybber> because you call it inside t
17:39:14	FromDiscord	<clyybber> t itself is just a list of statements
17:39:15	*	ertp07 joined #nim
17:39:16	FromDiscord	<Rika> t() == `echo("hi")()`
17:39:22	Yardanico	eyp
17:39:22	Yardanico	yep
17:40:05	FromDiscord	<codic> alright so assume i remove that. then I just get the output
17:40:08	FromDiscord	<codic> before
17:40:08	FromDiscord	<codic> after
17:40:08	Yardanico	@codic I'd suggest you to read manual a bit :P
17:40:18	FromDiscord	<codic> there's no hi in the output :(
17:40:18	FromDiscord	<codic> true
17:40:20	Yardanico	no you also get "hi"
17:40:30	FromDiscord	<codic> nope
17:40:32	Yardanico	yes
17:40:33	FromDiscord	<codic> https://play.nim-lang.org/#ix=2lzj
17:40:35	Yardanico	https://play.nim-lang.org/#ix=2lzk
17:40:41	Yardanico	@codic because you REMOVED t
17:40:42	Yardanico	t is CODE
17:40:49	Yardanico	if you don't insert code, it's not gonna be executed
17:40:52	Yardanico	templates are code substitution
17:40:56	FromDiscord	<codic> Ohhhhhhhhhhhhhhh I call it as `t` not `t()`
17:41:00	Yardanico	you don't call it
17:41:02	FromDiscord	<clyybber> its not called
17:41:03	Yardanico	you simply insert the code body
17:41:05	FromDiscord	<clyybber> its getting inserted
17:41:07	FromDiscord	<codic> It's inserted
17:41:09	FromDiscord	<clyybber> yeah
17:41:13	FromDiscord	<codic> Yeah
17:41:13	FromDiscord	<codic> it serves my purpose still :P
17:41:17	FromDiscord	<clyybber> :)
17:42:22	FromDiscord	<codic> i was hoping that would somehow double up as a pragma as well. but `proc thing() {.Lol.} = echo "O"` doesnt have the `O`. Oh well /shrug
17:42:32	FromDiscord	<codic> *reads the manual*
17:42:39	Yardanico	because you'll insert the proc itself
17:42:45	FromDiscord	<clyybber> because it inserts "proc thing() = echo "0"
17:43:05	FromDiscord	<codic> and never calls thing()
17:43:08	FromDiscord	<codic> i see
17:43:10	FromDiscord	<clyybber> yeah
17:43:25	FromDiscord	<clyybber> if you want to make that work, you have to do t() I think
17:43:30	Yardanico	leorize[m]: I'm still confused a lot, I install lcov as 1.14 but it shows version 1.0
17:43:31	Yardanico	gcov is 9.3
17:43:41	leorize	which lcov
17:43:49	FromDiscord	<codic> @clyybber nope didn't work :\
17:43:54	FromDiscord	<clyybber> let me try
17:44:00	FromDiscord	<codic> https://play.nim-lang.org/#ix=2lzm that's mine
17:44:05	Yardanico	I tried https://aur.archlinux.org/packages/lcov/ and https://github.com/void-linux/void-packages/blob/master/srcpkgs/lcov/template
17:44:14	Yardanico	(i'm on bedrock linux with void linux as main strata and arch additional)
17:44:16	*	krux02 joined #nim
17:44:25	FromDiscord	<codic> nooooo bedrock instability time
17:44:30	Yardanico	wat
17:44:46	FromDiscord	<Rika> wat
17:44:50	FromDiscord	<codic> having multiple package systems on one
17:44:56	FromDiscord	<codic> just... stop it :p
17:44:56	Yardanico	it works nice
17:45:06	FromDiscord	<codic> ~~until it breaks~~
17:45:10	Yardanico	except it doesn't
17:45:17	FromDiscord	<codic> ¯\_(ツ)_/¯
17:45:21	FromDiscord	<codic> maybe for you
17:45:21	leorize	it's breaking afaict :P
17:45:28	leorize	which lcov <- run that
17:45:36	Yardanico	/usr/bin/lcov
17:45:37	leorize	see what lcov is the primary
17:45:48	leorize	maybe do a readlink on it too
17:45:50	Yardanico	that's from void
17:45:53	FromDiscord	<codic> What is lcov?
17:46:05	Yardanico	readlink /usr/bin/lcov doesn't show anything
17:46:32	*	Trustable joined #nim
17:46:33	FromDiscord	<clyybber> @codic Oh, yeah, sorry that won't work. You need a macro here
17:46:58	FromDiscord	<codic> ah. trying that
17:47:03	leorize	Yardanico: well maybe try to disable void one?
17:47:09	Yardanico	I tried
17:47:18	Yardanico	lemme try building from github
17:47:29	FromDiscord	<codic> didn't exactly work
17:47:31	FromDiscord	<codic> https://play.nim-lang.org/#ix=2lzo
17:47:39	Yardanico	thats
17:47:39	Yardanico	not
17:47:41	Yardanico	how you use
17:47:42	Yardanico	macros
17:47:42	FromDiscord	<codic> Yardanico: why not use the aur one? since you have arch access anyways
17:47:49	Yardanico	@codic it doesn't work
17:47:49	FromDiscord	<codic> :|
17:47:51	FromDiscord	<codic> oh
17:47:52	Yardanico	the one from github worked
17:47:54	Yardanico	directly
17:47:56	FromDiscord	<codic> strange
17:48:18	FromDiscord	<codic> aur gets the release version from github anyways
17:48:19	FromDiscord	<codic> https://aur.archlinux.org/cgit/aur.git/tree/PKGBUILD?h=lcov
17:48:25	Yardanico	still
17:48:32	Yardanico	and I didn't do latest release
17:48:34	FromDiscord	<codic> Aaaanyways why is that not how you use macros?
17:48:34	FromDiscord	<codic> oh
17:48:43	Yardanico	@codic because they operate on AST
17:48:46	FromDiscord	<clyybber> @codic There you go: https://play.nim-lang.org/#ix=2lzp
17:49:32	*	luis_ joined #nim
17:49:40	FromDiscord	<codic> Ohhhhhh
17:49:40	FromDiscord	<codic> the backticks make sense but why t[0]? is t a seq?
17:49:48	FromDiscord	<clyybber> its a NimNode
17:49:50	Yardanico	read the macros docs really
17:49:58	Yardanico	i'm sorry but explaining every single thing is really not okay
17:50:05	FromDiscord	<clyybber> and the first child of a proc NimNode is its name
17:50:05	Yardanico	read manual and macros docs a bit
17:50:07	Yardanico	read tut3
17:50:16	Yardanico	(https://nim-lang.org/docs/tut3.html)
17:50:37	FromDiscord	<codic> alright thank you!
17:51:31	*	aeverr joined #nim
17:51:32	*	aeverr is now known as aEverr
17:51:41	Yardanico	ip leaked
17:51:54	leorize	bedrock is a weird distro
17:51:56	aEverr	hello, i closed discord because discord's very cpu intensive
17:52:05	FromDiscord	<codic> "ip leaked" wait whattt
17:52:07	Yardanico	leorize: I don't think it's related to bedrock really
17:52:09	aEverr	(i'm Rika if you still don't know)
17:52:15	Yardanico	i do
17:52:21	leorize	aEverr: you can use the name Rika here too
17:52:22	FromDiscord	<codic> is it?
17:52:27	FromDiscord	<codic> lemme see how much cpu it's taking for me
17:52:29	leorize	or does someone already take it lol
17:52:33	aEverr	i'm signed in already
17:52:50	leorize	/nick Rika
17:52:51	leorize	lol
17:52:54	aEverr	also "ip leaked"????
17:53:02	FromDiscord	<codic>
17:53:02	FromDiscord	<codic> https://cdn.discordapp.com/attachments/371759389889003532/709100699098677309/unknown.png
17:53:11	Yardanico	aEverr: yes of course
17:53:14	Yardanico	everyone can see your IP
17:53:19	Yardanico	if you're connected directly
17:53:20	leorize	it's not like leaking your ip is a problem anyway
17:53:24	Yardanico	yeah ik
17:53:34	*	poohman_1 quit (Ping timeout: 240 seconds)
17:53:40	aEverr	i was questioning why you exclaimed that
17:53:47	Yardanico	leorize: made it work finally :D
17:53:59	luis_	Hi everyone, where do I set --arc on a nimble file?
17:54:03	aEverr	hm, what clients do you use, i dont like kde's default client
17:54:09	Yardanico	luis_: can you show your .nimble file?
17:54:11	aEverr	--gc:arc
17:54:11	Yardanico	aEverr: Quassel
17:54:14	leorize	I use weechat
17:54:17	FromDiscord	<codic> polari
17:54:21	Yardanico	quassel is not an IRC client really
17:54:29	FromDiscord	<codic> Or usually a matrix client :P
17:54:38	Yardanico	the client is quassel client, the server is irc client and quassel server at the same time
17:54:40	FromDiscord	<codic> since it interfaces with irc
17:54:40	FromDiscord	<codic> but when i use irc directly https://wiki.gnome.org/Apps/Polari
17:54:55	FromDiscord	<codic> i like kde's default but it's too cluttered
17:55:05	FromDiscord	<codic> so i remove it on all my installs (which are all kde)
17:55:14	luis_	https://i.imgur.com/McnjWXv.png
17:55:15	leorize	weechat is my matrix and irc client :P
17:55:20	luis_	here is my nimble file
17:55:30	leorize	luis_: you don't set it in nimble
17:55:52	aEverr	weechat aint gui so eeeehhhh
17:56:10	leorize	luis_: make a file src/allocator.nim.cfg
17:56:17	FromDiscord	<codic> weechat has matrix?
17:56:18	luis_	amazing
17:56:20	luis_	thanks
17:56:22	leorize	add gc:arc there
17:56:32	FromDiscord	<codic> eh polari rules
17:56:41	leorize	codic: yea, as a python plugin
17:56:49	FromDiscord	<codic> or gnome fractal (matrix and irc)
17:56:50	FromDiscord	<codic> Oh nice
17:57:01	leorize	it's more featureful than most other matrix clients btw
17:57:13	leorize	the only client to have stable E2E support that's not riot.im
17:57:17	aEverr	i dont see how konversation is cluttered
17:57:29	aEverr	i'll stay here in konversation
17:57:37	liblq-dev	aEverr: have you tried polari?
17:57:42	FromDiscord	<codic> idk, last i tried it was honestly the kde 4 days
17:57:53	FromDiscord	<codic> like the really late ones when plasma next was about to be released
17:57:59	*	PMunch joined #nim
17:58:06	FromDiscord	<codic> redownloading it
17:58:15	FromDiscord	<codic> alot of kde was cluttered in those days
17:58:15	aEverr	its gnome, i dont want to try i
17:58:28	liblq-dev	oh no not the K desktop environment
17:58:29	FromDiscord	<codic> honestly for some apps gnome does it better than kde
17:58:30	FromDiscord	<codic> just some apps
17:58:49	Yardanico	leorize: thanks a lot, coverage is really cool stuff :)
17:58:50	liblq-dev	and for visual design and layout it's just 10000000x better
17:58:57	liblq-dev	KDE is so damn cluttered
17:59:00	FromDiscord	<codic> No u
17:59:03	FromDiscord	<codic> It's opinionated
17:59:05	liblq-dev	it makes my brain hurt
17:59:06	Yardanico	we have #nim-offtopic FYI
17:59:15	aEverr	clutter is what i like i think
17:59:17	FromDiscord	<codic> gnome makes my brain hurt
17:59:18	Yardanico	and its bridged to discord too
17:59:27	FromDiscord	<codic> for how complex it is to tweak the simplest thing
17:59:31	aEverr	i just dont like gnome's ui
17:59:40	aEverr	also yeah kde has p. good customization
17:59:46	FromDiscord	<codic> ^
17:59:56	liblq-dev	and that's about all it has :)
18:00:06	liblq-dev	"pretty good" is not good enough imo
18:00:21	liblq-dev	but whatever it's opinionated as codic said
18:00:28	FromDiscord	<codic> > "pretty good" is not good enough imo
18:00:31	liblq-dev	just use what you like
18:00:35	FromDiscord	<codic> but you are kinda implying that gnome has better customization
18:00:40	liblq-dev	nope
18:00:42	aEverr	i dont see how its cluttered really, but yeah w/e
18:01:00	FromDiscord	<codic> that's the beauty of linux
18:01:00	liblq-dev	I only said that GNOME has better HIG
18:01:10	FromDiscord	<codic> to each their own
18:01:10	FromDiscord	<codic> I partially agree
18:01:14	Yardanico	guys really
18:01:17	Yardanico	#nim-offtopic
18:01:27	FromDiscord	<codic> why isn't that mentioned in discord #offtopic btw?
18:01:30	aEverr	hey i didnt start it :P
18:01:48	leorize	well it doesn't really matter, we can move there :P
18:02:01	aEverr	im not interested in participating in that anymore
18:02:43	FromDiscord	<codic> ¯\_(ツ)_/¯
18:02:45	leorize	Yardanico: it's cool until it starts showing you all sorts of false negative :P
18:02:54	FromDiscord	<codic> Connecting from konversation to see if it's decent
18:03:03	*	codic^ joined #nim
18:03:07	leorize	go terminal master race
18:03:11	codic^	this actually
18:03:13	codic^	is pretty nice
18:03:24	aEverr	what the heck are yall moving to offtopic or not
18:03:30	codic^	sure
18:03:43	codic^	noow how do I join a channel after connection in konversation
18:03:53	leorize	/join
18:03:54	aEverr	same way you do it in any client
18:04:00	codic^	o
18:04:13	codic^	"[11:04] [Channel] Cannot join channel (+r) - you need to be identified with services - see https://freenode.net/kb/answer/registration"
18:04:17	codic^	How come that's r+ but this isn't
18:04:28	codic^	anyways lemme connect to my irc acc
18:04:35	aEverr	maybe spam? dunno\
18:04:42	leorize	yea, was added due to spam
18:04:46	codic^	Oof
18:05:03	leorize	this one have to stay -r since beginners join here
18:05:12	codic^	:\
18:05:30	Yardanico	leorize: well just to get a general idea it works pretty well https://i.imgur.com/jEfIw02.png
18:05:35	Yardanico	from nimes :D
18:05:45	leorize	nice :D
18:05:48	Yardanico	mappers thing is true because I only played 1 rom which obviously has only one mapper
18:06:08	aEverr	? wrong channel?
18:06:16	codic^	yay I'm in
18:06:21	leorize	aEverr: ?
18:06:26	Yardanico	but yeah due to inlining and stuff sometimes lines are red
18:06:27	codic^	In #nim-offtopic
18:06:30	Yardanico	even though they were probably called
18:06:43	Yardanico	i wonder how does it work that fast
18:06:45	Yardanico	with coverage
18:06:45	aEverr	leorize: "mappers thing..." where'd this come from
18:06:47	FromDiscord	<codic> argh this is gonna take forever
18:06:51	Yardanico	aEverr: nimes
18:06:52	FromDiscord	<codic> to customize all the colors to dark ones
18:06:53	Yardanico	NES mappers
18:06:56	aEverr	OH
18:07:00	Yardanico	https://wiki.nesdev.com/w/index.php/Mapper
18:07:04	leorize	we are talking code coverage lol
18:07:07	aEverr	sorry, i lost context i think
18:07:15	FromDiscord	<codic> lol
18:07:19	leorize	Yardanico: lines are red because of hidden conditionals
18:07:34	Yardanico	yeah I'm probably stupid for compiling with -d:danger, lemme try without it
18:07:51	FromDiscord	<codic> d:danger ftw!
18:07:52	*	codic_ joined #nim
18:07:56	codic_	Test.
18:07:59	codic_	nice
18:08:03	leorize	the compiler preserve line info well enough that inlining is not even remotely an issue
18:08:08	Yardanico	oh okay
18:09:01	leorize	for this kind of coverage, the better the debug info is, the more precise the coverage data
18:09:04	Yardanico	it's amazing how fast it is
18:09:14	Yardanico	I mean with -d:danger there are no slowdowns
18:09:42	leorize	well I do it with the compiler and it only have a 1 min overhead
18:09:48	Yardanico	eqdestroy be like https://i.imgur.com/9Ce7usD.png
18:09:52	leorize	the largest overhead is in lcov :)
18:10:11	Yardanico	I guess it can't understand if functions are being called by function pointers
18:10:26	leorize	functions are always called by function pointers
18:10:30	Yardanico	ah wait it can
18:10:36	Yardanico	i was looking at the wrong file
18:11:06	Yardanico	yeah https://i.imgur.com/Jb5LIB4.png
18:11:28	Yardanico	cool to find out " 15 0 : of 0x8000..0xBFFF: result = m.cartridge.prg[m.prgBank1*0x4000 + int(adr - 0x8000)]" is never reached :P
18:11:31	Yardanico	for super mario bros
18:11:46	Yardanico	ah nvm it is
18:11:50	Yardanico	line info is a bit off
18:11:50	*	codic_ quit (Client Quit)
18:11:54	*	codic^ quit (Quit: Konversation terminated!)
18:11:58	Yardanico	there weren't any exceptions
18:12:17	*	Romanson quit (Quit: Connection closed for inactivity)
18:13:16	krux02	Yardanico, yea line info is always a bit off.
18:13:53	aEverr	does nim have a convention for documenting arguments and returns?
18:13:57	leorize	we should consider changing Nim's name mangling scheme
18:14:13	leorize	aEverr: name them correctly and you avoided 70% of the docs
18:14:28	leorize	and no, we don't have a convention for this
18:14:43	krux02	aEverr, nope
18:14:50	aEverr	i see, i dont have the issue currently and was just asking
18:15:08	krux02	I prefer to mention them in a full sentence in double backticks
18:15:12	Yardanico	me too
18:15:29	leorize	same here
18:15:59	leorize	is there any reason why Nim use context-dependant name mangling?
18:16:16	leorize	debugging tooling would be easier if it was a bit deterministic...
18:16:17	krux02	leorize, not a technical one
18:16:18	PMunch	What do you ean?
18:16:53	krux02	nim does not have name mangeling, it hash signature hashing
18:16:57	leorize	proc foo() -> foo_<md5 of the environment around the symbol>
18:17:15	krux02	the difference is, name mangeling is revertable, signature hashing is not.
18:17:53	krux02	leorize, the environment is probably because of hygienic templates.
18:18:32	leorize	true, traditional mangling scheme wouldn't work here
18:18:38	krux02	a proc is module a should not name clash with the same proc in module b
18:19:26	leorize	pascal have name mangling and modules and this has never been an issue
18:19:31	leorize	same goes for rust
18:20:30	leorize	I guess only Araq can answer this one
18:20:40	leorize	probably because it's complicated to implement
18:21:32	*	ertp07 quit (Ping timeout: 260 seconds)
18:22:48	*	ertp07 joined #nim
18:22:52	leorize	well to revert nim signature hashes we currently have to read .ndi, then parse .nim to get the full signature
18:22:59	leorize	not a flow that's tooling friendly
18:25:05	*	chemist69 quit (Ping timeout: 252 seconds)
18:26:11	*	chemist69 joined #nim
18:33:29	*	ertp07 quit (Quit: Quit)
18:43:02	*	krux02 quit (Remote host closed the connection)
18:43:07	*	exelotl quit (Ping timeout: 260 seconds)
18:50:10	*	luis_ quit (Ping timeout: 272 seconds)
18:58:30	*	poohman_1 joined #nim
19:00:03	*	Vladar quit (Quit: Leaving)
19:02:55	*	NimBot joined #nim
19:12:46	*	Jesin joined #nim
19:17:35	*	luis_ joined #nim
19:19:43	*	luis_ quit (Client Quit)
19:20:25	FromGitter	<awr1> nim having a canonical form of name mangling seems like something that would be worthy of consideration as nim edges closer to having a specification as opposed to being impl defined
19:26:33	disruptek	mangling will be changing.
19:27:19	disruptek	it will be a function of past compilation state per the cache.
19:28:02	disruptek	but, generally speaking, it will be as succinct as possible. first, just foo_2, foo_3, and foo. next, scope-based foo_N and foo.
19:29:33	FromGitter	<awr1> hm
19:30:15	FromGitter	<awr1> when nim functions are exported in a DLL via `exportc` without a specified name, what is the behavior of that? are they mangled?
19:30:21	*	iws quit (Quit: Lost terminal)
19:30:41	FromDiscord	<Yardanico> They have the same name as in Nim
19:30:52	FromDiscord	<Yardanico> I mean in the code itself
19:31:20	FromGitter	<awr1> so you can't have overloads on DLL exports because there is no mangling?
19:31:33	FromGitter	<awr1> (without changing exportc to have specific names)
19:32:38	FromGitter	<awr1> (or well, if the backend was C++, you could use its mangling i suppose)
19:33:16	*	Hideki_ joined #nim
19:34:24	*	aEverr quit (Quit: Konversation terminated!)
19:34:37	*	luis_ joined #nim
19:37:25	*	Hideki_ quit (Ping timeout: 246 seconds)
19:44:10	*	exelotl joined #nim
19:44:23	*	Jesin quit (Ping timeout: 260 seconds)
19:44:25	*	luis_ quit (Quit: luis_)
19:48:26	*	waleee-cl quit (Quit: Connection closed for inactivity)
19:50:12	*	dadada joined #nim
19:50:36	*	dadada is now known as Guest80851
19:55:16	*	zacharycarter joined #nim
19:58:36	*	waleee-cl joined #nim
20:06:02	*	abm joined #nim
20:09:01	zacharycarter	well if https://github.com/nim-lang/Nim/issues/14260 - is the case - maybe just remove discardable?
20:09:05	disbot	➥ {.discardable.} on object fields causes compile error ; snippet at 12https://play.nim-lang.org/#ix=2lAd
20:12:42	zacharycarter	seems pretty half baked if I can't use it in a type definition for a function pointer
20:12:43	*	poohman_1 quit (Ping timeout: 260 seconds)
20:12:51	FromDiscord	<Rika> maybe he likes it enough to keep it but not change it? dunno, esp. that clyy fixed it on his side...
20:14:03	zacharycarter	I think I like discardable better than littering the codebase with `discard`
20:14:43	zacharycarter	but if I can't use it in a type definition then it's really not working for me and I just need to put discard everywhere
20:17:24	disruptek	use a template.
20:17:45	zacharycarter	that's not the point
20:18:00	disruptek	well, i don't agree that it should be extended.
20:18:09	zacharycarter	if I have a bunch of function pointers in a type and I want to discard their result
20:18:30	zacharycarter	the typer checker will complain whether I implement the function with discardable or not
20:18:31	disruptek	nothing is preventing you from doing this.
20:18:37	zacharycarter	from doing what?
20:18:44	disruptek	discarding proc results.
20:19:01	FromGitter	<Willyboar> dis we missed you
20:19:03	zacharycarter	I know nothing is - the point is I wanted to avoid writing `discard` everywhere
20:19:10	disruptek	use a template.
20:19:19	zacharycarter	I agree nothing is preventing me, I'm being forced to
20:19:27	zacharycarter	meh
20:19:35	disruptek	you're being forced to be more explicit.
20:19:37	zacharycarter	jumping through mroe hoops
20:19:57	zacharycarter	I figured the discardable pragma was explicit enough
20:20:00	disruptek	willyboar: it's mothers day.
20:20:03	zacharycarter	but I guess not
20:20:33	disruptek	it's comfort for ffi, and a dubious one at that.
20:20:53	zacharycarter	meh
20:20:53	FromGitter	<Willyboar> dis yes i know, my kids allday drawing things for my wife
20:20:58	disruptek	ie. there's more to debate wrt including it at all, versus including it everywhere.
20:22:05	disruptek	mothers day is very important to me; i always double up on hookers and add a hefty tip.
20:22:10	disruptek	moms gotta work too, y'know.
20:22:58	disruptek	does tsoding's stream have something to teach us?
20:26:00	*	ptdel quit (Quit: https://quassel-irc.org - Chat comfortably. Anywhere.)
20:27:42	solitudesf	no
20:28:40	disruptek	sux.
20:29:37	disruptek	maybe we need more graphics.
20:42:07	*	exelotl quit (Ping timeout: 260 seconds)
20:44:27	disruptek	anyone have any xp with io_uring?
20:46:23	*	Jesin joined #nim
20:47:57	*	Jjp137 quit (Ping timeout: 260 seconds)
20:48:13	*	Jjp137 joined #nim
20:49:29	FromDiscord	<Rika> how experimental are the for loop macros?
20:54:23	FromGitter	<sealmove> are arrays initialized to 0's?
20:54:28	FromGitter	<sealmove> I know seqs are
20:55:04	FromDiscord	<Rika> yes
20:59:02	FromGitter	<sealmove> :O nice
21:03:37	disruptek	ah, bad docs.
21:04:05	liblq-dev	can I `|` a proc type with a concept type?
21:04:29	liblq-dev	my intuition tells me it *should* be possible
21:04:48	disruptek	here are three pipes: |||
21:04:54	disruptek	try it and if you need more, lemme know.
21:05:39	FromGitter	<sealmove> liblq-dev why do you want to try something like this? :D
21:06:06	liblq-dev	@sealmove it's… complicated
21:06:45	liblq-dev	I want to provide the flexibility of being able to use a generator-style callback or an object iterable with pairs() as a GL uniform source
21:14:06	FromDiscord	<Rika> is there a better less technical explanation for what bindSym does?
21:17:29	liblq-dev	it binds the symbol from the scope it's called in instead of the scope the macro is called in
21:20:47	*	exelotl joined #nim
21:24:51	*	Trustable quit (Remote host closed the connection)
21:25:44	*	solitudesf quit (Remote host closed the connection)
21:31:41	*	krux02 joined #nim
21:41:36	*	tane quit (Quit: Leaving)
21:43:37	*	arecaceae quit (Remote host closed the connection)
21:44:00	*	arecaceae joined #nim
21:49:37	*	liblq-dev quit (Quit: WeeChat 2.8)
21:55:05	FromDiscord	<Rika> can i check if a `NimNode` i constructed is valid?
22:01:34	FromGitter	<timotheecour> can anyone please merge https://github.com/nim-lang/Nim/pull/14293 to unblock nim CI ?
22:01:35	disbot	➥ diable nimx (CI failure)
22:15:24	krux02	Rika: what do you mean with "is valid"?
22:15:38	krux02	you mean valid syntax?
22:16:09	krux02	I would say, return it from a macro and the compiler will tell you if it's valild.
22:16:42	FromDiscord	<Rika> basically something like compiles() i guess?
22:17:00	FromDiscord	<Rika> so i can make an if else for a case i'm thinking of
22:17:05	disruptek	if you can get some goats in there somehow, that'll help.
22:17:20	krux02	Rika: no there is not.
22:17:39	krux02	you have to construct a valid ast.
22:17:57	krux02	`compiles` is already bad enough and should be avoided.
22:19:19	krux02	The reason is, you never know why something does not compile in `compiles`. Sometimes it is because a symbol isn't found. Sometimes it is because some obscure compiler bug, and sometimes it works even though it should not because of an unexpected compiler behavior.
22:19:49	krux02	if an ast is valid also depends heavily on the context.
22:20:19	FromDiscord	<Rika> i see, then i'm gonna need to think of another solutionm
22:30:44	*	Guest80851 is now known as dadada
22:30:55	dadada	hey my peeps
22:30:59	FromDiscord	<Rika> hi
22:31:04	dadada	hi Rika
22:31:10	krux02	hi
22:31:55	dadada	is there something eval like for nim macros? I want to get the type that is returned by a call for example, when I have something like var.field.call() ... getType doesn't work for this
22:32:27	krux02	dadada, there is a pattern that you can use, but it is a bit complicated.
22:33:38	krux02	you have to write two macros, an outer untyped frontend macro that eventually generats the calls like `var.field.call()` as an argument to an inner typed macro.
22:33:54	krux02	the inner typed macro will then be able to call getTypeInst.
22:34:12	FromDiscord	<Rika> how do i check if a proc can be used with a var as its sole argument? let's say, len(someVarWithSomeType) or something, how do i see if that'd work?
22:34:22	FromDiscord	<Rika> i'm stuck T_T
22:34:38	dadada	krux02: I'll do so :D thanks mate
22:34:42	krux02	btw getTypeInst/getTypeImpl is the substitution for `getType` which should have been declared as deprecated when `getTypeInst` and `getTypeImpl` were introduced.
22:34:53	krux02	np
22:36:36	krux02	Rika: I think you can call getTypeInst on the proc symbol and see if the signature matches. But I honestly don't think you should go that route. It is error prone and there is probably a better solution to you problem.
22:36:41	krux02	I don't know your problem though.
22:38:11	FromDiscord	<Rika> my problem is actually the example i gave, i'm trying to see if i can get the length of a variable, if any
22:38:21	FromDiscord	<Rika> if none, i'd default to something
22:39:16	krux02	well I would have a solution to that, but it doesn't work for `len` because of an implementation detail in system.nim
22:39:28	FromDiscord	<Rika> w-what
22:39:44	FromDiscord	<Rika> do you mind elaborating? or is it gonna fly over my head if you tried?
22:39:45	krux02	The solution that I already tried was this:
22:40:32	krux02	I declared ``template len(arg: untyped): int = -1``
22:40:56	krux02	now you have a length for everything and -1 is the fallback if no precise overload exists.
22:41:10	krux02	now calling `len` on anything will work, no exception.
22:41:12	krux02	The problem:
22:41:18	FromDiscord	<Rika> cant be used in macros?
22:41:43	FromDiscord	<Rika> or rather code generated in macros? 🤔
22:42:03	skrylar[m]	oh yeah. i saw the lineinfo stuff earlier lastnight. should probably use it with my logstuff
22:42:43	krux02	This is the declaration that breaks this: proc len*(x: (type array)|array): int {.magic: "LengthArray", noSideEffect.}
22:43:01	FromDiscord	<Rika> why would it break?
22:43:47	krux02	because the argument are two typeclasses combined, overload resolution will pick len(arg: untyped) over len(x: (type array)|array)
22:44:16	krux02	and therefor `len(myarray)` will return -1
22:44:36	krux02	The only fix is to change system.nim
22:44:48	krux02	you see, the solution isn't perfect.
22:44:48	skrylar[m]	(also earlier, re: gtk) y'know its weird, BeOS is the only OS i've seen where the GUI is fully multithreaded
22:45:15	FromDiscord	<Rika> minor noob question, what would (type array) resolve to? xd
22:46:44	krux02	(type array) is an alternative syntax (that I do not approve) for typedesc[array]
22:47:14	FromDiscord	<Rika> thats funky and confusing...
22:47:27	krux02	yea
22:47:39	FromDiscord	<Rika> why would two typeclasses be less "precise" than `untyped` itself?
22:47:47	FromDiscord	<Rika> i thought untyped was the least precise one can be
22:47:51	krux02	If you care about it. I am currently working on a compiler branch that tries to unify this mess.
22:48:07	krux02	A compiler that only accepts typedesc[array] and not (type array).
22:48:26	krux02	and yes, I thought the same about typedesc.
22:48:31	leorize	@timotheecour: with packages CI moved away from the main pipeline, we don't have to care about making sure it's always green when it's not our fault anymore
22:48:36	krux02	I mean I thought the same about untyped
22:49:32	krux02	that is in my opinion incorrect overload resolution.
22:49:37	FromDiscord	<Rika> are there proposals to change it?
22:49:46	krux02	no
22:50:58	krux02	you first have to convince Ar4q though that there is an issue, though.
22:51:10	krux02	twice though
22:51:40	krux02	then you can start on a proposal to fix it.
22:52:03	leorize	you can just write an RFC saying how you want things to be improved
22:52:30	krux02	or do that.
22:53:16	FromDiscord	<Rika> but i wonder if this is objectively wrong though... no idea
22:54:17	krux02	it is objectively annoying, if you want to make the `templat fallback(arg: untyped)` a real pattern of the language.
22:55:27	FromDiscord	<Rika> is there a case where it's *useful* though
22:55:37	krux02	and combined type classes are less specific than `untyped` witch doesn't even need to be a valid expression.
22:55:53	*	PMunch quit (Quit: leaving)
22:56:00	krux02	Rika: the pattern?
22:56:15	krux02	Yes the pattern would be useful for many things.
22:56:17	FromDiscord	<Rika> i mean, type classes being less precise
22:56:39	krux02	It could replace many use cases of `when compiles(foo(x)): foo(x)`
22:56:59	krux02	or better said, it could replace all use cases of that.
22:57:11	krux02	but only if it works.
22:59:00	FromDiscord	<Rika> is the change to system.nim easy?
22:59:14	FromDiscord	<Rika> easy to perform i mean
22:59:15	leorize	touching system.nim is never easy
22:59:34	leorize	well it's easy to change things in there, but getting them approved is an another story
22:59:38	*	zacharycarter quit (Quit: Lost terminal)
22:59:46	leorize	though usually you just need to make sure that all tests passes
22:59:59	krux02	splitting the proc up into two different procs is trivial.
23:00:15	krux02	and it would solve this very specific problem with len.
23:00:22	krux02	But it does't fix the general problem.
23:04:55	FromGitter	<timotheecour> @leorize that defeats (to some extent) the purpose of important_packages which is to ensure a PR doesn’t break nim packages; if it’s always broken, it makes it harder to tell whether your PR broke it or whether it was already broken.
23:05:55	FromGitter	<timotheecour> The flakyness of important_packages are largely fixable and I have a patch in the pipeline to address its main problem, but until it’s submitted, https://github.com/nim-lang/Nim/pull/14293 un-breaks it
23:05:57	disbot	➥ diable nimx (CI failure)
23:06:00	leorize	yes, but it's not urgent anymore
23:06:31	leorize	we are disabling a lot of packages and the majority of them don't get re-enabled
23:06:36	FromGitter	<timotheecour> sure, urgent is subjective.
23:07:06	leorize	at this rate disabling just seem to be "making the result look good" and less "improve ci quality"
23:07:37	FromGitter	<timotheecour> (disabled packages means a PR could genuinely break it but we won’t know until it’s re-enabled)
23:07:55	leorize	do we ever re enable things?
23:08:01	FromGitter	<timotheecour> oh ya, all the time
23:08:06	FromGitter	<timotheecour> i do, and others too
23:08:08	FromDiscord	<Varriount> timotheecour: I really liked your issue detailing the bugs/problems with the findExe & getAppFilename procedures.
23:09:33	krux02	timotheecour: I think that important packages should not automatically update
23:10:04	FromGitter	<timotheecour> what do you mean by update?
23:10:07	krux02	if they stay at a fixed version, their maintainers can't break randomly
23:10:27	krux02	important packages are always on the latest commit.
23:10:43	krux02	that is a problem because the latest commit does not always work.
23:10:57	krux02	and with 100 packages, the propability that something breaks is very high.
23:11:08	leorize	we also have the problem that people don't tag releases
23:11:11	FromGitter	<timotheecour> I’m aware, that’s exactly what https://github.com/timotheecour/Nim/issues/167 is about
23:11:14	disbot	➥ important_packages keep breaking because tests run on HEAD but `nimble install` installs latest release ; snippet at 12https://play.nim-lang.org/#ix=2lAI
23:11:33	FromGitter	<timotheecour> (and that’s what I’m fixing in my branch)
23:15:55	*	Hideki_ joined #nim
23:17:36	FromGitter	<timotheecour> @Varriount ya, it’s been bugging me for a while; I’ll submit a `which` RFC/PR so you can chime in there too
23:19:12	FromDiscord	<Rika> new issue, how do i check if the node isnt an iterator...
23:20:54	FromDiscord	<Varriount> Rika: Get the type implementation?
23:21:26	FromGitter	<timotheecour> `n.typekind == nnkIteratorTy` or similar
23:24:05	FromDiscord	<Rika> @Varriount `ForLoopStmt`s seem to be untyped :/
23:25:13	skrylar[m]	leorize: have been entertaining the idea that this is a good use for branches tho
23:27:24	FromGitter	<deech> Is there a compile time conditional that checks for `nimscript`? `when nimvm: ...` doesn't seem to work.
23:28:23	leorize[m]	when nimvm works, but it's special
23:28:23	leorize[m]	read the docs
23:28:37	leorize[m]	it's a weird quirk of nim
23:28:53	FromGitter	<timotheecour> `when defined(nimscript)`
23:29:18	FromGitter	<timotheecour> `when nimvm:` is always true inside nimscript
23:30:02	FromGitter	<timotheecour> but also inside VM for non nimscript code
23:30:53	*	abm quit (Read error: Connection reset by peer)
23:31:23	FromGitter	<deech> Ah `when defined(nimscript):...` is what I need!
23:39:47	*	Jesin quit (Quit: Leaving)
23:42:34	*	Jesin joined #nim
23:48:46	skrylar[m]	i'm probably the only one with this problem, but eh. https://github.com/nim-lang/Nim/issues/14295
23:48:49	disbot	➥ [1.2.0] --genDeps:on implicitly turns off compilation ; snippet at 12https://play.nim-lang.org/#ix=2lAO
23:58:14	*	ftsf joined #nim