<< 12-12-2019 >>
00:05:09FromGitter<Clyybber> test
00:05:16FromDiscord<Clyybber> ok
00:11:57*endragor joined #nim
00:13:37*abm quit (Quit: Leaving)
00:14:00*endragor quit (Remote host closed the connection)
00:14:37*endragor joined #nim
00:31:21*ng0 quit (Quit: Alexa, when is the end of world?)
00:31:53*endragor quit (Remote host closed the connection)
00:36:15*krux02 joined #nim
00:41:29FromGitter<Asc2011> hey, i'm looking for some detailed docs that describe the usage of nimsuggest - maybe the VS-Code-thinggy ?
00:48:26*krux02_ joined #nim
00:52:29*krux02 quit (Ping timeout: 265 seconds)
01:04:44*krux02_ quit (Remote host closed the connection)
01:33:34*njoseph quit (Ping timeout: 265 seconds)
01:43:00FromGitter<Asc2011> öok, another question: Is the idetools-integrations dead ? [https://nim-lang.org/docs/idetools.html]
01:45:10FromGitter<Asc2011> is Britney Spears around here ?
01:47:52rayman22201only Christina Aguilera XP
01:48:05rayman22201nimsuggest is idetools. They are the same
01:48:39rayman22201for VS code, you probably want to check out this: https://github.com/PMunch/nimlsp
01:49:33rayman22201or this: https://github.com/pragmagic/vscode-nim
01:51:30FromGitter<Asc2011> rayman22201, thx, i looked at the VS-Code implementation, but i don't use VS-Code. Somehow i 'm not a VS-Code user. My editor-support uses the older idetools-protokol and that does not work too good at the moment. And now i'm wondering : there seems to be a version-1-protokol for nimsuggest, which is probably what i need - but it's not documented..
01:52:31FromGitter<Asc2011> BTW i'm on atom, so its the atom-nim-pack - which is a nice codebase.
01:53:18FromGitter<Asc2011> what do you mean they are the same ? the outputs are the same ?
01:53:34*endragor joined #nim
01:55:52leorize@Asc2011 I'd glad to answer any question about nimsuggest
01:56:07leorizeAlso, if you understand vimscript, you can take a look at my plugin: https://github.com/alaviss/nim.nvim
01:57:43*endragor_ joined #nim
01:59:36FromGitter<Asc2011> @leorize : thx, i'm just wondering if i need to update the Atom-implementation (current idetools) to nimsuggest or if i can get away with some cheaper efforts ?-) U know Britney writes here [ https://nim-lang.org/docs/idetools.html ] in 2019 that the idetolls-stuff should work :-))
02:01:36*endragor quit (Ping timeout: 265 seconds)
02:05:21FromGitter<Asc2011> @FromIRC leorize, nice video :-)
02:10:48leorize@Asc2011 you'll have to update things to nimsuggest
02:10:51leorizeidetools is dead
02:14:05*njoseph joined #nim
02:18:13FromGitter<cndkhuong> idetools not found
02:20:44*njoseph quit (Ping timeout: 276 seconds)
02:24:49FromGitter<Asc2011> May somebody point me to the place where the old idetools-protocol is described. invocation-switches etc. that would make it a bit easier ;) thxx..
02:25:04*beatmox quit (Ping timeout: 268 seconds)
02:25:35*beatmox joined #nim
02:32:34*dddddd quit (Remote host closed the connection)
03:00:34leorizeAsc2011 just look for idetools in the compiler
03:00:47leorizeeverything is still there, just disabled
03:01:00leorizebut it's pretty much dead code at this point and no one tested it
03:01:07leorizeyou are better off with something that supports nimsuggest
03:01:21disruptekhe speaks from experience.
03:01:25leorizethere's a reason why nimsuggest was created: idetools was too slow it's unusable
03:03:02FromGitter<Asc2011> hmm, i'm working on it. nimsuggest was already in the package. So it's in the compiler-souce on GitHub, i'll take a look, thx..
03:09:34*ponyrider joined #nim
03:12:22*njoseph joined #nim
03:12:38*rockcavera quit (Remote host closed the connection)
03:26:07FromGitter<Asc2011> nimsuggest says : usage: sug|con|def|use|dus|chk|mod|highlight|outline|known file.nim[;dirtyfile.nim]:line:col ?? is there a example ?
03:26:40leorizehttps://nim-lang.org/docs/nimsuggest.html
03:26:54FromGitter<Obround> Hello! How do you do this type of *thing* in nim? (Its hard to describe): ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5df1b37ef1cc6c27ea344c54]
03:28:14leorizetype test = object
03:28:19leorize meth_val: int
03:29:10leorizeproc init_test(x, y: int): test = result = test(meth_val: result.meth(x, y))
03:29:14FromGitter<Asc2011> sry, but thats insufficient ;( ,
03:29:44leorizeObround: Nim don't have any equivalent to `__init__`
03:29:54leorizethe convention is to create an init proc for users to call
03:30:13leorize@Asc2011: well, just try it
03:30:45leorizeto get suggestions at a position in a file: sug <filename>.nim:<line>:<col>
03:30:54FromGitter<Obround> Yes, but can't you do something like this: proc initTest(x, y: int):`
03:32:50FromGitter<Asc2011> nope, "cannot find file" when i add the line/col-info
03:33:45leorize[m]Asc2011: nimsuggest spawns a shell
03:33:56leorize[m]you interact with it via that shell
03:34:14FromGitter<Asc2011> sure,,
03:35:54FromGitter<Asc2011> "nimsuggest sug test.nim:10:4" fails , "nimsuggest sug test.nim" works ??=))
03:39:56FromGitter<Asc2011> thats why examples are so00oo important - try it yourself - nimsuggest is always unhappy and answers with the usage-message, wtf
03:40:17*kevin joined #nim
03:40:41*kevin is now known as Guest58692
03:47:35leorize[m]do `nimsuggest test.nim`, then the shell will spawn
03:49:34FromGitter<Asc2011> hmm, the code-base i'm working on tries to get this in one-go. And that simply doesnt work... as advertised ...
04:05:04*liam_ joined #nim
04:10:19liam_sashlick, gogolxdong: So on the topic of the OpenSSL stuff from yesterday, I have written my own wrapper that does stuff with execProcess. It's not pretty, but it works! Obviously a proper wrapper would be faster and more secure, so I'm still looking forward to that.
04:10:32*Guest58692 quit (Ping timeout: 268 seconds)
04:12:08leorizeI thought there's an openssl wrapper already?
04:13:19liam_It only wraps the AES and SHA parts of libcrypto
04:19:13*Kaynatwo joined #nim
04:22:48*Kanato quit (Ping timeout: 245 seconds)
04:40:17*nsf joined #nim
04:52:39*narimiran joined #nim
04:55:13*liam_ quit (Ping timeout: 250 seconds)
04:56:52*chemist69 quit (Ping timeout: 252 seconds)
04:58:50*chemist69 joined #nim
04:58:54shashlickCan you share what you have
05:00:50*madprops gives shashlick a slice of pumpkin pie
05:07:43shashlickgorgeEx
06:08:42FromGitter<phillvancejr> Do we have a GUI library that will allow a window to be always on top of others and provide non native controls such that they can have custom styles
06:16:58*noonien quit (Quit: Connection closed for inactivity)
06:19:33AraqI think you need to use the windows api directly for "window always on top of others" but it sounds really annoying, don't do that
06:29:41*ltriant quit (Ping timeout: 250 seconds)
06:32:27TangerHmmm, FileStream.getPosition() is throwing me `index out of bounds, the container is empty [IndexError]` when it's on the very last element of the stream :S
06:33:03TangerIt's in a not stream.atEnd() loop, surely it wouldn't enter the loop if there's no more data to get?
06:33:27TangerOh, I'm an idiot, disregard that
06:50:14*solitudesf joined #nim
07:12:44*ng0 joined #nim
07:13:38*ganderzz joined #nim
07:27:39*PMunch joined #nim
07:34:43*ganderzz quit (Remote host closed the connection)
07:34:48*Vladar joined #nim
07:42:59*solitudesf quit (Remote host closed the connection)
07:43:13*solitudesf joined #nim
07:48:49*solitudesf quit (Ping timeout: 268 seconds)
07:50:17*ng0 quit (Quit: Alexa, when is the end of world?)
08:00:00*gmpreussner quit (Quit: kthxbye)
08:04:36*gmpreussner joined #nim
08:09:54*gour joined #nim
08:11:06Araqoh my, it's gour. welcome back
08:14:21gourAraq: congrat for your nim accomplishments!! i'm back seeing that nim-gui worlds is becoming richer and now i'm in position to really start working on desktop gui project...before that i resumed my studies of the subject and some family-related issues are over (loss of family member...)
08:16:13PMunchSorry to hear that gour
08:16:34PMunch(the loss of a family member, not that you're ready to work on your project)
08:18:17gourPMunch: well, my mother was old, but suffered from dementia and i dedicate myself to take care about her until her final day, so, i guess you can imagine, that in such condition i did not have almost *any* free time during those years...
08:18:29gour*dedicated
08:20:40PMunchOh for sure, that takes up a lot of time/energy
08:22:06gourotoh, i can say for sure that no language comes close to Nim when it comes to performance, elegant syntax, licensing, features, (probably) community as well
08:25:09PMunchI can't argue with you there. Now we just need the immense catalogue of libraries that Python has and we're set!
08:43:27gourhow is memory managed in wxnim?
08:47:13Araqif PMunch didn't change it, it's not managed at all, you translate to C++ and so C++'s regime is used (destructors etc)
08:47:29PMunchI didn't touch it
08:48:14gourAraq: it means one has to manually take care about it?
08:49:02Araqno, the framework does it for you
08:50:06gourahh, ok. that's the beauty of C++ vs C?
08:50:44Araqsomewhat
08:51:11Araqit's also the beauty of "memory management for UIs is not really all that hard to do"
08:52:03PMunchHmm, implement a PNG reader for the Nim VM, how hard can that be?
08:55:15PMunchUgh, why is the VM only allowed to cast between integers and/or floats of the same size..
08:55:34PMunchcast[seq[byte]](staticRead("somefile.png")) doesn't work..
08:57:51AraqPMunch: PRs are welcome, the VM got several updates recently
09:13:47*floppydh joined #nim
09:33:51*akitoshi joined #nim
09:42:04zedeuswhat's the arc alternative to deepCopy?
09:51:50AraqdeepCopy...
09:52:05AraqI need to port it
09:52:29zedeusah ok, wasn't sure
09:53:49Araqwell it's to be decided but it looks like we'll get 100% interop including support for shallowCopy
09:54:26zedeusI tried arc with some of my advent of code solutions, for some it's 2.5x faster than boehm with ~40% less memory used, and ~10% faster than markAndSweep with 3x less memory used
09:54:33zedeusfuture is looking very bright
09:59:27Araqit just got slower thanks to cycle collection but I think I can optimize it away most of the time
10:00:02Araqwow these are great results, can you blog about them?
10:00:06Araq:-)
10:02:03madpropsbtw i love the nim logo
10:02:09zedeushah, I don't have a blog and they're pretty small programs so probably not too interesting
10:02:32Araqanyway, the point of arc is not even performance
10:02:57Araqit's much better interop with "I am doing MM my way, thanks"
10:03:08madpropshuh there's another nim org using the same colors in the logo
10:04:47Araqmratsim: are there intrinsics for STC and CMC?
10:09:57PMunchmadprops, which one?
10:11:26madpropshttp://i.imgur.com/TJ5iaOc.jpg
10:11:54PMunchWell, not really the same colours
10:12:01PMunchI don't think anyone will confuse the two :P
10:22:59*nc-x joined #nim
10:24:06nc-xAraq: https://irclogs.nim-lang.org/11-12-2019.html#11:02:47
10:25:03*nc-x quit (Remote host closed the connection)
10:25:46Araq> i.e. should I move the other procs too?
10:26:05AraqI think this means nimscript.nim is beyond repair and should be deprecated altogether
10:26:37Araqon the other hand, so far paramStr caused the most trouble so lets only fix the reported bug
10:44:48*lritter joined #nim
10:48:03FromGitter<zetashift> Shouldn't ARC be compared with the current deferred GC rather than boehm and mark and sweep?
10:54:13FromGitter<alehander92> Araq, what should `move` do in js: shouldn't it be just a id
10:54:14FromGitter<alehander92> op
10:54:18FromGitter<alehander92> (t -> t)
10:54:40FromGitter<alehander92> or should it directly be a compile error
11:01:44*abm joined #nim
11:02:55FromDiscord<Clyybber> Araq: What does STC and CMC mean in this context?
11:04:42FromDiscord<Clyybber> Carry flag manipulation?
11:09:59Araqalehander42: it should be a cheap assignment followed by reset() of the source
11:10:07Araqclyybber: yeah
11:11:54Araqzetashift: well it should be compared to the "fastest" GC we have, maybe if you ignore latency
11:14:08*solitudesf joined #nim
11:24:52FromDiscord<mratsim> it should be compared with the default one as well since 99.9% of the people are using it and would want to know if it's worth being the first debuggers or arc
11:28:14Araqyou will be among the first debuggers no matter what :P
11:28:36Araqthis is beta-quality or maybe alpha
11:28:55Araqroutinely we find terrible bugs
11:29:34Araqat least you can use valgrind with it though
11:31:25FromGitter<Asc2011> what exactly is the project-file or main-project-file and how can i make sure that nimsuggest inspects & sees into this very important-file ? Is there any docu that describes a typical project-layouts. I know there are some variations binary/hybrid/js of config-files. Brew has placed a nim-compiler-config-file deep-down in /usr/local/nim/lib. , which i have not touched. But as nimsuggest automagically uses this file,
11:31:25FromGitter... so it seems very important. The question comes from a code-base that i'm currently struggling with. This ide-integration-package tries all sorts of wiered things from searching / probing to guessing just to identify "the-project-folder". It even reads the nimble-config-file to figure this out ? Obviously this is n ... [https://gitter.im/nim-lang/Nim?at=5df2250d42784416abd3f20c]
11:34:11FromGitter<Araq> pass the directory of the file to nimsuggest and let it figure it out itself
11:39:15*ltriant joined #nim
11:39:40*ltriant quit (Client Quit)
11:40:29FromGitter<Asc2011> thats exactly what the code already does and by that it fails -?- nimsuggest quits when i try that in the shell ? nimsuggest --stdin --debug ./src/* !!with a Asterix!! works
11:41:38FromGitter<Asc2011> why has the author of the atom-package spend 500-loc to figure things out ?
11:42:16*ng0 joined #nim
11:48:15*rockcavera joined #nim
11:49:35FromGitter<Asc2011> @Araq thx, anyhow, that package is a mess and it parses-console-streams which never worked for me.
11:51:30FromDiscord<mratsim> VS code package is also so-so 😉
11:52:39*Hideki_ joined #nim
11:55:41AraqAsc2011: auto-detection of project files is rather new in nimsuggest
11:56:11FromGitter<Asc2011> @mratsim and does it in VS-land work on one file or does it inspect many source-files in "a-project" ? looking at import and maybe into the std-lib ?
11:57:18*lritter quit (Read error: Connection reset by peer)
11:57:40*lritter joined #nim
11:58:49Araqyou can configure it
11:59:42FromGitter<Asc2011> @Araq thats good news - so i can kick most of the old-code :-) would you recommend to have one demon per file or one demon for all files open in the editor ? Does nimsuggest follow imports or looks into the std-lib ?
12:00:08Araqavoid "onde demon per file" at all costs
12:00:21Araqit's not how nimsuggest was designed
12:01:46FromGitter<Asc2011> @Araq ahh, have i understood - you say 'better have one nimsuggest-demon for some open files' not a 1:1 solution ?
12:02:51Araqopen nimsuggest once, it's really like the compiler, you don't "compile" module dependencies, do you. the compiler figures it out
12:06:44FromDiscord<mratsim> unfortunately the VS code plugin spawns one daemon per file
12:06:57FromGitter<Asc2011> ;-)) right - understood,, how important is it to generate a temp-file and pass this along with the file from the text-buffer ? the atom-modul does that all the time - is the idea to pass a 'working/error-free'-source-file to nimsuggest ?
12:09:37Araqyes pass it temporary files if they were edited
12:12:31FromGitter<Asc2011> @mratsim maybe they chat with one another :-)
12:13:12FromDiscord<mratsim> More like they deide to hog your CPU, RAM or both
12:13:15FromDiscord<mratsim> decide*
12:14:43FromGitter<Asc2011> @mratsim this is the new cool - loaded with AI - maybe they observe your typing-pattern :-)
12:17:00*rockcavera quit (Read error: Connection reset by peer)
12:17:17*rockcavera joined #nim
12:17:17*rockcavera quit (Changing host)
12:17:17*rockcavera joined #nim
12:23:41*endragor_ quit (Remote host closed the connection)
12:24:09madpropsi think rust is a scheme to make me exercise my fingers
12:53:47PMunchCrap.. PNG uses the deflate algoritm, which I'm guessing no-one has implemented for the Nim VM yet?
13:06:18FromDiscord<mratsim> wouldn't nim png work in the VM?
13:07:13PMunchDoubt it
13:07:18PMunchHold on, let me try
13:08:15PMunchNope
13:08:24PMunchError: cannot 'importc' variable at compile time; fopen
13:08:39PMunchOr wait
13:09:52PMunchNope, even with decodePNG32 I get: Error: cannot call method parseChunk at compile time
13:10:59PMunchAh, because the VM doesn't support methods?
13:26:55PMunchOooh, but the nimz decompressor works
13:29:43shashlickPMunch why not build it into an exe and call with gorge instead
13:29:57PMunchWhere's the fun in that?
13:31:54shashlickVm isn't very fast comparatively
13:32:02*pbb quit (Ping timeout: 252 seconds)
13:32:33PMunchTrue, but these are very small images
13:33:14shashlickOf course depends what your requirement is but I have had to be careful with impacting compile time with nimterop
13:33:22*pbb joined #nim
13:35:46PMunchYeah I ran into that with protobuf as well
13:36:28PMunchThe parsing took so long that your editor would noticably hang if nimsuggest was running on the same thread
13:37:50FromGitter<alehander92> isn't there a simpler way to do all this
13:38:23FromDiscord<Rika> the algorithm im using for day 12 is making my computer's ram sweat
13:38:44FromGitter<alehander92> i feel like a background server which just scans for changing lines with signatures/declarations and only parsing them with language-specific parsers
13:39:04FromGitter<alehander92> should be like enough for most language usages
13:40:18FromGitter<alehander92> yeah, it doesnt count for some language-specific things like scoping/inheritance
13:40:29FromGitter<alehander92> but probably thats what editors did before lsp anyway
13:41:11FromGitter<alehander92> hm, and probably thats what my sublime is doing
13:57:25*rockcavera quit (Read error: Connection reset by peer)
13:57:42*rockcavera joined #nim
13:57:42*rockcavera quit (Changing host)
13:57:42*rockcavera joined #nim
14:02:34*krux02 joined #nim
14:04:56FromGitter<Asc2011> and how do you like the lsp-thing that evbr is talking about ?
14:07:32FromGitter<alehander92> who is evbr
14:07:42FromGitter<alehander92> ah everybody
14:08:13FromGitter<alehander92> i like it, i just feel that its hard to make it defect-free/fast in all cases when its part of compilers
14:09:56FromGitter<alehander92> afaik some people redesign compilers to act more like server loops these days but i still wonder if its worth it for 80% cases
14:11:43FromGitter<Asc2011> and the accuracy of lsp while developing - that should be by concept the best support one could get ?
14:16:26*endragor joined #nim
14:37:42*nsf quit (Quit: WeeChat 2.6)
14:41:02FromDiscord<Fern & Simula (They/Them)> i'm trying to use dot operators to create function calls. I've got the following code, but I always get an error message that the compiler can't instantiate `U`
14:41:05FromDiscord<Fern & Simula (They/Them)> template `.`[T,U](f: (proc(x:T): U), x: T): U = return f(x)
14:41:08*FromGitter quit (Remote host closed the connection)
14:41:26*FromGitter joined #nim
14:41:29FromDiscord<Fern & Simula (They/Them)> is the issue with what i'm writing, or is it just not do-able?
14:41:47PMunchHow are you calling that?
14:41:57FromDiscord<Fern & Simula (They/Them)> oh sorry, i'll just pastebin the whole thing, one sec
14:42:24FromDiscord<Fern & Simula (They/Them)> https://pastebin.com/znjY8XkQ
14:43:25PMunchAny particular reason you don't want to do it in the regular Nim way? https://play.nim-lang.org/#ix=24d7
14:44:05FromDiscord<Fern & Simula (They/Them)> yeah, just palying around with dotOperators
14:44:24FromDiscord<Fern & Simula (They/Them)> i'm most likely gonna use something similar to this for a project in the near future so i wanna figure it all out now
14:44:53PMunchWell, the problem is that Nim isn't able to figure out what U is supposed to be from your call
14:45:34FromDiscord<Fern & Simula (They/Them)> well yeah. i figured it would be pretty simple to figure it out, but i guess not
14:46:11PMunchAs you can see you can call it if you specify the parameters: https://play.nim-lang.org/#ix=24da
14:46:18PMunchAlthough that's not very pretty :P
14:47:45FromDiscord<Fern & Simula (They/Them)> huh, interesting
14:47:55FromDiscord<Fern & Simula (They/Them)> i can also do it with untyped parameters. thanks though, that's very handy!
14:48:11PMunchYeah you can do it like this: https://play.nim-lang.org/#ix=24db
14:48:40FromDiscord<Fern & Simula (They/Them)> oh wow, i can spcify f as just proc? that's even better than what i had lol
14:48:42FromDiscord<Fern & Simula (They/Them)> thanks!
14:48:59FromDiscord<Fern & Simula (They/Them)> ~~hacking a dynamic lang with a static lang is really weird~~
14:49:05PMunchYup, but that doesn't stop you from passing only one argument..
14:49:21PMunchErr, passing a procedure that takes more than one argument*
14:50:18FromDiscord<Fern & Simula (They/Them)> it's not client-facing code, it's probably fine lol
14:50:27FromDiscord<Fern & Simula (They/Them)> err, user-facing
14:50:39FromDiscord<Fern & Simula (They/Them)> i could also use varargs
14:54:22PMunchYou could also use a concept: https://play.nim-lang.org/#ix=24de
14:55:38PMunchI mean you would still just get an error when using the proc thing from earlier, since it expands and then realises that it can't actually compile that
14:59:52*endragor quit (Remote host closed the connection)
15:00:13PMunchOr if you want to create your own custom error: https://play.nim-lang.org/#ix=24dg
15:00:40*ng0_ joined #nim
15:01:11*ng0 quit (Disconnected by services)
15:01:14*ng0_ is now known as ng0
15:05:20FromGitter<alehander92> i guess proc is a typeclass?
15:05:27FromGitter<alehander92> so i wouldnt say this is dynamic
15:05:41FromGitter<alehander92> but i am not sure about the proc thing
15:05:57FromGitter<alehander92> the proc {.} things bited me some times
15:07:37*NimBot joined #nim
15:11:43*MarquisdeFalbala joined #nim
15:15:41*Hideki_ quit (Remote host closed the connection)
15:16:33*Hideki_ joined #nim
15:17:10*endragor joined #nim
15:23:59*Hideki_ quit (Ping timeout: 276 seconds)
15:24:48*endragor quit (Remote host closed the connection)
15:28:41*abm quit (Quit: Leaving)
15:33:29FromGitter<Asc2011> i'm gonna try the nimsuggest-socket version from node/atom - can anybody recommend a socket-lib that does the job best ? or just any plain and simple unix-domain-socket lib ?
15:35:08*dddddd joined #nim
15:38:32leorizenimsuggest communicates via tcp sockets
15:38:47leorizeto be fair it'll be faster if you just use nimlsp
15:39:09FromGitter<Asc2011> ups, tcp not upd
15:39:55leorizenimlsp does all the logic needed to properly use nimsuggest
15:40:00FromGitter<Asc2011> and why now nimlsp again ? it's compülicated
15:40:28leorizebecause nimlsp supports LSP, which should be available on editors like atom
15:41:07leorizeconfiguring an lsp client is much much easier than trying to write a full-fledged nimsuggest client
15:41:45FromGitter<Asc2011> well, maybe yes, there are things floating around, but i've not tested them yet,, so i count three construction sites on my side plus LSP makes four :-)
15:43:14FromGitter<Asc2011> i have a idetools/nimsuggest-client, ok a bit outdated, on my hands. I have no idea what LSP is for a beast :-)
15:44:20FromGitter<Asc2011> i bet you think this one https://github.com/PMunch/nimlsp. ?
15:44:33leorizehttps://github.com/atom/atom-languageclient
15:44:35leorizeyep
15:44:41leorizeconfiguring is rather easy ^
15:45:06leorizeno need to manually parse nimsuggest's output, no need to write project finding code
15:45:13leorizeit's a major upside, really
15:45:56FromGitter<Asc2011> the issue-36 says "nimsuggest crashes for specific file in projects" - what is going on here :--)
15:46:36leorizenimsuggest crashing is super normal by now :P
15:47:52FromGitter<Asc2011> hehe, well right-said-fred , but why is nimsuggest in the lsp-project in-the-game at all ?
15:49:46leorizenimlsp is basically strapping lsp support to nimsuggest
15:49:47FromGitter<Asc2011> is one complementing the other ? how does the info flow operate between IDE<->nimsuggest<->LSP ?
15:49:57leorizeIDE <-> nimlsp
15:50:10leorizenimlsp uses nimsuggest as a library
15:51:24*Hideki_ joined #nim
15:51:36FromGitter<Asc2011> so from the IDE-side of things, i'd configure the LSP-client which talks to nimlsp<- talks-to->nimsuggest ?
15:53:52leorizeya
15:54:02leorizeyou only have to care about the talk to nimlsp part
15:54:11leorizethe rest is automated and non-configurable anyway
15:55:17FromGitter<Asc2011> i'll think about it .... greets
16:01:01*floppydh quit (Quit: WeeChat 2.6)
16:06:23PMunchNot quite
16:06:32PMunchnimlsp doesn't talk to nimsuggest
16:06:53PMunchnimsuggest can be compiled into its own program, but it can also be imported as a library
16:07:20PMunchSo nimlsp imports nimsuggest and so it's just calling directly into the nimsuggest procedures
16:07:38PMunchInstead of communicating with a separate instance of nimsuggest
16:12:59FromGitter<Asc2011> aha, so protocol-wise its JSON-RPC to-and-from nimlsp<-JRPC->atom-lsp-client while nimlsp spares us anoother socket and loads nimsuggest as a lib :-) sounds interesting and here https://langserver.org is said that neither Nim nor Coffeescript are on the list, wheras COBOL & PHP enjoy full-support :) it's unfair
16:13:08*PMunch quit (Quit: Leaving)
16:19:32FromDiscord<mratsim> always assume that Nim is never on the list or on anyone Radar and you would be pleasantly surprised 😉
16:20:58*Hideki_ quit (Ping timeout: 265 seconds)
16:25:56FromGitter<Asc2011> hihi, i get the idea - keeping a low-profile - @PMunch looking at this list https://github.com/atom/ide-csharp/blob/master/package.json of lsp-services - which of these does nimlsp support at the moment ?
16:27:45FromGitter<Asc2011> and how can one properly address another person which is behind a chat-service-bridge ?=((
16:29:51ZevvI just call them all FromGitter
16:30:09ZevvFromGitter: just like you
16:30:10FromGitterZevv, I'm a bot, *bleep, bloop*. I relay messages between here and https://gitter.im/nim-lang/Nim
16:30:21Zevvha
16:30:51FromGitter<Asc2011> ;-)
16:34:21FromGitter<Asc2011> found the info on nimlsp-services
16:38:43leorize@Asc2011: well just type out the name in the brackets <> manually :p
16:40:36FromGitter<Asc2011> <PMunch> like so ? nimlsp doesn't compile :-) lets have a compile-party,,
16:40:52leorizenah, just the name :P
16:41:02leorizePMunch: <- like this
16:42:05leorizeanyhow, he went offline :p
16:42:33FromGitter<Asc2011> ic PMunch ? again ? leorize: :-=
16:55:50*MarquisdeFalbala quit (Ping timeout: 246 seconds)
17:06:06FromGitter<Varriount> @Araq So, what it the difference between the current GC and ARC?
17:06:26FromDiscord<Clyybber> Varriount: ARC is deterministic
17:06:27disruptekmore bugs in arc.
17:06:39FromDiscord<Clyybber> disruptek: I wouldn't be sooo sure
17:07:33FromDiscord<Clyybber> ok maybe more, but at least not this one: #12273
17:07:36disbothttps://github.com/nim-lang/Nim/issues/12273 -- 3Memory corruption ; snippet at 12https://play.nim-lang.org/#ix=24dJ
17:08:07disruptekyour cloud is gone.
17:08:36FromDiscord<Clyybber> ☁️
17:08:56FromDiscord<Clyybber> must have turned into dust
17:25:22AraqVarriount: ARC is reference counting with move semantics
17:26:21Araqsince the reference count values stay precise, you can do copy-on-write reliably
17:35:44FromGitter<phillvancejr> Does nim call default constructor's of C++ objects by default? I created a c++ class with a default constructor and an overloaded constructor and whenever I create on object of the class in Nim, it calls both constructors and calls the destructor twice. Does anyone have an Idea on what I'm doing wrong?
17:39:27FromDiscord<mratsim> Aren't constructors are always called in C++?
17:39:52FromDiscord<mratsim> how did you wrap the constructors?
17:40:22nisstyreSo, if I wanted to treat a string as a File object, the way to do that seems to be use StringStream, but what if I wanted to construct an actual file handle that I could give to a C library function expecting FILE* ? I looked into fmemopen for that, but it sounds like it's not portable
17:40:56FromDiscord<mratsim> here is an example of wrapping multiple constructors @Phillvancejr: https://github.com/status-im/nim-ttmath/blob/master/src/ttmath.nim#L84-L86
17:42:12FromDiscord<mratsim> @nisstyre: https://nim-lang.org/docs/io.html#stdout
17:43:07FromDiscord<mratsim> now I'm not sure how to transform a string in memory into a file handle
17:43:20nisstyremratsim: that's the part I'm wondering about mainly
17:44:13FromDiscord<mratsim> can you do that in C?
17:44:19nisstyremy use case for this is actually calling libssl functions related to x.509 that expect files
17:44:29nisstyremaybe there's a higher level library that allows access to these
17:44:45nisstyreyou can do it in C with fmemopen
17:44:55nisstyreI just wanted to know if there's a better way
17:45:00nisstyreit's not very portable
17:45:12nisstyrethere's also open_memstream, but it's a bit tricky
17:45:19nisstyreyou have to actually write to the file stream
17:45:20FromDiscord<mratsim> I see a bytes API here: https://www.openssl.org/docs/man1.0.2/man3/d2i_X509.html
17:45:36nisstyreah maybe that will work
17:45:39nisstyreI was reading something else
17:48:18nisstyrethe thing is, it's PEM encoded
17:48:32FromGitter<phillvancejr> @mratsim I used c2nim to generate the nim wrapping then I added importcpp pragmas
17:49:23nisstyrethese might work though
17:49:40FromDiscord<mratsim> @nisstyre, we use BearSSL for our SSL needs: https://github.com/status-im/nim-bearssl
17:50:12nisstyreok d2i_X509 might work for my uses
17:50:18nisstyremratsim: I'll look into that, thanks
17:50:24nisstyreI really just want to parse certificate chains
17:50:35nisstyreso I'm not tied to openssl
17:51:01FromDiscord<mratsim> @Phillvancejr mmm, it might need some further edit then. otherwise this might simplify things: https://github.com/fragcolor-xyz/fragments#cpp
17:54:39FromGitter<phillvancejr> @mratsim it seems like it should be simple. it is a very small class with only a default constructor and one overloaded to take a string. Even if I remove the wrapping for the default constructor and only use the overloaded version, Nim still calls the default constructor. It is very strange.
17:55:25FromDiscord<mratsim> do you have an example you can post?
17:58:10*Hideki_ joined #nim
18:02:35*Hideki_ quit (Ping timeout: 250 seconds)
18:04:43*rockcavera quit (Remote host closed the connection)
18:10:25*narimiran quit (Ping timeout: 268 seconds)
18:10:28FromGitter<phillvancejr> @mratsim ⏎ Person.h ⏎ ⏎ ```Bob says hello! ⏎ Phill says hello! ⏎ Phill says goodbye! ⏎ Phill says goodbye!``` [https://gitter.im/nim-lang/Nim?at=5df28294e72656230131be29]
18:12:01solitudesfanother bot https://forum.nim-lang.org/t/5675
18:12:04FromGitter<phillvancejr> Maybe it has to do with C++ move semantics? Maybe an object is being created then passed to the nim function, so there are actually two objects created?
18:13:04FromGitter<phillvancejr> I thought maybe it was the bycopy pragma, but changing it to byref didn't affect the output
18:16:27*dddddd quit (Ping timeout: 250 seconds)
18:19:08FromDiscord<mratsim> very strange bot @solitudesf, it actually replie thank you in the linked reddit thread
18:20:18FromDiscord<mratsim> mmm I never tried overloading classes, only basic struct and std::string or unique_ptr
18:21:17FromGitter<phillvancejr> Ah ok, thanks for taking a look. I'll post it in the forum so it doesn't get lost here. Thanks for your time!
18:21:39*Trustable joined #nim
18:22:02FromDiscord<mratsim> you're welcome
18:22:35FromDiscord<mratsim> if you ever need string/vector/unique pointers this is my code: https://github.com/numforge/agent-smith/blob/master/third_party/std_cpp.nim
18:24:59FromGitter<phillvancejr> oh nice, thanks!
18:25:11FromGitter<alehander92> hmm
18:27:23FromGitter<alehander92> ok @phillvancejr
18:27:39FromGitter<alehander92> i know what is happening
18:27:54FromGitter<alehander92> you create `object`
18:28:02FromGitter<alehander92> so in the resulting code its defined like Person p2__QSETErn3Fwo89a6F8hQtQKQ;
18:28:11FromGitter<alehander92> which i think calls
18:28:15FromGitter<alehander92> the default constructor
18:28:25FromGitter<alehander92> (this is a top level definition of the variable)
18:29:04FromGitter<alehander92> later its called again in the main
18:29:12FromGitter<alehander92> with the actual constructor call
18:29:47FromGitter<alehander92> this also happens only on top level it seems
18:30:40*dddddd joined #nim
18:32:50FromGitter<alehander92> `var p2: Person` also does this in any situation (which is probably expected
18:33:43FromGitter<alehander92> so its an interesting bug, and not sure how it can be solved easily
18:33:52FromGitter<alehander92> even more interesting is that if there is no default one
18:33:59FromGitter<alehander92> let p2 = constructPerson("Phill") is directly broken
18:34:06FromGitter<alehander92> var p2: Person as well
18:34:30FromGitter<alehander92> which is again probably obvious if one uses c++ not sure
18:38:09FromGitter<alehander92> however this seems to break
18:38:13FromGitter<alehander92> `default(Person)`
18:38:30FromGitter<alehander92> maybe thats fine
18:39:07FromGitter<alehander92> Araq ^ i guess you can clarify those, but the first one (top level let p = Person(..) calling twice seems as a small bug)
18:39:27FromGitter<alehander92> calling different constructors *
18:39:36FromGitter<phillvancejr> What is default(Person)? I haven't seen that yet
18:39:53FromGitter<phillvancejr> Also thank you Alexander for taking the time to look at this
18:40:05FromGitter<alehander92> well, there is this `default()` builtin that constructs a default value for a type
18:40:22FromGitter<alehander92> no problem, trying to learn more about c++ constructors myself
18:40:45FromGitter<phillvancejr> Ah ok, and it takes a type? But don't types in nim default to zero values? or are objects nil?
18:42:07FromGitter<alehander92> well yeah, but you still need an operation Type -> Value
18:42:42FromGitter<phillvancejr> The double constructor kind of makes sense if on the C++ side. If I was creating a Person by calling a function that takes a Person and returns a Person, I think this would result in double constructors/destructors because a temporary object is created, but then I would expect the Destructor to be called before the second constructor is called so I don't think that is what Nim is doing
18:42:43FromGitter<alehander92> ref objects i think are nil, objects should be just zero i guess
18:43:05FromGitter<alehander92> but this is not a double constructor bug
18:43:10FromGitter<alehander92> this is a different bug
18:43:16FromGitter<phillvancejr> I thought so
18:44:02FromGitter<alehander92> basically nim generates something like ⏎ ⏎ Person p; // top level if p was top level ⏎ ... ⏎ ... [https://gitter.im/nim-lang/Nim?at=5df28a7279534b3868787f8d]
18:44:11*nsf joined #nim
18:45:02FromGitter<alehander92> not really sure if there is an easy workaround tho
18:45:13FromGitter<alehander92> for the compiler
18:45:31FromGitter<alehander92> otherwise for the user, the obvious thing would be to just move this inside a function
18:49:47FromGitter<phillvancejr> move which part into the function? The construction of a Person is already inside of a function.
18:50:15FromGitter<alehander92> no
18:50:18FromGitter<alehander92> you call it on top level
18:52:42FromGitter<phillvancejr> how so? The only thing at top level is the type declaration right?
18:58:02FromGitter<alehander92> ok, i am talking about the example
18:58:08FromGitter<phillvancejr> Actually it looks like the function is the problem. When I create a person at the top level like this: ⏎ ⏎ ```let p = constructPerson()``` ⏎ ⏎ This actually results in the double constructors. So actually the function is the problem I think, it is creating two Person objects [https://gitter.im/nim-lang/Nim?at=5df28dc0ac14cc652c712b86]
19:05:44FromGitter<alehander92> yes
19:05:49FromGitter<alehander92> you're right!
19:06:02FromGitter<alehander92> but indeed
19:06:17FromGitter<alehander92> this is because it directly does p = {};
19:07:20*PMunch joined #nim
19:07:24FromGitter<alehander92> and when its a function
19:07:35FromGitter<alehander92> it doesnt do it really on top level, but in the NimMainModule
19:07:50FromGitter<alehander92> if you open
19:07:53FromGitter<alehander92> the nimcache cpp file
19:07:57FromGitter<alehander92> youll immediately see
19:08:01FromGitter<alehander92> what i am talking about
19:08:05FromGitter<alehander92> no need for guesses
19:14:01FromGitter<bung87> ```code paste, see link``` ⏎ ⏎ the r_load is nil, what’s wrong with the code [https://gitter.im/nim-lang/Nim?at=5df2917955d9392300fca512]
19:14:52FromGitter<bung87> the go version https://github.com/shirou/gopsutil/blob/master/cpu/cpu_darwin_cgo.go#L86
19:16:17PMunchHmm, it would be nice if choosenim show would show the time of compilation
19:16:22PMunchEspecially for devel
19:17:41FromGitter<bung87> hmm I’ve read python version and go version, but still have no idea
19:17:45*abm joined #nim
19:18:52*njoseph quit (Ping timeout: 268 seconds)
19:19:04disruptekAraq, shashlick: nimgit2 on windows doesn't support ssh; is that right?
19:20:57FromGitter<alehander92> @bung87 how is host_cpu_load_info_data_t
19:20:58FromGitter<alehander92> defined
19:21:49FromGitter<alehander92> some of your host_statistics definition sseems wrong, i saw a similar stack overflow answer and error
19:21:58FromGitter<bung87> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5df29355ac14cc652c7153d6]
19:22:12FromGitter<alehander92> look at go: you need to pass host_info_t
19:22:16FromGitter<alehander92> not ptr to host_info_t
19:22:33FromGitter<alehander92> host_info_t itself is a pointer type from what the SO answer said
19:25:30FromGitter<bung87> casthost_info_t (r_load) , casthost_info_t (r_load.addr), casthost_info_t (r_load.unsafeAddr) I ‘v tried doest not work
19:25:57*JustASlacker joined #nim
19:25:58FromGitter<bung87> r_load , r_load.addr , r_load.unsafeAddr
19:26:15FromGitter<alehander92> what do you need that fr
19:26:17FromGitter<alehander92> for*
19:26:21FromGitter<alehander92> ah, hm, strange
19:26:47FromGitter<bung87> I change host_info_t to `type host_info_t = ptr cint` right?
19:27:09JustASlackerhello
19:27:19JustASlackerhow can I convert a HashSet to json
19:27:25FromGitter<alehander92> dont really know, i dont use mac honestly but
19:27:44FromGitter<alehander92> maybe overload `%` for it
19:27:48FromGitter<alehander92> JustASlacker
19:28:19JustASlackerI tried var jsonout = %* vhosts , vhost being the Hashset
19:28:46FromGitter<alehander92> yeah, i think defining a proc `%` for HashSet[T]
19:28:48FromGitter<alehander92> should be ok
19:28:58FromGitter<alehander92> which e.g. returns an array or a custom object
19:29:07FromGitter<alehander92> the question is , how would you deserialize it later
19:29:13FromGitter<alehander92> and know that it was a set
19:29:25JustASlackerit doesnt need to be a set
19:29:28JustASlackerit can be a list
19:29:38JustASlackerits just a set so elemtns are unique
19:30:07*solitudesf- joined #nim
19:30:27JustASlackerwould it be easier to cast it to array first
19:30:29JustASlackeror something
19:30:51JustASlackernot sure I know how to overload %
19:31:47*solitudesf- quit (Client Quit)
19:33:34FromGitter<bung87> https://github.com/bung87/psutil-nim/blob/master/src/psutil/psutil_macosx.nim I’ve post it to github , if you would like to help.
19:34:52FromGitter<phillvancejr> @alehander92 Oh I see what you mean! That is strange. Hopefully that will be fixed. I guess the solution for now is to just create the objects at the top level without using the C++ constructor?
19:35:05FromGitter<phillvancejr> How do we make a note of this for a bug fix?
19:36:45*kungtotte quit (Ping timeout: 268 seconds)
19:39:45*njoseph joined #nim
19:44:43FromGitter<alehander92> @phillvancejr but .. how can we do that ? is it possible at all in c++?
19:44:54FromGitter<alehander92> please open a github issue
19:44:57FromGitter<alehander92> thanks!
19:45:05FromGitter<alehander92> @bunjg87 sorry no time and i dont use mac
19:45:12FromGitter<alehander92> but open in the forum maybe?
19:45:19FromGitter<alehander92> JustASlacker please
19:45:21FromGitter<alehander92> look at json.nim
19:45:25FromGitter<alehander92> in the standard library
19:45:28FromGitter<alehander92> lot of examples
19:45:30FromGitter<alehander92> Araq
19:45:39FromGitter<alehander92> Protocol* from posix
19:45:42*Marquisd_ joined #nim
19:45:46FromGitter<alehander92> leads to some strange check errors for me
19:45:52FromGitter<alehander92> because it seems it expects 6 to 23
19:45:57FromGitter<alehander92> and IP which is 0
19:45:59FromGitter<alehander92> fails it
19:46:03FromGitter<alehander92> it works with checks:off
19:46:20*Marquisd_ is now known as Marquisdefalbala
19:47:08FromGitter<bung87> ok , I’ll post it forum
19:48:30JustASlackeralehander92 right ho
19:54:49*narimiran joined #nim
19:57:12*Vladar quit (Quit: Leaving)
20:05:08*nsf quit (Quit: WeeChat 2.6)
20:08:45shashlickdisruptek ya I've not added it yet - on Linux, cmake finds libssh2 if it is already installed
20:09:55disruptekright, that's what i figured.
20:13:11*solitudesf- joined #nim
20:14:57*solitudesf quit (Ping timeout: 240 seconds)
20:19:27*krux02 quit (Remote host closed the connection)
20:26:08shashlickUnlike nimarchive, I might probably create a libssh2 wrapper and then have nimgit2 depend on it
20:27:08disruptekmight need to sort out where to stash the so first.
20:30:44FromGitter<phillvancejr> @alehander92 The solution is definitely a hack and won't work with libraries but the way to make it work would be to always define an empty default constructor and destructor in C++. These are both required by Nim even if you aren't using them. Also we shouldn't use RAII in the destructor because otherwise the resource will be freed twice in Nim, even in some cases if you're checking for null. Instead make a
20:30:44FromGitter... clean up function that will have to be called explicitly in Nim. ⏎ thing.h ⏎ ⏎ ```new thing! ⏎ deleting data! 10``` ⏎ ... [https://gitter.im/nim-lang/Nim?at=5df2a3737ce7ea1e35727627]
20:32:59FromGitter<phillvancejr> I'm definitely not a c++ expert but this is the solution I came up with. Its not great though ha ha, we should probably just not try to use C++ classes in Nim for now
20:33:14*Hideki_ joined #nim
20:38:06*Hideki_ quit (Ping timeout: 265 seconds)
20:38:56*rockcavera joined #nim
20:41:08*Trustable quit (Remote host closed the connection)
20:42:29*njoseph quit (Ping timeout: 276 seconds)
20:42:48*Trustable joined #nim
20:59:16*letto_ quit (Quit: Konversation terminated!)
21:01:14*narimiran quit (Ping timeout: 268 seconds)
21:01:55FromGitter<alehander92> honestly i am not sure
21:02:24FromGitter<alehander92> i think
21:02:25FromGitter<alehander92> a github issue
21:02:51FromGitter<alehander92> might help as i still think this is a bug
21:02:55FromGitter<alehander92> the let behavior
21:28:59FromDiscord<Clyybber> branching with a probability is not as easy as it seems (with inclusive ranges)
21:29:51FromDiscord<Clyybber> I wonder if xoroshiro truly is inclusive though
21:33:56*njoseph joined #nim
21:51:08*PMunch quit (Quit: leaving)
22:02:42*Kanato joined #nim
22:04:46*luis_ joined #nim
22:06:36*Kaynatwo quit (Ping timeout: 268 seconds)
22:12:39*Trustable quit (Remote host closed the connection)
22:20:04*gour quit (Remote host closed the connection)
22:21:56*solitudesf- quit (Ping timeout: 276 seconds)
22:23:52*JustASlacker quit (Ping timeout: 268 seconds)
22:32:29*clyybber joined #nim
22:32:39*liam_ joined #nim
22:35:55FromGitter<phillvancejr> For sure, should be a github issue. I'll upload the examples. It definitely should work as expected
22:36:38liam_So I'm using nimongo and json in the same project. I need nimongo's %* macro to create bson for manipulating documents, but I also need to create JsonNode objects for other stuff. Because the json library also uses %* to create it's own objects, I get ambiguity errors. Is there a way to differentiate the macros for use?
22:37:03disruptekjson.%*
22:38:19liam_Yeah I tried that but I got the same errors. I'm just now thinking that maybe I have to do json.%* and bson.%*
22:38:23liam_Ill try it
22:39:06*lritter quit (Quit: Leaving)
22:40:03liam_Now I'm getting " Error: undeclared identifier: '.%*' "
22:42:12*clyybber quit (Quit: WeeChat 2.7)
22:42:23disruptekodd.
22:42:31liam_Indeed
22:42:55liam_I have double checked my syntax and it looks correct
22:43:04liam_As in, no loose '.%*'
22:43:34disrupteki kinda get why it happens with a macro, but i feel like there should be a workaround.
22:43:41disruptekie. in the compiler.
22:44:40liam_So do you think this is an issue with how the compiler implements macros?
22:44:48disrupteknah.
22:45:22disrupteki just think we should be able to parse the macro symbols.
22:45:32disruptekie. tie them to a module.
22:46:07disruptekyou didn't find anything about this on the forum? i think i've heard of this problem before.
22:47:17disruptekjson.`%*` of course.
22:48:29disruptekcoulda sworn it was the first thing i tried.
22:50:23liam_If i use json.%* where I want JsonNode objects then errors are raised at the other %* macros, and if I put 'bson.' in front of them then I get that undeclared identifier error
22:50:28liam_I'll check the forum
22:51:03disruptekyou need to say json.`%*` because operators need stropping in this context.
22:52:01liam_How is that different?
22:52:16disrupteksignificantly, it features backquotes.
22:53:09liam_Ah I see now
22:54:30liam_Nah now I'm getting more errors
22:55:13liam_Is there a way to import macros with different identifiers?
22:55:45disruptekcan you put what you have in a paste? i'm not convinced.
22:56:47liam_In a paste as in paste it in here on the IRC or in a pastebin (or similar)?
22:56:56disruptekpastebin, ideally.
22:57:56liam_https://pastebin.com/a3cBN2wG
22:58:53disruptektry a space after the identifier and the data you're passing to it as a first argument.
22:59:01disruptekthat, or wrap the arguments in ().
22:59:24disruptekjson.`%*`({"some":"data"}) should work.
22:59:39disruptekjson.`%*` {"some":"data"} should work, too.
22:59:55liam_Same for the bson?
22:59:59disrupteksure.
23:00:58liam_Whey! That worked.
23:01:00liam_Thanks
23:01:11disruptek🎉
23:01:41liam_Could you enlighten me on why that worked?
23:02:09disruptekfunction syntax; this"foo" is a proc this() that takes a single argument of a string.
23:02:26disruptekthis("foo") also works. or this "foo".
23:02:52liam_I see
23:02:57liam_And the backticks?
23:03:24disruptekoperators have to be stropped when you want to refer to them by name, as when you define them.
23:03:37disruptekotherwise, the compiler thinks you want to evaluate them.
23:03:58disruptekbecause . is an operator character, we have to strop the remaining operator characters.
23:04:51liam_Ok, got it
23:07:39*ponyride1 joined #nim
23:09:30*ponyrider quit (Ping timeout: 268 seconds)
23:15:26FromDiscord<mratsim> @Clyybber, RNG just gives you uint64
23:16:02FromDiscord<mratsim> but rejection sampling that creates an uniform distribution for a range (from a full uint64 RNG) is exclusive
23:16:24FromDiscord<mratsim> see: https://github.com/mratsim/weave/blob/master/weave/random/rng.nim#L64-L72
23:38:51*clyybber joined #nim
23:43:31*clyybber quit (Client Quit)
23:44:38*ng0 quit (Quit: Alexa, when is the end of world?)
23:48:08*luis_ quit (Quit: luis_)
23:51:38*clyybber joined #nim
23:54:45*Hideki_ joined #nim
23:55:16FromDiscord<Clyybber> @mratsim Nice thanks!
23:57:18*clyybber quit (Quit: WeeChat 2.7)
23:59:27*Hideki_ quit (Ping timeout: 268 seconds)