<< 31-01-2020 >>
00:06:23*def- quit (Quit: -)
00:06:33*def- joined #nim
00:07:56*theelous3 joined #nim
00:11:07*ptdel joined #nim
00:15:06*theelous3 quit (Ping timeout: 268 seconds)
00:19:03*chemist69 quit (Ping timeout: 245 seconds)
00:21:18*nixfreak quit (Read error: Connection reset by peer)
00:26:54FromGitter<zetashift> I will do that thanks for help :D
00:45:01*tobbez quit (Quit: Lost terminal)
00:53:03*Trustable quit (Remote host closed the connection)
01:04:35*arecaceae quit (Remote host closed the connection)
01:04:58*arecaceae joined #nim
01:14:47*ftsf joined #nim
01:20:07*PMunch quit (Quit: Leaving)
01:57:21shashlick@zetashift - here's a working wrapper for https://github.com/orx/orx using nimterop - http://ix.io/28Lv/nim
01:57:33shashlicksorry, want to make it easier to find when i search back on google
01:57:48shashlicksome stuff is stubbed out and should be declared properly if you need it
02:04:39*lritter quit (Quit: Leaving)
02:05:04FromGitter<deech> When I have `static:` output to the console `nimble` doesn't seem to show it, is it logged somewhere?
02:24:17shashlickTry verbose
02:30:06FromGitter<deech> Doesn't work.
02:35:56shashlickUsing nimble build?
02:37:37FromGitter<deech> Yep
02:47:51*icebattle quit (Ping timeout: 268 seconds)
02:56:39leorize--debug
02:57:01leorizethat's the only way I ever get nimble to show anything
02:59:00FromGitter<deech> That worked great. Thanks!
03:16:33*muffindrake quit (Quit: muffindrake)
03:18:08*muffindrake joined #nim
03:21:43*leorize quit (Ping timeout: 240 seconds)
03:25:26*leorize joined #nim
03:32:03*rockcavera quit (Remote host closed the connection)
03:38:35*muffindrake1 joined #nim
03:38:51*muffindrake1 quit (Client Quit)
03:39:04*muffindrake1 joined #nim
03:41:53*muffindrake quit (Ping timeout: 272 seconds)
03:44:15*muffindrake1 is now known as muffindrake
04:28:02FromDiscord_<Elegant Beef> A uint is much like int in that it depends on system arch?
04:29:04*arecaceae quit (Remote host closed the connection)
04:29:22*arecaceae joined #nim
04:36:09leorizeyes
04:38:26FromDiscord_<Elegant Beef> Also is it just me that finds that `Filemode.fmReadWrite`, clears the files?
04:38:41FromDiscord_<Elegant Beef> Also is it just me that finds that `Filemode.fmReadWrite`, clears the file annoying?
04:38:48*leorize_ joined #nim
04:38:49FromDiscord_<Elegant Beef> It's so annoying that it clears the file
04:39:27*leorize is now known as Guest62666
04:39:27*leorize_ is now known as leorize
04:40:06leorizethat has been the case for a long long time now with almost all languages :P
04:40:26leorizeuse fmReadWriteExisting to not clear the file
04:40:32FromDiscord_<Elegant Beef> Almost every other language it's much more explicit, atleast in my knowledge
04:40:40FromDiscord_<Elegant Beef> read/write doesnt scream clear to me
04:40:43*Guest62666 quit (Ping timeout: 240 seconds)
04:41:56leorizeyou'll have to bear with it for now
04:42:08leorizeit's a legacy thing that stick around since the beginning
04:42:22FromDiscord_<Rika> what's the difference between fmAppend and fmReadWriteExisting?
04:42:24leorizeuntil nim 2.0 I don't think we can replace the io module with something better :P
04:42:36leorizeRika: append set the cursor at the end
04:42:37FromDiscord_<Elegant Beef> Ah, the `open()` function clearing the file is really weird
04:42:52FromDiscord_<Rika> it's the same in python
04:42:52FromDiscord_<Elegant Beef> Super counter intuitive imo
04:43:04FromDiscord_<Rika> so i think it's just a python concept mistakenly copied into nium
04:43:06FromDiscord_<Rika> nim
04:43:11FromDiscord_<Elegant Beef> Ok you werent defending it
04:43:20FromDiscord_<Elegant Beef> I was going to say "It's the same in language X" isnt a valid defence
04:44:05leorizeio module hasn't got a lot of love since it's creation
04:44:08FromDiscord_<Rika> eh i dont like it either
04:44:25FromDiscord_<Elegant Beef> I come C# and it's opening is vastly more explicit
04:44:41leorizebut it's now also a crucial module that we can't just swap with something else (yet) :(
04:45:05FromDiscord_<Elegant Beef> Well i'll be over there bitching if anyone needs me
04:45:11leorizeI'm planning to make an io module that no longer depends on C stdlib
04:45:22FromDiscord_<Elegant Beef> I fought for like 2 minutes going "Why the hell is this file im manually writting to empty"
04:45:25leorizeit'll have saner APIs, but that's a lot of work
04:45:35TangerHey folks, I recently took a dive into macros to write a DSL for something I'm working on. First large-ish kinda program and it's definitely not done as efficiently as it could be. Just wondering if anybody would be keen to review my nim code style (not the code, more the way the code is structured and where I could make the code clearer). https://gitlab.com/fight-for-tangaria/main/blob/25-ecs-system/src/fftpkg/components.nim
04:46:03FromDiscord_<Rika> i'll look, but i'm not great at writing clean macros (i've made some insane ones though)
04:46:29FromDiscord_<Rika> can you explain how it's used though? is that in the comments/docs
04:47:17FromDiscord_<Rika> nvm, it's in the code
04:47:39TangerRika: Yeah, should be in the tests. Thanks man, I appreciate it! Macros are a wily beast
04:47:54leorizecan you make lines shorter? :P
04:48:37leorizealso `baseProps: Option[BasePropertyDefs] = none(BasePropertyDefs)` can be `baseProps = none(BasePropertyDefs)`
04:49:17FromDiscord_<Rika> this is a lot of helper procs 👀
04:49:42FromDiscord_<Elegant Beef> Wait... nim requires a leading 0 for floats?
04:49:45FromDiscord_<Elegant Beef> Ok that's it
04:49:53FromDiscord_<Rika> yeah?
04:49:56FromDiscord_<Elegant Beef> I now dislike nim
04:49:59FromDiscord_<Rika> ok
04:49:59FromDiscord_<Elegant Beef> 😄
04:50:06FromDiscord_<Rika> see you
04:50:11FromDiscord_<Rika> thanks for trying it out though
04:50:20FromDiscord_<Elegant Beef> Im just joking
04:50:58FromDiscord_<Rika> i dont know why no-leading-0 is a thing though
04:51:05FromDiscord_<Rika> it's offputting for me
04:51:17*nsf joined #nim
04:51:31FromDiscord_<Elegant Beef> I mostly do math by hand, it gets in the way for multiplication
04:51:34FromDiscord_<Elegant Beef> Atleast i did mostly
04:51:45FromDiscord_<Elegant Beef> Also it's not needed
04:51:56FromDiscord_<Elegant Beef> i dont put x/1 either
04:52:59TangerRika, Yeah, the goal is to cut a lot of them out or move them if they're still necessary. leorize: Yeah, haha, I should change my character width to normal term size and probably expand some statements
04:53:35FromDiscord_<Rika> buildPropDef has one hell of a line length
04:53:56FromDiscord_<Rika> actually all of the helpers signatures are long asf
04:55:24FromDiscord_<Rika> `newIdentNode(str)` can be `str.ident`
04:55:34FromDiscord_<Rika> or whatever ufcs style you use
04:56:05TangerWhat would you suggest to shorten these function signatures? I've considered turning the signatures into an actual type def or an object or something similar
04:56:06TangerCool, thanks
04:56:20*marmotini_ joined #nim
04:56:52FromDiscord_<Rika> i usually newline them
04:57:24FromDiscord_<Rika> `proc buildInheritedComponentTypeDef(name: NimNode, base: NimNode, props: NimNode`... can be `proc buildInheritedComponentTypeDef(name, base, props: NimNode`...
04:57:53FromDiscord_<Rika> basically if you have args beside each other w/ the same type you can omit the ones before the last arg w/ the type
04:58:06TangerOh snap, what a silly bugger, my damn editor is wrapping the lines -_-
04:59:32FromDiscord_<Rika> LOL
05:00:04TangerDidn't *actually* commit the line breaks I see
05:00:20*marmotini_ quit (Remote host closed the connection)
05:00:20FromDiscord_<Rika> also a `for ... in ...` won't run if the iterable is empty i'm pretty sure, so you can remove L339
05:01:59FromDiscord_<Rika> L237 has a useless comment lol
05:02:11FromDiscord_<Rika> wait, not a comment
05:02:13FromDiscord_<Rika> disregard
05:07:32FromDiscord_<Rika> i'm having some sort of mental block right now so ill call it there, maybe someone else can continue reading the macro
05:07:57TangerRika: All good, thanks once again for that!
05:31:04*dddddd quit (Remote host closed the connection)
05:34:28*Romanson joined #nim
05:49:49*sagax joined #nim
05:53:10*narimiran joined #nim
06:23:49*actuallybatman quit (Quit: leaving)
06:25:25*ptdel quit (Quit: https://quassel-irc.org - Chat comfortably. Anywhere.)
07:12:20ZevvTo the Battrain!
07:14:54Zevvhttps://joellaity.com/2020/01/31/string.html
07:17:03FromDiscord_<Rika> great, he uses light green and yellow
07:17:14FromDiscord_<Rika> cant even see the difference
07:17:15FromDiscord_<Rika> rip
07:17:26Zevvyeah, but our strings are only green on black, so who cares
07:19:48*jjido joined #nim
07:20:03*theelous3 joined #nim
07:21:17*tefter quit (Quit: WeeChat 2.7)
07:29:17*Vladar joined #nim
07:34:22*solitudesf joined #nim
07:59:45*xet7 joined #nim
08:00:00*gmpreussner quit (Quit: kthxbye)
08:02:47*ljoonal quit (Quit: ljoonal.xyz)
08:04:36*ljoonal joined #nim
08:04:48*gmpreussner joined #nim
08:10:27*ljoonal quit (Quit: ljoonal.xyz)
08:14:47*ljoonal joined #nim
08:27:27*floppydh joined #nim
08:33:40*Romanson quit (Quit: Connection closed for inactivity)
08:38:56FromDiscord_<Rika> hello, i'm being dumb
08:39:03FromDiscord_<Rika> how do i use a code block to assign
08:41:38TangerRika: https://play.nim-lang.org/#ix=28Ml
08:43:02FromDiscord_<Rika> tried that but it's giving me an error when i use it, do i need type annotations?
08:43:42FromDiscord_<Rika> oh, apparently i commented out my else clause
08:43:43FromDiscord_<Rika> lol
08:45:12TangerXD
08:46:49FromDiscord_<Rika> i hate myself, i made this macro too insane lol
09:13:44Zevvyou must find your inner Zen - breathe the macro - grok the macro - you and the macro must become one
09:14:38*sunwukong joined #nim
09:15:09*sunwukong left #nim (#nim)
09:27:03*Pqzcih5- quit (Ping timeout: 240 seconds)
09:28:00*Pqzcih5 joined #nim
09:37:46*xet7 quit (Quit: Leaving)
09:38:07*xet7 joined #nim
09:44:03*Pqzcih5 quit (Ping timeout: 240 seconds)
09:54:11FromDiscord_<exelotl> You must channel your inner Zevv
09:54:39*chemist69 joined #nim
10:00:51*Zevv 's zen just slightly cracked when his Thalys was canceled
10:05:41FromGitter<alehander92> dont
10:05:51FromGitter<alehander92> so what happened
10:05:54FromGitter<alehander92> with arduino
10:10:42*m|b_ joined #nim
10:11:14Zevv"happened"
10:22:42FromGitter<alehander92> what did
10:22:51FromGitter<alehander92> the arduino people say
10:22:53FromGitter<alehander92> i remember
10:22:58FromGitter<alehander92> patching avr-g++
10:23:02FromGitter<alehander92> or something like this was the last idea
10:26:07Zevvthe arduino developers are not very vocal
10:26:24Zevvthe forum now started to focus on how Nim was no better then C and that I should not use the String (with a capital S) class on an arduino
10:26:27Zevvso that was very helpful
10:26:37Zevvanyway, stuff works, albeit with a hack or two
10:28:56*chemist69 quit (Quit: WeeChat 2.7)
10:32:56FromGitter<alehander92> well they are right
10:33:12*chemist69 joined #nim
10:33:45FromGitter<alehander92> C is a better fit for arduino just because of its massive popularity/ecosystem,obvious no-gc and popularity
10:34:01FromGitter<alehander92> but yeah other languages still might be supported
10:34:10FromGitter<alehander92> good, so i hope to see
10:34:19FromGitter<alehander92> some projects soon :P
10:34:24FromGitter<alehander92> :) *
10:35:58dom96C is so full of gotchas though
10:36:18FromDiscord_<Rika> which Nim usually doesnt have no?
10:36:23dom96It might be fine for a small script that blinks lights, but once you want to develop something larger it will be a massive PITA
10:36:40dom96Every language has gotchas, but Nim has far fewer than C I'd say
10:37:18FromGitter<alehander92> yeah but if they have to support one language
10:37:24FromGitter<alehander92> it makes sense it would be C
10:37:31FromGitter<alehander92> now multilanguage support is nice
10:37:41FromGitter<alehander92> but i dont know enough about arduino to talk here
10:38:39FromDiscord_<Rika> dom96, which is why i said usually, because you dont know what you dont know lol
10:42:04ZevvTheir argument was that a zero terminated array of chars is better
10:42:13Zevvok, then I rest my case
10:44:34FromDiscord_<Rika> arent nim strings just wrapping \0 chararrays anyway
10:49:52lqdev[m]@Rika true. they have a dedicated length field, though.
10:53:36*Pqzcih5 joined #nim
10:54:11*narimiran quit (Ping timeout: 265 seconds)
11:01:26FromDiscord_<Rika> you should probably signify that in the forum post then
11:18:12*abm joined #nim
11:18:13*cgfuh joined #nim
11:33:00*thomasross_ joined #nim
11:33:00*thomasross quit (Killed (orwell.freenode.net (Nickname regained by services)))
11:33:00*thomasross_ is now known as thomasross
11:39:36*belamenso joined #nim
11:43:54Zevvnah I dont care at all. i came to ask bout arduino, not to tell about nim
11:51:26*dddddd joined #nim
12:00:08*ng0 joined #nim
12:07:10Yardanicodoes "zip" module support uncompressing LZMA data?
12:13:32*rockcavera joined #nim
12:26:07*nsf quit (Quit: WeeChat 2.7)
12:27:55*m|b_ quit (Quit: Connection closed for inactivity)
12:37:02*krux02 joined #nim
12:47:29FromGitter<alehander92> just realized a loop until a very big value in a for
12:47:38FromGitter<alehander92> and again started to think
12:47:50FromGitter<alehander92> do we need int-s, or do we need ranges most of the time
12:48:01FromGitter<alehander92> is there somebody who uses a lot of range types
12:48:03FromGitter<alehander92> in nim
12:49:31*jjido quit (Quit: Connection closed for inactivity)
12:50:57*narimiran joined #nim
12:52:22*Trustable joined #nim
12:54:51*m|b_ joined #nim
13:09:29*Pqzcih5 quit (Remote host closed the connection)
13:10:52*nixfreak joined #nim
13:16:41*Zevv rarely. None of the languages i used befor nim had them, so they're simply not in my mental toolkit until i really need it
13:18:30FromDiscord_<sbr> Does someone know how I can serve static assets with Jester in production (Heroku + nginx)?
13:18:30FromDiscord_<sbr> My css, js and ico files are in the standard jester folder "public".
13:18:30FromDiscord_<sbr> With local development, Jester generates the main html file and also pulls the other static files from the "public" folder.
13:18:30FromDiscord_<sbr> In production, I get a 403 forbidden error for everything in the "public" folder.
13:18:31FromDiscord_<sbr> Here's the nginx config: https://gitlab.com/sophiabrandt/nimurlshortener/blob/master/services/nginx/default.conf
13:18:34FromDiscord_<sbr> Here's the main .nim file (I tried to enable CORS on all routes): https://gitlab.com/sophiabrandt/nimurlshortener/blob/master/services/url-shortener/src/urlShortener.nim
13:20:52*Vladar quit (Quit: Leaving)
13:22:53*kungtotte quit (Read error: Connection reset by peer)
13:23:58*kungtotte joined #nim
13:32:24shashlick@Yardanico use nimarchive
13:34:54Yardanicoshashlick: thanks
13:35:25Yardanicoit's just that i've used my own small binding for easylzma before, but i'd prefer to use something maintained :)
13:35:49shashlickNimarchive is in choosenim
13:36:13shashlickSo the more users, the more stable it can get
13:38:18Yardanicoshashlick: can I uncompress LZMA in memory with it?
13:38:18*nixfreak quit (Read error: Connection reset by peer)
13:39:30*nixfreak joined #nim
13:40:00Yardanicoah nvm it seems like I can
13:41:34*skellock joined #nim
13:43:20*marmotini_ joined #nim
13:43:56*krux02 quit (Remote host closed the connection)
13:55:06*belamenso quit (Remote host closed the connection)
14:09:25shashlicklibarchive is pretty powerful
14:09:40shashlickWhat you using this for
14:09:52Yardanicoshashlick: yeah, I see, I'm mostly done but I can't figure out how to read all data into my buffer :P
14:10:18YardanicoI probably need archive_read_data but it needs a bufsize argument, but I don't know the size of the uncompressed data
14:11:11Yardanicoshashlick: just for my toy replay reading lib for a game
14:11:19Yardanicoit stores game events in an LZMA stream and I've used easylzma before
14:13:01shashlickCool
14:13:06shashlickWhat os
14:13:10Yardanicolinux
14:13:22Yardanicoi mean the bindings themselves work just fine
14:13:31YardanicoI just need to figure out how to do it with libarchive :P
14:17:19Yardanicoah nvm there's archive_entry_size
14:19:09Yardanicoshashlick: one slight problem is that nimsuggest doesn't seem to pick up nimarchive definitions
14:19:27Yardanicoso in vscode I see "undeclared identifier" for all libarchive stuff even if it compiles fine
14:22:45Araqhttps://github.com/nim-lang/Nim/pull/13301 upvote/downvote please
14:22:46disbot➥ new `quoted`: outplace version of `addQuoted`
14:28:26Yardanicoshashlick: seems like I figured out how to uncompress raw LZMA data :)
14:28:29Yardanicodo you need it as an example?
14:30:09Yardanicoalso is doing "cast[pointer](cstring(data))" correct? when data is a nim string
14:30:23Yardanicowhen I want to use string as a buffer so C lib can write something in it
14:30:31Yardanico(well I mean it works but I don't know if it's correct)
14:36:32Yardanicoalthough with my usage it seems that using libarchive is much slower than using easylzma directly
14:46:10*Trustable quit (Remote host closed the connection)
14:51:57*nullwarp joined #nim
14:59:47*nixfreak quit (Ping timeout: 252 seconds)
15:12:28FromDiscord_<Clyybber> Araq: We should really give sugar.outplace a nice operator
15:14:40shashlickYardanico: if you compile at least once, suggestions will start again
15:14:52shashlickPlease share example as well
15:14:55Yardanicoshashlick: I compiled like 30 times already :P
15:15:05Yardanicoeven after restarting VSCode it's still the same
15:15:49shashlickDo you have latest nimterop?
15:16:04Yardanicoshashlick: 0.4.4
15:16:34Yardaniconim check doesn't find nimarchive stuff
15:17:09Yardanicocan this be because I'm not in the folder the nim file is in?
15:17:10Yardanico"Error: unhandled exception: /home/dian/.nimble/pkgs/nimterop-0.4.4/nimterop/build.nim(80, 18) `false` Results not cached - run nim c/cpp at least once"
15:17:21shashlickHmm
15:17:31Yardanicoit still doesn't work even if I'm in the same folder, hm
15:18:08Yardanicoshashlick: it doesn't work in one simple file as well
15:18:21shashlickStrange
15:18:36shashlickI'll have to check again
15:19:30Yardanicoshashlick: this is how I uncompress LZMA, although I don't know if it's correct or can be improved https://play.nim-lang.org/#ix=28Nt
15:21:21*ng0_ joined #nim
15:21:21*ng0_ quit (Changing host)
15:21:21*ng0_ joined #nim
15:24:53*ng0 quit (Ping timeout: 272 seconds)
15:27:55*m|b_ quit (Quit: Connection closed for inactivity)
15:29:19shashlickLooks like you can ship https://github.com/genotrance/nimarchive/issues/7
15:29:21disbot➥ Add archive writer example in the Readme. ; snippet at 12https://play.nim-lang.org/#ix=28Nv
15:32:51*floppydh quit (Quit: WeeChat 2.7)
15:37:18*ng0_ is now known as ng0
15:48:53*lritter joined #nim
15:53:15*icebattle joined #nim
15:58:56*ng0 quit (Quit: leaving)
16:01:48FromDiscord_<Rika> sometimes i wish nim code could also compile into lua, but lua is prolly as hard to convert to as js
16:03:09FromDiscord_<demotomohiro> In my old nim code, there are lines like "echo foo;", but Nim don't show any errors or warnings.
16:03:09FromDiscord_<demotomohiro> If there are ';' in all end of lines, code looks like C/C++.
16:03:21Araqfun fact: the jsgen was a luagen once
16:03:30Araqit never worked though
16:05:19nisstyreRika: it's probably not hard to embed a lua interpreter in your nim binary
16:05:24nisstyrebut...why
16:05:52nisstyreI guess I can see it being useful as a scripting language
16:05:57nisstyreif you're building a game engine
16:06:41Yardanico@demotomohiro yes you can use ; in Nim code
16:07:09Yardanicoso you can put multiple statements on one line, and it can also used as a delimiter in proc arguments (and it behaves differently than , in some cases)
16:08:08nisstyrealso in if statements
16:08:13Yardanicoyeah
16:08:22Yardanicolike if (let a = something(); a > 5)
16:08:34*marmotini_ quit (Remote host closed the connection)
16:08:36nisstyreI find that a great alternative to the C behaviour of assignment expressions
16:08:41Yardanicoor while (let read = readSomething(); read > 0)
16:08:44nisstyremuch more readable
16:08:58Yardanicoor proc b(x:int):int=(result=5;(if x>5:(if x<3:(if x==0:(echo x;echo x-1)))elif x==0:(echo 0;echo 5)else:return 5))
16:08:58YardanicoxD
16:09:41*a_b_m joined #nim
16:09:43FromGitter<erhlee-bird> @nisstyre embedding a small LuaJIT has been really helpful for embedded environments where you have binary size constraints and want to provide a "safe" layer above with C/C++ FFI
16:10:22nisstyrethat makes sense
16:11:45FromDiscord_<demotomohiro> @Yardanico in addtion to write multiple statements on one line, I can use ; to make code looks like C 🙂
16:11:59Yardanicowell, do it if you want :)
16:12:13Yardanicobut pls dont
16:12:24FromDiscord_<Rika> nisstyre, i mean nim to lua
16:12:39FromDiscord_<Rika> i want to code something in nim, but what i'm implementing should be in lua
16:12:48nisstyreyeah I know, I'm suggesting the alternative would be to embed the interpreter
16:12:49Yardanico@Rika while that might be useful in some rare cases, I don't really think it's worth the effort of maintaining a backend
16:12:54FromDiscord_<Rika> it's for aegisub automation scripting
16:12:56Yardaniconim even had a PHP backend :P
16:12:57FromDiscord_<Rika> i know it isnt
16:13:03FromDiscord_<demotomohiro> I removed `;` in my old Nim code so it doesn't looks like C.
16:13:03nisstyrecompiling to Lua, I don't know what the benefits would be
16:13:15FromDiscord_<Rika> i'm looking into how to extend aegisub without using lua
16:13:21nisstyrePHP backend is just monstrous
16:13:24*abm quit (Ping timeout: 248 seconds)
16:13:33FromDiscord_<Rika> god
16:14:06Yardaniconisstyre: well IIRC Ar*q said he was paid to make it
16:14:12nisstyreoh, ok
16:14:14Yardanicoand it was mostly unmaintained and got removed around 1.5-2 years ago
16:14:19nisstyrestill, that would horrify me
16:14:21FromDiscord_<demotomohiro> When `func` was added to Nim, I asked Araq whether I can use `func` in compiler or standard library code when I write PR for Nim.
16:14:21FromDiscord_<demotomohiro> iirc, he said no. But changed mind?
16:14:21FromDiscord_<demotomohiro> https://github.com/nim-lang/Nim/pull/13285
16:14:22disbot➥ use func keyword in system
16:14:37Yardanico@demotomohiro well, it's all about bootstrapping and stuff
16:14:59Yardanicoolder compiler version that doesn't know about "func" wouldn't be able to compile code with "func"
16:15:05Yardanicobut csources that we use now support "func"
16:15:10Yardanico(well, that's my vision on that problem)
16:16:09FromDiscord_<demotomohiro> @Yardanico I see
16:16:40Yardanicoit's the same reason of why new nim language features can't be used in the compiler/stdlib right away
16:17:48Araqthat and I don't fight these things forever
16:18:05Araqso apparently now 'func' is acceptable
16:19:44*dddddd quit (Remote host closed the connection)
16:19:56FromDiscord_<demotomohiro> there are many procs with `noSideEffect` pragma in Nim/lib
16:20:56Yardanicojust replace them all xD
16:23:37Yardanicoto be honest I don't really use the effect system though :P
16:24:15Yardanicowell I mean I don't use noSideEffect or tags
16:28:51FromDiscord_<demotomohiro> @Yardanico why? dont you use noSideEffect?
16:29:14Yardanicowell, because I don't really need it in my own code :P
16:29:25Yardanicoi mostly write code or procedures which mutate stuff
16:29:28Yardaniconot pure functions
16:30:10*a_b_m quit (Quit: Leaving)
16:31:54FromDiscord_<demotomohiro> pure function means if all inputs are constant, compiler can replace function call to constant.
16:32:25Yardanicowell, I thought a pure function is "Its return value is the same for the same arguments. "
16:32:30Yardanicoeven if the arguments are decided on runtime
16:32:55Yardanicoand ". Its evaluation has no side effects."
16:34:02Yardanicoah nvm I didn't understand your message
16:38:49*akitoshi joined #nim
16:40:33akitoshiHi, what’s the nim equivalent of D’s opDispatch?
16:40:46Yardanicowhat does that do?
16:41:28akitoshiMember names not found in a class or struct can be forwarded to a template function named opDispatch for resolution
16:42:17lqdev[m]akitoshi: experimental dot operators https://nim-lang.org/docs/manual_experimental.html#special-operators-dot-operators
16:42:51Yardanicoand really I don't think you'd need that in Nim with macros and stuff
16:43:53FromDiscord_<demotomohiro> that sound like this: https://nim-lang.org/docs/manual.html#procedures-properties
16:46:21leorizeYardanico: as long as the function don't touch states other than the ones it can reach via parameters, it's qualified for {.noSideEffect.}
16:46:28Yardanicoleorize: oh really?
16:46:34leorizeyes
16:46:38Yardanicowell, I also use "var Type" sometimes
16:46:43Yardanicothat doesn't count I guess?
16:46:48leorizeNim's definition is really lax
16:46:51leorizethat one is allowed
16:47:05Zevvleorize: afaik that's not true - it's not "no side effects" in the FP sense, right?
16:47:15leorizeyep
16:47:22FromDiscord_<Rika> in the fp sense, all args are immutable no?
16:47:58ZevvOh I'm wrong, what leorize says is what the manual says
16:47:59leorizehttps://play.nim-lang.org/#ix=28NP
16:48:28leorizevar T is supported
16:48:52leorizethe only thing Nim won't let you is when you access states that shouldn't be accessible via parameters
16:51:25akitoshilqdev[m]: thanks, dot operator works like a charm :)
16:55:48FromDiscord_<demotomohiro> oh, I can even call time.epochTime() in func.
16:55:48FromDiscord_<demotomohiro> https://play.nim-lang.org/#ix=28NQ
16:55:54Yardanicowaat
16:56:06Yardanicothat should certainly be disallowed
16:56:41lqdev[m]that seems broken
16:57:39FromDiscord_<demotomohiro> It seems func in Nim only means it doesn't have side effect. It doesn't mean same return value for same argument.
16:58:19Yardanicowell reading time from system shouldn't be allowed anyway
16:58:23Araqit means what the spec says it means
16:58:47Araqbut epochTime should have .sideEffect, it does have a .tag though, yay
17:06:33lqdev[m]not sure now reading time is a side effect, but ok
17:06:43lqdev[m]s/now/how/
17:07:03lqdev[m]it's not like reading time in a thread will break something
17:07:19Yardanicoit's about accessing global state and stuff like that
17:07:30Yardanico"echo" is a side effect too
17:07:37lqdev[m]then "side effect" is a bit of a misnomer.
17:07:45lqdev[m]echo does have side effects, ofc
17:07:51lqdev[m]because it writes global state
17:08:05lqdev[m]but reading global state shouldn't be a side effect imo
17:08:33Araqit works out if the .importc'ed procs are annotated properly
17:08:39lqdev[m]maybe I'm just dumb, idk.
17:08:43Araqbut often they are not, see epochTime
17:09:26AraqnoSideEffect = proc result only depends on its parameters = proc doesn't use global state
17:11:40Yardanicoyeah I understood now, seems like I could make most of my "proc" as "func"
17:13:27Yardanicois there an easy way to which side effects does my proc have?
17:13:30Yardanicoto see*
17:14:59Yardanicoah, well, one way is to use "nim doc" :P
17:15:19shashlickany other crucial choosenim issues open?
17:15:20leorize[m]just mark it with nosideeffects :p
17:15:28Yardanicoleorize: it doesn't say which side effects exactly
17:15:39leorize[m]iirc there's also an {.effects.} statement that print them
17:15:45Yardanicoah right
17:15:46FromDiscord_<demotomohiro> Should I make a issue about I can use epochTime in func?
17:15:50Yardanicoyeah
17:16:43*ptdel joined #nim
17:16:52leorize[m]yes
17:21:01*abm joined #nim
17:29:01*skellock quit (Ping timeout: 252 seconds)
17:38:44FromDiscord_<demotomohiro> done: https://github.com/nim-lang/Nim/issues/13306
17:38:47disbot➥ I can use times.epochTime() in func. ; snippet at 12https://play.nim-lang.org/#ix=28O8
17:45:41shashlickhow do you get osx to load a brew installed osx?
17:45:49shashlickugh brew installed openssl
17:45:54disruptekreading time is an effect because it helps make unreproducible builds.
17:46:14shashlicki want it to load a different libcrypto.dylib
17:51:52shashlickgot it - https://stackoverflow.com/questions/38670295/homebrew-refusing-to-link-openssl
17:52:10disruptekshashlick how come you aren't at fosdem?
17:52:16disruptekHOW COME
17:52:27shashlicktoo far man
17:52:31shashlicki'll come some day
17:52:42shashlickbundle some work travel with it perhaps
17:53:32FromDiscord_<Recruit_main_70007> i will go to Nimcom
17:56:00*skellock joined #nim
18:02:22*skellock quit (Ping timeout: 268 seconds)
18:02:27disruptekhmm, someone came from california, shashlick.
18:02:37disruptekwhat does that say about you?
18:03:03disruptekvery, very troubling. you just might be too old.
18:04:16FromDiscord_<kodkuce> old xD
18:07:02Zevvdisruptek: when do you travel?
18:07:12Zevvor are you in .be already
18:08:45disrupteki'm here; just took me awhile to find a quality source of glass without usng the french word for crystal.
18:08:54*nsf joined #nim
18:09:29Zevvyou'll manage. If they don't understand you, just try speaking louder
18:10:52disruptekturns out track marks are a form of written communication for more than just leo.
18:11:23shashlick😄
18:12:06disruptekwait, is .be not belize?
18:12:32disrupteki've made a terrible mistake.
18:12:42Zevvthe fun you'll have
18:13:27livcd.be is belgian
18:13:32disruptekbut i brought enough edibles for a week. 🤭
18:13:49Zevven we're all waiting for you here in belarus :(
18:14:04disrupteksumbitch
18:18:55shashlickDoes anyone know where the dynlib search code is in the compiler
18:19:54shashlickWhere stuff like https://github.com/nim-lang/Nim/blob/devel/lib/wrappers/openssl.nim#L68 gets resolved
18:21:06disrupteki know to embed the path you need to rpath because dumb.
18:21:25*disruptek out.
18:23:15FromDiscord_<dissociative Peach> Hello, I don't need help right now but I'd like to say that I am using nim to write a system tool for managing a tree of jails/vms/sessions, with helpers to change information for all or part of children. I'd wager there's probably tools like thiss existing, but I do not know what the name of them is, so I do not know how many there are.
18:23:48FromDiscord_<dissociative Peach> I will probably come for help in the near future.
18:25:45*moon-chilled joined #nim
18:29:45*krux02 joined #nim
18:50:50shashlickhttps://github.com/dom96/choosenim/issues/122#issuecomment-580862084
18:50:52disbot➥ SIGBUS: Illegal storage access. (Attempt to read from nil?) on macOS ; snippet at 12https://play.nim-lang.org/#ix=27Z9
18:56:51*akitoshi quit (Quit: Connection closed for inactivity)
19:06:10FromGitter<timotheecour> @shashlick ⏎ ⏎ > proc loadLibPattern*(pattern: string, globalSymbols = false): LibHandle =
19:08:42shashlickthanks, I found it
19:10:51*skellock joined #nim
19:16:03*skellock quit (Ping timeout: 272 seconds)
19:16:16FromGitter<timotheecour> @shashlick could that then simply be a case of choosenim built using version 1.0 of ssl and then at runtime it loads 1.1 and then crashes?
19:16:50shashlickcould be - not forward compatible?
19:17:06shashlickwell rather backward compatible
19:18:06FromGitter<timotheecour> it’s a real possibility; u could try to repro by building choosenim w 1.1 and see if problem persists for eg
19:18:13*ptdel quit (Ping timeout: 245 seconds)
19:21:08shashlickokay will try
19:21:14shashlickthen might need to build on travis with latest ssl
19:21:19FromDiscord_<Clyybber> timotheecour: In your make pr, you saw a 5 times speed up with make -j
19:21:21FromDiscord_<Clyybber> right?
19:21:36FromDiscord_<Clyybber> What was the speedup after doing make -j $(nproc) ?
19:22:02FromGitter<timotheecour> on my laptop (16 logical cpu), and for time(building csources), yes
19:22:41FromDiscord_<Clyybber> I think we should be fine with just make -j then
19:25:01FromDiscord_<Clyybber> for portabilities sake
19:26:01FromGitter<timotheecour> no, read PR history+comments, and read also https://unix.stackexchange.com/questions/316644/is-make-j-with-no-argument-dangerous + other threads, it may cripple performance as it can spawn unlimited jobs
19:26:05FromDiscord_<Clyybber> because afaict the script will fail unless you have nproc sysctl or getconf available
19:26:27FromDiscord_<Clyybber> timotheecour: It will just spawn as many jobs as they are files to process independently
19:27:21FromGitter<timotheecour> yes, that can cripple a machine with low number of logical cpu’s; and it all depends on load average and other factors, so i changed it from my original version which just had `make -j`
19:27:43FromDiscord_<Clyybber> but its not portable
19:27:47FromDiscord_<Clyybber> and that is worse IMO
19:28:02FromDiscord_<Clyybber> csources doesn't contain that many files anyways
19:28:44FromGitter<timotheecour> if we ever find a machine where `logicalCpus=$(nproc 2>/dev/null || sysctl -n hw.logicalcpu 2>/dev/null || getconf _NPROCESSORS_ONLN 2>/dev/null)` fails, it’s trivial enough to add some default (say, 1) in a future PR
19:29:11FromDiscord_<Recruit_main_70007> sorry to interrupt, but i was wondering if they were any active ML framework, i am unable to properly intall Nimtorch and Arraymancer needs a dll i cant provide
19:29:32FromDiscord_<Clyybber> Why can you not provide it?
19:29:54FromDiscord_<Clyybber> timotheecour: Right. I'll add a default
19:30:05FromDiscord_<Recruit_main_70007> i just tried everything i could, and didnt work
19:30:58FromDiscord_<Clyybber> timotheecour: Also that nproc + 1 thing, its a stupid heuristic. On my machine when having firefox running it makes stuff slower
19:31:14FromGitter<timotheecour> clybber cool, thanks; please also add a comment in your PR linking to something like https://unix.stackexchange.com/questions/316644/is-make-j-with-no-argument-dangerous so that the next person doesn’t wonder about the same thing
19:31:15FromDiscord_<Clyybber> But I know its used a lot :p
19:31:50FromDiscord_<Clyybber> timotheecour: I'll add it in the commit message, that seems cleaner
19:32:49FromDiscord_<Clyybber> And normally schedulers should be smart enough not to lock up your system with make -j
19:33:43FromGitter<timotheecour> well commit messages are not as visible as in the code, but up to u; the +1 heuristic should make a benefitial difference for machienes with low logical cpu count (eg 1 turns into 2 etc) ; if it’s detrimental i’d like to look into it, mind sharing more details (either PM or as a comment in PR where i introduced it)
19:34:38FromDiscord_<Clyybber> its detrimental if your cpu is already on load
19:34:45FromDiscord_<Clyybber> so that cpu is the bottleneck
19:34:47FromDiscord_<Clyybber> and not io
19:35:10FromDiscord_<Clyybber> which I think might be on CIs but I havent measured it
19:35:26FromGitter<timotheecour> there’s also: ⏎ ⏎ > `-l [load], --load-average[=load]`
19:35:30FromGitter<timotheecour> in make
19:36:08FromDiscord_<Clyybber> Oh, I see.
19:36:34FromDiscord_<Clyybber> That could allow us to get rid of nproc || ....
19:36:40FromDiscord_<Clyybber> Right?
19:39:14*onionhammer quit (Ping timeout: 240 seconds)
19:46:07*skellock joined #nim
19:48:05FromGitter<timotheecour> not really; you’d have to carefully craft that load average; I tried on my system `make -j -l X` with different values of X and it’s slower than with `-j $nproc`unless X is large enough, and u’d need nproc to set X anyway
19:50:41*skellock quit (Ping timeout: 265 seconds)
19:52:25FromDiscord_<Clyybber> just set x to 90?
19:52:35shashlickgosh nim.cfg is too brittle
19:52:42FromDiscord_<Clyybber> use nims
19:52:44FromDiscord_<Clyybber> : )
19:52:55FromGitter<timotheecour> ya use nims, i hope nim.cfg can die
19:53:02shashlicktrying to fix some mingw issues in the repo
19:53:03FromDiscord_<Clyybber> same
19:54:08shashlickeven if I set amd64.windows.gcc.cpp.exe, nim still uses g++
19:54:28shashlickalso unclear how to @if cpu = x
19:54:51FromGitter<timotheecour> `just set x to 90?` that would behave essentially like unlimited; crafting `-l` is ok on one fixed machine, but having a heuristic that works in a range of machines (from low powered build bots to hight powered local dev machienes) is a much harder task than the simple nproc+1 heuristic; YMMV
19:56:23FromDiscord_<Clyybber> but low powered build bots will reach 90 way faster so it *balances* itself out
19:56:31FromDiscord_<Clyybber> If I understood it right
19:57:01shashlicksame stuff works on osx but not on linux
19:57:21*skellock joined #nim
19:58:20FromDiscord_<Clyybber> shashlick: Weird
19:59:47shashlickhttps://github.com/nim-lang/Nim/compare/devel...genotrance:mingwbugs?expand=1
20:00:27shashlickso `nim c -d:mingw` and `nim cpp -d:mingw` now work correctly on osx but on linux, only `nim c -d:mingw` works but `nim cpp -d:mingw` uses /usr/bin/g++
20:02:11*skellock quit (Ping timeout: 268 seconds)
20:02:29FromGitter<timotheecour> @shashlik i had a PR in progress that would output the command line flags (joined all the various from nim.cfg, config.nims, command line flags); would that help?
20:03:00FromGitter<timotheecour> (would output it to `nim dump` ’s json output that is)
20:03:13shashlickwould nim dump -d:mingw work
20:03:53FromGitter<timotheecour> it’d output exactly what the compiler sees after it processes all configs
20:05:17shashlickbesides config/nim.cfg, what else affects nim's selection of gcc, etc
20:06:19*letto_ quit (Ping timeout: 252 seconds)
20:08:42FromGitter<kaushalmodi> config.nims
20:09:50shashlickgosh what dumb stuff, i was thinking I was testing nimdevel while nim = 1.0.4
20:13:40shashlickphew - https://github.com/nim-lang/Nim/pull/13307
20:13:41disbot➥ Fix #10717, fix #13284
20:21:55*ftsf quit (Ping timeout: 268 seconds)
20:22:50*skellock joined #nim
20:22:58*icebattle quit (Ping timeout: 268 seconds)
20:27:42*skellock quit (Ping timeout: 265 seconds)
20:37:11*letto_ joined #nim
20:38:25FromDiscord_<Clyybber> timotheecour: FWIW -j -l is exactly as fast as -j (nproc)
20:38:44FromDiscord_<Clyybber> -l 80 that is
20:39:07FromDiscord_<Clyybber> (1 sec faster actually, not that it matters)
20:40:15FromDiscord_<Clyybber> I meant "as fast as -j (nproc + 1)"
20:40:20FromGitter<timotheecour> Yes on my dev machine, but u’d need to try it on other dev machines too (eg CI machiens with low logical cpu count) and different environment (eg w resource contention)
20:40:42FromDiscord_<Clyybber> I think its even better there
20:40:48FromDiscord_<Clyybber> since it takes into account the actual load
20:40:54FromDiscord_<Clyybber> and not just the core count
20:41:43FromGitter<timotheecour> with `-l 80` u’d allow spawn ~80 jobs and run into issues mentioned in https://unix.stackexchange.com/questions/316644/is-make-j-with-no-argument-dangerous
20:42:41FromGitter<timotheecour> (ie: 80 probably too high, but then picking a good value that works across the baord is not easy at all, compared to simple nproc heuristic)
20:44:45FromGitter<timotheecour> point is that `nproc+1` heuristic is commonly used so it’s a safe choice; if u somehow find an article that shows a better multi-scenario setting involving `-j -l X` then that would be a convincing argument.
20:45:25*icebattle joined #nim
20:46:19FromDiscord_<Clyybber> timotheecour: I'm not doing -j 80
20:46:28FromDiscord_<Clyybber> -l 80 takes into account the load
20:47:05FromDiscord_<Clyybber> That article would be `man make` :p
20:47:15FromDiscord_<Clyybber> I mean, its there for exactly this purpose
20:47:47FromGitter<timotheecour> I know; my point is `-j -l 80` means: spawn infinitely many jobs until the load average reaches 80.0
20:48:05FromDiscord_<Clyybber> yep, and thats a much better heuristic
20:49:00FromGitter<timotheecour> feel free to open a PR with your suggestion, I’m not convinced at all
20:49:39FromDiscord_<Clyybber> Alright
20:50:58*theelous3 quit (Ping timeout: 260 seconds)
20:51:53FromDiscord_<Clyybber> So I tried it on my raspberry pi 3 now too
20:52:09FromDiscord_<Clyybber> And it is as fast as -j (nproc + 1)
20:52:32FromDiscord_<Clyybber> It doesn't matter wether -l 80 or 90
20:55:43FromGitter<timotheecour> right, the value is so high it won’t be actually used; the only issue is with machines with very low logical cpu count ⏎ ⏎ > CPU: 4× ARM Cortex-A53, 1.2GHz ⏎ ⏎ has 4 (maybe 8 logical, i dunno); u shd try also on machines with just 1 (eg u can try on AWS using the free tier machines) [https://gitter.im/nim-lang/Nim?at=5e34944f73ddad4acd8536fe]
20:57:37FromDiscord_<Clyybber> No need to try
20:57:46FromDiscord_<Clyybber> On 1 core its just: start job
20:58:07FromDiscord_<Clyybber> if we are not loaded (io bottleneck) spawn another
20:58:39FromDiscord_<Clyybber> I'm sure you'll agree that its a much better heuristic than nproc + 1
20:59:15FromDiscord_<Clyybber> Btw timotheecour what does `echo_run` do exactly?
20:59:39FromGitter<timotheecour> prints cmd then runs cmd. very useful.
20:59:58FromDiscord_<Clyybber> Ah, right. $* are all args, cool thanks
21:00:39FromDiscord_<Clyybber> Any reason to use printf "/n.../n" instead of just echo?
21:02:22FromDiscord_<Clyybber> ping timotheecour
21:02:30*skellock joined #nim
21:04:40FromDiscord_<Clyybber> also, why can we only run parallell with no options given?
21:04:47*skellock quit (Client Quit)
21:04:47shashlickyay I don't need to do anything for Dash docsets
21:05:18*skellock joined #nim
21:09:09FromGitter<timotheecour> 1) printf is like echo but more formatting can be done; i’ve adapted it from a personal script where it did make a difference; here it maybe doesn’t make a difference ⏎ 2) also, why can we only run parallell with no options given? ⏎ because the makefile doesn’t accept options (eg `--cpu`)
21:10:17*skellock quit (Quit: WeeChat 2.7)
21:14:03*narimiran quit (Ping timeout: 245 seconds)
21:18:47*dddddd joined #nim
21:21:16FromDiscord_<Clyybber> alright
21:41:28shashlickwhat do you do with this - cannot bind another '=destroy' to: SvnzFileObj; previous declaration was constructed here implicitly
21:41:43shashlickmy destroy does specific stuff
21:47:40*luis_ joined #nim
21:55:09FromGitter<timotheecour> @shashlick for ssl stuff, u may wanna look at very similar past issues related to SSL in nim eg https://github.com/nim-lang/Nim/pull/11272 ⏎ ⏎ > Other dynamic libraries (like libpg) load diffetent openSSL version then what nim loads. ⏎ > Having two different openSSL loaded version causes a crash. ⏎ ... [https://gitter.im/nim-lang/Nim?at=5e34a23d594a0517c25ca1af]
21:55:12disbot➥ Add force openSSL version with -d:sslVersion=x.x.x ; snippet at 12https://play.nim-lang.org/#ix=28PS
21:57:57*Ven`` joined #nim
22:01:33FromDiscord_<Clyybber> shashlick: You gotta put your destroy definition more to the top
22:02:08*Ven`` is now known as Ven_de_Thiel
22:02:44FromDiscord_<Clyybber> timotheecour: Here it is https://github.com/nim-lang/Nim/pull/13308
22:02:45disbot➥ Make build_all.sh more portable and a bit simpler
22:04:32FromGitter<deech> Is there a stdlib function that translates from `bool` to `cint` and vice versa?
22:05:16FromDiscord_<Clyybber> deech: Try ord
22:05:36FromDiscord_<Clyybber> someBool.ord.cint
22:05:50FromDiscord_<Clyybber> cint(ord(someBool))
22:06:11FromGitter<Clyybber> @deech
22:06:38FromGitter<deech> Nice, thanks!
22:07:21*NimBot joined #nim
22:07:24shashlickCool I'll
22:07:24FromGitter<Clyybber> np
22:08:56FromGitter<Clyybber> Cool I'll too
22:14:02*krux02 quit (Remote host closed the connection)
22:14:37*krux02 joined #nim
22:15:26shashlickWorked!
22:16:05*ptdel joined #nim
22:16:39FromGitter<Clyybber> Nice
22:18:02shashlick@timotheecour I agree with dom96 though - for the end user, it doesn't help since the version we pick at build time may not be present on his system at all
22:19:38FromDiscord_<Recruit_main_70007> is it faster to compile without gc despite warnings comming up?
22:20:22FromGitter<Clyybber> faster to compile?
22:20:25FromGitter<Clyybber> or faster to run?
22:20:54FromDiscord_<Recruit_main_70007> both
22:21:16FromGitter<Clyybber> to compile no difference
22:21:47FromGitter<Clyybber> but running it with no gc (gc:none)
22:21:57FromGitter<Clyybber> means that seqs, refs and strings will leak
22:26:25FromDiscord_<Recruit_main_70007> so, should i use none?
22:27:06FromGitter<Clyybber> nope
22:27:10FromGitter<Clyybber> you should simply use a gc
22:27:17FromGitter<Clyybber> or --gc:arc
22:27:38FromDiscord_<Recruit_main_70007> ok
22:28:23FromDiscord_<Recruit_main_70007> i think that the none garbage collector option should be improved, it would attract some C/C++ programmers
22:28:57FromGitter<Clyybber> --gc:arc is not a garbage collector
22:29:13FromGitter<Clyybber> and we are actively working on improving it :)
22:29:37FromDiscord_<Recruit_main_70007> i thought it said, a garbage collector to rule them all or something like that
22:29:55FromGitter<Clyybber> its not a real garbage collector
22:29:58FromGitter<Clyybber> its deterministic
22:30:07FromGitter<Clyybber> for refs it uses refcounting though
22:30:30FromDiscord_<Recruit_main_70007> hmm, how new is it, 1.0.4 seems to not have it
22:30:39FromGitter<timotheecour> @shashlick could we add logic to check at runtime which version of sll is loaded, then proceed accordingly using runtime dispatch
22:30:48FromGitter<Clyybber> There its probably called --gc:destructors
22:30:58FromGitter<Clyybber> arc is just an abbreviation
22:31:12FromGitter<Clyybber> but if you plan on using it
22:31:19FromGitter<Clyybber> using devel is best
22:31:44FromDiscord_<Recruit_main_70007> explain me,
22:33:18FromGitter<Clyybber> what?
22:33:29FromDiscord_<Recruit_main_70007> the dvel thing
22:33:39FromDiscord_<Clyybber> its the latest version
22:33:42Zevvyou get the latest source from github
22:33:48Zevvand build from source
22:33:59Zevvthat is what all the true hackers do
22:35:12FromDiscord_<Clyybber> what Zevv said :p
22:36:30FromDiscord_<Recruit_main_70007> i would love to, (i am a maniac of having the latest version), but the thing is that Nim is the lowest level ive got (without counting .Net since it already has all the compilers and shit needed) and i have 0 experience with C/C++ compilers, so will just wait until an stable release comes pre compiled
22:36:55*abm quit (Quit: Leaving)
22:37:02FromDiscord_<Clyybber> you don't have to have much experience
22:37:12FromDiscord_<Clyybber> you don't even have to build it
22:37:17FromDiscord_<Clyybber> one sec
22:38:34FromDiscord_<Clyybber> try this: https://github.com/nim-lang/nightlies/releases/download/2020-01-23-version-1-0-89b39ee/nim-1.0.6-windows_x64.zip
22:38:53FromDiscord_<Clyybber> or in general look here for precompiled nightlies: https://github.com/nim-lang/nightlies/releases
22:39:44FromDiscord_<Clyybber> the last ones don't seem to have windows though, and I assume you are on windows
22:39:51FromDiscord_<Clyybber> but the one I sent you should be recent enough
22:40:02FromDiscord_<Recruit_main_70007> i will have a look at them, thanks
22:40:05FromDiscord_<Clyybber> np
22:43:18*abm joined #nim
23:06:53shashlick@timotheecour - that will be a special exception for openssl
23:09:02*icebattle quit (Ping timeout: 268 seconds)
23:15:25*nsf quit (Quit: WeeChat 2.7)
23:29:36*skellock joined #nim
23:40:16*skellock quit (Quit: WeeChat 2.7)
23:43:57*theelous3 joined #nim
23:44:27*oprypin quit (Quit: Bye)
23:44:35*oprypin joined #nim
23:44:45*icebattle joined #nim
23:48:12*ng0 joined #nim
23:48:12*ng0 quit (Changing host)
23:48:12*ng0 joined #nim
23:49:10*ng0 quit (Client Quit)
23:50:58*icebattle quit (Ping timeout: 268 seconds)
23:51:52*theelous3 quit (Ping timeout: 265 seconds)