<< 06-02-2020 >>
00:03:01*krux02 quit (Remote host closed the connection)
00:06:38FromDiscord_<Clyybber> krux02: Running `bin/nim doc --warning[LockLevel]:off --hint[Conf]:off --hint[Path]:off --hint[Processing]:off -d:boot --putenv:nimversion=1.1.1 --git.commit:devel --git.url:https://github.com/nim-lang/Nim -o:web/upload/1.1.1/json.html --index:on lib/pure/json.nim` is what happens there, so you should be able to reproduce locally via that command
00:06:47*krux02 joined #nim
00:07:05FromDiscord_<Clyybber> krux02: Heh you just missed my answer: https://irclogs.nim-lang.org/06-02-2020.html
00:07:15*ltriant_ quit (Ping timeout: 240 seconds)
00:07:52*ltriant joined #nim
00:08:19FromDiscord_<Clyybber> Its that line here: https://builds.sr.ht/~araq/job/147327#task-test-525
00:10:26krux02Clyybber: thank you
00:13:54krux02Clyybber: I fixed it
00:14:03krux02I hope CI is running through now
00:14:04krux02good night
00:14:07*krux02 quit (Remote host closed the connection)
00:14:11FromDiscord_<Clyybber> nice,
00:14:14FromDiscord_<Clyybber> good night
00:15:56*blueberrypie5 joined #nim
00:16:41*disbot_ joined #nim
00:17:45*MightyJoe joined #nim
00:19:30*WilhelmV1nWeiner joined #nim
00:21:10*voltist joined #nim
00:21:23*icebattle quit (Ping timeout: 252 seconds)
00:21:28*blueberrypie quit (Quit: Ping timeout (120 seconds))
00:21:29*disbot quit (Remote host closed the connection)
00:21:36*cyraxjoe quit (Quit: No Ping reply in 180 seconds.)
00:21:36*gsingh93 quit (Ping timeout: 272 seconds)
00:21:36*lritter quit (Ping timeout: 272 seconds)
00:21:36*WilhelmVonWeiner quit (Ping timeout: 272 seconds)
00:21:36*blueberrypie5 is now known as blueberrypie
00:24:24*gsingh93 joined #nim
00:25:14*hax-scramper quit (Ping timeout: 240 seconds)
00:25:19*ikan-keli_ quit (Quit: ZNC 1.8.x-git-125-e415d9f5 - https://znc.in)
00:25:21*MD87 quit (Read error: Connection reset by peer)
00:25:21*npgm quit (Read error: Connection reset by peer)
00:25:21*hpyc9 quit (Quit: ZNC 1.7.5 - https://znc.in)
00:25:23*noonien quit (Read error: Connection reset by peer)
00:25:23*matti quit (Read error: Connection reset by peer)
00:25:24*nuxdie quit (Read error: Connection reset by peer)
00:25:34*ikan-keli_ joined #nim
00:25:43*MD87 joined #nim
00:25:48*noonien joined #nim
00:25:49*npgm joined #nim
00:25:58*nuxdie joined #nim
00:26:02*Yardanico quit (Quit: No Ping reply in 180 seconds.)
00:26:02*hpyc9 joined #nim
00:26:06*matti joined #nim
00:26:17*disrupteq joined #nim
00:26:35*lritter joined #nim
00:27:23*gangstacat quit (Ping timeout: 272 seconds)
00:28:01*EastByte quit (Ping timeout: 272 seconds)
00:28:07*Yardanico joined #nim
00:28:11*muffindrake quit (Remote host closed the connection)
00:28:19*Hideki_ joined #nim
00:28:50*muffindrake joined #nim
00:30:09*gangstacat joined #nim
00:31:49*EastByte joined #nim
00:52:54*leorize quit (Quit: WeeChat 2.6)
00:53:43*Hideki_ quit (Remote host closed the connection)
00:55:24*arecaceae quit (Remote host closed the connection)
00:56:07*arecaceae joined #nim
00:56:29voltistAnybody know of an easy way to find all text in a string that isn't inside brackets
00:56:49voltist?
00:59:31*nullwarp quit (Quit: ZNC 1.7.5 - https://znc.in)
01:00:39*nullwarp joined #nim
01:08:01*hax-scramper joined #nim
01:09:40FromDiscord_<Winton> Spanish?
01:10:41*cornfeedhobo quit (Quit: ZNC - https://znc.in)
01:18:42*cornfeedhobo joined #nim
01:30:56*Hideki_ joined #nim
01:40:28FromGitter<deech> Araq, great talk!
01:48:11*lritter quit (Ping timeout: 260 seconds)
01:54:33*lritter joined #nim
01:56:47*Hideki_ quit (Ping timeout: 265 seconds)
01:59:28*joshbaptiste joined #nim
02:13:36*icebattle joined #nim
02:18:31*icebattle quit (Ping timeout: 260 seconds)
02:25:31*chemist69_ joined #nim
02:28:47*chemist69 quit (Ping timeout: 260 seconds)
02:32:10*s87651[m] joined #nim
02:57:19*johnd joined #nim
02:57:34*johnd is now known as Guest53574
03:01:11*Guest53574 quit (Client Quit)
03:01:40*johnd_ joined #nim
03:03:59*johnd_ is now known as JohnAD
03:04:49*JohnAD quit (Client Quit)
03:05:41*johnd_ joined #nim
03:06:00*johnd_ is now known as JohnAD
03:09:40*casaca joined #nim
03:10:49*valit joined #nim
03:22:29*lritter quit (Quit: Leaving)
03:25:30*valit quit ()
03:29:27FromGitter<iffy> @deech link?
03:29:55*hax-scramper quit (Ping timeout: 260 seconds)
03:31:04FromGitter<iffy> If I have a Future with an error, how do I get the string message out of that error? `getCurrentExceptionMsg()` doesn't apply, right? I just want to log the message
03:31:44disruptekif it raises, you ... except ValueError as e: echo e.msg
03:31:44leorize[m]@iffy https://fosdem.org/2020/schedule/event/nimmovesemantics/
03:31:50*ptdel quit (Quit: https://quassel-irc.org - Chat comfortably. Anywhere.)
03:32:28disruptekif it doesn't, you can use the /is/ operator.
03:32:56FromGitter<iffy> Thanks! (I was looking for `.msg`)
03:33:09disruptekright. 👍
03:33:11FromGitter<iffy> so `myfut.readError().msg`
03:33:37disruptekyeah, or whatever.
03:33:59disrupteki don't use that api, but.. yeah.
03:35:02*muffindrake quit (Ping timeout: 260 seconds)
03:36:32*muffindrake joined #nim
03:38:25voltistI'm trying to comment this Nim code I'm working on well because it's going to accompany a paper I'm going to publish later this year
03:38:31voltistI fear the judgement of the masses :)
03:38:44*hax-scramper joined #nim
03:39:47disruptekdude, that's awesome.
03:39:53disruptekwhat's the paper on?
03:45:27voltistIt's on simulating the nervous system of the C. elegans worm, which is entirely mapped. I'm trying to simulate the whole connectome 90% biologically accurate rather than part of it 100% accurate (like the amazing work of OpenWorm) in order to demonstrate the research potential of simulation.
03:46:05disruptekwow, amazing.
03:47:56disruptekwhen is it due out?
03:50:12*endragor joined #nim
03:50:15*letto joined #nim
03:50:44voltistA few months after I finish writing/editing, which shouldn't be to far away now since I've almost finished the programming and I just need to fine-tune/test
03:52:41disruptekthat sounds like about 8mos.
03:53:12voltistSomething like that yeah
03:53:19voltistAssuming everything goes to plan
03:55:46disrupteksuper cool.
03:56:39voltistIndeed
03:57:50voltistI'll make a web demo at some point with the JS compilation target
03:58:12voltistAnd my limited knowledge of JS/browser stuff
04:00:06disruptekpeople seem to like karax.
04:01:53voltistIf it can do simple graphics stuff then I'll make sure to give it a try
04:03:31*theelous3 quit (Ping timeout: 260 seconds)
04:03:55*arecaceae quit (Remote host closed the connection)
04:04:17*arecaceae joined #nim
04:07:45*theelous3 joined #nim
04:31:26*rockcavera quit (Remote host closed the connection)
04:31:31*dddddd quit (Ping timeout: 260 seconds)
04:33:55TangerHeya folks, I'm not sure what this error means: `Error: template instantiation too nested`. This is my code: https://play.nim-lang.org/#ix=2aSA
04:34:22TangerI think I'm maybe defining the template incorrectly
04:34:30disruptekit's just a warning, right?
04:34:44disruptekit just annoys me. doesn't prevent compilation.
04:35:14TangerHuh...
04:35:24TangerHaha, was thrown off by the "Error" label
04:40:46*nsf joined #nim
04:45:09TangerIs there a function in macros or something that gets the AST/code that a template would substitute into?
04:46:21disrupteki dunno. i've done that with macros but i guess not a template.
04:46:44TangerNo worries, thanks mate!
04:48:20disrupteklet us know what you end up using. 😉
04:49:17voltistDoes httpClient work with the JS backend?
04:50:01voltistHmmm, maybe not
04:50:17disruptekno, but there's a pr pending that provides it.
04:50:36disruptek!pull fetch js
04:50:37disbot_https://github.com/nim-lang/Nim/pull/12531 -- 3Add module jsfetch 7& 2 more...
04:51:46voltistAh neat
04:54:13voltistOh wonderful now tables isn't working
04:54:22voltistThe JS backend is giving me a wonderful time
04:57:20voltistAre tuples known to not work with the JS backend? I can't use them in tables or as sequence items
04:58:25disrupteki dunno, but it sounds surprising, alright.
05:20:21*adalricus joined #nim
05:22:59*ltriant quit (Quit: leaving)
05:33:15*leorize joined #nim
05:43:33*narimiran joined #nim
05:53:24*Hideki_ joined #nim
05:54:06*theelous3 quit (Ping timeout: 265 seconds)
05:57:59*Hideki_ quit (Ping timeout: 272 seconds)
06:28:51FromDiscord_<Chiqqum_Ngbata> Do tables become Map or a regular JS object
06:30:30leorizenope
06:31:13leorizeit's just a regular object
06:31:34FromDiscord_<Chiqqum_Ngbata> Ah. We should probably use Map eh?
06:31:40leorizenope
06:31:47FromDiscord_<Chiqqum_Ngbata> Why do you figure
06:32:01voltistI'm now having trouble getting strings from JS into Nim. I call a function from a JS file that returns a string and it returns a weird string of non-ascii chars
06:32:02leorizewe compile to js, not translating to js equivalent :P
06:32:14leorizeso we need to make sure that the code work the same way as with native targets
06:32:38leorizevoltist: tell nim that it returns a cstring
06:33:06FromDiscord_<Chiqqum_Ngbata> Is there some way to represent arbitrary-type key in a regular JS object
06:33:29voltistleorize: That works. Thanks!
06:33:54leorizeChiqqum_Ngbata: iirc there is the `js()` macro that let you do so
06:33:58leorizeI don't know where it is though
06:35:32FromDiscord_<Chiqqum_Ngbata> I mean, in JavaScript, object keys can only be string. Maps can have an arbitrary key. Kind of curious as to how it compiles; maybe I'll check it out later
06:37:47leorizethis module should have what you need I think: https://nim-lang.github.io/Nim/jsffi.html
06:58:47*hax-scramper quit (Ping timeout: 272 seconds)
06:59:26*hax-scramper joined #nim
07:05:27*hax-scramper quit (Read error: Connection reset by peer)
07:05:40*hax-scramper joined #nim
07:20:04*solitudesf joined #nim
07:24:48*hax-scramper quit (Read error: Connection reset by peer)
07:24:58*hax-scramper joined #nim
07:28:44*hax-scramper quit (Read error: Connection reset by peer)
07:28:55*hax-scramper joined #nim
07:29:44*Sembei joined #nim
07:45:14*hax-scramper quit (Ping timeout: 240 seconds)
07:45:58*hax-scramper joined #nim
07:48:24*ftsf joined #nim
08:00:00*gmpreussner quit (Quit: kthxbye)
08:02:07*ng0 quit (Quit: leaving)
08:05:07*gmpreussner joined #nim
08:07:07*endragor quit (Ping timeout: 260 seconds)
08:07:34*endragor joined #nim
08:13:23*hax-scramper quit (Read error: Connection reset by peer)
08:13:37*hax-scramper joined #nim
08:21:45*Vladar joined #nim
08:27:29*voltist quit (Remote host closed the connection)
08:28:56*floppydh joined #nim
08:28:57*hax-scramper quit (Read error: Connection reset by peer)
08:29:21*hax-scramper joined #nim
08:32:14*rhenvar joined #nim
08:32:32*Guest89399 quit (Ping timeout: 260 seconds)
08:32:32rhenvarhello
08:33:06rhenvaris this a good place to discuss support for nimble?
08:33:14*SunDwarf joined #nim
08:34:30lqdev[m]yes
08:35:53rhenvaris there an accepted way of configuring nimble install and nimble build to accept flags implicitly?
08:36:14rhenvarsuch that nimble build would do, say, nimble build --app:lib
08:37:11rhenvaror, if possible, should this be done in the Dockerfile with nimlang/nim:onbuild - if so, how can I add flags to this hook
08:39:28*pvn joined #nim
08:39:31*marmotini_ joined #nim
08:39:31*hax-scramper quit (Read error: Connection reset by peer)
08:39:35lqdev[m]it should be possible using nim.cfg or config.nims for your nim project file https://nim-lang.org/docs/nimc.html#compiler-usage-configuration-files
08:39:47*hax-scramper joined #nim
08:40:56*chemist69_ quit (Ping timeout: 246 seconds)
08:42:01*chemist69 joined #nim
08:42:56*marmotini_ quit (Read error: Connection reset by peer)
08:43:19*marmotini_ joined #nim
08:45:19rhenvarworked like a charm, thanks
09:20:17*krux02 joined #nim
09:27:46*paper_ joined #nim
09:28:47*hax-scramper quit (Read error: Connection reset by peer)
09:30:26*hax-scramper joined #nim
09:55:06*Hideki_ joined #nim
09:59:35*Hideki_ quit (Ping timeout: 260 seconds)
10:13:08*onionhammer quit (Ping timeout: 260 seconds)
10:13:24*marmotini_ quit (Remote host closed the connection)
10:13:28*natrys joined #nim
10:13:32*onionhammer joined #nim
10:14:35ZevvSoooo, how are we nimmin' today?
10:15:57*marmotini_ joined #nim
10:19:15FromGitter<Varriount> Zevv: Looking through PRs, making comments
10:19:33*couven92 joined #nim
10:19:55ZevvYeah that was my new years resolution, to do that more often :/
10:21:01FromGitter<Varriount> I usually just go over the ones proposing new language features or standard library comes
10:21:13FromGitter<Varriount> *standard library changes
10:22:21*marmotini_ quit (Ping timeout: 265 seconds)
10:25:37*marmotini_ joined #nim
10:31:05*marmotini_ quit (Remote host closed the connection)
10:31:21FromGitter<Varriount> Zevv: How is everything?
10:31:44*Vladar quit (Quit: Leaving)
10:31:51ZevvSlowly but surely recovering from the post-FOSDEM flu :)
10:32:24couven92Zevv, Huh, @PMunch also said he git sick when we got home... :O
10:32:25ZevvI felt it coming, although I did a good job of hitting the sack pretty early there was no avoiding
10:32:30couven92I am fine though :P
10:32:52Zevv*still* fine :)
10:33:08couven92Oh ye of little confidence!
10:33:31Zevvspeak to you in 10-14 days, muha!
10:40:41*marmotini_ joined #nim
10:45:02FromGitter<Varriount> Zevv: I get a sinus infection and/or cold last week
10:45:37ZevvSo, there is a strong correlation between influenza and Nim usage. This can't be right!
10:45:52FromGitter<Varriount> No coughing, just major fatigue, sinus pressure, and both a runny and stopped up nose
10:46:36ZevvYou should try C++ instead, see if that helps
10:46:56Araqhey
10:47:03FromGitter<Varriount> What, it'll confuse the viruses and bacteria away?
10:51:52*ng0 joined #nim
11:00:43FromGitter<Varriount> Araq: Any thoughts on my comments for https://github.com/nim-lang/Nim/pull/13305 ?
11:00:45disbot_➥ [pending #13303] new: typetraits.getTypeId
11:01:26*NimBot joined #nim
11:03:31Araqwhat you wrote makes sense but I fear its implementation
11:11:21*krux02 quit (Remote host closed the connection)
11:14:39*marmotini_ quit (Ping timeout: 272 seconds)
11:15:56*krux02 joined #nim
11:18:11*krux02 quit (Remote host closed the connection)
11:18:29*Ckatt joined #nim
11:21:39*Ckat quit (Ping timeout: 268 seconds)
11:21:39*Ckatt is now known as Ckat
11:26:57FromDiscord_<jjude> is it possible to change server name from httpbeast while using jester?
11:27:13FromGitter<Varriount> What do you mean by server name?
11:28:45FromDiscord_<jjude> say you write a REST server using jester; while invoking the api, jester will output content-type, content-size, and server name in addition to whatever response your REST server outputs. I want to change this server name
11:32:54*marmotini_ joined #nim
11:33:16*Zectbumo joined #nim
11:38:13*marmotin_ joined #nim
11:40:38*marmotini_ quit (Ping timeout: 260 seconds)
11:45:41*marmotini_ joined #nim
11:47:00*marmotin_ quit (Read error: Connection reset by peer)
11:47:08*marmotini_ quit (Remote host closed the connection)
11:47:13*abm joined #nim
11:49:49*marmotini_ joined #nim
11:56:18*Tongir joined #nim
12:00:15*solitudesf quit (Ping timeout: 272 seconds)
12:04:24*Hideki_ joined #nim
12:04:57*Zectbumo quit (Remote host closed the connection)
12:05:13*Zectbumo joined #nim
12:06:38*hax-scramper quit (Read error: Connection reset by peer)
12:06:54*hax-scramper joined #nim
12:16:42Zevvits a const in httpbeast, so no
12:17:20Zevvotoh, you have the source :)
12:18:24*icebattle joined #nim
12:20:41*marmotini_ quit (Remote host closed the connection)
12:23:03*icebattle quit (Ping timeout: 272 seconds)
12:25:47*Hideki_ quit (Remote host closed the connection)
12:26:37*Hideki_ joined #nim
12:26:59*PMunch_ joined #nim
12:31:15*Hideki_ quit (Ping timeout: 260 seconds)
12:46:28*paxis quit (Ping timeout: 248 seconds)
12:48:46*nsf quit (Quit: WeeChat 2.7)
12:49:03*Hideki_ joined #nim
12:53:23*endragor quit (Remote host closed the connection)
12:54:10FromGitter<alehander92> hmm
12:55:05FromGitter<alehander92> @iffy hm i didnt know about the error api thanks
13:00:41*paxis joined #nim
13:01:19*solitudesf joined #nim
13:03:01FromGitter<alehander92> anybody needs help? jjude look at the server api
13:03:03FromGitter<alehander92> e.g. asynchttpserver
13:03:07FromGitter<alehander92> or httpbeast
13:05:29*marmotini_ joined #nim
13:11:32*npgm quit (Ping timeout: 248 seconds)
13:11:41*LyndsySimon quit (Ping timeout: 265 seconds)
13:11:48FromDiscord_<highDevGuy> hello again. I'm a bit lost again. I wrote a nim script that works great on windows where I develop and test it, but now I need to run it on a linux server and I don't know how to compile it for linux and then run it there. Any help would be appreciated ❤️ Even just a link to somewhere where I could learn more
13:12:21Yardanicowell, the easiest way would be do use WSL if you run win10
13:12:29Yardanicoand what do you mean by a "nim script"?
13:12:32*d10n-work quit (Ping timeout: 260 seconds)
13:12:40Yardanicoa program written in Nim which you compile, or literally NimScript (.nims)?
13:13:38FromDiscord_<highDevGuy> 😄 I guess just a program written in nim 🙂 I'm using jester and building a simple web service (if that is the proper terminology)
13:15:01FromDiscord_<highDevGuy> and yes, I'm developing on win10 but I have no clue what it means to use WSL 🙂
13:15:04Yardanicowell yeah, then you need to compile it as a binary if you don't want to share the source code
13:15:28YardanicoWSL is windows subsystem for linux, basically a linux kernel running inside windows NT kernel, search for it on the internet, there's quite a lot of info about it
13:15:44Yardanicoit's not emulation, it's something like wine but other way
13:15:53FromDiscord_<highDevGuy> ok, thank you
13:18:28*noonien quit (Ping timeout: 248 seconds)
13:27:17*Kaivo joined #nim
13:30:44*marmotini_ quit (Remote host closed the connection)
13:32:36*marmotini_ joined #nim
13:33:18*kungtotte quit (Read error: Connection reset by peer)
13:33:52*kungtotte joined #nim
13:34:03FromDiscord_<treeform> I would just ssh into the server and compile it there.
13:34:28Yardanicowell I don't know if they have a server :)
13:34:35FromDiscord_<treeform> get choosenim working on linux, install all dependencies and compile.
13:34:42YardanicoWSL is not that hard tho
13:34:51Yardanicoit works ootb on windows with default ubuntu distro, and you can use choosenimt here just fine
13:35:06FromDiscord_<treeform> is it binary compatible with all linux distros?
13:35:33Yardanicoyes, there are custom releases for distros like Arch Linux so you can run them in WSL too
13:35:47Yardanicoof course if you want to make your binary truly compatible with all distros, you'd have to statically compile it with musl :)
13:35:49FromDiscord_<treeform> what about glibc vs muslc?
13:36:05Yardanicoalmost all modern distros are glibc, aren't they
13:36:18FromDiscord_<treeform> thats true
13:36:20Yardanicoseems like alpine works in WSL though
13:36:36Zevvping PMunch
13:36:47Yardanicoor https://github.com/am11/VoidMuslWSL for void linux with musl on WSL
13:36:55FromDiscord_<treeform> For many people linux=Ubantu
13:37:10Araqjust compile against glibc, it's the standard on Linux for a reason. nothing works with "all distros" anyway, that's the point of having so many distros.
13:37:17FromDiscord_<treeform> Ubuntu*
13:37:31FromGitter<kaushalmodi> @timotheecour Please see this: https://github.com/nim-lang/Nim/commit/b20d7e2110d3d8ab406d44f2231120847e6fea05#commitcomment-37152517 (about Travis failing on Nim devel)
13:38:09Araqthere is always something that doesn't run on distro X and "doesn't run" is equal to "secure"
13:38:24Zevv--os:any! \o/
13:38:37FromDiscord_<treeform> --os:every
13:39:05Yardanicowell I think that static musl build will work on almost any distro you'd find out there
13:39:06FromDiscord_<treeform> does --os:any means it will run on any of the OSes but it does not know which one?
13:39:10FromGitter<kaushalmodi> Araq: I don't know anything about netdial or ipv6, but that test is try to open a ::1 port which the Linux on travis does not like
13:40:07FromGitter<kaushalmodi> hmm, looks like your "doesn't run on distro X" comment was in a different context; nothing to do with that failing test :P
13:40:11Zevvtreefrom: nah, --os:any is a misleading name anyway, but my intention is to have it no dependencies on the underlying target. The original name was os:ansic, but that was voted out
13:40:22FromDiscord_<treeform> I remember when everyone was going run out of ipv4, and it was like a huge problem. Then we all got NATs and like it more now... ipv6 is not really needed.
13:40:57Yardanicowell I'd like to be able to open ports :)
13:41:11Yardanicoand host stuff without any other servers, and NATs are bad for that
13:41:12Araqipv6 is our only option for interstellar travel so that every toaster in alpha centauri can have an IP
13:41:27Zevvevery toaster in Alpha Centauri can get its own /32
13:41:50FromDiscord_<treeform> I think Alpha Centauri will be behind a NAT.
13:42:07FromGitter<alehander92> but where would we get all that bread for those toasters
13:42:26FromDiscord_<treeform> Zevv, what so --os:any is a thing, and it only works on linux?
13:42:54FromGitter<alehander92> os:any is for bare metal apps/linux kernels/toy os-es right?
13:42:59FromGitter<alehander92> linux kernel modules*
13:43:15Zevvtreeform: it should run on - tadaaa - *anything*!
13:43:33FromDiscord_<treeform> no that is --os:standalone
13:43:51Zevvit works for me on linux, on atmel AVR, in the linux kernel, on bare metal ARMs, etc. It generates the least amount of stuff, so a typical hello world is mostly nothing.
13:43:54FromGitter<alehander92> oh treeform
13:43:58FromGitter<alehander92> thats the story :)
13:44:00ZevvI aim for 0% overhead compared t C
13:44:17Zevvthis might obsolete os:standalone - in the documentation it already did
13:44:19FromGitter<alehander92> it is the better os:standalone
13:44:51FromDiscord_<treeform> Does any one have a 64 bit toy kernal in nim? Dom's is only 32bit.
13:45:19FromGitter<alehander92> this one is a bit more advanced https://github.com/samanthadoran/Mero
13:45:23FromGitter<alehander92> but also not x64
13:49:12FromDiscord_<treeform> that is more advanced.
13:51:59*dddddd joined #nim
13:52:09FromGitter<alehander92> i also started one based on Mero and nimkernel to experiment with random ideas
13:52:16FromGitter<alehander92> but have no time for it lately
13:52:33FromGitter<alehander92> plan to port it to os:any and clean it up a bit later
13:58:47*theelous3 joined #nim
14:02:11*LyndsySimon joined #nim
14:03:43*d10n-work joined #nim
14:07:38*nullwarp quit (Ping timeout: 240 seconds)
14:08:30*noonien joined #nim
14:09:16*npgm joined #nim
14:09:42*nullwarp joined #nim
14:14:23*lritter joined #nim
14:17:09*abm quit (Read error: Connection reset by peer)
14:19:14*Tongir quit (Ping timeout: 240 seconds)
14:22:52*rockcavera joined #nim
14:28:17*marmotini_ quit (Remote host closed the connection)
14:30:54*marmotini_ joined #nim
14:40:54*PMunch__ joined #nim
14:42:17*PMunch_ quit (Quit: Leaving)
14:43:39*PMunch quit (Disconnected by services)
14:43:43*PMunch__ is now known as PMunch
14:44:08*PMunch_ joined #nim
14:46:20*nsf joined #nim
14:47:17*cgfuh joined #nim
14:51:07*tefter joined #nim
15:02:03*CcxWrk quit (Ping timeout: 240 seconds)
15:03:09*CcxWrk joined #nim
15:07:17FromGitter<nixfreakz_twitter> good morning , I am looking for information on how to create a program that allows me to find a specific string 16 chars (numeric only) and grep though multiple directories
15:13:35Araqlook at nimgrep's source code?
15:17:31FromGitter<zetashift> Araq, I'm watching your talk atm and if I understand correctly if you annotate a parameter with `sink` it tells the compiler that value won't be used after it leaves the function scope and this is can be moved?
15:22:46*ftsf quit (Ping timeout: 268 seconds)
15:24:19*ng0_ joined #nim
15:24:19*ng0_ quit (Changing host)
15:24:19*ng0_ joined #nim
15:27:40*krux02 joined #nim
15:27:49*tane joined #nim
15:27:53*ng0 quit (Ping timeout: 268 seconds)
15:29:28*tane quit (Client Quit)
15:31:09*Hideki_ quit (Remote host closed the connection)
15:31:46*Hideki_ joined #nim
15:34:17*marmotini_ quit (Remote host closed the connection)
15:36:31*Hideki_ quit (Ping timeout: 260 seconds)
15:37:50*lritter quit (Quit: Leaving)
15:38:13leorize[m]that's how it works
15:39:34FromGitter<zetashift> Clears up a lot for me, can't believe I couldn't infer this from destructors.rst
15:40:03leorize[m]Yardanico: yes, a statically compiled musl libc based library will run pretty much anywhere
15:40:28*dddddd quit (Remote host closed the connection)
15:40:30leorize[m]binary*
15:43:18*sammich quit (Ping timeout: 260 seconds)
15:46:48*lritter joined #nim
15:48:02FromGitter<nixfreakz_twitter> *<Araq>* look at nimgrep's source code? , Thanks
15:50:40*Pqzcih5 joined #nim
15:51:54*sammich joined #nim
15:52:59*icebattle joined #nim
15:56:40FromGitter<zetashift> great talk anyway Araq :D
15:58:56*Vladar joined #nim
16:00:16FromGitter<nixfreakz_twitter> so with nimgrep , just trying it out and looking at source, does includeFile:<pat> support wildcards ? for instance I am looking for specific string in *.xml files
16:01:57*krux02 quit (Remote host closed the connection)
16:02:12*theelous3 quit (Read error: Connection reset by peer)
16:02:39*theelous3 joined #nim
16:13:01shashlick@leorize: no luck with package testing on travis
16:14:28lqdev[m]@nixfreakz_twitter wildcards are provided by the shell, not nimgrep. considering how such a wildcard is usually expanded, it *might* support them if it supports passing multiple `includeFile`s.
16:14:50lqdev[m]oh wait, that's not how it'd work
16:15:00*Hideki_ joined #nim
16:15:41disruptektoday we build something amazing.
16:15:46lqdev[m]looking at the source code, `includeFile` actually accepts regexes
16:16:21lqdev[m]so just use `--includeFile:".+?\.xml"
16:25:29FromGitter<alehander92> look at walkPattern docs
16:25:31FromGitter<alehander92> in the os module
16:25:37FromGitter<alehander92> NiX ^
16:28:11*avsej joined #nim
16:28:11*avsej quit (Changing host)
16:28:11*avsej joined #nim
16:28:27*Hideki_ quit (Ping timeout: 240 seconds)
16:30:39*fredrik92 joined #nim
16:33:52*couven92 quit (Ping timeout: 268 seconds)
16:35:58Araqzetashift: that's correct
16:36:50*fredrik92 is now known as couven92
16:39:09*shadowbane quit (Quit: Konversation terminated!)
16:39:47*shadowbane joined #nim
16:40:37leorizeshashlick: guess we should try something better than travis
16:41:35*hax-scramper quit (Quit: Quit)
16:41:53*hax-scramper joined #nim
16:42:19*shadowbane quit (Client Quit)
16:43:52*shadowbane joined #nim
16:44:16*hax-scramper quit (Client Quit)
16:44:32*hax-scramper joined #nim
16:47:37*shadowbane quit (Remote host closed the connection)
16:47:42*nsf quit (Quit: WeeChat 2.7)
16:48:11*shadowbane joined #nim
16:50:52*tane joined #nim
16:56:52*tane quit (Quit: Leaving)
16:57:53FromDiscord_<clyybber> ooh, we are getting a native libp2p it seems
17:14:02FromGitter<kristianmandrup> Thanks @
17:24:09shashlick@leorize we can leave the current testing as is, can you setup just package testing on azure
17:25:00disrupteki'm kinda curious what you guys are working on. i thought i understood it but now i'm not so sure.
17:25:15disruptekyou want to test nimble against important packages, right?
17:26:27leorizeyep
17:30:17disrupteki think the way it works is, nigel creates unreproducible builds that connect a unique child app with a parent app. that is, the uuid of each is encoded in the binary.
17:32:16disruptekwhen you run the child, it talks to the parent "kernel" to instantiate itself and alloc cloud resources. this way, allocators can be scoped or inherited, children can be recursively freed, and so on.
17:34:56leorizeshashlick: so how should we do package testing? for stable and devel on linux only?
17:35:39disruptekimo, you have to do it on multiple platforms.
17:37:41leorizethis is only to test nimble correctness, so I don't think we should do it for all platforms
17:37:46leorizethat task can be left to the compiler
17:38:01shashlickYa starting small
17:38:19disruptekhow does it work? you just test `nimble install`?
17:38:20shashlickWill need cross platform eventually
17:38:26shashlickAnd just devel is fine
17:38:46disruptekyeah.
17:40:04*shadowbane quit (Remote host closed the connection)
17:40:44*shadowbane joined #nim
17:41:38FromDiscord_<kodkuce> so when zero cost async, and this new gc ready 🙂
17:42:25disrupteki dunno, we never did talk about whether to add a closure cursor.
17:43:42disruptekAraq: also, i was thinking maybe we should have a way to upgrade/downgrade cursors.
17:45:00disruptekwhat's important about cursors is the guarantee, not the underlying machinery.
17:49:14*fredrik92 joined #nim
17:49:21*couven92 quit (Disconnected by services)
17:49:25*fredrik92 is now known as couven92
17:52:22FromGitter<Varriount> Can anything go wrong if the cursor pragma is misapplied?
17:52:36leorizememory leak
17:53:19FromGitter<Varriount> What about if it's not used at all?
17:53:54leorizeyou mean the variable is not used after declaration?
17:54:16Zevvunneeded copies I guess?
17:56:16*marmotini_ joined #nim
18:00:12*marmotini_ quit (Remote host closed the connection)
18:02:33*pvn quit (Quit: Leaving.)
18:03:36*marmotini_ joined #nim
18:04:05disruptekno, it's fine.
18:04:22disruptekyou can't leak memory with cursors.
18:04:51*Vladar quit (Quit: Leaving)
18:08:13ZevvI can do `f = Foo(val:1)` to make a Foo. If I want to do a new ref Foo I need to do `f = new Foo; f.val=1`. Why can I not do construction with initial values in one go without having to declare a dedicated type for `ref Foo`?
18:08:41FromGitter<nixfreakz_twitter> referring to this https://github.com/nim-lang/Nim/blob/devel/lib/pure/oswalkdir.nim
18:09:52rayman22201We only need @mratsims talk from fosdem now! Congratulations @araq excellent talk!
18:10:28leorizeZevv: make your own `new` proc, then propose it into the stdlib :)
18:11:14ZevvYeah I need it. Becasue with =destroy you must pass the base type, so for every Foo I do need to create a FooRef in able to use destructors
18:12:16*adalricus quit (Quit: ERC (IRC client for Emacs 26.3))
18:12:20leorizeI think not having an easy way to define a finalizer is an oversight that should be fixed
18:14:29*marmotini_ quit (Remote host closed the connection)
18:17:27leorizeZevv: https://play.nim-lang.org/#ix=2aVo
18:18:30*marmotini_ joined #nim
18:18:33Zevvyes but the problem is the other way around, you're solving the other end
18:18:36Zevvlemme write it up
18:19:07Zevvnot completely true though
18:20:31Zevvthere is two problems:
18:20:40Zevv- you can't create a destructor for a ref type
18:20:58Zevv- there is no shortcut for initializing a ref type (which your example fixes)
18:21:56Zevvso your thing helps with the other problem
18:22:02Zevvbut still I need an additional type
18:23:48*floppydh quit (Quit: WeeChat 2.7)
18:23:53leorizeyep, we should ping Araq on this
18:24:18leorizethere should be an easy syntax to declare destructors or finalizers for ref types
18:25:36*Hideki_ joined #nim
18:26:27ZevvAnd while we're at it, I have yet another arc question because I do not understand stuff: This leaks: http://ix.io/2aVq
18:26:56leorizeglobal variables are not destroyed
18:27:11Zevvnot true. The 'two' destroy gets called. But not the ones.
18:27:25disruptekwell, you don't disarm them.
18:27:40leorizelooks like you found a bug
18:27:47Zevvno, I don't disarm them
18:28:47disruptekyes, and you need to.
18:29:18Zevvsomehow it makes sense, but somehow it still feels alien
18:30:02*icebattle quit (Ping timeout: 240 seconds)
18:30:07*Hideki_ quit (Ping timeout: 260 seconds)
18:30:18disruptekit's kinda like "cooperative". you sorta share some of the responsibility with arc.
18:32:02Zevvsure, I get that. But it feels not intuitive that the fact that having a destuctor for a type changes the cleanup behaveour
18:32:06Zevvof the other
18:34:14disrupteki guess the key here is that you have to explicitly set those refs to nil (read: disarm) because simply zero'ing out the memory of Two isn't enough to show arc that the refs shall be destroyed.
18:34:47disruptekthat's the only association that exists, right?
18:34:58Zevvyes - but
18:35:03*marmotini_ quit (Read error: Connection reset by peer)
18:35:21Zevvif two does not have a destructor, the default behaviour is that under water it gets zeroed out
18:35:28Zevvso that the references to one go away
18:35:37Zevvbut if I *do* have a destructor, it does not do that
18:35:59*marmotini_ joined #nim
18:36:05disruptekyeah; because you wouldn't write a destructor unless you had specific behavior you wanted to happen.
18:36:16disruptekby creating the proc, it becomes your problem to solve.
18:36:23ZevvThe reason I have a destructor, in practice, is that I usually want to do work to cleanup stuff. Closing a socket, saying by to the other end, stuff like that
18:36:48leorize> A =destroy hook frees the object's associated memory and releases other associated resources. Variables are destroyed via this hook when they go out of scope or when the routine they were declared in is about to return.
18:36:50ZevvI do not want to take care to clean out every pointer there. Is there a good technical reason why the default behaviour can not also apply when I *do* have a destructor
18:36:53*icebattle joined #nim
18:37:07leorizeso if you declare a destroy hook it becomes your problem
18:37:31disruptekthere's a performance issue, for one thing.
18:37:33Zevvleorize: where does it say that in the contract?
18:38:34disrupteki really don't think it makes sense to confuse the issue by trying to automatically run the default destructor.
18:38:40disruptekyou can always call it yourself, right?
18:38:42leorizeZevv: > For object however, the compiler generated hooks can be overridden. This can also be important to use an alternative traversal of the involved datastructure that is more efficient or in order to avoid deep recursions.
18:38:47leorizethis is by design
18:39:19disruptekit's hard to see how it could work otherwise, honestly.
18:39:33leorizesee https://nim-lang.github.io/Nim/destructors#hook-lifting
18:40:46Zevvleorize: I read that, but I was too naive - I assumed the 'hook' was there for my convenience, as if to use as a regular destructor. I can do some of my own stuff in there, but Nim will make sure nothing leaks in the end
18:41:21Zevvbut now I am responsible for that. As if I still need to free() my C allocations
18:42:03disruptekdid you try calling the default destructor?
18:42:11ZevvHow does one do that?
18:42:24disruptek`system.destroy=`()?
18:42:35leorizenah, it doesn't work like that
18:42:38disrupteksomething to that effect?
18:42:49Zevvnope
18:43:03ZevvError: call depth limit reached in a debug build (2000 function calls).
18:43:06Zevvyou broke my computer
18:43:11*tane joined #nim
18:43:44disruptekthat could be a bug, but let's consult the great tomb of arc wisdom.
18:43:48*marmotin_ joined #nim
18:43:53disruptektome, too.
18:43:57*nsf joined #nim
18:44:03leorizeshashlick: you need to add nimble to the nim project on azure pipelines
18:44:09leorizenarimiran could do that I think
18:44:13Zevvanyway, it is not that I do not understand all this - I'm doing the devils advocate thing. You give the user a tool to do cleanup, but this brings responsibility again. That needs to be very thoroughly documented. Saying you do GC of refcounting makes people thing everything will be all right in the end
18:44:23*marmotini_ quit (Read error: Connection reset by peer)
18:44:26disruptekyou're not wrong.
18:45:12Zevvbut I'm not sure how much of this is all final design
18:45:28shashlick@leorize: please use my updated script from my PR
18:45:28ZevvI asked some things earlier and got answers indicating things might change
18:45:29*marmotini_ joined #nim
18:46:04*marmotin_ quit (Read error: Connection reset by peer)
18:46:20*marmotini_ quit (Read error: Connection reset by peer)
18:46:54disruptekthere could be a pragma macro added to impl the default destructor; you'd attach this to your existing destructor.
18:47:14Zevvwould the other way around not be the right way?
18:47:25disruptekwhat?
18:47:28ZevvIf you have nothing to clean up, there is no need to put in a destructor anyway
18:47:46disruptekwell, it won't hurt in that event.
18:47:47Zevvbut if you want a destructor, you do that to cleanup resources 99% of the time
18:48:02ZevvBut still I would very much like Nim to take care of the memory for me
18:48:13Zevvand not ask me to nil my pointers.
18:48:15disruptekright. that's what the macro does.
18:48:24disruptekanything you haven't disarmed, it will disarm at the end of the proc.
18:48:34ZevvDo that by default!
18:48:42Zevv*unless* I ask not to
18:48:43disruptekfeels too implicit to me.
18:49:02disruptekhow do i escape that, for example?
18:49:07Zevvwith your macro!
18:49:14disruptekhmmph.
18:49:15Zevv{.letMeDoThatMyselfPlease.}
18:49:20disruptekbananas.
18:49:22Zevvbecause who is ever going to use that
18:49:34leorizeshashlick: why do you need portable git?
18:49:40ZevvHow often do you make data stuctures that are so nasty that you can not trust the default arc to clean them up for you?
18:49:59Zevvand arc can do that perfectly fine, but you might be able to do that more effectively.
18:50:13disruptekwe're talking about a proc which literally wouldn't exist if that were the case, so it's kinda pointless to consider that scenario.
18:50:14ZevvSo that's an optimization. If you don't ask for it, stuff works just fine and it all gets cleaned up
18:50:16disruptekit's not part of the problem domain.
18:50:43Zevv?
18:50:50disrupteki would just rather see a pragma that represents code versus seeing no pragma and assuming magic.
18:51:10Zevvhow is this magic? If I do *not* specify my own destructor, it also works. Is that magic?
18:51:27Zevvno, that is what I would expect from a pragmatic point of view: clean up my shit after me Nim please thank you
18:51:27disruptekwell, it varies with the gc, doesn't it?
18:51:36leorizeshashlick: also, environment variables don't passthrough between steps in azure-pipelines, so I can't use that script
18:52:14ZevvOk again - when and why does one need destructors/finalizers?
18:52:17disruptekyou might be right, zevv.
18:52:30shashlickthat's fine - portable git is because cmake fails if the path to git (which carries bash.exe) has spaces in it
18:52:42disruptekZevv: araq's favorite example is auto-closing files.
18:53:00Zevvdisruptek: is that a "can you please shut the f up" kind of "you may be right"?
18:53:05*marmotini_ joined #nim
18:53:14Zevvright: that is what I mean. That is what your everyday programmer uses those for
18:53:18shashlickleorize: also i confused your azure yml with travis
18:53:30Zevvclose a file. Close a socket. Delete a file from disk. stuff like that
18:53:35disruptekno, you're kinda convincing me simply because it should be semantically similar/identical across GCs.
18:53:50ZevvBut when I want to use it for that stuff, I suddenly am resonsible for managing memory - which otherwise was never my problem!
18:54:09disruptekyeah, i heard you the first time. 😄
18:55:13ZevvYeah but I was still in WHY DOES HE NOT UNDERSTAND ME mode
18:55:27Zevvtakes a few sentences to cool down from that
18:55:32disruptekeveryone makes mistakes. 😉
18:55:46disrupteki just think that argument trumps mine.
18:56:17Zevvwell, anyway, it is nice to know that I'm not being just plain stupid
18:56:27disrupteknot /this/ time.
18:57:20ZevvI think this will just need some time to get the dust settled, but I will do some inquiring with araq one day to see how this is supposed to work in the end
18:59:26leorizeshashlick: https://dev.azure.com/alaviss/Nim/_build/results?buildId=277&view=results
18:59:33leorizehopefully it finishes :P
18:59:57shashlicknice you got it working 🙂
19:00:38*Pqzcih5 quit (Remote host closed the connection)
19:11:33*icebattle quit (Ping timeout: 260 seconds)
19:14:13AraqZevv: I'm back
19:14:42*marmotini_ quit (Remote host closed the connection)
19:15:06ZevvHi there. So I was rambling about, not sure if you care to read all that back.
19:16:01AraqI did read it but I'm not sure I understood it
19:16:19Zevvsorry about that :)
19:18:05*icebattle joined #nim
19:18:15ZevvIts a usability thing with finalizers. When I make a finalizer for an object, I get the responsibility to disarm anyting that might be in there.
19:18:36Araqhmm tell me more
19:18:41ZevvIn practice though, regular users will use destructors mainly for cleanup tasks like sockets and files. But now they also get the responsibility for cleaning up pointers.
19:19:01*marmotini_ joined #nim
19:19:02ZevvIf I do not make an explicit =destroy for an object, Nim will do the right thing and clean up my memory for me
19:19:18Zevvif I *do* make an =destroy, it will often be for pragmatical reasons of RAII
19:19:28Araqit's for x in fields(obj): `=destroy`(x)
19:19:43Araqinside your custom =destroy, but yeah
19:19:49Zevvsure. But now that has to go in every destroy.
19:19:58ZevvWhy could nim not do that for my anyway?
19:20:03Araqcooldome's patch is overly aggressive
19:20:27ZevvI do see that in some cases complex data structures can be cleaned up better with custom algorithms, but I guess that for the majority of the use cases Nim should just do The Right Thing here
19:20:33Araqin my mind the body should be close(self.file); system.`=destroy`(self)
19:20:53ZevvIn my mind the body should be `close(self.file)` only
19:21:09Araqnah, that's a pretty bad solution
19:21:21Zevvwhat use case do you see where I would *not* want to do a system.=destroy(self)?
19:21:22Araqbecause then you lose the ability to do custom traversals
19:21:31ZevvBut these are the exceptions. Make a pragma for that
19:21:42ZevvNow you put the burden on everyone who wants to use a destructor
19:21:45Zevvand the price is high
19:21:54disruptekthe price is one line of explicit code.
19:22:03ZevvYes, I mean the price of not putting that in
19:22:17Araqwell I disagree for two reasons:
19:22:23disruptekshould we warn on destructors that have no pragma macro for default cleanup?
19:22:28disruptekif not, then it's no different.
19:22:28Araq1. custom destructors should be rare
19:23:08Araq2. trying to do the "best" thing and then opting-out with yet another pragma is 2 features where previously we only had 1
19:23:18FromDiscord_<sveri> Hi, I am trying to get into nim and wanted to parse some yaml. I found this library: https://nimyaml.org/api.html and installed it with `nimble install yaml`
19:23:18FromDiscord_<sveri> Now when I try to use that and import it with `import yaml/serialization, streams` I get an error while compiling:
19:23:18FromDiscord_<sveri> `specparser.nim(6, 12) Error: cannot open file: yaml/serialization`
19:24:10Zevvfair enought. But up to now I could always trust Nims memory management to do the right thing
19:24:26Zevvand now - because I want to close a file - I need to be explicit about these things, or leak
19:24:40disruptekyeah?
19:24:47*marmotini_ quit (Remote host closed the connection)
19:24:57disruptekwhy do you want to leak so bad, zevv?
19:25:02disrupteklotta drinking?
19:25:05Araqsveri: I'm not sure either, I'm not familiar with yaml
19:25:19AraqZevv: there is one good argument in your favour though
19:25:21Zevvdisruptek: pay attention dude, I do *not* want to leak
19:25:26Araqand that is that we already have the pragma
19:25:39Araqand it's called .nodestroy to skip destructor injections
19:25:51Araqand already you kinda need it for custom traversals...
19:26:04Zevvright
19:26:52Araqstill not sure if there is a real problem here. if you do close(x.file) in =destroy you already have a weak abstraction
19:27:07Zevvalso - my final reason: it is just surprising. I do not have a destructor, everything gets cleaned up. I add a destructor to close a file, and suddenly stuff leaks * that did not leak before that *
19:27:13Araqas the close should be in the =destroy for the File object itself
19:28:14Zevvwell, maybe I want to remove a temp file from disk, or unlock a mutex, or remove a shmem IPC, whatever
19:28:35Zevvthe fact is that first there was no leak, and now there is. It catches me by surprise
19:31:57AraqI understand
19:31:59disruptekit'll be surprising if it doesn't leak.
19:32:03Araqbut do you understand my point?
19:32:06*nsf quit (Quit: WeeChat 2.7)
19:32:41Araqthe way this system is setup is that you have an AutoFile that has a =destroy
19:32:42Zevvsure I do
19:33:32Zevvbut people will go use this for other things, even if it is bad design
19:33:55disrupteknim is pointy, yes.
19:34:39Araqwell I know what C++ does
19:34:54AraqI'm not sure what Rust does, somebody should find out for us
19:35:07Araqsame for Swift which also has this feature
19:35:37Araqif we do it like the 3 major competitors it's fair to not expect too many complaints about it :P
19:35:58Zevvanyway, just my two cents
19:36:08FromDiscord_<sveri> Araq: do you have an idea how I can figure out what the problem is?
19:36:08Araqyeah, yeah, "copying bad design is not good design", but what if we don't know what good design looks like ;-)
19:36:32Araqsveri: which OS are you on?
19:36:39FromDiscord_<sveri> Windows
19:36:51Araqok, one sec
19:38:42disrupteki'm still not convinced it can even work if you auto-invoke a default destructor.
19:39:29Araqsveri: works for me
19:39:47Araqmaybe you have a problem with your configuration?
19:39:56Araqyou can also do
19:40:07*solitudesf- joined #nim
19:40:30Araqnim c --path:here_is_the_fun/.nimble/pkgs/yaml-0.12.0 mymodule.nim
19:40:47FromDiscord_<sveri> Araq: I will try that, thank you
19:42:17disruptekwhat about a hook that runs at the top of the default destructor?
19:42:48disruptekthen you can define `=alsodestroy`() for file.close and rely on the normal object teardown for the remainder.
19:43:08Araqdisruptek, Zevv I think it's fine to say that right now we need another round of bugfixes before we can tweak the existing language rules
19:43:13*solitudesf quit (Ping timeout: 272 seconds)
19:43:24Zevvfair enough :)
19:43:33disruptekyeah, i'm not worried.
19:48:51FromDiscord_<sveri> Araq: it works using nim. But running nimble test it cannot find the yaml parser. When I start the tests with `nim c -r tests` it works.
19:48:51FromDiscord_<sveri> Does nimble have known problems on windows? Is it better to just use nim for running stuff in a project / library?
19:50:44*dddddd joined #nim
19:54:11Zevvdisruptek: are you ever?
19:54:19shashlicksveri: what does `nimble path yaml` say
19:54:28shashlickalso what's the directory contents of that path
19:54:40shashlicki just tried as well and it works fine
19:54:41shashlickw
19:54:47shashlickversions of nim/nimble will help
19:55:52FromDiscord_<sveri> ```
19:55:53FromDiscord_<sveri> nimble path yaml
19:55:53FromDiscord_<sveri> C:\Users\sveri\.nimble\pkgs\yaml-0.12.0
19:55:53FromDiscord_<sveri> nimble -v
19:55:53FromDiscord_<sveri> nimble v0.11.0 compiled at 2020-01-23 17:35:57
19:55:53FromDiscord_<sveri> git hash: 4007b2a778429a978e12307bf13a038029b4c4d9
19:55:53FromDiscord_<sveri> ```
19:55:58shashlick~paste
19:55:59disbot_paste: 11a frowned-upon behavior in chat; please use a service such as https://play.nim-lang.org/ or http://ix.io/ or https://gist.github.com/ and supply us a URL instead.
19:56:02FromDiscord_<sveri> The path resolves correctly, the package exists
19:56:44shashlickwhat is in `C:\Users\sveri\.nimble\pkgs\yaml-0.12.0\yaml`
19:57:40*ptdel joined #nim
19:57:59FromDiscord_<sveri> https://paste.ee/p/r4u6F
20:01:15shashlickokay, did you add `yaml` to your `.nimble` file with a `requires` statement?
20:15:37FromDiscord_<sveri> ah no, I did not do that. I thought it does that automatically.
20:23:02*Zectbumo quit (Remote host closed the connection)
20:23:18shashlickwithout that, nimble won't include it in the requirements
20:26:50*letto quit (Quit: Konversation terminated!)
20:27:32*letto joined #nim
20:31:23*Zectbumo joined #nim
20:37:38Zevv1
20:37:38Zevv1
20:37:51*dadada joined #nim
20:37:57dadadakotlin has some nice features, positively surprised as I thought all the hype around that lang was mostly buzz
20:38:01dadadathere's still a significant number of things I prefer about nim
20:38:06dadadabut if I had to do something with JVM I think I wouldn't mind kotlin
20:47:43PMunchClojure is another thing to try on the JVM
20:47:47PMunchReally nice language
20:47:54dadadaI think this introductory video by the lang's dev is well worth watching: https://www.youtube.com/watch?v=6P20npkvcb8
20:47:57disrupteknim'll run in the jvm via graal.
20:48:09dadadadisruptek: cool
20:48:24PMunchWhy though?
20:48:48disruptekinterop with other graal languages.
20:49:00PMunchAaah rigt
20:49:12FromDiscord_<sveri> shashlick: that did it, thank you 🙂 In hindsight it's obvious, but the docs said that nimble install is enough.
20:49:19dadadaPMunch: I have a weak spot for anything lisp like :D
20:51:20PMunchSame here, just don't tend to use them much :P Had a look at Janet recently: https://janet-lang.org/
20:51:27PMunchLooks a lot like Clojure
20:51:37*krux02 joined #nim
20:51:39*revere quit (*.net *.split)
20:51:39*d-nice2[m] quit (*.net *.split)
20:51:40*planetis[m] quit (*.net *.split)
20:51:54*revere joined #nim
20:54:03*gsingh93 quit (Ping timeout: 240 seconds)
20:55:29*MyMind joined #nim
20:55:47disruptek~motd is janet versus fennel... fight!
20:55:47disbot_motd: 11janet versus fennel... fight!
20:56:35*gsingh93 joined #nim
20:57:11*Sembei quit (Ping timeout: 268 seconds)
21:02:15*narimiran quit (Ping timeout: 260 seconds)
21:26:47*oculux quit (Ping timeout: 268 seconds)
21:28:25FromDiscord_<Auriel> didn't use fennel yet but I think it would have a case of usage for me for scripts in one game; the output is clearly readable so it could be easily used if somebody prefer lua
21:30:56*oculux joined #nim
21:38:29*oculux quit (Ping timeout: 272 seconds)
21:39:51*oculux joined #nim
21:40:46disruptekwell, that's just it, right?
21:41:15disruptekisn't #13203 actually fixed now?
21:41:17disbot_https://github.com/nim-lang/Nim/issues/13203 -- 3realloc() performance: -d:useMalloc vs Nim default allocator ; snippet at 12https://play.nim-lang.org/#ix=2aWu
21:46:46*tiorock joined #nim
21:46:46*tiorock quit (Changing host)
21:46:46*tiorock joined #nim
21:46:46*rockcavera is now known as Guest17347
21:46:46*tiorock is now known as rockcavera
21:50:39*Guest17347 quit (Ping timeout: 268 seconds)
21:53:28Zevvmooi man bij $klant krijgt iedereen een mooie bluetooth keyfinder
21:53:34Zevvdan moet je een app installeren van 23 MB
21:53:51Zevvik heb nu een scriptje waarmee ik alle keychains in het pand kan laten bleeepen op de alarmstand
21:53:55Zevven een pret dat we hebben
21:54:18ZevvOops, EWRONGWIN again :/
21:54:19disrupteki don't believe it.
21:55:07Zevvof course you don't
21:55:35*nullwarp quit (Ping timeout: 272 seconds)
21:56:20Zevvbut to stay on topic - what would have fixed 13203 then?
21:56:36disrupteki thought you made a realloc.
21:56:49*PMunch quit (Quit: leaving)
21:57:05Zevvnope. my realloc is the -d:useMalloc realloc
21:57:16Zevvbut the default Nim allocator does not properly relalloc
21:57:33disruptekeven in refc?
21:57:46Zevvthere is simply no efficient realloc implementation in the allocator
21:57:57Zevvit falls back to malloc()/memcpy()/free()
22:00:39disruptekaha, i'm an idiot. i get it.
22:06:09*arecaceae quit (Remote host closed the connection)
22:07:04*arecaceae joined #nim
22:08:07Araqthere is one for TLSF
22:08:18Araqbut before ARC we couldn't use it anyway
22:08:37Araqif I find the algorithm again, we can port it
22:08:53disruptekneat.
22:16:06*nullwarp joined #nim
22:17:05leorizehow does devel even pass CI?
22:17:07leorizehttps://dev.azure.com/alaviss/Nim/_build/results?buildId=280&view=logs&j=12f1170f-54f2-53f3-20dd-22fc7dff55f9&t=356bb04c-cb4a-5f04-82ca-d3b102917eba
22:18:01leorizeI'm adding important packages testing to nimble and I can't make it pass
22:19:04disrupteksumthin's diff'ent.
22:21:24leorizeand even more interesting it seems, I could replicate the results easily
22:22:25*nullwarp_ joined #nim
22:24:07*nullwarp quit (Ping timeout: 240 seconds)
22:24:55*natrys quit (Quit: natrys)
22:27:23*Hideki_ joined #nim
22:28:06dadadadon't know if you know, but I just tested it, and you can apparently always get the latest version of nim with homebrew
22:28:16dadadawhich I'm using on linux
22:28:45dadadamy distro Fedora also packages Nim, but they take a little longer than homebrew for getting updates
22:29:01dadadaFedora's nim version is at 1.0.4, while homebrew has 1.0.6
22:29:04*solitudesf- quit (Ping timeout: 268 seconds)
22:29:40dadadainstall brew is basically a one-liner and then some changes to PATH
22:29:51dadadas/install/installing
22:29:59dadadathen
22:30:02dadadabrew install nim
22:30:04dadadaand you're done
22:30:26*oculux quit (Ping timeout: 240 seconds)
22:30:36leorizeusually we use choosenim :P
22:30:37dadadabrew upgrade nim
22:30:48dadadafor always getting the latest and greatest
22:31:27*Hideki_ quit (Ping timeout: 240 seconds)
22:31:41dadadahomebrew works on all linux distros and macos :D probably also Window's subsystem for Linux, but I've not investigated this
22:31:43leorizeshashlick: I figured out the problem
22:31:54leorizethe lack of CI for nimble is the problem
22:32:24leorizenimble master branch has been broken since after 0.11.0
22:32:38leorizeit might be more recent but I can't bisect rn
22:33:02dadadaleorize: wanted to present a great alternative, of course you can use homebrew for much more than just install Nim
22:34:21*Zectbumo quit (Remote host closed the connection)
22:35:51*tane quit (Quit: Leaving)
22:38:22*krux02 quit (Remote host closed the connection)
22:38:36*nullwarp_ is now known as nullwarp
22:39:34leorize[m]yea, though I'm not a fan of running more than one package manager in a system
22:40:29disruptekoh, i dunno... i was sad that i couldn't get guix working.
22:40:43disrupteki'd like to use it for my toolchains.
22:41:48dadadahomebrew restricts itself to a hidden directory in your home folder, which why it has very little chance of creating a mess, you could simply wipe it with one simple rm -rf
22:42:03dadadawhich is
22:46:46*shadowbane quit (Quit: Konversation terminated!)
22:50:02*shadowbane joined #nim
22:50:36dadadadisruptek: homebrew is dead simple to get working
22:51:01disruptekyeah, i've used it for years.
22:51:10disruptekbut homebrew != nix/guix.
22:51:55*acidx quit (Ping timeout: 260 seconds)
22:52:00dadadado you know about bedrocklinux?
22:52:35dadadaI've not yet tried it, but it's an intriguing project
22:52:50*acidx joined #nim
22:54:40disrupteknope. what's it about?
22:56:14dadadahave you ever felt that there are multiple good distributions with pros and cons, but you have to make a decision on what to settle for?
22:56:21dadadait's the solution to that issue
22:57:17dadadamaybe not a perfect solution, but pretty cool "meta distribution", which can hijack your running distro to become "bedrocklinux"
22:57:29disruptekhmm.
22:57:55disruptekis it container-based?
22:57:58dadadano
22:59:32disruptekhow much memory do you expect llvm to consume during a linux build?
23:02:42dadadadisruptek: it's an unrelated question, right?
23:02:47disruptekyep.
23:04:25dadadamore than 64k, I can not give a serious answer to this question :D
23:04:57disrupteki'm at about 40gig and, i dunno, that's starting to feel like real memory.
23:06:07dadadatrue, I think it reserves more than it needs to use
23:07:14dadadacan linux be compiled with llvm now? it didn't use to work, special extensions to gcc and all
23:07:28disruptekdunno.
23:08:29FromDiscord_<Faster Speeding> anybody got an idea what would be causing `Error: internal error: genTypeInfo(tyNil)`?
23:10:27FromDiscord_<Faster Speeding> https://gist.github.com/FasterSpeeding/09e6a28a08e5c50a30433ff509e6b2ed is the generated traceback
23:16:07*rhenvar quit (Quit: Leaving)
23:19:11leorize[m]disruptek: llvm took 2-4gb to build
23:19:37disruptekthis is working, which is perhaps even more surprising.
23:19:53leorize[m]@Faster Speeding can you file an issue on github? internal errors are bugs
23:20:11FromDiscord_<Faster Speeding> ok, will do thx
23:27:18shashlickLeorize: what's the bug in nimble
23:34:17*dawkot joined #nim
23:34:53*dawkot left #nim (#nim)
23:56:10disruptekno alarms and no surprises.
23:56:16*couven92 quit (Quit: Client Disconnecting)