<< 24-11-2018 >>
00:04:05*kapil____ joined #nim
00:04:12*enthus1ast joined #nim
00:32:19*Ven`` joined #nim
01:11:38*zachk quit (Quit: Leaving)
01:13:30*Ven`` quit (Ping timeout: 250 seconds)
01:20:10*theelous3_ quit (Ping timeout: 268 seconds)
01:21:09*lordwarnut joined #nim
01:22:12FromGitter<Varriount> zachcarter: If I recall correctly, the type becomes opaque - it can't be referenced or modified directly.
01:32:51*theelous3_ joined #nim
02:03:19zachcarter@Varriount: good to know- thanks
02:03:48zachcarterjust not sure how it is useful in that case here
02:09:27*MightyJoe quit (Remote host closed the connection)
02:35:51*craigger_ quit (Quit: bye)
02:39:29*craigger joined #nim
02:43:17*craigger quit (Client Quit)
02:45:38*craigger joined #nim
02:55:22xacefinding getEnv was hard, by chance i was in os docs and found that it exports getEnv, it linked me to: https://nim-lang.org/docs/ospaths.html#getEnv , where i checked the source, to later find out it's include 'inlcudes/osenv'.
02:56:07leorizeit's fixed in devel
02:56:08leorizedon't worry
02:56:17leorizenow it's in `os`
02:56:20xaceah, cool :)
03:06:03*cyraxjoe joined #nim
03:07:40*banc quit (Quit: ZNC - http://znc.in)
03:11:30*lordwarnut quit (Quit: Leaving)
03:23:52*banc joined #nim
03:32:02*MightyJoe joined #nim
03:32:45*cyraxjoe quit (Ping timeout: 268 seconds)
03:33:03*MightyJoe is now known as cyraxjoe
03:38:56*vlad1777d quit (Ping timeout: 268 seconds)
03:43:08*kapil____ quit (Quit: Connection closed for inactivity)
03:45:00*theelous3_ quit (Ping timeout: 244 seconds)
04:14:53*kapil____ joined #nim
05:54:57*rlr quit (Quit: quitting)
06:04:35*dddddd quit (Remote host closed the connection)
06:05:10*rlr joined #nim
06:08:20*narimiran joined #nim
06:53:08*nsf joined #nim
06:59:20*mech422__ joined #nim
07:02:33*mech422_ quit (Ping timeout: 245 seconds)
07:13:07*kapil____ quit (Quit: Connection closed for inactivity)
08:14:10narimiran@alehander42 are you here maybe?
08:38:51*stefanos82 joined #nim
09:08:36*Vladar joined #nim
09:24:19Araqzachcarter: I'm here to talk about regions
09:53:11FromGitter<alehander42> @narimiran yes
09:53:27narimirangreat, i have some questions about gara
09:54:46narimirani started using it, for now just as "prettier `case`
09:56:37narimiranbut i didn't manage to 'translate' multiple possibilities in one case, e.g. `case of foo, bar, baz: do something`
10:09:45*kapil____ joined #nim
10:10:46narimiran@alehander42 is that possible with gara? if it is not currently, would you consider it?
10:12:51FromGitter<alehander42> Just added it, thank you for pointing it out: https://github.com/alehander42/gara/commit/bdd0f13aae49df76f56c97d5adf3388a19f13831#diff-d0513ee9b0c694f63209651ddf4bc2c7R486
10:13:18FromGitter<alehander42> it should be possible now to do something like
10:13:40FromGitter<alehander42> ```match a: ⏎ foo or bar: ⏎ do stuff ⏎ baz: ⏎ ..``` [https://gitter.im/nim-lang/Nim?at=5bf92454f59704348ef6c75d]
10:15:55narimiranand you solved my next question before i even asked (about those echoes :))
10:16:36FromGitter<alehander42> yeah, sorry about that :D, I should add a test maybe to check out the stdout
10:16:47narimiranno problem :)
10:18:06narimirani know i already said that couple of times, but i really like the syntax. i plan to use it instead of `case` in my code, even when there is no pattern matching going on
10:18:57*rlr left #nim ("Be back later...")
10:20:20FromGitter<alehander42> yeah, it's pretty nice
10:21:05FromGitter<alehander42> what's mostly left is fixing remaining bugs and improving error checking (and maybe optimizing it a bit)
10:23:04*ng0 joined #nim
10:23:05FromGitter<alehander42> e.g. currently it doesn't actually generate `case` , but sometimes it's possible, so it should do that
10:27:11narimiranyeah, sounds good
10:30:03zachcarterAraq: sweet - I'm up
10:30:06*rlr joined #nim
10:30:50zachcarterif you're still around and able to tell me more about regions
10:31:11zachcarterokay
10:31:16zachcarterI see your forum response
10:32:09*Ven`` joined #nim
10:32:15zachcarterthanks for posting that - I will play around with this.
10:45:29zachcarterreally excited about the hotreloading work
10:49:05*vlad1777d joined #nim
10:52:56*NimBot joined #nim
11:10:30Araqzachcarter: well ... anything unclear?
11:13:35*Snircle joined #nim
11:16:59*Ven`` quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
11:21:34*vlad1777d quit (Ping timeout: 246 seconds)
11:24:09*narimiran quit (Remote host closed the connection)
11:25:35FromDiscord_<niv> hi. im looking for a way to customise json string generation; specifically, i want to adjust how floats are emitted. currently i have this working: https://gist.github.com/niv/d4f2e9c5c41ee1be3d76d8bbe178d633 obviously, that's not what i want to do, since this breaks interop with any other json datatype. is there a better way to do this that i'm unaware of?
11:44:11*dddddd joined #nim
11:51:39*kungtotte quit (Quit: leaving)
11:57:33*kungtotte joined #nim
12:29:19Araqniv: looks ok to me, there are no good solutions as json doesn't really allow "patching"
12:29:42Araqbut you could also create a PR to allow json more flexibility
12:38:54*vlad1777d joined #nim
12:39:14*zachcarter quit (Ping timeout: 272 seconds)
12:51:01FromDiscord_<niv> thanks, araq. im not sure i want to PR any presentation-related PR to json, as it'll likely be a performance hit with another indirection. my use case is fringe enough that i don't think i should impose this on stdlib just because it'd make my life easier
12:52:34Araqalso try my packedjson, perhaps.
12:53:59FromDiscord_<niv> ah, will have a look, thanks. in case you are wondering, the truncated float rendering is for making diffs between changes on the source data i work less churny.
13:02:26*vlad1777d quit (Ping timeout: 250 seconds)
13:12:18*zachcarter joined #nim
13:13:25*nsf quit (Quit: WeeChat 2.3)
13:13:54*enthus1ast quit (Ping timeout: 252 seconds)
13:16:47*zachcarter quit (Ping timeout: 240 seconds)
13:21:38*zachcarter joined #nim
13:22:20zachcarterAraq: apologies - I went back to sleep. No I understand everything I think.
13:27:27zachcarterI guess even if a global is allocated using regions - it still needs to have access synchronized via locks?
13:31:00*enthus1ast joined #nim
13:43:40*wildlander quit (Quit: Konversation terminated!)
14:04:08Araqwell no.
14:04:38*theelous3_ joined #nim
14:04:44Araqthat depends on whether multiple threads can access it at the same time
14:11:25*rokups joined #nim
14:54:24*nsf joined #nim
15:01:30FromGitter<Clyybber> ```Error: no symbol to borrow from found```
15:02:21FromGitter<Clyybber> when using `==` with `{.borrow.}`
15:02:32FromGitter<Clyybber> Is that expected to work?
15:03:21*ghost64 quit (Quit: See you!)
15:05:17leorizewhat are you borrowing from?
15:05:27leorizeany snippets?
15:05:32FromGitter<Clyybber> cint
15:05:37*narimiran joined #nim
15:05:44leorizea snippet would be helpful
15:05:47FromGitter<Clyybber> yes, a second
15:07:42*ghost64 joined #nim
15:09:17FromGitter<Clyybber> Nvrmind
15:09:26FromGitter<Clyybber> I was using a template
15:09:33FromGitter<Clyybber> should have used a proc
15:09:44FromGitter<Clyybber> Thanks leorize
15:18:41narimiranAraq: version-0-19 branch `./koch boot`, iteration: 2, throws "Error: system module needs: raiseExceptionEx" - any idea what that could be and how to solve it?
15:19:50FromGitter<arnetheduck> nirimiran, I've had random errors like that caused by `choosenim` - if you have it in your path (or `.nimble/bin`), try without
15:22:16*miran joined #nim
15:23:09miranthis is on CI tests. but i think i found what is causing that
15:23:54*narimiran quit (Ping timeout: 272 seconds)
15:24:27*miran is now known as narimiran
15:24:59FromGitter<Clyybber> leorize Wierd, It works when in an extracted snippet, but not in my code
15:25:30leorizetry making a copy of your code
15:25:38leorizethen start trimming unnecessary things
15:25:46leorizeeventually you'll get to the problem
15:27:01narimirannow the question is: do we also backport https://github.com/nim-lang/Nim/commit/dfb8730f51a022ff02bdd973e6c5946cfdcdc3c6 ? (cc Araq)
15:27:10*Sembei quit (Ping timeout: 250 seconds)
15:31:29FromGitter<alehander42> Araq, did you base the effect system on a previous existing one, or is it more homegrown? I am playing with the protodesign my own effect/tag for pseudo and it got crazy pretty quickly
15:31:43FromGitter<alehander42> effect/tag system
15:31:44FromGitter<alehander42> *
15:35:07FromGitter<Clyybber> leorize I can't figure it out.
15:35:17FromGitter<Clyybber> Can you try to reproduce? https://gist.github.com/Clyybber/46a7123c65658aca15401900bce379e1
15:35:22FromGitter<alehander42> I didn't realize `ignoreEffect`, `noEffect` and `noSideEffect`should be different in such a system (as you can have effects which are not side effects )
15:35:24FromGitter<Clyybber> Importing it should be enough
15:39:31FromGitter<Clyybber> Nvrmind, only importing it doesn't trigger the error
15:46:11FromDiscord_<DanielVF> I'm running into something I don't understand
15:46:25FromDiscord_<DanielVF> I have a seq of a non-ref object
15:46:36FromDiscord_<DanielVF> I'm trying to return one of these from a proc
15:46:56narimiran@DanielVF code sample, please
15:47:17FromDiscord_<DanielVF> And I'm getting a "Error: unhandled exception: invalid object assignment [ObjectAssignmentError]"
15:47:20narimiranshort, self-contained, example we can run
15:47:23FromDiscord_<DanielVF> Code sample coming up!
15:47:39leorizeClyybber: I tried compiling your sample
15:47:44leorizeit's perfectly fine here
15:50:24FromGitter<Clyybber> Wierd, the error disappeared now, and I don't know what I changed
15:50:25FromDiscord_<niv> i have a weird issue where, on windows only, someFileStreamContainingBinaryCrap.readStr(16) would return data of length 16, but the file position would only seek +15, thus misaligning further reads. replacing filestream with memfilestream fixes it. i'm having a really hard time tracking the root cause down as it only happens when i do this with multiple files. the issue only manifests on windows, not posix.
15:50:34FromGitter<Clyybber> Perhaps my editor was writing incorrectly
15:50:51FromDiscord_<niv> any ideas/takers on what could be wrong?
15:52:49leorizeniv: maybe check out the windows implementation?
15:53:12FromDiscord_<niv> i did, it just calls fread.
15:53:18FromDiscord_<niv> hence why i'm so confused.
15:53:47leorizeif you could provide a self-contained sample, please file an issue on Github
15:54:48FromDiscord_<DanielVF> Must be some kind of compiler issue. Identical code runs fine on the nim sandbox and errors at runtime localy.
15:54:53FromDiscord_<niv> ok, so, i just tried --cpu:i386 and that "fixes" it.
15:55:59FromDiscord_<niv> im so confused.
15:56:10leorizea code sample would help. Araq is a windows user and might be able to help you
15:56:32FromDiscord_<niv> yeah, unfortunately, the sample is >200MB.
15:56:43leorizecan you create a small one?
15:56:50FromDiscord_<niv> it doesn't trigger if i give it less data
15:57:10leorizemaybe the idea of it being a Windows bug could be considered...
15:57:17FromDiscord_<niv> compiling a 32bit binary seems to not trigger it either.
15:57:31FromDiscord_<niv> but i'm not sure i should be shipping 32bit exe files.
15:57:37leorizeDanielVF: please note that the playground is using Nim 0.18.0 and not the latest 0.19.0
15:57:41leorizeso it might be a regression
16:00:01*enthus1ast quit (Ping timeout: 260 seconds)
16:00:47FromDiscord_<DanielVF> https://gist.github.com/DanielVF/8b8254600e5747d234944c37b4bfc577
16:00:59FromDiscord_<DanielVF> Good to know, leorize
16:01:13FromDiscord_<DanielVF> Here's the code, with my local error output above
16:01:29narimiranok, i got the same error
16:01:56narimiranbut before i see what's going on, do you know you can do `for i in 0 ..< size`? :)
16:02:21FromDiscord_<DanielVF> '..<' eh? That's great.
16:02:28FromDiscord_<DanielVF> I was sure there had to be a better way
16:02:38Araqnarimiran, the problem is that 0.19 can build 0.19.2 but devel can't
16:03:07Araqthere is nothing to backport
16:04:29narimiranAraq: are you sure? see lin 592 here: https://github.com/nim-lang/Nim/commit/83a36ccb6cf949c5ea164e6f47904e771da1be77#diff-215b3bc37e90d2aef8648b3948377de5R592
16:05:01narimiranDanielVF i think i found the problem, give me a sec
16:05:04leorize@niv: can you provide a code sample and some precise requirements?
16:05:18leorizelike that you'd need to feed it a file > 200MB?
16:05:44narimiranDanielVF you didn't (correctly) initialize your returning ref objects
16:07:50AraqAlehander, I read a couple of papers but there is not much to it
16:08:18FromDiscord_<DanielVF> narimiran: Doesn't creating a seq of value objects of certain size, create those objects?
16:09:00AraqI stayed away from any effect system with parametrization and went for a special type inference rule for callback parameters
16:09:30Araqthis rule has since then been re-invented by two independent groups and so it seems sound
16:09:38FromDiscord_<DanielVF> narimiran: You are right that it does work if I initialize the object
16:09:44Araqcooldome wants me to remove it though, too subtle.
16:10:15AraqHaskell was a minor source of inspiration too
16:10:28Araq("yeah... that's not how to do it...")
16:12:09FromDiscord_<DanielVF> Is there anyway to define my own proc/operator to act as a "[]" square brackets accessor?
16:12:38FromDiscord_<niv> proc `[]`*(self: MyType,..)
16:12:41*enthus1ast joined #nim
16:12:56narimiranDanielVF: `proc '[]'(a: MyType)` (change ' to backticks)
16:13:05FromDiscord_<niv> yes, discord ate the backticks in mine.
16:13:49FromDiscord_<DanielVF> Thanks!
16:14:17FromDiscord_<niv> you can actually add multiple "indices", i.e. myType[5, "test", someVar]
16:14:55FromDiscord_<niv> example, in case it's still unclear: https://github.com/niv/neverwinter.nim/blob/master/neverwinter/twoda.nim#L45
16:14:55FromDiscord_<DanielVF> Sweet.
16:15:29FromDiscord_<DanielVF> I've got the basic case of [] working.
16:15:36FromDiscord_<DanielVF> Much thanks.
16:16:32FromDiscord_<DanielVF> FYI: I've currently got the only bot written in nim in Halite.io competition
16:16:35FromDiscord_<DanielVF> https://halite.io/
16:16:53narimiranAraq: i'm not sure that @yglukhov's today's fix for #9716 will work in 0.19 branch without the change in lib/system/excpt.nim introduced here: https://github.com/nim-lang/Nim/pull/9480 (but i might be wrong, of course ;))
16:22:23Araqyeah but what devel should do
16:22:44Araqis to check whether there is a compilerProc of the name raiseExceptionEx and if not, fallback to the old version
16:22:55Araqthen devel can compile an older version of Nim
16:24:37narimiranwhat about just backporting #9480?
16:30:13Araqit's not a bugfix, it's a new feature and shouldn't be backported
16:33:00narimiranok, i'll try to fix it in my branch and see what the tests say....
16:37:19narimiranwell, the fix is much easier than i thought. i just needed to resolve the merge conflict the other way.
16:37:46narimiranAraq: can i force push to 0.19 branch? (it will change commit hash of the last commit)
16:37:47Araqare you updating the 0.19 branch?
16:37:53narimirani am
16:38:07Araqcool, we can build the installers, tonight
16:38:23narimirannice
16:38:30Araqwhat's there to "force" push?
16:39:28narimiranjust to change the way i cherry-picked yuriy's PR (to leave the original 0.19 version which has "raiseException", and not "raiseExceptionEx")
16:39:46Araqok, do it
16:41:39narimirandone. now (i hope) it should have no problem with building nim, let's see what CI says
16:41:54*enthus1ast quit (Ping timeout: 264 seconds)
16:54:56*enthus1ast joined #nim
16:55:22*Snircle quit (Quit: Textual IRC Client: www.textualapp.com)
16:59:44narimiranDanielVF: i took a look at halite. seems like fun, but i didn't see nim in available languages :/
17:03:32narimiranDanielVF: but now i see on the leaderboard that you are indeed the best nim player in the world :D
17:03:38FromDiscord_<DanielVF> Lol
17:04:04FromDiscord_<DanielVF> narimiran: They just added official support, though no code libraries
17:04:46FromDiscord_<DanielVF> narimiran: I snuck my code in before official support by writing a javascript wrapper that passes messages to the nim program
17:05:03narimiranwe should advertise that, maybe somebody else would decide to join and do it in nim, too
17:05:27*enthus1ast quit (Ping timeout: 252 seconds)
17:05:38FromDiscord_<DanielVF> narimiran: Now though, I think you just have to have a MyBot.nim file
17:05:47FromDiscord_<DanielVF> And it will compile and run it.
17:06:16FromDiscord_<DanielVF> narimiran: You get 2 seconds per turn, plus 30 seconds at startup
17:06:49FromDiscord_<DanielVF> narimiran: My bot currently play both sides of a 400 turn game in 0.45 seconds total. 😄
17:07:06FromDiscord_<DanielVF> narimiran: My bot currently plays both sides of a 400 turn game in 0.45 seconds total. 😄
17:07:58narimiranand what heuristic/logic do you use?
17:11:38FromDiscord_<DanielVF> Here's a gist with my code to talk to the game engine: https://gist.github.com/DanielVF/e7a481c402b5e6dbac1a91a2251dfb09
17:11:53FromDiscord_<DanielVF> Here's a gist with my code to talk to the halite engine: https://gist.github.com/DanielVF/e7a481c402b5e6dbac1a91a2251dfb09
17:12:19narimiran(if you edit your messages, we on IRC receive them twice (the original + edit)
17:17:34narimiranDanielVF: here is another "did you know?": you can define your arrays like this: `neighbors*: array[Direction, Cell]` and then you don't have to convert your enum to str
17:17:37narimiran*to int
17:18:31*enthus1ast joined #nim
17:23:35shashlickWhat's faster - string to enums or using tables
17:23:54FromDiscord_<DanielVF> narimiran: Thanks for the tip. This is my first nim code - it's great to learn the things I didn't know.
17:26:03narimiranthis is your first nim code? wow! keep up the good work! (from which language do you come?)
17:27:41FromDiscord_<DanielVF> I've updated the gist with an anti-collision and getting from point a to b module
17:28:14FromDiscord_<DanielVF> narimiran: I write C, Ruby, Python, JS, and PHP every week.
17:28:33FromDiscord_<DanielVF> narimiran: Plus c# most weeks, and Go most months.
17:30:53narimiranshashlick: i don't know exactly, but my experience with tables is such that i try to avoid them whenever i can
17:30:54FromDiscord_<DanielVF> narimiran: Nim really hits a sweet spot for me. I get to write C fast code with a syntax I love better than python/js and near ruby.
17:32:33GertmGo would be for the libraries. Nim for the fun. :)
17:32:45FromDiscord_<DanielVF> narimiran: I love syntax of defining procs that operate on their first argument when used in an object context.
17:33:12FromDiscord_<DanielVF> Gertm: Haha, yeah.
17:35:16GertmDanielVF: You mean Uniform Function Call Syntax ?
17:35:20GertmD has that too.
17:35:56GertmIt's a cool way of doing things, flexible.
17:39:47narimiranyeah, i use that very often and would like to see that in python too ;)
17:40:37narimiranspeaking of which, nim usually uses 2 spaces for indentation. it takes some time to get used to it, but now looking at 4 spaces feels weird :)
17:43:10narimiranDanielVF: you know about let/var blocks? and that you can define more variables of the same type in one row?
17:43:38narimiranwell, and type blocks too ;)
18:00:32narimiranDanielVF: i've sent you some comments under your gist
18:00:36zachcarterbrought the allegro5 bindings up to date
18:00:41zachcarteror at least I got them to compile and run the example
18:03:00FromDiscord_<DanielVF> narimiran - Thanks very much for the comments!
18:03:57narimiranall of those are just cosmetics, to make your code more nim-like
18:04:01AraqI still want you to use "method call syntax" :P
18:04:26Araqthere is nothing "uniform" about putting the first arg in a special position, it's the opposite of "uniform"
18:04:47Araqor maybe "chainable call syntax"
18:06:00FromDiscord_<DanielVF> I like that I can use the same proc in either style, depending on the needs of the code.
18:06:00narimiranAraq: version 0.19 now passes on travis, and on appveyor is my old friend error after xmltree.nim
18:06:30leorizeDanielVF: There's even the command call syntax
18:06:39leorizebasically paren-free syntax :P
18:07:39FromDiscord_<DanielVF> Haha, sweet.
18:08:01FromDiscord_<DanielVF> Can I do `obj.foo arg` ?
18:08:31narimirantry it ;)
18:26:57shashlickThanks narimiran
18:27:10shashlickHow to get file date in vm?
18:27:35shashlickgetFileInfo gives the importc error
18:28:24narimiranshashlick: i confess i haven't tried `strtabs`, but when i used regular tables, they were noticeably slower than using arrays (i had a situation where i had unique strings and i could easily convert them to indices)
18:31:57shashlickI'm trying to improve the performance of nimterop - lots going on there, plus running in the vm so slower regardless
18:32:19shashlickIs the vm slower than compiled? By what order of magnitude
18:38:30*stefantalpalaru joined #nim
18:48:49*leorize quit (Quit: WeeChat 2.2)
18:49:44*theelous3_ quit (Ping timeout: 268 seconds)
18:53:57FromGitter<alehander42> Araq, haha I see, i was trying to think of non-haskell sources too
18:54:11FromGitter<alehander42> i actually jumped into parametrized effects
18:54:53FromGitter<alehander42> which basically can act as compile time functions returning list of tags
18:54:59FromGitter<alehander42> but have to think about it more
19:03:40Araqshashlick, factor of 10-20
19:04:59Araqnarimiran, what's the error?
19:05:35narimiran"Command exited with code 1"
19:05:45narimiranthat's after "PASS: xmltree.nim C"
19:05:57Araqtstreams stuck?
19:06:04narimiranthis was happening for a lots of PRs in the past month or so
19:07:31*nimrod joined #nim
19:08:55narimiranit might be this: "foreign_thr" — this is what is after xmltree.nim in travis
19:09:23*nimrod quit (Client Quit)
19:09:25Araqhttps://ci.appveyor.com/project/Araq/nim/builds/20526793/tests
19:09:33Araqit's tasyncssl.nim
19:10:02Araqyou can disable it on appveyor via
19:10:05Araqdisabled: "appveyor"
19:10:14Araqin the discard spec section of the test
19:10:26narimiranok, i'll do that
19:11:32narimiranbtw, this is the first time i notice (well, you pointed me directly at it) that there are different tabs in appveyor. i was always looking only at the default one (console) :D
19:12:01Araqappveyor's UI is superior to travis'
19:13:36narimiranbtw, devel branch has no problem with that test. will we remember to enable it later? ;)
19:15:40narimiranok, i guess it is too late for me. this will be disabled only on 0.19 branch
19:15:53Araqyes, that's what I meant
19:17:07narimiranpushed a new commit
19:29:06*narimiran quit (Remote host closed the connection)
19:30:49*nsf quit (Quit: WeeChat 2.3)
19:32:26*narimiran joined #nim
19:47:55FromGitter<arnetheduck> if you're doing backports, https://github.com/nim-lang/Nim/commit/c7eba64dee9a8e686fde03ee2a2992e93a084927 fixes a pretty nasty bug
19:48:53FromGitter<arnetheduck> that one causes anything compiled with nim to explode and take up all memory when run as a child process (say, in a makefile)
19:55:39Araqhmm ok, narimiran please backport that one too then
19:57:30*wildlander joined #nim
20:09:48FromGitter<alehander42> i had a similar bug
20:10:03FromGitter<alehander42> where an exception in quit handler can create a stack explosion
20:10:09FromGitter<alehander42> not sure if it was the same
20:27:58*smt joined #nim
20:29:43*mech422 joined #nim
20:32:40*mech422__ quit (Ping timeout: 250 seconds)
20:36:52*mech422_ joined #nim
20:37:40*mech422 quit (Ping timeout: 244 seconds)
20:45:46*shpx joined #nim
20:49:27FromGitter<Varriount> Hi everyone, how are things?
20:50:46zachcarterhey @Varriount: good - how are you?
20:50:57zachcarterhave a good turkey day?
20:51:12*rokups quit (Quit: Connection closed for inactivity)
20:55:56narimiran@arnetheduck, Araq: backported
21:00:31*vlad1777d joined #nim
21:17:29*miran joined #nim
21:19:13*narimiran quit (Ping timeout: 245 seconds)
21:31:48*shpx quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
21:41:16*craigger quit (Quit: bye)
21:45:57*craigger joined #nim
21:52:11*narimiran joined #nim
21:54:44*miran quit (Ping timeout: 268 seconds)
22:02:04*zachk joined #nim
22:02:21*zachk quit (Changing host)
22:02:21*zachk joined #nim
22:21:15*mech422 joined #nim
22:24:20*mech422_ quit (Ping timeout: 268 seconds)
22:27:33*mech422 quit (Ping timeout: 245 seconds)
22:36:59*narimiran quit (Remote host closed the connection)
22:50:17*stefanos82 quit (Quit: Quitting for now...)
22:52:57*druonysus quit (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.)
23:02:07FromGitter<zacharycarter> https://imgur.com/a/oZAe5UV - latest version of bgfx
23:05:27zachcarterwould be very nice to automate these bindings though with nimgen - there are quite a few things that need fixing after c2nim though
23:14:22zachcartershashlick: maybe if I can come up with the manual steps I take to produce them - you could possibly help me produce the nimgen cfg?
23:14:48zachcarterI'm just using the C99 API at this point - and expecting the user to have the shared library in the linker's lib path
23:14:57zachcarteror search path rather
23:17:16zachcartergoing to try on my own first to reproduce my manual c2nim steps
23:19:56shashlicksure
23:27:13*Vladar quit (Remote host closed the connection)
23:50:23zachcartershashlick: it's difficult - there are quite a few enums which evaluate entries in other enums
23:50:33zachcarteractually - not really enums
23:50:39zachcarterbut stuff like -
23:51:33FromGitter<zacharycarter> ```code paste, see link``` ⏎ ⏎ Where these constants are already declared, and they might be unsigned 64 bit ints or signed 64 bit ints [https://gitter.im/nim-lang/Nim?at=5bf9e405662131389412ae51]
23:52:25FromGitter<zacharycarter> doesn't make sense to hard-code them in any automated solution
23:52:44FromGitter<zacharycarter> but I'm not sure how best to figure out whether to change that 0 or a 0'i64 or 0'u64
23:52:46FromGitter<zacharycarter> depending on context