<< 31-12-2022 >>
00:00:03*neceve quit (Quit: ZNC - https://znc.in)
00:03:40*wallabra quit (Ping timeout: 272 seconds)
00:06:12*nick2 joined #nim
00:06:47*azimut quit (Ping timeout: 255 seconds)
00:35:29*dom96 quit (Quit: Nim is King)
01:01:36*nick2 quit (Quit: WeeChat 3.7.1)
01:08:10NimEventerNew thread by Akito: Async Socket with SSL Context throws "error:1420C114:SSL routines:ssl_write_internal:uninitialized", see https://forum.nim-lang.org/t/9783
03:07:01FromDiscord<_ 🎄> I defined my ast as a bunch of nested objects but idk if this is necessarily the best way
03:07:48FromDiscord<_ 🎄> I may need to refactor
03:40:50FromDiscord<jos> how can i read a list returned from a c ffi
03:41:04FromDiscord<jos> like the normal `Element firstElm; int numElms;` type thing
04:07:08FromDiscord<Elegantbeef> `ptr UncheckedArray[T]`
04:12:16FromDiscord<jos> thanks!
04:12:18FromDiscord<jos> it's working
04:14:25*arkurious quit (Quit: Leaving)
04:20:04*Batzy quit (Quit: https://quassel-irc.org - Chat comfortably. Anywhere.)
04:20:32*Batzy joined #nim
04:20:52*Batzy quit (Client Quit)
04:21:15*Batzy joined #nim
04:30:09*azimut joined #nim
04:36:50FromDiscord<that_dude> Is end actually used for anything? I don't remember it being used https://media.discordapp.net/attachments/371759389889003532/1058604621536444518/image.png
04:37:39FromDiscord<that_dude> I did find a forum post talking about source code filters, but given it's the first time I'm hearing about it, is it commonly used?
04:40:48FromDiscord<Bung> identifier preserved aren't checked everywhere.
04:58:00FromDiscord<jos> i have a weird issue with a variable captured in a closure
04:58:10FromDiscord<jos> sent a code paste, see https://play.nim-lang.org/#ix=4k0Y
04:58:39FromDiscord<jos> accessing any captured var in `requestAdapaterCallback` gives me an `EXC_BAD_ACCESS`
04:59:02FromDiscord<jos> i'm sure it's not off the stack-- `wgpuInstanceRequestAdapter` invokes the callback before it exits
04:59:26FromDiscord<jos> does this have to do with the c ffi or something, because i can't replicate with a trivial case
05:49:32FromDiscord<Elegantbeef> `requestAdtaperCallback` should be probably `cdecl` or similar @jos
05:53:32*azimut quit (Remote host closed the connection)
05:54:18*azimut joined #nim
06:02:55FromDiscord<jos> https://forum.nim-lang.org/t/1488
06:03:00FromDiscord<jos> i found this which is pretty cool
06:03:10FromDiscord<jos> In reply to @Elegantbeef "`requestAdtaperCallback` should be probably": i did exportc
06:03:48FromDiscord<jos> i think its a limitation of the language impl, the link above explains i think
06:04:49FromDiscord<Elegantbeef> exportC isnt a calling convention
06:05:01FromDiscord<Elegantbeef> is `test` global or local?
06:06:17FromDiscord<jos> local
06:06:25FromDiscord<jos> i just realized that was unclear, sorry
06:07:10FromDiscord<Elegantbeef> what's the typedefinition for the parameter you're passing that to?
06:07:15FromDiscord<jos> yeah setting it to cdecl doesn't even allow the capture to happen, which makes sense
06:07:21FromDiscord<jos> i can pass it in via userdata instead
06:07:26FromDiscord<Elegantbeef> I have a feeling it's supposed to be `cdecl` or `nodecl`
06:07:34FromDiscord<jos> i was writing my own ffi, i changed it to cdecl
06:07:45FromDiscord<Elegantbeef> Ah yea that seems right
06:07:45FromDiscord<jos> it's working now, without the capture (which makes sense)-- really appreciate the help
06:08:18FromDiscord<Elegantbeef> I assume you resolved it before i tried to help, so i guess cheers 😛
06:22:25FromDiscord<jos> i was narrowing it down 🙂 i didn't know the name cdecl yet but i was close
06:52:25FromDiscord<brendo-m> sent a code paste, see https://paste.rs/HMQ
06:53:43FromDiscord<Elegantbeef> I dont think so
06:54:35FromDiscord<Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4k1h
06:57:02FromDiscord<brendo-m> unfortunately I can't because the call site is inside the `json.%` macro https://github.com/nim-lang/Nim/blob/devel/lib/pure/json.nim#LL436C8-L436C10
06:57:17FromDiscord<Elegantbeef> I'm uncertain your issue
06:57:46FromDiscord<Bung> can you constrain your generic type to your custom types ?
06:57:48FromDiscord<brendo-m> sent a code paste, see https://paste.rs/tSr
06:58:12FromDiscord<Elegantbeef> Use `std/jsonutils` instead
06:59:06FromDiscord<Bung> it's truely ambiguous to me
07:00:29FromDiscord<Elegantbeef> Yea you have two procedures with the exact same definition, you have to manually export all the procedures you want to except the one you override
07:01:00FromDiscord<Elegantbeef> Though i dont think you can
07:01:21FromDiscord<spetum> I'm kind of new to nim-- What am I doing wrong here?↵`var history = array[16, array[127, int]]`↵`Error: invalid type: 'typedesc[array[0..15, array[0..127, int]]]' for var`
07:01:29FromDiscord<Elegantbeef> Really i dont get why this module exports `json`
07:01:37FromDiscord<Elegantbeef> `var history: array[16, array[127, int]]`
07:01:50FromDiscord<spetum> :facepalm:
07:01:51FromDiscord<spetum> thank you
07:02:43FromDiscord<Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4k1j
07:02:50FromDiscord<brendo-m> it exports json just for convenience. it doesn't matter if I don't though
07:03:14FromDiscord<Elegantbeef> It does matter if you dont, cause you can just import json directly and then dont have this ambiguity since both `%` are not from the same module
07:03:33FromDiscord<Elegantbeef> There is a language design issue that there is no way to select a procedure from a root module
07:04:16FromDiscord<Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4k1k
07:04:22FromDiscord<Elegantbeef> A macro could be used to resolve this, but that feels just as dirty
07:06:00FromDiscord<brendo-m> ``echo `%`(Foo(bar: 42))`` - this is my ultimate goal. this macro doesn't know about module `b`
07:06:27FromDiscord<Elegantbeef> Are you trying to override the `%` for types?
07:06:29FromDiscord<brendo-m> basically I'm trying to replicate https://github.com/nim-lang/Nim/pull/11416/files
07:07:01FromDiscord<Elegantbeef> Use `std/jsonutils` and then export your `toJsonHook` and `fromJsonHook`
07:07:15FromDiscord<Elegantbeef> You dont need `json`/`jsonutils` inside the consume module
07:07:17FromDiscord<Elegantbeef> consuming\
07:08:22FromDiscord<Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4k1l
07:08:46FromDiscord<brendo-m> I tried that but it excludes all of the `%` procs
07:08:58FromDiscord<brendo-m> there aren't that many, so I could just copy paste them all
07:09:08FromDiscord<brendo-m> `jsonutils` looks promising though
07:09:20FromDiscord<Elegantbeef> Depends what you're after i guess
07:09:25FromDiscord<Elegantbeef> I personally never use `%`
07:12:23FromDiscord<Elegantbeef> I do have a semi dumb solution
07:14:39FromDiscord<brendo-m> I love semi dumb solutions
07:15:23FromDiscord<Elegantbeef> Shit nevermind
07:15:31FromDiscord<Elegantbeef> a generic typeclass and a concept typeclass are equal
07:16:14FromDiscord<Elegantbeef> Thought a concept was higher
07:18:35FromDiscord<Elegantbeef> This is where I say "use patchfile"
07:22:11FromDiscord<Elegantbeef> Though this might be a good issue since there is literally not way to disambiguate
07:30:20*derpydoo joined #nim
07:52:53*pro joined #nim
09:05:16*rockcavera quit (Remote host closed the connection)
09:21:11*azimut quit (Ping timeout: 255 seconds)
09:22:32*pro quit (Quit: pro)
09:22:40*azimut joined #nim
09:43:06*Batzy quit (Quit: https://quassel-irc.org - Chat comfortably. Anywhere.)
09:44:21*Batzy joined #nim
10:16:59*azimut quit (Ping timeout: 255 seconds)
10:17:03*azimut_ joined #nim
10:27:41*jmdaemon quit (Ping timeout: 246 seconds)
10:48:32*azimut_ quit (Remote host closed the connection)
10:49:01*azimut joined #nim
10:57:57FromDiscord<voidwalker> https://media.discordapp.net/attachments/371759389889003532/1058700532849582130/image.png
10:58:16FromDiscord<voidwalker> https://media.discordapp.net/attachments/371759389889003532/1058700612558135406/image.png
11:10:07FromDiscord<lignus> sent a code paste, see https://paste.rs/REJ
11:10:33FromDiscord<lignus> (edit) "https://play.nim-lang.org/#ix=4k1W" => "https://play.nim-lang.org/#ix=4k1V"
11:22:28FromDiscord<planetis> there is no interface, how the api looks, depends on the allocator you implemented.
11:30:53FromDiscord<lignus> sent a long message, see https://paste.rs/HNt
11:35:28FromDiscord<planetis> > I see some people casting from ptr T to ref T↵that pbl wouldn't work. There is a header before what the ref pointer points to, that contains a refcounter, rti for object inheritance, a destructor pointer, etc, you can read the code to confirm this.
11:36:07NimEventerNew thread by jasonfi: -d:useMalloc fixes memory leak, but I don't know why, see https://forum.nim-lang.org/t/9784
11:36:30FromDiscord<planetis> there is an example pools allocator https://github.com/nim-lang/fusion/blob/master/src/fusion/pools.nim
11:37:28FromDiscord<lignus> Thank you, that was exactly what I was looking for! I think I was confused because I found someone monkey patching the standard library implementation.↵↵<https://github.com/vitreo12/nim_arc_custom_alloc/blob/master/custom_malloc.nim>
11:38:14FromDiscord<planetis> You can attach destructors https://nim-lang.github.io/Nim/destructors.html to the types you allocate
11:39:25FromDiscord<lignus> Why would I attach a destructor?
11:39:33FromDiscord<planetis> oh you want to write an allocator that replaces the std one, better see how it's done in https://github.com/Yardanico/mimalloc_nim
11:40:17FromDiscord<lignus> no
11:40:56FromDiscord<planetis> But for std containers the is no way to replace the allocator they use
11:40:58FromDiscord<lignus> I'm actually writing an arena allocator. I was confused becuase the allocator I looked at had a different goal (replacing std) than mine (a secondary allocator).
11:42:03FromDiscord<planetis> Yeah, you might end up handrolling your own container types that use that allocator and have destructors for RAII style memory management
11:42:39FromDiscord<lignus> Ah, okay! All allocations done within an arena are freed when the arena is deallocated, though.
11:43:14FromDiscord<planetis> for refs you might need types like https://github.com/nim-lang/threading/blob/master/threading/smartptrs.nim
11:43:40FromDiscord<planetis> (edit) "for refs you might need types like https://github.com/nim-lang/threading/blob/master/threading/smartptrs.nim ... " added "to replace them"
11:44:01FromDiscord<planetis> In reply to @lignus "Ah, okay! All allocations": so just like the fusion pools allocator
11:45:48FromDiscord<lignus> Maybe, I haven't read through it yet. Thanks for the great examples!
11:46:00FromDiscord<planetis> you're welcome!
11:53:04*pro joined #nim
12:01:53FromDiscord<Phil> Without fully getting the context, new should ideally always return ref types
12:02:01FromDiscord<Phil> And man my discord is slow
12:02:39FromDiscord<Phil> There I go, reading through the messages, see a question about return type of "new", think it's the most current one, I type and BAM discord loads like 50 new messages since then
12:03:50FromDiscord<Phil> In reply to @Elegantbeef "Shit nevermind": You still around? I tinkered a bit further since then and got stuck with a NimNode that's a proc being demanded by the macro to be the proc-type and I'm not sure what to do there
12:03:57FromDiscord<Phil> (edit) "around?" => "around or already gone?"
12:04:19FromDiscord<Phil> (edit) "then" => "our last exchange"
12:04:42*pro quit (Quit: pro)
12:06:03FromDiscord<Phil> sent a code paste, see https://play.nim-lang.org/#ix=4k2a
12:06:53FromDiscord<Phil> I'm.... reasonably sure that the proc within the NimNode is exactly that kind of proc, I'm mostly not sure what the "proper" way to extract it is
12:07:11FromDiscord<Phil> (edit) "I'm.... reasonably sure that the proc within the NimNode ... isis." added " `copyProc`" | "is" => "is."
12:49:09FromDiscord<Bung> I think you can check the async pragma implmentation.
13:04:46*pro joined #nim
13:18:06FromDiscord<Phil> In reply to @Bung "I think you can": This? https://github.com/nim-lang/Nim/blob/8c0f2f01527dbeba071c6478d6d38c9cf841ba71/lib/pure/asyncmacro.nim#L253
13:19:18FromDiscord<Bung> yes, as I remenber it also do copy body and replace proc things.
13:20:20FromDiscord<Phil> It's got a 120 line proc `asyncSingleProc` for the most part
13:54:23NimEventerNew thread by jerous: Nimqt - bindings to Qt, see https://forum.nim-lang.org/t/9785
14:04:45FromDiscord<ShalokShalom> @Phil Oh, that is a binding to qt-widgets, the other Qt toolkit
14:05:04FromDiscord<ShalokShalom> Its more the traditional style of an UI toolkit.
14:05:23FromDiscord<Phil> Wait a sec, there are ways that you can do debugging like jvm debugging, with stopping code while executing, seeing variables and their current values and stuff?
14:06:15FromDiscord<ShalokShalom> You mean something likr Dtrace?
14:06:34FromDiscord<Phil> I have no idea of anything I've only ever seen that style of debugging experience in Java based IDEs
14:07:03FromDiscord<Phil> Correction: IDEs while using java
14:07:14FromDiscord<Phil> And python... and while in the browser also JS
14:07:18FromDiscord<ShalokShalom> Well, Bryan Cantrill kinda brought this kind of advanced debugging to C
14:07:25FromDiscord<ShalokShalom> And generally to non VM languages.
14:07:40FromDiscord<ShalokShalom> His videos are worth watching, he is a hilarious genius.
14:07:57FromDiscord<ShalokShalom> In reply to @Isofruit "Correction: IDEs while using": That actually comes from Smalltalk 😛
14:08:07FromDiscord<ShalokShalom> As the whole IDE/VM stuff in general.
14:08:24FromDiscord<ShalokShalom> Idk if Dtrace would make much sense on Nim. I kinda doubt
14:08:31*xet7 quit (Quit: Leaving)
14:08:47FromDiscord<ShalokShalom> Since you cant see, where an issue arised anymore. The compilation is already too advanced. I assume.
14:08:55FromDiscord<enthus1ast> dtrace is the solaris thing right?
14:09:23FromDiscord<ShalokShalom> Yes
14:09:53FromDiscord<jtv> If you’re talking about things like ebpf that isn’t stopping a program or anything and is mainly used for observability
14:09:58FromDiscord<ShalokShalom> It kinda brought other projects into rolling
14:09:59FromDiscord<Phil> In reply to @ShalokShalom "His videos are worth": That may be true, but I'd mostly be looking for ways that I could just plug this kind of stuff into nim for a similar blessed experience. Though I still only read here and there that it has support on some level but never see anyone actually ever doing it
14:10:32FromDiscord<ShalokShalom> You would need to ask the devs
14:10:50FromDiscord<ShalokShalom> Idk, if that is still traceable after compilation
14:11:18FromDiscord<enthus1ast> maybe the nim embedded debugger is worth some love
14:11:20FromDiscord<Phil> I stumble over this kind of stuff from time to time when I read that C has a bunch of tooling on that level
14:11:24FromDiscord<jtv> But debuggers have long used breakpoints to allow you to stop execution and observe the program. Nim does allow you to use debuggers commonly used in C like gdb
14:11:44FromDiscord<jtv> Most languages do, even Python has pdb
14:12:02FromDiscord<jtv> 100% worth your time to learn
14:12:05FromDiscord<enthus1ast> and python has IPython.embed()
14:12:11FromDiscord<ShalokShalom> Afaik, thats a different kind of breed
14:12:15FromDiscord<ShalokShalom> And not really comparable
14:12:19FromDiscord<ShalokShalom> I forgot why :0
14:12:25FromDiscord<ShalokShalom> (edit) ":0" => "😛"
14:13:23FromDiscord<ShalokShalom> Ah yeaz
14:13:26FromDiscord<ShalokShalom> Trace runs on a different thread thus it does not impact the main code thread.There is also a fundamental difference in thinking when we want to use trace and when want to debug. Tracing is a process about getting information regarding program's execution. On the other hand debugging is about finding errors in the code.
14:13:30FromDiscord<ShalokShalom> (edit) "yeaz" => "yeah"
14:13:38FromDiscord<ShalokShalom> (edit) "Trace" => ""Trace" | "code." => "code.""
14:14:11FromDiscord<enthus1ast> ah me have not read the whole conversation but just talked random stuff \:)
14:14:37FromDiscord<ShalokShalom> That's me, usually
14:14:46FromDiscord<<She>Horizon</Her>> Anyone know if there's any existing WebKit bindings for C (or Nim, Nim is preferable)
14:15:10FromDiscord<jtv> Right he was asking about debuggers, trace (which is obsolete) dtrace (which isn’t) and ebpf are observability tools, not really used for debugging.
14:15:28FromDiscord<ShalokShalom> In reply to @Isofruit "I stumble over this": I think, this is likely to be one of the cases, at which compiling to horribly unreadable C code does kill the fun.
14:15:59FromDiscord<jtv> What do you mean??
14:16:09FromDiscord<ShalokShalom> In reply to @jtv "Right he was asking": Well, they argubly make common debugging tools superfluous
14:16:22FromDiscord<ShalokShalom> In reply to @jtv "What do you mean??": Using Dtrace for Nim
14:16:29FromDiscord<jtv> Super false.
14:16:36FromDiscord<ShalokShalom> I would wonder, if that would be working.
14:17:02FromDiscord<ShalokShalom> In reply to @Event Horizon "Anyone know if there's": There is one for C++
14:17:07FromDiscord<jtv> Have you ever tried to set a uprobe in a containerized environment? They don’t work well even and it’s not an easy thing to do
14:17:12FromDiscord<ShalokShalom> Qt has it, called Qt WebKit
14:17:32FromDiscord<<She>Horizon</Her>> In reply to @ShalokShalom "There is one for": That's fine for me
14:17:41FromDiscord<ShalokShalom> In reply to @jtv "Have you ever tried": I meant effectively, not technically
14:17:41FromDiscord<jtv> And dtrace only gives you access at the system call level and below
14:18:02FromDiscord<ShalokShalom> In reply to @Event Horizon "That's fine for me": Oh, and there are also Webkit based C browsers
14:18:07FromDiscord<ShalokShalom> Those for Gnome
14:18:24FromDiscord<<She>Horizon</Her>> In reply to @ShalokShalom "Oh, and there are": Oh? That's probably more useful for me, I'd prefer to just keep my code at C
14:19:12FromDiscord<ShalokShalom> Although idk how much they depend on gtk
14:19:23FromDiscord<ShalokShalom> And I might just confuse something 😅
14:19:30FromDiscord<ShalokShalom> What do you want to do?
14:19:41FromDiscord<ShalokShalom> In reply to @jtv "And dtrace only gives": I see
14:19:49FromDiscord<ShalokShalom> Do you think, we can use it for Nim?
14:20:07FromDiscord<jtv> In reply to @ShalokShalom "I meant effectively, not": What does that even mean? I have used debuggers and tracing tools both for 30 years and built a very successful product around ebpf and don’t understand.
14:21:16FromDiscord<ShalokShalom> I assumed they would cover similar use cases
14:21:40FromDiscord<ShalokShalom> Since tracing is still used to understand your program, in case you encounter an issue
14:22:07FromDiscord<ShalokShalom> May be performance related, may be related to memory issues or something else
14:22:22FromDiscord<ShalokShalom> But I always assumed, they cover kinda the same space
14:22:45FromDiscord<Bung> In reply to @Event Horizon "Anyone know if there's": there's https://github.com/juancarlospaco/webgui
14:22:50FromDiscord<<She>Horizon</Her>> In reply to @ShalokShalom "What do you want": I want to use WebKit as the driving force behind a basic/crappy TUI browser
14:23:38FromDiscord<jtv> I mean, a debugger makes it a lot easier to understand what’s going on inside a program but is much more heavyweight, as it’s basically rewriting your code. It isn’t suitable for production at scale
14:24:02FromDiscord<<She>Horizon</Her>> In reply to @Bung "there's https://github.com/juancarlospaco/webgui": Oh?
14:24:44FromDiscord<jtv> But still it’s generally going to be used to get metrics data out, they’re hard to use for debugging
14:25:20FromDiscord<ShalokShalom> And do you think we could use dtrace for Nim?
14:25:51FromDiscord<jtv> Dtrace is language independent
14:26:03FromDiscord<jtv> It’s more a system-level tool
14:26:28FromDiscord<ShalokShalom> Nice :R
14:26:34FromDiscord<ShalokShalom> (edit) ":R" => "😄"
14:28:23FromDiscord<jtv> Your program uses system calls under the hood, you can see those get invoked, the data that gets past to it, etc. but those tools aren’t meant for app debugging
14:30:19FromDiscord<Phil> sent a long message, see http://ix.io/4k2H
14:30:50FromDiscord<jtv> Nim program compile w that flag by default
14:31:05FromDiscord<jtv> And it’s much simpler than the ebpf and dtrace debugging
14:32:20FromDiscord<Phil> I have no experience with those either so I can't compare to those.↵My baseline I'm comparing to is pretty much the java dev workflow:↵Click round button next to line of code to set a breakpoint there, click shiny play button to run program, see loc highlighted where you're currently at and all available variables and their values
14:32:20FromDiscord<jtv> With ebpf, if your program already compiles directly to an elf file, you can get flame graphs easily. Doing real state inspection dynamically? Nope
14:32:38FromDiscord<jtv> You need to pre-build and hook your probes
14:32:50FromDiscord<jtv> Yeah gdb does that
14:33:53FromDiscord<Phil> Then this is likely me not having gdb experience and not knowing how to properly use it in nim
14:34:01FromDiscord<jtv> Things like VSCode wrap more UI around it all.
14:34:47FromDiscord<jtv> I’m an old command line guy so I tend to use my debuggers directly but I’d assume the vscode experience is similar to what you’d see in Java environments
14:47:39*pro quit (Quit: pro)
14:48:24*derpydoo quit (Ping timeout: 260 seconds)
14:49:12*derpydoo joined #nim
14:58:23*nick1 joined #nim
14:58:35*azimut quit (Remote host closed the connection)
14:59:23*azimut joined #nim
15:02:34NimEventerNew thread by archnim: A serious documentation for Nim, see https://forum.nim-lang.org/t/9786
15:04:55FromDiscord<jtv> Someone on my team who is a VSCode user says he uses the CodeLLDB extension for debugging on compiled languages, @Phil.
15:08:08FromDiscord<ezquerra> The debugger is another example of nim’s tooling being subpar compare to the language itself
15:09:13FromDiscord<ezquerra> Ideally, you should just be able to click on a line in VS Code, launch the debugger, and when the line is hit view the state of all the local variables and al start stepping through the code
15:10:02*arkurious joined #nim
15:12:45FromDiscord<jtv> You can
15:13:43FromDiscord<jtv> It's a compiled language. The CodeLLDB extension works just as well for Nim as it does for C.
15:15:08FromDiscord<jtv> I don't know about when you're targeting the JS backend, but I'd assume even there you'd have full traceability back to the source.
15:23:44FromDiscord<Piqueiras> if I have both an iterable and an enum with 10 elements, is there a way I can work out a pairs iterator for those
15:25:13FromDiscord<Phil> In reply to @Piqueiras "if I have both": Without having tried this myself, have you looked into `std/sequtils.zip` yet?
15:25:41FromDiscord<Piqueiras> tbh no lol
15:26:21FromDiscord<Piqueiras> i just found it kinda weird cause enums should be known at compile time (as far as i know) but I want to iterate over a seq
15:33:22*derpydoo quit (Ping timeout: 268 seconds)
15:33:29FromDiscord<sOkam!> In a manual memory situation, how do you access a specific location of a ptr uint8 or pointer?
15:34:03FromDiscord<jtv> I would assume you cast it to a ptr array
15:34:06FromDiscord<sOkam!> I'm trying:↵https://www.gingerbill.org/article/2019/02/08/memory-allocation-strategies-002/↵But stuck at the `void ptr = &arena_buffer[arena_offset];` part
15:34:08FromDiscord<willm> sent a code paste, see https://play.nim-lang.org/#ix=4k34
15:34:41FromDiscord<Phil> sent a code paste, see https://play.nim-lang.org/#ix=4k35
15:34:48FromDiscord<willm> sent a code paste, see https://play.nim-lang.org/#ix=4k36
15:34:54FromDiscord<willm> (edit) "https://play.nim-lang.org/#ix=4k36" => "https://play.nim-lang.org/#ix=4k37"
15:38:18FromDiscord<Phil> Not one of the guys that uses nimlsp myself. PMunch is fairly active here though, he likely knows more.↵What OS are you installing from?
15:38:25FromDiscord<willm> In reply to @Isofruit "Not one of the": NixOS
15:38:46FromDiscord<Phil> `nimble -v` output?
15:39:01FromDiscord<willm> it's the same, doesn't give any more clues
15:39:20FromDiscord<willm> https://media.discordapp.net/attachments/371759389889003532/1058771344759738450/screenshot-1672501143.jpg
15:39:23FromDiscord<willm> there's the full output
15:39:27FromDiscord<Phil> That's for getting the nimble version, not the nimble version
15:39:32FromDiscord<Phil> (edit) "nimble" => "nim"
15:39:33FromDiscord<willm> oh my bad
15:39:47FromDiscord<willm> sent a code paste, see https://play.nim-lang.org/#ix=4k39
15:41:35FromDiscord<Phil> Hmm can't think of any further debugging information for when the guy arrives. I can only tell you it works on arch =/
15:42:07FromDiscord<willm> yeah unfortunate that the output doesn't say anything more... thanks for the help
15:42:47FromDiscord<willm> In reply to @Isofruit "Hmm can't think of": can you think of another package that installs an executable? Testing to see if that works would narrow it down to either a nixos issue or nim issue
15:43:38FromDiscord<planetis> There are two ways, use ptr UncheckedArray and [] or declare a += on pointers and get the result. Example code does both
15:44:18FromDiscord<planetis> In reply to @sOkam! "In a manual memory": There are two ways, use ptr UncheckedArray and [] or declare a += on pointers and get the result. Example code does both
15:45:14FromDiscord<Piqueiras> and, is it me or i can use pair iterator on string.split(char)
15:45:18FromDiscord<Piqueiras> (edit) "can" => "cant"
15:47:01FromDiscord<Piqueiras> no, it wont let me
15:50:31FromDiscord<Bung> oh it compiled before I was born ..
15:52:02FromDiscord<planetis> In reply to @planetis "There are two ways,": my bad it's usually `+!` https://github.com/nim-lang/Nim/blob/devel/lib/std/private/dragonbox.nim#L1047
15:57:55FromDiscord<Piqueiras> oh, so i cant iterate pairs on split unless i save it on a variable
15:59:39FromDiscord<sOkam!> sent a code paste, see https://play.nim-lang.org/#ix=4k3d
16:00:51FromDiscord<sOkam!> oh, i might have forgotten to init the arena. whoppsss 🙈
16:07:36FromDiscord<ezquerra> In reply to @jtv "It's a compiled language.": What I’d like is that the out of the box debugger experience worked well in all OS supported by nim
16:08:32*jvinet joined #nim
16:21:23FromDiscord<planetis> In reply to @sOkam! "I might not be": your translation is correct, but I wonder why it uses the pointer to a pointer
16:28:51FromDiscord<ambient> In reply to @jtv "I don't know about": Do these also work in Windows? I've found Nim doesn't seem to generate debug info that many tools are able to read
16:28:58FromDiscord<haxscramper> documentation
16:29:24FromDiscord<haxscramper> In reply to @NimEventer "New thread by archnim:": @archnim are you here?
16:29:47FromDiscord<ambient> for example i have not successfully generated suitable labels for Nim programs for Superluminal sampling profiler
16:30:18FromDiscord<ambient> all function names are just some hex values
16:32:12*rockcavera joined #nim
16:54:24FromDiscord<ShalokShalom> https://github.com/google/rune/issues/38
16:54:35FromDiscord<ShalokShalom> Google loves Nim
16:54:39FromDiscord<ShalokShalom> Its official
16:54:59FromDiscord<archnim> I'm here @haxscramper
16:56:35FromDiscord<haxscramper> Ok, so I wanted to elaborate a bit on the forum post - while the idea to have a dedicated team for the docs is ok, generally speaking it must be built on some.kind of foundation of processes which are hardly present here
16:57:37FromDiscord<haxscramper> And considering the attitude (trying to push 62$ book as the first sentence) of the core devs ... basically the docs would be improved incrementally and this is the only solution I've seen work over the years
16:58:33FromDiscord<haxscramper> So I was speaking more from my experience and directly applicable actions rather than more abstract general view of the problem with documentation
16:59:29FromDiscord<archnim> I understand.
17:00:50FromDiscord<archnim> Concerning the core devs, I understand them too, I know they are not selfish, the just need to live too. And they do a lot of sacrifice to Nim.
17:01:35FromDiscord<haxscramper> Araq is paid full-time
17:02:58FromDiscord<haxscramper> The content on books is ok, but I still can't understand why not put it in docs
17:04:34FromDiscord<Phil> I think actionable statements could be that exported symbol should have docs and exported procs runnable examples. Those often exist, but not always and could be a measurable metric
17:05:24FromDiscord<haxscramper> This is a tooling question but yes, some boilerplate requirements can move things into a better direction
17:05:54FromDiscord<Phil> That includes docs at the start of libs/modules including concepts or larger examples
17:06:18FromDiscord<ringabout> I don't know whether a new book is needed. I would be in favor of the improvement of the current documentation. `runnableExamples` additions or perhaps a specific or some handy documentation on specific topics.
17:07:31FromDiscord<haxscramper> Someone writing module docs should write a piece of code that uses every proc/type in this module
17:07:39FromDiscord<haxscramper> And then explain why everything is used this way
17:07:49FromDiscord<Phil> I agree with favouring module/symbol docs over a new book
17:08:05FromDiscord<haxscramper> Wrote an "except" somewhere? Why code can raise? Is it important?
17:08:27FromDiscord<haxscramper> Called something? What does it do? Can I get this from docs?
17:09:08FromDiscord<haxscramper> What was the original task for this piece of code? Take examples that people write on the the forum
17:09:22FromDiscord<haxscramper> That's exactly what must be in docs
17:09:50FromDiscord<haxscramper> People go elsewhere to ask questions and then they are satisfied with things they get on forum
17:10:18FromDiscord<haxscramper> Therefore documentation implementers should look very closely on forums
17:11:37FromDiscord<ringabout> As to the multilingual version of Nim documentation, If only we could set a workflow for it, I will definitely contribute to it. I tried, but I couldn' figure out how to work with https://weblate.org/ See also https://github.com/nim-lang/Nim/issues/18821
17:11:48FromDiscord<Vindaar> as an aside: whenever ideas for docs come up, let me plug the Clojure docs again https://clojuredocs.org/clojure.core/map user addable examples, which can be voted on is imo a great idea (generally the structure is also pretty nice imo)
17:11:58FromDiscord<ringabout> (edit) "couldn'" => "couldn't"
17:16:16FromDiscord<haxscramper> haxe docs can be commented on using github, this might also be useful
17:16:27FromDiscord<haxscramper> right, there is no shortage of technical things that can be implemented
17:20:51FromDiscord<haxscramper> https://forum.nim-lang.org/t/3766#23446 but with things like these
17:21:53FromDiscord<haxscramper> https://forum.nim-lang.org/t/4331#26963
17:22:16FromDiscord<haxscramper> https://forum.nim-lang.org/t/3308#20858
17:25:55FromDiscord<ShalokShalom> Yeah, I actually love that you can comment on Haxe docs on the website
17:30:40FromDiscord<haxscramper> In reply to @haxscramper "And considering the attitude": > User archnim2 shows up a couple of months later and complains our docs are bad, including every written book about Nim, even those freely available for the poor people who are happy to pay for everything else in their lives.
17:31:32FromDiscord<haxscramper> In reply to @ShalokShalom "Yeah, I actually love": as you can see the goal is to just dodge complaints
17:32:02FromDiscord<willyboar> In reply to @haxscramper "haxe docs can be": You can use utterances to do that
17:36:21FromDiscord<archnim> I just published a post on the forum thread to describe in details how I think the docs could be remastered.
17:40:03FromDiscord<jtv> In reply to @ambient "Do these also work": Not sure, I'm not a Windows person unfortunately. But I'd be surprised... the default flow is that Nim produces C or C++ that gets compiled. And there, it's easy to add #line pragmas to get the line info back. If the debugger expects C++, maybe make sure the backend is generating C++ in case it's a name mangling issue or somesuch?
17:40:45FromDiscord<Gumbercules> In reply to @haxscramper "The content on books": The book literally reads like the manual
17:41:31FromDiscord<jtv> Yeah, neither nim book really helps much vs the online docs, and the online docs aren't awesome
17:41:59FromDiscord<Gumbercules> Stefans book actually tries
17:42:46FromDiscord<haxscramper> In reply to @Gumbercules "The book literally reads": the "mastering nim" one?
17:43:01FromDiscord<haxscramper> well, looking at TOC that is 1:1 with nim manual this is not surprising
17:43:40FromDiscord<haxscramper> IIRC technically MIT license allows you to take community-contributed content and sell it
17:43:43FromDiscord<haxscramper> or something like that
17:45:32*derpydoo joined #nim
17:46:58FromDiscord<willyboar> Cool!
17:48:09FromDiscord<jtv> There's a bit more depth on macros and such, but a lot of it is basically the language reference
17:50:29FromDiscord<Phil> In reply to @haxscramper "IIRC technically MIT license": MIT allows you everything except for suing the person writing the code if it was bad code is the short version that I think of when it comes up
17:52:01FromDiscord<haxscramper> whether it is an ok thing to do or not is another question
17:52:46FromDiscord<haxscramper> but since not a lot of people have this book anyway it is hard to judge how similar is it anyway
18:02:53FromDiscord<Bung> speaking of documentation, first idea coming to my mind is split the official site "Documentation" to 3 parts "Documentation" , "References" and "Learn", I dont want "Getting started with Nim" every time.
18:07:27FromDiscord<ambient> In reply to @Isofruit "MIT allows you everything": MIT requires that you also disclose where the code is from, license (MIT) and who wrote it. People often completely ignore this part.
18:10:50FromDiscord<answer> sent a code paste, see https://play.nim-lang.org/#ix=4k3J
18:12:56FromDiscord<haxscramper> Not sure if you can use concepts as direct fields, IIRC they are meant to be used as type constraints, so `T: Contains` or something similar
18:13:40FromDiscord<Gumbercules> In reply to @haxscramper "well, looking at TOC": Which is annoying because if I had known this was the case I would have never ordered it. I've already thrown more than enough $ at Nim.
18:15:51FromDiscord<haxscramper> maybe that's just TOC, I can't really see what's inside for obvious reasons
18:16:10FromDiscord<haxscramper> are there any text pieces that also come directly from the manual?
18:20:03FromDiscord<Gumbercules> I'll have a look later today
18:20:20FromDiscord<Gumbercules> I'm fairly certain the answer is yes, but I'll verify
18:23:16FromDiscord<jtv> @answer concepts aren't concrete types, they're saying what a concrete type must have to be considered "Contains" at compile time
18:23:32FromDiscord<jtv> You still need a concrete type in order to be able to instantiate
18:23:47FromDiscord<jtv> Even if it's just an object ref you cast to
18:26:41FromDiscord<answer> trying to figure out if there's some other way to describe it since i see some other code use concepts as constraints but maybe it's a little ambitious... um looking at https://github.com/mratsim/weave right now and seeing how they use concepts and how they contain things with concepts
18:26:44FromDiscord<Gumbercules> https://media.discordapp.net/attachments/371759389889003532/1058813474123284591/rn_image_picker_lib_temp_04e08726-c649-4230-9a84-5b85c1cd1729.jpg
18:27:08FromDiscord<Gumbercules> https://nim-lang.org/docs/manual.html#lexical-analysis-identifiers-amp-keywords
18:27:30FromDiscord<jtv> One sec
18:27:56FromDiscord<haxscramper> https://media.discordapp.net/attachments/371759389889003532/1058813774225743942/image.png
18:27:58FromDiscord<haxscramper> yeah xxxxd
18:28:25FromDiscord<Gumbercules> Most of the book (part two) is the spec, so it seems a copy paste of the manual probably with some additions / omissions
18:29:39FromDiscord<voidwalker> I got the book as well, I am pissed there is no pdf
18:29:44FromDiscord<voidwalker> you can't ctrl+f a book
18:30:09FromDiscord<Gumbercules> I mean, I'd just be happy with some more original content
18:30:20FromDiscord<Gumbercules> More than 40ish pages anyway
18:30:27FromDiscord<jtv> @answer are you just trying to keep run-time type info around, like a box type?
18:30:35FromDiscord<jtv> I have one, let me make it public for you, one second.
18:31:07FromDiscord<voidwalker> yeah the book is quite .. non homogenous. starts out like it would be a book for beginners, then copy pastes the manual, then some template/macros
18:31:08FromDiscord<answer> i have a friend exploring nim right now and they're trying to use concepts like traits
18:31:12FromDiscord<answer> (edit) "i have a friend exploring nim right now and they're trying to use concepts like traits ... " added "from rust"
18:31:25FromDiscord<answer> maybe there's a better equivalent for this?
18:32:24FromDiscord<jtv> I mean, give me more detail about what you're actually trying to do in the code snippet and I can help
18:32:48FromDiscord<jtv> For instance, check out box.nim in the repo I just made public: https://github.com/crashappsec/nimutils
18:33:26FromDiscord<jtv> It uses a concept to do what I think is what you're trying to do
18:33:26FromDiscord<answer> it isn't public btw
18:34:33FromDiscord<jtv> Whoops, try it now
18:37:09FromDiscord<jtv> So you can see the big difference is, you need a concrete type to instantiate, and I use Box as that type.
18:37:16FromDiscord<answer> ohh interesting
18:38:01FromDiscord<jtv> As long as you know the type at compile time, it's not going to need any runtime checking, but if you need to figure out the type at runtime, the info is there.
18:41:16FromDiscord<answer> this is neat but i assume they are looking for some compiletime way to do this like rust does with traits. this does do the job though
18:43:02FromDiscord<jtv> Well, if you don't want the runtime type info around, you don't need the recursion when boxing/unboxing. I did it that way b/c of my specific use case
18:43:29FromDiscord<jtv> And then everything is static, but you still need some container that is a concrete type
18:43:48FromDiscord<jtv> Whether it's RootObj, or...
18:44:19FromDiscord<jtv> A concept is nothing but an interface, it's not something you can instantiate, you can only instantiate things that implement the interface
18:44:51FromDiscord<jtv> So in order to instantiate your Pair type, you need to have something concrete at some point 🙂
18:44:56FromDiscord<answer> yeah
18:45:13FromDiscord<jtv> Hope that helps
18:46:40FromDiscord<answer> i'll try playing around with it with that knowledge
18:47:00FromDiscord<jtv> Cool, lmk if you have any more questions
18:48:02FromDiscord<Gumbercules> In reply to @answer "i have a friend": C++ concepts
18:48:20FromDiscord<Gumbercules> https://en.cppreference.com/w/cpp/language/constraints
19:28:58*xet7 joined #nim
19:32:36FromDiscord<ShalokShalom> In reply to @Isofruit "MIT allows you everything": MIT also gives a waver to patents, I think
19:32:56FromDiscord<ShalokShalom> Since that is, what the CC 0 doesnt, and what makes Fedora reject it as a license
19:33:13FromDiscord<ambient> Public domain does not waver patents?
19:33:29FromDiscord<ShalokShalom> In reply to @haxscramper "as you can see": Oh, I know that so well
19:33:51FromDiscord<ShalokShalom> Not from this project (only) but this is generally how people react to critiqe
19:34:00FromDiscord<ShalokShalom> In reply to @ambient "Public domain does not": CC 0 doesnt
19:34:30FromDiscord<ShalokShalom> https://hackaday.com/2022/07/29/why-fedora-decided-to-give-cc0-licensed-code-the-boot/
19:48:46*joast quit (Quit: Leaving.)
19:55:00FromDiscord<System64 ~ Flandre Scarlet> Hi, does Nim support QT?
19:57:02FromDiscord<ShalokShalom> Yes
19:57:06FromDiscord<ShalokShalom> Since today 😛
19:57:36FromDiscord<ShalokShalom> The Nim Qt Widget library has actually been posted today
19:58:15FromDiscord<ShalokShalom> @System64 ~ Flandre Scarlet https://forum.nim-lang.org/t/9785
19:59:06FromDiscord<System64 ~ Flandre Scarlet> In reply to @ShalokShalom "The Nim Qt Widget": Oh what a coincidence lol↵And does it support that? https://media.discordapp.net/attachments/371759389889003532/1058836716355584021/image.png
19:59:17FromDiscord<ShalokShalom> The Qt Quick binding is already in use for a couple while https://github.com/filcuc/NimQml
19:59:56FromDiscord<System64 ~ Flandre Scarlet> In reply to @ShalokShalom "The Qt Quick binding": Oh so I can quickly dev GUI apps with NimQml and NimQT?
20:00:07FromDiscord<ShalokShalom> @System64 ~ Flandre ScarletWell, I am not the right person to ask. Generally: Nim has some support in the Qt designer, for the language itself. Its a bit unmaintained, though.
20:00:32FromDiscord<ShalokShalom> In reply to @System64 "Oh so I can": You can. If you want to design the GUI in the visual designer, I think only QML is an option.
20:00:54FromDiscord<ShalokShalom> Since it produces own files. Qt widgets gets implemented in the Nim logic code.
20:01:30FromDiscord<ShalokShalom> You know the difference between Widgets and Quick in terms of use case and so on?
20:03:18FromDiscord<ShalokShalom> I actually like the API, that the new Qt Widget binding brings
20:03:22FromDiscord<System64 ~ Flandre Scarlet> In reply to @ShalokShalom "You know the difference": No↵I just want to create an UI like that↵https://editor.p5js.org/system64MC/sketches/k8vzhPDLZ↵But I like declarative syntax like XAML, FXML, etc and with an editor
20:03:22FromDiscord<ShalokShalom> And I might even use it
20:04:05FromDiscord<ShalokShalom> I would suggest, you look into both
20:04:16FromDiscord<ShalokShalom> QML offers hardware acceleration and mobile support
20:04:57FromDiscord<System64 ~ Flandre Scarlet> But does QML have event trigger when I click on a button, move my mouse and click on a canvas, etc?
20:05:00FromDiscord<ShalokShalom> Considering this one binding came out today, practice is basically not established with it
20:05:11FromDiscord<ShalokShalom> And talking purely theoretically is tricky.
20:05:24FromDiscord<ShalokShalom> In reply to @System64 "But does QML have": Yes
20:05:37FromDiscord<ShalokShalom> Its actually a superset of Javascript
20:05:55FromDiscord<System64 ~ Flandre Scarlet> Oh, I know JS quite well
20:06:05FromDiscord<ShalokShalom> You shouldnt overuse it
20:06:22FromDiscord<ShalokShalom> But a QML file is a fully fledged Javascript plus the QML parts
20:06:45FromDiscord<ShalokShalom> You can can combine UI element stuff with Javascript logic
20:06:49FromDiscord<ShalokShalom> To describe your UI
20:07:19FromDiscord<System64 ~ Flandre Scarlet> And can I set UI elements (slider values for exemple) from Nim backend?
20:07:35FromDiscord<ShalokShalom> And there is an equivalent to the Qt Designer, for Quick: https://www.qt.io/product/ui-design-tools
20:07:42FromDiscord<ShalokShalom> Qt Designer is for Qt Widgets
20:07:52FromDiscord<ShalokShalom> Qt Design Studio is for Qt Quick
20:08:11FromDiscord<System64 ~ Flandre Scarlet> I'm on Qt Designer
20:08:17FromDiscord<ShalokShalom> In reply to @System64 "And can I set": I did not ask with it yet. @Phil knows more.
20:08:28FromDiscord<System64 ~ Flandre Scarlet> https://media.discordapp.net/attachments/371759389889003532/1058839075395747970/image.png
20:08:56FromDiscord<ShalokShalom> In reply to @System64 "I'm on Qt Designer": Yeah, I see. They have it a bit weird. Design Studio used to be paid, and now its free and still closed source. And they have a weird way of distributing it
20:09:00FromDiscord<ShalokShalom> At least to Linux.
20:09:23FromDiscord<ShalokShalom> (edit) "ask" => "code"
20:09:39FromDiscord<System64 ~ Flandre Scarlet> Oh alright↵Qt is weird
20:09:57FromDiscord<ShalokShalom> The point is, with the Qt Design Studio could you export QML files
20:10:07FromDiscord<ShalokShalom> And then referencing them in your Nim backend
20:10:37FromDiscord<System64 ~ Flandre Scarlet> In reply to @ShalokShalom "The point is, with": is Design Studio still free,
20:10:38FromDiscord<System64 ~ Flandre Scarlet> (edit) "free," => "free?"
20:10:43FromDiscord<System64 ~ Flandre Scarlet> (edit) "free?" => "free for Windows?"
20:10:47FromDiscord<ShalokShalom> I think so
20:10:48FromDiscord<ShalokShalom> https://discord.com/channels/371759389889003530/371759607934353448/1046505560763269250
20:10:51*kenran joined #nim
20:10:51*kenran quit (Remote host closed the connection)
20:10:55FromDiscord<ShalokShalom> This gives you a good idea
20:11:31FromDiscord<System64 ~ Flandre Scarlet> Seems interesting
20:13:46*jmdaemon joined #nim
20:14:39FromDiscord<System64 ~ Flandre Scarlet> And is there another alternative that uses a declarative syntax for UI design? (except HTML)
20:15:06FromDiscord<ShalokShalom> Yes
20:15:51FromDiscord<ShalokShalom> https://github.com/can-lehmann/owlkettle
20:16:16FromDiscord<ShalokShalom> Know, that GTK is mainly used for Linux
20:16:27FromDiscord<ShalokShalom> It does look not native on other platforms
20:16:52FromDiscord<ShalokShalom> And is generally horrible. This just happens to be a good implementation and abstraction, it seems.
20:17:13FromDiscord<ambient> GTK/Nim works pretty well on Windows too in my experience https://ssalewski.de/gtkprogramming.html
20:18:15FromDiscord<ambient> although I haven't tried shipping
20:18:43FromDiscord<ShalokShalom> Yeah, it works.
20:18:49FromDiscord<ShalokShalom> It just doesnt look native.
20:19:01FromDiscord<ShalokShalom> Qt looks on any platform like a native applications.
20:19:06FromDiscord<ShalokShalom> (edit) "applications." => "application."
20:19:31FromDiscord<System64 ~ Flandre Scarlet> Owlkettle seems good and easy to use↵And less hard to understand than QT
20:19:38FromDiscord<ambient> Windows apps don't have a real native look
20:19:41FromDiscord<ambient> anyways...
20:20:02FromDiscord<ambient> I use Windows 11 and the user interface is still a complete mess
20:20:22FromDiscord<ambient> They don't know if they want to be a desktop UI or a tablet UI
20:20:30FromDiscord<ShalokShalom> In reply to @System64 "Owlkettle seems good and": I like its API
20:20:58FromDiscord<ShalokShalom> @System64 ~ Flandre Scarlet You can find here a few options more
20:20:58FromDiscord<ShalokShalom> https://github.com/ringabout/awesome-nim#user-interface
20:21:13FromDiscord<ShalokShalom> And then there is Godot 😛
20:21:26FromDiscord<ShalokShalom> Although I find it hard to get running, I havent really tried.
20:21:50FromDiscord<System64 ~ Flandre Scarlet> https://editor.p5js.org/system64MC/sketches/k8vzhPDLZ↵Can I do that easily with Owlkettle?
20:21:55FromDiscord<ShalokShalom> You might find this interesting: https://medium.com/swlh/what-makes-godot-engine-great-for-advance-gui-applications-b1cfb941df3bhttps://medium.com/swlh/what-makes-godot-engine-great-for-advance-gui-applications-b1cfb941df3b
20:22:01FromDiscord<ShalokShalom> (edit) "https://medium.com/swlh/what-makes-godot-engine-great-for-advance-gui-applications-b1cfb941df3bhttps://medium.com/swlh/what-makes-godot-engine-great-for-advance-gui-applications-b1cfb941df3b" => "https://medium.com/swlh/what-makes-godot-engine-great-for-advance-gui-applications-b1cfb941df3b"
20:22:32FromDiscord<ShalokShalom> In reply to @System64 "https://editor.p5js.org/system64MC/sketches/k8vzhPD": I have not used it. A power user from our community did and he was very happy with it.
20:22:38FromDiscord<ambient> I think there are OpenGL or SDL based UI libraries that are better than Godot, if you don't want to use GDscript
20:22:43FromDiscord<ShalokShalom> And I guess you can. 🙂
20:23:04FromDiscord<ShalokShalom> In reply to @ambient "I think there are": You can use Nim with Godot.
20:23:22FromDiscord<ShalokShalom> And I would rather cut my foot of, before I use OpenGL or SDL instead of Godot. ^^
20:23:24FromDiscord<ambient> In reply to @ShalokShalom "You can use Nim": Last I looked the bindings were not actively maintained
20:23:40FromDiscord<ShalokShalom> You mean those of Godot 4?
20:23:52FromDiscord<ambient> Well, either 3.5 or 4+
20:24:08FromDiscord<ShalokShalom> The Godot 3 ones are a bit outdated, thats correct. And as said, I am not sure it was me and just the binding.
20:24:12FromDiscord<ambient> They changed how they do FFI in Godot 4
20:24:16FromDiscord<ShalokShalom> Yep.
20:25:26FromDiscord<ShalokShalom> I maintain that repo. ↵https://github.com/Vivraan/godot-lang-support
20:25:58FromDiscord<ambient> In reply to @ShalokShalom "I maintain that repo.": That's cool of you
20:26:22FromDiscord<ShalokShalom> Its not much work
20:26:27FromDiscord<ShalokShalom> And I like documentation 😄
20:26:29FromDiscord<ShalokShalom> https://github.com/pragmagic/godot-nim
20:26:38FromDiscord<ShalokShalom> Seems at somewhat maintained.
20:26:50FromDiscord<ShalokShalom> I guess at one point, you dont have much left to do
20:27:05FromDiscord<ShalokShalom> Particularly with a language like Nim.
20:27:16FromDiscord<ShalokShalom> Many languages tried to create bindings, same with Qt.
20:27:26FromDiscord<ShalokShalom> Both dont make it easy (Godot 4 does)
20:27:45FromDiscord<ShalokShalom> And Nim proved in both camps, that decent bindings could be created and maintained with it.
20:27:55FromDiscord<ShalokShalom> It definitely has an advantage there. 🙂
20:28:07FromDiscord<ShalokShalom> I also maintain the list of Qt bindings 😄
20:31:24*arik_ joined #nim
20:33:45arik_Anyone using Mummy?
20:40:40FromDiscord<ShalokShalom> Its quite popular 😄
20:42:28FromDiscord<arik> Indubitably
20:43:05FromDiscord<arik> I had a question about querying with Hoppscotch
20:45:23FromDiscord<spoon> bruh, windows defender really trying to block choosenim from downloading the nim installer because it's somehow android malware
20:45:41FromDiscord<spoon> already solved it but really
20:45:56FromDiscord<ShalokShalom> Nim got quite popular among malware peeps
20:46:09FromDiscord<arik> Probably still better than what McAfee would do
20:46:09FromDiscord<ShalokShalom> So Windows decided at one point, to flag all Nim programs
20:46:24FromDiscord<ShalokShalom> Since this is apparently the obvious thing to do ^^
20:47:16FromDiscord<spoon> has anything ever happened with removing the symbols after compiling?
21:12:15FromDiscord<! Nilts> sent a code paste, see https://play.nim-lang.org/#ix=4k4p
21:12:32FromDiscord<! Nilts> (edit)
21:37:35FromDiscord<System64 ~ Flandre Scarlet> sent a code paste, see https://play.nim-lang.org/#ix=4k4s
21:47:50FromDiscord<huantian> Uh did you try doing a system update first? `sudo pacman -Syu` I’d assume
21:48:53FromDiscord<huantian> It looks an error caused by a partial upgrade
21:53:18FromDiscord<ShalokShalom> @System64 ~ Flandre Scarlet I actually have a small script for that to avoid. It updates, before I install anything, and does a couple of other neat tricks.
21:54:35FromDiscord<ShalokShalom> Its designed for Garuda, but I assume you could adjust it 😛
21:55:08FromDiscord<ShalokShalom> https://codeberg.org/shalokshalom/add-and-friends
22:06:13*arik_ quit (Ping timeout: 256 seconds)
22:07:10FromDiscord<System64 ~ Flandre Scarlet> In reply to @ShalokShalom "<@380360389377916939> I actually have": I tried to update, doesn't work
22:09:15FromDiscord<huantian> In reply to @ShalokShalom "<@380360389377916939> I actually have": if you want to update before you isntall with pacman can't you just `pacman -Syu package-name`
22:09:51FromDiscord<System64 ~ Flandre Scarlet> In reply to @huantian "if you want to": Also tried, still have the errors
22:10:42FromDiscord<ShalokShalom> In reply to @huantian "if you want to": that would also install, if the update fails, as an example
22:10:54FromDiscord<ShalokShalom> there are many small details to respect 🙂
22:10:56FromDiscord<huantian> are you sure? https://packages.msys2.org/package/mingw-w64-x86_64-gcc?repo=mingw64 says that `mingw-w64-x86_64-gcc` is on version 12
22:11:14FromDiscord<ShalokShalom> In reply to @System64 "Also tried, still have": Do a -Syy
22:11:22FromDiscord<System64 ~ Flandre Scarlet> In reply to @ShalokShalom "Do a -Syy": I did -Syyu
22:11:37FromDiscord<ShalokShalom> I would go to the Arch channel and ask there
22:11:54FromDiscord<System64 ~ Flandre Scarlet> I'm not on Arch, I'm on MSys2
22:11:57FromDiscord<System64 ~ Flandre Scarlet> on Windows
22:13:05FromDiscord<ShalokShalom> Then I would ask them 😄
22:54:42FromDiscord<System64 ~ Flandre Scarlet> @ShalokShalom did they said something?
23:06:47FromDiscord<beans> does nim have a cross platform library for creating desktop notifications?
23:06:57FromDiscord<beans> or really just for windows, on linux I can just call notify-send
23:12:20*oprypin quit (Quit: Bye)
23:12:29*oprypin joined #nim
23:42:44FromDiscord<_ 🎄> does Nim have pattern matching?
23:43:00FromDiscord<_ 🎄> I really require it because this code is a mess without it as compared to the previous Ocaml code
23:52:23*azimut quit (Ping timeout: 255 seconds)
23:56:41FromDiscord<System64 ~ Flandre Scarlet> In reply to @_ 🎄 "does Nim have pattern": Regex?
23:57:45FromDiscord<amadan> In reply to @_ 🎄 "does Nim have pattern": Few libraries have implemented it like https://github.com/haxscramper/hmatching