<< 01-04-2024 >>
00:17:47*Mister_Magister quit (Quit: bye)
00:30:39FromDiscord<zumi.dxy> In reply to @tixonochek "i definetely have a": yeah... so there's a protip for any new programming language: get mainstream or FAANG recognition before red-teamers do lol
01:25:27*SchweinDeBurg quit (Quit: WeeChat 4.3.0-dev)
01:25:52*SchweinDeBurg joined #nim
02:19:38*SchweinDeBurg quit (Quit: WeeChat 4.3.0-dev)
02:43:02*une left #nim (#nim)
03:06:40*lucasta quit (Quit: Leaving)
04:01:44*SchweinDeBurg joined #nim
04:27:21*pmp-p quit (Ping timeout: 252 seconds)
04:29:56*pmp-p joined #nim
05:07:51NimEventerNew thread by dwhall256: A template to declare an enum, see https://forum.nim-lang.org/t/11331
05:42:54FromDiscord<.bobbbob> uh, is it normal for the compiler to take 3 gigs of ram when doing a sort of complex regular expression with nim-regex?
05:44:32*pmp-p quit (Ping timeout: 260 seconds)
05:45:09FromDiscord<.bobbbob> and to take 20 minutes with no end in sight?
05:46:51FromDiscord<.bobbbob> that is it compiles fine with debug flag but release flag takes forever, guess I need to rework the logic
05:48:10*pmp-p joined #nim
06:59:29FromDiscord<odexine> In reply to @albassort "i wanna just throw": and what happens when they fail?
07:09:30*junaid_ joined #nim
07:17:09Amun-Ra.bobbbob: I experienced something similar with pegs
07:33:41NimEventerNew thread by Araq: NVIDIA uses Nim!, see https://forum.nim-lang.org/t/11332
07:37:19FromDiscord<odexine> "bait used to be believable"
07:38:09FromDiscord<Elegantbeef> It's better than his last year attempt at humour
07:39:21FromDiscord<odexine> i dont remember what last years was
07:41:04FromDiscord<Elegantbeef> It was as funny as a funeral
07:41:24FromDiscord<Elegantbeef> https://forum.nim-lang.org/t/10058#66293
07:43:22FromDiscord<odexine> oh this
07:43:40FromDiscord<odexine> hella repressed memory i guess
08:00:59*entikan quit (Quit: bye!)
08:07:59*Amun-Ra looks at the date
08:36:34*ntat joined #nim
08:45:47*ntat quit (Quit: Leaving)
09:12:21*pmp-p quit (Ping timeout: 268 seconds)
09:19:57*ntat joined #nim
09:46:58*pmp-p joined #nim
10:47:39*krux02 joined #nim
11:11:24FromDiscord<sOkam! 🫐> sent a code paste, see https://play.nim-lang.org/#pasty=tVvFYsOOVURt
11:13:05FromDiscord<sOkam! 🫐> (edit) "https://play.nim-lang.org/#pasty=HhYZLwAghMrX" => "https://play.nim-lang.org/#pasty=hJGsPrBpGahC"
11:13:20FromDiscord<solitudesf> fmt works on static strings
11:13:23FromDiscord<odexine> Format strings must be compile
11:13:23FromDiscord<odexine> Fuck
11:13:24FromDiscord<solitudesf> known at compiletime
11:13:37FromDiscord<solitudesf> In reply to @odexine "Fuck": get obliterated
11:13:57FromDiscord<sOkam! 🫐> so I cannot read it from a file unless its compiletime? 😔
11:14:02FromDiscord<odexine> Yes
11:14:10FromDiscord<sOkam! 🫐> that sucks 😦
11:14:23FromDiscord<odexine> fmt parses and works on the string on CT
11:14:34FromDiscord<sOkam! 🫐> ic
11:14:42FromDiscord<sOkam! 🫐> I assume `&` does the same?
11:14:50FromDiscord<odexine> You’re gonna need another thing to do it on runtime
11:14:52FromDiscord<odexine> In reply to @heysokam "I assume `&` does": ?
11:14:54FromDiscord<odexine> Oh
11:14:55FromDiscord<odexine> Yes
11:15:30FromDiscord<sOkam! 🫐> wait, does `std/strutils %` work at runtime? that could solve it
11:15:53FromDiscord<solitudesf> it does, since you explicitly pass it the data needed
11:16:18FromDiscord<sOkam! 🫐> zenks
11:32:51FromDiscord<ebahi> sent a code paste, see https://play.nim-lang.org/#pasty=gauGmEEfseJR
11:36:07FromDiscord<ebahi> i tried making a type called mfaMethod and adding it as an argument but i felt like it added more complexity
11:36:52FromDiscord<sOkam! 🫐> sent a code paste, see https://play.nim-lang.org/#pasty=jWYrXvJwgxGd
11:37:08FromDiscord<sOkam! 🫐> (edit) "https://play.nim-lang.org/#pasty=yPQUyLjOmzCU" => "https://play.nim-lang.org/#pasty=waTbVzVIKPoc"
11:37:20FromDiscord<ebahi> ill try it
11:37:49FromDiscord<sOkam! 🫐> you can make assign it a value into the argument, to make it a default argument, incase you only choose the style ocassionally
11:37:59FromDiscord<sOkam! 🫐> (edit) removed "make" | removed "it"
11:42:51*Mister_Magister joined #nim
12:05:54FromDiscord<norax.2> how do i handle like ctrl +c
12:05:56FromDiscord<norax.2> on nim
12:06:07FromDiscord<norax.2> whitout putting a whole try arround my code
12:06:23FromDiscord<norax.2> (edit) "whitout putting a ... whole" added " try around my" | removed "try arround my"
12:09:49FromDiscord<norax.2> In reply to @ebahi "hello, im having a": but the real thing is
12:09:54FromDiscord<norax.2> why are you making a log in
12:10:05FromDiscord<norax.2> when literally you can use a simple url whit the bot token
12:10:12FromDiscord<norax.2> thats for a user bot
12:14:20FromDiscord<odexine> Or a third party client
12:18:33FromDiscord<norax.2> In reply to @odexine "Or a third party": yeah i dont mind(at all) but i whant to just warning that the thing breaks discord tos
12:18:50FromDiscord<odexine> Okay
12:18:51FromDiscord<norax.2> i also did one
12:19:03FromDiscord<norax.2> but yeah they will prb ban the account that uses the client
12:19:19FromDiscord<norax.2> so use an alt
12:35:30FromDiscord<whisperecean> I am using parseopt and I'd like to have something like this --file (or -f) and then /tmp/file.txt and not -f="/tmp/file.txt"
12:35:33FromDiscord<whisperecean> How to achieve this?
12:35:47FromDiscord<ebahi> In reply to @norax.2 "but the real thing": its a simple project i made as a wrapper for discord's private user api
12:55:30FromDiscord<nnsee> In reply to @whisperecean "How to achieve this?": https://nim-lang.org/docs/parseopt.html#nimshortnoval-and-nimlongnoval
12:57:06FromDiscord<nnsee> alternatively, use something like cligen. imo just much nicer to work with
13:05:52*Mister_Magister quit (Excess Flood)
13:07:07*Mister_Magister joined #nim
13:10:48FromDiscord<odexine> me not wondering what a noval was, thinking its some sort of strange oval
13:21:03FromDiscord<nnsee> i like my novals long
13:50:14*def- quit (Quit: -)
13:50:32*def- joined #nim
13:55:31FromDiscord<whisperecean> In reply to @nnsee "alternatively, use something like": I have noticed...this is really terrible to work with
13:57:08FromDiscord<nnsee> sent a code paste, see https://play.nim-lang.org/#pasty=IUjLqkkElagO
13:58:59FromDiscord<odexine> it depends on how you want to model it
13:59:11FromDiscord<odexine> some people dont want that (i personally dont)
14:04:18FromDiscord<whisperecean> But I dont get the doc. I want -f file.txt but I cant find that option
14:04:27NimEventerNew thread by Charles: How to use dynamic dispatch with inheritance without converting to the appropriate subtype by hand?, see https://forum.nim-lang.org/t/11335
14:06:56FromDiscord<whisperecean> Or should I just specify longNoVal for f,file?
14:07:25FromDiscord<nnsee> In reply to @whisperecean "Or should I just": hm, it looks like you maybe can't do that with short opts
14:08:21FromDiscord<nnsee> In reply to @whisperecean "Or should I just": no, you'd want the opposite
14:08:38FromDiscord<nnsee> `f` and `file` are both _expected_ to have a value
14:09:16FromDiscord<whisperecean> So what should I use?
14:11:22FromDiscord<whisperecean> I rarely see utils which work like this ... -option="test" --file="/tmp/file.txt"
14:11:45FromDiscord<nnsee> https://play.nim-lang.org/#pasty=RdTpZfMTshtD it looks like the short option does not like spaces. with no space, it works as expected
14:15:34FromDiscord<nomad0529> sent a code paste, see https://play.nim-lang.org/#pasty=TaxVHBbMtouz
14:16:10FromDiscord<whisperecean> @nnsee So there are 2 options to do this?
14:17:27FromDiscord<_nenc> In reply to @whisperecean "So what should I": I think maybe parseopts can't do that↵Nim has it's custom in cli options `--a:v -b:v`
14:18:24FromDiscord<whisperecean> I see. Ill try cligen then
14:19:53FromDiscord<_nenc> i am afraid maybe cligen also can't do that \:(
14:20:31FromDiscord<whisperecean> wait what?
14:20:35FromDiscord<whisperecean> this is basic functionality....
14:21:40FromDiscord<solitudesf> it can. you can even use cligen/parseopt3 to have drop-in alternative to stdlib parseopt that works like you expect.
14:22:40FromDiscord<whisperecean> I just have a simple -a int -f file.txt
14:22:47FromDiscord<whisperecean> (edit) "I just ... have" added "want to"
14:23:01FromDiscord<solitudesf> okay
14:23:08FromDiscord<whisperecean> That's all
14:23:12FromDiscord<solitudesf> good
14:23:22FromDiscord<_nenc> I am not very familiar with cli libs in Nim, but I don't know any lib that majorly works like that.
14:23:45FromDiscord<whisperecean> So what do I do to achieve this result?
14:23:57FromDiscord<solitudesf> In reply to @whisperecean "So what do I": use the hecking cligen?
14:24:06FromDiscord<solitudesf> the guy is feeding you misinfo
14:24:29FromDiscord<_nenc> you can implement it \:D↵or cligen/parseopt3, idk that but I hope it supports that
14:25:14FromDiscord<_nenc> yeah, I checked it, I think it works like what you want
14:25:43FromDiscord<_nenc> sent a code paste, see https://play.nim-lang.org/#pasty=JCZPEZrrTMnN
14:33:50FromDiscord<treeform> In reply to @chronos.vitaqua "Should I just use": If you don't have to use JWT, don't. It's better to just use our own random token that you compare with the database. I find JWT just stupid.
14:34:30FromDiscord<treeform> Also guzba/jwttea is a pure nim lib that does not require openssl.
14:34:31FromDiscord<whisperecean> But from the dispatch proc fun how do I assign my variables?
14:34:41FromDiscord<treeform> (edit) "guzba/jwttea" => "guzba/jwtea"
14:35:01FromDiscord<Robyn [She/Her]> In reply to @treeform "If you don't have": Fair enough, what's the stupid thing about JWT? The vulnerabilities it exposed?
14:35:25FromDiscord<treeform> no vulnerabilities, its just complexity overkill for most projects.
14:36:57FromDiscord<Robyn [She/Her]> Fair enough
14:37:08FromDiscord<treeform> I always want to avoid complexity, I can't avoid using JWT because google requires it.
14:37:30FromDiscord<treeform> A random auth token is way simpler
14:38:28FromDiscord<treeform> then JWT (which requires openSSL and eliptic key cryptography) and double rounds to take your long live token into a short lived token etc...
14:38:56FromDiscord<treeform> I wasted days dealing with that, don't make your users waste their time!
14:39:16FromDiscord<treeform> use random auth token
14:39:42FromDiscord<Robyn [She/Her]> Fair enough, my token format should be secure enough either way since it only stores minimal data
14:39:53FromDiscord<Robyn [She/Her]> And none of it is private data really
14:40:15FromDiscord<Robyn [She/Her]> Just an encrypted string of `{user id}.{token id}.{timestamp}`
14:42:47FromDiscord<treeform> why not just "token id"
14:42:50FromDiscord<treeform> then you look it up in the db?
14:44:28FromDiscord<Robyn [She/Her]> That's a good point actually
14:44:35FromDiscord<Robyn [She/Her]> Originally it was gonna be unencrypted
14:45:36FromDiscord<odexine> depends on you, "looking up a db" could be too much complexity compared to jwt. but in most cases it is not
14:45:39FromDiscord<Robyn [She/Her]> Actually, unencrypted would probably be easier to do and it's not like it'd cause any issues
14:45:57FromDiscord<odexine> i dont think theres a reason to encrypt random gen tokens
14:46:03FromDiscord<Robyn [She/Her]> In reply to @odexine "depends on you, "looking": Looking it up is easy and painless so not a problem
14:46:14FromDiscord<Robyn [She/Her]> In reply to @odexine "i dont think theres": Yeah I'm realising that now
14:48:11FromDiscord<treeform> less complexity is my goal when I build stuff
14:48:28FromDiscord<treeform> you shoud be running all your trafic though https so everything is always encrypted
14:48:50FromDiscord<Robyn [She/Her]> In reply to @treeform "you shoud be running": Also true yeah
14:49:13FromDiscord<treeform> (https is very easy to do with services like cloud flare or let's encrypt plugin for nginx)
14:50:56FromDiscord<Robyn [She/Her]> Yeah I'm planning on using Cloudflare so no problems here
14:53:24FromDiscord<odexine> ~~caddy~~
14:54:07FromDiscord<Robyn [She/Her]> Also probably Caddy, yeah-
14:55:42FromDiscord<whisperecean> In reply to @whisperecean "But from the dispatch": Any hints?
14:56:33*pmp-p quit (Ping timeout: 252 seconds)
14:57:39*pmp-p joined #nim
14:58:49FromDiscord<vindaar> Can you clarify? When you have some argument `file` of the proc you dispatch via cligen, the `file` argument will contain the data you pass via `-f FOO` or `--file FOO`↵(@whisperecean)
15:00:07FromDiscord<whisperecean> That argument will contain the data but based on that argument I want to call a different proc or rather pass that data to that proc
15:00:26Amun-Ranim arg parser does not support posix style -f foo / --file foo
15:01:49FromDiscord<vindaar> then inside of the proc you simply call another proc with the `file` passed on? I'm confused↵(@whisperecean)
15:02:06FromDiscord<whisperecean> Like inside of the fun proc?
15:02:09FromDiscord<whisperecean> I dont get it
15:03:21FromDiscord<vindaar> sent a code paste, see https://play.nim-lang.org/#pasty=rGDimBHJpkIg
15:03:49Amun-Raah, cligen
15:07:32FromDiscord<whisperecean> sent a code paste, see https://play.nim-lang.org/#pasty=VwezYPIAKWyR
15:08:32FromDiscord<vindaar> feel free to ping me if you have further questions about cligen
15:09:07FromDiscord<whisperecean> Just one more thing. How do I make parameters optional and how do I just print version after that?
15:10:07FromDiscord<whisperecean> Ah I think I got it
15:17:00FromDiscord<whisperecean> Can I overwrite the usage as well?
15:26:11NimEventerNew thread by ninovanhooff: Am iterating over this sequence correctly?, see https://forum.nim-lang.org/t/11336
15:31:01*pmp-p quit (Ping timeout: 272 seconds)
15:42:33FromDiscord<vindaar> optional\: just give the argument a default value.↵overwrite the usage\: what do you mean?
15:46:41FromDiscord<whisperecean> This is by default generated: Usage:↵ fun [REQUIRED,optional-params]
15:52:18FromDiscord<vindaar> You want to customize the autogenerated help? You can pass a `help` argument, see here for example\: https://github.com/c-blake/cligen/blob/master/examples/dups.nim#L164-L180
15:52:47*pmp-p joined #nim
15:55:03FromDiscord<anuke> How do I get the defined value for a symbol passed to the compiler `-d:foo:value` in a source file?
15:55:15FromDiscord<whisperecean> In reply to @vindaar "You want to customize": I think that help argument does not modify that usage info
15:58:02FromDiscord<vindaar> So what exactly do you want to change? The word 'usage'?
15:58:10FromDiscord<norax.2> In reply to @whisperecean "I think that help": try macro lib?
15:58:16FromDiscord<norax.2> ops
15:58:17FromDiscord<norax.2> wrong reply
15:58:40FromDiscord<norax.2> In reply to @anuke "How do I get": try macro lib
16:00:47FromDiscord<nomad0529> how do i get a string from a FlowVar[system.string]?
16:01:10FromDiscord<whisperecean> In reply to @vindaar "So what exactly do": Correct. That whole helper usage string
16:07:29FromDiscord<vindaar> See the wiki for the config file\: https://github.com/c-blake/cligen/wiki/Example-Config-File and at the bottom the `useHdr` line etc
16:08:11FromDiscord<norax.2> sent a code paste, see https://play.nim-lang.org/#pasty=trYeIuZNGJVc
16:08:25FromDiscord<norax.2> tried it
16:08:26FromDiscord<norax.2> works
16:09:43FromDiscord<whisperecean> In reply to @vindaar "See the wiki for": I see. Does that mean it can only be overwritten through that config file?
16:12:07FromDiscord<anuke> sent a code paste, see https://play.nim-lang.org/#pasty=kWhPlzameNKK
16:13:00FromDiscord<norax.2> but
16:13:04FromDiscord<norax.2> use = and not :
16:13:07FromDiscord<norax.2> else is buggy
16:13:17FromDiscord<norax.2> (edit)
16:15:06FromDiscord<whisperecean> sepChars = "=" # Do not use Araq's weird : convention
16:15:09FromDiscord<whisperecean> lol love this
16:15:12FromDiscord<vindaar> sent a code paste, see https://play.nim-lang.org/#pasty=TLLQNuNGGncy
16:15:22FromDiscord<norax.2> In reply to @norax.2 "or else is buggy": also
16:15:35FromDiscord<norax.2> it doesnt work like you set it to a value and it spits the value
16:15:39FromDiscord<norax.2> jsut says true or false
16:15:48FromDiscord<norax.2> that need to be handled in the script too
16:16:00FromDiscord<norax.2> (edit) "that need to be handled in the script too ... " added "if you need the settage"
16:17:30NimEventerNew thread by Nlits: How to properly make a static type-checking macro?, see https://forum.nim-lang.org/t/11337
16:21:56FromDiscord<whisperecean> In reply to @vindaar "It depends on your": Thanks that works. Did you know this already or you had to check it in the code?
16:28:30FromDiscord<solitudesf> In reply to @anuke "How do I get": `const foo {.strdefine.} = ""`
16:29:03FromDiscord<solitudesf> nvm, you already found it
16:30:11*xet7 quit (Remote host closed the connection)
16:31:11FromDiscord<vindaar> roughly I knew, but I had to check to not tell you rubbish↵(@whisperecean)
16:38:24*xet7 joined #nim
16:48:13FromDiscord<Robyn [She/Her]> I'm trying to provide an abstraction layer over types of storage providers but I want to avoid using generic types everywhere, but it's also the only way I can correctly use the right procs for the storage provider implementation... :/
16:48:32FromDiscord<Robyn [She/Her]> sent a code paste, see https://play.nim-lang.org/#pasty=uNvXipMEdQkJ
16:49:13FromDiscord<Robyn [She/Her]> I don't know if `methods` are suitable here? Iirc generics with methods were deprecated
16:50:08FromDiscord<Robyn [She/Her]> Generics are also the reason why I can't store the procedures in the obejct itself
16:50:14FromDiscord<Robyn [She/Her]> (edit) "`methods`" => "`method`s"
16:50:33FromDiscord<Robyn [She/Her]> sent a code paste, see https://play.nim-lang.org/#pasty=INYBTOvPUPJj
16:50:37FromDiscord<Robyn [She/Her]> Example procedure there
16:51:20FromDiscord<Phil> Honestly I'd rather name what you call StorageProvider "Repository", but that's besides the fact I guess
16:51:43FromDiscord<Phil> In your case I don't see the benefit of using methods tbh. Why not use normal procs and generics?
16:52:33FromDiscord<Phil> Ah, I assume you want to be able to store Providers in a hashmap or seq and pass them around to do things with them
16:52:47FromDiscord<Phil> In that case you can still use methods and write yourself generic helper procs that they call.
16:52:51FromDiscord<Robyn [She/Her]> In reply to @isofruit "In your case I": `MineHandle` would need to store the type everywhere, no?
16:53:18FromDiscord<Robyn [She/Her]> In reply to @isofruit "Honestly I'd rather name": Oh? Why `Repository`? Seems like a better name but what's the reasoning?
16:54:09FromDiscord<Phil> In reply to @chronos.vitaqua "`MineHandle` would need to": `MineHandle` can only have one behaviour in one instance. If you want different instances of `MineHandle` to have different storage behaviour you'll need to give `MineHandle` fields to store e.g. an `extract` proc
16:54:18FromDiscord<Phil> (edit) "instance." => "instance in either scenario."
16:54:48FromDiscord<Phil> In reply to @chronos.vitaqua "Oh? Why `Repository`? Seems": Repository is just the default name for database abstractions that I'm aware of.
16:54:57FromDiscord<Phil> Used by e.g. "Spring"
16:55:00FromDiscord<Phil> (edit) ""Spring"" => "Spring"
16:55:59FromDiscord<Robyn [She/Her]> I don't think I understand? I was planning of making it so `mineHandle.extract(1, myVarObj)` call `extractImpl(mineHandle.provider, 1, myVarObj)`
16:56:03FromDiscord<Phil> Or rather I think java's general DB abstraction just calls them all repository, not even Spring specific
16:56:30FromDiscord<Robyn [She/Her]> Just wondering if there's a way to make it so the associated type info isn't important
16:57:30FromDiscord<Phil> In reply to @chronos.vitaqua "I don't think I": Now I don't understand what you're going for, what calls what?
16:58:17FromDiscord<Phil> Ohhh wait, sorry, now I see
16:58:34FromDiscord<Phil> You just don't want to be bothered about having to specify the ProviderType all the time
16:58:38FromDiscord<Robyn [She/Her]> Glad I didn't have to form a cohesive sentence xD
16:58:43FromDiscord<Robyn [She/Her]> In reply to @isofruit "You just don't want": Yep!
16:58:56FromDiscord<Phil> I misread your code while glancing over it as MineHandle being a custom type of provider
16:59:02FromDiscord<Phil> Thus the hickup, well let's see
16:59:06FromDiscord<Robyn [She/Her]> Aaah okay!
16:59:14*SchweinDeBurg quit (Quit: WeeChat 4.3.0-dev)
16:59:19FromDiscord<Robyn [She/Her]> `MineHandle` is basically a reference to a table
16:59:37FromDiscord<Robyn [She/Her]> sent a code paste, see https://play.nim-lang.org/#pasty=LuSXvHDFnkrl
16:59:39FromDiscord<Phil> Okay fair, if that's how you want to roll with it then methods are the only thing you can do. Use methods that call generic procs then I guess
17:01:50FromDiscord<Robyn [She/Her]> Makes sense, oh well, thanks Phil!
17:02:03FromDiscord<Phil> Personally I'd still go with generics. But methods with what you want seem to be the correct way to go about it.
17:02:41FromDiscord<Robyn [She/Her]> I'll probably use generics then, seems like a better situation and I don't need to think about overhead from dynamic dispatch (even if it wouldn't be much either way)
17:04:43FromDiscord<Phil> I tend to just not go OO the majority of the time.↵The mental model of an object that defines its own thing for encapsulation rarely sits too well with me. I prefer using modules for that.
17:05:32FromDiscord<Phil> But that also is a very easy way of building yourself a ball of mud. Then again, that's mostly a concern when there's multiple devs in the project or if you yourself are undisciplined
17:06:30FromDiscord<Robyn [She/Her]> Fair
17:27:22FromDiscord<mratsim> In reply to @isofruit "But that also is": If the language doesn't support OO, like Go, Rust or C, there is no concern.
17:34:08*Guest5853 joined #nim
17:34:19*ntat quit (Remote host closed the connection)
17:34:21*Guest5853 is now known as ExclamationPoint
17:35:28ExclamationPointwhat's nim's use-case? is it supposed to replace something or connect something? why would someone replace c++ with it? or use it alongside c++, for example?
17:41:54*junaid_ quit (Remote host closed the connection)
17:42:35FromDiscord<Robyn [She/Her]> In reply to @ExclamationPoint "what's nim's use-case? is": Imo it's just much cleaner, I've never used C++ tbh, but Nim lets you use libraries from C++ if you use the C++ backend which is always a plus
17:43:29FromDiscord<Phil> In reply to @mratsim "If the language doesn't": I mean yeah but I also don't know how those structure their projects
17:43:45FromDiscord<Phil> And C gave rise to C++ because people felt a need for it in the sense that C seemingly didn't provide decent enough structuring capabilities
17:44:36FromDiscord<odexine> In reply to @isofruit "I mean yeah but": PSST
17:44:58FromDiscord<Phil> Post super stress torture?
17:45:23FromDiscord<Phil> Practically stupidly simple teachings?
17:45:43FromDiscord<Phil> Pirating simple software terribly?
17:46:06FromDiscord<Phil> Cmon rika, reply, I'm starting to slowly run out of stupid versions of this acronym
17:46:31FromDiscord<Robyn [She/Her]> Peanut Shaped Silly Tincan
17:47:26FromDiscord<Phil> Basically, nim was my first primarily OO language
17:47:49FromDiscord<Phil> And I never used it in a work setting with multiple devs, only in small scale projects
17:48:30FromDiscord<Phil> So I'm not 100% how well you can constrain structure to prevent other devs from fucking up.↵You have the advantage over me there @mratsim as afaik you did work in a nim team before
17:48:35FromDiscord<Phil> or right now? not sure
17:50:12NimEventerNew thread by Nlits: How to remove the N_LIB_PRIVATE from exported c code., see https://forum.nim-lang.org/t/11338
17:53:44ExclamationPointRobyn, if it's a replacement for C++, what's cleaner about the language? it looks like Pascal to me. is it the indent scope vs braces? does the time spent learning a new language eventually pay off in some measurable way?
17:55:26ExclamationPointPhil, if C++ came about because C was not well structured, what was the motivation behind Nim?
17:56:12FromDiscord<Phil> In reply to @ExclamationPoint "<@180601887916163073>, if C++ came": Be an actually easily graspable language with an actual type system as opposed to whatever silly joke C has
17:56:47FromDiscord<Phil> Like, I respect C for its capabilities and to a degree I'm at this point capable of reading C code
17:57:02FromDiscord<Phil> But that also means I was exposed to how they name and structure APIs and that just makes me want to go bald
17:57:10FromDiscord<Phil> To avoid the pain of pulling my hair out.
17:57:39FromDiscord<Phil> "Just pass a pointer, what's behind the pointer? Who knows, not me!"
17:57:41FromDiscord<odexine> In reply to @isofruit "Cmon rika, reply, I'm": More like “psst use a language without OOP wink wink wink wink”
17:58:16*krux02_ joined #nim
17:59:01FromDiscord<Phil> I started with nim in webdev and found it pretty easy to write code that was easily understandable and despite having no knowledge of metaprogramming and compile-time programming I started dealing with both in my very first nim project basically.↵↵Nim has a way of making various programming paradigms accessible that most other languages just don't.
17:59:18FromDiscord<Phil> Except for OO. If you want to see the pinacles of OOP then nim won't lead you there
18:00:29*krux02 quit (Ping timeout: 240 seconds)
18:00:42FromDiscord<Robyn [She/Her]> In reply to @ExclamationPoint "Robyn, if it's a": I came from Python originally, compared to when I tried to use C++ (and Java), it feels less verbose in general, the lack of classes is comfortable after I got used to it, and even the relatively low-level code with pointers written in Nim still looks readable imo
18:00:43FromDiscord<Phil> Nim as a language provides a lot of tools that allow you to dive into things but you're not forced to use any of them.↵At this point I've appreciated a lot of the features enough to contribute myself.
18:01:13FromDiscord<Phil> Likehell, there's a GTK wrapper with an actually declarative DSL to write Widget trees with event listeners etc. included (owlkettle)
18:02:19FromDiscord<Phil> The only other wrapper I'm aware of that is capable of that is relm4
18:08:19FromDiscord<Robyn [She/Her]> Hey Phil, you know more SQL than me, is it possible to check whether a table was created or if it already exists?
18:08:32ExclamationPointPhil, pointers have types for a reason. nothing forces you to point to memory of that type, but that's because memory management is optional. if you use managed memory constructs in C/C++, you would always point to memory of that type. this is kind of important when you have a systems programming language, so i'm not sure what Nim does to make this work, but it wouldn't be easy to do systems programming without it.
18:08:52FromDiscord<Phil> In reply to @chronos.vitaqua "Hey Phil, you know": This will blow your mind, check for SQL syntax "IF EXISTS" or the like 😛
18:09:01FromDiscord<Robyn [She/Her]> In reply to @ExclamationPoint "<@180601887916163073>, pointers have types": A lot of libs seem to do `void` which is something I've seen
18:09:30FromDiscord<Robyn [She/Her]> In reply to @isofruit "This will blow your": I know I can do `CREATE TABLE IF NOT EXISTS`, but I want to see if a table already existed or if it was created :P
18:09:35FromDiscord<Phil> In reply to @ExclamationPoint "<@180601887916163073>, pointers have types": It is possible to pass pointers in nim, it's just not the "default way" that you do things
18:10:16FromDiscord<Phil> Like how rust has unsafe, nim typically guides you towards using value-types, then ref types and only as a last resort actual pointers
18:10:21ExclamationPointPhil, that's like blaming the language for the way it's used. if I used arrays of characters instead of strings, it would be my fault, no?
18:11:57FromDiscord<Phil> sent a long message, see https://pasty.ee/JzwDxrmbzOhf
18:11:59ExclamationPointPhil, C++ also has value and reference passing, and actively discourages pointers. even dynamic memory allocation is poor form. it's available, but no one promotes that kind of programming
18:13:50ExclamationPointsizeof() def returns the allocated size of an array in bytes in C.
18:13:56FromDiscord<Phil> I'm likely not a good person to give an unbiased view on this one.↵To me, nim as a language just got more "out of my way" than other languages I've used so far (which is only marginally more than half a dozen tbf).↵A lot of "opt-in" features that are useful everywhere and allows you to unlock concepts when you want to deal with them.
18:14:18FromDiscord<Phil> (edit) "them." => "them, not up-front."
18:14:45FromDiscord<Phil> The topic comes up regularly enough even on the subreddit, I'll look for a link to the last 1-2 times I wrote a reply to it
18:15:11Amun-RaRobyn: hmm, do "select 1 from table" and check for error?
18:15:23ExclamationPointsorry, I didn't join to debate which is better, but it seems that there are a lot of misconceptions about the way C and C++ work, and I don't see a big motivation for creating another language when using the existing ones correctly make things simpler. thanks for the answers tho
18:17:30ExclamationPointRobyn, what you want is to check the schema for a table/view/procedure/whatever you're checking. this doesn't exist in formalized SQL but can be done different ways depending on which RDBS you're using. search for 'how to check if table exists in X' where X is mysql or mssql or oracle or sqlite or whatever
18:20:22FromDiscord<Phil> https://www.reddit.com/r/nim/comments/167qzdl/why_to_learn_nim/jyu23vl/↵https://www.reddit.com/r/nim/comments/17j3833/for_all_the_nim_devs_here_whats_a_killer_feature↵↵Those are 2 topics that deal with the question that likely give you a better idea than I can during dinner prep 😛
18:22:15FromDiscord<Phil> In reply to @ExclamationPoint "<@180601887916163073>, C++ also has": I was honestly only mentioning C++ once in the beginning due to the entire OOP topic with Robyn earlier, everything after mostly relates to C (which I have passing experience with due to having to write bindings to it for a wrapper) and java
18:22:49nisstyreyou can basically write C in Nim if you want, and I have done that a bunch of times because you sometimes have to wrap C libraries
18:23:11nisstyrezero cost iterators are useful to wrap C libraries because you can hide all the nasty pointer arithmetic in it, then free your resources if you have to at the end
18:23:31nisstyretoo bad more C libraries don't let you pass in a custom allocator though
18:24:22FromDiscord<Phil> As for why not just use C++ - You'd need to ask somebody that uses both to get decent arguments.↵I've read mostly that due to feature bloat it's a very footgun language due to the dozen different standards you may encounter in a given codebase. The many different C++ variants are what made me go to other languages for learning first.
18:24:48FromDiscord<Phil> Its OOP focus also didn't particularly peak my interest when I already dealt with OOP focused languages before
18:25:19FromDiscord<Phil> (edit) "Its OOP focus also didn't particularly peak my interest when I already dealt with OOP focused languages before ... " added "and didn't find the paradigm too enticing, nor to lead to particularly flexible code."
18:26:29nisstyreyou should watch https://www.youtube.com/watch?v=rX0ItVEVjHc
18:35:50FromDiscord<tapeda> In reply to @ExclamationPoint "sorry, I didn't join": In an interview the creator of nim mentions his original motivation was seeing a a gap in the language landscape that was a systems programming language that combined the best parts of Ada(really cool), python(really useable), and lisp(really extensible) I hope that helps get past the more superficial parts about other languages sharing similar features, and nim might be older than yo
18:36:41FromDiscord<Robyn [She/Her]> In reply to @Amun-Ra "Robyn: hmm, do "select": How about if there's no entries? :p
18:38:59Amun-RaRobyn: you still have "1"
18:41:11Amun-Raand no entries - no error
18:42:48FromDiscord<Robyn [She/Her]> Oh okay then, thanks!
18:48:44*xutaxkamay_ joined #nim
18:49:01*xutaxkamay quit (Read error: Connection reset by peer)
18:49:36*xutaxkamay_ is now known as xutaxkamay
18:51:15FromDiscord<Phil> sent a long message, see https://pasty.ee/DCFmutIJlOOw
18:54:10*xutaxkamay quit (Read error: Connection reset by peer)
18:54:15FromDiscord<Phil> I agree with the statement though that trying to model data in objects with specific interactions adds unnecessary constrains and complexity that ultimately make your life harder
18:54:22*xutaxkamay_ joined #nim
18:55:05*xutaxkamay_ is now known as xutaxkamay
18:55:22nisstyrePhil: you can safely assume your user will be using your web app on either a modern phone, or relatively modern desktop or laptop PC usually
18:55:37nisstyrewhen I did webdev stuff, there was a range of user agents that we bothered to support
18:55:44FromDiscord<Phil> That's fair, but that's a software platform
18:55:54FromDiscord<Phil> That being the browser
18:56:11nisstyreOk yeah, you assume a version of the browser that assumes a hardware platform, so it's transitive
18:57:10nisstyreIf you removed all the workarounds in some web apps to support older platforms I bet they'd be faster
18:57:12FromDiscord<Phil> Do browsers assume specific hardware platforms?↵Even then, that's only Frontend webdev, backend webdev also runs on hardware but there you can only assume "I have a CPU with multiple threads"
18:57:25nisstyreThey do yes
18:57:44nisstyrefor backend, I'm sure you can make some assumptions? Like you're not gonna suddenly be running on ARM or MIPS when you didn't expect it
18:57:47FromDiscord<Phil> I mean, beyond just "X86-64 level 0"
18:58:02FromDiscord<Phil> Or whatever the level range was called that went from 1-4
18:58:07FromDiscord<Phil> (edit) "0"" => "1""
18:58:14nisstyreah ok, well I'd say that's a problem with modern web backend development. "We must slice this up to run on tiny serverless workers"
18:58:32nisstyreyou can do the lowest common denominator though
18:59:20FromDiscord<Phil> I feel like I can agree with you there, but also that doesn't feel like it applies the constrains like the statement "Your platform is the hardware, not software" to me implies that there are
18:59:31nisstyrehe also makes the point about people assuming things are done one at a time, which I see a lot in that type of code
19:01:09ExclamationPointPhil, thanks for the links!
19:01:23nisstyrehow come nobody has done any sort of frontend framework that exploits multiple cores for rendering stuff
19:01:28nisstyreor have I missed that
19:01:44FromDiscord<Phil> In reply to @nisstyre "how come nobody has": I mean, can you? The browser is who renders things, not you, you just provide the HTML representation
19:01:49nisstyreDOM isn't threadsafe which makes it harder
19:02:11nisstyreI was thinking about all these libraries that do their own shadow DOM and then sync them up
19:02:15ExclamationPointtapeda, I'm so old when I started using the internet it only came in black and white text
19:02:31nisstyrebut yeah you're really limited by the browser engine
19:02:48FromDiscord<Phil> You can still multithread in the browser, but that really only pays off when you do a lot of computation in the browser
19:02:55FromDiscord<Phil> And the vast majority of websites just don't
19:03:01FromDiscord<Phil> So nobody bothers
19:03:50nisstyreold graphics rendering systems used to work by you setting up a scene and then you send it to the renderer and it renders it
19:03:59nisstyrethat's pretty much how the web works
19:04:41FromDiscord<Phil> Which is a shame because with the reactive programming tooling out there it's really easy to just integrate a thread into a system that does a whole lot of message passing and to do an action you just pass a message with a specific type to a webworker from a global message queue and it messages the result back into the global message queue which other places in your codebase then react to that specific message type
19:05:31FromDiscord<Phil> Which is a setup you'll already have in place in a lot of places that write Single-page-applications, so integrating is easy - just never necessary and thus never useful =/
19:05:41FromDiscord<.bobbbob> In reply to @.bobbbob "uh, is it normal": I reworked the code to use std/re instead of nim-regex and it compiles normally and is immensely faster so am I doing something wrong, is there a bug or what? Should I make a bug report?
19:07:10FromDiscord<Phil> In reply to @nisstyre "how come nobody has": Sidenote, even then, everything I've seen around for frontend has always been frontend rendering being single threaded
19:08:07FromDiscord<Phil> You might pass of backend-computation to another thread (actor-model or the like), but the logic of how things look and should be rendered I've always encountered as conceptually being single threaded.
19:08:39nisstyrepart of the problem too is the bloated event system because it has to trigger so many JavaScript events
19:08:49nisstyreand anything can then block the entire thing
19:09:35nisstyreso you're just supposed to not be stupid and run expensive computations
19:10:35FromDiscord<Phil> But even outside of that, GTK I've seen the same
19:10:56FromDiscord<Phil> Single frontend thread, for backend computation use their API to spawn one-off threads or use their queue if you want multithreading ala actor-model
19:11:41nisstyreyes GTK is trash too, can't do anything safely with multiple threads. I've tried it before.
19:11:48FromDiscord<Phil> That's my only three points of reference:↵- colleagues talking about iOS mobile dev using mostly single frontend thread and however many backend-threads as needed, likely just one↵- GTK↵- Web
19:11:49nisstyreyou have to do expensive polling with queues and stuff
19:11:56FromDiscord<Phil> (edit) "reference:↵-" => "reference tbh:↵-"
19:11:59FromDiscord<tixonochek> Hey, why there isnt a ukrainian language-specific channel? I mean, there are a lot of languages there.
19:13:00nisstyrePhil: https://en.wikipedia.org/wiki/PHIGS this is basically how web rendering works, at least if you're just doing basic HTML stuff
19:13:36nisstyrewhich makes sense since the WWW came about in the late 80s right
19:13:43nisstyreso they were probably inspired by that
19:14:25nisstyreSo we're all using a system that was obsolete by the mid 90s in how it's architected...
19:15:11FromDiscord<Phil> In reply to @nisstyre "you have to do": Coooouuuuuldn't you use stuff like epoll to let the thread sleep and wake it up via a posix signal to avoid unnecessary polling?
19:15:23nisstyreI didn't try, but yeah maybe
19:15:38FromDiscord<Phil> Note I only barely know what I'm talking about here, I roughly have an understanding of how asyncdispatch work, I just started implementing an actor-model-like multithreading runtime
19:15:47FromDiscord<Phil> Which uses chronos and a mechanism like that
19:16:18nisstyrehttps://github.com/weskerfoot/Schemekit/blob/b6a3ab3b82c835ae8582708fc0a9a93f8b37c7ac/browser.c#L21
19:16:25nisstyreit was this function that was causing me problems
19:16:37nisstyreI'm sure there's a less dumb way of using it though
19:17:42NimEventerNew thread by ggibson: Help with channels and threads, see https://forum.nim-lang.org/t/11339
19:17:49nisstyreok there is a blocking version, I don't remember why but it wouldn't have worked for some reason
19:18:53nisstyreprobably because I wanted the same thread to be responsive to other stuff
19:20:09FromDiscord<Phil> Good to see that good ol print debugging is alive and well in C as well 😄
19:20:37nisstyreof course
19:21:36FromDiscord<Robyn [She/Her]> In reply to @tixonochek "Hey, why there isnt": I don't know anything about Ukrainian or other languages, but this looks similar to the script Ukrainian uses? https://discord.com/channels/371759389889003530/771116956484894770
19:22:32FromDiscord<Robyn [She/Her]> https://discord.com/channels/371759389889003530/725426142454087761 this one too
19:22:45FromDiscord<nnsee> In reply to @chronos.vitaqua "I don't know anything": that's bulgarian
19:22:49FromDiscord<nnsee> In reply to @chronos.vitaqua "https://discord.com/channels/371759389889003530/725": that's russian
19:22:51FromDiscord<tixonochek> What do you mean by the script
19:22:53FromDiscord<Robyn [She/Her]> Ooh
19:22:54FromDiscord<tixonochek> In reply to @nnsee "that's bulgarian": Exactly
19:23:05FromDiscord<Robyn [She/Her]> In reply to @tixonochek "What do you mean": Script as in the alphabet it uses
19:23:12FromDiscord<Robyn [She/Her]> Latin languages share the same script
19:23:14FromDiscord<nnsee> it's cyrillic
19:23:17FromDiscord<tixonochek> In reply to @chronos.vitaqua "Script as in the": Like the letters?
19:23:21FromDiscord<Robyn [She/Her]> Yeah
19:23:22FromDiscord<Phil> We just don't have languages for everything. Given the activity in those channels is basically 0 I'm tempted to keep ignoring those.
19:23:27FromDiscord<Phil> (edit) "languages" => "language channels"
19:23:51FromDiscord<tixonochek> I mean to be fair this server is flooded with unnecessary channels from what I can see
19:23:59FromDiscord<Phil> All they have managed to do so far is make my life harder when a discord spambot manages to post in all channels simultaneously
19:24:09FromDiscord<nnsee> there's definitely too many imo
19:24:12FromDiscord<Robyn [She/Her]> The Russian channel seems to have activity at least, I know griff is Ukrainian
19:24:57FromDiscord<tixonochek> In reply to @nnsee "there's definitely too many": I know servers a lot bigger than this one, but the channels there are organized and there are a lot fewer than here
19:25:29FromDiscord<tixonochek> I was just asking about the language channel because if we have stuff like bulgarian, why not
19:25:55FromDiscord<nnsee> well, i used to mod the python discord, and it was huge, and a considerable amount of effort went into planning and organizing the channels in a way that made sense
19:26:34FromDiscord<tixonochek> In reply to @nnsee "well, i used to": An that rises the question of why there is a gaming channel.
19:26:37FromDiscord<nnsee> we had three different help channels for a while and people were encouraged to grab a vacant one and ask for help there, but that system immediately stopped working once we listed the server on the public guilds thing
19:26:41FromDiscord<tixonochek> (edit) "An" => "And"
19:27:01FromDiscord<nnsee> i stopped being a mod pretty soon after
19:27:59FromDiscord<Robyn [She/Her]> In reply to @nnsee "we had three different": Oh? How did it stop working?
19:28:05nisstyrePhil: this looks promising https://gitlab.gnome.org/GNOME/libdex
19:28:19FromDiscord<nnsee> there were far too many people for a free for all system
19:28:25FromDiscord<nnsee> this was also way before threads were a thing
19:28:56FromDiscord<Robyn [She/Her]> In reply to @nnsee "there were far too": Fair
19:29:02FromDiscord<nnsee> people generally just used the first help channel regardless of whether it was inactive or not, and the second and third ones were used far far less
19:29:17FromDiscord<nnsee> then, once an extremely large amount of people started pouring in, it just turned into mayhem
19:29:33FromDiscord<nnsee> it was hard to keep track of questions that had been asked but not received an answer, etc
19:29:34FromDiscord<Phil> In reply to @nisstyre "<@180601887916163073>: this looks promising": So async in C? I'd mostly shrug. Like sure, but I don't wish to write C 😅
19:29:44FromDiscord<nnsee> we experimented with various bot-based systems
19:29:54FromDiscord<Robyn [She/Her]> In reply to @nnsee "it was hard to": Fair enough, do you think forum channels helped at all or?
19:30:22nisstyrePhil: yep, futures in C, and yeah if you can just do it in Nim then it's irrelevant
19:30:33FromDiscord<nnsee> i don't really remember. i remember that people didn't really use those channels at all compared to the help channels
19:31:52FromDiscord<Phil> You can, there's 2 async runtimes. neither fully work without cycles so you have to use nim's default memory management strategy (orc) or the old default (refc) which can collect cycles, but they work well enough and didn't produce memory leaks nor data races for me so far when experimenting with them in my runtime
19:32:27FromDiscord<tixonochek> In reply to @nnsee "i don't really remember.": On the php discord as a help channel they have only 1 forum channel. And I hate it, because its kind of weird to create a whole "thread" for some extremely simple question. And you need to wait a few days to get an answer anyway.
19:32:49FromDiscord<Phil> (which is unfinished and I'll only manage to get back to once work gives me more breathing room to not have it in my head as much)
19:33:02FromDiscord<Robyn [She/Her]> In reply to @isofruit "You can, there's 2": How about: Cps :>
19:33:27FromDiscord<Phil> Playing around with that was the next step after
19:33:34FromDiscord<Robyn [She/Her]> I'm looking into cps but just trying to get a mvp before actually integrating it into my project
19:33:37FromDiscord<Phil> But this product launch just takes too much out of me as lead dev
19:33:42FromDiscord<Robyn [She/Her]> In reply to @tixonochek "On the php discord": That's fair oof
19:33:48FromDiscord<Robyn [She/Her]> In reply to @isofruit "But this product launch": Understandable
19:34:11FromDiscord<Phil> "lead dev" with fat quotation marks since it's mostly 3 frontend devs that I'm guiding
19:34:16FromDiscord<Phil> Myself included
19:34:22nisstyrePhil: actor model style concurrency would be nice to have in Nim. More complex though. For parallelism I've tried using Weave and it's pretty good so far.
19:34:57FromDiscord<Phil> In reply to @nisstyre "<@180601887916163073>: actor model style": Is it? In my head it was always super simple because I just see it as a webserver that coincidentally runs on my machine that I have a websocket-like connection to
19:35:28nisstyrethe complexity is in managing queues, backpressure, etc, and exceptions of course
19:35:31FromDiscord<Phil> Or at least that's the style of concurrency that I was writing a lib for
19:35:41nisstyreif your actors can raise exceptions you have to somehow handle that
19:35:44FromDiscord<Phil> Ahhhhhh okay yeah that's brutal as hell
19:36:51FromDiscord<Phil> As in, managing queues, raising exceptions has not been that big a problem for me yet
19:36:56FromDiscord<Phil> (edit) "raising" => "dealing with"
19:38:14nisstyreyeah if an actor's mailbox just fills up with stuff you have to somehow deal with that
19:38:29nisstyreit's supposed to guarantee that messages get sent, but I guess not necessarily received
19:42:59FromDiscord<Phil> I think for MVP I'm more going the route of allowing deadlocks
19:43:06FromDiscord<Phil> Can't finish sending if the mailbox is full
19:43:42FromDiscord<Phil> Because everything else potentially has performance implications
19:44:07FromDiscord<Phil> (As in, my mind is going persisting the messages)
19:44:53FromDiscord<Phil> Or re-allocating channels which are static-size as provided by the lib I use
19:45:52nisstyreyou should probably arena allocate them
19:48:04FromDiscord<Phil> Eh, problems for later.↵The lib is supposed to first provide sth because right now we have nothing that allows this kind of mental model of just multithreading with message passing
19:48:23FromDiscord<Phil> Right now it's all just allocating the channels on startup and after that fixed size.
19:48:43nisstyrethat's probably the best way tbh
19:49:47FromDiscord<Phil> Got to have sth to be able to iterate on it.↵And getting that to work without any tsan/asan/valgrind errors is already a challenge and a half
19:51:35*pmp-p quit (Quit: No Ping reply in 180 seconds.)
19:52:58*pmp-p joined #nim
20:14:38FromDiscord<nitely_> @.bobbbob report the nim-regex issue pls
20:15:13FromDiscord<requiresupport> question about macros, do they take effect even for the modules im importing, or nim's stdlib?
20:20:06FromDiscord<Phil> Kind of? If the stuff you're importing calls macros, those will get resolved and thus affect your compile time
20:25:06FromDiscord<Elegantbeef> If a macro is uses it's used
20:26:04FromDiscord<Elegantbeef> Deep philosophical takes with beef
20:32:05FromDiscord<michaelb.eth> sent a long message, see https://pasty.ee/QLstmjRKFYZL
20:33:44FromDiscord<Elegantbeef> > There I was, this self-taught programmer that had thought I was "hot shit"↵I have the exact opposite problem 😄
20:34:13FromDiscord<michaelb.eth> In reply to @Elegantbeef "> There I was,": You are the living mountain
20:34:22FromDiscord<michaelb.eth> you have become the climb
20:39:58FromDiscord<Phil> I still think I'm hot shit
20:40:08FromDiscord<Elegantbeef> You got one of those words right
20:40:09FromDiscord<Phil> Not because I don't think that I've climbed the mountain
20:40:18FromDiscord<Phil> (edit) removed "don't"
20:40:23FromDiscord<Phil> But more because I see my colleagues
20:40:48FromDiscord<Phil> Or rather I've seen multiple colleagues. And if that's my baseline I'm hot shit.
20:41:31FromDiscord<Phil> Still means I'm basically entirely outclassed by people actually having a grasp on low-level stuff like mratsim, jtv or the folks holding C/C++ talks
20:46:22FromDiscord<Phil> Or anyone having done any kind of actual performance optimization on the low-level front
20:46:45FromDiscord<Elegantbeef> We get it, you think you're better than me! 😛
20:47:06FromDiscord<Phil> Only it being a douchebag fruit
20:47:22FromDiscord<Phil> I have you beat there hands down since your PFP is a cute doggo
20:47:45FromDiscord<Phil> (edit) "it" => "at"
20:47:58FromDiscord<Phil> And at being old, definitely at being old
20:48:37FromDiscord<Phil> Potentially also on height
20:49:28FromDiscord<michaelb.eth> In reply to @isofruit "Potentially also on height": are you in the "tall programmers" club/
20:49:33FromDiscord<michaelb.eth> (edit) "club/" => "club?"
20:49:53FromDiscord<Phil> "Tall programmers whose height will give them back problems" club
20:50:04FromDiscord<michaelb.eth> me 2 😅
20:50:16FromDiscord<Elegantbeef> Could not be me
20:51:05FromDiscord<leorize> you're just telling me that your posture suck xd↵(@Phil)
20:51:09FromDiscord<Phil> Of course not, you have beef.↵Only evil people have beef.↵And evil people are small since that brings them closer to hell.↵This is definitely how everything works
20:51:26FromDiscord<Elegantbeef> I mean I'm 5'10"(177cm)
20:51:32FromDiscord<Elegantbeef> I'm not that small
20:51:45FromDiscord<Phil> In reply to @leorize "you're just telling me": That could be a pretty fair assessment
20:52:05FromDiscord<Phil> I'm just hoping the rock climbing delays when it eventually fucks me over
20:52:27FromDiscord<Elegantbeef> Not all of us can look giraffes in the eyes but at least I don't have to duck to enter doors
20:52:44FromDiscord<leorize> depends on which country you visit
20:52:53FromDiscord<Elegantbeef> Me vist?
20:53:04FromDiscord<Elegantbeef> I don't even visit Canada!
20:53:21FromDiscord<leorize> just make sure your rock climbing posture is good phil
20:53:47FromDiscord<Phil> It is, I've been told just hanging generally is pretty good to decompress the spine
20:53:55FromDiscord<Elegantbeef> With the way he speaks rock climbing is just pulling himself up the cliffside
20:55:27FromDiscord<michaelb.eth> is it true that a true Canadian outdoorsman can climb a cliffside with only his lips?
20:56:28FromDiscord<Elegantbeef> Close it's cheeks
21:37:05FromDiscord<spivee> Is it normal to take 18.4 seconds to compile `echo "hello world!"` on vcc?
21:37:50FromDiscord<spivee> Not trying to throw shade, genuinely curious whether there is a config problem I should be chasing down or if it's just not a prioritised target or what is going on.
21:38:33FromDiscord<Elegantbeef> I'm going to say no
21:38:37FromDiscord<Elegantbeef> But I've never used vcc
21:49:06FromDiscord<spivee> ah antivirus
21:49:54FromDiscord<spivee> put an 'exclusion' in Windows Security for the whole `nim-2.0.2` folder and now it's 0.05 seconds
21:50:31FromDiscord<requiresupport> this issue with AVs and nim is so weird
21:52:07FromDiscord<grumblygibson> `tryRecv` polling every thread in a group of threads for a 'Ready' message should be a decent way to find threads ready for another work item, no? Trying to find error in my logic. context: https://forum.nim-lang.org/t/11339
21:56:01FromDiscord<grumblygibson> I've been at this for days tracking down the issue. Would love a channels expert to weigh in.
22:07:21FromDiscord<griffith1deadly> In reply to @spivee "Is it normal to": on vcc yes
22:16:52FromDiscord<Robyn [She/Her]> sent a code paste, see https://play.nim-lang.org/#pasty=XDJgNKAhEidv
22:16:59FromDiscord<Robyn [She/Her]> Why am I getting an error here? :/
22:17:44FromDiscord<Robyn [She/Her]> Removing `T: BaseProvider` and replacing it with `T` works but, that's a bit annoying? Since then I could pass anything in
22:18:23FromDiscord<Robyn [She/Her]> I can do `T: ref object` though??
22:23:18NimEventerNew thread by RegularAlias: Attempting a DSL on top of htmlgen for fun/learning, see https://forum.nim-lang.org/t/11340
22:26:48FromDiscord<Robyn [She/Her]> `T: BaseProvider` also works if I pass `BaseProvider` when defining `accs`, why can't I use `SQLiteHandler`? It inherits from `BaseProvider` so it makes no sense
22:26:48FromDiscord<airdroplivelinkinbi04221> Hot Teen & Onlyfans Leaks :underage: :peach: https://discord.gg/esexx @everyone
22:26:49FromDiscord<Robyn [She/Her]> <@&371760044473319454>
22:32:39FromDiscord<Robyn [She/Her]> Is there a way for me to get the fields of a type via the typedesc?
22:32:53FromDiscord<Robyn [She/Her]> I don't want my types to require `partial` on all of them :/
22:41:43FromDiscord<spivee> I have a C header I use that defines `main` on linux and `WinMain` on Windows, and then calls `entry_point(int argc, char argv)`, which my projects are expected to define themselves. (like SDL/Sokol/etc.) Is there a way I can instruct Nim to use `entry_point(int argc, char argv)` as an entry point?
22:51:21FromDiscord<Elegantbeef> sent a code paste, see https://play.nim-lang.org/#pasty=GcYvMCkHCyMY
22:53:00FromDiscord<spivee> sweeet, thanks!
22:53:23FromDiscord<bosinski2023> sent a code paste, see https://play.nim-lang.org/#pasty=xXjgsBIZtdOT
22:55:19FromDiscord<Elegantbeef> `T: BaseProvider` inside a typedef does not work how you'd imagine
22:55:35FromDiscord<Elegantbeef> sent a code paste, see https://play.nim-lang.org/#pasty=GelgNdteviFN
22:55:54FromDiscord<Robyn [She/Her]> sent a code paste, see https://play.nim-lang.org/#pasty=rWMETIYzaxhN
22:56:12FromDiscord<Robyn [She/Her]> In reply to @Elegantbeef "`T: BaseProvider` inside a": How does it work? Feels silly
22:56:16Amun-Raptr UncheckedArray[cstring] == cstringArray
22:56:18FromDiscord<Robyn [She/Her]> But thanks Beef!
22:56:27FromDiscord<Elegantbeef> It only accepts `BaseProvider`
22:56:57FromDiscord<Elegantbeef> `provider` works cause `provider` is implicitly convertible to \`BaseProvider
22:57:31FromDiscord<Elegantbeef> But `typedesc[SqLiteProvider]` is not implicitly convertible
23:02:26FromDiscord<Robyn [She/Her]> Aaah okay, thank you!
23:13:13*SchweinDeBurg joined #nim
23:14:29*pmp-p quit (Ping timeout: 240 seconds)