<< 23-10-2022 >>
00:03:39*wallabra_ joined #nim
00:04:38*wallabra quit (Ping timeout: 276 seconds)
00:05:29*wallabra_ is now known as wallabra
00:06:10FromDiscord<Elegantbeef> Yea i imagine that's easier
01:11:39FromDiscord<sOkam!> sent a code paste, see https://play.nim-lang.org/#ix=4dT8
01:11:52FromDiscord<sOkam!> they output `ptr 30` and `30`
01:12:09FromDiscord<Elegantbeef> `cast[ByteAddress](num.addr)`
01:12:22FromDiscord<Elegantbeef> `repr` changes its behaviour depending on mm afaik
01:12:36FromDiscord<Elegantbeef> One version prints `addressInHex => data`
01:13:03FromDiscord<Elegantbeef> Yea refc's version shows the address
01:13:23FromDiscord<Elegantbeef> The RCs do not
01:13:57FromDiscord<sOkam!> ah i must be on rc, and the tutorial was on refc
01:14:04FromDiscord<guttural666> is there an easy/ideomatic way to check if a given int is a valid value of a given enum?
01:17:53arkanoidhow can I get the typeImpl of a parameter type in a macro? https://play.nim-lang.org/#ix=4dTc I've tried with generics and "static typedescr" but it doesn't work. I just get it working by using the type inside the macro
01:40:05FromDiscord<exelotl> In reply to @guttural666 "is there an easy/ideomatic": I'd do `n in ord(Foo.low)..ord(Foo.high)`
01:40:33FromDiscord<exelotl> Not sure if there's a better way
01:42:24arkanoida clearer example of what I'm trying to achieve. Should be simple, but I'm failing hard in passing type argument to macro https://play.nim-lang.org/#ix=4dTf
01:42:40arkanoiddo I have to pass a static default object instead and extract type from there?
01:43:17arkanoidbut the macro should get an object as argument, instead of a generic type or a typedesc
01:45:18*arkurious quit (Quit: Leaving)
01:47:04FromDiscord<Elegantbeef> you can do `t.getTypeImpl[^1].getTypeImpl()`
01:48:01arkanoidit works! ... how? is typedesc not a type?
01:48:17arkanoidthanks!
01:48:28FromDiscord<Elegantbeef> typedesc is a type which is the issue
01:48:36FromDiscord<Elegantbeef> `getTypeImpl` of a typedesc is the typedesc
01:51:30*wallabra quit (Ping timeout: 250 seconds)
01:52:10arkanoidok? and according to this, also the generic argument is kinda like typedesc
01:53:52*wallabra joined #nim
02:01:29NimEventerNew thread by snej: Nim 2: What’s special about Task?, see https://forum.nim-lang.org/t/9543
02:02:56FromDiscord<Elegantbeef> If you have a generic type you're not going to have fun with `getTypeImpl`
02:03:12FromDiscord<Elegantbeef> The typed AST is woefully underspecified
02:11:25arkanoidI'll stay far from it, then
02:11:39FromDiscord<Elegantbeef> Arkanoid it doesnt emit an object but https://play.nim-lang.org/#ix=4dTm does work
02:11:52arkanoiddamn, nimpretty fails pretty badly in formatting the output of dumpAstGen
02:11:52FromDiscord<Elegantbeef> Yes it's a hack 😄
02:12:43arkanoidmacro in template, mixins, whoa, this is hardcore
02:17:17arkanoidok, you went down the route of creating an object off type typ and use fields
02:19:25arkanoidI don't understand how you are building a type with a nnkTupleConstr
02:20:02arkanoidthere are no nnkTuple nodes in dumpAstGen of a type block
02:21:48FromDiscord<Elegantbeef> It's a tuple alias
02:21:56FromDiscord<Elegantbeef> I'm making a tuple of the fields in this hack
02:23:30arkanoidoh! you're making an unnamed tuple there
02:23:31arkanoidgot it
02:23:34arkanoidmakes sense now
02:24:00FromDiscord<Elegantbeef> https://play.nim-lang.org/#ix=4dTo is a bit less hacky
02:24:02arkanoidreading the manual about mixin now
02:24:27FromDiscord<Elegantbeef> Only used due to the hack of using a macro inside the template
02:24:37FromDiscord<Elegantbeef> I still just say use a generic instead
02:26:04arkanoidthis is rocket science for my understanding
02:26:24FromDiscord<Elegantbeef> The entire point of mixin in that code was just to make it so you dont need to export `macros` or `import macros` in modules that use it
02:26:46FromDiscord<Elegantbeef> You can use `mixin` and `bind` to fix issues you'd otherwise have when the template was expanded at the callsite
02:27:25FromDiscord<Elegantbeef> The macro there is much less complicated, but still a tinge complicated
02:29:09arkanoidthanks for all of this
02:46:18*Lord_Nightmare quit (Quit: ZNC - http://znc.in)
02:47:02arkanoidfunny, path concatenation "/" happens to change sep character bases on os. But it does also when doing cross compilation, leading to a failure
02:47:34FromDiscord<Generic> oh yeah I know this issue
02:47:39FromDiscord<Elegantbeef> Yep it's not fun
02:47:43arkanoidso nim cpp -d:mingw would use build path separated by "\" even on linux
02:47:49FromDiscord<Elegantbeef> also `getSourcePath` or w/e doesnt play well with mingw
02:48:48arkanoidexactly. I'm trying to remove absolute paths from my config.nims replacing them by currentSourcePath() / ...
02:49:03arkanoidbut apparently I have to put them back
02:50:26*Lord_Nightmare joined #nim
03:05:28FromDiscord<Elegantbeef> Even more fun when i was trying to get miniaudio to compile with mingw
03:05:55FromDiscord<Elegantbeef> Tried like thirty things then i just decided "you know copying this to a `.c` extension is probably what i have to do"
03:06:51FromDiscord<Elegantbeef> Fun thing is Nim's `compile` pragma doesnt seem to work with `.h` files
03:27:42*vicecea quit (Remote host closed the connection)
03:28:30*vicecea joined #nim
04:50:38FromDiscord<sOkam!> Is there a way to get the name of a value in string form, instead of the value itself converted to string?
04:50:50FromDiscord<sOkam!> (edit) "value" => "variable"
04:51:06FromDiscord<Elegantbeef> `astToStr`
04:57:29FromDiscord<sOkam!> how do you declare a generic proc to be part of a type?
04:58:21FromDiscord<Elegantbeef> What?
04:59:25FromDiscord<sOkam!> sent a code paste, see https://play.nim-lang.org/#ix=4dTR
04:59:52FromDiscord<Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4dTS
05:01:22FromDiscord<sOkam!> not sure if this is the way to do it, btw↵just trying to figure out how to map GLFWKey codes to procedures, in a way that they can be remapped
05:03:55FromDiscord<sOkam!> sent a code paste, see https://play.nim-lang.org/#ix=4dTU
05:04:13FromDiscord<Elegantbeef> `` goes on the symbol number one
05:04:21FromDiscord<Elegantbeef> Number to can you not just do `Table[GlfwKey, proc()]`
05:04:21FromDiscord<Elegantbeef> two even
05:04:41FromDiscord<Elegantbeef> Or `array[GlfwKey, proc()]` though know Enums are wacky in C land
05:04:47FromDiscord<sOkam!> probably can. just wondering how to do it
05:05:34FromDiscord<Elegantbeef> Well the above should work so...
05:07:55FromDiscord<sOkam!> sent a code paste, see https://play.nim-lang.org/#ix=4dTV
05:08:23FromDiscord<Elegantbeef> What's the type of `GlfwKey`?
05:08:28FromDiscord<sOkam!> enum
05:10:33FromDiscord<Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4dTW
05:12:30FromDiscord<sOkam!> Oh i wasn't exporting tables from the other module 🤦‍♂️
05:12:45FromDiscord<sOkam!> nice 🐠 memory
05:45:58FromDiscord<ChocolettePalette> We will not back down 💪🏻💪🏻💪🏻
05:46:18FromDiscord<Elegantbeef> I'm going to need a map to explain this
06:19:20FromDiscord<Rika> You didn’t register the s key I assume
06:25:04FromDiscord<sOkam!> #noskey #willnevergoback 🙂
07:36:23FromDiscord<Require Support> anyone have an example of creating/read/write into async namedpipe on windows? saw this https://github.com/cheatfate/asynctools but not sure if there are examples out there that I could learn from
07:49:40*vicecea quit (Remote host closed the connection)
07:50:11*vicecea joined #nim
08:30:35FromDiscord<enthus1ast> we do not have a json schema to nim object converter right?
08:36:48FromDiscord<enthus1ast> @Require Support\: did you see the short example at the start of https://github.com/cheatfate/asynctools/blob/master/asynctools/asyncpipe.nim
08:36:49FromDiscord<enthus1ast> ?
08:37:30FromDiscord<enthus1ast> this uses waitFor though, you prolly want to use await in an async proc
08:42:48FromDiscord<Phil> In reply to @enthus1ast "we do not have": Could you define the problem a bit clearer for me? Because when I hear "json to nim object converter" I think jsony, but I know you know of jsony so I'm pretty sure that "schema" in there changes things
08:43:52FromDiscord<Elegantbeef> Take a json scheme output a Nim object
08:43:59FromDiscord<Elegantbeef> I think there are some
08:44:15FromDiscord<enthus1ast> i have a bunch of json schemas, i want to build a client for a protocol (nats.io) and one of their api's heavily uses json, so they document their api through json schema, i do not want to handle json in my code, but nim objects
08:44:35FromDiscord<Phil> I don't get the difference between a json string and a json schema
08:44:36FromDiscord<enthus1ast> so i want to have nim objects and use json.to etc
08:44:52FromDiscord<Elegantbeef> Emit the type definitions from json schemes
08:44:54FromDiscord<enthus1ast> json schema IS a json \:)
08:44:58FromDiscord<Elegantbeef> Not just convert data to file
08:45:03FromDiscord<Phil> Is JSON schema just what you'd copy paste into the browser to get a js object?
08:45:05FromDiscord<enthus1ast> but it describes other json messages
08:45:22FromDiscord<enthus1ast> like a xml schema
08:45:36FromDiscord<Phil> So you want to generate a nim type from a provided json schema?
08:45:41FromDiscord<enthus1ast> yes
08:45:46FromDiscord<Phil> That you can then just jsony convert objects into?
08:46:02FromDiscord<enthus1ast> in my case std json, but yes
08:46:04FromDiscord<Phil> (edit) "That you can then just ... jsonyto" added "use" | "usejsony ... convertactual" added "to" | "objects" => "actual json"
08:46:29FromDiscord<Phil> So what you need is a utility acts before you actually compile your project, right?
08:46:35FromDiscord<enthus1ast> yes
08:46:46FromDiscord<Phil> Okay, I think I've begun to understand the problem domain
08:46:50FromDiscord<enthus1ast> i mean, i could just do it by hand, but yeah, lazy
08:47:02FromDiscord<Require Support> In reply to @enthus1ast "we do not have": https://github.com/treeform/jsony maybe this?
08:47:02FromDiscord<carmellajones34> sent a long message, see http://ix.io/4dUv
08:47:19FromDiscord<Phil> In reply to @carmellajones34 "Win up to $1000": <@&371760044473319454> I am hostile to this trash person!
08:47:35FromDiscord<Phil> I am now no longer hostile
08:47:52FromDiscord<Elegantbeef> Again jsony doesnt take in a json file and emit a type definition
08:48:19FromDiscord<enthus1ast> imho jsony cannot act on schema's but not 100% sure↵(@Require Support)
08:48:22FromDiscord<Phil> In reply to @Elegantbeef "Again jsony doesnt take": Yeh, it's for when you have the type definition and want to convert actual JSON, got that so far
08:48:38FromDiscord<Require Support> oh sorry didnt notice
08:48:41FromDiscord<Elegantbeef> https://github.com/jiro4989/nimjson something like
08:49:02FromDiscord<enthus1ast> yes
08:49:03FromDiscord<PMunch> It's so confusing when another mod swoops in and deletes a message
08:49:40FromDiscord<Elegantbeef> The consensus in other servers is to 👍️ the summoning message to indicate someone took action
08:51:01FromDiscord<enthus1ast> we need a moderator bot
08:51:17FromDiscord<Elegantbeef> That's been said like 300 times at this point
08:51:23FromDiscord<enthus1ast> yeah can imagine
08:52:19FromDiscord<Phil> Couldn't a utility that generates type definitions as a one of be written in any language since you just need a script / binary that does the job for you once, generates the nim file with the types and you're settled for life (unless the json changes a week later) ?
08:52:27FromDiscord<Phil> (edit) "one of" => "one-of"
08:52:35FromDiscord<enthus1ast> yes sure
08:52:44FromDiscord<Elegantbeef> Well the benefit of doing it inside Nim is that if you do it at compile time you always have up to date to your json templates
08:53:08FromDiscord<enthus1ast> Elegantbeef\: this MIGHT be possible, but i would rather gen the nim code
08:53:16FromDiscord<enthus1ast> i do not like wrappers that does this
08:53:21FromDiscord<Elegantbeef> Regardless nimjson exists 😄
08:53:30FromDiscord<Phil> In reply to @Elegantbeef "Well the benefit of": If you turn that into part of your compile-pipeline you're making compiling longer though and need to lug around json example messages forever
08:53:32FromDiscord<Elegantbeef> Fork if needbe
08:53:39FromDiscord<Elegantbeef> Oh noes↵(@Phil)
08:53:52FromDiscord<enthus1ast> (i fact, the other nats wrapper is a nimgen wrapper, which is broken and completely useless)
08:54:14FromDiscord<Elegantbeef> Have known truths isnt a big deal to me, the CT is a tinge of a deal
08:54:16FromDiscord<Phil> In reply to @Elegantbeef "Oh noes (<@180601887916163073>)": Oi, I discovered that a prologue extension of mine literally doubled to tripled compile-time since I'm generating so much code through it, so I've become a bit more mindful
08:54:23FromDiscord<Phil> (edit) "In reply to @Elegantbeef "Oh noes (<@180601887916163073>)": Oi, I discovered that a prologue extension of mine literally doubled to tripled compile-time since I'm generating so much code through it, so I've become a bit more mindful ... " added "of prolonging compiletime"
08:55:04FromDiscord<Elegantbeef> I do like concrete definitions, but i also do like having code that doesnt fail cause i forgot to run "jsongenerator thingy"
08:55:32FromDiscord<Phil> I mean, if your json-schema changes that much you have different problems
08:55:34FromDiscord<Elegantbeef> You can always do similar to futhark where you only regenerate when you need to so then CTE isnt expensive
08:55:34FromDiscord<Elegantbeef> It's just a simple check
08:55:44FromDiscord<enthus1ast> Elegantbeef\: i think this particular projects i want to speak to, is quite stable already
08:55:45FromDiscord<Require Support> is there a way to install nim 2 within a script like `curl https://nim-lang.org/choosenim/init.sh -sSf | bash -s -- -y`
08:56:25FromDiscord<enthus1ast> with choosenim you can switch between version
08:56:30FromDiscord<Require Support> or i should ask how would one install nim 2
08:56:39FromDiscord<Phil> In reply to @Require Support "is there a way": If that script installs choosenim you can use a choosenim command to swap to devel (which is nim 2)
08:56:43FromDiscord<enthus1ast> nim 2 ?
08:56:49FromDiscord<Require Support> oh
08:56:51FromDiscord<Require Support> its devel
08:56:57FromDiscord<enthus1ast> was not gone for soo long \:)
08:56:58FromDiscord<Require Support> sorry just saw the nim conf presentation
08:57:11FromDiscord<Phil> Devel is what will be released as nim 2 at... some point
08:57:24FromDiscord<Require Support> 2022 (some point)
08:57:31FromDiscord<Elegantbeef> Araq says 'soon'
08:57:33FromDiscord<Phil> Been using it for half a year now or so, literally can not move back to 1.68
08:57:38FromDiscord<Elegantbeef> Sooner than one would expect
08:57:46FromDiscord<Phil> Like, very literally, I'd get compiler errors
08:58:16FromDiscord<Require Support> ye decided to move to it now so i dont cry when it eventually gets released
08:58:24FromDiscord<Elegantbeef> Yea my code doesnt work on devel, though only cause i have a distinct type being passed a nil lit, so it's about 3 seconds to fix
08:58:29FromDiscord<Elegantbeef> Just havent got around to it
08:59:07FromDiscord<Elegantbeef> Hell my text editor is borked and it's code suggestions are hidden but doc comments arent
08:59:13FromDiscord<Elegantbeef> Debian sid was a choice
09:00:22FromDiscord<Phil> That kinda reminds me I should get around to making use of nim merging in default-values for objects
09:00:34FromDiscord<Elegantbeef> That was removed
09:00:39FromDiscord<Phil> Wait, it was?
09:00:45FromDiscord<enthus1ast> i learned to just use nim stable
09:00:46FromDiscord<Phil> Still addicted to constructor then
09:01:02FromDiscord<enthus1ast> for me its too annoying when my code breaks all the time \:)
09:01:30FromDiscord<Elegantbeef> I often fix things that i want to toy with so i ping pong back and forth
09:01:31FromDiscord<Phil> Well, mine hasn't broken, though I guess it would've if I had started incorporating the default value thing
09:02:16FromDiscord<enthus1ast> Elegantbeef\: its good that you can do this, i'm more than often overwhelmed by stuff i actually wanna do \:)
09:02:27FromDiscord<enthus1ast> without fixing compiler stdlib etc
09:02:30FromDiscord<Phil> I just want my compiler-warnings to work as expected and the bugfixes in the stdlib are absolutely essential
09:03:04FromDiscord<Elegantbeef> Yea i had to implement `toOpenArray` on the VM to get my current project of "Replace all string libraries to openArray[char]"
09:04:48FromDiscord<Phil> Updates devel↵> /home/philipp/.nimble/pkgs/regex-0.19.0/regex.nim(856, 18) Error: Exception can raise an unlisted exception: Exception↵Never mind, I spoke to soon, code did break
09:05:14FromDiscord<Elegantbeef> Super strict funcs
09:05:14FromDiscord<enthus1ast> haha \:)
09:05:43FromDiscord<Phil> When it's not in my own packages is the most annoying thing
09:06:07FromDiscord<Elegantbeef> Update your regex phil
09:06:07FromDiscord<enthus1ast> i know i'm completely not meant
09:06:13FromDiscord<Elegantbeef> https://github.com/nitely/nim-regex/commit/0333e127b37e499663320145aa36cd8a6ad7d6ea
09:07:19FromDiscord<Phil> Aaaaaaaand next it broke prologue↵> /home/philipp/.nimble/pkgs/prologue-0.6.0/prologue/core/context.nim(213, 8) Error: Base method 'extend' requires explicit '{.gcsafe.}' to be GC-safe
09:07:38FromDiscord<Phil> In reply to @Elegantbeef "https://github.com/nitely/nim-regex/commit/0333e127": I'll need to see where the old nim-regex dependency came in from and suggest a PR for updating the nimble file
09:08:09FromDiscord<Elegantbeef> I mean you can just do `nimble uninstall nim-regex` and then `nimble install -d` which 'should' make it work?
09:08:28FromDiscord<Elegantbeef> But yea people need to update nimble files for packages
09:08:38FromDiscord<Elegantbeef> Semver only works if people actually respect it
09:12:28FromDiscord<Phil> Alrighty, the prologue dependency was also fixable by just updating prologue, next an ndb error
09:12:50FromDiscord<Phil> Fuck, that one doesn't have an update that fixes the error
09:17:23FromDiscord<Phil> Question, if a proc is annotated with `raises: [Exception]`↵Doesn't that just mean it can raise any exception, so one might as well remove the "raises" pragma?
09:17:34FromDiscord<Elegantbeef> Correct
09:20:09FromDiscord<ringabout> In reply to @Isofruit "Question, if a proc": which line?
09:20:49FromDiscord<Phil> In reply to @flywind "which line?": That's in ndb↵↵> /.nimble/pkgs/ndb-0.19.9/ndb/sqlite.nim(314, 5) Error: tryWithStmt(db, query, @args, proc (stmt: PStmt): bool = result =↵> isSome(tryNext(stmt))) can raise an unlisted exception: Exception
09:22:15FromDiscord<ringabout> https://github.com/xzfc/ndb.nim/blob/f9c85711ffc2ba350fb3c64e5ce38ada47380742/ndb/sqlite.nim#L242 looks suspicious
09:22:44FromDiscord<Require Support> when compiling with devel breaks because of chronos 😢
09:23:01FromDiscord<Phil> sent a code paste, see https://play.nim-lang.org/#ix=4dUC
09:23:30FromDiscord<Phil> (edit) "https://play.nim-lang.org/#ix=4dUC" => "https://play.nim-lang.org/#ix=4dUD"
09:23:39FromDiscord<Phil> What I don't get is that tryNext doesn't raise an exception either as per raises pragma annotation
09:23:44FromDiscord<ringabout> In reply to @flywind "https://github.com/xzfc/ndb.nim/blob/f9c85711ffc2ba": You can add `effectsOf: body`
09:24:14FromDiscord<ringabout> sent a code paste, see https://play.nim-lang.org/#ix=4dUE
09:24:14FromDiscord<Phil> And tryWithStmt also doesn't raise anything
09:24:17FromDiscord<Phil> What does that do?
09:25:08FromDiscord<ringabout> `stricteffects` is enabled on the devel branch which required the effects of callback should be considered too.
09:25:42FromDiscord<ringabout> `effectsOf` takes the effects of `body` into consideration in this case.
09:26:03FromDiscord<ringabout> https://github.com/nitely/nim-regex/commit/0333e127b37e499663320145aa36cd8a6ad7d6ea is an example showing its usage.
09:26:26FromDiscord<Phil> ` Error: undeclared identifier: 'body'`
09:26:52FromDiscord<Phil> For ↵> proc tryWithStmt(db: DbConn, query: SqlQuery, args: seq[DbValue],↵> body: proc(stmt: Pstmt): bool {.raises: [], tags: [], effectsOf: body.}): bool =
09:28:49FromDiscord<ringabout> sent a code paste, see https://paste.rs/yL7
09:33:15FromDiscord<Phil> Alright, one update of my constructor dependency from 1.0.3 to 1.0.5 later I am now at a compiler error in my own code!
09:33:26FromDiscord<Phil> One that I don't really understand, but at least we got past the dependencies
09:33:48FromDiscord<ringabout> In reply to @Isofruit "Alright, one update of": What's the error?
09:33:59FromDiscord<Phil> sent a code paste, see https://play.nim-lang.org/#ix=4dUG
09:34:48FromDiscord<Phil> sent a code paste, see https://play.nim-lang.org/#ix=4dUH
09:35:02FromDiscord<Phil> Ohhh I'm an idiot, forgot a default
09:35:15FromDiscord<Phil> No idea how that even compiled before
09:35:51FromDiscord<ringabout> Yeah, the compiler may accept invalid code but shouldn't.
09:39:41FromDiscord<Phil> sent a code paste, see https://play.nim-lang.org/#ix=4dUK
09:43:30FromDiscord<ringabout> Is this a compiler bug or an constructor issue?
09:44:03FromDiscord<ringabout> I mean does 1.0.3 work?
09:44:54FromDiscord<Phil> It's 1.0.5
09:45:21FromDiscord<Phil> Oh you mean the constructor update might be the cause?
09:45:27FromDiscord<ringabout> Yeah
09:46:34FromDiscord<Phil> Yeah, same error
09:47:04FromDiscord<Phil> Fascinatingly I can't manage to generate a minimal example it seems
09:49:59FromDiscord<ringabout> It's fine to be an issue
10:02:26FromDiscord<Phil> Without recreating it though there can be no troubleshooting
10:08:24*fanta1 joined #nim
10:14:14FromDiscord<Phil> Its the macro cache table screwing up somehow
10:17:10FromDiscord<Phil> Found the issue, just not sure what the fix here is
10:19:10FromDiscord<Phil> sent a code paste, see https://play.nim-lang.org/#ix=4dUV
10:19:57FromDiscord<Phil> (edit) "https://play.nim-lang.org/#ix=4dUV" => "https://play.nim-lang.org/#ix=4dUW"
10:31:25FromDiscord<Phil> What I don't get is that in my "playground" project it's not translating the type into a string incorrectly, aka there it removes the default value correctly
10:32:10FromDiscord<Phil> And it keeps repeating this issue across all my types that use stuff like `0..9` or `Natural` instead of `int`
10:41:49FromDiscord<Phil> At the very least I can compile my project again
12:23:44*jmdaemon quit (Ping timeout: 250 seconds)
13:08:00*xet7 joined #nim
13:19:37FromDiscord<untoreh> sent a code paste, see https://play.nim-lang.org/#ix=4dVC
13:38:54FromDiscord<auxym> not sure. For starters you could try looking at the output of `expandArc` (assuming you are using arc/orc)? If that doesn't yield anything, mighht be valgrind time.
13:40:07FromDiscord<auxym> Nim's stdlib async implementation doesn't exactly have a great reputation though. Even Araq, in yesterday's talk, called it "crazy"
13:41:41FromDiscord<auxym> you could try https://github.com/status-im/nim-chronos as an alternative
13:50:25FromDiscord<untoreh> I tried asyncdispatch after chronos, because chronos also behaves similarly
13:54:21FromDiscord<ringabout> Are you using ARC or ORC?
13:58:45FromDiscord<Tanguy> sent a code paste, see https://play.nim-lang.org/#ix=4dVO
13:59:02FromDiscord<Tanguy> How do you measure the used ram?
14:00:23FromDiscord<untoreh> look at the process? also valgrind
14:01:01FromDiscord<Rika> You need the use malloc define flag to accurately measure memory usage with external (non Nim( tools
14:01:06FromDiscord<Rika> (edit) "Nim(" => "Nim)"
14:05:11FromDiscord<untoreh> oh it seems disabling `threadsafe` flag the memory bloat doesn't happen anymore
14:06:03FromDiscord<untoreh> no i lied wtf
14:12:42FromDiscord<untoreh> so it happens with `-d:threadsafe` with the GC and regardless of threadsafe with `--mm:orc`
14:17:54FromDiscord<Tanguy> The memory is used here (in chronos): https://github.com/status-im/nim-chronos/blob/master/chronos/ioselects/ioselectors_epoll.nim#L96↵It's necessary to use epoll to have a buffer with room for many FDs, though the default (1024) may be a bit high for your usage↵↵But all of this should be cleared by the GC at the end of the thread
14:20:30FromDiscord<Tanguy> Mhh, actually with threads "allocShared" is used and doesn't seem to be free-able
14:39:12FromDiscord<Phil> Did they buff compile-time speed in the last 6 months on devel?
14:39:32FromDiscord<Phil> Because I don't think my project compiled quite as fast before I updated my devel branch
14:40:57FromDiscord<Bung> this month as speed improvement, before I dont know
14:42:03FromDiscord<auxym> araq said in his nimconf talk that bootstrapping the compiler with orc went from 16s to 12s on his machine
14:42:22FromDiscord<auxym> (not sure exactly when this happened, I don't use devel)
14:43:09FromDiscord<Phil> I mean I'd happily support any claim of around 50-75% compile-speed improvement
14:43:33FromDiscord<Phil> I don't have any truly hard numbers but I know compilation of a release build before took around 30-45 seconds, let me see how long it is now
14:44:36FromDiscord<Phil> let me correct that, was 45-60 seconds
14:45:06FromDiscord<Phil> Hmmm release builds haven't gotten that much faster then, just the compile builds
14:45:17FromDiscord<Phil> Well, those are the ones that mostly matter for me
14:49:39FromDiscord<Phil> Yeah, debug builds build around 50% faster
14:49:50FromDiscord<Phil> (edit) "Yeah, debug builds build around 50% faster ... " added "for me, gut estimate"
14:54:48*derpydoo joined #nim
14:55:01FromDiscord<Tanguy> @untoreh https://github.com/status-im/nim-chronos/pull/326 you can use this branch, and call `closeThreadDispatcher()` at the end of each thread, which fixes the leak in chronos
14:56:02FromDiscord<Tanguy> We could probably add it to the dispatcher destructor, would have to think about it
14:59:25FromDiscord<ringabout> Or trying running `GC_runOrc` at the end of a thread proc.
14:59:37FromDiscord<ringabout> (edit) "trying" => "try"
15:03:14FromDiscord<untoreh> with `GC_runOrc` leaks almost nothing
15:03:23FromDiscord<ringabout> If thread exit may not trigger collection process if it does not exceed threshold of garbage.
15:03:27FromDiscord<ringabout> (edit) "If thread exit may not trigger ... collection" added "a"
15:03:41FromDiscord<ringabout> (edit) "If thread" => "Thread" | "Threadexit may not trigger a collection process if it does not exceed ... threshold" added "the"
15:04:05FromDiscord<ringabout> It can be possibly leaking if you don't call `GC_runOrc` manually.
15:04:52FromDiscord<ringabout> It is kind of a semi-automatical memory management.
15:05:52FromDiscord<ringabout> I have a PR to clean up before thread exit, but it was rejected => https://github.com/nim-lang/Nim/pull/19146
15:07:06FromDiscord<Tanguy> But it will be collected eventually by another thread, since memory is shared?
15:07:31FromDiscord<Tanguy> (not familiar with memory scheme of orc)
15:07:49FromDiscord<ringabout> I don't think so, the collector of ORC is thread local.
15:08:05FromDiscord<ringabout> That's why you cannot share refs between threads.
15:09:26FromDiscord<Tanguy> Uh, that's unfortunate
15:09:29FromDiscord<ringabout> And it goes for `std/threadpool` which doesn't work for ORC.
15:12:36FromDiscord<Require Support> anyone try compiling with binaries that use chronos async + `--mm:orc` ?
15:13:19FromDiscord<Tanguy> chronos is tested with orc in the CI, but I don't know if anyone already used it in production (at status we still use refc)
15:13:19FromDiscord<Require Support> getting: `Error: unhandled exception: index out of bounds, the container is empty [IndexDefect]` but having a hard time finding whats causing it
15:13:40FromDiscord<Require Support> ah i see
15:14:12FromDiscord<Tanguy> Are you using a recent chronos version? There was a recent PR which fixed a similar issue
15:14:57FromDiscord<Require Support> I think yes but ill double check
15:15:00FromDiscord<Phil> ringabout, do you perchance know if something was changed with std/macros?↵Specifically the `getImpl` proc?
15:16:22FromDiscord<Phil> sent a code paste, see https://play.nim-lang.org/#ix=4dWk
15:17:49FromDiscord<ringabout> I cannot think of something particular. You can check https://github.com/nim-lang/Nim/pulls?q=is%3Apr+author%3Ametagn+is%3Amerged or run a bisect
15:18:19FromDiscord<Phil> I have no yet suceeded in figuring out how to do a minimal replica, all I know is that the second I use a custom range (e.g. 0..9) or a somewhat abnormal int type (such as Natural) the representation of my type definition doesn't remove the assignment (`= 0` in this case)
15:18:55FromDiscord<Phil> Enums do not have this issue
15:19:11FromDiscord<ringabout> Oh, sorry. It seemed to be introduced by my object defaults PR.
15:19:58FromDiscord<ringabout> Could you possibly try `--undef:nimPreviewRangeDefault`.
15:20:08FromDiscord<ringabout> (edit) "`--undef:nimPreviewRangeDefault`." => "`--undef:nimPreviewRangeDefault`?"
15:20:24FromDiscord<ringabout> (edit) "seemed" => "seems" | "be" => "have been"
15:20:52FromDiscord<Phil> So just define that compile flag?
15:21:14FromDiscord<ringabout> It undefines the `nimPreviewRangeDefault` flag.
15:21:26FromDiscord<Phil> Yep, compiles
15:21:27FromDiscord<ringabout> Yeah
15:21:44FromDiscord<Phil> So that flag fixes my issue surprisingly enough
15:22:36FromDiscord<ringabout> Ranges in the object definition now have a default value, so `default(object)` will work correctly.
15:23:21FromDiscord<ringabout> I think constructor probably should fix this problem.
15:23:53FromDiscord<ringabout> You can open an issue for constructor.
15:24:23*nisstyre quit (Changing host)
15:24:23*nisstyre joined #nim
15:26:20FromDiscord<ringabout> You can use `--undef:nimPreviewRangeDefault` as a workaround. It is unstable. There is a critical issue to be fixed. I may figure something out in the future.
15:28:26FromDiscord<ringabout> I can add your library to important packages so that we cannot fix it at the first time, if it has a working CI and no external deps except nimble packages.
15:29:49FromDiscord<ringabout> You can also test against the devel branch using a cron CI which runs everyday.
15:30:37FromDiscord<ringabout> (edit) "using" => "setting"
15:41:41*arkurious joined #nim
15:41:52*xaltsc quit (Quit: WeeChat 3.6)
15:42:51FromDiscord<Phil> In reply to @flywind "I can add your": Sadly this is not for my lib, but rather my backend application, which isn't really going to be of use for anyone but me (unless somebody wants to throw up their own dnd wiki webpage with ridiculously fast loading times)
15:51:18*derpydoo quit (Ping timeout: 250 seconds)
16:07:47*krux02 joined #nim
16:16:54*arkanoid quit (Ping timeout: 264 seconds)
16:45:21FromDiscord<Goel> Why can't an identifier (`var a: int`) have the same name of a proc? I don't understand since they have separate structure and syntax that should be possible. `a` should be seen as identifier and `a()` always as proc/func call, no?
16:54:54*_alexr_ joined #nim
16:57:35*_alexr_ left #nim (#nim)
17:01:02FromDiscord<exelotl> sent a code paste, see https://play.nim-lang.org/#ix=4dWW
17:01:39FromDiscord<exelotl> (edit) "https://play.nim-lang.org/#ix=4dWW" => "https://play.nim-lang.org/#ix=4dWX"
17:01:40FromDiscord<Bung> a is variable now
17:02:08FromDiscord<Bung> semcheck will treat it as variable
17:05:23FromDiscord<Goel> sent a code paste, see https://play.nim-lang.org/#ix=4dX0
17:05:59FromDiscord<Goel> (edit) "https://play.nim-lang.org/#ix=4dX0" => "https://play.nim-lang.org/#ix=4dX1"
17:06:52FromDiscord<Bung> I think they are follow identity rules, nothing special
17:26:43FromDiscord<Phil> In reply to @flywind "Ranges in the object": Question, is the entire thing with enabling default values for objects still in devel or has it been removed? (since beef said something in that regard earlier)
17:27:23FromDiscord<ringabout> It is not
17:28:10FromDiscord<guttural666> hey, can sb give me a hint why my call of createDir gives me the run time errors seen below (Linux): https://media.discordapp.net/attachments/371759389889003532/1033793969458524273/unknown.png https://media.discordapp.net/attachments/371759389889003532/1033793969940865154/unknown.png
17:29:01FromDiscord<guttural666> ahhhh, fuck, this is the naming thing of the compiler
17:29:21FromDiscord<Phil> In reply to @guttural666 "hey, can sb give": Allow me to introduce you to nim's style indifference
17:29:29FromDiscord<guttural666> since it ignores certain writing styles
17:29:38FromDiscord<Phil> where `is_potato` counts as the same symbol as `isPotato`
17:29:53FromDiscord<guttural666> yeeeah, right, saw that in a video 2 days ago haha
17:30:10FromDiscord<Elegantbeef> `os.createFile(dir)`
17:30:19FromDiscord<guttural666> a dedicated error message for that would be cool
17:30:22FromDiscord<guttural666> yeah
17:30:28FromDiscord<Elegantbeef> It's not an error though to be fair
17:30:53FromDiscord<guttural666> technically not, but a warning would be approp
17:30:55FromDiscord<Phil> In reply to @guttural666 "a dedicated error message": I mean, the compiler can't know if you did or did not intend this, often enough it is perfectly valid for a proc to call itself to some degree
17:30:57FromDiscord<Elegantbeef> The declaration are from two seperate modules
17:30:58FromDiscord<Elegantbeef> Nim chooses the closest to the call site
17:31:27FromDiscord<Phil> If you want warnings about this stuff there's a compiler flag btw
17:31:39FromDiscord<Phil> ` --styleCheck:usages`
17:32:02FromDiscord<guttural666> ah okay, thanks for the info, nah was just confused, but now its clear 😄
17:32:19FromDiscord<Phil> You can turn that into an error via `--styleCheck:error`
17:32:35FromDiscord<Phil> Which would give you behaviour like in other languages where style insensitivity is not a thing
17:32:54FromDiscord<guttural666> thanks for the info
17:32:56FromDiscord<Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4dXh
17:33:08FromDiscord<guttural666> agreed eleg
17:33:30FromDiscord<Elegantbeef> You know the aft end of my name is also 4 letters and is a word! 😛
17:33:48FromDiscord<Phil> Eleg the beef
17:33:51FromDiscord<guttural666> 4k monitor and Discord scaling is brutal haha
17:34:04FromDiscord<guttural666> have to use binos to see Discord chat haha
17:34:33FromDiscord<Elegantbeef> Get a crt that's 4\:3
17:34:47FromDiscord<Elegantbeef> Certainly will not have that issue
17:35:41FromDiscord<guttural666> indeed 😛
17:36:36FromDiscord<guttural666> anybody experience in linking nim source with Qt stuff? C++ is giving me testicular cancer
17:38:39FromDiscord<Elegantbeef> FFI with QT lies madness
17:40:19FromDiscord<guttural666> have you tried it and is it hard?
17:41:52FromDiscord<Phil> sent a code paste, see https://play.nim-lang.org/#ix=4dXm
17:42:05FromDiscord<Phil> (edit) "https://play.nim-lang.org/#ix=4dXm" => "https://play.nim-lang.org/#ix=4dXn"
17:42:08FromDiscord<ringabout> let x = default(A)
17:42:27FromDiscord<Phil> And `new(A)` for ref types?
17:42:32FromDiscord<Phil> or default for all?
17:43:04FromDiscord<ringabout> Yeah, use `new A`, default(A) is still a nil
17:43:22FromDiscord<ringabout> Don't use `let x = A()` for now because there is a bug to be fixed.
17:43:32FromDiscord<Elegantbeef> Constructor is now dead to Phil
17:43:49FromDiscord<Phil> In reply to @Elegantbeef "Constructor is now dead": Not yet confirmed
17:44:00FromDiscord<Phil> For now I still can't compile my project without it
17:44:30FromDiscord<ringabout> It needs to fix a problem because ranges in object get a default value by default on the devel branch.
17:44:32FromDiscord<Phil> And get strange errors like `/home/philipp/.choosenim/toolchains/nim-#devel/lib/pure/collections/tables.nim(269, 6) Error: the field 'data' is not accessible.`
17:46:56*arkanoid joined #nim
17:47:01arkanoidwhat does "--app:console|gui|lib|staticlib" do?
17:50:24FromDiscord<Phil> sent a code paste, see https://play.nim-lang.org/#ix=4dXp
17:52:46FromDiscord<ringabout> What's the type of key and value?
17:55:27FromDiscord<Phil> sent a code paste, see https://play.nim-lang.org/#ix=4dXr
17:55:53FromDiscord<Phil> (edit) "https://play.nim-lang.org/#ix=4dXr" => "https://play.nim-lang.org/#ix=4dXs"
17:57:14FromDiscord<ringabout> Could you possibly provide something I can run? I could give a try tomorrow.
17:57:40FromDiscord<Phil> I'll try to do a minimal example in a bit, I'm pretty close to get the project to compile
17:58:02FromDiscord<Phil> Huh, and pretty close means it just compiled, to the drawing board with me then
17:58:45FromDiscord<Elegantbeef> I mean you need to remove `defaults`
17:58:50FromDiscord<Elegantbeef> My macro axes all `=` cause they're invalid code
17:59:31FromDiscord<Phil> Did that, didn't change the issue
18:00:09FromDiscord<Phil> sent a code paste, see https://play.nim-lang.org/#ix=4dXt
18:00:11FromDiscord<Elegantbeef> Likely is an issue with the new default intializers with an unexported error
18:00:29FromDiscord<ringabout> I agree so I will check soon
18:00:31FromDiscord<Phil> I'll cut it down a bit more but that was the quickest example I could whip up
18:00:38FromDiscord<Elegantbeef> also `newMemberShipTable` likely will not be allowed
18:00:52FromDiscord<Elegantbeef> iirc initalisers have a static requirement
18:00:59FromDiscord<Elegantbeef> Though dont know if the present implementation does
18:01:50FromDiscord<ringabout> It is probably fine
18:02:05FromDiscord<Phil> sent a code paste, see https://play.nim-lang.org/#ix=4dXw
18:02:47FromDiscord<Phil> In reply to @Elegantbeef "also `newMemberShipTable` likely will": For now I've received neither compiler warnings nor errors for it, so I'm assuming so far I'm safe
18:02:51FromDiscord<ringabout> In reply to @flywind "It is probably fine": You can use refs in the static context after all.
18:03:20FromDiscord<ringabout> sent a code paste, see https://play.nim-lang.org/#ix=4dXx
18:03:52FromDiscord<Phil> Wait a sec, sth's fishy, why does it still know of the "CampaignMemberships" type when I deleted that
18:03:53FromDiscord<Phil> one sec
18:04:42FromDiscord<Phil> Apparently it creates an alias for you, that's...kinda crazy, but also cool
18:08:06FromDiscord<guttural666> is it idiomatic to call the main file project_name.nim? or does one use main.nim?
18:09:51FromDiscord<Phil> In reply to @guttural666 "is it idiomatic to": Generally you'll find more camel case than snake case, and if you use `nimble init` you'll get `projectName.nim`
18:10:04FromDiscord<Phil> nimble init to initialize your project
18:10:58FromDiscord<guttural666> so projectName would be the main file typically?
18:11:06FromDiscord<guttural666> not main.nim
18:12:41FromDiscord<Phil> pretty much
18:13:05FromDiscord<Phil> Well, `projectName.nim`, in my case e.g. `nimstoryfont.nim` or `snorlogue.nim` or `tinypool.nim`
18:14:09FromDiscord<Patitotective> In reply to @guttural666 "is it idiomatic to": for my gui apps i use `main.nim` so its easier to rename the app name :p
18:16:58FromDiscord<wiga> I have a binary that i need to use in my project, is there a way to embed it and execute it when i run the program?
18:17:35FromDiscord<wiga> I found "staticRead" and "staticExec" but im not sure if its what i have to use
18:17:42FromDiscord<guttural666> In reply to @Patitotective "for my gui apps": like that more as well tbh, uniform entry point, easy to find for others
18:17:45FromDiscord<wiga> And im not sure how i can use it
18:19:38FromDiscord<guttural666> what's the syntax for importing my own nim files in my project? trying import src/dir_ops.nim atm
18:22:21FromDiscord<Patitotective> In reply to @guttural666 "what's the syntax for": https://nim-lang.org/docs/manual.html#modules
18:24:33FromDiscord<Patitotective> In reply to @wiga "I have a binary": do you need to execute it at run time?
18:31:16FromDiscord<wiga> In reply to @Patitotective "do you need to": Yes
18:31:36FromDiscord<wiga> Smth like upx does
18:31:44FromDiscord<wiga> Or any packer/embedder
18:34:07FromDiscord<guttural666> In reply to @Patitotective "https://nim-lang.org/docs/manual.html#modules": okay, it worked, the language server was just confused
18:40:07FromDiscord<guttural666> In reply to @exelotl "I'd do `n in": "for e in enum_type" also works mate, this is amazing, sadly no dot contains(int32): bool function I reckon
18:41:21FromDiscord<guttural666> is there an equiv to cpp lambdas in Nim?
18:42:12FromDiscord<guttural666> I think I tried defining a proc in a proc last night with let pls = proc ... but it said something about "has to be generic" or smth?
18:43:56FromDiscord<wiga> In reply to @wiga "Or any packer/embedder": Nvm found how
18:44:25*xaltsc joined #nim
18:46:30FromDiscord<guttural666> how do I escape carriage return in code for a string to stop it getting too long https://media.discordapp.net/attachments/371759389889003532/1033813683790422106/unknown.png
18:46:35FromDiscord<Phil> ElegantBeef , it appears I remain a fan of constructor
18:47:15FromDiscord<Phil> sent a code paste, see https://play.nim-lang.org/#ix=4dXK
18:48:08FromDiscord<Patitotective> sent a code paste, see https://play.nim-lang.org/#ix=4dXL
18:49:07FromDiscord<guttural666> sent a code paste, see https://play.nim-lang.org/#ix=4dXM
18:49:28FromDiscord<Phil> I assume that is because the fields get values associated with them at compile-time, which is not going to work with stuff like DateTime-fields
18:50:14FromDiscord<Phil> In reply to @guttural666 "doesn't seem efficient to": I mean, most of the time just throwing in `\n` into my echo's worked for me
18:50:54FromDiscord<Phil> https://media.discordapp.net/attachments/371759389889003532/1033814793313857566/unknown.png
18:51:28FromDiscord<guttural666> In reply to @Isofruit "I mean, most of": I mean formatting source code to be able to comply with 80 column width limitations, not formatting program output 😄
18:52:06FromDiscord<guttural666> (formatting source code without having to rely on ones own editor, which of course would be able to auto line break what you see)
18:52:24FromDiscord<Phil> sent a code paste, see https://play.nim-lang.org/#ix=4dXN
18:52:34FromDiscord<Phil> And for carrying over lines...errr that one's a good question
18:52:36FromDiscord<guttural666> ahhhhh
18:52:55FromDiscord<guttural666> I'll try this
18:53:28FromDiscord<Phil> Like, you can break stuff in between brackets in between multiple lines, but for those times where you chain like 5 proc calls... well I think you can break them into separate lines as well but you need to be really damn careful with indentation
18:54:35FromDiscord<guttural666> the syntax checker seems to be happy with this 🙂 thanks https://media.discordapp.net/attachments/371759389889003532/1033815715951685642/unknown.png
18:55:11FromDiscord<Phil> In reply to @guttural666 "the syntax checker seems": If you have types whose definition is that long I'd recommend a type alias tbh
18:55:19FromDiscord<Phil> For the tuple at least
18:55:23FromDiscord<Horizon [She/Her]> Anyone know if there's a simple-ish way to define a library for async and synchronous use?
18:55:24FromDiscord<Patitotective> sent a code paste, see https://play.nim-lang.org/#ix=4dXO
18:55:33FromDiscord<Patitotective> In reply to @guttural666 "the syntax checker seems": though those are two lines, not one
18:55:54FromDiscord<Horizon [She/Her]> Would a macro be suitable for that maybe?
18:56:14FromDiscord<guttural666> In reply to @Isofruit "If you have types": if it's something I would use more often, then I agree yes
18:56:50FromDiscord<Phil> Fair
18:57:05FromDiscord<Horizon [She/Her]> `when defined(async`LibraryName`): #do async code`↵`else: #do synchronous code`
18:58:11FromDiscord<Phil> In reply to @Event Horizon "`when defined(async`LibraryName`): #do async": I'm pretty sure prologue swaps out its entire HTTP server via a when statement, should be feasible.↵If necessary write your own module that has the same public procs wherever you need it
18:58:19FromDiscord<guttural666> In reply to @Patitotective "though those are two": 2 lines in the output as well you mean? that would not be what I wanted
18:58:42FromDiscord<Horizon [She/Her]> In reply to @Isofruit "I'm pretty sure prologue": Probably gonna do that yeah
18:59:04FromDiscord<Horizon [She/Her]> I wonder if there's a way to fuck with macros to define something for async and synchronous use
18:59:11FromDiscord<Horizon [She/Her]> There'd be no point though lmao
18:59:49FromDiscord<Phil> In reply to @Event Horizon "I wonder if there's": I mean, you could write a template that tries to compile a piece of code that only works if the async lib is there and otherwise assumes the sync one
19:00:00FromDiscord<Phil> and then do a when statement around that
19:00:19FromDiscord<Phil> Though that feels like it's going into Frankenstein territory tbh
19:00:21FromDiscord<Patitotective> In reply to @guttural666 "2 lines in the": if you use multiline strings, two lines mean two lines
19:01:40FromDiscord<Elegantbeef> Nim optimises constants so it'll emit `echo "hello world"`↵(@guttural666)
19:01:47FromDiscord<Horizon [She/Her]> In reply to @Isofruit "I mean, you could": Yeahh fair enough lol
19:02:41FromDiscord<Horizon [She/Her]> May need a template anyway for deciding against `proc method() {.async.}: Future[T]` `proc method(): T`
19:02:43FromDiscord<Elegantbeef> https://nim-lang.org/docs/asyncmacro.html#multisync.m%2Cuntyped
19:02:44FromDiscord<Elegantbeef> `{.multisync.}`↵(@Horizon [She/Her])
19:02:44FromDiscord<guttural666> In reply to @Elegantbeef "Nim optimises constants so": I figured as much yeah, but no inbuild way to escape a line break?
19:02:50FromDiscord<Horizon [She/Her]> Though I'd imagine a macro is probably better
19:02:58FromDiscord<Horizon [She/Her]> In reply to @Elegantbeef "`{.multisync.}` (<@909883978717204561>)": Oh? Will read
19:03:14FromDiscord<Elegantbeef> What are you after guttural?
19:04:14FromDiscord<Horizon [She/Her]> That's neat! But how do you know if you're using the async or sync variant? :P does it change the name at all?
19:04:23FromDiscord<Elegantbeef> Same name
19:04:32FromDiscord<guttural666> In reply to @Elegantbeef "What are you after": line break in a string in code, so that the string (with 1000 chars) will be displayed in a pleasant and readable manner no matter what your editor is capable of auto formatting for you
19:05:08FromDiscord<Horizon [She/Her]> In reply to @Elegantbeef "Same name": Ah so just doing `await` triggers the asynchronous proc, and otherwise it calls the synchronous proc? Useful
19:05:16FromDiscord<guttural666> (edit) "In reply to @Elegantbeef "What are you after": ... line" added "escaped"
19:05:23FromDiscord<Elegantbeef> It uses generic dispatch to distinguish between the two
19:05:23FromDiscord<Elegantbeef> https://github.com/nim-lang/Nim/search?q=multisync
19:07:41FromDiscord<Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4dXT
19:10:17FromDiscord<guttural666> okay, thanks for the info!
19:11:08FromDiscord<guttural666> I think a lot of these noob question should be dumped on the net to be discovered by search engine crawlers
19:11:27FromDiscord<Horizon [She/Her]> In reply to @Elegantbeef "It uses generic dispatch": Good to know!
19:11:45FromDiscord<guttural666> + the offical docs need unique adresses for all the entries, like cppreference.com
19:12:06FromDiscord<guttural666> (edit) "adresses" => "addresses"
19:12:26FromDiscord<Elegantbeef> 'PRs welcome' 😄
19:12:50FromDiscord<guttural666> what's PR?
19:13:01FromDiscord<Elegantbeef> Pull request
19:13:04FromDiscord<Phil> In reply to @guttural666 "I think a lot": Do what I do, open SO questions and answer them with what you found out!
19:13:12FromDiscord<Phil> SO is always nice for searchability
19:13:40FromDiscord<guttural666> ah 😄 I'm just a noob trying to learn, I don't think I can be of much help, other than being an annoying user asking and complaining 😄
19:14:05FromDiscord<guttural666> please no abbrev.
19:14:08FromDiscord<guttural666> SO?
19:14:12FromDiscord<Phil> I mean, we all learned at one point in time, figuring things out and then documenting it somewhere is always nice ^^↵SO = Stack Overflow
19:14:25FromDiscord<Phil> The holy bible of coding knowledge
19:14:32FromDiscord<Phil> Just holier and more useful
19:14:33FromDiscord<guttural666> Stack Overflow is a cesspool
19:14:41FromDiscord<Phil> Absolutely disagreed
19:15:03FromDiscord<Phil> 9/10 I find the answer on there that I was looking for
19:15:29FromDiscord<guttural666> yes, I'm talking about UX for people who aren't super pros at everything
19:15:38FromDiscord<Patitotective> In reply to @guttural666 "please no abbrev.": wdym
19:15:39FromDiscord<guttural666> just reading, sure
19:15:39FromDiscord<Phil> Even makes for a pretty damn good webpage to improve coding style (stackexchange.codereview)
19:15:56FromDiscord<guttural666> In reply to @Patitotective "wdym": SO = stack overflow, didn't get the abbbrevation
19:16:11FromDiscord<Patitotective> (edit) "In reply to @guttural666 "please no abbrev.": wdym ... " added "/j"
19:16:15FromDiscord<Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4dXW
19:17:04FromDiscord<guttural666> sent a code paste, see https://play.nim-lang.org/#ix=4dXX
19:17:16FromDiscord<guttural666> (edit) "https://play.nim-lang.org/#ix=4dXX" => "https://play.nim-lang.org/#ix=4dXY"
19:17:31FromDiscord<Phil> In reply to @guttural666 "yes, I'm talking about": I entirely disagree with SO's general UI around questions being in any way complicated (for user accounts I agree, but that doesn't detract from the point that the main UI is solid), though I guess that debate belongs in #offtopic
19:17:59FromDiscord<guttural666> yeah, didn't want to make a fuss about it
19:18:11FromDiscord<Phil> Though I guess I did accept that Stack-Overflow's internal search is atrocious and you should use google at all times ever
19:18:27FromDiscord<Phil> or related search engines
19:18:29FromDiscord<guttural666> duckduckgo/swisscows...
19:18:31FromDiscord<guttural666> haha
19:18:48FromDiscord<Phil> Ecosia is giving me pretty solid results lately, likely staying there
19:19:17FromDiscord<Patitotective> the best search engine is #main
19:19:18FromDiscord<guttural666> heard of that when I researched alts. but as you said this is offtopic
19:19:52FromDiscord<Phil> ~~Disagreed for I can't even find discussions where I know I was involved in them from 6 months ago~~
19:22:36FromDiscord<Patitotective> In reply to @Isofruit "~~Disagreed for I can't": Search > before:june from:Isofruit#5120
19:30:15FromDiscord<guttural666> any Vimmers hier?
19:30:22FromDiscord<guttural666> Denglisch
19:30:27FromDiscord<guttural666> -> here
19:31:09FromDiscord<guttural666> just wondered if there was an auto format tool for nim in vim
19:34:24FromDiscord<guttural666> + (lots of question): is there an optimal /preferred way in Nim to define Flags like bitsets in C/CPP?
19:34:39FromDiscord<Generic> set
19:34:53FromDiscord<guttural666> thanks, will look into that!
19:35:33FromDiscord<Patitotective> its all in the manual https://nim-lang.org/docs/manual.html#set-type-bit-fields
19:37:08FromDiscord<guttural666> I have that open at all times, just a bit tricky to a) find out the equiv. nomenclature in Nim and b) find out what "you do" in case xy in Nim yknow
19:38:09FromDiscord<guttural666> some things to get used to after years of ABAP and hobby C++ 😄
19:38:39FromDiscord<Generic> https://github.com/nim-lang/Nim/wiki/Nim-for-C-programmers
19:38:52FromDiscord<Generic> also my condolences for having to use ABAP
19:39:08FromDiscord<Elegantbeef> "Hobby C++" isnt that akin to sticking your hand in a meat grinder as a passtime?
19:40:19FromDiscord<guttural666> hahahahahaha
19:40:19FromDiscord<Phil> In reply to @Elegantbeef ""Hobby C++" isnt that": TBH I have heard more brutal horror stories around ABAP than C++
19:40:31FromDiscord<Phil> In fact, I've heard that C++ with smart pointers is said to be more bearable
19:40:54FromDiscord<guttural666> ABAP is a fucking breeze, such great UX
19:41:14FromDiscord<guttural666> C++ on the other hand
19:41:33FromDiscord<Elegantbeef> That's what big C++ wants you to think
19:41:38FromDiscord<guttural666> cock & ball torture, mainly because everything is just painful, i.e. reflection
19:41:43Amun-Raabap? are all cobol programmers are stone cold by now?
19:41:55Amun-Ras/are//
19:42:10FromDiscord<Phil> I'm pretty sure they just became rarer, I can not imagine all of them being down under by now
19:42:24FromDiscord<Phil> The power of compelling towards legacy maintenance is too strong for that
19:42:28FromDiscord<guttural666> In reply to @Amun-Ra "abap? are all cobol": looking at code in my company, apparently not, because they still use cobol idioms to this day
19:42:46Amun-Ra<;
19:42:58FromDiscord<guttural666> MOVE x TO pls.
19:43:01FromDiscord<guttural666> hell yeah
19:43:42FromDiscord<guttural666> ADD 1 to i. bahahahaha 😄
19:43:48Amun-RaMOVE x TO pls sound very canadian ;)
19:43:56FromDiscord<guttural666> hahahahaha 😄
19:44:00FromDiscord<Phil> Given that even relatively "modern" german Software companies may still stick with a 10 year old java version, I foresee a still long life for cobol
19:44:04Amun-Rasounds*
19:44:27FromDiscord<Phil> Because from what I've heard, banks are even more conservative
19:44:52Amun-Ra(sorry for those errors, sleep deprivation takes its toll)
19:44:53FromDiscord<guttural666> yeah, never change a running system ey? 😄
19:45:33FromDiscord<Xenon> In reply to @Isofruit "Given that even relatively": COBOL is never gonna die
19:45:40FromDiscord<guttural666> hahahaha 😄
19:45:41FromDiscord<Xenon> it's like C but more tenacious
19:45:51FromDiscord<Phil> In reply to @guttural666 "yeah, never change a": That works as long as your product isn't a webpage that needs to provide new features all the time, at that point you're starting to marry a frontend framework and then you're in a half-year-update cycle
19:46:12FromDiscord<guttural666> In reply to @Isofruit "That works as long": hear hear
19:47:06FromDiscord<guttural666> nah, ABAP is really beautiful tbh. the case insentivieness and preferred writing (keywords upper, other lower) makes it super readable
19:47:27FromDiscord<guttural666> it's just super ergonomical and well suited for the job
19:47:38FromDiscord<Elegantbeef> But but this is why people say Nim is unusable, the compiler confuses `word` for `wOrd`↵(@guttural666)
19:47:59FromDiscord<guttural666> hahahaha
19:48:06FromDiscord<guttural666> only plebs like me
19:48:17FromDiscord<Phil> If you ever want to read a book about the wars around what is called "style insensitivity", boy do I have an RFC for you
19:48:37FromDiscord<guttural666> HAHAHAHAHA 😄
19:48:52FromDiscord<Horizon [She/Her]> In reply to @guttural666 "cock & ball torture,": Aren't C++ looking to add reflection soon? Also iirc someone made a clang plugin for reflection too
19:48:53FromDiscord<Phil> People still write in that thing
19:49:07FromDiscord<Elegantbeef> C++ has reflection already
19:49:13FromDiscord<guttural666> In reply to @Event Horizon "Aren't C++ looking to": yeh, 20 years after cool kids did
19:49:21FromDiscord<Xenon> In reply to @Elegantbeef "C++ has reflection already": uh
19:49:22FromDiscord<Horizon [She/Her]> In reply to @Elegantbeef "C++ has reflection already": Oh already? Since when?
19:49:23FromDiscord<Xenon> kinda
19:49:25FromDiscord<Xenon> not really
19:49:27FromDiscord<Xenon> like
19:49:28FromDiscord<Xenon> at all
19:49:38FromDiscord<guttural666> In reply to @Elegantbeef "C++ has reflection already": not gooood enough!
19:49:50FromDiscord<guttural666> loop over struct members or GTFO haha
19:49:51FromDiscord<Elegantbeef> You shouldnt need runtime reflection like ever
19:50:00FromDiscord<ezquerra> One thing that @Araq did not mention in his 2.0 talk is incremental compilation. Is that totally out of the scope of nim 2.0?
19:50:34FromDiscord<Horizon [She/Her]> That's already done iirc
19:50:37FromDiscord<Horizon [She/Her]> For Nim source files
19:50:54FromDiscord<guttural666> is Araq the bulgarian guy who did some Nim talks, he's really cool
19:50:55FromDiscord<Horizon [She/Her]> Just use a C compiler that supports incremental compilation and you're gucci
19:51:10FromDiscord<Horizon [She/Her]> In reply to @guttural666 "is Araq the bulgarian": Araq is one of the founders of Nim
19:51:21FromDiscord<Phil> In reply to @guttural666 "is Araq the bulgarian": IIRC Araq is a prof in germany as a sidenote
19:51:25FromDiscord<ezquerra> Is it? If so, I'm surprised it would not be worth mentioning in a talk about what's coming in 2.0...
19:52:00FromDiscord<guttural666> In reply to @Isofruit "IIRC Araq is a": things to be proud of here: Nim and KDE
19:52:08FromDiscord<Elegantbeef> No Nim's IC is not done
19:52:08FromDiscord<Elegantbeef> It can work, but it very much is still not finished
19:52:45FromDiscord<ezquerra> @ElegantBeef, but it won't be part of 2.0?
19:53:08FromDiscord<Elegantbeef> Araq is the language creator, and is not bulgarian
19:53:22FromDiscord<ezquerra> And by _can_ work you mean that you think it is possible to make it work at some point, or that it already works in some cases?
19:53:46FromDiscord<Phil> Moving this towards more nim technical bits, Beef are you familiar on whether the default-value feature in nim devel is built to assign values at compile-time?
19:54:29FromDiscord<guttural666> oh, so it's an alias of Andi Rumpf, got it
19:54:48FromDiscord<Patitotective> isnt it andreas
19:54:53FromDiscord<Elegantbeef> Let's hope for the worlds sake that's where the german pride ends
19:54:54FromDiscord<Elegantbeef> No clue i'm not apart of the core dev team
19:54:56FromDiscord<guttural666> yeh
19:55:01FromDiscord<Phil> Andi is a common shorthand for Andreas in germany
19:55:12FromDiscord<guttural666> In reply to @Isofruit "Andi is a common": so it is
19:55:15Amun-Raand iirc dom is my countryman
19:56:22FromDiscord<Generic> Andi Scheuer 🤮
19:56:26FromDiscord<Phil> sent a code paste, see https://play.nim-lang.org/#ix=4dY5
19:56:39FromDiscord<Phil> In reply to @Generic "Andi Scheuer 🤮": Moderator I am hostile to having this name mentioned in my presnece
19:56:41FromDiscord<Elegantbeef> By can work it means it works in some isntances
19:56:42FromDiscord<Elegantbeef> From what araq has said he thinks that only static values should be allowed inside default values
19:56:42FromDiscord<Phil> (edit) "presnece" => "presence"
19:56:42FromDiscord<Elegantbeef> It should work regardless of backend
19:56:58FromDiscord<Phil> (jk)
19:57:09FromDiscord<guttural666> Andi B. Scheuert
19:57:35FromDiscord<Phil> In reply to @Elegantbeef "From what araq has": If that's how it'll be implemented constructor is pretty much embedded in my projects forever
19:58:11FromDiscord<Phil> Because I need solutions that allow me to instantiate objects also with DateTime fields
19:59:15FromDiscord<guttural666> what is this called in Nim? https://media.discordapp.net/attachments/371759389889003532/1033831991805759590/unknown.png
19:59:25FromDiscord<Patitotective> pragma
19:59:36FromDiscord<Patitotective> https://nim-lang.org/docs/manual.html#pragmas
19:59:37FromDiscord<guttural666> thanks
19:59:50FromDiscord<guttural666> it's like talking to the compiler yeah?
19:59:57FromDiscord<Patitotective> yea
20:00:00FromDiscord<Elegantbeef> `now` isnt static↵(@Phil)
20:00:07FromDiscord<guttural666> yeah
20:00:25FromDiscord<Phil> In reply to @guttural666 "it's like talking to": In a sense, you can provide meta information of all kinds, You can even fairly easily add your own pragmas and access them, it requires neither macros nor is it that hard
20:00:35FromDiscord<Elegantbeef> Araq's reasoning is that "It's better to have the constraint of statics and to back off if it's desired"
20:01:32FromDiscord<Phil> Meta information can include stating whether certain exceptions will be raised, whether something is gcsafe, telling the compiler to shut up and accept something is gcsafe when it disagrees, enabling or disabling certain compiler checks locally and more
20:02:51FromDiscord<guttural666> In reply to @Isofruit "Meta information can include": yeah, the power of these tools is just bonkers, that's what I could surmise from the tutorials, still learning the basics now tho
20:03:27FromDiscord<Phil> In reply to @guttural666 "yeah, the power of": Aye, makes sense, just want to state that these things are easier to grasp than macros ^^
20:03:43FromDiscord<Phil> Though there's a lot of things that can be done with them
20:03:53FromDiscord<Elegantbeef> sent a code paste, see https://paste.rs/ccl
20:04:09FromDiscord<Elegantbeef> Easy as hell serialisers that skip fields
20:04:54FromDiscord<guttural666> macros are just a Doom level power fantasy, not sure when I will be able to grasp the real potential of those, I already have real probs grasping what even is possible with C++ templates and compile time evaluations
20:05:17FromDiscord<Elegantbeef> If you can write code, you can write a macro to automate it
20:05:29FromDiscord<guttural666> not the primitive examples, I get those, but what even is possible with those tools
20:05:43FromDiscord<Elegantbeef> Async is implemented in user space with macros
20:05:52FromDiscord<guttural666> sure, but at that point imagination and creativity is required
20:06:01FromDiscord<Phil> In reply to @guttural666 "not the primitive examples,": Enthus1ast wrote a templating engine in those called nimja for example
20:06:12FromDiscord<Phil> Ormin as an ORM makes heavy use of them IIRC
20:06:40FromDiscord<Phil> Beef's constructor package generates object-construction procs with default values for you which has been a godsend for me and such a valuable feature they added it in nim 2
20:07:04FromDiscord<Phil> (which is currently the devel branch of the compiler)
20:07:53FromDiscord<guttural666> what does devel mean
20:08:23FromDiscord<Phil> devel is the name of a git branch where the current in-development version of nim is
20:08:34FromDiscord<guttural666> okay thanks
20:21:32FromDiscord<guttural666> if anybody's bored, you can look at my 100loc Num noob code to extract files from subfolders command line tool 😄 very proud indeed haha
20:21:34FromDiscord<guttural666> https://gitlab.com/hansdampfinger666/flubber
20:22:19FromDiscord<Elegantbeef> Enums dont need commas or `()` around strings
20:25:14FromDiscord<guttural666> you mean it could be ↵none = "none"↵pls = "pls"
20:25:30FromDiscord<guttural666> if I want to keep the 0...n numbering?
20:26:10FromDiscord<Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4dYc
20:26:30FromDiscord<guttural666> looks good, syntax checker doesn't complain, thanks for the heads up
20:26:43FromDiscord<Elegantbeef> sent a code paste, see https://paste.rs/PVi
20:26:47FromDiscord<Elegantbeef> Whoops missed a `(`
20:26:52FromDiscord<guttural666> got it
20:26:56FromDiscord<guttural666> thanks!
20:27:23FromDiscord<guttural666> okay, haven't dealt with exceptions yet
20:27:28FromDiscord<guttural666> gonna save that snippet
20:29:31FromDiscord<guttural666> do you use empty lines to relax the visual strain? the pure indent based syntax makes it a bit tricky, but I can't help myself throwing new lines in between single line statement blocks and multi line blocks for readability
20:29:36FromDiscord<Elegantbeef> I swear kate really just doesnt like me, downloaded the nightly appimage, my code suggestions work but the project plugin doesnt. Built from source, the project plugin works, but i have the same broken code suggestions
20:29:57FromDiscord<Elegantbeef> Of course you use emtpy lines to reduce visual noise
20:30:04FromDiscord<Elegantbeef> Just cause it's singificant white space doesnt mean you cannot have insignificant whitespace 😛
20:30:05FromDiscord<guttural666> Vim works great with the plugins and all, really recommended
20:30:17FromDiscord<guttural666> hehehe
20:30:17FromDiscord<guttural666> okay
20:30:21FromDiscord<Elegantbeef> Cool if i want to go back to 1993 i'll consider it
20:30:37FromDiscord<Elegantbeef> I use vim for one off things, not for a serious editor
20:31:35FromDiscord<auxym> vscode with vim keybindings is king
20:31:51FromDiscord<guttural666> if this doesn't whet your toungue I dunno what will https://media.discordapp.net/attachments/371759389889003532/1033840198519361536/unknown.png
20:32:26FromDiscord<Elegantbeef> I cannot stand the fact everything is alligned to a text grid
20:32:31FromDiscord<guttural666> https://media.discordapp.net/attachments/371759389889003532/1033840365955985479/unknown.png
20:32:44FromDiscord<auxym> vim isn't the only editor that can do pretty colors and a translucid bg 😉
20:33:08FromDiscord<Elegantbeef> image.png https://media.discordapp.net/attachments/371759389889003532/1033840521594024047/image.png
20:33:09FromDiscord<Elegantbeef> nice
20:33:25FromDiscord<auxym> +1 for fish indeed
20:33:25FromDiscord<guttural666> In reply to @auxym "vim isn't the only": heresy
20:33:34FromDiscord<guttural666> fish is just so guuuuud
20:33:48FromDiscord<guttural666> + Garuda Linux KDE, pure luv
20:33:56FromDiscord<Elegantbeef> But yea this is what i'm dealing with
20:33:57FromDiscord<Elegantbeef> image.png https://media.discordapp.net/attachments/371759389889003532/1033840725357506651/image.png
20:34:18FromDiscord<Elegantbeef> The suggestion window is gone, but the description window is there
20:34:18FromDiscord<Elegantbeef> Fuck me amirite
20:34:43FromDiscord<guttural666> heheheh
20:35:25FromDiscord<Elegantbeef> The best part is that it works properly briefly, but then suddenly stops
20:35:27FromDiscord<Elegantbeef> image.png https://media.discordapp.net/attachments/371759389889003532/1033841104317063269/image.png
20:35:36FromDiscord<Elegantbeef> I blame debian sid
20:35:47FromDiscord<guttural666> that looks pretty solid
20:36:11FromDiscord<guttural666> with the icons and all, very nice indeed
20:36:27FromDiscord<Elegantbeef> Yea kate works well, when it doesnt shit on your door step and run away
20:36:59FromDiscord<guttural666> yeah, Kate is pretty good, but KDevelop is not so good imo
20:37:49FromDiscord<guttural666> wonder why, must be sharing some of their code bases
20:38:03FromDiscord<guttural666> Debugger in KDevelop is a joke
20:38:12FromDiscord<Elegantbeef> They're both based on KTextWrite or w/e the KDE writing stuff
20:40:29FromDiscord<guttural666> okay
20:41:04*qwestion joined #nim
20:42:54FromDiscord<guttural666> just can't stand using the mouse anymore, just makes everything slow and painful, browsing with the Firefox Vim short cut extension is amazing btw.
20:43:14FromDiscord<guttural666> also Eclipse Vim Mode for ABAP uuuuuiiii
20:43:20FromDiscord<guttural666> chef's kiss
20:52:49*xaltsc quit (Quit: WeeChat 3.7.1)
20:53:07*xaltsc joined #nim
20:56:38*xaltsc quit (Client Quit)
20:57:03*xaltsc joined #nim
21:11:41FromDiscord<guttural666> how do I only extract the nth value of a return tuple? https://media.discordapp.net/attachments/371759389889003532/1033850221341462618/unknown.png
21:16:13FromDiscord<Patitotective> In reply to @guttural666 "how do I only": indexing, just like an array
21:16:30FromDiscord<Elegantbeef> `let (_, _, ext) = f.splitFile()` or `let ext = f.splitFile()[2]`
21:18:53FromDiscord<guttural666> ident through name alone would not be possible?
21:19:20FromDiscord<guttural666> position may be subject to change
21:21:03FromDiscord<guttural666> the API will prob never break, so anyway
21:21:05FromDiscord<guttural666> thanks! 😄
21:21:18FromDiscord<guttural666> hardcoded [2] it is
21:21:55FromDiscord<guttural666> should probably not complain, I wish I had index access to members in C/CPP structs
21:22:22FromDiscord<Patitotective> `f.splitFile().ext` (?↵)
21:22:24FromDiscord<Patitotective> (edit) "(?↵)" => "(?)"
21:22:42FromDiscord<guttural666> Bjarne why have you forsaken me
21:22:57FromDiscord<guttural666> In reply to @Patitotective "`f.splitFile().ext` (?)": what is this black magic
21:23:32FromDiscord<Patitotective> you can access named tuples just like object fields
21:23:55FromDiscord<guttural666> oh!
21:23:58FromDiscord<Patitotective> you should read some tutorial https://nim-lang.org/documentation.html
21:24:11FromDiscord<Patitotective> (edit) "tutorial" => "tutorials"
21:24:22FromDiscord<wiga> What IDE you guy using?
21:24:24FromDiscord<wiga> For nim
21:24:29FromDiscord<guttural666> suuure
21:24:43FromDiscord<Patitotective> In reply to @wiga "What IDE you guy": does sublime count?
21:24:43FromDiscord<guttural666> yeah, thanks mate!!
21:24:46FromDiscord<guttural666> Vim
21:24:50FromDiscord<wiga> In reply to @Patitotective "does sublime count?": Yes
21:25:10FromDiscord<wiga> Anything that has code completion
21:25:18FromDiscord<guttural666> Vim haha
21:25:49FromDiscord<wiga> Very Idiot Man (sry i had to say it)
21:34:51*fanta1 quit (Quit: fanta1)
21:37:06FromDiscord<guttural666> it's super effective
21:40:36FromDiscord<guttural666> also it's Voluptuous Incessant Manliness
21:56:29FromDiscord<Elegantbeef> Incestuous\
21:58:57FromDiscord<Bung> oh, three word dont in my programming english dictionary.
22:04:06FromDiscord<Bung> https://github.com/nim-lang/Nim/pull/20636 Elegantbeef how's this look
22:14:33FromDiscord<Elegantbeef> I'm uncertain what a 'forwardGeneric' means
22:19:19FromDiscord<Bung> hmm, CI fails, doesn't work, think the behavior should same as arc does
22:20:35FromDiscord<Bung> I thought when formal type allow generic and given tyForward that can return isGeneric
22:20:52FromDiscord<albassort> Nim works on microprocessors right
22:20:58FromDiscord<albassort> As long as I have GC off
22:21:11FromDiscord<Elegantbeef> you can just use `--mm:arc`
22:21:18FromDiscord<albassort> Ahh
22:21:28FromDiscord<albassort> Now I just need to figure out how microprocessors work
22:22:24FromDiscord<albassort> I want to store my code on nand and on init, it reads from nand and enters main
22:22:41FromDiscord<albassort> And you update the firmware by flashing nand with a new binary
22:22:56FromDiscord<albassort> Inb4 That's how it already works
22:44:15*krux02 quit (Read error: Connection reset by peer)
23:45:16*jmdaemon joined #nim
23:56:28FromDiscord<auxym> assuming you mean microcontrollers, #embedded . FYI microprocessor == "CPU" as in any normal computer
23:57:14*wallabra quit (Ping timeout: 276 seconds)
23:58:11*wallabra joined #nim