| 00:01:48 | FromDiscord | <sOkam!> sent a code paste, see https://play.nim-lang.org/#ix=4mEJ |
| 00:02:08 | FromDiscord | <Elegantbeef> Make an iterator |
| 00:02:54 | FromDiscord | <sOkam!> how do you yield a modifiable item in an iterator? |
| 00:03:01 | FromDiscord | <sOkam!> like mitems does |
| 00:03:31 | FromDiscord | <Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4mEK |
| 00:03:35 | FromDiscord | <leorize> you can do exactly what mitems does \:p |
| 00:03:53 | FromDiscord | <Elegantbeef> For all of Nim's jagged edges it's very user space extensible |
| 00:04:40 | FromDiscord | <leorize> unless there's a `magic` somewhere in a nim proc/iter/whatever definition, it's likely the exact code you need to reimplement the same thing |
| 00:05:38 | FromDiscord | <sOkam!> kk |
| 00:05:49 | FromDiscord | <sOkam!> where does magic process come from, btw? |
| 00:06:30 | FromDiscord | <sOkam!> i assume the code must be somewhere, but always wondered where that is stored (out of curiosity) |
| 00:06:41 | FromDiscord | <leorize> it's usually in the compiler |
| 00:06:52 | FromDiscord | <Elegantbeef> magic is implemented in the compiler, cause it cannot be done in userspace |
| 00:06:53 | FromDiscord | <leorize> search `m<Magic name>` to find it |
| 00:07:25 | FromDiscord | <Elegantbeef> Generally it's just base operations or backend specific operations |
| 00:08:54 | FromDiscord | <sOkam!> sent a code paste, see https://play.nim-lang.org/#ix=4mEL |
| 00:10:25 | FromDiscord | <Elegantbeef> `0..<height` but yes |
| 00:25:31 | * | genpaku quit (Remote host closed the connection) |
| 00:28:43 | * | genpaku joined #nim |
| 00:29:44 | * | lumo_e joined #nim |
| 00:32:30 | FromDiscord | <PunchCake> how would i program the esp32 with nim? |
| 00:33:55 | FromDiscord | <Elegantbeef> #embedded would help more. But you can use the same C compiler required for the chip and faff with cmake if you want |
| 00:33:59 | FromDiscord | <Elegantbeef> No clue if ratel supports that chip |
| 00:34:29 | FromDiscord | <sOkam!> sent a code paste, see https://play.nim-lang.org/#ix=4mEP |
| 00:34:49 | FromDiscord | <Elegantbeef> "sub iterators"? |
| 00:35:10 | FromDiscord | <Elegantbeef> You cannot chain iterators if that's what you mean |
| 00:35:14 | FromDiscord | <PunchCake> sent a code paste, see https://play.nim-lang.org/#ix=4mEQ |
| 00:43:35 | FromDiscord | <sOkam!> In reply to @PunchCake "are you making a": game of life |
| 00:43:44 | FromDiscord | <sOkam!> In reply to @Elegantbeef "You cannot chain iterators": kk ✍️ |
| 01:21:22 | FromDiscord | <Pusha> Could someone give me a simple explanation of the main difference between templates and macros in Nim? |
| 01:21:54 | FromDiscord | <Elegantbeef> Templates are code substitution which means they just replace the AST where your parameters are. Macros evaluate code on the VM and can operate on the input. |
| 01:22:02 | FromDiscord | <Elegantbeef> This means template are vastly simpler to read/write/reason |
| 01:37:23 | FromDiscord | <Goat> How do I read data thats being piped or redirected into my program? |
| 01:37:43 | FromDiscord | <Elegantbeef> `stdin` |
| 01:37:47 | FromDiscord | <Elegantbeef> I'd imagine |
| 01:52:25 | FromDiscord | <Goat> Yeah, I figured that, but I don't understand how I use it for that purpose |
| 01:54:20 | FromDiscord | <Elegantbeef> Should just be `readAll()` |
| 01:55:45 | FromDiscord | <Elegantbeef> Or `readLine` depending |
| 01:57:56 | FromDiscord | <auxym> hm, any way to list files in a directory at compile-time? `listFiles` doesn't seem to work (Error: undeclared identifier: 'listFiles') |
| 01:58:09 | FromDiscord | <auxym> https://nim-lang.org/docs/nimscript.html#listFiles%2Cstring |
| 01:58:24 | FromDiscord | <Elegantbeef> If you can `import std/os` you can use `walkFiles` |
| 01:58:30 | FromDiscord | <Elegantbeef> `listFiles` is only for nimVm |
| 02:01:23 | FromDiscord | <auxym> walkFiles was giving me cannot importc at compile-time |
| 02:02:29 | FromDiscord | <auxym> /home/francis/.choosenim/toolchains/nim-1.6.10/lib/pure/os.nim(2124, 11) Error: cannot 'importc' variable at compile time; glob |
| 02:03:09 | FromDiscord | <Elegantbeef> Walkdirs perhaps |
| 02:04:46 | FromDiscord | <Elegantbeef> Dont recall which are supported at CT |
| 02:05:59 | FromDiscord | <auxym> k, ill try |
| 02:26:32 | FromDiscord | <tfp> is there a way to turn an X to a ref X |
| 02:27:27 | FromDiscord | <Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4mF7 |
| 02:27:37 | FromDiscord | <Elegantbeef> `ref` in Nim is heap allocated so no there is not a way to make a stack value a ref |
| 02:31:05 | FromDiscord | <tfp> the syntax around ref objects is pretty janky ngl |
| 02:31:25 | FromDiscord | <Elegantbeef> I find it fine |
| 02:31:42 | FromDiscord | <tfp> it's definitely fine but it's dangerous |
| 02:32:00 | FromDiscord | <tfp> at least by most modern lang standards |
| 02:32:22 | FromDiscord | <tfp> the constructor syntax for non ref types with noInit is nice |
| 02:32:32 | FromDiscord | <Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4mF8 |
| 02:32:50 | FromDiscord | <tfp> ye but standards help a language scale u know |
| 02:32:50 | FromDiscord | <Elegantbeef> ref objects have the same semantics though |
| 02:32:54 | FromDiscord | <tfp> and then everyone benefits from network effects |
| 02:32:57 | FromDiscord | <Elegantbeef> `(ref MyType)(a: 100)` works as does `MyRefType(a: 100)` |
| 02:33:16 | FromDiscord | <Elegantbeef> The aforementioned ref syntax is only needed for primitives or similar really |
| 02:34:31 | FromDiscord | <tfp> the first one looks nice |
| 02:34:33 | FromDiscord | <tfp> let me tryt hat |
| 02:34:50 | FromDiscord | <Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4mF9 |
| 02:35:06 | FromDiscord | <tfp> i mean it's ok |
| 02:35:16 | FromDiscord | <tfp> but also won't benefit from network effect because it's not really standard |
| 02:35:29 | FromDiscord | <Elegantbeef> I dont see the network effect you're talking about |
| 02:35:38 | FromDiscord | <tfp> if nim was cut down and had syntax that was more appealing to systems level programmers i think it would take off |
| 02:35:52 | FromDiscord | <Elegantbeef> It'd also be Zig 😄 |
| 02:36:06 | FromDiscord | <tfp> kind of, zig is okay |
| 02:36:06 | FromDiscord | <Elegantbeef> Or Odin more accurately |
| 02:36:34 | FromDiscord | <tfp> i think orc with move semantics plus the strong metaprogramming is the killer app |
| 02:36:38 | FromDiscord | <Elegantbeef> but Nim is not C or C derived so it will be odd to people that havent written pascallian languages |
| 02:37:05 | FromDiscord | <tfp> but what i mean by network effect is, common patterns should be kind of standardized, so they can be operated over generically + understood immediately by other programmers |
| 02:37:17 | FromDiscord | <tfp> that enables network effect and helps the lang scale up better with libs and such |
| 02:37:22 | FromDiscord | <tfp> rust nailed that imo |
| 02:38:39 | FromDiscord | <Elegantbeef> I mean most people have a `type MyRef = ref object` or `type MyType = object` and unless you really care it you dont need any other semantics |
| 02:38:48 | FromDiscord | <tfp> well |
| 02:38:52 | FromDiscord | <Elegantbeef> A single type alias is pretty conventional |
| 02:38:54 | FromDiscord | <tfp> like for example none of these patterns work except the first one because patty generates the constructors for my enums |
| 02:39:05 | FromDiscord | <Elegantbeef> Why wouldnt they work? |
| 02:39:40 | FromDiscord | <Elegantbeef> Ah nvm i see |
| 02:39:47 | FromDiscord | <tfp> because i can't replace the constructors with versions that give me a RefTy |
| 02:40:10 | FromDiscord | <Elegantbeef> That's on patty imo |
| 02:40:14 | FromDiscord | <tfp> nooo way |
| 02:40:16 | FromDiscord | <tfp> it's on the language |
| 02:40:37 | FromDiscord | <Elegantbeef> It's on patty, it doesnt have a mechanism to define a `ref` constructor if you want, or a ref variant |
| 02:41:12 | FromDiscord | <tfp> why should it have to think about that? should every lib have to think about that? |
| 02:41:18 | FromDiscord | <Elegantbeef> Nim doesnt even have the concept of constructors aside from `MyType(a: ..., b: ...)` |
| 02:41:44 | FromDiscord | <Elegantbeef> There isnt even any desire from an large amount of the community for them aside from default values |
| 02:42:05 | FromDiscord | <tfp> well i am definitely spoiled by rust |
| 02:43:01 | FromDiscord | <Elegantbeef> Araq does seem interested in type bound operations, so perhaps one day user defined constructors will be a feature tacked onto types, but presently there isnt much but making constructors more succinct |
| 02:44:08 | FromDiscord | <Elegantbeef> I mean in rust how do you make a new RC with a given value? |
| 02:44:09 | FromDiscord | <Elegantbeef> I assume it has a `RC<T>.new(args)`? |
| 02:44:32 | FromDiscord | <tfp> Rc::new(value on the stack) |
| 02:44:44 | FromDiscord | <tfp> ::new is a pretty standard pattern |
| 02:44:59 | FromDiscord | <Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4mFb |
| 02:45:07 | FromDiscord | <Elegantbeef> Ok so the same thing i provided |
| 02:45:25 | FromDiscord | <Elegantbeef> So make an PR/RFC to make it apart of the stdlib is all i can really say |
| 02:45:49 | FromDiscord | <tfp> i also kind of think newX initX is a worse version |
| 02:45:57 | FromDiscord | <tfp> because it also suffers from no network effect |
| 02:45:59 | FromDiscord | <Elegantbeef> Most sane people agree |
| 02:46:02 | FromDiscord | <Elegantbeef> It's not generic |
| 02:46:04 | FromDiscord | <tfp> ye |
| 02:46:23 | FromDiscord | <Elegantbeef> I still think network effect is a silly term. 😄 |
| 02:46:45 | FromDiscord | <tfp> i like it because it encompasses the ability to operate over it generically + programmers grokking it immediately |
| 02:47:08 | FromDiscord | <tfp> the term is used a lot in other contexts but i bring it here because i like it |
| 02:47:28 | FromDiscord | <Elegantbeef> I prefer just talking about sane design 😛 |
| 02:47:43 | FromDiscord | <tfp> mm |
| 02:47:43 | FromDiscord | <Elegantbeef> embedding the type into the procedure name means you have 0 capability to compose |
| 02:48:06 | FromDiscord | <Elegantbeef> It neuters one of Nim's best features it's generic interfaces |
| 02:48:35 | FromDiscord | <tfp> i think the reason rust didn't make it a trait or something is because the argument list can be different |
| 02:48:43 | FromDiscord | <tfp> so it can't really be operated on generically anyway |
| 02:48:53 | FromDiscord | <tfp> but they do have Default which basically means a 0-arg ctor, which can be operated on generically |
| 02:49:03 | FromDiscord | <Elegantbeef> It can be |
| 02:49:17 | FromDiscord | <tfp> ye like for initializing empty lists, things like that, it's a quite nice trait to have |
| 02:49:27 | FromDiscord | <Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4mFc |
| 02:49:36 | FromDiscord | <Elegantbeef> Like you can require a specific type of constructor |
| 02:49:45 | FromDiscord | <tfp> ah yeah i think concepts are interesting |
| 02:49:51 | FromDiscord | <tfp> no dynamic dispatch for them is disappointing but i get it |
| 02:50:11 | FromDiscord | <tfp> honestly the main reason i'm using nim is because the metaprogramming is insanely powerful and i like ORC |
| 02:50:15 | FromDiscord | <Elegantbeef> Insert my dumb traitor package here |
| 02:50:22 | FromDiscord | <tfp> i'm using iface 🙂 |
| 02:50:27 | FromDiscord | <tfp> i think i looked at traitor though |
| 02:50:47 | FromDiscord | <tfp> nim is the closest thing to a language i would have made myself, but i don't really have the skills or resources to make one yet |
| 02:50:53 | FromDiscord | <Elegantbeef> Yea it's basically traits as the name implies, but i think i need to rewrite it |
| 02:51:44 | FromDiscord | <Elegantbeef> I very much love concepts, I'm quite annoyed by libraries that force using specific types |
| 02:52:09 | FromDiscord | <Elegantbeef> https://github.com/beef331/truss3d/blob/master/src/truss3D/shaders.nim#L132-L225 |
| 02:52:23 | FromDiscord | <Elegantbeef> I do want to expand this to my entire API |
| 02:52:46 | FromDiscord | <tfp> https://media.discordapp.net/attachments/371759389889003532/1069812454034526312/image.png |
| 02:52:51 | FromDiscord | <tfp> 😭 |
| 02:53:01 | FromDiscord | <Elegantbeef> lol |
| 02:53:42 | FromDiscord | <Elegantbeef> Think a normal variant makes more sense there 😛 |
| 02:53:54 | FromDiscord | <Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4mFe |
| 02:54:38 | FromDiscord | <tfp> oh i see |
| 02:54:48 | FromDiscord | <tfp> i figured that would have the same problem |
| 02:55:37 | FromDiscord | <Rika> it doesnt |
| 02:55:43 | FromDiscord | <Rika> theyre all in the same branch |
| 02:55:51 | FromDiscord | <Rika> the field isnt redefined in such case |
| 02:58:10 | FromDiscord | <tfp> i still have the issue between VecX and Mat now but its ok |
| 02:58:43 | FromDiscord | <tfp> one day i will rewrite everything in my own language, one day 🙂 |
| 03:15:04 | FromDiscord | <sOkam!> sent a code paste, see https://play.nim-lang.org/#ix=4mFi |
| 03:16:51 | FromDiscord | <sOkam!> i guess i would need to continue up towards the first loop to do that↵is it possible to go out of the outer in some way, from the inner? |
| 03:17:28 | FromDiscord | <sOkam!> (edit) "go" => "continue to next step" | "continue to next stepout of the outer ... ininside" added "loop" | "out of the outerloopin some way, from ... the" added "inside" |
| 03:17:39 | FromDiscord | <sOkam!> (edit) removed "out" |
| 03:17:53 | FromDiscord | <Elegantbeef> `break` |
| 03:18:08 | FromDiscord | <sOkam!> doesn't break break both? |
| 03:18:25 | FromDiscord | <Elegantbeef> Breaks out of the nearest `block` `while` or `for` |
| 03:18:32 | FromDiscord | <sOkam!> kk, that solves it then |
| 04:12:42 | * | LuxuryMode joined #nim |
| 04:20:48 | * | lumo_e quit (Quit: Quit) |
| 04:52:40 | * | arkurious quit (Remote host closed the connection) |
| 05:00:03 | * | derpydoo quit (Remote host closed the connection) |
| 05:16:06 | * | GreaseMonkey quit (Remote host closed the connection) |
| 05:17:12 | * | greaser|q joined #nim |
| 06:13:21 | * | deadmarshal_ quit (Ping timeout: 265 seconds) |
| 06:22:26 | * | LuxuryMode quit (Quit: Connection closed for inactivity) |
| 06:27:41 | * | deadmarshal_ joined #nim |
| 06:35:11 | * | LuxuryMode joined #nim |
| 06:44:24 | * | junaid_ joined #nim |
| 06:46:51 | * | junaid_ quit (Remote host closed the connection) |
| 07:04:21 | * | kenran` joined #nim |
| 07:47:36 | * | PMunch joined #nim |
| 08:53:31 | * | greaser|q quit (Changing host) |
| 08:53:31 | * | greaser|q joined #nim |
| 08:53:36 | * | greaser|q is now known as GreaseMonkey |
| 09:05:59 | FromDiscord | <UrNightmaree> is there a way to set `--os` flag through environment variable? |
| 09:06:36 | FromDiscord | <UrNightmaree> (edit) "is there a way to set `--os` flag ... through" added "option" |
| 09:12:38 | FromDiscord | <Phil> In reply to @UrNightmaree "is there a way": Yes.↵I assume you created your project with nimble init?↵Then you'll have noticed that you have a config.nims file in there |
| 09:13:40 | FromDiscord | <Phil> That's where you can write nimscript that will execute every time you compile the project.↵If you set a compiler flag in there (basically by just writing it as `--os:<whatevervalue>`, no frills) it'll get applied to the subsequent compilation command. |
| 09:14:32 | FromDiscord | <Phil> Nimscript is a subset of nim, so it can do a lot (though not everything) from nim.↵To read in an environmental variable there, use `getEnv` (https://nim-lang.org/docs/nimscript.html#getEnv%2Cstring%2Cstring) |
| 09:14:54 | FromDiscord | <UrNightmaree> In reply to @Isofruit "Yes. I assume you": > I assumed you created your project with nimble init?↵↵nope, i tried to package up Arturo which uses Nim to compile |
| 09:15:23 | FromDiscord | <UrNightmaree> i tried to add Arturo into Termux-packages repository |
| 09:15:41 | FromDiscord | <Phil> Ahh, check. See if there's a `config.nims` file.↵Alternatively, see if in the `.nimble` file there is a nimble task defined that defines how it should get compiled |
| 09:15:47 | * | azimut quit (Ping timeout: 255 seconds) |
| 09:17:29 | FromDiscord | <Phil> Nimble tasks are basically nimscript excerpts that you can use like functions and call from the outside with `nimble <task>`, so you can just change the compilation command based on if/when in the nimble task |
| 09:18:07 | FromDiscord | <Phil> (edit) "outside" => "shell" |
| 09:18:59 | FromDiscord | <UrNightmaree> In reply to @Isofruit "Ahh, check. See if": here, i found `config.nims` file↵<https://github.com/arturo-lang/arturo/blob/master/config.nims> |
| 09:19:51 | FromDiscord | <UrNightmaree> i only found switch statement for checking if `cc` environment variable is valid CC compiler |
| 09:20:38 | FromDiscord | <UrNightmaree> i might want to check `build.nims` as well |
| 09:20:44 | FromDiscord | <Phil> At a glance that cc looks more like it's for checking if the commandline parameter cc is set with gcc |
| 09:20:49 | FromDiscord | <UrNightmaree> since its a script for building the binary |
| 09:21:21 | FromDiscord | <Phil> The `get` proc is not the same as `getEnv`:↵> Retrieves a configuration 'key' like 'gcc.options.always'. |
| 09:22:48 | FromDiscord | <Phil> sent a long message, see http://ix.io/4mGd |
| 09:23:11 | FromDiscord | <Phil> (edit) "http://ix.io/4mGd" => "http://ix.io/4mGe" |
| 09:23:31 | FromDiscord | <Phil> (edit) "http://ix.io/4mGe" => "http://ix.io/4mGf" |
| 09:23:41 | FromDiscord | <UrNightmaree> im new to Nim (and it's ecosystem) 😅 |
| 09:24:00 | FromDiscord | <UrNightmaree> anyway, i found this <https://github.com/arturo-lang/arturo/blob/v0.9.80/build.nims#L51> |
| 09:24:41 | FromDiscord | <UrNightmaree> seems its for targeting armv7/aarch64 |
| 09:24:58 | FromDiscord | <UrNightmaree> instead x86 and x86_64 |
| 09:25:01 | * | redj quit (Ping timeout: 252 seconds) |
| 09:25:12 | FromDiscord | <Phil> God damn that's a complex build setup |
| 09:25:47 | * | redj joined #nim |
| 09:26:04 | FromDiscord | <Elegantbeef> Techincally arturo's build also installs it |
| 09:26:23 | FromDiscord | <Elegantbeef> It's actually quite nice to develop on, though it's a big script |
| 09:27:01 | FromDiscord | <Phil> I'm not seeing it set any os flag |
| 09:27:21 | FromDiscord | <UrNightmaree> i only found the `arm` and `arm64` argument |
| 09:27:28 | FromDiscord | <UrNightmaree> i also checks it workflow for Raspberry Pi |
| 09:27:36 | FromDiscord | <UrNightmaree> also uses the argument above |
| 09:27:57 | FromDiscord | <Phil> That's for the CPU flags. You wanted to set the os flag, which seems unset so you're... slightly less likely to break something by manipulating this |
| 09:28:33 | FromDiscord | <Phil> You should be able to get away with just adding an if/when statement near the bottom that sets that flag |
| 09:28:41 | FromDiscord | <UrNightmaree> i could just create `.patch` for it |
| 09:28:53 | FromDiscord | <UrNightmaree> patching some parts of the script |
| 09:29:08 | FromDiscord | <UrNightmaree> anyway thanks for the help |
| 09:29:14 | FromDiscord | <Phil> .patch files are a thing? |
| 09:29:19 | FromDiscord | <UrNightmaree> In reply to @Isofruit ".patch files are a": yes |
| 09:29:21 | FromDiscord | <Phil> For nimscript I mean |
| 09:29:38 | FromDiscord | <Elegantbeef> They're trying to package arturo for termux |
| 09:29:46 | * | tinytoast joined #nim |
| 09:29:48 | FromDiscord | <UrNightmaree> In reply to @Elegantbeef "They're trying to package": that's correct |
| 09:30:36 | FromDiscord | <Phil> Yeah but wasn't it "compile arturo to C , proceed with compiling and packaging with normal C tools" ? |
| 09:30:42 | FromDiscord | <Phil> (edit) "," => "then," |
| 09:30:51 | FromDiscord | <Phil> (edit) "C then," => "C,then" |
| 09:31:02 | FromDiscord | <Elegantbeef> I'm uncertain what's wrong with the current setup though |
| 09:31:37 | FromDiscord | <UrNightmaree> In reply to @Elegantbeef "I'm uncertain what's wrong": i need to get it working on all architecture that Termux are supported |
| 09:32:01 | FromDiscord | <UrNightmaree> currently Termux supports arm, aarch64, x64_86 and i686 |
| 09:32:10 | * | Batzy quit (Ping timeout: 260 seconds) |
| 09:32:10 | * | tinystoat quit (Ping timeout: 260 seconds) |
| 09:32:19 | * | Batzy joined #nim |
| 09:33:02 | FromDiscord | <Elegantbeef> Is this cross compiled? |
| 09:34:56 | FromDiscord | <Elegantbeef> On second though given i have 0 clue the path required, i'll shush |
| 09:42:05 | FromDiscord | <UrNightmaree> In reply to @Elegantbeef "Is this cross compiled?": yes |
| 10:12:26 | * | LuxuryMode quit (Quit: Connection closed for inactivity) |
| 11:15:28 | FromDiscord | <Phil> Can you alias a proc? |
| 11:16:38 | FromDiscord | <Phil> Nevermind, doing a variable assignment is essentially aliasing of a proc |
| 11:16:46 | FromDiscord | <Phil> (edit) "Nevermind, doing a variable assignment ... is" added "to that proc" |
| 11:28:55 | PMunch | @Phil, except for generics |
| 12:08:06 | FromDiscord | <turbo> For strutils, why does `val.parseEnum[MyType]()` not compile, but `parseEnum[MyType](val)` does? |
| 12:10:54 | FromDiscord | <Phil> Huh that one I didn't have before, maybe a generic limitation |
| 12:11:09 | FromDiscord | <ringabout> Documented in the Nim Manual, the limitation of method call syntax; use `val.parseEnum[:MyType]()` instead |
| 12:11:49 | FromDiscord | <Phil> https://nim-lang.org/docs/manual.html#templates-limitations-of-the-method-call-syntax↵The section for it |
| 12:12:15 | FromDiscord | <Phil> As well as: https://nim-lang.org/docs/manual.html#procedures-method-call-syntax |
| 12:13:10 | FromDiscord | <ringabout> > The method call syntax conflicts with explicit generic instantiations: p[T](x) cannot be written as x.p[T] because x.p[T] is always parsed as (x.p)[T]. |
| 12:13:49 | FromDiscord | <ringabout> > The [: ] notation has been designed to mitigate this issue: x.p[:T] is rewritten by the parser to p[T](x), x.p[:T](y) is rewritten to p[T](x, y). Note that [: ] has no AST representation, the rewrite is performed directly in the parsing step. |
| 12:28:39 | * | jmdaemon quit (Ping timeout: 252 seconds) |
| 12:33:48 | * | dnh joined #nim |
| 12:47:28 | * | ltriant quit (Ping timeout: 248 seconds) |
| 12:49:13 | * | ltriant joined #nim |
| 12:54:19 | * | ltriant quit (Ping timeout: 252 seconds) |
| 12:59:09 | FromDiscord | <sOkam!> If I have a `type Thing = ref object of RootObj`, and I want it to be linked in two spots, where one modifies the other, do I need to define it inside the other places as `ref Thing` or can it just be `Thing` and be linked with the others directly? 🤔 |
| 13:26:14 | PMunch | Nope, Thing is now a ref object and will be passed by reference (what you call linking) |
| 13:30:43 | * | ltriant joined #nim |
| 13:32:36 | * | fallback quit (Remote host closed the connection) |
| 13:43:47 | FromDiscord | <ringabout> sent a code paste, see https://play.nim-lang.org/#ix=4mHd |
| 13:46:24 | * | fallback joined #nim |
| 14:23:38 | FromDiscord | <fabricio> is this a bug? |
| 14:25:59 | FromDiscord | <fabricio> sent a code paste, see https://play.nim-lang.org/#ix=4mHh |
| 14:28:50 | FromDiscord | <ringabout> `static: main()` works. Looks like a known bug or needs a better error message. |
| 14:30:05 | FromDiscord | <ringabout> In reply to @fabricio "is this a bug?": See also https://github.com/nim-lang/Nim/issues/20545 |
| 14:32:02 | FromDiscord | <fabricio> ah yes I see, I forgot `const`s are a compile time thing |
| 14:36:41 | FromDiscord | <pietroppeter> there might be a nim related post (barely) on HN front page... |
| 14:38:17 | FromDiscord | <ringabout> sent a code paste, see https://play.nim-lang.org/#ix=4mHj |
| 14:39:45 | PMunch | @pietroppeter, neat. I might've commented |
| 14:39:58 | PMunch | Of course not the happiest topic.. |
| 14:40:43 | * | advesperacit joined #nim |
| 14:42:59 | FromDiscord | <pietroppeter> yeah, although a topic which might actually benefit from the additional visibility... |
| 14:54:00 | FromDiscord | <fabricio> sent a code paste, see https://play.nim-lang.org/#ix=4mHo |
| 14:54:20 | FromDiscord | <ringabout> No problem |
| 15:10:48 | PMunch | @pietroppeter, most definitely |
| 15:12:11 | FromDiscord | <auxym> this I assume? https://news.ycombinator.com/item?id=34594743 |
| 15:12:24 | FromDiscord | <auxym> IMO it's a very good thing that it gets more visibility |
| 15:12:33 | PMunch | @pietroppeter, by the way I believe Nim should compile under MSVC, at least my friend spent a lot of time making it work a while ago |
| 15:13:43 | FromDiscord | <ringabout> The compile time of VCC is much more than mingw for the same Nim program. |
| 15:14:38 | FromDiscord | <ringabout> So sometimes I don't even want to debug VCC related issues reported. |
| 15:14:49 | FromDiscord | <ringabout> (edit) "want to debug" => "feel like debugging" |
| 15:15:54 | * | PMunch quit (Quit: Leaving) |
| 15:28:40 | FromDiscord | <pietroppeter> In reply to @auxym "this I assume? https://news.ycombinator.com/item?id": yeah, we do not usually post the link here to avoid votes coming from clicking being flagged in some way |
| 15:29:27 | FromDiscord | <pietroppeter> In reply to @PMunch "<@869267093110026320>, by the way": good to know, in desperate case having an additional option to test is good |
| 15:46:03 | * | LuxuryMode joined #nim |
| 15:49:29 | * | arkurious joined #nim |
| 15:52:31 | FromDiscord | <auxym> ah sorry, wasn't aware |
| 16:05:00 | FromDiscord | <Nerve> Need some static introspection help. I want to compare the fields of two similar types, and transfer the fields with the same name and type. How would I go about it? |
| 16:05:08 | FromDiscord | <Nerve> (edit) "Need some static introspection help. I want to compare the fields of two similar ... types," added "object" |
| 16:08:07 | FromDiscord | <Phil> Why not just 2 nested loops using the fieldPairs operator and a when statement that assigns one field to the other if the names are identical? |
| 16:08:35 | FromDiscord | <Nerve> Is that really the simplest way to do it? Alright then |
| 16:11:12 | FromDiscord | <Nerve> How do I do the assignment? I have a string of a field name, how do I key into the object field? |
| 16:11:30 | FromDiscord | <Hourglass [She/Her]> In reply to @Isofruit "Nevermind, doing a variable": Templates can always be used for one, and then there's also `const myAlias = myProc` I think |
| 16:11:56 | FromDiscord | <Hourglass [She/Her]> Oh was already answered |
| 16:13:26 | FromDiscord | <Phil> I'm on the go, so somebody else will have to do if you want explicit examples, but basically you can just take the field variable from the iterator and assign to it like it was a normal mutable variable |
| 16:13:44 | FromDiscord | <Nerve> Okay I'll give it a try |
| 16:14:07 | FromDiscord | <Nerve> Interesting, `nim check` appears to like it |
| 16:14:20 | FromDiscord | <Nerve> So `fieldPairs` does indeed yield mutable references |
| 16:15:09 | arkanoid | https://nim-lang.org/docs/manual.html#foreign-function-interface-packed-pragma == packed with 1-byte alignment? |
| 16:16:56 | FromDiscord | <Phil> In reply to @Nerve "So `fieldPairs` does indeed": It's strictly speaking the compiler writing obj.fieldname for you for every field on the object |
| 16:17:05 | FromDiscord | <Phil> That's why it works and is mutable |
| 16:17:39 | FromDiscord | <Phil> So in truth there isn't even iteration going on, the compiler unrolls that loop at compile time |
| 16:21:28 | FromDiscord | <huantian> it might be worth it to make a macro though |
| 16:21:53 | FromDiscord | <huantian> actually hmm I assumed it would be O(n^2) even thought it's rolled out but actually it's probably not |
| 16:22:06 | FromDiscord | <huantian> because you use compile time comparisons to check if the fields are equal |
| 16:29:21 | * | kenran` quit (Remote host closed the connection) |
| 17:33:05 | * | azimut joined #nim |
| 17:45:38 | * | jmdaemon joined #nim |
| 18:17:07 | FromDiscord | <Nerve> sent a long message, see http://ix.io/4mIj |
| 18:17:15 | FromDiscord | <Nerve> (edit) "http://ix.io/4mIj" => "http://ix.io/4mIk" |
| 18:17:45 | FromDiscord | <Nerve> (edit) "http://ix.io/4mIk" => "http://ix.io/4mIm" |
| 18:19:36 | FromDiscord | <Nerve> The advantages of Nim are obvious here, I can distribute a dependency-free executable we can use on servers or hand around for local use on workstations. The Python story was woefully inadequate on that front; Nuitka amazingly worked but it did not produce particularly fast code, and the binaries were quite large. |
| 18:20:04 | FromDiscord | <Nerve> (edit) "The advantages of Nim are obvious here, I can distribute a dependency-free executable we can use on servers or hand around for local use on workstations. The ... Pythonpackaging" added "standard" | "standardPython ... story" added "packaging" |
| 18:20:28 | FromDiscord | <Nerve> I have a roughly 10x reduction in speed and binary size with Nim |
| 18:20:40 | FromDiscord | <Nerve> (edit) "reduction" => "increase" | "increasein speed and ... binary" added "reduction in" |
| 18:20:53 | FromDiscord | <Nerve> (edit) "10x" => "10-100x" | "10-100xincrease in speed and ... reduction" added "10x" |
| 18:28:58 | FromDiscord | <Phil> sent a long message, see http://ix.io/4mIp |
| 18:29:32 | FromDiscord | <Arathanis> In reply to @Nerve "Software engineering philosophy question.": It sounds like you could leverage `std/jsonutils` and its `fromJsonHook` and `toJsonHook` (de)serialization hooks to get to the concrete types you want.↵↵Combined with `fusion/matching` you can have a lot of expressive power to convert to concrete types. |
| 18:29:38 | FromDiscord | <Phil> (edit) "http://ix.io/4mIp" => "http://ix.io/4mIq" |
| 18:30:31 | FromDiscord | <Phil> In reply to @Nerve "Software engineering philosophy question.": I also generally do not recommend messing around with JsonNode.↵You just lose out on one of the things that makes nim beneficial over python: You get free sanity checking via the types.↵JsonNode takes that away from you. |
| 18:30:58 | FromDiscord | <Phil> That's why I wouldn't go with std/json in the first place and go straight to jsony since I'd want to avoid JsonNode anyway and have a bunch of static types everywhere. |
| 18:32:01 | FromDiscord | <Phil> (edit) "In reply to @Nerve "Software engineering philosophy question.": I also generally do not recommend messing around with JsonNode.↵You just lose out on one of the things that makes nim beneficial over python: You get free sanity checking ... via" added "at compile time" |
| 18:32:45 | FromDiscord | <ShalokShalom> @Phil in F#, you get typed JSON, just by reading it from the file ↵↵does something exist for Nim too?↵We call it type providers |
| 18:33:13 | FromDiscord | <Phil> In reply to @ShalokShalom "<@180601887916163073> in F#, you": I'm not familiar with such a thing existing.↵Really tbh, I just beat all my JSON problems with jsony and writing a lot of types |
| 18:33:34 | FromDiscord | <ShalokShalom> yeah, you get all the types inferred, this way |
| 18:34:05 | FromDiscord | <ShalokShalom> typescript has this also, I just see |
| 18:34:11 | FromDiscord | <Phil> Yeah, that would be pretty neat to have more convenience that way.↵Writing the types out imo is an acceptable amount of work.↵Yeah it's a lot of typing, but in the end the clarity is worth it. |
| 18:34:30 | FromDiscord | <Phil> In reply to @ShalokShalom "typescript has this also,": ? |
| 18:34:57 | FromDiscord | <ShalokShalom> although only for JSON, and not yet for html, csv, xml, custom ones, etc |
| 18:35:39 | FromDiscord | <ShalokShalom> In reply to @Isofruit "?": Typescript has this feature also, called Type Providers |
| 18:35:46 | FromDiscord | <Phil> Ohhhh in the sense that you can basically "import json" and it imports the json as an object |
| 18:35:49 | FromDiscord | <ShalokShalom> Although just for one data type |
| 18:36:02 | FromDiscord | <Phil> (edit) "Ohhhh in the sense that you can basically "import json" and it imports the json as an object ... " added "with inferred types and IDE type hints etc." |
| 18:36:33 | FromDiscord | <ShalokShalom> In reply to @Isofruit "Ohhhh in the sense": yeah, you get autocompletion when you work with the document and it is just fully typed out of the box |
| 18:37:05 | FromDiscord | <ShalokShalom> FSharp has that for basically any important data type and then some |
| 18:39:07 | FromDiscord | <Arathanis> sent a code paste, see https://play.nim-lang.org/#ix=4mIv |
| 18:39:50 | FromDiscord | <Arathanis> (edit) "https://play.nim-lang.org/#ix=4mIv" => "https://play.nim-lang.org/#ix=4mIw" |
| 18:40:05 | FromDiscord | <Arathanis> (edit) "https://play.nim-lang.org/#ix=4mIw" => "https://play.nim-lang.org/#ix=4mIx" |
| 18:40:17 | FromDiscord | <Arathanis> (edit) "https://play.nim-lang.org/#ix=4mIx" => "https://play.nim-lang.org/#ix=4mIv" |
| 18:42:15 | FromDiscord | <Phil> sent a long message, see http://ix.io/4mIA |
| 18:42:50 | FromDiscord | <Phil> (edit) "http://ix.io/4mIA" => "http://ix.io/4mIC" |
| 18:43:56 | FromDiscord | <Phil> (edit) "http://ix.io/4mIC" => "http://ix.io/4mID" |
| 18:45:18 | FromDiscord | <Phil> And the nice side-effect of jsony is that it's even faster than std/json by a pretty significant margin (like, roughly factor 5-6x faster) |
| 18:45:34 | FromDiscord | <Arathanis> How does it handle heterogeneous data? |
| 18:45:40 | FromDiscord | <Arathanis> I am looking at it now, it looks nice. |
| 18:46:21 | FromDiscord | <Phil> define heterogeneous data?↵It basically forces you into "static" json-patterns.↵You want a different field, you'll need a different type |
| 18:46:51 | FromDiscord | <Phil> That's because it parses json directly to a type, it does not use object variants as an intermediary the way std/json does |
| 18:47:00 | FromDiscord | <Phil> (edit) "That's because it parses json directly to a type, it does not use object variants ... as" added "(JsonNode)" |
| 18:47:17 | FromDiscord | <Arathanis> sent a code paste, see https://play.nim-lang.org/#ix=4mIE |
| 18:47:31 | FromDiscord | <jmgomez> In reply to @ShalokShalom "<@180601887916163073> in F#, you": Nim is like in another league in that regard :nim1: |
| 18:47:52 | FromDiscord | <jmgomez> You can do a "type provider" from chat gpt if you wish so 😛 |
| 18:48:17 | FromDiscord | <Phil> sent a code paste, see https://play.nim-lang.org/#ix=4mIF |
| 18:48:52 | FromDiscord | <Arathanis> In reply to @Isofruit "Hmm let me see": sweet, I appreciate it |
| 18:48:58 | FromDiscord | <Phil> Basically I'd need to deal with "oh god" having possibly either string, int or float values? |
| 18:49:40 | FromDiscord | <Arathanis> sent a code paste, see https://play.nim-lang.org/#ix=4mIG |
| 18:49:42 | FromDiscord | <Phil> Wait, that's not valid json! |
| 18:49:48 | FromDiscord | <Arathanis> (edit) "https://play.nim-lang.org/#ix=4mIG" => "https://play.nim-lang.org/#ix=4mIH" |
| 18:49:50 | FromDiscord | <Phil> because string and 10 aren't key value pairs |
| 18:50:00 | FromDiscord | <Arathanis> oh sorry it was supposed to be a list! |
| 18:50:01 | FromDiscord | <Arathanis> let me edit it |
| 18:50:16 | FromDiscord | <Arathanis> still consuming my morning coffee haha |
| 18:50:49 | FromDiscord | <Phil> Ohhhh in that case I get it more |
| 18:51:28 | FromDiscord | <Arathanis> sent a code paste, see https://play.nim-lang.org/#ix=4mII |
| 18:51:36 | FromDiscord | <Arathanis> (edit) "https://play.nim-lang.org/#ix=4mII" => "https://play.nim-lang.org/#ix=4mIJ" |
| 18:51:40 | FromDiscord | <Arathanis> i would never structure my JSON this way if I could help it |
| 18:51:51 | FromDiscord | <Arathanis> but the world doesn't follow what i would consider best practices |
| 18:51:59 | FromDiscord | <Arathanis> ive had to ingest some nonsense haha |
| 18:52:06 | FromDiscord | <Arathanis> its important that my JSON library be able to handle this |
| 18:52:14 | FromDiscord | <Phil> In reply to @Arathanis "still consuming my morning": Yeah that's the kind of thing where std/json basically does better or rather in a smaller-scale you'll partially rebuilding a less "liberal" version of std/json↵You'll have to define an object variant type for that list, which defines what is allowed in there and what isn't, and then you can define json hook and parsehook for it for to-json and from-json conversions |
| 18:52:31 | FromDiscord | <Phil> "Does better" in the sense of being more convenient |
| 18:52:36 | FromDiscord | <Arathanis> i appreciate you taking the time to look at this. |
| 18:52:59 | FromDiscord | <Phil> jsony likely will still be faster because of the direct conversion, but you'll end up with a type containing an object variant |
| 18:52:59 | FromDiscord | <Arathanis> If you can find a good way to leverage jsony to do this id love to see it. |
| 18:54:16 | FromDiscord | <alexb> sent a code paste, see https://play.nim-lang.org/#ix=4mIL |
| 18:54:17 | FromDiscord | <alexb> sent a code paste, see https://paste.rs/35R |
| 18:54:40 | FromDiscord | <alexb> Could you please help me to fix this |
| 18:55:59 | FromDiscord | <Ayano> https://media.discordapp.net/attachments/371759389889003532/1070054857702113390/image.png |
| 18:56:19 | FromDiscord | <Ayano> (edit) "" => "what am i doing wrong" |
| 18:56:27 | FromDiscord | <Hourglass [She/Her]> sent a long message, see http://ix.io/4mIM |
| 18:56:40 | FromDiscord | <Hourglass [She/Her]> In reply to @Ayano "what am i doing": `r` needs to be defined in a variable |
| 18:56:57 | FromDiscord | <Hourglass [She/Her]> `reader: var Reader` means that it needs a mutable `Reader` instance |
| 18:57:19 | FromDiscord | <Hourglass [She/Her]> In reply to @Hourglass, When the Hour Strikes "`r` needs to be": (I mean, `var` instead of using `let` or `constant`) |
| 18:57:30 | FromDiscord | <Ayano> https://media.discordapp.net/attachments/371759389889003532/1070055237836091392/image.png |
| 18:57:57 | FromDiscord | <Ayano> https://media.discordapp.net/attachments/371759389889003532/1070055351761764492/image.png |
| 18:58:12 | FromDiscord | <Ayano> it should be var right |
| 18:58:29 | FromDiscord | <Hourglass [She/Her]> Huh yeah it should be... I'm not sure then, sorry |
| 18:58:37 | FromDiscord | <Ayano> yea same here |
| 18:59:07 | FromDiscord | <Nerve> In reply to @Isofruit "I also generally do": Part of my issue is that I'm converting from a binary format which somewhat messily and ambiguously stores these variant formats (values that serve double uses, arrays that have varying lengths depending), to well-structured JSON that reflects the variant of data it contains. |
| 18:59:10 | FromDiscord | <Phil> In reply to @Ayano "": Does that error also pop up when compiling? |
| 18:59:48 | FromDiscord | <Ayano> yes |
| 18:59:52 | FromDiscord | <Ayano> https://media.discordapp.net/attachments/371759389889003532/1070055832693243914/image.png |
| 19:00:06 | FromDiscord | <Arathanis> In reply to @Nerve "Part of my issue": does my example help you at all with this? It sounds like it might be just what you need. |
| 19:00:07 | FromDiscord | <ShalokShalom> In reply to @jmgomez "You can do a": how do you mean? |
| 19:01:30 | FromDiscord | <jmgomez> In reply to @ShalokShalom "how do you mean?": type providers are code generators, in Nim you can modify the AST therefore generate code based on any input |
| 19:01:40 | FromDiscord | <Nerve> In reply to @Arathanis "does my example help": Like I just mentioned, my issue isn't switching from existing JSON, it's switching from an existing binary format; your hook solution only comes in after the data is already in JSON format |
| 19:01:47 | * | sommerjan joined #nim |
| 19:02:02 | FromDiscord | <Arathanis> In reply to @Nerve "Like I just mentioned,": ohhh i understand now |
| 19:02:44 | FromDiscord | <ShalokShalom> In reply to @jmgomez "type providers are code": sure, and does this include some type provider like behavior is already present, or just relatively simple to make |
| 19:02:51 | FromDiscord | <ShalokShalom> like, can I use this now? |
| 19:03:57 | FromDiscord | <jmgomez> `import std/[macros, genasts]` should have you covered 😛 |
| 19:04:10 | FromDiscord | <jmgomez> what do you want to do? |
| 19:05:58 | FromDiscord | <ShalokShalom> just the default behavior of a type provider |
| 19:06:23 | FromDiscord | <ShalokShalom> without me having to write a macro for it. i |
| 19:06:34 | FromDiscord | <ShalokShalom> would imagine, there is already one, maybe? |
| 19:06:43 | sommerjan | Hi, I'm trying to do something like let x = case optional_result of Some(@x): "some" of None(): "none". I'm using fusion/matching with {.experimental: "caseStmtMacros".}. The error I'm getting is that expression '"some"' is of type 'string' and has to be used (or discarded). Can anyone point me in the right direction? |
| 19:11:24 | FromDiscord | <Arathanis> In reply to @Hourglass, When the Hour Strikes "Hey y'all, I'm struggling": One thing I've noticed out of the gate is your catch for semicolon appear to be matching colon: https://media.discordapp.net/attachments/371759389889003532/1070058738129575996/image.png |
| 19:19:05 | FromDiscord | <crow> In reply to @Hourglass, When the Hour Strikes "Hey y'all, I'm struggling": you're not gonna get the text found a semicolon, since semicolon is in Symbols, so it will go down the if in 53 and skip over the elif at 73 |
| 19:19:39 | FromDiscord | <crow> In reply to @Hourglass, When the Hour Strikes "Hey y'all, I'm struggling": I generally ignore whitespace before tokens by the way |
| 19:19:42 | FromDiscord | <crow> when writing lexers |
| 19:20:20 | FromDiscord | <crow> it makes more sense to have a lex function that matches exactly one token, which I achieve by always skipping whitespace before trying to lex, and eventually creating an EOF token after the whitespace after the last token |
| 19:21:27 | FromDiscord | <crow> (edit) "that" => "where every step of the loop, it" |
| 19:28:35 | FromDiscord | <jmgomez> In reply to @ShalokShalom "would imagine, there is": not sure, I bet people just write then down if they need to. I can give you an example if you give me a few minutes. Im finishing something else |
| 19:28:47 | FromDiscord | <jmgomez> but it isnt really too much code, you will see |
| 19:31:35 | FromDiscord | <Phil> sent a code paste, see https://paste.rs/cnW |
| 19:31:47 | FromDiscord | <Hourglass [She/Her]> In reply to @Arathanis "One thing I've noticed": That was just a test, I forgot to remove it after the commit |
| 19:32:39 | FromDiscord | <Hourglass [She/Her]> It's the `case` block that should actually be catching the semicolon but unsure why it isn't when it's at the wnd of the file (which seems to be the only issue it has?) |
| 19:32:52 | FromDiscord | <Phil> Actually, my browser just corrected me, the string `"{3: 4}"` is valid json, I'm not sure how one would parse that tbh |
| 19:33:07 | FromDiscord | <Phil> (edit) "Actually, my browser just corrected me, the string `"{3: 4}"` is valid json, I'm not sure how one would parse that ... tbh" added "in general" |
| 19:35:23 | FromDiscord | <Phil> sent a code paste, see https://play.nim-lang.org/#ix=4mJ0 |
| 19:36:15 | FromDiscord | <Hourglass [She/Her]> In reply to @Isofruit "Actually, my browser just": In Python it'd just be `myDict[3]` |
| 19:36:18 | FromDiscord | <Hourglass [She/Her]> It's just key values |
| 19:36:21 | FromDiscord | <Phil> In reply to @alexb "Hello everyone. I'm absolute": Hmmm there's nothing fundamentally wrong with this code.↵I can run your copy-pasted example just fine, the route even responds correctly.↵Does the error immediately happen after running the binary? |
| 19:36:27 | FromDiscord | <Hourglass [She/Her]> It should be the same in JS |
| 19:36:48 | FromDiscord | <alexb> In reply to @Isofruit "Hmmm there's nothing fundamentally": Yes, the error happens immediately. I showed the full output |
| 19:36:56 | FromDiscord | <ShalokShalom> In reply to @jmgomez "but it isnt really": yeah, I am sure homoiconicity helps |
| 19:37:09 | FromDiscord | <huantian> what if you just have jsony parse to a JsonNode and then do the rest of the parsing yourself |
| 19:37:25 | FromDiscord | <ShalokShalom> i still think, it could help the ecosystem to provide such macros out of the box |
| 19:37:42 | FromDiscord | <ShalokShalom> people coming to the language will hardly start to write macros |
| 19:37:56 | FromDiscord | <Phil> In reply to @Ayano "yes": okay, make a `var reader2 = reader` assignment before that line and use that, does it still error out? Because now I'm curious |
| 19:38:39 | FromDiscord | <Phil> (edit) "that," => "that in `newPackage`," |
| 19:39:15 | FromDiscord | <huantian> then you don't have to do any manual string parsing yourself |
| 19:39:53 | FromDiscord | <Phil> In reply to @alexb "Yes, the error happens": Huh, you're using nim 1.9.1? ↵Try 1.6.10, does that resolve the issue?↵(If you have choosenim that should be `choosenim 1.6.10` |
| 19:40:43 | FromDiscord | <Phil> In reply to @huantian "what if you just": Was that to me? I'm going through the backlog of messages while I was dealing with the "mixed datatype example" |
| 19:41:22 | FromDiscord | <alexb> Let me try |
| 19:41:23 | FromDiscord | <alexb> sent a code paste, see https://play.nim-lang.org/#ix=4mJ2 |
| 19:41:26 | FromDiscord | <Arathanis> In reply to @Isofruit "If that's deemed unacceptable,": FWIW that kind of heinous structure will get me to look for an alternative service before trying to parse that nonsense lol |
| 19:41:33 | FromDiscord | <Arathanis> so i dont mind glossing over it a bit |
| 19:41:48 | FromDiscord | <Arathanis> if a service I was actually trying to use returned data that way i'd give it a 😬 |
| 19:42:07 | FromDiscord | <Phil> In reply to @Arathanis "if a service I": I mean it's basically telling you that it wants nothing to do with Java 😛 |
| 19:42:21 | FromDiscord | <Arathanis> In reply to @Isofruit "I mean it's basically": > my life is painful and I want company |
| 19:42:35 | FromDiscord | <Phil> (Which I'm bringing up because java is the most wide-spread backend language I'm aware of) |
| 19:42:49 | FromDiscord | <Arathanis> In reply to @Isofruit "(Which I'm bringing up": this server is supposed to be safe for work |
| 19:43:07 | FromDiscord | <Arathanis> you can't bring up cosmic horrors like that |
| 19:43:08 | FromDiscord | <Arathanis> 😉 |
| 19:43:09 | FromDiscord | <Phil> In reply to @Arathanis "this server is supposed": See, it is safe for work, java is type safe! |
| 19:44:01 | FromDiscord | <Phil> But yeah, I'm all for shitting on java, though tbh I find the community it cultivated and the mindset that lead to things like spring much more offensive than the language itself |
| 19:44:41 | FromDiscord | <Arathanis> In reply to @Isofruit "But yeah, I'm all": I got an ultrawide screen monitor so I can finally read java class names on a single line w/o scrolling |
| 19:45:05 | FromDiscord | <Phil> I don't know why but these jokes still make me laugh out loud xD |
| 19:45:34 | FromDiscord | <Array in ∀ Matrix> kek↵(@Arathanis) |
| 19:46:06 | FromDiscord | <Array in ∀ Matrix> java is verbose for what? |
| 19:46:14 | FromDiscord | <Array in ∀ Matrix> zig lang also is similar |
| 19:46:24 | FromDiscord | <Arathanis> im convinced java class names are descided on by throwing 5~10 numbered darts at a dart board covered in buzz words and then determining the name in dart order.↵↵AbstractClassFactoryAbstractionLayerFactory |
| 19:46:54 | FromDiscord | <Phil> sent a code paste, see https://play.nim-lang.org/#ix=4mJ5 |
| 19:47:12 | FromDiscord | <Ayano> In reply to @Isofruit "okay, make a `var": this doesnt work either https://media.discordapp.net/attachments/371759389889003532/1070067745560985671/image.png |
| 19:47:39 | FromDiscord | <Phil> now that is fascinating, are you sure both of those Reader types are the same and you don't have a name collision somewhere? |
| 19:48:12 | FromDiscord | <Phil> Like, is there possibly a second type called `Reader` somewhere and newPacket expects that instead of the type you're generating? |
| 19:48:24 | FromDiscord | <Phil> (edit) "generating?" => "providing?" |
| 19:49:22 | FromDiscord | <Arathanis> GOOD question |
| 19:49:23 | FromDiscord | <Ayano> In reply to @Isofruit "now **that** is fascinating,": https://media.discordapp.net/attachments/371759389889003532/1070068295086129192/image.png |
| 19:49:25 | FromDiscord | <Arathanis> excellent thing to check |
| 19:49:32 | FromDiscord | <Ayano> when i control click both in vs code |
| 19:49:33 | FromDiscord | <Phil> In reply to @sommerjan "Hi, I'm trying to": You know how in a lot of other languages you can have a function that returns 5, then you call that in your code like `get5()` and it doesn't matter if you make use of it or not? |
| 19:49:38 | FromDiscord | <Ayano> they get to the same class |
| 19:50:13 | FromDiscord | <Ayano> but i include it like this https://media.discordapp.net/attachments/371759389889003532/1070068504092491857/image.png |
| 19:50:15 | FromDiscord | <Ayano> could that be it? |
| 19:50:20 | FromDiscord | <Phil> INCLUDE |
| 19:50:25 | FromDiscord | <Phil> The evil has arisen! |
| 19:50:49 | FromDiscord | <Arathanis> i didnt even know you can do that, what does it do semantically? |
| 19:50:53 | FromDiscord | <Phil> Include is a way to copy-paste the code that is in those packages.↵Meaning if you included that file elsewhere as well, you now have 2 reader types |
| 19:51:01 | FromDiscord | <Arathanis> OH GOD |
| 19:51:14 | FromDiscord | <Phil> Like, include is basically "copy paste the contents of that file into my file" |
| 19:51:15 | FromDiscord | <Arathanis> its just like C/C++ include? |
| 19:51:52 | FromDiscord | <Phil> And that's, ladies and gentlemen, is why you should avoid includes unless you know what you're doing and are building your project structure around it or sth. |
| 19:52:05 | FromDiscord | <Phil> `import` is the way to go, 99.999% of the time |
| 19:52:15 | FromDiscord | <Arathanis> im struggling to tihnk of a reason to use include at all |
| 19:52:23 | sommerjan | @Phil yes, but I'd like the value to get assigned to my variable without having to create a procedure |
| 19:52:26 | FromDiscord | <Phil> Me as well, maybe c++ interop or sth |
| 19:52:47 | FromDiscord | <Arathanis> sent a code paste, see https://play.nim-lang.org/#ix=4mJ6 |
| 19:52:50 | FromDiscord | <Phil> In reply to @sommerjan "<@180601887916163073> yes, but I'd": Ah, shit, I forgot to finish my reply to you |
| 19:53:03 | FromDiscord | <Phil> Nim doesn't like calling procedures and then ignoring what they return, at all |
| 19:53:36 | FromDiscord | <Phil> sent a code paste, see https://play.nim-lang.org/#ix=4mJ7 |
| 19:53:55 | FromDiscord | <Arathanis> i love that feature actually, you must explicitly ignore a value, usually from a procedure |
| 19:53:57 | FromDiscord | <Phil> In line 2 I called get5, I told that proc to give me whatever it calculates so the compiler assumes for me that I wanted to do something with what it returns |
| 19:54:42 | FromDiscord | <Phil> sent a code paste, see https://paste.rs/AqU |
| 19:55:25 | FromDiscord | <Phil> In reply to @Array in ∀ Matrix "java is verbose for": java is verbose for everything ever |
| 19:55:37 | FromDiscord | <Ayano> https://media.discordapp.net/attachments/371759389889003532/1070069865211252976/image.png |
| 19:55:39 | FromDiscord | <Ayano> uhm |
| 19:55:44 | FromDiscord | <Ayano> (edit) "uhm" => "umm" |
| 19:55:49 | FromDiscord | <Phil> sent a code paste, see https://play.nim-lang.org/#ix=4mJ8 |
| 19:55:59 | sommerjan | @Phil The problem isn't that I want values that should be discarded. My example works if I don't use Optional and just initialize based on a boolean value. Something like let x = case bool_value of true: "works" of false: "all good" |
| 19:56:03 | FromDiscord | <Phil> In reply to @Ayano "": New error means progress! |
| 19:56:20 | FromDiscord | <Ayano> yes... |
| 19:56:34 | FromDiscord | <Phil> In reply to @sommerjan "<@180601887916163073> The problem isn't": Your example code sadly gets completely mangled to me, I'll need a bit to even parse that as it all gets pumped out as a single-line string to me |
| 19:57:35 | FromDiscord | <Ayano> In reply to @Ayano "": @Jonas, his code btw |
| 19:58:27 | FromDiscord | <Arathanis> @Hourglass [She/Her] do you have example input / bootstrapping code so I can execute your lexer/ |
| 19:58:29 | FromDiscord | <Arathanis> (edit) "lexer/" => "lexer?" |
| 19:58:35 | FromDiscord | <Arathanis> I'd like to see the error you are encountering |
| 19:59:38 | FromDiscord | <Hourglass [She/Her]> Yep I do up here ^^ also in the `tests` folder |
| 19:59:48 | FromDiscord | <Phil> In reply to @sommerjan "Hi, I'm trying to": Ohhh you're ... trying to do a case macro magic thing?↵Well, I assume that based on https://nim-lang.org/docs/manual_experimental.html#case-statement-macros↵That's elegantbeef teritory |
| 19:59:48 | FromDiscord | <Hourglass [She/Her]> Uh i forgot to reply |
| 19:59:51 | FromDiscord | <Hourglass [She/Her]> In reply to @Hourglass, When the Hour Strikes "Hey y'all, I'm struggling": Here: |
| 20:00:26 | FromDiscord | <Hourglass [She/Her]> Probably fixable by duplicating the last character in the source but that's just a hack |
| 20:00:52 | FromDiscord | <Phil> In reply to @Ayano "yes...": For the most part that means you have some funky async code somewhere.↵That one's a bit trickier.↵When I get these kinds of errors it's typically because I'm not using async/await/waitFor correctly |
| 20:01:27 | sommerjan | @Phil bingo |
| 20:01:52 | sommerjan | @Phil It's likely not ready for primetime yet |
| 20:02:34 | FromDiscord | <Phil> In reply to @Ayano "<@473566840132730889>, his code btw": Could you copy paste the full stacktrace?↵Generally reading the text properly is easier on the eyes + the full stacktrace (and seeing the lines around the code-sections that are blowing up) is pretty helpfull |
| 20:03:42 | FromDiscord | <Phil> In reply to @sommerjan "<@180601887916163073> It's likely not": To be fair, that stuff is in experimental ^^ |
| 20:04:58 | FromDiscord | <Hourglass [She/Her]> In reply to @Arathanis "<@909883978717204561> do you have": Also, to compile the code, you should do `nim c --run src/main.nim tests/test1.mlx -p` |
| 20:05:12 | FromDiscord | <Hourglass [She/Her]> The `-p` flag makes it print out the tokens |
| 20:05:29 | FromDiscord | <Arathanis> In reply to @Hourglass, When the Hour Strikes "Also, to compile the": i see what the issue is |
| 20:05:41 | FromDiscord | <Arathanis> sorry ive been parsing the code, i actually like lexing and parsing a lot |
| 20:05:46 | * | LuxuryMode quit (Quit: Connection closed for inactivity) |
| 20:05:48 | FromDiscord | <Hourglass [She/Her]> Nah dw xD |
| 20:06:06 | FromDiscord | <Arathanis> or maybe I should say |
| 20:06:09 | FromDiscord | <Arathanis> "i think I see the problem" |
| 20:06:13 | FromDiscord | <Arathanis> let me keep looking briefly |
| 20:06:13 | FromDiscord | <Arathanis> :P |
| 20:06:30 | FromDiscord | <Hourglass [She/Her]> Alright aha |
| 20:06:33 | FromDiscord | <alexb> sent a code paste, see https://play.nim-lang.org/#ix=4mJc |
| 20:06:52 | FromDiscord | <alexb> macos runs some shit on port 5000 on my macbook |
| 20:16:52 | FromDiscord | <Arathanis> @Hourglass [She/Her] the problem is two fold. The reason you are missing the final ":" is because your notAtEndOfLine check is using `len - 1` instead of `len` so it will always cause you to skip the final character. If you remove the `-1` you will be able to see the final character. If you do that now though, you will endlessly loop any Symbols because the Symbol path does not call `Lexer.increment` |
| 20:17:23 | FromDiscord | <Arathanis> (edit) "@Hourglass [She/Her] the problem is two fold. The reason you are missing the final ":" is because your notAtEndOfLine check is using `len - 1` instead of `len` so it will always cause you to skip the final character. If you remove the `-1` you will be able to see the final character. If you do that now though, you will endlessly loop any Symbols because the Symbol path does not call `Lexer.increment` ... " added "so it never |
| 20:17:25 | FromDiscord | <Phil> In reply to @alexb "Configured choosenim and finally": Less for you specifically, just need a link to your message there for @ringabout ↵Is that known behaviour?↵Should an issue be opened? |
| 20:17:44 | FromDiscord | <Phil> (edit) "↵Is" => "↵For ringabout:↵Is" |
| 20:17:55 | FromDiscord | <Arathanis> (edit) "notAtEndOfLine check" => "`notAtEndOfLine` procedure" |
| 20:18:14 | FromDiscord | <Hourglass [She/Her]> In reply to @Arathanis "<@909883978717204561> the problem is": Aaaah, that explains it, thank you! ^^ |
| 20:18:20 | FromDiscord | <Arathanis> You are very welcome. |
| 20:18:22 | FromDiscord | <Phil> sent a code paste, see https://play.nim-lang.org/#ix=4mJh |
| 20:18:33 | FromDiscord | <Phil> (edit) "https://play.nim-lang.org/#ix=4mJh" => "https://play.nim-lang.org/#ix=4mJi" |
| 20:22:01 | FromDiscord | <Hourglass [She/Her]> In reply to @Arathanis "<@909883978717204561> the problem is": God this was such a simple fix i didn't notice xD |
| 20:22:28 | FromDiscord | <Arathanis> its one of the easiest issues to look passed. off by one errors |
| 20:31:48 | FromDiscord | <jmgomez> sent a code paste, see https://play.nim-lang.org/#ix=4mJm |
| 20:32:26 | FromDiscord | <jmgomez> (edit) "https://play.nim-lang.org/#ix=4mJm" => "https://play.nim-lang.org/#ix=4mJp" |
| 20:33:10 | FromDiscord | <jmgomez> (edit) "https://play.nim-lang.org/#ix=4mJp" => "https://play.nim-lang.org/#ix=4mJq" |
| 20:36:14 | FromDiscord | <Hourglass [She/Her]> In reply to @Arathanis "its one of the": Yepp |
| 20:53:57 | FromDiscord | <ShalokShalom> sent a code paste, see https://play.nim-lang.org/#ix=4mJy |
| 20:54:02 | FromDiscord | <ShalokShalom> if it really works this way |
| 20:54:19 | FromDiscord | <ShalokShalom> do you have experience with tp? |
| 20:56:30 | FromDiscord | <jmgomez> https://media.discordapp.net/attachments/371759389889003532/1070085184696893530/Screenshot_2023-01-31_at_20.55.11.png |
| 20:57:42 | FromDiscord | <jmgomez> Nim suggest doesnt suggest though, it may be have a bug somewhere because that should definitely work. I have way more complicated ASTs with suggestions |
| 21:01:31 | FromDiscord | <deech> I did something like this for a talk about type introspection I gave a few years ago. https://github.com/deech/LambdaWorldCadiz2019-WhatFPCanLearnFromStaticIntrospection/blob/master/json_reflection.nim |
| 21:01:38 | FromDiscord | <jmgomez> In reply to @ShalokShalom "do you have experience": I used them in the past and did one, like 10+ years ago |
| 21:03:55 | FromDiscord | <jmgomez> In reply to @deech "I did something like": I think I told you but I was present there on that talk and I was the guy that said that yes, F# type providers are similar. First time I heard about Nim was there. Last year I was looking for a low level lang to do the UE plugin and after trying Rust I remembered Nim, rewatched the talk and started to play with it |
| 21:07:41 | FromDiscord | <deech> Oh haha, I totally forgot. Sorry. |
| 21:12:00 | FromDiscord | <Nerve> In reply to @deech "Oh haha, I totally": Oh, look who it is! Do you have the code lying around for the macro that diffs types from your "Nim Nuggets" presentation? |
| 21:15:24 | FromDiscord | <deech> In reply to @Nerve "Oh, look who it": This is the demo code (https://github.com/deech/NimNuggets/blob/master/backup/migration.nim) and this is the implementation (https://github.com/deech/NimNuggets/blob/master/backup/migrationmacros.nim). |
| 21:15:43 | FromDiscord | <Nerve> Thanks much |
| 21:25:06 | * | LuxuryMode joined #nim |
| 21:26:44 | FromDiscord | <Nerve> Does anyone have experience cross-compiling with Nimja? |
| 21:27:05 | FromDiscord | <Nerve> sent a code paste, see https://play.nim-lang.org/#ix=4mJK |
| 21:27:54 | FromDiscord | <Nerve> Trying to instantiate templates for a Windows build using mingw leads to this error |
| 21:28:56 | FromDiscord | <Nerve> My template call, which works on Linux, is↵`compile_template_file(get_script_dir() / "company_project" / "templates" / "a_template.nimja")` |
| 21:30:20 | FromDiscord | <Nerve> I'd imagine it has something to do with backslashes and the assumed mingw drive root |
| 21:42:54 | FromDiscord | <ShalokShalom> In reply to @jmgomez "Nim suggest doesnt suggest": well, we know nimsuggest |
| 21:44:45 | FromDiscord | <ShalokShalom> In reply to @Nerve "Does anyone have experience": dont try to hijack your request, did you know you can use Zig to cross compile? |
| 21:45:04 | FromDiscord | <ShalokShalom> it seems to be the most successful way |
| 21:45:20 | FromDiscord | <ShalokShalom> https://github.com/enthus1ast/zigcc |
| 22:14:15 | FromDiscord | <T0lk1en> yoo |
| 22:14:29 | FromDiscord | <T0lk1en> does anyone wanna help me out rq. |
| 22:14:43 | FromDiscord | <T0lk1en> need to see if my server is working. |
| 22:16:10 | FromDiscord | <ShalokShalom> how? |
| 22:17:41 | FromDiscord | <T0lk1en> ncat into it |
| 22:17:47 | FromDiscord | <T0lk1en> ill share ip and port |
| 22:18:18 | FromDiscord | <Hourglass [She/Her]> In reply to @T0lk1en "ncat into it": Sure |
| 22:18:22 | FromDiscord | <T0lk1en> ncat 192.168.1.76 443 |
| 22:18:32 | FromDiscord | <Arathanis> that definitely isnt going to work |
| 22:18:42 | FromDiscord | <T0lk1en> how so? |
| 22:18:45 | FromDiscord | <Arathanis> that is a local IP |
| 22:18:47 | FromDiscord | <Arathanis> on your LAN |
| 22:18:50 | FromDiscord | <T0lk1en> ah |
| 22:19:04 | FromDiscord | <Arathanis> only machines on your network will see it |
| 22:19:11 | FromDiscord | <Arathanis> we would need your external IP |
| 22:19:14 | FromDiscord | <Hourglass [She/Her]> Yeah lmao |
| 22:19:20 | FromDiscord | <Hourglass [She/Her]> And you'd need to port forward it |
| 22:19:20 | FromDiscord | <Arathanis> and you need to forward port 443 to that machine in your router |
| 22:19:32 | FromDiscord | <T0lk1en> and then ittl work? |
| 22:19:41 | FromDiscord | <Arathanis> that gets us to "maybe" |
| 22:19:46 | FromDiscord | <T0lk1en> ty |
| 22:19:48 | FromDiscord | <Arathanis> yup |
| 22:19:52 | FromDiscord | <Arathanis> good luck, my friend |
| 22:19:58 | FromDiscord | <Arathanis> networking is... something else |
| 22:20:14 | FromDiscord | <Hourglass [She/Her]> Doesn't IPv6 solve this stuff? Or? |
| 22:20:22 | FromDiscord | <T0lk1en> gotta learn it. Thats the whole point of this project |
| 22:20:51 | FromDiscord | <Arathanis> In reply to @Hourglass, When the Hour Strikes "Doesn't IPv6 solve this": pretty sure it still does local IPs |
| 22:21:13 | FromDiscord | <Hourglass [She/Her]> I thought external IPv6 addresses didn't require port forwarding |
| 22:21:14 | FromDiscord | <Arathanis> I think its because its a little easier to configure things with some network abstractions |
| 22:21:43 | FromDiscord | <Arathanis> oh i tihnk that is true |
| 22:22:24 | FromDiscord | <Arathanis> cause a device can basically have as many ipv6 addresses as it wishes |
| 22:22:36 | FromDiscord | <Arathanis> and you can use that in lieu of ports |
| 22:25:26 | FromDiscord | <T0lk1en> are there any default forworded ports? |
| 22:25:43 | FromDiscord | <Arathanis> depends on your router |
| 22:25:48 | FromDiscord | <Arathanis> (edit) |
| 22:25:52 | FromDiscord | <Arathanis> but you best hope the answer to that is "no" |
| 22:26:16 | FromDiscord | <Arathanis> you do not want your computer exposed to the internet "by default" |
| 22:26:23 | FromDiscord | <T0lk1en> 🥲 |
| 22:26:34 | FromDiscord | <T0lk1en> gotcha |
| 22:26:54 | FromDiscord | <Arathanis> you can ping your own server using `localhost:443` or `127.0.0.1:443` or w/ your local ip you provided earlier |
| 22:27:02 | FromDiscord | <Arathanis> if it works for you, it would work for us with the proper external network config |
| 22:27:32 | FromDiscord | <T0lk1en> yeah ive been doing that |
| 22:27:45 | FromDiscord | <T0lk1en> what about free online hosting? any places like that> |
| 22:27:47 | FromDiscord | <T0lk1en> ? |
| 22:27:51 | FromDiscord | <crow> In reply to @Hourglass, When the Hour Strikes "I thought external IPv6": I think it depends on the contract with your ISP |
| 22:28:00 | FromDiscord | <crow> and firewall settings on your router too |
| 22:28:22 | FromDiscord | <crow> actually, im pretty sure port forwarding is only about firewall settings of the router |
| 22:28:23 | FromDiscord | <Arathanis> In reply to @T0lk1en "what about free online": im sure some still exist but my old go-to answer for this recently discontinued free hosting cause it was being abused |
| 22:28:35 | FromDiscord | <Arathanis> you would have to look around |
| 22:28:41 | FromDiscord | <T0lk1en> gotcha |
| 22:28:52 | FromDiscord | <Arathanis> cant get into the router, huh? |
| 22:28:53 | FromDiscord | <crow> (edit) "actually, im pretty sure port forwarding is only about firewall settings of the router ... " added "mainly (and of course which computer on LAN to forward to)" |
| 22:29:37 | FromDiscord | <T0lk1en> family router. Dont know password and owner is out of town |
| 22:29:42 | FromDiscord | <Arathanis> ahh |
| 22:29:53 | FromDiscord | <Arathanis> you could have them do it when they get back |
| 22:29:57 | FromDiscord | <Arathanis> otherwise you can look for hosting options. |
| 22:30:02 | FromDiscord | <T0lk1en> ofc but im impatient lol |
| 22:31:14 | FromDiscord | <crow> there might be "tunnels" you can use to connect a local address with a (sub)domain |
| 22:31:33 | FromDiscord | <crow> there are services that offer that, even some freemium ones possibly |
| 22:31:45 | FromDiscord | <T0lk1en> gotcha |
| 22:31:58 | FromDiscord | <crow> I won't advertise any of them though |
| 22:32:09 | FromDiscord | <T0lk1en> what about suggest one? |
| 22:32:18 | FromDiscord | <crow> I don't know of a non shady one actually |
| 22:32:19 | FromDiscord | <crow> sorry |
| 22:32:32 | * | dnh quit (Ping timeout: 248 seconds) |
| 22:32:36 | FromDiscord | <crow> but I haven't really looked |
| 22:32:57 | FromDiscord | <T0lk1en> goctha |
| 22:33:50 | FromDiscord | <crow> alternatively technically tor hidden services are the same things, but then at an onion address |
| 22:34:09 | FromDiscord | <crow> but it's probably more trouble than good i guess |
| 22:34:43 | FromDiscord | <T0lk1en> ive used tor before. Even spun up vanity addresses just never hosted |
| 22:34:57 | FromDiscord | <crow> idk if hosting is different traffic than just using it |
| 22:35:08 | FromDiscord | <crow> I would assume not |
| 22:35:22 | FromDiscord | <T0lk1en> maybe eepsite |
| 22:35:38 | FromDiscord | <crow> but if hosting is identifiable by ISPs, might not be a good idea |
| 22:36:42 | FromDiscord | <T0lk1en> dont plan too host forever. Just as a proof of concept |
| 22:39:10 | FromDiscord | <crow> it's still more trouble than the proxies i mentioned |
| 22:39:22 | FromDiscord | <T0lk1en> gotcha |
| 22:39:55 | FromDiscord | <T0lk1en> ill just wait to port forword |
| 22:53:08 | FromDiscord | <Nilts> how do i slice a cstring like `[1..^1]` |
| 22:57:03 | FromDiscord | <Elegantbeef> You dont |
| 22:57:57 | FromDiscord | <Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4mK3 |
| 22:59:27 | FromDiscord | <Nilts> In reply to @Elegantbeef "If you know the": i mean, couldn't you just use `len`? |
| 22:59:40 | FromDiscord | <Elegantbeef> `len` on a cstring is got iteratively |
| 22:59:50 | * | advesperacit quit () |
| 23:00:22 | FromDiscord | <Elegantbeef> So... you could, but in that case you might aswell just copy to the string |
| 23:00:26 | Amun-Ra | .len works unless there's '\x00' somewhere in the middle |
| 23:00:49 | FromDiscord | <Nilts> In reply to @Elegantbeef "So... you could, but": so is there a better way to slice it? |
| 23:01:01 | Amun-Ra | Nilts: ($s)[1..^1] |
| 23:01:18 | FromDiscord | <Elegantbeef> Only better if you dont care about performance |
| 23:01:23 | Amun-Ra | mhm |
| 23:01:27 | FromDiscord | <Elegantbeef> That allocates multiple strings |
| 23:01:45 | FromDiscord | <Nilts> In reply to @Elegantbeef "Only better if you": Well, since i'm using the js backend, which one would assume to be slower, i do care |
| 23:02:14 | FromDiscord | <Elegantbeef> Wait this is JS, if so len is free |
| 23:02:24 | FromDiscord | <Elegantbeef> so then `myCstring[i..^1]` is cheap as |
| 23:02:32 | Amun-Ra | writing JS backend is real fun; check this: var s = newstring 5; echo s.repr |
| 23:03:23 | Amun-Ra | hmm |
| 23:04:06 | FromDiscord | <Nilts> In reply to @Elegantbeef "so then `myCstring[i..^1]` is": how do i implement this, or do i use native imports |
| 23:04:16 | Amun-Ra | nevermind, I had a weird problem with that yesterday; I have to find what was the real problem |
| 23:04:43 | Amun-Ra | ah, right, I forgot to set js backend ;) |
| 23:05:20 | Amun-Ra | it fails to run on nodejs |
| 23:05:20 | FromDiscord | <Elegantbeef> The best thing to do in JS land is what amun gave i guess |
| 23:05:27 | FromDiscord | <Elegantbeef> Since JS strings are not utf8 encoded |
| 23:05:56 | Amun-Ra | cstrings on JS are kind of easier to work on than strings |
| 23:06:49 | FromDiscord | <Elegantbeef> Unlike C they're not encoded the same as Nim strings so in many cases you want to avoid converting to string if you can |
| 23:08:35 | FromDiscord | <Elegantbeef> C strings can atleast be converted using `toOpenArray` if you know the length and `cstring myString` is free |
| 23:09:34 | Amun-Ra | and JS target does not have problem with null terminators inside cstring and getting the length |
| 23:10:07 | FromDiscord | <Elegantbeef> Yep the joy of pascal strings |
| 23:24:11 | * | oprypin quit (Quit: Bye) |
| 23:24:21 | * | oprypin joined #nim |
| 23:29:04 | FromDiscord | <Nilts> sent a code paste, see https://play.nim-lang.org/#ix=4mKc |
| 23:30:11 | FromDiscord | <Elegantbeef> Your list is empty |
| 23:31:39 | FromDiscord | <Nilts> In reply to @Elegantbeef "Your list is empty": but, i made a if stmt ensuring that it was not. |
| 23:32:53 | FromDiscord | <Nilts> nvm, i just needed to re-arrange the checks |
| 23:33:29 | FromDiscord | <Elegantbeef> Hey you can argue with me all day long, the code says you're wrong 😛 |
| 23:34:59 | FromDiscord | <Nilts> In reply to @Elegantbeef "Hey you can argue": nice rhyme |
| 23:53:04 | * | jmdaemon quit (Ping timeout: 248 seconds) |
