<< 26-06-2026 >>
00:41:54*xet7 joined #nim
00:43:06*xet7 quit (Remote host closed the connection)
00:43:30*xet7 joined #nim
00:57:00*xet7 quit (Remote host closed the connection)
01:03:56*disso-peach quit (Remote host closed the connection)
01:18:33*britney joined #nim
01:24:42*rockcavera quit (Remote host closed the connection)
02:31:44FromDiscord<Laylie> include is evil
02:31:56FromDiscord<Laylie> try delayed imports first
02:34:00*hygo quit (Read error: Connection reset by peer)
02:40:03*hygo joined #nim
03:08:43FromDiscord<[Next-System] systemblue> hey please help me
03:13:23FromDiscord<[Next-System] systemblue> I want to use↵zeroMem/copyMem in Nim-JS↵↵please help me
03:13:58FromDiscord<[Next-System] systemblue> sent a code paste, see https://play.nim-lang.org/#pasty=oSKqwApc
03:14:03FromDiscord<[Next-System] systemblue> and inline JS is very hard
03:50:48FromDiscord<__jont__> I love Nim have been using Nim for a while. But honestly, I’m a bit concerned on how deeply the core developers of the language are becoming reliant on LLMs for the next iteration of the language, and how ready they are to squash dissent when concerns are brought up. ↵↵Sucks to see, I hope Nimony is good when it's complete though. And I hope this comment doesn't get me banned
03:51:19*cm quit (Quit: Bye.)
03:52:32FromDiscord<__jont__> (edit) "I love Nim ... haveit" added "and" | "Nim" => "it"
03:57:06FromDiscord<Laylie> not sure about the squashing dissent but yes it is quite unfortunate
03:58:06*cm joined #nim
04:01:34FromDiscord<gesee37> sent a code paste, see https://play.nim-lang.org/#pasty=zFLdtwNh
04:09:33FromDiscord<__jont__> In reply to @Laylie "not sure about the": https://discord.com/channels/371759389889003530/1515616139927883948/1519674492777926727↵↵This is what made me think of "squashing dissent." In fairness the person was being a little abrasive, but I think the action taken was excessive
04:11:33FromDiscord<__jont__> I don't think they were inherently breaking any rules for disagreeing. Unless the rules state that we need to fall in line on our opinion of AI
04:14:43FromDiscord<Laylie> wow...
04:15:59FromDiscord<gesee37> In reply to @__jont__ "https://discord.com/channels/371759389889003530/151": Yeah, moderators have admitted it was excessive, a warning would have suffice
04:18:03FromDiscord<gesee37> In reply to @__jont__ "I don't think they": The pb was more that it was being a bit (if I can say that) agggressive on people using it.↵The same way he has the right not to like it, the same way people have the right to like it.
04:18:20FromDiscord<gesee37> (edit) "agggressive" => "aggressive"
04:23:14FromDiscord<__jont__> I get that this was recognized publicly, but the user is no longer in the server and it sends a message that conflicting opinions on the role of LLMs in architecting language infrastructure will not be tolerated. Which is dangerous when considering the inherent risk one accepts when having such a hands-off approach to "development" (if we still allow ourselves to call it that)
04:28:35FromDiscord<gesee37> In reply to @__jont__ "I get that this": Yeah and they already recognized it so we can only hope for them to do better
04:30:10FromDiscord<__jont__> Yeah, hope so
04:45:30*xet7 joined #nim
04:59:33*Goodbye_Vincent1 quit (Quit: )
05:01:09*Goodbye_Vincent1 joined #nim
05:15:33FromDiscord<im_rob> In reply to @__jont__ "I love Nim and": wait please share with ignorant me the source of your fears with LLM reliance
05:16:20FromDiscord<im_rob> Though I do lurk the server and the forum, I'm out of the loop from Nimony development
05:16:41FromDiscord<im_rob> (edit) "development" => "development."
05:20:46FromDiscord<__jont__> In reply to @im_rob "wait please share with": https://addyosmani.com/blog/cognitive-surrender/↵↵This is a pretty good post. This explains "comprehension debt", which metastasizes into technical debt when you're not just the users of a new tech but the architects of it
05:29:42FromDiscord<__jont__> As for Nimony, the use of AI on its development is very public ↵↵<https://forum.nim-lang.org/t/13948>
05:40:55FromDiscord<gesee37> In reply to @__jont__ "As for Nimony, the": There is no problem with the fact AI is used for this.↵Nimony is a project Araq planned for years and designed it way before that.↵I don't see why using it to execute it is a problem
05:42:04FromDiscord<gesee37> Araq already proven that he knows what he does, after all, he made nim in the first place
05:42:40FromDiscord<gesee37> (edit) "Araq ... already" added "has"
05:44:48FromDiscord<__jont__> I'm fairly biased but honestly I don't have a huge issue with rapid prototyping, assuming that everything will probably be handwritten over several times before release. I don't have an issue with Araq doing this of course, he's a very skilled developer and has made many many contributions to the Nim ecosystem. It's more so about the social implications of encouraging people who don't have the same expertise to follow suit I guess
05:45:04FromDiscord<__jont__> (edit) "prototyping," => "prototyping using AI,"
05:46:29FromDiscord<__jont__> (edit) "ecosystem." => "ecosystem as its creator."
05:56:00FromDiscord<gesee37> oh I see
05:56:40FromDiscord<tempestro> Why should Araq be responsible for how others use a tool
05:57:29FromDiscord<tempestro> I think all C++ veterans should feel ashamed for encouraging other, less skilled people to use C++
05:58:05FromDiscord<__jont__> Lol, that's a fair point
06:00:05FromDiscord<nnsee> In reply to @gesee37 "Araq has already proven": from seeing the people around me, people who know their skillset is above their peers' (and thus think they are not affected) are usually the ones that are hit hardest with the debilitating effects of heavy LLM usage
06:00:05FromDiscord<__jont__> I guess I just don't like how it was handled. But again, I understand that Araq recognized it probably should've gone another way. It being over AI use just touched a nerve with me because of my personal bias against it
06:00:07FromDiscord<amun0666> In reply to @gesee37 "You should avoid doing": zeromem and copymem are undefined for js backend (those `when`s need to be toplevel). I don't see anything wrong with the code on the first glance. I need to write something like that for my project.
06:00:48FromDiscord<im_rob> In reply to @tempestro "I think all C++": no the student must engage in the same torture I once endured. alone, afraid, and nothing making sense
06:01:03FromDiscord<tempestro> oooh, edgy
06:01:17FromDiscord<im_rob> In reply to @__jont__ "As for Nimony, the": in any case, appreciate the links
06:01:23FromDiscord<gesee37> In reply to @amun0666 "zeromem and copymem are": not the code, the signature
06:01:30FromDiscord<gesee37> of the functions
06:01:46FromDiscord<amun0666> That signature doesn't exist in JS target.
06:02:31FromDiscord<gesee37> In reply to @amun0666 "That signature doesn't exist": The when(js) is in the function there
06:02:57FromDiscord<amun0666> Yes. It needs to be toplevel. `when`: `func`. Not the other way around.
06:04:02FromDiscord<amun0666> I'd ass when declared too, just in case it appears in the next nim release.
06:04:16FromDiscord<amun0666> (edit) "ass" => "add"
06:04:34FromDiscord<amun0666> (edit) "when declared" => "`when declared`"
06:10:05FromDiscord<gesee37> sent a long message, see https://pasty.ee/QoJcPTCX
06:10:58FromDiscord<gesee37> In reply to @amun0666 "zeromem and copymem are": But why would you need this ? Isn't JS already sandboxed ? (if you are also in security stuffs)
06:13:00FromDiscord<amun0666> That depends on what you want to do. I need to clear the buffer from time to time in a `proc`. Btw. `{.noinit.}` in JS is a no-op as all arrays are initialized to zero.
06:14:55FromDiscord<gesee37> In reply to @amun0666 "That depends on what": That op would probably not be called zeroMem for JS↵As you can't really zeroMem something in JS
06:15:27FromDiscord<amun0666> sent a long message, see https://pasty.ee/cpFOlDwQ
06:16:10FromDiscord<amun0666> (edit) "https://pasty.ee/NEJfZFtk" => "https://pasty.ee/QdMxkgQP"
06:16:35FromDiscord<amun0666> Hmm, I don't know how to force discord to display nim code properly.
06:16:55FromDiscord<nnsee> use three backticks with the nim identifier
06:17:04FromDiscord<nnsee> \`\`\`nim↵code here↵\`\`\`
06:17:08FromDiscord<amun0666> (edit) "long message," => "code paste," | "https://pasty.ee/DBFeVFCt" => "https://play.nim-lang.org/#pasty=hSKilsbq"
06:17:13FromDiscord<amun0666> I see. Thank you.
06:19:17FromDiscord<gesee37> sent a code paste, see https://play.nim-lang.org/#pasty=uTEwbVoT
06:21:42FromDiscord<amun0666> This is my global booldefine that's true when the target I build for doesn't offer `copymem`.
07:27:43FromDiscord<amun0666> sent a code paste, see https://play.nim-lang.org/#pasty=OJHCxktU
07:28:17FromDiscord<amun0666> This is kind of ugly, because it relies on internal array start index naming implementation, but works.
07:28:59FromDiscord<amun0666> A pointer in JS target is really passed as a pair of arguments: object itself and start index.
07:30:39FromDiscord<amun0666> The implementation of the `moveMem` would look similar.
07:36:22FromDiscord<amun0666> sent a code paste, see https://play.nim-lang.org/#pasty=QPzeXzgw
07:36:48FromDiscord<amun0666> sent a code paste, see https://play.nim-lang.org/#pasty=HlRVFRbP
07:37:17FromDiscord<amun0666> I tested `copymem` and `zeromem`, `movemem`- not.
07:39:03FromDiscord<gesee37> sent a code paste, see https://play.nim-lang.org/#pasty=hfHSIdfc
07:40:32FromDiscord<gesee37> (edit) "https://play.nim-lang.org/#pasty=vVtNBCRu" => "https://play.nim-lang.org/#pasty=SlFvuxUD"
07:42:49FromDiscord<amun0666> That's true, it'll only work with array-like types, but you shouldn't use those functions on other types anyway.
07:43:52FromDiscord<amun0666> For example, zeromem an object. That what reset is for.
07:51:48FromDiscord<gesee37> In reply to @amun0666 "That's true, it'll only": In C i can totally use those on other types↵You can copyMem whethever you like but not for JS↵That's why you can't add the `Mem`↵It's copyArray but not copyMem
07:53:25FromDiscord<amun0666> Those are low level functions and one needs to know when and when not to use those on. As for the naming, that's what they are named in the nim stdlib and I just follow the standard. No need for the extra when is a plus.
07:53:49FromDiscord<amun0666> There are no pointer types in js either.
07:53:54FromDiscord<amun0666> (edit) "either." => "neither."
07:56:54FromDiscord<gesee37> sent a long message, see https://pasty.ee/jIhKLuRZ
07:58:20FromDiscord<amun0666> My point is one shouldn't use low level zeromem on any native nim types.
07:58:42FromDiscord<amun0666> Unless one accepts the possibility of being shot on the foot.
07:59:01FromDiscord<amun0666> (edit) "on" => "oneself in"
08:00:29FromDiscord<gesee37> In reply to @amun0666 "Those are low level": Yeah the pb is that those functions can't exist for JS in the same sense↵CopyMem in C, C++, Rust means the same things, copying memory↵But in JS if it just means "copying arrays) they it's not copyMem, it's a subset of copyMem and thus can't be called copyMem because it can't do everything copyMem does.↵But sequtils probably has the functions you are trying to make
08:01:39FromDiscord<amun0666> I don't think there is something like that in sequtils.
08:02:49FromDiscord<amun0666> I mean something that applies to my use case.
08:05:38FromDiscord<amun0666> Hmm, that `zeromem` function I pasted above is not correct, it has to use raw index, too.
08:06:34FromDiscord<gesee37> In reply to @amun0666 "Hmm, that `zeromem` function": Just do an iteration ?↵Is there a need for inline JS ?
08:07:47FromDiscord<amun0666> Iteration is fine, copy/move/zeromem are better for hot paths.
08:09:20*tokyovigilante quit (Ping timeout: 245 seconds)
08:10:15FromDiscord<gesee37> In reply to @amun0666 "Iteration is fine, copy/move/zeromem": When It's JS target just do iteration ↵It's not like It's even the worst thing you could do
08:11:17FromDiscord<amun0666> That's what I have now; additional `when`s for JS target. I'll add those functions above to my code and remove all unnecessary `when` for JS target.
08:20:06*tokyovigilante joined #nim
08:50:03*Skippy8 joined #nim
08:54:15*Skippy8 quit (Ping timeout: 246 seconds)
09:03:05*Skippy8 joined #nim
09:05:10*Skippy8 quit (Read error: Connection reset by peer)
09:06:16*Skippy8 joined #nim
09:20:26*beholders_eye joined #nim
09:23:55FromDiscord<amun0666> There's an obvious error in functions I pasted above for `mem` functions take size argument and not the length.
09:49:54*kick455 quit (Ping timeout: 245 seconds)
09:50:09*oculux joined #nim
09:55:34*kick455 joined #nim
09:58:39*oculux quit (Ping timeout: 246 seconds)
10:10:34*Vyrnexis quit (Ping timeout: 256 seconds)
10:11:12*Vyrnexis joined #nim
11:05:59*beholders_eye quit (Ping timeout: 260 seconds)
11:20:16*Vyrnexis quit (Quit: The Lounge - https://thelounge.chat)
11:20:41*Vyrnexis joined #nim
11:34:54FromDiscord<0ffh> In reply to @__jont__ "I don't think they": He was being an ass, and intentionally offensive, and he wasn't banned just kicked.
11:51:04FromDiscord<0ffh> And he definitely broke the rules.↵The only thing that arguably went wrong is he didn't get an official warning first.
12:13:18FromDiscord<m091834> So with IC in devel branch it does work in nim now?
12:26:06FromDiscord<tauruuuuuus> compiling it right now to see owo
12:26:12FromDiscord<tauruuuuuus> (edit) "compiling it right now to see owo ... " added "if it does"
12:26:23FromDiscord<tauruuuuuus> (edit) removed "owo" | "if it does ... " added "owo"
12:54:40*beholders_eye joined #nim
14:10:03*Skippy8 quit (Ping timeout: 252 seconds)
14:34:00*Skippy8 joined #nim
15:57:32FromDiscord<tvoyglaza> Nimmmm thanks for existing!!! ₍₍⚞(⸝⸝>⸝⸝<⸝⸝)⚟⁾⁾
15:57:52FromDiscord<tvoyglaza> Bro, Nim is so cool!
16:02:40Amun-Ra
16:04:08*deavmi quit (Quit: No Ping reply in 180 seconds.)
16:08:14*britney left #nim (WeeChat 4.9.2)
16:09:09*deavmi joined #nim
16:09:18FromDiscord<gesee37> In reply to @tvoyglaza "Nimmmm thanks for existing!!!": I would genuinely like to know what made you so excited 🙂↵Care to share ?
16:11:29*Skippy8 quit (Read error: Connection reset by peer)
16:36:25*Skippy8 joined #nim
16:44:38FromDiscord<tvoyglaza> In reply to @gesee37 "I would genuinely like": Yeah!↵Well, I liked so much the ease of doing metaprogramming on this language, and it's easy to learn like Python, and also can run natively in most machines. And I liked that the documentation is really well done.
16:45:26FromDiscord<tvoyglaza> Bro, this language is underrated
16:51:10FromDiscord<gesee37> In reply to @tvoyglaza "Yeah! Well, I liked": Same here even though python have a cold place in a dark little corner in my heart, I find Nim fun while powerful to use
16:51:19FromDiscord<gesee37> In reply to @tvoyglaza "Bro, this language is": Litterally
16:53:05*Skippy8 quit (Ping timeout: 245 seconds)
16:53:17*Skippy8 joined #nim
17:12:18*Skippy8 quit (Ping timeout: 246 seconds)
17:12:54FromDiscord<jubalrahl> I am still fairly new to Nim and was playing around with NiGui. To avoid Windows compile issues, I am compiling with Zigcc. The question I have is, how can I compile it and not have the exe open a command line at the same time? I tried adding --app:gui to the command line arguments that I am using to compile the .nim file. What am I doing wrong? Do I need to have a .cfg file and put -app:gui there? Is the fact that I am using zigcc to c
17:21:45FromDiscord<jubalrahl> oh, my command line string is: nim c --app:gui --cc:clang --clang.exe="zigcc.cmd" --clang.linkerexe="zigcc.cmd" --forceBuild:on --opt:speed --run NiCalc.nim
17:51:21*Skippy8 joined #nim
17:52:27FromDiscord<DetermiedNim1> You might have to see if zigcc has settings for whether to open a window or not
17:59:35*Skippy8 quit (Ping timeout: 244 seconds)
18:09:18*beholders_eye quit (Ping timeout: 252 seconds)