| 00:28:38 | * | jmd_ quit (Ping timeout: 246 seconds) |
| 00:50:15 | * | jmd_ joined #nim |
| 00:52:36 | FromDiscord | <uninnocent> How do I communicate with sockets in python and net in nimlang? |
| 00:52:54 | FromDiscord | <uninnocent> When I just try to connect the two, it doesn't do anything. The message is blank. |
| 00:58:41 | FromDiscord | <demotomohiro> If you use TCP protocol, one socket have to listen, and another socket connect to it. |
| 01:00:41 | FromDiscord | <uninnocent> I know |
| 01:05:00 | * | nyeaa492842301 joined #nim |
| 01:05:21 | FromDiscord | <uninnocent> sent a code paste, see https://play.nim-lang.org/#ix=4Aop |
| 01:05:46 | FromDiscord | <uninnocent> sent a code paste, see https://play.nim-lang.org/#ix=4Aoq |
| 01:05:52 | FromDiscord | <heysokam> In reply to @uninnocent "How do I communicate": do you -need- raw sockets?↵there is `treeform/netty` which is really good, so you might find it useful |
| 01:05:57 | FromDiscord | <uninnocent> (edit) "https://play.nim-lang.org/#ix=4Aoq" => "https://play.nim-lang.org/#ix=4Aor" |
| 01:06:10 | FromDiscord | <uninnocent> In reply to @heysokam "do you -need- raw": Not sure what that is |
| 01:06:18 | FromDiscord | <uninnocent> I just want fast communication between devices |
| 01:06:36 | FromDiscord | <heysokam> https://github.com/treeform/netty/ |
| 01:06:50 | FromDiscord | <uninnocent> In reply to @heysokam "https://github.com/treeform/netty/": will this work with python |
| 01:06:51 | FromDiscord | <heysokam> (edit) "https://github.com/treeform/netty/" => "Netty - reliable UDP connection library for games in Nim.↵https://github.com/treeform/netty/" |
| 01:07:01 | FromDiscord | <uninnocent> like if I set up that server |
| 01:07:02 | FromDiscord | <uninnocent> I sent |
| 01:07:09 | FromDiscord | <uninnocent> and use that library for my nim client |
| 01:07:12 | FromDiscord | <heysokam> In reply to @uninnocent "will this work with": it will work with anything you can communicate from with nim |
| 01:07:28 | FromDiscord | <heysokam> (edit) "In reply to @uninnocent "will this work with": it will work with anything you can communicate ... from" added "with" | removed "with" |
| 01:07:31 | FromDiscord | <uninnocent> In reply to @heysokam "it will work with": Im trying to do python + nim cross-communication |
| 01:07:53 | FromDiscord | <heysokam> how would you communicate with a C backend? |
| 01:07:58 | FromDiscord | <heysokam> apply that to nim, its the same answer |
| 01:07:59 | FromDiscord | <uninnocent> c? |
| 01:08:04 | FromDiscord | <uninnocent> Ive never written in c |
| 01:08:31 | FromDiscord | <heysokam> rust, C, any-other-systems-language-backend |
| 01:08:36 | FromDiscord | <heysokam> its all the same |
| 01:09:19 | FromDiscord | <uninnocent> I haven't used anything other than python. Im just looking for a straight answer as to how I can communicate between python / nim, not references to things that I've never done / dont know about |
| 01:10:26 | FromDiscord | <heysokam> i see. i haven't communicated from python with any backend, so can't guide you on that myself. but its basically the same as interoperating with C, which there should be a lot of information in the python community, since its backend is technically C |
| 01:10:26 | FromDiscord | <heysokam> thats why i assumed you could know about it already |
| 01:11:28 | FromDiscord | <graveflo> so this code doesn't even print "connected by" on the python scripts end? |
| 01:11:36 | FromDiscord | <uninnocent> it does, it prints connceted |
| 01:11:41 | FromDiscord | <uninnocent> But after I send a command, no response |
| 01:11:44 | FromDiscord | <graveflo> well you said that nothing is printed |
| 01:11:59 | FromDiscord | <uninnocent> In reply to @graveflo "well you said that": I meant to say no response from client |
| 01:11:59 | FromDiscord | <uninnocent> mb |
| 01:12:51 | FromDiscord | <graveflo> ok print the size of the cmd recved from nims end. The first one and see how much data is in the buffer |
| 01:15:26 | FromDiscord | <graveflo> you could also try using this version of `recv` since it doesn't try as hard to fill the buffer: https://nim-lang.org/docs/net.html#recv%2CSocket%2Cstring%2Cint%2Cint |
| 01:16:10 | FromDiscord | <graveflo> (edit) "you" => "~~you" | "https://nim-lang.org/docs/net.html#recv%2CSocket%2Cstring%2Cint%2Cint" => "https://nim-lang.org/docs/net.html#recv%2CSocket%2Cstring%2Cint%2Cint~~↵actually nvm that" |
| 01:17:11 | * | nyeaa492842301 quit (Read error: Connection reset by peer) |
| 01:17:36 | * | nyeaa492842301 joined #nim |
| 01:18:53 | FromDiscord | <demotomohiro> sent a code paste, see https://play.nim-lang.org/#ix=4Aos |
| 01:19:48 | FromDiscord | <demotomohiro> So I think `var cmd: string = socket.recv(1024)` waits for receiving 1024 byte data. |
| 01:20:39 | FromDiscord | <demotomohiro> Probably `data = conn.recv(1024)` in python code also doesn't return until it receive 1024 byte data. |
| 01:32:44 | * | disso-peach joined #nim |
| 01:34:55 | * | lucasta joined #nim |
| 01:41:27 | FromDiscord | <spotlightkid> For any socket based protocol you'll have to either agree on fixed length messages on each side or define some sort of message framing. E.g. sending a separator between each message (that doesn't occur in messages, ofc). Like `\n` or `\r\n` for example. The recipient can then read until it finds the separator, so that it knows a message is complete. |
| 01:43:29 | FromDiscord | <spotlightkid> No, it reads up to 1024 at maximum. With sockets you can not rely on reading the desired number of bytes in one `recv` call.↵(@demotomohiro) |
| 01:51:21 | FromDiscord | <sl0vak_> sent a code paste, see https://play.nim-lang.org/#ix=4Aow |
| 01:51:32 | FromDiscord | <spotlightkid> @uninnocent\: a few comments on you code\: in the Python code your `reponse` variable has a confusing name. It should really be called `request` imo, even if the server sends it. On line 21 you may receive the whole response or not (see my message abaove).↵In the nim code, on line 19 you need to check whether you received a complete command or otherwise repeat the rad until you do. How do you know whether the command is complete? |
| 01:55:20 | FromDiscord | <demotomohiro> In reply to @sl0vak_ "Hi, does anyone know": How about to use `os.copyFile`?↵https://nim-lang.org/docs/os.html#copyFile%2Cstring%2Cstring |
| 01:55:43 | FromDiscord | <arathanis> In reply to @sl0vak_ "Hi, does anyone know": what does reduce mean in this context? |
| 01:57:00 | FromDiscord | <arathanis> you write the character 4 to stdout, then copy the contents of the current file to a file called "4"? |
| 01:57:05 | FromDiscord | <sl0vak_> In reply to @arathanis "what does reduce mean": text |
| 01:57:14 | FromDiscord | <arathanis> In reply to @sl0vak_ "text": like code golf? |
| 01:57:18 | FromDiscord | <sl0vak_> In reply to @arathanis "you write the character": Yeah |
| 01:57:26 | FromDiscord | <sl0vak_> In reply to @arathanis "like code golf?": Yeah |
| 01:57:36 | FromDiscord | <arathanis> any particular reason why? and why on a single line? |
| 01:57:57 | FromDiscord | <sl0vak_> In reply to @arathanis "any particular reason why?": Bytes size |
| 01:58:53 | FromDiscord | <arathanis> it seems pretty reduced to me tbh |
| 01:59:07 | FromDiscord | <arathanis> why do you need to reduce the # of bytes? |
| 01:59:38 | FromDiscord | <sl0vak_> In reply to @arathanis "why do you need": For fun, a simple challenge |
| 02:00:03 | FromDiscord | <demotomohiro> `copyFile` doesnt reduce code size? |
| 02:00:22 | FromDiscord | <spotlightkid> It's not a really challenge if you ask other how to do it, isn't it? |
| 02:00:28 | FromDiscord | <sl0vak_> In reply to @demotomohiro "`copyFile` doesnt reduce code": I'll try it |
| 02:00:37 | FromDiscord | <arathanis> In reply to @spotlightkid "It's not a really": damn, true |
| 02:00:42 | FromDiscord | <sl0vak_> In reply to @spotlightkid "It's not a really": True |
| 02:01:04 | FromDiscord | <arathanis> code golf is evil outside of code golf tournaments anyway |
| 02:01:30 | FromDiscord | <sl0vak_> In reply to @arathanis "code golf is evil": Yep |
| 02:12:23 | * | Ekho quit (Quit: CORE ERROR, SYSTEM HALTED.) |
| 02:13:03 | FromDiscord | <gogolxdong666> sent a code paste, see https://play.nim-lang.org/#ix=4AoC |
| 02:13:38 | FromDiscord | <gogolxdong666> any solution to fix this up? |
| 02:14:07 | FromDiscord | <swordtrait> Can this code be made faster? https://media.discordapp.net/attachments/371759389889003532/1128509560307585114/image_7.png |
| 02:15:41 | * | Ekho joined #nim |
| 02:15:56 | FromDiscord | <swordtrait> This program now takes around 50ms |
| 02:16:16 | FromDiscord | <Elegantbeef> Did you compile it to release? |
| 02:16:25 | FromDiscord | <swordtrait> yes |
| 02:16:33 | FromDiscord | <swordtrait> -d:danger |
| 02:21:34 | FromDiscord | <swordtrait> Can it add parallel iterators, but I won't |
| 02:22:06 | FromDiscord | <swordtrait> (edit) "iterators, but I won't" => "iterators?" |
| 02:22:45 | FromDiscord | <Elegantbeef> If you write multithreaded iterators |
| 02:25:00 | FromDiscord | <swordtrait> I don't have the ability to use parallel iterators, I just started learning nim🤣 |
| 02:26:29 | FromDiscord | <arathanis> In reply to @Elegantbeef "If you write multithreaded": is this just an iterator that is safe to share between threads? |
| 02:26:43 | FromDiscord | <Elegantbeef> Who knows you'd be writing it |
| 02:33:14 | FromDiscord | <heysokam> sent a code paste, see https://play.nim-lang.org/#ix=4AoH |
| 02:33:26 | FromDiscord | <Elegantbeef> Cause you don't want it to |
| 02:33:40 | FromDiscord | <ringabout> In reply to @gogolxdong666 "any idea to fix": Did you use the latest nightlies? what becomes of `nim -v`? |
| 02:33:56 | FromDiscord | <Elegantbeef> sent a code paste, see https://paste.rs/v0x5t |
| 02:34:08 | FromDiscord | <gogolxdong666> sent a long message, see http://ix.io/4AoJ |
| 02:34:34 | FromDiscord | <gogolxdong666> Is this a relative issue? |
| 02:34:51 | FromDiscord | <heysokam> sent a code paste, see https://play.nim-lang.org/#ix=4AoK |
| 02:35:34 | FromDiscord | <gogolxdong666> latest devel |
| 02:41:04 | * | rockcavera quit (Remote host closed the connection) |
| 02:52:01 | FromDiscord | <uninnocent> In reply to @graveflo "ok print the size": How would I do that? |
| 02:53:40 | FromDiscord | <graveflo> just echo the `len` of data. Don't just look at the string, See if there is anything there. If it blocks then it might be trying to fill the buffer. |
| 02:53:47 | FromDiscord | <uninnocent> Alright |
| 02:54:36 | FromDiscord | <uninnocent> sent a long message, see http://ix.io/4AoM |
| 02:54:51 | FromDiscord | <uninnocent> No output at all |
| 02:54:52 | FromDiscord | <graveflo> then it is blocking. It might be trying to fill the buffer |
| 02:56:32 | FromDiscord | <ringabout> @graveflo hi, would you mind fixing the render bug mentioned in the internals channel https://discord.com/channels/371759389889003530/768367394547957761/1125529732855906375 ? |
| 02:57:00 | FromDiscord | <uninnocent> Oh, it actually does print something when I close the server @graveflo |
| 02:57:51 | FromDiscord | <uninnocent> https://media.discordapp.net/attachments/371759389889003532/1128520566203224115/image.png |
| 02:58:34 | FromDiscord | <uninnocent> https://media.discordapp.net/attachments/371759389889003532/1128520747812393132/image.png |
| 02:58:42 | FromDiscord | <uninnocent> window 1 = client↵window 2 = server |
| 02:59:06 | FromDiscord | <uninnocent> echo len(data) only runs when I kill the servre |
| 02:59:13 | FromDiscord | <graveflo> In reply to @ringabout "<@200775012796334081> hi, would you": oh sure, I forgot about that. Thanks |
| 02:59:22 | FromDiscord | <uninnocent> if I run a command and THEN kill the server, 0 isnt echoed |
| 02:59:45 | FromDiscord | <uninnocent> (edit) "servre" => "server" |
| 03:00:45 | FromDiscord | <graveflo> In reply to @uninnocent "if I run a": does the python script print "Sent to... " |
| 03:01:13 | FromDiscord | <uninnocent> https://media.discordapp.net/attachments/371759389889003532/1128521415595921458/image.png |
| 03:01:17 | FromDiscord | <uninnocent> this is all t gets up to every time |
| 03:01:22 | FromDiscord | <uninnocent> (edit) "t gets" => "igets" |
| 03:01:27 | FromDiscord | <uninnocent> (edit) "igets" => "it gets" |
| 03:07:42 | FromDiscord | <uninnocent> @graveflo forgot to mention this, but the client can send data to the server fine |
| 03:07:50 | FromDiscord | <uninnocent> It just cant read the server's responded |
| 03:07:53 | FromDiscord | <uninnocent> (edit) "responded" => "responses" |
| 03:08:58 | FromDiscord | <graveflo> ok try using the `recvLine` instead, just for now. And make sure the python script is using a line termination character |
| 03:10:00 | FromDiscord | <uninnocent> I did something to fix it but idk what |
| 03:10:04 | FromDiscord | <uninnocent> its not fully fixed |
| 03:12:40 | FromDiscord | <uninnocent> https://media.discordapp.net/attachments/371759389889003532/1128524296071946312/image.png |
| 03:13:02 | FromDiscord | <uninnocent> if I make the server error at a certian point the command goes through |
| 03:13:25 | FromDiscord | <uninnocent> sent a code paste, see https://play.nim-lang.org/#ix=4AoQ |
| 03:14:13 | FromDiscord | <graveflo> you can try using `send` instead of `sendall` the python script. I think you might be able to `flush` in python too if that is the issue |
| 03:15:41 | FromDiscord | <uninnocent> send instead of sendall didnt do anything. |
| 03:15:44 | FromDiscord | <uninnocent> And what's flush |
| 03:16:56 | FromDiscord | <graveflo> on buffered streams it tries to dump everything from the buffer to the device |
| 03:18:42 | FromDiscord | <graveflo> well idk if this is just a typo or something but you have `conn.recv` instead of `conn.recv(1024)` for example. So you are not getting bytes.. `data` in this case would be a callable object |
| 03:19:52 | FromDiscord | <uninnocent> In reply to @graveflo "well idk if this": I know, I put a comment there. |
| 03:20:08 | FromDiscord | <uninnocent> In reply to @uninnocent "": If I intentonally make that line error, it shows this |
| 03:20:16 | FromDiscord | <uninnocent> if I add the 1024 thing, it just hangs |
| 03:25:09 | FromDiscord | <graveflo> it works if you add `os.linesep` on pythons side and use `recvLine` on nims side. This seems to be a buffering issue. You need to maybe use an unbuffered socket in Nim or look into how to dump the buffer on demand. It seems Nim is trying to fill the entire 1024 bytes when you call `recv`. That is my guess |
| 03:26:27 | * | Onionhammer quit (Quit: The Lounge - https://thelounge.chat) |
| 03:28:51 | FromDiscord | <graveflo> yea looks like `newSocket(AF_INET, SOCK_STREAM, buffered=false)` works if you dont want to use `recvLine` |
| 03:31:58 | * | Onionhammer joined #nim |
| 03:36:23 | FromDiscord | <demotomohiro> https://stackoverflow.com/questions/9563563/what-is-a-message-boundary |
| 03:38:09 | FromDiscord | <demotomohiro> https://stackoverflow.com/questions/17446491/tcp-stream-vs-udp-message |
| 05:07:02 | FromDiscord | <ri> v2.0? |
| 05:07:29 | FromDiscord | <ri> 🧐 |
| 05:29:18 | FromDiscord | <uninnocent> In reply to @graveflo "yea looks like `newSocket(AF_INET,": Perfect, thank you so much. This worked. |
| 06:01:41 | * | jmdaemon- joined #nim |
| 06:02:32 | * | jmd_ quit (Ping timeout: 246 seconds) |
| 06:09:49 | FromDiscord | <PingusMcDuck> You guys see this Mojo language, supposed to be pythons exact syntax with C performance |
| 06:13:05 | FromDiscord | <Elegantbeef> "pythons exact syntax" |
| 06:13:05 | FromDiscord | <Elegantbeef> It's like people don't even look at things |
| 06:16:54 | * | lucasta quit (Remote host closed the connection) |
| 06:19:43 | FromDiscord | <huantian> They probably just don’t know how to phrase mojo’s supersettyness of python |
| 06:20:03 | FromDiscord | <Elegantbeef> Awful |
| 06:20:08 | FromDiscord | <Elegantbeef> That's how one phrases it |
| 06:26:33 | FromDiscord | <PingusMcDuck> "Embracing Python massively simplifies our design efforts, because most of the syntax is already specified"↵↵As far as syntax goes am I reading ir incorrect? |
| 06:26:40 | FromDiscord | <Elegantbeef> Sure they support python, but they also introduced their own superset language |
| 06:28:16 | FromDiscord | <Elegantbeef> sent a code paste, see https://play.nim-lang.org/#ix=4Apa |
| 06:30:15 | FromDiscord | <Elegantbeef> or `fn doThing(inout a: MyType)` |
| 06:30:19 | FromDiscord | <Elegantbeef> Apparently they changed that I could've sworn that used to be `a: &MyType` |
| 06:32:18 | * | ntat joined #nim |
| 06:41:10 | * | PMunch joined #nim |
| 07:02:16 | FromDiscord | <uninnocent> sent a code paste, see https://play.nim-lang.org/#ix=4Apd |
| 07:03:24 | FromDiscord | <Elegantbeef> by using`startProcess` |
| 07:03:31 | FromDiscord | <Elegantbeef> Hmm is this on the edge of malware... I cannot tell |
| 07:04:44 | Amun-Ra | this and the nickname… ;) |
| 07:05:03 | FromDiscord | <uninnocent> In reply to @Amun-Ra "this and the nickname…": Lol |
| 07:11:08 | * | Jjp137 quit (Ping timeout: 246 seconds) |
| 07:20:18 | * | Jjp137 joined #nim |
| 08:00:01 | FromDiscord | <System64 ~ Flandre Scarlet> Hi, do you know why I have those errors with Clang please?↵Everything is fine wigth gcc https://media.discordapp.net/attachments/371759389889003532/1128596612789309450/message.txt |
| 08:00:05 | FromDiscord | <System64 ~ Flandre Scarlet> (edit) "wigth" => "with" |
| 08:03:15 | FromDiscord | <System64 ~ Flandre Scarlet> I also use C++ backend |
| 08:04:54 | PMunch | Seems like you're missing a linker flag |
| 08:07:48 | FromDiscord | <System64 ~ Flandre Scarlet> In reply to @PMunch "Seems like you're missing": Ah alright, which one? |
| 08:09:13 | PMunch | No idea, but those are linker errors which says it can't find various functions. So either you're trying to link against a different version of the library which is missing those functions, or you're missing a linker flag |
| 08:28:01 | FromDiscord | <System64 ~ Flandre Scarlet> In reply to @PMunch "No idea, but those": But why GCC has no problem with that? |
| 08:28:39 | NimEventer | New thread by grd: Does anyone knows how good github copilot is when..., see https://forum.nim-lang.org/t/10334 |
| 08:28:50 | PMunch | Different search paths? Different default options? Could be anything |
| 08:30:44 | * | ovenpasta joined #nim |
| 08:47:22 | FromDiscord | <System64 ~ Flandre Scarlet> Will try to solve this↵Btw I have a question↵Is it possible to generate code with a macro? |
| 08:47:50 | Amun-Ra | I can see no contents in this message.txt |
| 08:49:24 | FromDiscord | <System64 ~ Flandre Scarlet> In reply to @Amun-Ra "I can see no": You need to download it because too big for Discord to display it |
| 08:49:48 | Amun-Ra | I see "content-length: 0" |
| 08:51:31 | FromDiscord | <System64 ~ Flandre Scarlet> sent a code paste, see https://play.nim-lang.org/#ix=4Apr |
| 09:06:07 | FromDiscord | <ieltan> In reply to @sys64 "Will try to solve": I think it's literally what they are meant to be used for but I am a dunce at macros |
| 09:07:08 | FromDiscord | <ieltan> In reply to @sys64 "For example I want": Shouldn't the optimizing compiler do this for you ? |
| 09:12:10 | FromDiscord | <ambient3332> TIL: nimble generateApi |
| 09:12:48 | FromDiscord | <Yardanico> In reply to @sys64 "Will try to solve": > Is it possible to generate code with a macro?↵that's literally what they're meant for, so yes |
| 09:13:10 | FromDiscord | <Yardanico> in fact there's at least one package that does some unroll at compile time thing, and i've made a toy macro for unrolling loops with static iteration expressions a while ago |
| 09:13:36 | FromDiscord | <Yardanico> you just loop over the values in the macro yourself and generate the code repeatedly with the changed `i` in the code |
| 09:14:47 | FromDiscord | <ambient3332> (it was actually just a rule for exec("nim r bindgen/bindgen.nim")) |
| 09:15:54 | FromDiscord | <ambient3332> Wouldn't it be wonderful if you could just press a button and a Nim API magically appeared for any C++ library or software you want |
| 09:15:55 | * | lukasp joined #nim |
| 09:16:04 | lukasp | Dear all. Since Nim errs on funcs with side effects, it apparently has knowledge about side effects. I was wondering if Nim automatically turns a proc into a func when it has no side effects. In other words, if func just exists for developers to express their assumptions but given the assumptions hold, it would not make a difference to use proc. |
| 09:16:31 | FromDiscord | <Yardanico> In reply to @ambient3332 "Wouldn't it be wonderful": Futhark is kind of meant for this, but it's not always 100% automatic :) |
| 09:16:36 | FromDiscord | <Yardanico> before futhark there was also nimporter |
| 09:17:08 | FromDiscord | <ambient3332> Yeah, but is it actually robust enough to be used or do I end up chasing some rabbit hole for a month? |
| 09:17:38 | PMunch | ambient3332, I've used it for production software |
| 09:17:40 | FromDiscord | <System64 ~ Flandre Scarlet> In reply to @ieltan "Shouldn't the optimizing compiler": I'm not sure, since a is a variable |
| 09:17:55 | PMunch | I wrote it because I was tired of chasing rabbit holes.. |
| 09:18:10 | FromDiscord | <System64 ~ Flandre Scarlet> In reply to @yardanico "> Is it possible": How can I do that in the case of my for loop? |
| 09:18:14 | PMunch | Be warned though, it doesn't do C++ at the moment. Only C |
| 09:18:25 | PMunch | But it has been really solid for me |
| 09:18:25 | FromDiscord | <Yardanico> In reply to @sys64 "How can I do": make a macro? :P |
| 09:18:46 | PMunch | Even managed to wrap Gtk with GtkWebView without fuzz |
| 09:19:08 | FromDiscord | <ambient3332> I can write a simple C API for a C++ lib myself as long as the process is stable |
| 09:19:21 | FromDiscord | <System64 ~ Flandre Scarlet> In reply to @PMunch "Be warned though, it": Ah so I compile to C++, the compiler will yield an error? |
| 09:19:41 | FromDiscord | <Yardanico> In reply to @sys64 "Ah so I compile": no, it just won't work, futhark only does C |
| 09:20:28 | FromDiscord | <Yardanico> In reply to @yardanico "make a macro? :P": i've had one simple one, but I don't know where to find it now |
| 09:20:32 | PMunch | System64_~_Flandre_Scarlet, that was meant for ambient. Futhark will work just fine if you're importing a C library but compiling for C++ |
| 09:21:15 | FromDiscord | <System64 ~ Flandre Scarlet> Oh, srry |
| 09:21:46 | FromDiscord | <Yardanico> In reply to @sys64 "Oh, srry": https://forum.nim-lang.org/t/9504 |
| 09:22:28 | FromDiscord | <Yardanico> <https://github.com/juancarlospaco/nodejs/blob/main/src/nodejs/jsunrolls.nim#L4> should work too (it doesn't seem to be js specific at all) |
| 09:22:49 | FromDiscord | <Yardanico> although it has an extra runtime variable, while it could just inline the variable itself to the call |
| 09:23:05 | FromDiscord | <Yardanico> but this does make the implementation much easier |
| 09:23:15 | FromDiscord | <System64 ~ Flandre Scarlet> Ah thanks! |
| 09:29:46 | FromDiscord | <System64 ~ Flandre Scarlet> In reply to @yardanico "https://forum.nim-lang.org/t/9504": https://github.com/schneiderfelipe/unrolled↵Found this, but doesn't work |
| 09:30:34 | FromDiscord | <ieltan> In reply to @yardanico "before futhark there was": Btw I heard very little things about nimporter, what gives ? |
| 09:30:48 | FromDiscord | <Yardanico> In reply to @ieltan "Btw I heard very": wait sorry |
| 09:30:57 | FromDiscord | <Yardanico> that's the wrong name, nimporter is just a small wrapper on top of nimpy for nim |
| 09:30:59 | FromDiscord | <Yardanico> i meant nimterop :) |
| 09:31:07 | FromDiscord | <ieltan> Oh I see haha |
| 09:31:14 | FromDiscord | <Yardanico> nimterop is unmaintained now, the maintained alternative is futhark |
| 09:31:16 | FromDiscord | <demotomohiro> sent a code paste, see https://play.nim-lang.org/#ix=4Apw |
| 09:31:40 | FromDiscord | <Yardanico> sent a code paste, see https://play.nim-lang.org/#ix=4Apx |
| 09:32:00 | FromDiscord | <Yardanico> it's usually not hard to see that |
| 09:33:09 | PMunch | Personally I could never get nimterop to work. Some people seemed to get it working though |
| 09:33:54 | PMunch | Granted I couldn't get c2nim to work either, so maybe I was the problem |
| 09:35:18 | FromDiscord | <demotomohiro> In reply to @yardanico "> Translater has AI": If you know C lang and read C library documents, it would be not hard.↵But how automatic translater like futhark or c2nim do that? |
| 09:35:26 | FromDiscord | <Yardanico> they don't do that, they do best effort |
| 09:38:12 | FromDiscord | <ambient3332> How I use c2nim is just run it on the C API and do a couple of manual fixes. Worked well so far |
| 09:38:18 | FromDiscord | <demotomohiro> Then, if you want good translated Nim code, you need to know C lang, read C library documents and fix translated code. |
| 09:39:05 | * | lukasp quit (Quit: Client closed) |
| 09:39:08 | FromDiscord | <ambient3332> Couldn't I just write a C api for a C++ lib with Nim directly? |
| 09:39:16 | FromDiscord | <ambient3332> Just do .emit. |
| 09:39:33 | FromDiscord | <ambient3332> or something |
| 09:41:10 | FromDiscord | <ambient3332> The way I figure out is that you have 1. C++ lib that's C++ object file (mangled names), 2. C API for C++ lib that just calls C++ funcs and gives C ABI, 3. Nim lib that consumes the C ABI. But could the 2. just not be replaced directly with Nim instead? |
| 09:43:06 | FromDiscord | <ambient3332> So I would not have to mess with any C/C++ build systems and could stay inside Nim the whole time |
| 09:43:41 | FromDiscord | <demotomohiro> Nim has pragmas to import C++ classes or functions:↵https://nim-lang.org/docs/manual.html#implementation-specific-pragmas-importcpp-pragma |
| 09:44:04 | FromDiscord | <ambient3332> Indeed, but I need to do separate `nim cpp` |
| 09:44:11 | FromDiscord | <ambient3332> (which is fine) |
| 09:44:31 | FromDiscord | <ambient3332> Just looking the least painful proces to import complex C++ libs so I can run `nim c` |
| 09:46:31 | FromDiscord | <demotomohiro> Maybe, import C++ lib to Nim code with importcpp pragma, export C functions and compile it as static or dynamic link library? |
| 10:06:01 | * | Guest99 joined #nim |
| 10:17:22 | PMunch | ambient, using c2nim and manual fixes is fine if you only have a few fixes to make and you don't ever update what you're interfacing against. Problem is if you wrap a large library there is a lot of stuff to manually fix, and each fix has the potential to introduce hard-to-spot bugs. Then the library updates and you have to do everything all over again, which is just a world of pain. |
| 10:17:56 | PMunch | And Futhark could support C++, it's just a matter of figuring out and writing the rules for how to wrap C++ stuff in Nim. |
| 10:18:31 | PMunch | Say you have something which returns a std::vector, what does that map to in C? What should it map to in Nim? |
| 10:39:30 | FromDiscord | <System64 ~ Flandre Scarlet> How can I make my Nim program not detected as a virus please? |
| 10:42:50 | PMunch | Plead with AV vendors to stop their poor fingerprinting of Nim binaries |
| 10:43:48 | FromDiscord | <System64 ~ Flandre Scarlet> In reply to @PMunch "Plead with AV vendors": Even worse : This is Windows Defender's antivirus |
| 10:44:14 | FromDiscord | <spotlightkid> then plead wih MS |
| 10:45:07 | FromDiscord | <System64 ~ Flandre Scarlet> MS sucks |
| 10:45:09 | PMunch | Point still stands, MS makes AV = MS is an AV vendor |
| 10:52:36 | FromDiscord | <System64 ~ Flandre Scarlet> would signing the program work? |
| 10:55:15 | FromDiscord | <Andreas> In reply to @PMunch "Point still stands, MS": maybe, join with people from other affected groups first, collect some money, hire a FSF-lawyer and hope for the best.. |
| 11:03:54 | FromDiscord | <demotomohiro> Or somehow convince people that your program is not virus and ask people to ignore AV.↵Or stop using windows and use Linux, ReactOS or other OS. |
| 11:17:32 | * | Guest99 quit (Quit: Client closed) |
| 11:20:56 | * | Guest99 joined #nim |
| 11:21:04 | * | PMunch_ joined #nim |
| 11:22:13 | * | Guest99 is now known as m2a2 |
| 11:23:01 | FromDiscord | <Andreas> In reply to @demotomohiro "Or somehow convince people": no, it the AV-Vendors duty to deliver a proof, that binaries generated by Nim|D|other-languages perform malicious actions. What they do is -> there has been a malware made by nim, so we block anything and that is imo not fair or legal. Any affected company would take this to court. |
| 11:24:35 | * | PMunch quit (Ping timeout: 264 seconds) |
| 11:31:01 | FromDiscord | <System64 ~ Flandre Scarlet> The open source community is affected by this too |
| 11:33:19 | FromDiscord | <Andreas> In reply to @sys64 "The open source community": correct, even some Python-packages seem to be affected - i wonder how the Python-Foundation would react if anything-python would be detected as malware ? |
| 11:34:21 | FromDiscord | <System64 ~ Flandre Scarlet> In reply to @Andreas "correct, even some Python-packages": I can't distribute my synthesizer without anyone saying "uh oh! Detected as virus" |
| 11:35:31 | FromDiscord | <System64 ~ Flandre Scarlet> And I can't download an open source program without SmartScreen saying "uh oh! It can be dangerous! Be careful!" |
| 11:38:52 | * | jmdaemon- quit (Ping timeout: 240 seconds) |
| 11:40:29 | FromDiscord | <spotlightkid> welcome to corporate walled gardens! |
| 11:42:56 | FromDiscord | <Andreas> In reply to @sys64 "I can't distribute my": mybe try to follow https://forum.nim-lang.org/t/9850 pietpropers suggestions and have your music-app tested and signed to get around the virus-detection ? |
| 11:43:37 | FromDiscord | <System64 ~ Flandre Scarlet> In reply to @spotlightkid "welcome to corporate walled": I think you can sign your app but obviously... It costs moneeeey!↵I doubt it's still about security anymore, more like it's about money |
| 11:43:48 | FromDiscord | <System64 ~ Flandre Scarlet> In reply to @Andreas "mybe try to follow": Gotta read that |
| 11:44:59 | FromDiscord | <Andreas> In reply to @sys64 "I think you can": if thats really true, then the FSF should be interested to step-in, because this hinders free-software-developments.. |
| 11:51:06 | * | ntat quit (Quit: Leaving) |
| 11:52:48 | FromDiscord | <jmgomez> In reply to @PMunch "Say you have something": why just dont use the cpp backend instead? |
| 11:54:09 | FromDiscord | <jmgomez> In reply to @demotomohiro "Maybe, import C++ lib": this is the way to go about it but @ambient3332 if `nim c` is not a hard requirement, just use the cpp backend |
| 11:56:34 | FromDiscord | <dlesnoff> sent a code paste, see https://play.nim-lang.org/#ix=4ApP |
| 11:58:41 | FromDiscord | <jviega> Well, almost certainly libclang.a is not in your path. Add a C flag, -L/path/to/where/lib/clang/is |
| 11:59:12 | FromDiscord | <pmunch> Yup, this has been mentioned in several issues already: https://github.com/PMunch/futhark/issues/27 |
| 12:00:14 | FromDiscord | <pmunch> It should be fixed once I merge this: https://github.com/PMunch/futhark/pull/78 |
| 12:06:04 | FromDiscord | <dlesnoff> Thanks. Sorry, I could have searched a little harder. |
| 12:08:33 | FromDiscord | <spotlightkid> Debian/Ubuntu and their stupid messing with installation paths. I had fun with this with a Python project too. They patch their python's `sys.path` to change the install location for third-party packages but then don't fix the stdlib `sysconfig` module. |
| 12:28:57 | FromDiscord | <fabricio> I don't know if there is a better place to ask this, but has anyone tried to integrate nim.nvim or nimsuggest autocompletion with coq_nvim? |
| 12:51:36 | FromDiscord | <djazz> In reply to @jiezron "Thanks. Sorry, I could": I opened that PR just the other day 😄 |
| 12:52:43 | FromDiscord | <djazz> Meanwhile use nimble --passL:-L/usr/lib/llvm-14/lib install futhark |
| 12:58:19 | FromDiscord | <ambient3332> In reply to @PMunch "Say you have something": Probably something like this https://github.com/SciNim/impulse/blob/26e25e701be75446ad2b91403a4538465f44f1b5/impulse/fft/std_cpp.nim |
| 13:11:01 | FromDiscord | <takemichihanagaki3129> In reply to @Andreas "correct, even some Python-packages": I had this problem twice on the same year.↵A nim version was detected by my Windows Defender as a malware. ↵And in other occasion, an application that I developed using Python and compiling via Nuitka was flagged the same way too... |
| 13:16:01 | FromDiscord | <Andreas> In reply to @takemichihanagaki3129 "I had this problem": i just emailed to FSF and brought the case of "false positives' to their attention. Currently there is no 'campain' covering this topic. I'm sure they are already aware - lets wait a bit and maybe the FSF surveys their members and starts acting upon it.. |
| 13:16:56 | * | rockcavera joined #nim |
| 13:41:22 | PMunch_ | ambient, yeah probably. I basically just need someone with enough C++ knowledge (and preferably wrapper knowledge) to help out with Futhark |
| 13:58:11 | FromDiscord | <jmgomez> In reply to @PMunch_ "ambient, yeah probably. I": not a cpp expert but feel free to ask, learned a thing or two about it by doing NUE 🙂 |
| 13:59:58 | PMunch_ | jmgomez, yeah I've been meaning to get in touch about it :) But I've been very busy at work and with renovating the house we just bought |
| 14:01:49 | FromDiscord | <blashyrk> Is every source file a separate module in nim? If I have multiple source files and `import X` in one or another, I can't access "private" fields. Is there a way to have multiple source files in a single module? |
| 14:02:08 | PMunch_ | You could use include |
| 14:02:21 | PMunch_ | include basically copy-pastes one file into another |
| 14:02:37 | PMunch_ | While import treats it as a separate module |
| 14:03:32 | FromDiscord | <ieltan> Is it possible for a macro to extract type information about a closure and manipulate it ? |
| 14:03:55 | FromDiscord | <ieltan> also wish to do the same for an object field |
| 14:05:23 | * | PMunch_ quit (Quit: Leaving) |
| 14:10:02 | * | lucasta joined #nim |
| 14:11:33 | * | lucasta quit (Remote host closed the connection) |
| 14:11:57 | * | lucasta joined #nim |
| 14:12:25 | * | disso-peach quit (Quit: Leaving) |
| 14:13:15 | FromDiscord | <jmgomez> In reply to @PMunch_ "<@726017160115126333>, yeah I've been": cool, take your time and congrats for the house! |
| 14:50:55 | FromDiscord | <bung8954> sent a code paste, see https://play.nim-lang.org/#ix=4Aqw |
| 15:16:53 | NimEventer | New thread by beckx: GUI-LIB nip - button callback and object data, see https://forum.nim-lang.org/t/10335 |
| 16:15:17 | * | lucasta quit (Remote host closed the connection) |
| 17:25:21 | * | m2a2 quit (Quit: Client closed) |
| 17:26:37 | * | Guest99 joined #nim |
| 17:27:54 | * | Guest99 is now known as m2a2 |
| 17:45:46 | FromDiscord | <eliot> hello, i've recently installed nim using choosenim on windows but i seem to be getting an error related to git when installing a package thru nimble. heres the output of when i try to install winim: |
| 17:45:59 | FromDiscord | <eliot> sent a long message, see http://ix.io/4Arw |
| 17:46:20 | FromDiscord | <eliot> (edit) "http://ix.io/4Arw" => "http://ix.io/4Arx" |
| 17:46:29 | FromDiscord | <eliot> it then shows every git argument |
| 17:46:40 | FromDiscord | <eliot> (edit) "http://ix.io/4Arx" => "http://ix.io/4Ary" |
| 17:47:19 | FromDiscord | <graveflo> its a bug |
| 17:47:33 | FromDiscord | <graveflo> your username has a space and the command is formatted improperly because of this |
| 17:47:39 | FromDiscord | <eliot> ooh |
| 17:47:47 | * | ntat joined #nim |
| 17:56:50 | * | anddam quit (Quit: WeeChat 4.0.1) |
| 18:02:11 | * | anddam joined #nim |
| 18:04:33 | * | anddam quit (Client Quit) |
| 18:05:33 | * | anddam joined #nim |
| 18:30:38 | FromDiscord | <bostonboston> When generating a dll I specify the stdcall calling conv but the produced symbols are not prefixed with an underscore, infact no matter what calling conv I specify the symbols do not seem to change? |
| 18:32:20 | FromDiscord | <bostonboston> sent a code paste, see https://play.nim-lang.org/#ix=4ArM |
| 18:46:55 | FromDiscord | <bostonboston> Okay I figured out why my symbols weren't changing based on the calling conv but I still don't know why stdcall doesn't prepend an underscore |
| 18:52:29 | FromDiscord | <System64 ~ Flandre Scarlet> sent a code paste, see https://play.nim-lang.org/#ix=4ArP |
| 18:53:05 | FromDiscord | <System64 ~ Flandre Scarlet> Because it struggles with high window lengths |
| 19:04:04 | FromDiscord | <juan_carlos> In reply to @bostonboston "Okay I figured out": `{.exportc: "_foo".}` can take a name. |
| 19:05:37 | FromDiscord | <bostonboston> Yes but I shouldn't have to do that if the stdcall standard is to do it by default right? |
| 19:06:18 | * | azimut_ quit (Ping timeout: 240 seconds) |
| 19:07:39 | FromDiscord | <juan_carlos> In reply to @bostonboston "Yes but I shouldn't": If you think its a bug, then should be reported. |
| 19:20:00 | FromDiscord | <bostonboston> I don't necessarily think it's a bug, I was assuming I'm doing something wrong |
| 19:20:08 | * | m2a2 quit (Quit: Client closed) |
| 19:37:49 | FromDiscord | <nnsee> In reply to @sys64 "Because it struggles with": profile to find out? |
| 19:37:55 | FromDiscord | <nnsee> check out https://nim-lang.org/blog/2017/10/02/documenting-profiling-and-debugging-nim-code.html#profiling-your-code |
| 19:43:53 | * | ntat quit (Quit: Leaving) |
| 19:47:42 | * | ntat joined #nim |
| 21:10:12 | * | ntat quit (Quit: Leaving) |
| 21:10:23 | FromDiscord | <ambient3332> sent a code paste, see https://play.nim-lang.org/#ix=4Asb |
| 21:10:37 | FromDiscord | <ambient3332> for example instead of dividing by a constant, you can put var foo = 1/CONSTANT outside the loop |
| 21:10:41 | FromDiscord | <ambient3332> then multiply by it |
| 21:11:27 | FromDiscord | <Elegantbeef> Also not using recursive logic |
| 21:11:44 | FromDiscord | <ambient3332> Yeah, I was thinking the same but not knowing anything of the outside logic, can't say much |
| 21:11:53 | FromDiscord | <Elegantbeef> Also not using methods |
| 21:12:18 | FromDiscord | <meriem> Hi |
| 21:12:38 | FromDiscord | <ambient3332> Generally as your synth is in dataflow format, it should be unnecessary to do anything recursive, just do a toposort |
| 21:13:09 | FromDiscord | <ambient3332> (although as I know very little who know, just as general guidelines) |
| 21:28:22 | FromDiscord | <System64 ~ Flandre Scarlet> In reply to @ambient3332 "Generally as your synth": toposort?↵Also output call a module that call a module that call a module and so on |
| 21:28:48 | FromDiscord | <System64 ~ Flandre Scarlet> so it's like↵function(function(function(function(function(function(function(...))))))) |
| 21:47:48 | FromDiscord | <Elegantbeef> Recursion bad |
| 21:56:14 | FromDiscord | <System64 ~ Flandre Scarlet> In reply to @Elegantbeef "Recursion bad": I don't see any other way to make a modular synth like this |
| 21:56:55 | FromDiscord | <Elegantbeef> Well profile it first |
| 22:18:04 | * | jmdaemon joined #nim |
| 22:21:41 | * | ovenpasta quit (Ping timeout: 245 seconds) |
| 23:04:24 | FromDiscord | <Elegantbeef> @ringabout\: an inane wiki page was created. |
| 23:04:25 | FromDiscord | <Elegantbeef> https://github.com/nim-lang/Nim/wiki/World |
| 23:24:16 | * | ormiret_ joined #nim |
| 23:26:22 | * | blackbeard420_ joined #nim |
| 23:26:29 | * | blackbeard420_ quit (Remote host closed the connection) |
| 23:26:53 | * | blackbeard420_ joined #nim |
| 23:29:13 | * | lumidify_ joined #nim |
| 23:32:30 | * | ormiret quit (*.net *.split) |
| 23:32:30 | * | lumidify quit (*.net *.split) |
| 23:32:30 | * | mahlon quit (*.net *.split) |
| 23:32:31 | * | blackbeard420 quit (*.net *.split) |
| 23:32:31 | * | ormiret_ is now known as ormiret |
