| 00:00:11 | AlexMax | overview.c is 1306 |
| 00:00:40 | AlexMax | lines |
| 00:20:33 | * | jimi|ansible joined #nim |
| 00:25:10 | * | jimi|ansible quit (Ping timeout: 246 seconds) |
| 00:28:27 | * | junland joined #nim |
| 00:32:15 | * | nephyrin11 joined #nim |
| 00:33:20 | * | krux02 quit (Remote host closed the connection) |
| 00:36:35 | * | nephyrin11 quit (Remote host closed the connection) |
| 00:39:49 | * | m27frogy23 joined #nim |
| 00:43:22 | * | m27frogy23 quit (Remote host closed the connection) |
| 00:54:02 | AlexMax | hrm, c2nim seems to have skipped over something |
| 00:54:49 | AlexMax | ohhhhhhhhhhhh |
| 00:54:55 | AlexMax | that's becuase it's a #define |
| 00:55:09 | * | kmehall8 joined #nim |
| 00:55:18 | * | redlegion_ joined #nim |
| 00:57:47 | * | kmehall8 quit (Remote host closed the connection) |
| 00:59:14 | * | PrimHelios quit (Quit: Leaving) |
| 01:03:52 | AlexMax | Is there an equivalent of __LINE__ or __FILE__ in Nim? |
| 01:04:37 | AlexMax | actually hrm, that's why this is a macro...it uses the line number and file name to initialize a unique hash |
| 01:05:32 | * | dddddd quit (Ping timeout: 252 seconds) |
| 01:05:57 | * | gitness20 joined #nim |
| 01:06:11 | AlexMax | so wrapping that function will be difficult |
| 01:09:58 | * | gitness20 quit (Remote host closed the connection) |
| 01:14:53 | * | redlegion_ quit (Ping timeout: 252 seconds) |
| 01:15:51 | leorize | AlexMax: you can use `{.emit.}` |
| 01:16:13 | leorize | or just replicate the macro in nim |
| 01:18:47 | FromGitter | <zacharycarter> there's ways to replace C macros in the c2nim header section |
| 01:18:55 | FromGitter | <zacharycarter> if they're simple |
| 01:19:29 | FromGitter | <zacharycarter> if they're not - I'd recommend using a C/C++ pre-processor before running your code through c2nim |
| 01:19:49 | FromGitter | <zacharycarter> or you could use nimgen - which does a lot of this for you I believe AlexMax ^ |
| 01:20:05 | AlexMax | I'm already using nimgen. That's why the original macro isn't even the c2nim'ed source |
| 01:20:31 | FromGitter | <zacharycarter> weird |
| 01:20:42 | FromGitter | <zacharycarter> I don't know the internals of nimgen well enough to comment |
| 01:20:57 | AlexMax | Well, it's obvious why it's gone, the macro is defined, but never "called" |
| 01:21:06 | * | stefanos82 quit (Quit: Quitting for now...) |
| 01:21:12 | FromGitter | <zacharycarter> all I know is - when I run into macros in code I'm trying to bind to - I usually run `gcc -E` |
| 01:21:27 | AlexMax | and even if it was called, it would be expanded into a form of nk_tree_push_hashed |
| 01:21:30 | FromGitter | <zacharycarter> ah - I solve this problem with Nuklear I believe |
| 01:21:34 | FromGitter | <zacharycarter> in the header files |
| 01:21:38 | FromGitter | <zacharycarter> oh |
| 01:21:52 | AlexMax | What I'm trying to figure out is if I can recreate the intent of the macro from nothing |
| 01:22:00 | AlexMax | the macro is |
| 01:22:01 | AlexMax | #define nk_tree_push_id(ctx, type, title, state, id) nk_tree_push_hashed(ctx, type, title, state, NK_FILE_LINE,nk_strlen(NK_FILE_LINE),id) |
| 01:22:13 | AlexMax | and |
| 01:22:14 | AlexMax | #define nk_tree_push(ctx, type, title, state) nk_tree_push_hashed(ctx, type, title, state, NK_FILE_LINE,nk_strlen(NK_FILE_LINE),__LINE__) |
| 01:22:39 | AlexMax | It uses __FILE__ and __LINE__ to come up with a hash unique to that invocation |
| 01:23:02 | AlexMax | and I bet some other mechanism could replace it |
| 01:23:45 | * | adelrune quit (Ping timeout: 244 seconds) |
| 01:27:53 | FromGitter | <zacharycarter> HRM |
| 01:28:03 | FromGitter | <zacharycarter> sorry - I wouldn't try to replace the macros like that |
| 01:28:16 | FromGitter | <zacharycarter> I'd use the compiler pre-processor to process them |
| 01:28:32 | FromGitter | <zacharycarter> I think that's your problem tbh |
| 01:30:10 | FromGitter | <zacharycarter> I ran into this issue when originally trying to write the bindings for nuklear |
| 01:30:50 | FromGitter | <zacharycarter> the solution isn't to try to reproduce the macro / code in Nim - it's to get the C pre-processor to spit out the C code so that you can create Nim bindings to it |
| 01:31:14 | FromGitter | <zacharycarter> if that makes any sense AlexMax ^ |
| 01:35:27 | FromGitter | <zacharycarter> the best approach to writing bindings (in a manual way) is this - ⏎ ⏎ 1) Find the header file that is the source of the library and start from there - run this through c2nim and see where things start erroring out. ⏎ ⏎ 2) Most likely it's going to be due to some C construct Nim doesn't support - so start commenting these out. ... [https://gitter.im/nim-lang/Nim?at=5bac33dfaedb375b9c514fb9] |
| 01:39:14 | AlexMax | I think you think I'm further behind the eight-ball than I am :P |
| 01:39:52 | AlexMax | I'm fixing the nimgen'ed bindings |
| 01:39:58 | AlexMax | It already runs the preprocessor on it |
| 01:40:36 | AlexMax | And like I said, the macro disappears because when you run the preprocessor and define a "function" macro with no calls to it, there's no actual code to convert |
| 01:41:03 | AlexMax | And inserting a call to convert would be useless because it would just be whatever __FILE__ and __LINE__ evaluate to at that point |
| 01:41:47 | FromGitter | <zacharycarter> yeah - I guess I don't understand the problem - I didn't mean any offense |
| 01:42:04 | AlexMax | No worries. |
| 01:42:05 | FromGitter | <zacharycarter> or like - to insinuate you didn't know what you were doing - sorry if it came off that way |
| 01:42:30 | AlexMax | I didn't take it like that at all. There was just a clear miscommunication and I wanted to clear the air. :) |
| 01:42:31 | * | darithorn quit (Quit: Konversation terminated!) |
| 01:42:33 | FromGitter | <zacharycarter> maybe it's a c2nim problem then? |
| 01:42:49 | FromGitter | <zacharycarter> I guess the question is - what's generating the bad macro code? |
| 01:42:52 | AlexMax | It's a byproduct of nimgen's running the preprocessor |
| 01:42:56 | FromGitter | <zacharycarter> ah |
| 01:43:37 | FromGitter | <zacharycarter> so probably Nimgen needs to incorporate some more clean-up logic after the preprocessor step |
| 01:43:37 | AlexMax | The macro function vanishes |
| 01:43:45 | FromGitter | <zacharycarter> which makes sense |
| 01:43:53 | FromGitter | <zacharycarter> since it'd only be there at compile-time |
| 01:43:58 | FromGitter | <zacharycarter> and if nimgen's not evaluating it |
| 01:44:06 | AlexMax | It's a catch 22 |
| 01:44:14 | FromGitter | <zacharycarter> yeah - I see now |
| 01:44:15 | AlexMax | It needs to run the preprocessor in order to sort out some things |
| 01:44:32 | AlexMax | but it also makes macro functions vanish |
| 01:44:40 | AlexMax | I think the only option is a reimplementation |
| 01:47:33 | AlexMax | It needs some way to get a guaranteed unique string, along with a guaranteed unique number |
| 01:48:37 | AlexMax | What's the intended way to create a multiline if statement? |
| 01:48:59 | FromGitter | <zacharycarter> do you mean via a Nim macro? |
| 01:49:29 | FromGitter | <zacharycarter> also - is there an issue or any reference I can look at to further understand this problem? |
| 01:49:32 | AlexMax | nah, i just mean how do people use whitespace on multiline if statements |
| 01:49:40 | AlexMax | it's a separate question |
| 01:49:46 | FromGitter | <zacharycarter> oh |
| 01:49:55 | FromGitter | <zacharycarter> typically - I do something like this: |
| 01:50:03 | AlexMax | it's been bugging me for a while |
| 01:50:37 | AlexMax | offsetting the second line of a two line if statement by two spaces looks like it melds into the conditional block itself |
| 01:50:43 | * | hassan_ joined #nim |
| 01:50:49 | AlexMax | Four spaces is...eh |
| 01:50:53 | AlexMax | I was thinking three spaces |
| 01:51:15 | AlexMax | it's not aligned with anything, so it should stick out more |
| 01:52:03 | AlexMax | And if you want to understand the macro problem I'm having, take a look at this |
| 01:52:05 | AlexMax | https://github.com/vurtun/nuklear/blob/master/src/nuklear.h#L2552 |
| 01:52:12 | FromGitter | <zacharycarter> I don't know - it's typically up to the individual programmer ⏎ all I know is whenever I encounter libraries that don't rely two-spaces, I want to pull my hair out |
| 01:52:38 | * | Tef61 joined #nim |
| 01:52:51 | AlexMax | Well, I indent everything by two spaces |
| 01:53:05 | AlexMax | But I was talking about how many spaces to indent the second line of a multi-line if statement by |
| 01:53:32 | FromGitter | <zacharycarter> I ported this in three(ish) days - https://github.com/zacharycarter/litz/blob/master/src/litz.nim |
| 01:53:33 | AlexMax | indenting _that_ by two makes the second line of the if statement blend in with the conditional block below it |
| 01:53:44 | FromGitter | <zacharycarter> so that's how I write Nim code when I'm just in a groove I guess |
| 01:54:02 | FromGitter | <zacharycarter> I get what you're saying - but my response is - every Nim programmer has their own way of doing it |
| 01:54:15 | FromGitter | <zacharycarter> and lining up the next line of code |
| 01:54:24 | FromGitter | <zacharycarter> some people - indent it by four spaces for some reason |
| 01:55:23 | FromGitter | <zacharycarter> others try to line it up parameter wise - I'm not sure there is a real existing standard for this - I know there's a Nim style-guide but I don't think it's really up-to-date or followed |
| 01:55:28 | * | hassan_ quit (Remote host closed the connection) |
| 01:56:23 | FromGitter | <zacharycarter> also - that is a port straight from JS |
| 01:57:24 | * | Tef61 quit (Remote host closed the connection) |
| 01:57:50 | AlexMax | hrm |
| 01:57:54 | AlexMax | https://nim-lang.org/docs/oids.html |
| 01:58:04 | AlexMax | this could work, possibly |
| 01:58:18 | FromGitter | <zacharycarter> I don't think you'll find too much difference from this code I wrote a while back either though - https://github.com/zacharycarter/zengine/blob/master/src/zengine/zgl.nim - in terms of styling |
| 01:59:47 | FromGitter | <zacharycarter> I think you can line up the code any way that makes sense for you |
| 02:00:29 | AlexMax | hrm, damn, except that isn't consistent from function call to function call |
| 02:00:32 | FromGitter | <zacharycarter> https://github.com/flyx/emerald/blob/master/src/emerald/filters.nim |
| 02:00:35 | FromGitter | <zacharycarter> drives me crazy |
| 02:01:07 | FromGitter | <zacharycarter> well - these are mostly ports of libraries that I never went back and spent a lot of time cleaning up |
| 02:01:13 | FromGitter | <zacharycarter> zenigne = raylib |
| 02:01:25 | FromGitter | <zacharycarter> and the other code is hyperHTML |
| 02:01:34 | FromGitter | <zacharycarter> which I plan to actually turn into a usable library |
| 02:01:47 | AlexMax | damn son |
| 02:02:02 | FromGitter | <zacharycarter> but I'm just giving you the basic gist of how, when I'm porting / writing a large library - I start out structuring my code |
| 02:02:22 | FromGitter | <zacharycarter> I write a TON of throw-away Nim code |
| 02:02:47 | FromGitter | <zacharycarter> but there's no other language I'd rather program in - it's the only language I feel okay about throwing away code in |
| 02:03:03 | AlexMax | I...wish Nim had destructors as part of the type |
| 02:03:12 | AlexMax | otherwise, yeah, this langauge is pretty fun |
| 02:03:18 | FromGitter | <zacharycarter> other languages make me take way to long to think about how to structure / write my code or worry about memory allocation / deallocation |
| 02:03:31 | FromGitter | <zacharycarter> they're coming - according to our BDFL anyway |
| 02:03:55 | leorize | There's even an issue to track it, and they are already somewhat working |
| 02:03:59 | FromGitter | <zacharycarter> most of my code is just hobby shit anyway - I don't need it to work in any professional sense |
| 02:04:00 | AlexMax | I don't like how right now you gotta remember to pass the finalizer as part of `new` |
| 02:04:15 | AlexMax | but yeah, I saw the work on destructors n stuff |
| 02:04:20 | FromGitter | <zacharycarter> well get in on the RFC then! |
| 02:04:23 | AlexMax | But I thought it was a ways off... |
| 02:04:29 | AlexMax | like, Nim 2.0 land |
| 02:04:40 | leorize | it's in v1 milestone so... |
| 02:04:45 | FromGitter | <zacharycarter> I have no idea - ever - what is going on with this language from a future perspective |
| 02:04:54 | FromGitter | <zacharycarter> there are constantly too many things up in the air |
| 02:05:05 | * | redlegion_ joined #nim |
| 02:05:19 | FromGitter | <zacharycarter> but I know that I can A) write a front-end web app with Nim and generally keep up with the JS backend changes to not worry about it being a big deal |
| 02:05:45 | FromGitter | <zacharycarter> and B) at the same time - experiment with game programming and learn lower level programming concepts via Nim - since I come from a zero comp-sci background |
| 02:06:02 | AlexMax | i like this language because it's not C++-levels of complexity |
| 02:06:06 | FromGitter | <zacharycarter> so for me - Nim is THE perfect language - except for the fact it leads to zero job opportunities in the space |
| 02:06:15 | FromGitter | <zacharycarter> yeah - C++ just infuriates me |
| 02:06:18 | AlexMax | but it's also not C's level of lack of complexity |
| 02:06:21 | FromGitter | <zacharycarter> I'd rather program in C every day |
| 02:06:34 | AlexMax | well, maybe it's not even complexity, it's just the endless minutea you gotta know |
| 02:06:41 | FromGitter | <zacharycarter> yeah |
| 02:06:52 | FromGitter | <zacharycarter> not that the language isn't designed well |
| 02:06:56 | FromGitter | <zacharycarter> froma . feature perspective |
| 02:07:19 | FromGitter | <zacharycarter> it's just - my god - I don't want to deal with another JavaScript in the native programming world |
| 02:07:33 | FromGitter | <zacharycarter> I feel like I'm dealing w/ babel and ECMAScript these days when it comes to C++ |
| 02:07:54 | FromGitter | <zacharycarter> regardless of the fact that the code actually matters |
| 02:08:26 | FromGitter | <zacharycarter> just in terms of having to learn different specifications and then on top of it worry about platform compatability |
| 02:09:31 | FromGitter | <zacharycarter> seems like the pattern these days though - there's like Rust 2015 or something ow |
| 02:09:33 | FromGitter | <zacharycarter> now* |
| 02:09:41 | FromGitter | <zacharycarter> and then a new dated spec of Rust coming out soon |
| 02:10:15 | FromGitter | <zacharycarter> https://blog.rust-lang.org/2018/07/27/what-is-rust-2018.html |
| 02:10:34 | FromGitter | <zacharycarter> hey - if we just do a Nim 2018 - we never need a Nim 1.0 |
| 02:11:12 | AlexMax | I don't want the mental overhead Rust would bring |
| 02:11:36 | FromGitter | <zacharycarter> oh - I'm just saying from a versioning perspective :P |
| 02:11:43 | FromGitter | <zacharycarter> we can just cheat |
| 02:12:54 | FromGitter | <zacharycarter> I mean really - wtf - how corporate can you get from a language perspective beyond releasing yearly candidates of your softare |
| 02:12:58 | FromGitter | <zacharycarter> software* |
| 02:13:18 | FromGitter | <zacharycarter> I can only see things going into supported releases and enterprise support etc |
| 02:14:27 | AlexMax | Well, hoenstly, I wish Nim had that kind of corporate support |
| 02:14:36 | AlexMax | of a Mozilla or Google |
| 02:15:29 | FromGitter | <zacharycarter> I think we all do |
| 02:15:45 | FromGitter | <zacharycarter> but - we're attracted to Nim for some reason or another, and it factually lacks all of that |
| 02:16:24 | shashlick | Looks like I missed some nimgen fun |
| 02:16:30 | FromGitter | <zacharycarter> so IMO - Nim is worth exploring, evangelizing, improving, and donating to |
| 02:16:56 | FromGitter | <zacharycarter> and hopefully one day, it becomes the better alternative to Rust / C++ |
| 02:17:21 | FromGitter | <zacharycarter> but for me - it's already better than TypeScript |
| 02:20:28 | shashlick | @AlexMax: the #defines basically implement function shortcuts - no way to get those into nim without reimplement |
| 02:21:37 | AlexMax | shashlick: Exactly my suspicion. ;) |
| 02:21:55 | AlexMax | Only trouble is, what is __FILE__ and __LINE__ in nim? |
| 02:22:11 | AlexMax | Right now I'm progressing just fine by creating a unique string per invocation of nuklear.tree_push_hashed |
| 02:22:27 | * | redlegion_ quit (Ping timeout: 240 seconds) |
| 02:22:28 | AlexMax | but it'd be nice to turn it into a plain old function |
| 02:25:07 | FromGitter | <achou11> Has anyone run into this issue when drying to do a simple `nim c -r filename.nim`? ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bac3f83fbe23c730fa64886] |
| 02:25:33 | FromGitter | <rayman22201> @AlexMax https://nim-lang.org/docs/system.html#instantiationInfo |
| 02:26:09 | leorize | achou11: What OS are you running? |
| 02:27:12 | FromGitter | <achou11> @FromIRC just updated my post |
| 02:28:03 | leorize | you would have to type my name out if you wanted to tag me :) |
| 02:28:22 | FromGitter | <achou11> @leorize my bad haha |
| 02:29:30 | FromGitter | <achou11> I found this (https://stackoverflow.com/questions/19580758/gcc-fatal-error-stdio-h-no-such-file-or-directory) and tried the command but it didn’t help |
| 02:30:57 | * | craigger joined #nim |
| 02:31:14 | FromGitter | <zacharycarter> @achou11 - you're going to have to tell the Nim compiler what gcc version you want to use - if you want to override the default |
| 02:31:23 | FromGitter | <zacharycarter> funny thing is - clang = gcc on osx |
| 02:31:33 | FromGitter | <zacharycarter> because Apple loves to eff with things |
| 02:32:19 | FromGitter | <zacharycarter> soooo |
| 02:33:19 | FromGitter | <achou11> @zacharycarter what does that flag look like? |
| 02:33:34 | FromGitter | <zacharycarter> one sec - trying to find it now |
| 02:34:00 | FromGitter | <achou11> also I never had to do it before, so I’m wondering if all of these updates came back to bite me haha |
| 02:36:11 | FromGitter | <zacharycarter> @achou11 - check out - https://github.com/zacharycarter/zapp/blob/master/nim.cfg |
| 02:36:28 | FromGitter | <zacharycarter> maybe clang / llvm had an update? can always refer to Nim release notes |
| 02:36:40 | FromGitter | <zacharycarter> some things that gcc supports / llvm doesn't |
| 02:37:07 | FromGitter | <zacharycarter> I've found myself having to install gcc on my osx machine before and switching the compiler to use the gcc binary |
| 02:37:30 | * | craigger quit (Quit: bye) |
| 02:37:40 | FromGitter | <zacharycarter> for instance - llvm doesn't support all of the intrinsics gcc does - certain atomic operations that work on gcc won't work on clang |
| 02:38:24 | shashlick | @AlexMax: that function is only using __LINE__ as a unique seed |
| 02:38:30 | shashlick | you can set it to some other random value |
| 02:38:48 | FromGitter | <zacharycarter> this section probably explains it better than I can - https://nim-lang.org/docs/nimc.html#cross-compilation |
| 02:38:59 | AlexMax | shashlick: Yeah, but I wonder if there's a way to do that automatically |
| 02:39:07 | AlexMax | instead of having to define a cstring ahead of time like I'm doing now |
| 02:39:09 | FromGitter | <zacharycarter> and then of course - just search on github for examples if you need more specifics |
| 02:39:19 | AlexMax | also I'm having a bizarre problem |
| 02:39:22 | AlexMax | I can't index an array |
| 02:40:06 | AlexMax | code that works fine in the playgournd is totally broken in the context of the nimgenned nuklear library |
| 02:40:24 | FromGitter | <zacharycarter> pg is old |
| 02:40:34 | FromGitter | <zacharycarter> like 18.0 I think? |
| 02:40:35 | AlexMax | It specifically takes issue with the index operator, saying "type expected" |
| 02:40:43 | FromGitter | <zacharycarter> maybe older |
| 02:40:52 | FromGitter | <achou11> @zacharycarter thanks for the help! I see that nim 0.19.0 was just released, so maybe it has fixes for this? |
| 02:41:21 | AlexMax | Oh man I'm an idiot |
| 02:41:25 | shashlick | you can use instantiationInfo() |
| 02:41:26 | FromGitter | <zacharycarter> @achou11 I wouldn't call what you're describing a bug - it's a fact of software |
| 02:41:27 | shashlick | https://forum.nim-lang.org/t/3221 |
| 02:41:34 | AlexMax | i was using : by accident |
| 02:41:36 | AlexMax | instead of = |
| 02:41:38 | AlexMax | lollll |
| 02:41:49 | AlexMax | what |
| 02:41:51 | AlexMax | 0.19 is out? |
| 02:41:51 | * | Taylor21 joined #nim |
| 02:41:54 | shashlick | bung87: I'm testing on mac and I see the issue |
| 02:41:58 | * | Taylor21 quit (Remote host closed the connection) |
| 02:42:00 | shashlick | ya just updated and I see 0.19.0 |
| 02:42:16 | FromGitter | <achou11> also how do I update to latest nim version with choosenim? |
| 02:42:22 | AlexMax | no blog post about it? |
| 02:42:24 | FromGitter | <zacharycarter> @achou11 - gcc and llvm are totally out of Nim's control - Nim allows you to override the executable you use to compile the C code it produces |
| 02:42:27 | FromGitter | <zacharycarter> that's pretty much it |
| 02:42:51 | shashlick | @AlexMax: see https://forum.nim-lang.org/t/3221 for getting file info at compile time |
| 02:42:59 | FromGitter | <zacharycarter> @achou11 - I believe the command is `choosenim update devel` but you can refer to `choosenim --help` for specifics |
| 02:43:16 | FromGitter | <achou11> @zacharycarter i see. I guess I’ll have to dig then. Thanks much! |
| 02:43:27 | FromGitter | <zacharycarter> np! |
| 02:43:44 | AlexMax | shashlick: nice |
| 02:46:01 | shashlick | bung87: please try compiling now - should work correctly with clang itself |
| 02:46:31 | shashlick | you needed -std=c++0x for clang++ to understand the new enum syntax |
| 02:48:04 | FromGitter | <zacharycarter> @achou11 feel free to post a gist / whatever summarizing your problem in here - it's just difficult to guess what your issue might be regarding compiler implementations and their inconsistencies - if you post your entire project on some public SCM solution, and then a public link with the error message - that's even better, as people can try to compile it themselves (I have a mac book pro that I generally develop |
| 02:48:04 | FromGitter | ... on Nim with) |
| 02:48:22 | FromGitter | <achou11> @zacharycarter just tried specifying the compiler and it worked, even if I specify the default one 🙄 |
| 02:48:33 | * | craigger joined #nim |
| 02:48:54 | FromGitter | <zacharycarter> do you have xcode dev tools - or whatever - installed? |
| 02:49:05 | FromGitter | <achou11> Yeah I’m pretty sure I do |
| 02:49:26 | leorize | AlexMax: since Araq haven't merged back to master, I would assume the release is still in unofficial state, hence no blog post (yet) |
| 02:49:45 | FromGitter | <zacharycarter> please keep in mind - apple likes to do everything differently from everyone else - and at the same time make its dev users live's hell |
| 02:49:51 | FromGitter | <achou11> Is that what `xcode-select —install` does? |
| 02:49:57 | FromGitter | <zacharycarter> yup |
| 02:50:02 | FromGitter | <achou11> damn |
| 02:50:32 | FromGitter | <zacharycarter> if you're working on apple specific software - life is fine |
| 02:50:38 | FromGitter | <zacharycarter> use XCode and that whole eco-system |
| 02:51:29 | FromGitter | <zacharycarter> if you want to go cross-platform with a OSX as your dev machine - it's not quite as simple |
| 02:52:21 | FromGitter | <achou11> where would I find the config settings with the defaults? |
| 02:52:33 | FromGitter | <zacharycarter> heh there are none |
| 02:52:38 | FromGitter | <zacharycarter> apple just overrides things |
| 02:52:44 | FromGitter | <zacharycarter> like try typing this into your terminal |
| 02:52:46 | FromGitter | <achou11> for nim? |
| 02:53:15 | FromGitter | <zacharycarter> `gcc --version` |
| 02:53:36 | FromGitter | <achou11> I get this: ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bac462ffbe23c730fa6488a] |
| 02:53:36 | FromGitter | <zacharycarter> one sec - I'll point you to Nim's |
| 02:53:47 | FromGitter | <zacharycarter> yeah |
| 02:53:50 | FromGitter | <zacharycarter> notice how gcc |
| 02:53:53 | FromGitter | <zacharycarter> points to clang |
| 02:53:55 | FromGitter | <zacharycarter> :) |
| 02:54:20 | FromGitter | <zacharycarter> a totally different compiler frontend (LLVM) |
| 02:54:32 | FromGitter | <zacharycarter> gotta love apple |
| 02:54:37 | FromGitter | <achou11> yeah I see that. So which one is nim using??? |
| 02:54:50 | FromGitter | <zacharycarter> https://github.com/nim-lang/Nim/blob/devel/config/nim.cfg |
| 02:54:56 | FromGitter | <zacharycarter> Nim will be using your OS's default I believe |
| 02:55:09 | FromGitter | <zacharycarter> if you want Nim to use actual gcc - you need to specify it in a config file |
| 02:55:29 | FromGitter | <zacharycarter> like here - https://nim-lang.org/docs/nimc.html#cross-compilation |
| 02:55:45 | FromGitter | <zacharycarter> so you can use brew to install bcc |
| 02:56:30 | FromGitter | <zacharycarter> and then link to it in your project with a cfg file - or override it with a more global cfg file somewhere - I forget where you put those - but the manual outlines that as well |
| 02:57:43 | FromGitter | <achou11> hmm makes sense |
| 02:57:56 | FromGitter | <zacharycarter> I think if you just put a .cfg file in your root project dir with `gcc.exe=whatever` it will work |
| 02:58:31 | FromGitter | <zacharycarter> but by default it's just going to invoke gcc on your system |
| 02:58:34 | FromGitter | <zacharycarter> and if that points to clang |
| 02:58:44 | leorize | you can put a global config file in ~/.config/nim/nim.cfg |
| 02:59:01 | FromGitter | <zacharycarter> you'll run into whatever compiler inconsistencies exisit - but this is how C++ started out too - so... |
| 02:59:10 | FromGitter | <zacharycarter> C++ started out by compiling to C |
| 02:59:13 | FromGitter | <zacharycarter> you gotta start somewhere |
| 02:59:32 | leorize | achou11: or just `nim c` something, the path to the configs will appear on top |
| 03:00:09 | FromGitter | <achou11> leorize I see that now. Very helpful :) |
| 03:00:27 | FromGitter | <achou11> @zacharycarter is it a problem if I have 2 paths for C?? |
| 03:00:42 | AlexMax | GNU Nim in our future? |
| 03:00:43 | AlexMax | :P |
| 03:01:00 | leorize | gnc - GNU Nim Compiler :P |
| 03:01:30 | FromGitter | <zacharycarter> @achou11 what do you mean two paths for C? |
| 03:01:33 | FromGitter | <zacharycarter> clang? |
| 03:01:48 | FromGitter | <zacharycarter> or two entries to a C compiler on your path? |
| 03:01:57 | * | Haydos23 joined #nim |
| 03:02:03 | FromGitter | <zacharycarter> as long as you're not overriding the default gcc symlink to clang on your system |
| 03:02:06 | FromGitter | <zacharycarter> you should be fine |
| 03:02:07 | * | Haydos23 quit (Remote host closed the connection) |
| 03:02:19 | FromGitter | <zacharycarter> generally you'll want to install gcc like `brew install gcc-8` or whatever |
| 03:02:35 | FromGitter | <zacharycarter> and then in your .cfg file - provide a path to wherever gcc-8 gets installed |
| 03:02:49 | FromGitter | <zacharycarter> like - `/usr/local/bin/gcc-8` |
| 03:03:01 | FromGitter | <achou11> yeah. So I have these two paths: ⏎ ⏎ ```clang is /usr/local/bin/clang ⏎ clang is /usr/bin/clang``` ⏎ ⏎ The second one is obviously the system one. The first seems to be the one that was installed with my LLVM homebrew installation (maybe???) [https://gitter.im/nim-lang/Nim?at=5bac4865fbe23c730fa6488c] |
| 03:03:13 | FromGitter | <zacharycarter> don't muck with the default paths / symlinks that OSX sets up for you |
| 03:03:34 | FromGitter | <zacharycarter> hrm - possibly |
| 03:03:59 | FromGitter | <achou11> and runnning clang points to the first one. not sure if that would be a problem for nim though |
| 03:04:53 | FromGitter | <zacharycarter> not sure either - I generally just use whatever version of clang is installed w/ OSX |
| 03:05:08 | FromGitter | <zacharycarter> and then if I need GCC, or another compiler backend I link to it |
| 03:05:18 | FromGitter | <zacharycarter> I've never tried to point to a different version of clang on OSX |
| 03:05:29 | FromGitter | <achou11> I didn’t either. I have no idea how this happened tbh |
| 03:05:49 | FromGitter | <zacharycarter> but I mean - I think it's all about just making sure the clang.exe / gcc.exe whatever points to the right location |
| 03:06:00 | FromGitter | <zacharycarter> and that location resolves (whether it's a symlink or whatever) to the correct executable |
| 03:06:22 | FromGitter | <zacharycarter> you can even get so specific w/ the compiler settings as to override the linkker |
| 03:06:27 | FromGitter | <zacharycarter> etc |
| 03:08:13 | FromGitter | <achou11> lol |
| 03:08:25 | * | skellock quit (Quit: WeeChat 1.9.1) |
| 03:08:31 | FromGitter | <achou11> so I just removed that homebrew installed clang binary from my path, and now nim is working again |
| 03:08:41 | FromGitter | <achou11> i hate computers |
| 03:08:55 | FromGitter | <zacharycarter> :) sounds about right with OSX |
| 03:08:56 | FromGitter | <achou11> i’ll be needing that table-flipping meme |
| 03:09:07 | FromGitter | <zacharycarter> please don't blame this on computers in general |
| 03:09:15 | FromGitter | <achou11> fair |
| 03:09:20 | FromGitter | <zacharycarter> just blame it on apple - the fact that they're calling clang gcc |
| 03:09:24 | FromGitter | <achou11> my only experience with computers for dev stuff is osx |
| 03:09:26 | FromGitter | <zacharycarter> is beyond stupid in the first place |
| 03:09:44 | FromGitter | <zacharycarter> well - in case you're interested in why this is so confounding |
| 03:09:49 | FromGitter | <achou11> i don’t know enough about C and compilers to understand |
| 03:09:57 | FromGitter | <zacharycarter> gcc = https://gcc.gnu.org/ |
| 03:10:14 | FromGitter | <achou11> ah |
| 03:10:18 | FromGitter | <achou11> i had an idea |
| 03:10:21 | FromGitter | <zacharycarter> llvm / clang = https://clang.llvm.org/ & https://llvm.org/ |
| 03:10:31 | FromGitter | <achou11> but when it comes to troubleshooting, I’m a sitting duck |
| 03:10:41 | FromGitter | <zacharycarter> and the latter - of course - is spear headed by apple |
| 03:11:40 | FromGitter | <achou11> so does this mean i have two installations of Clang on my computer?? |
| 03:12:03 | FromGitter | <achou11> one from system and the other from LLVM via Homebrew? |
| 03:12:22 | FromGitter | <zacharycarter> well if you installed any other version of clang on your computer, via homebrew - then yes |
| 03:12:56 | FromGitter | <achou11> daaahh |
| 03:12:59 | FromGitter | <zacharycarter> OSX comes with a default installation of llvm / clang because that compiler infrastructure is relied upon for building a lot of the internals the OS needs |
| 03:13:20 | FromGitter | <zacharycarter> if you install another version on top of that via a package manager - the default OS installation will likely get confused |
| 03:13:25 | FromGitter | <zacharycarter> which is probably what led to your issue |
| 03:13:26 | FromGitter | <achou11> yeah |
| 03:13:35 | FromGitter | <achou11> happens to me with other homebrew stuff |
| 03:13:49 | FromGitter | <zacharycarter> yeah - every modern operating system is a mess tbh |
| 03:13:53 | FromGitter | <zacharycarter> there are no great options |
| 03:14:10 | FromGitter | <zacharycarter> windows takes longer to update than a fresh install |
| 03:14:57 | FromGitter | <zacharycarter> almost every linux distro is a security mess and plus - linux itself has become extremely bloated over time and filled with unnecessary cruft |
| 03:15:08 | FromGitter | <zacharycarter> OSX is... OSX |
| 03:15:21 | FromGitter | <achou11> ah geez. I was thinking of switching to a Linux distro sometime soon :/// |
| 03:15:38 | FromGitter | <zacharycarter> well - certain linux distros are worse than others |
| 03:15:47 | FromGitter | <zacharycarter> would probably pay off to ask for advice in that realm |
| 03:16:14 | FromGitter | <zacharycarter> I just use what my work gives me for free - if I actually had to choose my own OS - I'd probably write my own and then never end up having access to a usable computer ever again :P |
| 03:17:28 | FromGitter | <achou11> what’s funny about all of this is that I’m *obviously* new to Nim and I was planning to go through the 2nd part of the official guide for funsies |
| 03:17:36 | FromGitter | <achou11> and then I ended up with this mess |
| 03:18:00 | FromGitter | <achou11> and what’s even funnier is that I’m used to it hahaha |
| 03:18:04 | FromGitter | <zacharycarter> well - this is kind of the bridge I try to inform my co-workers of regarding native vs non-native programming |
| 03:18:13 | FromGitter | <zacharycarter> and most of them don't get it at all |
| 03:18:56 | FromGitter | <achou11> Yeah I usually don’t have to deal with very low-level issues on my computer. Tonight was a little new for me, that’s for sure |
| 03:19:22 | FromGitter | <zacharycarter> you're not writing JS or HTMl that goes through an engine - you're writing code and saving it to a source file and then that's going through a compiler that has several phases to it, along with a linker, and other processes that you're probably not used to. |
| 03:19:33 | FromGitter | <achou11> Like once I see Clang in the error message, I start sweating lol |
| 03:19:35 | FromGitter | <zacharycarter> Best thing to do - is learn C |
| 03:19:35 | FromGitter | <achou11> yeah definitely |
| 03:19:46 | FromGitter | <zacharycarter> or at least get a semi-solid understanding of it |
| 03:20:12 | FromGitter | <achou11> one thing at a time. would be nice to have a general knowledge of that. at least learn some of the terminology |
| 03:20:32 | FromGitter | <zacharycarter> well - I think that C is a good bridge to Nim |
| 03:20:37 | FromGitter | <zacharycarter> and any other systems programming language |
| 03:20:48 | FromGitter | <zacharycarter> especially if you're starting from a high level of understanding regarding programming |
| 03:20:58 | FromGitter | <zacharycarter> it's the complete opposite of the spectrum |
| 03:21:14 | FromGitter | <zacharycarter> without diving into assembly / circuitry and logic gates and all of that |
| 03:21:38 | FromGitter | <zacharycarter> https://www.amazon.com/Programming-Language-2nd-Brian-Kernighan/dp/0131103628 - is probably a book every person studying comp-sci should ready anyway |
| 03:22:03 | AlexMax | ....this is weird |
| 03:22:04 | AlexMax | var x: array[3, cint] = [1'i32, 2, 3] |
| 03:22:08 | AlexMax | why does this compile? |
| 03:22:24 | AlexMax | Only the first literal is annotated as a 32-bit integer |
| 03:22:28 | FromGitter | <zacharycarter> and then if you really want to understand how computers / programs work there is - https://www.amazon.com/Code-Language-Computer-Hardware-Software/dp/0735611319 |
| 03:22:36 | AlexMax | if you remove that annotation, the compilation fails |
| 03:22:44 | FromGitter | <zacharycarter> and then from there you can move onto some book about a particular machine / instruction set |
| 03:22:51 | AlexMax | because you can't convert a literal to a cint |
| 03:23:10 | AlexMax | But shouldn't I be annotating EVERY literal? |
| 03:23:23 | AlexMax | var x: array[3, cint] = [1'i32, 2'i32, 3'i32] |
| 03:24:04 | FromGitter | <achou11> Yeah I’ve heard great things about that book. I wasn’t forced to do the low-level stuff in school since I wasn’t a CS major, so definitely started high-level. I’m a few weeks from starting my first full-time job, so I’ll have to put the low-level deep-dive on hold for now |
| 03:24:33 | FromGitter | <achou11> Seriously thanks for all of the help and conversation @zacharycarter ! Really appreciate it |
| 03:25:34 | FromGitter | <zacharycarter> like - https://www.amazon.com/Computer-Organization-Design-ARM-Architecture/dp/0128017333 |
| 03:25:37 | FromGitter | <zacharycarter> yeah - np |
| 03:25:54 | FromGitter | <zacharycarter> that's another good book once you read the code book |
| 03:26:28 | FromGitter | <zacharycarter> @achou11 - just FYI - I started at a high level and now am having to work through a lot of adversity getting to a lower level / more challenging position |
| 03:27:26 | FromGitter | <zacharycarter> I was a history major in college and am tired of doing the web dev thing (started by teaching myself HTMl / JS / CSS when I was like 12 - and then taught myself basic programming, and yadda yadda yadda - now my job is boring AF and my peers are all like me when I was just starting out / there's no where for me to go) |
| 03:28:04 | FromGitter | <zacharycarter> so - just be wary of the skillset you go after - sometimes learning the harder stuff / lower-level stuff even though it's not immediately practical, is the way to go |
| 03:28:16 | FromGitter | <zacharycarter> especially if you are passionate about software / not just web dev |
| 03:28:42 | FromGitter | <zacharycarter> but either way - good luck! I wish you the best and hope you get to use Nim in your pursuits! |
| 03:28:52 | FromGitter | <achou11> @zacharycarter Thanks for the book links! I’ll check them out some time in the near future. ⏎ ⏎ I definitely understand what you’re talking about with the whole web dev thing. While I really like the web dev stuff, I find the lower-level topics fascinating and always wished that I devoted time to doing that stuff first, whether on my own time or in school |
| 03:30:45 | FromGitter | <achou11> Hopefully I’ll make more progress with learning Nim after this snafu hahaha. I find that the ease of learning is great, especially from a Python background. |
| 03:32:02 | FromGitter | <zacharycarter> I'm sure you will - and trust me - there will be plenty more head scratchers if you stick around and use Nim :) I think I ran into a similar problem with GCC / Clang inconsistencies regarding atomic operations only a few months ago - you're doing great! :P |
| 03:33:21 | FromGitter | <zacharycarter> and the fact that you find the lower-level stuff fascinating and are drawn to it is great! (not that I can speak from being one of them) - but I think those with that kind of curiosity and dedication make the best engineers - they don't get complacent after they learn a framework and can pump out the same app over and over with it |
| 03:33:40 | FromGitter | <zacharycarter> which is what I find the goal of a lot of software orgs is these days |
| 03:36:16 | FromGitter | <zacharycarter> also - this is, IMO, a fantastic read on why being a software dev (who cares) in enterprise IT today - sucks: http://tonsky.me/blog/disenchantment/ |
| 03:39:20 | * | darithorn joined #nim |
| 03:51:18 | * | chemist69 quit (Ping timeout: 252 seconds) |
| 03:53:02 | * | chemist69 joined #nim |
| 04:05:13 | * | zachcarter quit (Quit: Lost terminal) |
| 04:22:50 | FromGitter | <zacharycarter> http://skiplang.com/ |
| 04:24:52 | FromGitter | <zacharycarter> https://github.com/skiplang/skip/blob/master/src/native/coroutine.sk |
| 04:25:15 | FromGitter | <zacharycarter> pretty interesting |
| 04:29:35 | FromGitter | <zacharycarter> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bac5caf7bd81c5b9dbd61e4] |
| 04:33:01 | FromGitter | <zacharycarter> haha - beat me to it of course - https://news.ycombinator.com/item?id=18080755 |
| 04:51:47 | FromGitter | <zacharycarter> how is one to manage dependencies w/o nimble? |
| 04:51:49 | FromGitter | <zacharycarter> nawabs? |
| 04:54:52 | FromGitter | <zacharycarter> I'm going to give it a shot - nimble is starting to annoy me as well in certain circumstances - in others it's fine |
| 05:10:16 | * | cswang10 joined #nim |
| 05:12:30 | FromGitter | <kayabaNerve> @zacharycarter What about Parabola (Libre Arch) with hardening ? |
| 05:13:03 | FromGitter | <zacharycarter> I haven't tried it |
| 05:13:33 | FromGitter | <kayabaNerve> They have a couple of wiki pages on making it more secure. It's fully open and Arch is pretty minimal. |
| 05:14:35 | * | cswang10 quit (Remote host closed the connection) |
| 05:14:46 | FromGitter | <zacharycarter> I'll have to check it out |
| 05:16:25 | FromGitter | <kayabaNerve> I have never installed Arch; I know for security, Hardened Gentoo is what most people go to. That said, for the three criteria you said, since usability and user experience wasn't one, Parabola is likely your best bet. |
| 05:18:17 | ldlework | NixOS is my favorite these days |
| 05:18:26 | ldlework | truly distinct linux distribution |
| 05:18:38 | FromGitter | <kayabaNerve> "distinct" ⏎ ⏎ "linux" |
| 05:19:31 | ldlework | Well, it and Guix |
| 05:20:16 | FromGitter | <kayabaNerve> I like Ubuntu Mate. |
| 05:20:47 | ldlework | None of those distros are very different. Different package repos, different update schedules, different package manager which all basically work the same way |
| 05:22:05 | ldlework | Cryptographically assured package reporducability using a single config language for the whole linux system, X, firewalls, boot options, services, everything - can stick my whole config in git and know that if my machine crashes I can reproduce my whole environment no problem. |
| 05:22:25 | ldlework | If I build a configuration one time that does the wrong thing, and say my window manager wont start, i can just roll back to the last generation in grub. |
| 05:23:16 | ldlework | the configuration language is almost a real language so you can make your config from modular abstractions and reusable bits that you can share with others. |
| 05:24:21 | FromGitter | <kayabaNerve> Parabola is Arch, a minimal distro that uses pacman, with no binary blobs and only free software in its package repos. ⏎ ⏎ I like Nix as a package manager but I question the idea of a whole OS around it. That said, the main thing between distros is the package manager/used Desktop... |
| 05:24:57 | ldlework | I just highlighted just a few of the great benefits of having an OS utilize the technology in Nix |
| 05:25:19 | ldlework | But you're right, another great thing about the ecosystem is it can be used on any linux distro and even OSX |
| 05:26:00 | FromGitter | <kayabaNerve> Yes but that alone doesn't make it an OS. |
| 05:26:16 | ldlework | "that alone"? what alone doesn't make what an OS? |
| 05:26:32 | FromGitter | <kayabaNerve> Distro, package manager, desktop, supporting suite... |
| 05:26:37 | FromGitter | <kayabaNerve> I would love to see Parabola with Guix and Mate. |
| 05:26:38 | ldlework | what? |
| 05:27:04 | ldlework | Guix is an operating system that uses Nix under the hood. |
| 05:27:15 | ldlework | So you mean Parabola with Nix pointed at the Guix repositories. |
| 05:27:32 | FromGitter | <kayabaNerve> An OS is a multitude of things. I'm not sure a OS built around Nix is the best option. I'm questioning NixOS versus another distro with Nix. That said, I acknowledge the major differences between distros is the package manager. |
| 05:27:41 | ldlework | I don't know why you'd want to use Nix as a package manager |
| 05:27:51 | ldlework | But then not have it for system configuration |
| 05:28:03 | FromGitter | <kayabaNerve> Guix is a package manager; GuixSD is the distro. |
| 05:29:08 | ldlework | "I'm not sure" is a very vague form of argument that nobody can do anything with. |
| 05:29:13 | ldlework | Have you heard of the term "FUD" |
| 05:29:24 | FromGitter | <kayabaNerve> Fear, uncertainty, and doubt. |
| 05:29:48 | FromGitter | <kayabaNerve> It's used to call out unfounded claims against a project. |
| 05:30:28 | FromGitter | <kayabaNerve> I like Nix. I 'm not sure about creating an entire distro around it. That said, most distros have their differences in what package manager is used. Therefore, I don't have a problem with NixOS. |
| 05:30:56 | FromGitter | <kayabaNerve> If I say Parabola + Nix that can be defined as Paranix (a new distro) so... |
| 05:30:57 | ldlework | Most OSes do not use their package manager technology for system configuration. |
| 05:31:06 | FromGitter | <kayabaNerve> True. |
| 05:31:14 | ldlework | OK I only pointed this out i think three times. |
| 05:31:36 | FromGitter | <kayabaNerve> Dude. I get you. I think you get me. Just a disconnect. Best thing is to move on. |
| 05:31:40 | ldlework | like i said, most linux distributions are barely distinct, until nixos and guixsd |
| 05:32:01 | ldlework | and maybe those distros that run literally every process in a container |
| 05:32:57 | FromGitter | <kayabaNerve> Arch is a very different thing from Ubuntu which is a very different thing from Qubes which is a different thing from TAILS. |
| 05:33:19 | ldlework | Qubes, sure. TAILS and ubuntu, not so much. |
| 05:33:27 | FromGitter | <kayabaNerve> Arch is a minimal OS. Ubuntu is an user friendly OS. Qubes is a secure OS. TAILS is an amnesiac system that resets on each launch. |
| 05:33:38 | ldlework | you can turn any of those into any of the other |
| 05:33:52 | FromGitter | <kayabaNerve> You can configure Nix to be Ubuntu |
| 05:33:58 | ldlework | nope |
| 05:34:09 | ldlework | i can install the same DE sure |
| 05:34:12 | FromGitter | <kayabaNerve> You're saying I can't use Nix to install Apt/GNOME and more? |
| 05:34:19 | ldlework | that's not what makes Ubuntu |
| 05:34:21 | ldlework | as you yourself pointed out |
| 05:34:24 | FromGitter | <kayabaNerve> Then what does? |
| 05:34:35 | ldlework | I can take all the linux distros and install any DE I want |
| 05:34:38 | FromGitter | <kayabaNerve> Kernel ⏎ Supporting Stack ⏎ Package Manager ⏎ DE [https://gitter.im/nim-lang/Nim?at=5bac6bee7bd81c5b9dbdbc68] |
| 05:35:26 | FromGitter | <kayabaNerve> So if I have my Nix script install the stack Ubuntu uses, install Apt, install the DE, install the software it ships with... |
| 05:35:28 | FromGitter | <kayabaNerve> It's all Linux. |
| 05:35:43 | ldlework | You'd still be using Nix |
| 05:35:46 | ldlework | And if you were on NixOS |
| 05:35:50 | ldlework | Nix for all system configuration |
| 05:36:01 | FromGitter | <kayabaNerve> Can Nix be used to uninstall Nix? |
| 05:36:04 | ldlework | I don't see how or why you want to continuously miss the point about how system confgiuration is done |
| 05:36:04 | FromGitter | <kayabaNerve> You can argue every Linux distro is not distinct or a lot are. |
| 05:36:09 | FromGitter | <kayabaNerve> I don't |
| 05:36:22 | FromGitter | <kayabaNerve> I'm commenting that it's not the only distinct Linux distro |
| 05:36:26 | ldlework | Maybe it is easier to think in terms of supersets of functionality |
| 05:36:32 | FromGitter | <kayabaNerve> I'm done... |
| 05:37:01 | ldlework | I already mentione two others that area also distinct, so maybe your frustration comes from fighting a strawman |
| 05:37:32 | ldlework | And I said "not that distinct" compared to the great distinctions with container OSes and functionally pure OSes |
| 05:37:55 | * | nsf joined #nim |
| 05:38:25 | ldlework | There are distinctions even between different flavours of ubuntu, but it is magnitude that was the crux of everything said |
| 06:07:49 | * | darithorn quit (Quit: Konversation terminated!) |
| 06:22:46 | * | jjido joined #nim |
| 06:27:11 | * | jjido quit (Client Quit) |
| 06:42:19 | FromGitter | <vivekimsit> I tried `nimble init` and its really cool |
| 06:42:47 | FromGitter | <vivekimsit> I was thinking if its a good idea to generate minimal `.gitignore` as well? |
| 06:43:11 | FromGitter | <codenoid> hi |
| 06:43:33 | FromGitter | <codenoid> last night try to combine nim mongo |
| 06:46:44 | * | faildrone2 joined #nim |
| 06:48:53 | FromGitter | <codenoid> AFAIK |
| 06:49:23 | Araq | is it too late to remove the 0.19 tag... |
| 06:49:46 | * | skrylar joined #nim |
| 06:51:01 | FromGitter | <gogolxdong> what's wrong. |
| 06:51:19 | Araq | the windows zips ship with a finish.exe that is wrong |
| 06:52:18 | FromGitter | <gogolxdong> I published a post of Nim 0.19 changelog on https://www.oschina.net/news . |
| 06:52:21 | * | faildrone2 quit (Ping timeout: 244 seconds) |
| 06:53:28 | Araq | before the official news entry is published? |
| 06:53:52 | Araq | oh well, I will just rebuild the windows zips with 0.19.1 |
| 06:54:09 | Araq | they are checksum'ed so people can still be sure these are not corrupted |
| 06:56:17 | * | unlink2 joined #nim |
| 06:56:21 | FromGitter | <narimiran> tags can be remove than added on a different commit ;) |
| 06:56:30 | FromGitter | <narimiran> *removed then |
| 06:58:25 | * | unlink2 quit (Remote host closed the connection) |
| 06:58:29 | FromGitter | <gogolxdong> where are you codenoid? |
| 06:59:09 | FromGitter | <bung87> shashlic I just wake up, I check it later. |
| 06:59:17 | FromGitter | <codenoid> i'm malaysia |
| 07:00:18 | FromGitter | <bung87> ah ,same timezone |
| 07:00:37 | FromGitter | <gogolxdong> We went to Kuala Lumpur two months ago. |
| 07:00:37 | FromGitter | <codenoid> where are you ? |
| 07:01:09 | FromGitter | <bung87> I didnot sleep last night , I’m in beijing of china |
| 07:01:44 | FromGitter | <gogolxdong> so you can read Chinese codenoid? |
| 07:01:54 | FromGitter | <codenoid> i love china |
| 07:01:58 | FromGitter | <codenoid> i have qq account |
| 07:02:01 | FromGitter | <codenoid> http://cd1.sportsdiscourse.com/uploads/2018/01/WANG-Manyu-ITTF.jpg |
| 07:03:08 | FromGitter | <gogolxdong> Genting Highlands is great. |
| 07:05:07 | FromGitter | <codenoid> ups, weibo too |
| 07:06:00 | FromGitter | <codenoid> 👍 |
| 07:06:53 | FromGitter | <gogolxdong> https://discord.gg/WxVRN8 we are in a discord group of most chinese ,there isn't many members atm. |
| 07:09:50 | FromGitter | <gogolxdong> another is from beijing too, didn't see him for a while. |
| 07:10:41 | FromGitter | <gogolxdong> bad English, haven't seen him for a while. |
| 07:46:23 | FromGitter | <gogolxdong> not sure how to translate `breaking changes` to Chinese exactly. |
| 07:57:24 | * | flusicaptain joined #nim |
| 08:00:29 | * | flusicaptain quit (Remote host closed the connection) |
| 08:03:06 | * | camu17 joined #nim |
| 08:03:45 | * | gmpreussner quit (Ping timeout: 252 seconds) |
| 08:03:48 | * | gmpreussner_ joined #nim |
| 08:07:31 | * | camu17 quit (Ping timeout: 246 seconds) |
| 08:16:32 | * | Vladar joined #nim |
| 08:22:57 | * | jvdmr27 joined #nim |
| 08:23:09 | FromGitter | <alehander42> @arnetheduck thank you, I am not absolutely sure, as mostly @michael72 works on zero-functional these days |
| 08:25:33 | * | PMunch joined #nim |
| 08:29:29 | * | jvdmr27 quit (Ping timeout: 244 seconds) |
| 08:33:30 | * | Vladar quit (Remote host closed the connection) |
| 08:33:54 | * | Vladar joined #nim |
| 08:42:26 | * | hackoon8 joined #nim |
| 08:45:19 | * | hackoon8 quit (Remote host closed the connection) |
| 08:53:32 | * | SupaHam9 joined #nim |
| 08:55:08 | * | SupaHam9 quit (Remote host closed the connection) |
| 09:00:59 | * | seni joined #nim |
| 09:01:59 | * | mpjetta25 joined #nim |
| 09:06:18 | * | mpjetta25 quit (Remote host closed the connection) |
| 09:17:18 | * | Rickta599 joined #nim |
| 09:22:05 | * | Rickta599 quit (Remote host closed the connection) |
| 09:27:39 | FromGitter | <7sDream> Does Nim have a release schedule table/web page that show our planning in recent years? |
| 09:36:07 | FromGitter | <gogolxdong> https://github.com/nim-lang/Nim/milestones |
| 09:36:57 | FromGitter | <gogolxdong> Are you still in Beijing? |
| 09:40:11 | * | geordi joined #nim |
| 09:41:10 | * | geordi quit (Remote host closed the connection) |
| 09:46:04 | FromGitter | <7sDream> Thx. |
| 09:46:13 | FromGitter | <7sDream> Yes~ |
| 09:47:32 | FromGitter | <gogolxdong> There is another one from Beijing , you can talk. |
| 09:48:56 | * | xet7 quit (Quit: Leaving) |
| 09:49:11 | * | Calinou quit (Quit: http://quassel-irc.org - Discuter simplement. Partout.) |
| 09:49:25 | * | xet7 joined #nim |
| 09:52:04 | FromGitter | <7sDream> I saw a programming language performance benchmark project on weibo this morning, just for share |
| 09:52:14 | FromGitter | <7sDream> (https://files.gitter.im/nim-lang/Nim/TOiv/image.png) |
| 09:52:17 | FromGitter | <7sDream> https://github.com/drujensen/fib |
| 09:53:28 | * | Calinou joined #nim |
| 09:57:15 | * | Pinapl2 joined #nim |
| 09:57:44 | FromGitter | <7sDream> happy to see more and more people notice Nim~ |
| 10:00:04 | * | jjido joined #nim |
| 10:01:13 | PMunch | Huh, tried to compile his fib.nim example with "nim cpp -d:release fib.nim" runs for me 16.8x faster |
| 10:01:20 | PMunch | From 11.3s to 0.68s |
| 10:02:03 | PMunch | I wonder what causes that.. |
| 10:05:35 | FromGitter | <narimiran> can i be the one that says that this benchmark is pointless? |
| 10:05:38 | * | Pinapl2 quit (Ping timeout: 252 seconds) |
| 10:06:18 | PMunch | Oh yeah, it's completely pointless. |
| 10:07:00 | * | jjido quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 10:08:07 | PMunch | This is very strange.. |
| 10:08:42 | PMunch | On my machine the C++ given by the benchmark runs about as fast as the Nim code compiled to C, but the Nim code compiled to C++ runs 16x faster than both of them |
| 10:09:20 | FromGitter | <narimiran> there are people who would say that all benchmarks are pointless (khmaraqkhm), but this one takes the cake :D |
| 10:11:39 | FromGitter | <gogolxdong> takes a cake , learned , lol |
| 10:11:48 | FromGitter | <gogolxdong> takes the cake |
| 10:12:19 | FromDiscord | <Shield> wow it took 130secs for me |
| 10:13:08 | PMunch | Shield, the Nim version? |
| 10:13:17 | PMunch | Could you try it with cpp instead of c? |
| 10:14:00 | PMunch | I just want to see if it's just something on my machine |
| 10:14:33 | * | preludedrew_ joined #nim |
| 10:15:50 | * | Calinou quit (Quit: http://quassel-irc.org - Discuter simplement. Partout.) |
| 10:16:23 | FromDiscord | <Shield> what's the correct way of putting compileToCpp in the cfg file? |
| 10:16:58 | * | noonien joined #nim |
| 10:17:08 | * | preludedrew_ quit (Remote host closed the connection) |
| 10:18:03 | FromDiscord | <Shield> it takes 3secs, the binary is small too |
| 10:18:23 | PMunch | So down from 130secs to 3 secs? |
| 10:18:28 | FromGitter | <narimiran> @gogolxdong i don't know if it is "a" or "the", i have a problem with those as they don't exist in my mother tongue |
| 10:19:17 | FromDiscord | <Shield> yeah, i'm on nim 0.18.0 |
| 10:19:24 | FromGitter | <narimiran> oh, it is indeed "the", i've checked :) |
| 10:19:36 | FromDiscord | <Shield> it seems that using --opt:speed isn't doing much when exporting to cpp |
| 10:19:43 | PMunch | Had a look at the C++ source code it generates, and it appears to actually calculate it on runtime |
| 10:19:50 | FromDiscord | <Shield> while for c it takes about 40secs |
| 10:19:55 | PMunch | Shield, -d:release implies --opt:speed |
| 10:20:52 | FromDiscord | <Shield> really? it went from 130secs to 40secs with --opt:speed, let me recheck |
| 10:25:48 | FromGitter | <gogolxdong> If your mother language derived from Germanic languages. |
| 10:27:43 | FromGitter | <gogolxdong> Is there a library implemented HAMT? |
| 10:28:48 | FromGitter | <gogolxdong> I found go uses de Bruijn (64 bit) constant |
| 10:29:28 | FromDiscord | <Shield> okay, the corrects times are: c takes 14.7s while cpp takes 1.93s |
| 10:30:10 | FromDiscord | <Shield> i had to clear nimcache, didn't expect it to cause problems |
| 10:30:29 | PMunch | Crazy that the C++ target is so much faster than the C target |
| 10:30:58 | FromGitter | <gogolxdong> It is if it is. |
| 10:34:06 | FromDiscord | <Shield> my machine is certainly not powerful to beat the 5.751s for C++ in the benchmark, i guess it may be closer to C++(constexpr) |
| 10:35:14 | FromGitter | <Bennyelg> Hey Im reviewing nim-days for the first time and I came across this: ⏎ ⏎ ```for i, l in pairs(lines):``` ⏎ ⏎ I know that we dont need use pairs. ```for idx, l in lines:``` will work just fine right? [https://gitter.im/nim-lang/Nim?at=5bacb262aedb375b9c544882] |
| 10:35:54 | FromGitter | <gogolxdong> no |
| 10:37:02 | FromGitter | <Bennyelg> work like charm to me |
| 10:37:06 | FromDiscord | <Shield> still, from the generated cpp code it doesn't seem like using it |
| 10:37:09 | FromGitter | <Bennyelg> ```var s = @["a", "b", "c"] ⏎ for i, x in s: ⏎ echo(i ,x)``` [https://gitter.im/nim-lang/Nim?at=5bacb2d553c31c11117104d0] |
| 10:37:46 | FromGitter | <bung87> the type must implements the pairs interface |
| 10:38:06 | FromGitter | <bung87> for builtins you already have |
| 10:38:22 | FromGitter | <Bennyelg> I c, |
| 10:39:11 | leorize | There's an implicit rule for items/pairs https://nim-lang.org/docs/manual.html#iterators-and-the-for-statement-implict-items-pairs-invocations |
| 10:42:42 | FromGitter | <gogolxdong> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bacb4214d320a463bffa283] |
| 10:45:56 | FromGitter | <vinodvinu> Hi, this code is from nim manual. ⏎ iterator countupT (a, b: T; step = 1): T {..} ⏎ Whats the meaning of "[T]" in this code ?, AFAIK, we are note using this. Since the code is countup(0, 10). |
| 10:46:55 | FromGitter | <bung87> called generic |
| 10:47:11 | FromGitter | <bung87> you can search in manual |
| 10:47:56 | FromGitter | <vinodvinu> Means, we can use this countup for all generic types, Am i right ? |
| 10:48:27 | FromGitter | <vinodvinu> But conting is only possible for numbers |
| 10:49:02 | FromGitter | <bung87> thats depens on the implements |
| 10:50:03 | FromGitter | <vinodvinu> So "[T]" means generic types |
| 10:50:42 | FromGitter | <bung87> if the implements requres some method of the type it must have |
| 10:51:01 | FromGitter | <bung87> otherwise you cannot do this |
| 10:51:26 | FromGitter | <vinodvinu> could you pls point me to an example ? |
| 10:51:28 | * | leru joined #nim |
| 10:52:36 | FromGitter | <bung87> sorry I am one my phone better search nim manual for generic |
| 10:52:59 | FromGitter | <vinodvinu> Ok |
| 10:53:50 | FromGitter | <vinodvinu> I saw this "[T]" in lot of pros in system module. |
| 10:54:28 | FromGitter | <vinodvinu> So i thought how to use this "[T]" in my procs |
| 10:56:18 | FromGitter | <bung87> yeah you can implements one ,if you implements same interfaces for not just one type. |
| 10:57:16 | FromGitter | <bung87> I thought it as c++ template |
| 10:57:52 | FromGitter | <moigagoo> Hi! How do I pass and mutate state inside an AyncHttpServer callback? I'm getting "Illegal capture" error. |
| 10:58:05 | * | jjido joined #nim |
| 10:58:23 | FromGitter | <vinodvinu> ha ha, i thought that i can declare a "seq[T]" as it is in my code :) |
| 10:59:04 | FromGitter | <moigagoo> I'm writing an app that uses websockets and I need it to mutate the state when certain payload is received on a certain path. |
| 10:59:14 | FromGitter | <bung87> no it ’s not the `any` |
| 11:00:08 | FromGitter | <vinodvinu> yes, later i learned that i can only declare a seq[] with a concrete type. |
| 11:00:27 | * | dddddd joined #nim |
| 11:00:38 | FromGitter | <bung87> @moigagoo I guess you pass variable from outer scope to a inner scope. |
| 11:00:39 | FromGitter | <vinodvinu> And the "any", "auto" are not concrete types |
| 11:01:44 | * | leru quit (Ping timeout: 252 seconds) |
| 11:02:38 | FromGitter | <moigagoo> @bung87 Well, yes. First, I tried mutating a global variable, then I tried wrapping the callback and server call in a closure and call that instead. |
| 11:05:24 | FromGitter | <bung87> well ,I also tried many times then give up, better redesign the process. |
| 11:06:52 | FromGitter | <bung87> closure proc and inline proc are different types |
| 11:07:18 | FromGitter | <moigagoo> What's the proper way to do it? Since I can't pass the state to the callback proc. I thought using a closusre was my only option. |
| 11:07:30 | FromGitter | <bung87> it’s not like you programming in js or python in that way |
| 11:08:20 | FromGitter | <bung87> no , you can using template or macro |
| 11:08:38 | FromGitter | <moigagoo> Sorry, I don't think I'm following you. Also, I think I've formulated the question wrong. |
| 11:09:05 | FromGitter | <moigagoo> I'm sure there's a simpler way, without templates or macros. |
| 11:09:35 | FromGitter | <bung87> then you need put you code |
| 11:09:58 | FromGitter | <bung87> otherwise no body knows what you exactly want to do |
| 11:10:31 | FromGitter | <moigagoo> Exactly. I'll tinker with it myself and come back with a concrete example if I fail. |
| 11:16:35 | * | Calinou joined #nim |
| 11:20:38 | * | Remram0 joined #nim |
| 11:22:47 | FromGitter | <Vindaar> @moigagoo (speaking as someone who's not very experienced with this sort of thing either) You should be able to use a channel to send something e.g. to the callback like I do here: https://github.com/brentp/nim-plotly/blob/master/src/plotly/image_retrieve.nim |
| 11:24:28 | * | Remram0 quit (Remote host closed the connection) |
| 11:26:59 | FromGitter | <Bennyelg> Mimicking python decorator , is templates are fine ? |
| 11:28:26 | FromGitter | <gogolxdong> it's from http:#supertech.csail.mit.edu/papers/debruijn.pdf |
| 11:29:09 | FromGitter | <gogolxdong> de Bruijn Sequences |
| 11:31:16 | * | francisl joined #nim |
| 11:33:51 | FromGitter | <mratsim> Nim also uses deBruijn for bit counting/log2: https://github.com/nim-lang/Nim/blob/master/lib/pure/bitops.nim#L56 |
| 11:35:03 | * | krux02 joined #nim |
| 11:59:21 | * | Calinou quit (Quit: http://quassel-irc.org - Discuter simplement. Partout.) |
| 12:04:00 | * | jjido quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 12:07:52 | * | francisl quit (Quit: francisl) |
| 12:10:01 | FromGitter | <gogolxdong> oh, great , seems it's a standardized implementation. Just finished the port from Go to Nim. |
| 12:12:19 | * | bloodygo1zo joined #nim |
| 12:15:01 | * | bloodygo1zo quit (Remote host closed the connection) |
| 12:17:32 | * | dade`22 joined #nim |
| 12:23:23 | FromGitter | <mratsim> Just use countTrailingZeroBits in the bitops module: https://github.com/nim-lang/Nim/blob/master/lib/pure/bitops.nim#L315 |
| 12:23:42 | FromGitter | <mratsim> the only issue I have with it is how the “noundefined” work |
| 12:23:48 | * | dade`22 quit (Ping timeout: 245 seconds) |
| 12:24:00 | FromGitter | <mratsim> Feel free to add your thoughts in this PR: https://github.com/nim-lang/Nim/pull/9029 |
| 12:35:28 | * | NimBot joined #nim |
| 12:36:20 | * | SenasOzys quit (Ping timeout: 252 seconds) |
| 12:38:23 | * | stefanos82 joined #nim |
| 12:43:36 | * | alephnull17 joined #nim |
| 12:44:10 | * | alephnull17 quit (Remote host closed the connection) |
| 12:47:11 | * | SenasOzys joined #nim |
| 12:48:28 | FromGitter | <gogolxdong> np |
| 12:49:29 | * | skrylar quit (Remote host closed the connection) |
| 12:51:37 | * | vlad1777d joined #nim |
| 12:56:55 | FromDiscord | <Shield> is using addr() the safest way to avoid object.field.subfield everytime you want to access the subfield? |
| 12:58:36 | def- | Shield: Template |
| 12:58:54 | * | emberquill joined #nim |
| 12:58:59 | * | francisl quit (Quit: francisl) |
| 12:59:30 | * | emberquill quit (Remote host closed the connection) |
| 13:00:02 | def- | template s: auto = obj.field.subfield |
| 13:00:55 | FromDiscord | <Shield> thanks |
| 13:03:30 | FromGitter | <zacharycarter> I got custom-tags working w/ the emerald DSL now |
| 13:03:41 | FromGitter | <zacharycarter> so the first custom element I'm building, is a router |
| 13:07:44 | FromGitter | <rufusroflpunch> ```proc sequence[T : any; R : set](ps: T): R = {1,2,3} ⏎ sequence(5)``` [https://gitter.im/nim-lang/Nim?at=5bacd62098245f4ae0908aca] |
| 13:07:53 | FromGitter | <rufusroflpunch> can someone explain why that results in a compilation error? |
| 13:09:16 | * | francisl joined #nim |
| 13:14:19 | FromGitter | <rufusroflpunch> scratch that, wrong snippet |
| 13:14:39 | * | ritchie_ joined #nim |
| 13:14:49 | FromGitter | <rufusroflpunch> ```proc sequence[T: any; R : seq](ps: T): R = @[1,2,3] ⏎ discard sequence[int, seq[int]](5)``` [https://gitter.im/nim-lang/Nim?at=5bacd7c901fb5f4ae12e919c] |
| 13:15:20 | FromGitter | <rufusroflpunch> that' s the one giving me trouble. I'm getting a "can't instantiate" error |
| 13:15:43 | * | chintimin joined #nim |
| 13:15:53 | FromGitter | <zacharycarter> I don't think you can use seq as a generic type |
| 13:16:02 | * | chintimin quit (Remote host closed the connection) |
| 13:16:04 | FromGitter | <zacharycarter> you'd need like |
| 13:16:38 | FromDiscord | <demotomohiro> Shield: |
| 13:16:39 | FromDiscord | <demotomohiro> https://wandbox.org/permlink/gKPvJgf8bB7NX46j |
| 13:17:13 | FromGitter | <zacharycarter> ```discard foo[int, seq[int]](5)`````` |
| 13:17:19 | FromGitter | <zacharycarter> also - like any isn't a concrete type |
| 13:17:25 | FromGitter | <zacharycarter> so I don't think it works w/ generics like that |
| 13:17:35 | FromGitter | <zacharycarter> you could use like SomeNumber or some other type that represents all numeric types |
| 13:17:39 | FromGitter | <zacharycarter> but I don't think you can use any like that |
| 13:18:36 | FromGitter | <rufusroflpunch> ah, I see. I think the issue was the any, I misunderstood its meaning |
| 13:19:10 | FromGitter | <rufusroflpunch> it actually works if I remove the `any` but leave it with `seq` |
| 13:19:11 | FromGitter | <ephja> yes, that works if you remove "any" |
| 13:19:16 | * | leru joined #nim |
| 13:19:34 | FromGitter | <rufusroflpunch> beautiful, thanks guys |
| 13:19:51 | FromGitter | <zacharycarter> sure thing |
| 13:20:12 | FromDiscord | <Shield> thanks for the snippet demotomohiro |
| 13:21:17 | * | TrentP8 joined #nim |
| 13:22:49 | * | TrentP8 quit (Read error: Connection reset by peer) |
| 13:26:03 | * | chemist69 quit (Ping timeout: 252 seconds) |
| 13:26:30 | * | chemist69 joined #nim |
| 13:26:33 | krux02 | yay it works: http://ix.io/1nGE |
| 13:26:35 | * | Riviera22 joined #nim |
| 13:26:39 | krux02 | this is fun |
| 13:26:57 | Araq | import segfaults |
| 13:27:03 | Araq | would have accomplished the same |
| 13:27:07 | krux02 | hmm |
| 13:27:31 | * | PrimHelios joined #nim |
| 13:27:49 | FromDiscord | <Shield> ouch |
| 13:27:56 | krux02 | can't find segfaults |
| 13:28:30 | * | Riviera22 quit (Remote host closed the connection) |
| 13:28:35 | * | leru quit (Ping timeout: 252 seconds) |
| 13:28:57 | krux02 | ok I did |
| 13:29:00 | krux02 | cool |
| 13:29:16 | Araq | it's terrible :-) |
| 13:29:27 | krux02 | why? |
| 13:29:42 | krux02 | actually I have no idea what I am doing, I just ported code from stack overflow |
| 13:29:48 | krux02 | and it works |
| 13:30:04 | krux02 | I don't know what a longjmp really does |
| 13:30:17 | krux02 | there will be a lot of unwinded stuff on the stack |
| 13:30:26 | * | Calinou joined #nim |
| 13:31:21 | krux02 | the idea is that this could be somewhen integrated in nimsuggest, so that even when it crashes with a segfault, it doesn't really crash, it restarts itself. |
| 13:32:02 | krux02 | Because there are bugs in nim, and with this safetyguard they don't need to be fixed for nimsuggest to become a stable process. |
| 13:32:35 | FromGitter | <zacharycarter> may want to be a bit more careful w/ the comparisons to Rust in Nim's twitter |
| 13:32:47 | krux02 | Araq: did you discuss with yuuta yamada the epc protocol of nimsuggest? |
| 13:32:50 | FromGitter | <zacharycarter> the whole - Rust type-safety comment - is getting a lot of flack from rustaceans |
| 13:33:02 | krux02 | I currently try to get the problems I have with nim-mode resolved. |
| 13:33:30 | krux02 | one is that sometimes emacs becomes unresposive, and when nimsuggest crashes, emacs really gets confused. |
| 13:34:01 | krux02 | I don't want to fix all the bugs in the compiler, I want emacs to be able to deal with imperfect nimsuggest, but the complexity of nim-mode is killing my progress. |
| 13:34:29 | FromGitter | <zacharycarter> VSCode / NimSuggest have the same issue |
| 13:34:30 | krux02 | I think yuuta yamada likes too much to add new fancy technology |
| 13:34:56 | krux02 | zacharycarter[m], I wonder where to fix the issue |
| 13:34:56 | FromGitter | <zacharycarter> Nimsuggest starts leaking and VSCode eventually needs a restart / you have to through your task manager and kill all the orphan nimsuggest processes |
| 13:35:18 | dom96 | https://nim-lang.org/blog/2018/09/26/version-0190-released.html 👑🎉👑🎉 |
| 13:35:20 | krux02 | on nimsuggest could handle it's own segfaults with the pattern I just posted (on posix) |
| 13:36:00 | krux02 | there is a buzz word missing |
| 13:36:29 | krux02 | metaprogramming |
| 13:36:41 | FromGitter | <kaushalmodi> @dom96 yay! |
| 13:37:17 | Araq | restarting after a segfault should be a watchdog's job, or supervisor, or whatever |
| 13:37:43 | krux02 | a "yay" from me as well :P |
| 13:38:03 | FromGitter | <zacharycarter> Also - just FYI Araq - I literally had to change zero things in my code to go from a relatively recent version of devel to 0.19.0 last night - at least for the JS target |
| 13:38:49 | * | ap0calypse22 joined #nim |
| 13:39:05 | FromGitter | <zacharycarter> great reply on the tweeters! |
| 13:39:09 | FromGitter | <zacharycarter> :D |
| 13:39:11 | dom96 | pff zacharycarter, you need more faith in my marketing :P |
| 13:39:16 | Araq | that doesn't count though, devel == 0.19 |
| 13:39:16 | krux02 | zacharycarter[m], how extensively do you use the js target? |
| 13:39:17 | dom96 | (regarding your tweet) |
| 13:40:21 | FromDiscord | <Shield> yay from me as well, waiting for the docs to catch up |
| 13:40:41 | dom96 | docs should be live already |
| 13:40:43 | * | ap0calypse22 quit (Killed (Sigyn (Spam is off topic on freenode.))) |
| 13:40:46 | FromDiscord | <Shield> the link for for-loop macros seems to be broken |
| 13:42:05 | Araq | yeah, the link structure isn't what I anticipated |
| 13:42:10 | Araq | these links are thus all broken |
| 13:42:18 | Araq | :-/ |
| 13:42:19 | FromDiscord | <demotomohiro> yay! for Version 0.19.0! |
| 13:42:28 | Araq | told you I hate directories with a passion :P |
| 13:43:26 | * | seni quit (Ping timeout: 252 seconds) |
| 13:43:46 | dom96 | When in doubt, use absolute links |
| 13:43:48 | * | Calinou quit (Quit: http://quassel-irc.org - Discuter simplement. Partout.) |
| 13:44:13 | dom96 | Should be fixed now |
| 13:44:24 | FromGitter | <survivorm> Congratulations on new release! |
| 13:45:02 | FromGitter | <codenoid> ✨ |
| 13:45:06 | FromGitter | <kaushalmodi> dom96: what's fixed? |
| 13:45:07 | FromGitter | <codenoid> 🌽 |
| 13:45:23 | FromGitter | <kaushalmodi> I was commenting on this in parallel: https://github.com/nim-lang/Nim/issues/8974#issuecomment-425096067 |
| 13:45:24 | FromDiscord | <Shield> it is fixed |
| 13:45:50 | shashlick | Is there a lib to make cpp interop easier |
| 13:46:01 | shashlick | Working with cpp data types |
| 13:46:23 | FromGitter | <kaushalmodi> nvm.. the broken link discussion was about the link in the blog post |
| 13:46:48 | * | nc-x joined #nim |
| 13:48:38 | dom96 | Araq: we need to merge devel into master |
| 13:48:50 | krux02 | there is a master? |
| 13:48:52 | dom96 | I'll do it |
| 13:49:35 | krux02 | I am so used that the master in Nim is called devel |
| 13:49:39 | FromGitter | <timotheecour> @dom96 @araq could that be the perfect time to do https://github.com/nim-lang/Nim/issues/8576 ? |
| 13:51:11 | FromGitter | <rufusroflpunch> > Added system.toOpenArray in order to support zero-copy slicing operations. This is currently not yet available for the JavaScript target. |
| 13:51:13 | FromGitter | <rufusroflpunch> wow! |
| 13:51:43 | FromGitter | <mratsim> if only \[\] defaulted to openarray now ;) |
| 13:52:11 | dom96 | kaushalmodi: that should now be fixed too |
| 13:54:00 | krux02 | regarding openarray, I think it is possible to tread any seq/string in the argument list of a function like an openarray. |
| 13:54:21 | * | vlad1777d quit (Ping timeout: 252 seconds) |
| 13:54:31 | krux02 | meaning people can write functions like ``proc foobar(a: string)`` and it takes no copy substrings |
| 13:55:18 | krux02 | but I would put it in the "it's neat if it's done" bucket. But it's not important for the moment at all. |
| 13:55:21 | * | nc-x quit (Quit: Page closed) |
| 13:56:33 | FromDiscord | <Zireael> I'm getting mad. I wanted to try out nim for an engine I'm using, but I can't grab nake... nimble install nake -n returns a git tlsv1 error 😦 |
| 13:56:48 | FromDiscord | <Zireael> first day using nim btw, so how would I grab nake manually? |
| 13:56:50 | FromGitter | <timotheecour> > and it takes no copy substrings ⏎ ⏎ seems incompatible with nil-terminated strings (which is why I suggested nil-terminated strings should only be for string literals) |
| 13:56:58 | krux02 | Araq: do you have any rough estimate on how many distinct symbols there are in the compilation of the nim compiler? I mean how many symbol IDs are there? |
| 13:56:59 | FromGitter | <mratsim> nake is usually not needed anymore @ Zireael |
| 13:56:59 | FromDiscord | <2vg> version 0.19.0! 🎉 |
| 13:57:15 | FromGitter | <mratsim> you can just use nimble if you need tasks |
| 13:57:30 | Araq | krux02, iirc about a million |
| 13:57:51 | FromDiscord | <Zireael> the bindings https://pragmagic.github.io/godot-nim/master/index.html tell me to use nake, so I want to use nake |
| 13:57:59 | FromDiscord | <Shield> nake didn't work for me |
| 13:58:07 | FromGitter | <kaushalmodi> @dom96 I am talking about the broken source/edit links on https://nim-lang.org/docs/cpuinfo.html |
| 13:58:11 | FromGitter | <kaushalmodi> they are still broken |
| 13:58:15 | krux02 | I have a rough idea in my head to implement responsive completion like colorpicking in computer graphics. |
| 13:58:20 | FromGitter | <kaushalmodi> I thought docgen would need to be fixed to fix those |
| 13:58:25 | FromDiscord | <Zireael> anyway if it blows up due to tls it means any other git libraries won't work either, which is a main problem |
| 13:58:32 | FromDiscord | <Shield> oh right, just read what the nake script does and make a nim program instead, i did exactly that to generate godot's bindings |
| 13:58:49 | FromGitter | <codenoid> we need a corn |
| 13:59:02 | dom96 | kaushalmodi: This file is missing in devel too |
| 13:59:05 | dom96 | so I dunno what's going on |
| 13:59:19 | krux02 | color picking is a technique where every distinct object is rendered in its unique color to an image that is usually not displayed. |
| 13:59:30 | dom96 | Zireael: You don't need nake, just use Nimble files |
| 13:59:40 | dom96 | or nimscript |
| 13:59:41 | FromDiscord | <Zireael> anyway I can't get the bindings themselves either, nimble install godot blows up with the same tls error |
| 13:59:49 | krux02 | then when you interact with the scene, like lick on an object, the pixel is read from that image and the color says exactly what object it is. |
| 13:59:56 | FromDiscord | <Zireael> I'm on Nim 0.18.0 |
| 13:59:57 | krux02 | this can be done with compilation, too |
| 14:00:04 | FromGitter | <mratsim> compile with -d:ssl maybe |
| 14:00:09 | dom96 | Zireael: this might help: https://github.com/nim-lang/nimble#troubleshooting |
| 14:01:03 | FromDiscord | <Zireael> I'm on Windows, so MacOS mumbo-jumbo doesn't need apply |
| 14:01:15 | FromGitter | <zetashift> ` Success: godot installed successfully.` @Zireael it's currently working for me I'm also on 0.18 |
| 14:01:33 | krux02 | the project is compiled and every character is mapped to a symbol id. Then the editor only needs to read the symbolid for that byte position in the file. |
| 14:01:52 | krux02 | plus some diffing this could really be super responsive and reliable |
| 14:02:09 | dom96 | Zireael: it might, if you have an old openssl dll |
| 14:02:14 | FromGitter | <kaushalmodi> @dom96 ⏎ ⏎ > so I dunno what's going on ⏎ ⏎ the paths are wrong. ... [https://gitter.im/nim-lang/Nim?at=5bace2e541177e0bc7b573c9] |
| 14:02:18 | FromDiscord | <Zireael> C:\Users\AdmKasia>nimble install godot |
| 14:02:18 | FromDiscord | <Zireael> Downloading https://github.com/pragmagic/godot-nim using git |
| 14:02:18 | FromDiscord | <Zireael> Error: unhandled exception: Unable to query remote tags for https://github.com/p |
| 14:02:18 | FromDiscord | <Zireael> ragmagic/godot-nim. Git returned: fatal: unable to access 'https://github.com/pr |
| 14:02:19 | FromDiscord | <Zireael> agmagic/godot-nim/': error:1407742E:SSL routines:SSL23_GET_SERVER_HELLO:tlsv1 al |
| 14:02:20 | FromDiscord | <Zireael> ert protocol version |
| 14:02:22 | FromDiscord | <Zireael> [OSError] |
| 14:02:39 | FromGitter | <zetashift> yikes pastebin it next time gitter/irc doesn't play well with discord code blocks |
| 14:02:40 | krux02 | one million symbol ids still fit in one 32 bit integer. |
| 14:02:41 | dom96 | kaushalmodi: that's not easily fixable |
| 14:02:41 | FromDiscord | <Zireael> C:\Users\AdmKasia>nimble install godot |
| 14:02:42 | FromDiscord | <Zireael> Downloading https://github.com/pragmagic/godot-nim using git |
| 14:02:42 | * | ash_worksi27 joined #nim |
| 14:02:42 | FromDiscord | <Zireael> Error: unhandled exception: Unable to query remote tags for https://github.com/pragmagic/godot-nim. Git returned: fatal: unable to access 'https://github.com/pragmagic/godot-nim/': error:1407742E:SSL routines:SSL23_GET_SERVER_HELLO:tlsv1 alert protocol version |
| 14:02:42 | FromDiscord | <Zireael> [OSError] |
| 14:02:53 | dom96 | zireael: please use a pastebin |
| 14:02:53 | * | ash_worksi27 quit (Remote host closed the connection) |
| 14:03:19 | FromGitter | <kaushalmodi> dom96: But should be fixed right? :) I didn't say "easily". |
| 14:03:35 | FromGitter | <kaushalmodi> It leaves a bad taste for new adopters if they find such broken links |
| 14:03:40 | dom96 | yes, it should |
| 14:03:44 | FromDiscord | <Zireael> pastebin: https://pastebin.com/CeJ8iv59 |
| 14:03:47 | FromGitter | <kaushalmodi> I believe almost all source/edit links are broken |
| 14:04:02 | dom96 | oh, that's bad |
| 14:04:03 | FromGitter | <kaushalmodi> (or at least they should be removed till they start getting pointed correctly) |
| 14:04:45 | FromGitter | <zacharycarter> omg - gitter just caught up - for some reason I haven't seen any messages for the past 20 mins or so |
| 14:05:06 | FromGitter | <zacharycarter> krux02: I'm building quite a bit with the JS target these days - while I work on a new version of the playground and my personal website |
| 14:05:29 | FromDiscord | <Zireael> so the reason tlsv1 doesn't work is this: https://blog.github.com/2018-02-23-weak-cryptographic-standards-removed/ |
| 14:05:41 | FromGitter | <zetashift> @Zireael this might be something with Git: https://github.com/glennhickey/progressiveCactus/issues/93 |
| 14:05:55 | FromDiscord | <Zireael> unfortunately it was so long ago I forgot what changes I had to make to my sourcetree to make it work |
| 14:05:59 | Araq | @Zireael: 'nimble install godot' works for me |
| 14:06:10 | Araq | you seem to be using an outdated git version |
| 14:06:33 | * | elrood joined #nim |
| 14:07:11 | dom96 | Zireael: You need to update git and/or your openssl |
| 14:07:55 | FromDiscord | <Zireael> ouch, indeed - sourcetree seems to have embedded its own newer version |
| 14:08:00 | FromDiscord | <Zireael> rectifying it now |
| 14:08:16 | FromDiscord | <Shield> system docs still only have two atomic functions |
| 14:08:16 | FromDiscord | <Zireael> sorta glad I picked up Nim today, otherwise the problem would still be there undetected 😛 |
| 14:10:37 | FromGitter | <kaushalmodi> @dom96 I have opened https://github.com/nim-lang/Nim/issues/9083 to focus on just fixing the Source/Edit links. Should such issues be tagged with a 0.19.1 milestone or similar? |
| 14:12:00 | Araq | these links are not wrong, so they haven't been fixed |
| 14:12:48 | Araq | oh now they are, ok :P |
| 14:13:44 | dom96 | kaushalmodi: showstopper label will do |
| 14:13:56 | FromGitter | <kaushalmodi> ok, I cannot add it, please do it |
| 14:14:06 | FromGitter | <kaushalmodi> Araq: That's what I have been saying since 12 days! :P |
| 14:14:23 | Araq | I thought it's about the devel vs master |
| 14:14:35 | Araq | edit should point to devel, view source to master, that one is not a bug |
| 14:15:34 | FromGitter | <kaushalmodi> Araq: Right, fixing this should fix both source/edit assuming master and devel have the same path hierarchy |
| 14:16:21 | FromGitter | <kaushalmodi> This is another regression I found yesterday in regenerating my Nim notes (the only regression from my snippet set): https://github.com/nim-lang/Nim/issues/9079 |
| 14:18:56 | dom96 | If someone wants a challenge: write a script that will find all such links in our docs and fix them :) |
| 14:19:57 | * | nc-x joined #nim |
| 14:21:03 | FromGitter | <kaushalmodi> dom96: I can easily do that (find the broken links, not fix :P) using a cli tool called `htmltest`: https://github.com/nim-lang/Nim/issues/8974#issuecomment-421579314 |
| 14:21:17 | FromGitter | <kaushalmodi> Last time I ran it, I had 20k+ broken links |
| 14:21:38 | FromGitter | <kaushalmodi> probably half were because of not finding dochack.js on local doc builds |
| 14:21:49 | * | SenasOzys quit (Remote host closed the connection) |
| 14:22:09 | * | SenasOzys joined #nim |
| 14:23:08 | dom96 | nc-x: You should now be able to speak |
| 14:23:22 | elrood | noble attempt, kaushalmodi, i've mentioned broken links in the docs numerous times in the past, and ways to automatically check for those, but nobody seemed to really care |
| 14:23:45 | nc-x | Thank you dom96 |
| 14:23:47 | nc-x | Araq: "edit should point to devel". But well then the line numbers might change if file gets modified/removed/something else. Maybe point it at a commit in devel (The url you get when you press 'y') |
| 14:24:19 | FromGitter | <kaushalmodi> nc-x: need to consider line number change, file path change too |
| 14:24:32 | dom96 | elrood: Mentioning ways is great, what's better is preparing a PR that will perform those checks :) |
| 14:25:10 | FromGitter | <kaushalmodi> dom96: I already got autocheck working.. https://github.com/nim-lang/Nim/issues/8974#issuecomment-421579314 |
| 14:25:23 | FromGitter | <kaushalmodi> the autocheck crashes because of unescaped % in the URLs |
| 14:25:43 | FromGitter | <kaushalmodi> while most browsers support those, you don't typically put bare symbols in URLs |
| 14:26:39 | FromGitter | <kaushalmodi> For example, to have `%` literally in the link, you need to encode that as `%25` |
| 14:26:46 | FromGitter | <kaushalmodi> `%` happens to be the URL encoding prefix |
| 14:26:49 | FromGitter | <kaushalmodi> https://www.w3schools.com/tags/ref_urlencode.asp |
| 14:29:34 | dom96 | Yeah, we should fix that |
| 14:29:45 | FromDiscord | <Zireael> @Shield so what did you do to eliminate using nake? I have nake installed but any nake commands complain that nake is unrecognized |
| 14:29:56 | dom96 | yay for early morning US HN traffic |
| 14:29:56 | stefanos82 | congrats to all folks who contributed to this new Nim release. Well done everyone +1 |
| 14:30:51 | Araq | the problem is that the docgen is freaking mess |
| 14:31:11 | Araq | so "don't have broken links in the docs" is easy to say, hard to fix |
| 14:31:54 | Araq | and the core problem here is that everything needs to be "configurable" so I have to read 3 to 5 different places at once to figure out how these links are produced |
| 14:32:25 | Araq | the fact that I'm not the original programmer responsible for this code doesn't help |
| 14:33:02 | Araq | the doc config system needs to be replaced with NimScript IMO |
| 14:34:03 | Araq | and ideally the documentation is full of hyperlinks to external projects like Nimble, popular Nimble packages... |
| 14:34:12 | FromDiscord | <Shield> @Zireael I meant look at the nake script and try to do the things it does manually, it's pretty simple, "godot.exe --gdnative-generate-json-api api.json" then call the genApi on it |
| 14:34:26 | FromDiscord | <Zireael> way over my head |
| 14:34:50 | FromDiscord | <Shield> it's really odd that the library needs to autogen the bindings insead of offering them |
| 14:34:55 | FromDiscord | <Zireael> anyway manually doing nim -c -r nakefile works, so why doesn't the 'nake' shortcut work? |
| 14:35:07 | * | Enigmagic11 joined #nim |
| 14:35:42 | FromDiscord | <Shield> nake exists in another folder, you need to set the PATH variable for it to call it like a command |
| 14:35:55 | FromDiscord | <Shield> use the full path |
| 14:36:14 | livcd | func is now an alias for proc <- the biggest change :D |
| 14:37:06 | FromDiscord | <Shield> nake is in "C:\Users\user name\.nimble\bin" |
| 14:37:53 | * | Enigmagic11 quit (Remote host closed the connection) |
| 14:38:05 | FromGitter | <kaushalmodi> > and ideally the documentation is full of hyperlinks to external projects like Nimble, popular Nimble packages ⏎ ⏎ Yeah, seems like the docgen needs to be a good static site generator |
| 14:38:55 | * | sooda22 joined #nim |
| 14:39:43 | dom96 | So many spammers |
| 14:42:47 | PMunch | Haha, did you recognize him by his nick? |
| 14:43:07 | dom96 | No, I can only see their messages |
| 14:43:11 | dom96 | Because I have +o |
| 14:43:17 | dom96 | They are constantly posting crap |
| 14:43:30 | PMunch | Ah right |
| 14:43:34 | dom96 | And now they're starting by choosing some random legit-sounding question |
| 14:43:49 | dom96 | Like this one joined and said: "are there any means to get `std::hash` without including <functional> or <unordered_{multi,}{set,map} > ????" |
| 14:44:12 | FromGitter | <zetashift> The bots are getting smarter |
| 14:44:22 | PMunch | Haha, yeah that seems like bot speak |
| 14:44:29 | leorize | they're copying stuff from other channels actually |
| 14:44:43 | * | Guest60373 joined #nim |
| 14:44:46 | PMunch | Aah, that kinda makes sense |
| 14:44:56 | FromGitter | <zetashift> I just read the HN topic, the stdlib of Nim isn't that bad is it now? |
| 14:45:13 | PMunch | What would be smarter would lurk in a channel for a couple of days or grab something from reddit.com/r/<channel> |
| 14:45:20 | PMunch | Then it would at least be the same topic |
| 14:45:45 | PMunch | zetashift, it's a bit eclectic |
| 14:48:34 | * | Guest60373 quit (Remote host closed the connection) |
| 14:48:37 | * | onionhammer1 quit (Ping timeout: 246 seconds) |
| 14:49:10 | PMunch | So now I should be able to see the spammers to? |
| 14:51:17 | * | tzui joined #nim |
| 14:52:06 | * | jonadab4 joined #nim |
| 14:52:21 | * | jonadab4 quit (Remote host closed the connection) |
| 14:52:47 | planetis[m] | congrats on the new release :) |
| 14:53:25 | * | francisl quit (Quit: francisl) |
| 14:53:42 | FromGitter | <cpunion> Can’t install nim 0.19.0 with choosenim. |
| 14:53:56 | FromGitter | <cpunion> (https://files.gitter.im/nim-lang/Nim/RaVl/image.png) |
| 14:54:28 | FromGitter | <cpunion> macOS 10.14 |
| 14:54:41 | FromDiscord | <Shield> removing 0.18.0 from PATH and installing 0.19.0 still call the old version |
| 14:54:41 | dom96 | weird |
| 14:54:49 | dom96 | I built it successfully using 10.11 |
| 14:55:09 | PMunch | Oh cool, didn't notice we had a new release |
| 14:55:22 | dom96 | cpunion: pretty sure this isn't a choosenim bug, report on Nim's github please |
| 14:55:33 | FromGitter | <cpunion> OK. |
| 14:55:53 | leorize | someone came here earlier today and complain about nim can't build a simple hello world on macOS Mojave |
| 14:55:58 | * | francisl joined #nim |
| 14:56:14 | dom96 | cpunion: you can double check by following this: https://nim-lang.org/install_unix.html#manual-installation-from-source |
| 14:56:31 | livcd | leorize: i can check later at home |
| 14:56:31 | FromDiscord | <demotomohiro> I found typo in nim manual: |
| 14:56:32 | FromDiscord | <demotomohiro> https://nim-lang.github.io/Nim/manual.html#types-preminusdefined-integer-types |
| 14:56:32 | FromDiscord | <demotomohiro> uintXX |
| 14:56:32 | FromDiscord | <demotomohiro> additional signed integer |
| 14:56:32 | FromDiscord | <demotomohiro> This should be "additional unsigned integer" |
| 14:56:45 | dom96 | PR :) |
| 14:56:49 | FromDiscord | <demotomohiro> ok |
| 14:56:51 | * | Praise20 joined #nim |
| 14:57:03 | dom96 | leorize: That would suck :( |
| 14:57:24 | leorize | I think it's a macOS problem... |
| 14:58:08 | FromGitter | <mratsim> I suddenly don’t want to update ... |
| 14:58:13 | livcd | :D |
| 14:58:50 | FromGitter | <Bennyelg> any thoughts ? https://github.com/nim-lang/Nim/issues/9082 |
| 14:58:51 | * | Praise20 quit (Read error: Connection reset by peer) |
| 14:58:57 | FromGitter | <7sDream> Can we install choosenim via brew? |
| 14:59:23 | FromGitter | <mratsim> It’s probably better to install it in your user directory |
| 14:59:47 | FromGitter | <cpunion> Aha, workaround: `C_INCLUDE_PATH=/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/usr/include/ choosenim update stable` |
| 14:59:57 | FromGitter | <cpunion> (https://files.gitter.im/nim-lang/Nim/pbBp/image.png) |
| 15:00:01 | FromDiscord | <Shield> so calling nim -v reports 0.19 but calling it in my project folder reports 0.18, god I hate windows |
| 15:00:21 | leorize | cpunion: definitely macOS bug :P |
| 15:00:34 | FromGitter | <kaushalmodi> Shield: why not delete the old version first? |
| 15:01:54 | FromGitter | <cpunion> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bacf0e2aedb375b9c561fc7] |
| 15:02:12 | FromGitter | <cpunion> @dom96 |
| 15:02:20 | leorize | try to build a simple "Hello world" in C |
| 15:02:35 | livcd | hmm I just git pull and go through the build process. Am I doing it wrong ? |
| 15:03:02 | FromGitter | <kaushalmodi> livcd: that would work too. I have been always building that way |
| 15:04:36 | * | m4ho3 joined #nim |
| 15:05:46 | FromGitter | <kdheepak> Hi all, when I run `choosenim update stable` I get the following: ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bacf1ca56877c463a921a5a] |
| 15:06:32 | FromGitter | <cpunion> ```code paste, see link``` ⏎ ⏎ @leorize `cc` works, `clang` fail. [https://gitter.im/nim-lang/Nim?at=5bacf1f8eba8e60bc6538717] |
| 15:07:19 | FromGitter | <cpunion> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bacf22753c31c111172c84d] |
| 15:07:47 | FromDiscord | <Shield> windows is slow with updating the path variable |
| 15:07:59 | FromGitter | <kaushalmodi> Shield: restart? |
| 15:08:03 | FromGitter | <kaushalmodi> :P |
| 15:08:19 | * | m4ho3 quit (Remote host closed the connection) |
| 15:08:22 | FromGitter | <kaushalmodi> it's funny and sad at the same time that restarts "fix" things in Windows |
| 15:08:51 | nc-x | Hmm.. usually you only need to open and close the terminal for path changes to take effect. Not restart PC |
| 15:08:54 | FromGitter | <kaushalmodi> on the other hand, my old RHEL 6.8 has had an uptime of may be a year now |
| 15:09:00 | FromGitter | <cpunion> Maybe it’s my problem. |
| 15:09:06 | FromGitter | <7sDream> It is the new MacOS that cause this compile failed? Just test on 10.13, works like a charm. |
| 15:09:08 | FromGitter | <cpunion> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bacf294aedb375b9c562ac0] |
| 15:09:39 | leorize | update your clang then |
| 15:09:41 | * | MSHughes joined #nim |
| 15:10:58 | * | stefantalpalaru joined #nim |
| 15:11:13 | livcd | kaushalmodi: that's a bit unfair. Restarts fix issues because people do not know and cant be bothered to investigate where the issue is :P |
| 15:12:11 | FromDiscord | <Shield> it works now, i recompiled fib.nim, it's really weird, compiling it the first time makes the program really slow, compiling it again will make it run the normal speed, these kind of things are what drive me mad |
| 15:12:17 | FromGitter | <kaushalmodi> livcd: I installed lastpass binary for Firefox, and the plugin asked me to restart |
| 15:12:35 | FromGitter | <kaushalmodi> It's not uncommon for software installs/updates to request restart on Windows |
| 15:12:38 | * | MSHughes quit (Killed (e (spam))) |
| 15:12:50 | stefantalpalaru | I'm getting a compile error for 0.19.0: ../lookups.nim(239, 35) Error: type mismatch: got <string> but expected one of: proc getIdent(ic: IdentCache; identifier: string): PIdent [...] expression: getIdent(toLowerAscii(x.s[0]) & substr(x.s, 1)) |
| 15:13:55 | livcd | kaushalmodi: I suspect devs opt for an easy restart instead of trying to fix the experience. Easier than trying to cover all the edge cases. |
| 15:13:57 | FromGitter | <dom96> @cpunion yeah, that's a regression in our new release I guess |
| 15:14:06 | FromGitter | <dom96> or maybe the same happens for Nim 0.18.0 |
| 15:14:14 | FromGitter | <dom96> in any case, report it on the Nim github repo if you haven't already |
| 15:14:41 | FromGitter | <dom96> include your workaround too |
| 15:15:01 | livcd | kaushalmodi: btw thanks for your nim notes :) |
| 15:15:44 | FromGitter | <kdheepak> I'm using a fresh install of nim ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bacf42001fb5f4ae12f66b7] |
| 15:15:57 | FromGitter | <dom96> @kdheepak that sucks :( try looking for a libcrypto on your system, add a symlink that matches one of those names, for example: libcrypto.1.1.dylib or install it if it hasn't been installed yet |
| 15:16:32 | FromDiscord | <Shield> targetting C++ using nim 0.19 makes the program ask about libstdc++-6.dll |
| 15:16:39 | FromDiscord | <Shield> and where's nimcache |
| 15:17:08 | FromGitter | <kaushalmodi> livcd: Thanks! :) btw they are guaranteed to work 100% on 0.19.0 :) |
| 15:18:44 | livcd | I like your site. It's very very readable |
| 15:18:59 | nc-x | Shield: You can use msvc compiler if you don't want dependency on libstdc++. Or maybe mingw has some switch for not having a dependency, idk. |
| 15:19:04 | * | SenasOzys quit (Remote host closed the connection) |
| 15:19:17 | * | SenasOzys joined #nim |
| 15:19:40 | FromGitter | <cpunion> @dom96 I installed `llvm` with `Homebrew`, so `clang` in llvm was called and failed. Just make `/usr/bin` first in `PATH`, the compilation succeeded. I think it’s not a bug of Nim. |
| 15:20:17 | nc-x | Shield: nimcache path - https://nim-lang.org/docs/nimc.html#compiler-usage-generated-c-code-directory |
| 15:20:44 | FromGitter | <kaushalmodi> livcd: Thanks. Really happy to hear that (btw the way, I have got reports that the site crashes Android Chrome as it has a lots of DOM elements.. so might split up that page into sub pages when I get time) |
| 15:21:11 | FromGitter | <kdheepak> Shouldn't this be updated too: https://github.com/Homebrew/homebrew-core/blob/master/Formula/nim.rb |
| 15:21:55 | * | Trustable joined #nim |
| 15:22:17 | * | PMunch quit (Quit: Leaving) |
| 15:23:14 | FromDiscord | <Shield> thanks nc-x, I prefered when nimcache was local |
| 15:23:32 | * | Zevv joined #nim |
| 15:23:35 | Zevv | Congratulations! |
| 15:24:03 | FromDiscord | <Shield> what did happen between 0.18 and 0.19 to make it depend on libstdc++ tho |
| 15:24:32 | * | teus10 joined #nim |
| 15:24:33 | FromGitter | <kdheepak> I have the following file: /usr/local/lib/libcrypto.1.0.0.dylib |
| 15:24:35 | FromGitter | <kaushalmodi> Shield: You can force the nimcache switch in your nim.cfg |
| 15:24:52 | FromGitter | <kdheepak> so I'm not sure why it's not finding it |
| 15:26:33 | * | teus10 quit (Remote host closed the connection) |
| 15:29:13 | FromDiscord | <Shield> everything's fine now |
| 15:29:14 | nc-x | Shield: So `nim cpp` on 0.18 did not depend on libstdc++? https://stackoverflow.com/questions/31449769/distribute-program-compiled-with-mingw-g I think mingw g++ always depends on them |
| 15:30:10 | FromDiscord | <Shield> yeah it didn't depend on it for the fib executable on nim 0.18, otherwise it'll complain like it did now |
| 15:30:10 | nc-x | -static-libstdc++ works maybe (https://stackoverflow.com/questions/14225083/linking-with-static-libstdc-flag-on-mingw-4-7-1) |
| 15:30:11 | * | nsf quit (Quit: WeeChat 2.2) |
| 15:31:06 | stefantalpalaru | the "../lookups.nim(239, 35) Error: type mismatch: got <string>" error appears when compiling nimfix |
| 15:31:37 | nc-x | isn't nimfix dead? |
| 15:32:08 | FromDiscord | <Shield> https://github.com/nim-lang/Nim/blob/devel/lib/system.nim#L3243 |
| 15:32:08 | FromDiscord | <Shield> only two atomic functions are forward declared and atomics module has no docs |
| 15:32:15 | nc-x | haven't heard about it since I started using nim which is about atleast half a year now. |
| 15:32:33 | livcd | hmm how do i pass the mingw switches in nim.cfg ? |
| 15:33:16 | FromGitter | <kaushalmodi> livcd: `nim --fullhelp | grep pass` |
| 15:33:46 | FromGitter | <kaushalmodi> sorry, nvm |
| 15:33:58 | FromGitter | <kaushalmodi> I guess you already know the switches and want to know how to set them in cfg? |
| 15:35:07 | livcd | yup |
| 15:35:32 | FromGitter | <kaushalmodi> I don't use nim.cfg, but I know how to set it in project directory config.nims |
| 15:35:33 | FromGitter | <kaushalmodi> https://nim-lang.org/docs/nims.html |
| 15:36:23 | FromGitter | <iffy> Does Nim have a combination grammar/parser? I've played with `pegs` and `strscans` and am now looking at `lexbase`, but I'm wondering if there's something like Python's excellent parsley library: https://parsley.readthedocs.io/en/latest/tutorial.html |
| 15:37:37 | FromGitter | <iffy> Also, the `Source` links from the docs to GitHub seem to be broken today: For instance, click on the `Source` link here: https://nim-lang.org/docs/parsesql.html#newNode,SqlNodeKind |
| 15:38:25 | FromGitter | <kaushalmodi> livcd: Here's an example of setting switches in nim.cfg: https://nim-lang.org/docs/nimc.html#cross-compilation-for-nintendo-switch |
| 15:38:50 | FromGitter | <kaushalmodi> @iffy The issue is known: https://github.com/nim-lang/Nim/issues/9083 |
| 15:39:04 | FromGitter | <iffy> k |
| 15:39:41 | livcd | kaushalmodi: thanks. weird i found --pasC:".." but dunno if it actually works |
| 15:40:19 | FromGitter | <kaushalmodi> is that a typo here in IRC or you are quoting it verbatim from docs? |
| 15:40:26 | FromGitter | <mratsim> passC |
| 15:40:33 | livcd | typo |
| 15:40:55 | livcd | and it works |
| 15:41:09 | FromGitter | <kaushalmodi> 👍 |
| 15:41:11 | livcd | was just passing linker flags to --passC:...ooops |
| 15:42:23 | livcd | just wanted to try the --passL:"-static -static-libgcc -static-libstdc++" |
| 15:43:14 | * | shashlick quit (Remote host closed the connection) |
| 15:45:06 | FromDiscord | <Shield> how to make it compile with C++ from nim.cfg? |
| 15:47:33 | nc-x | Well atleast as a hack, you can modify this https://github.com/nim-lang/Nim/blob/devel/config/nim.cfg#L141 (I think) |
| 15:48:13 | nc-x | search for `@if windows` in the file. There are many such locations where you can do the changes |
| 15:50:43 | FromDiscord | <Shield> it'll be nicer if you can change that in the configs, they're more convenient |
| 15:52:33 | FromDiscord | <Shield> btw, using -d:release still show line tracing in the source file |
| 15:52:34 | nc-x | You might also be able to use `passc` in .nims using `switch(...)`. Dunno about .nim.cfg though |
| 15:53:20 | nc-x | Maybe create an issue on github for static linking libstdc++ in mingw. That might become the default configuration then, |
| 15:54:25 | FromDiscord | <Shield> --passL:"-static -static-libgcc -static-libstdc++" did the job |
| 15:54:27 | nc-x | Also, I think nimscript, nimble, .cfg. There are too many ways for configuration. One needs to become the default and better get rid of the others. |
| 15:55:08 | FromGitter | <kaushalmodi> nc-x: I think nimscript is the recommended way (just that we cannot yet? create a global Nim config using .nims) |
| 15:55:14 | FromDiscord | <Shield> I find nim.cfg the simplest since it can affect a whole folder and works with sublime text |
| 15:55:27 | FromGitter | <kaushalmodi> and .nimble use NimScript |
| 15:55:41 | FromGitter | <kaushalmodi> config.nims affects the whole folder too |
| 15:56:11 | FromGitter | <kaushalmodi> Shield: See https://nim-lang.org/docs/nims.html |
| 15:56:12 | FromDiscord | <Shield> I'll try them |
| 15:56:54 | * | pink_mist4 joined #nim |
| 15:57:02 | * | Calinou joined #nim |
| 15:57:10 | * | pink_mist4 quit (Remote host closed the connection) |
| 15:57:22 | FromGitter | <kaushalmodi> Shield: The reason I love .nims is that I can code just as I do in Nim without .cfg syntax overhead |
| 15:57:25 | FromGitter | <kaushalmodi> example: https://github.com/kaushalmodi/hello_musl/blob/master/config.nims |
| 15:58:22 | * | TuxedoJack joined #nim |
| 15:58:29 | * | francisl quit (Quit: francisl) |
| 15:58:52 | FromDiscord | <Shield> i want somebody else to confirm this, -d:release does not remove line tracing |
| 15:59:01 | * | PMunch joined #nim |
| 16:00:31 | nc-x | Well according to https://github.com/nim-lang/Nim/blob/devel/config/nim.cfg#L55 , linetrace:off |
| 16:01:32 | * | TuxedoJack quit (Remote host closed the connection) |
| 16:02:11 | FromGitter | <kaushalmodi> Shield: How are you setting -d:release? |
| 16:02:12 | nc-x | I dunno what is linetrace so can't check any further :) |
| 16:02:25 | FromGitter | <kaushalmodi> are you typing it out manually on the command line? |
| 16:03:22 | FromDiscord | <Shield> it's more than that, -d:release in the config file will show you that it's the release build in the compilation message, but it's actually debug, using -d:release in the command line works the right way |
| 16:03:48 | cavariux | I'm happy about the new release, my only doubt is, what was the reasoning behind moving the nimcache? |
| 16:04:47 | FromGitter | <kdheepak> @dom96 `nim compile ...` works but nimble doesn't |
| 16:05:22 | FromGitter | <kdheepak> Any help here will be appreciated. ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bacffc153c31c1111732a17] |
| 16:05:35 | FromGitter | <kdheepak> I've installed `brew install openssl` |
| 16:05:41 | FromGitter | <kaushalmodi> Shield: `-d:release` cannot work in .cfg or .nims |
| 16:06:11 | FromGitter | <kaushalmodi> See the *Note* in that nims docs I linked above (I learned that recently too) |
| 16:06:35 | FromGitter | <kaushalmodi> > Note: In general, the define switches can also be set in NimScripts using switch or --, as shown in above examples. Only the release define (-d:release) cannot be set in NimScripts. |
| 16:07:20 | FromGitter | <kaushalmodi> cavariux: it doesn't add bloat to the src dirs |
| 16:07:27 | * | bezaban2 joined #nim |
| 16:07:54 | FromDiscord | <Shield> so release doesn't work in nims and cfg, but it still shows that it's making a release build, and all this time I thought I was doing the release build for my projects... |
| 16:07:57 | FromGitter | <kaushalmodi> you can easily `rm -rf $XDG_CACHE_HOME/nim` |
| 16:08:08 | cavariux | kaushalmodi, what I'm more worried about is forgeting aboug it and nimcache using too much space in my home folder :/ |
| 16:08:15 | FromGitter | <kaushalmodi> Shield: One apparent different you will see is the change in binary size |
| 16:08:28 | FromGitter | <kaushalmodi> do `-d:release` from command-line vs in .cfg |
| 16:08:45 | FromGitter | <kaushalmodi> set --nimcache to /tmp/nim in you .cfg |
| 16:08:59 | FromGitter | <kaushalmodi> s/you/your |
| 16:09:05 | FromGitter | <kaushalmodi> (i haven't tried it though) |
| 16:09:07 | nc-x | Shield: Maybe open an issue for this as well. It might be possible to detect `release` defined in your config and maybe show a warning or atleast not print that it is working in release mode |
| 16:09:12 | cavariux | I'm in windows :p |
| 16:09:13 | * | bezaban2 quit (Remote host closed the connection) |
| 16:09:19 | FromGitter | <kaushalmodi> I have actually just set XDG_CACHE_HOME to `/tmp/$USER/.cache` |
| 16:09:54 | FromGitter | <kaushalmodi> cavariux: you can dedicate any dir you like for cache |
| 16:10:42 | FromGitter | <kaushalmodi> haven't tried.. |
| 16:10:55 | FromGitter | <kaushalmodi> can you set XDG_CACHE_HOME env var on Windows? |
| 16:11:03 | FromGitter | <kaushalmodi> does Nim respect that env var if you set it? |
| 16:11:59 | FromGitter | <kaushalmodi> nvm I see that for Windows it goes to $HOME/cache |
| 16:12:00 | FromGitter | <kaushalmodi> https://nim-lang.org/docs/nimc.html#compiler-usage-generated-c-code-directory |
| 16:12:24 | FromGitter | <kaushalmodi> s/$HOME\/cache/$HOME\/nimcache |
| 16:13:29 | cavariux | I guess I could move it to the temo folder of windows |
| 16:13:54 | cavariux | but I'm not sure if windows cleans that folder periodically |
| 16:15:15 | * | Calinou quit (Quit: http://quassel-irc.org - Discuter simplement. Partout.) |
| 16:18:05 | * | tzui quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 16:21:52 | * | jjido joined #nim |
| 16:23:27 | * | jjido quit (Client Quit) |
| 16:24:25 | * | tzui joined #nim |
| 16:26:08 | * | miran joined #nim |
| 16:28:29 | FromGitter | <kaushalmodi> cavariux: If Windows doesn't do it automatically, check out CCleaner |
| 16:29:21 | FromGitter | <kaushalmodi> https://www.ccleaner.com/ccleaner/features |
| 16:31:16 | * | gertvdijk joined #nim |
| 16:31:22 | FromDiscord | <demotomohiro> I also use windows and using ram disk. All files created on ram disk are gone everytime I reboot PC. |
| 16:31:42 | FromDiscord | <demotomohiro> It is like a tmpfs or /tmp on Linux |
| 16:31:42 | FromGitter | <dom96> Araq: this probably shouldn't 404: https://nim-lang.org/docs/c2nim.html |
| 16:32:41 | * | SenasOzys__ joined #nim |
| 16:32:51 | * | hethkar10 joined #nim |
| 16:33:20 | * | SenasOzys quit (Read error: Connection reset by peer) |
| 16:33:44 | * | revprez_stg1 joined #nim |
| 16:34:13 | * | hethkar10 quit (Remote host closed the connection) |
| 16:34:31 | * | tzui quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 16:36:00 | * | gertvdijk quit (Remote host closed the connection) |
| 16:38:03 | * | revprez_stg1 quit (Remote host closed the connection) |
| 16:38:10 | * | SenasOzys__ quit (Ping timeout: 246 seconds) |
| 16:39:34 | FromGitter | <dom96> @kdheepak you need to check what .dylib file brew installed |
| 16:41:34 | FromGitter | <codenoid> hi @dom96 long time no c |
| 16:42:10 | FromGitter | <Bennyelg> @dom96 any idea why this happen ? https://github.com/nim-lang/Nim/issues/9082 |
| 16:44:47 | * | SenasOzys joined #nim |
| 16:45:15 | * | PrimHelios quit (Quit: Leaving) |
| 16:45:17 | FromGitter | <dom96> no idea |
| 16:45:19 | FromGitter | <dom96> Check blame :) |
| 16:46:24 | FromGitter | <kdheepak> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bad096053c31c111173779f] |
| 16:46:51 | FromGitter | <kdheepak> How do I check all the paths nimble searches? |
| 16:48:18 | krux02 | I just looked at the round function, it it takes an argument for decimal rounding |
| 16:48:29 | krux02 | floating point numbers aren't stored decimal |
| 16:48:45 | krux02 | that argument doesn't make a lot of sence for a core low level math operation. |
| 16:50:05 | FromGitter | <Bennyelg> Anyone tried ⏎ https://clickhouse.yandex/ ? |
| 16:50:46 | * | stefantalpalaru quit (Quit: Leaving) |
| 16:51:03 | FromDiscord | <Shield> I think it's because of the floating point precision, since it's just "round0(x*mult)/mult" |
| 16:51:59 | * | darithorn joined #nim |
| 16:53:11 | FromGitter | <kdheepak> Is this right? ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bad0af77bd81c5b9dc20aed] |
| 16:56:05 | * | nc-x quit (Quit: Page closed) |
| 16:58:14 | FromGitter | <kdheepak> Hmm. I tried this on my other mac, and it seems to work fine. |
| 16:58:23 | FromGitter | <kdheepak> I'm not sure what is going on on this machine. |
| 16:58:36 | federico3 | Araq: there's no mention of RISC-V on https://nim-lang.org/blog/2018/09/26/version-0190-released.html - is it officially supported? |
| 16:58:59 | FromGitter | <mratsim> The fact that this concept works is just plain awesome: ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bad0c5301fb5f4ae1300f8a] |
| 16:59:36 | FromGitter | <kdheepak> @dom96 can you tell me where nimble searches for libcrypto? |
| 16:59:53 | FromGitter | <kdheepak> My other machine doesn't even have libcrypto installed from brew, and it seems to work on there. |
| 16:59:57 | FromGitter | <kdheepak> Not sure what is going on. |
| 17:02:20 | * | dkp2 joined #nim |
| 17:04:38 | * | dkp2 quit (Read error: Connection reset by peer) |
| 17:05:04 | * | riidom joined #nim |
| 17:06:56 | * | adelrune joined #nim |
| 17:07:16 | * | floppydh quit (Quit: WeeChat 2.2) |
| 17:07:59 | FromGitter | <Bennyelg> 1) 19.0 fantastic release. |
| 17:09:26 | FromGitter | <dom96> @kdheepak what macOS versions are you running on those machines? |
| 17:09:41 | FromGitter | <dom96> Typically the .dylibs are found in your PATH |
| 17:11:17 | FromGitter | <dom96> try `export DYLD_LIBRARY_PATH="/usr/local/lib"` |
| 17:11:40 | FromGitter | <dom96> It's up to the OS to find these, not Nimble |
| 17:12:01 | FromGitter | <dom96> Nimble just says "give me libcrypto.dylib or libcrypt.1.1.dylib ... etc" |
| 17:12:50 | * | Calinou joined #nim |
| 17:14:30 | cavariux | dom96 nim should gc the nimcache folder :p |
| 17:14:59 | cavariux | serious note: where do you find more apropiate to put the cache path in windows? |
| 17:16:37 | FromGitter | <kaushalmodi> why not just use the temp files dir? |
| 17:16:54 | FromGitter | <kaushalmodi> there are many ways online that show how to auto-delete stuff from there: https://www.easeus.com/recover-deleted-files/automatically-delete-temp-files.html |
| 17:17:22 | cavariux | I'm currently using it, and also nim and nimble uses the temp dir for some common compilations |
| 17:17:32 | cavariux | but wanted to know if dom96 had other idea |
| 17:18:13 | FromGitter | <kaushalmodi> understood |
| 17:18:25 | FromGitter | <tim-st> I think golang doesnt gc the cache files too |
| 17:18:55 | miran | shouldn't nimcache files be in ~/.cache/nim ? |
| 17:18:58 | FromGitter | <tim-st> well they have a `clean` command |
| 17:19:16 | cavariux | tim-st I was joking there, there's no point in a compiler gc the cache files haha, but I agree a clean command would be neat |
| 17:19:16 | miran | because now i see that i have lots of nim-related stuff all over ~/.cache |
| 17:19:28 | * | mr_yogurt quit (Max SendQ exceeded) |
| 17:19:53 | * | mr_yogurt joined #nim |
| 17:19:58 | * | tzui joined #nim |
| 17:20:03 | FromGitter | <tim-st> on windows it's `$HOME\nimcache\` |
| 17:20:05 | miran | for example, now it is not easy to delete all nim-related files at once. |
| 17:20:10 | cavariux | miran you can change it changing the $XDG_CACHE_HOME variable in posix |
| 17:20:30 | miran | i must carefully open every folder to see if it is or isn't related to nim |
| 17:20:48 | FromGitter | <tim-st> on windows it's easy, I can just delete `nimcache` |
| 17:21:28 | cavariux | miran also the docs says that it should be ~/.cache/nim/$projectname(_r|_d) on Posix |
| 17:21:41 | miran | these are some relatively old files/folders. is maybe this sorted out in v0.19 so it leaves more tidy ~/.cache? |
| 17:21:48 | FromGitter | <tim-st> maybe it's a bug on posix |
| 17:22:11 | FromGitter | <mratsim> wow people are creating impressive stuff in Nim in stealth mode: https://news.ycombinator.com/item?id=18085914 |
| 17:22:13 | FromGitter | <tim-st> but not sure why people always figure these things out after the release :\ |
| 17:22:20 | FromGitter | <tim-st> :D |
| 17:22:39 | miran | cavariux: i have ~/.cache/nim and inside of it are folders from today when i updated to v0.19 |
| 17:22:56 | miran | let me see if i run some file from my project where it will finish |
| 17:23:37 | miran | nope, still a mess |
| 17:24:03 | miran | i've compiled called `day01.nim`, and it created ~/.cache/day01_d folder |
| 17:24:10 | FromGitter | <tim-st> maybe on posix a trailing `\` is needed to be recognized as a folder, but just a guess |
| 17:24:34 | cavariux | hmm that's weird, I would expect some bugs in windows but irc dom96 and araq work on macos |
| 17:24:43 | FromGitter | <mratsim> no Araq uses windows |
| 17:25:13 | FromGitter | <mratsim> @tim-st we should have a path distinct string ... |
| 17:25:18 | miran | hmmmm https://github.com/nim-lang/Nim/issues/8599 |
| 17:25:44 | miran | is should be fixed but it seems it isn't |
| 17:26:26 | FromGitter | <mratsim> works for me, but I’m still on a devel from last week |
| 17:26:53 | * | Jesin quit (Quit: Leaving) |
| 17:27:14 | FromGitter | <kaushalmodi> miran: what' |
| 17:27:21 | FromGitter | <kaushalmodi> what's your OS? |
| 17:27:29 | FromGitter | <kaushalmodi> cache works as expected for me: https://ptpb.pw/6r9l/text |
| 17:28:08 | * | vlad1777d joined #nim |
| 17:28:17 | FromGitter | <tim-st> I compiled my code with 0.19 and had no problems :) |
| 17:28:56 | * | SenasOzys quit (Ping timeout: 252 seconds) |
| 17:29:18 | miran | hmmm, so nobody can reproduce this :/ |
| 17:29:33 | FromGitter | <tim-st> which os do you have? linux? |
| 17:29:35 | miran | i'll open an issue regardless |
| 17:29:40 | miran | yeah, manjaro linux |
| 17:30:02 | FromGitter | <tim-st> manjaro is good, would use it too, if they recognize energy is a thing |
| 17:30:19 | miran | energy? what do you mean? |
| 17:30:25 | FromGitter | <tim-st> battery |
| 17:30:35 | miran | eh, i'm using desktop |
| 17:30:45 | FromGitter | <tim-st> I can run manjaro inside windows 7 and it uses less energy than running it native |
| 17:30:48 | deepend | browsing nim-lang.org/docs, the source links all seem to be dead |
| 17:31:12 | deepend | https://github.com/nim-lang/Nim/tree/master/terminal.nim#L127 |
| 17:31:20 | FromGitter | <kaushalmodi> miran: Can you check if setting $XDG_CACHE_HOME helps? |
| 17:31:22 | miran | deepend: there is already open issue about that |
| 17:31:43 | deepend | ok |
| 17:32:01 | miran | @kaushalmodi please tell me exactly what to check and what to do. i'm not very familiar with env variables |
| 17:32:13 | FromGitter | <kaushalmodi> `echo $XDG_CACHE_HOME` |
| 17:32:21 | FromGitter | <kaushalmodi> looking at this code: https://github.com/nim-lang/Nim/blob/devel/compiler/options.nim#L516 |
| 17:32:30 | cavariux | miran I was about to link you to an issue related until I saw narimiran was the author haha |
| 17:32:37 | FromGitter | <kaushalmodi> it should have created the `/nim` subdir.. so I am surprised |
| 17:32:43 | miran | cavariux: ;) |
| 17:32:55 | miran | @kaushalmodi nothing is printed! |
| 17:33:10 | FromGitter | <kaushalmodi> hmm, so it is not set |
| 17:33:20 | FromGitter | <codenoid> is nim have something like http://crystalshards.xyz/ |
| 17:33:22 | FromGitter | <kaushalmodi> so it should default to the second arg of `getEnv` |
| 17:33:32 | FromGitter | <kaushalmodi> wondering if there's a bug there.. |
| 17:33:42 | FromGitter | <kaushalmodi> are you using bash? |
| 17:33:49 | FromGitter | <kaushalmodi> ^ miran |
| 17:33:56 | miran | fish |
| 17:34:13 | FromGitter | <kaushalmodi> hmm, does it understand `export` command? |
| 17:34:28 | FromGitter | <kdheepak> @dom96 I'm running High Sierra on the one that works (10.13.6) |
| 17:34:34 | FromGitter | <kaushalmodi> try `export XDG_CACHE_HOME=~/.cache` |
| 17:34:36 | FromGitter | <kdheepak> I'm running Sierra on the one that doesn't work. |
| 17:34:48 | FromGitter | <kaushalmodi> from that same terminal then run a nim compilation |
| 17:34:57 | FromGitter | <kaushalmodi> does it *then* create the `nim/` subdir? |
| 17:36:33 | FromGitter | <kaushalmodi> miran: well, that logic works fine.. ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bad1521aedb375b9c572629] |
| 17:36:45 | FromGitter | <kdheepak> @dom96 There are no matching results for crypto in nimble? I used github search and found no results (https://github.com/nim-lang/nimble/search?q=crypto&unscoped_q=crypto) |
| 17:36:48 | miran | i exported it, but now there are no cache files, not in .cache, not in .cache/nim !? |
| 17:37:16 | FromGitter | <kaushalmodi> what does `echo $XDG_CACHE_HOME` say now? |
| 17:37:29 | miran | ~/.cache |
| 17:37:35 | FromGitter | <kaushalmodi> what do you get when you eval the above snippet? |
| 17:38:46 | miran | `/home/miran/.cache/nim` (twice) |
| 17:39:02 | * | Jesin joined #nim |
| 17:39:06 | FromGitter | <kaushalmodi> umm.. check your nim.cfg |
| 17:39:10 | FromGitter | <kaushalmodi> anything funny in there? |
| 17:39:23 | FromGitter | <kaushalmodi> betting that that's the case :P |
| 17:39:36 | FromGitter | <mratsim> @codenoid nimble.directory |
| 17:39:47 | FromGitter | <mratsim> https://nimble.directory/ |
| 17:39:55 | miran | there is no nim.cfg file in that folder |
| 17:40:09 | miran | in that project folder where i run those files |
| 17:40:27 | FromGitter | <kaushalmodi> I mean in your `~/.config/nim` (if you have any) |
| 17:40:55 | FromGitter | <kaushalmodi> if not that, then I am out of ideas.. I'll follow that issue of yours as I am curious |
| 17:40:58 | miran | and now the file i've used to run that `getenv` is put once again in .cache |
| 17:41:11 | * | SenasOzys joined #nim |
| 17:43:14 | FromGitter | <codenoid> umm actualy, nimble init ask me a lot of question |
| 17:43:30 | FromGitter | <kaushalmodi> @codenoid that's an update! |
| 17:44:10 | FromGitter | <codenoid> so, how i can update nim with nimble |
| 17:44:20 | FromDiscord | <Shield> I'd like some help with macros |
| 17:44:21 | FromGitter | <mratsim> e choosenim update stable |
| 17:44:26 | FromGitter | <mratsim> use* |
| 17:44:30 | FromGitter | <codenoid> i just install nim and it's still 0.19 |
| 17:44:35 | FromGitter | <codenoid> ok |
| 17:44:42 | FromGitter | <mratsim> but 0.19 is the latest |
| 17:44:46 | FromGitter | <Clyybber> lol |
| 17:44:48 | FromGitter | <mratsim> it’s out since today |
| 17:45:17 | FromGitter | <kaushalmodi> lol |
| 17:45:21 | FromGitter | <codenoid> ```$ choosenim stable ⏎ Info: Version 0.19.0 already selected ⏎ ``` [https://gitter.im/nim-lang/Nim?at=5bad1730aedb375b9c5734d2] |
| 17:45:27 | FromGitter | <mratsim> @Shield can’t help will be back in 50 min |
| 17:45:40 | FromGitter | <mratsim> @codenoid you have the latest version already |
| 17:45:54 | FromDiscord | <Shield> okay I'll wait |
| 17:46:01 | FromGitter | <codenoid> '-' |
| 17:46:33 | FromGitter | <codenoid> i thought the latest version is 0.19.1 |
| 17:46:43 | FromGitter | <mratsim> 1) 19.1 is the current devel |
| 17:46:57 | FromGitter | <mratsim> odd minor for devel and even for stable release |
| 17:47:09 | miran | no, 0.xx.1 is devel version. until today it was 0.18.1 |
| 17:47:55 | FromGitter | <mratsim> @Shield you should ask away anyway, I’ll see the history when I’m back, maybe someone will unblock you in the mean time |
| 17:49:40 | krux02 | dom96, I saw some activity from you in the design how math.round behaves today so maybe some attention from you on this PR: https://github.com/nim-lang/Nim/pull/9089 |
| 17:50:32 | * | tzui quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 17:50:58 | * | tzui joined #nim |
| 17:51:57 | FromGitter | <codenoid> is nim have mongo driver |
| 17:52:26 | * | Trustable quit (Remote host closed the connection) |
| 17:52:35 | FromDiscord | <Shield> I want a macro which will add blocks to the definition of a procedure, something like add_block("proc_name"):block which will modify the proc accordingly, preferably with an extra parameter "priority" which will be used to sort the blocks before inserting them |
| 17:52:58 | FromGitter | <Clyybber> @mratsim Do you know why exactly https://github.com/nim-lang/Nim/issues/6348 was closed? |
| 17:53:05 | FromGitter | <Clyybber> It seems like it still fails |
| 17:54:18 | * | tzui quit (Client Quit) |
| 17:57:10 | FromGitter | <kdheepak> @dom96 I solved it by adding the following: `export DYLD_FALLBACK_LIBRARY_PATH=$DYLD_FALLBACK_LIBRARY_PATH:$(brew --prefix openssl)/lib` to my path |
| 17:57:42 | FromDiscord | <niv> hello! great work on shipping 0.19. really nice release. |
| 17:58:28 | FromGitter | <codenoid> NiGui is cool |
| 17:59:17 | * | francisl joined #nim |
| 18:00:47 | * | metsjeesus25 joined #nim |
| 18:01:20 | FromDiscord | <niv> i have a question. `($Args["--add-restypes"]).split(",").mapIt(a.split(":"))` worked fine on 0.18. 0.19 now requires `($Args["--add-restypes"]).split(",").map do (a: string) -> seq[string]: a.split(":")` .. why? |
| 18:01:49 | FromGitter | <zacharycarter> does anyone know if gara - is a replacement for ast_pattern_matching? |
| 18:01:59 | * | wh0pie joined #nim |
| 18:02:00 | FromGitter | <zacharycarter> in other words - can gara be used to pattern match AST as well? |
| 18:02:44 | * | metsjeesus25 quit (Remote host closed the connection) |
| 18:03:47 | miran | those of you who like the new release: https://old.reddit.com/r/programming/comments/9jdfwe/nim_version_0190_released/ please write it publicly ;) |
| 18:04:35 | FromGitter | <dom96> @kdheepak glad you got it fixed! Please report this as an issue on the Nimble repo, include your macOS version and any other details you think are relevant. |
| 18:06:10 | * | wh0pie quit (Remote host closed the connection) |
| 18:08:11 | FromDiscord | <niv> also, another question: i've had the habit of creating custom `write` procs for my own types (i.e. proc write*(io: Stream, t: MyType)). that used to work in 0.18, but now the generic one in streams.nim appears to shadow my custom writers. is this intended? have i been doing things wrong all the time? |
| 18:13:18 | * | pjb16 joined #nim |
| 18:14:37 | FromGitter | <kaushalmodi> niv: If there's a regression, then you should open an issue I think. btw did `a.split` in mapIt work before? or did you mean `it.split`? (that doesn't work either) |
| 18:15:06 | FromGitter | <kaushalmodi> niv: Are you basically trying to parse `"a:bc,d:ef,g:hi"` ? |
| 18:15:27 | * | abm joined #nim |
| 18:15:48 | FromGitter | <vivekimsit> I would like to know which editor you all use? |
| 18:15:49 | * | pjb16 quit (Read error: Connection reset by peer) |
| 18:15:54 | FromGitter | <vivekimsit> I use vscode and vim |
| 18:15:54 | Araq | niv: please open bug reports, even if we tell you "It's now this way" this change needs to be communicated/documented |
| 18:15:55 | FromGitter | <kaushalmodi> Emacs |
| 18:16:06 | FromGitter | <vivekimsit> :( |
| 18:16:09 | FromDiscord | <niv> sorry, it.split, yes. typo when i made that example. that used to work before. |
| 18:16:16 | miran | vscode with vim keybindings :) |
| 18:16:17 | FromDiscord | <niv> and correct on what i'm parsing. |
| 18:16:30 | FromGitter | <vivekimsit> @miran that would work |
| 18:16:33 | FromDiscord | <niv> i'll create tickets then |
| 18:16:44 | FromGitter | <vivekimsit> does it support autocompletion? |
| 18:16:57 | miran | @vivekimsit yes it does |
| 18:17:14 | FromGitter | <kaushalmodi> niv: yeah, looks like a bug from this example: ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bad1eaa7bd81c5b9dc2989b] |
| 18:17:16 | miran | you install Nim extension and you're ready to go |
| 18:17:35 | FromGitter | <kaushalmodi> uncommenting the last echo in there gives: ⏎ ⏎ > Error: type mismatch: got <seq[string]> but expected 'string = Alias' |
| 18:17:59 | FromGitter | <kaushalmodi> but.. the type of `it` is string, not `seq[string]` |
| 18:18:37 | FromGitter | <vivekimsit> installing... |
| 18:19:11 | miran | @kaushalmodi @niv put it inside of parantheses and it works |
| 18:19:18 | miran | `echo ("a:bc,d:ef,g:hi".split(",")).mapIt(it.split(":"))` |
| 18:19:27 | * | krako joined #nim |
| 18:19:47 | FromGitter | <vivekimsit> ```code paste, see link``` ⏎ ⏎ this compiles [https://gitter.im/nim-lang/Nim?at=5bad1f43691dc567dbb264e3] |
| 18:20:04 | FromGitter | <vivekimsit> but not this `this.properties.getOrDefault(name, newSection())` |
| 18:20:07 | FromGitter | <kaushalmodi> miran: I get the same error |
| 18:20:26 | FromGitter | <kaushalmodi> I am on Nim built with f6c5c636bb1a1f4e1301ae0ba5a8afecef439132 |
| 18:20:27 | FromGitter | <vivekimsit> sorry |
| 18:20:35 | FromGitter | <vivekimsit> this is correct method: ⏎ ⏎ `````` [https://gitter.im/nim-lang/Nim?at=5bad1f7341177e0bc7b73a87] |
| 18:20:38 | miran | ouch, i have used devel, not 0.19 |
| 18:20:41 | FromGitter | <vivekimsit> proc getSection*(this: Ini, name: string): Section = ⏎ return this.sections.getOrDefault(name) ⏎ ⏎ `````` [https://gitter.im/nim-lang/Nim?at=5bad1f79fe3778111008c5da] |
| 18:20:44 | miran | it fails in 0.19 indeed |
| 18:21:11 | FromGitter | <kaushalmodi> when did you last build devel? |
| 18:22:05 | miran | aaaand, that explains my error with cache |
| 18:22:21 | miran | i have installed the newest stable, but i didn't use it. time to close that issue |
| 18:22:22 | FromGitter | <kaushalmodi> :P |
| 18:22:41 | * | shashlick joined #nim |
| 18:23:16 | FromGitter | <mratsim> @Clyybber use —newruntime (or whatever the flag is called now) |
| 18:23:26 | miran | the moral of the story: kids, don't be stupid like miran! |
| 18:23:33 | krako | Hi everyone ! I'm new to nim-lang and nim community. To start learning the nim language, I want to code a CLI app like fzf (https://github.com/junegunn/fzf). Can it be done with Nim ? |
| 18:23:36 | FromGitter | <vivekimsit> @miran how do you run tests on saving changes? |
| 18:23:57 | miran | @vivekimsit which tests? |
| 18:24:22 | FromGitter | <mratsim> @vivekimsit use Travis/Appveyor, and trigger a build at each commit |
| 18:24:41 | FromGitter | <kaushalmodi> he's asking how to run nimble test or whatever on local buffer saves |
| 18:24:43 | miran | there is some "nim: check project" function of the nim extension, but i have never used it, i don't know what does it do |
| 18:25:17 | FromGitter | <kaushalmodi> *could have simply triggered a fn on `after-save-hook` in Emacs.. just saying* |
| 18:25:18 | FromGitter | <mratsim> @krako sounds like a nice project. The basic rule is that if it can be done in C it can be done in Nim |
| 18:25:55 | FromGitter | <vivekimsit> @mratsim I am talking about running unit tests on my local machine |
| 18:26:03 | FromDiscord | <niv> kaushalmodi: wandered off to create a ticket, sorry. didn't follow chat. https://github.com/nim-lang/Nim/issues/9093 |
| 18:26:03 | FromGitter | <vivekimsit> sorry but its a vscode specific issue |
| 18:26:21 | FromGitter | <mratsim> @Shield if you still have your macro issue, ask away |
| 18:26:21 | FromDiscord | <niv> and https://github.com/nim-lang/Nim/issues/9091 |
| 18:27:50 | * | TheLemonMan joined #nim |
| 18:28:07 | miran | @vivekimsit sorry i cannot help you with that, i've never had a code where i would have to run unit tests after each save |
| 18:28:47 | shashlick | hello |
| 18:28:56 | FromGitter | <mratsim> @vivekimsit the easiest is to define a `nimble test` task and run it. |
| 18:29:57 | TheLemonMan | yo Araq I revived nimfix in #9094 |
| 18:30:05 | FromGitter | <mratsim> Something like this: https://github.com/numforge/loop-fusion/blob/master/loopfusion.nimble though I have to update to 0.19 due to nimcache change |
| 18:30:18 | FromGitter | <iffy> Where are the docs for the `do` notation? |
| 18:30:42 | FromGitter | <mratsim> do just replaces a comma |
| 18:30:58 | FromGitter | <vivekimsit> @mratsim let me check that, I just shifted from vim to vscode so still new to this ecosystem |
| 18:31:02 | FromGitter | <mratsim> you can do foo(a, b) or foo(a) do: b |
| 18:31:06 | krako | Do I need two threads for the job or async code could do it ? (One thread for the main script and one that filter a list based on user input) |
| 18:31:29 | FromGitter | <iffy> @mratsim is that documented somewhere? |
| 18:31:30 | FromGitter | <kaushalmodi> @iffy https://nim-lang.org/docs/manual.html#procedures-do-notation |
| 18:31:35 | FromGitter | <mratsim> @vivekimsit it’s not vscode specific, it works in all editors, ef you hav an integrated terminal |
| 18:31:36 | FromDiscord | <Shield> how do I convert an untyped body to a NimNode? |
| 18:31:42 | FromGitter | <iffy> ah, not the tutorial |
| 18:31:44 | FromGitter | <iffy> thanks |
| 18:32:05 | FromGitter | <mratsim> @Shield, macros automatically converts all parameters to NimNode except the static one |
| 18:32:09 | FromGitter | <vivekimsit> I then don't know about it |
| 18:32:29 | FromGitter | <kaushalmodi> @iffy also see my notes https://scripter.co/notes/nim/#do-notation |
| 18:32:33 | FromGitter | <mratsim> `macro foo(a: untyped): untyped` will give you a in NimNode form |
| 18:32:36 | * | umaxx19 joined #nim |
| 18:32:58 | FromGitter | <mratsim> use a.treerepr to make sense of it |
| 18:33:13 | TheLemonMan | getting nimfix to replace all the isNil for you would be nice |
| 18:33:21 | krako | Do you think I could code my fzf project with just the standard lib or I'll need external libs ? |
| 18:35:59 | * | umaxx19 quit (Remote host closed the connection) |
| 18:36:34 | FromDiscord | <Shield> I can't seem to get this right, I want to create an stmtList(), then add the untyped body to it |
| 18:36:58 | FromGitter | <mratsim> @krako checking the go dependencies: https://github.com/junegunn/fzf/blob/master/glide.yaml isatty is probably something related to terminal, you have the terminal module in stdlib, runewidth related to utf8, you can import unicode it has Rune type, shellwords: not sure, tcell: not sure, crypto: not sure what it is used for in a fuzzer, if it’s murmurhash or minxish, check what @bung87 released just yesterday |
| 18:36:58 | FromGitter | ... https://github.com/Nim-NLP/minhash or use nimcrypto for more standard hash: https://github.com/cheatfate/nimcrypto |
| 18:37:42 | FromGitter | <kaushalmodi> niv: about that mapIt thing, sugar `=>` comes to the rescue |
| 18:37:46 | FromGitter | <mratsim> @Shield: macro foo(a: untyped): untyped = result = newStmtList(); result.add a |
| 18:37:58 | FromGitter | <kaushalmodi> ```code paste, see link``` ⏎ ⏎ > a of type seq[seq[string]] = @[@["a", "bc"], @["d", "ef"], @["g", "hi"]] [https://gitter.im/nim-lang/Nim?at=5bad2386691dc567dbb2833b] |
| 18:38:36 | FromGitter | <mratsim> mapIt(it.split(“:”)) would avoid allocating the anonymouse function though |
| 18:39:00 | FromDiscord | <niv> i don't really mind changing the code to suit a new paradigm. i'm more worried about it being a undocumented/unknown regression |
| 18:39:10 | FromGitter | <kaushalmodi> niv: I agree |
| 18:39:21 | FromDiscord | <niv> not really enthusiastic about surprises in software development :p |
| 18:39:22 | FromGitter | <kaushalmodi> I was just trying to find the most concise way that works right now |
| 18:39:47 | FromGitter | <mratsim> choosenim 0.18.0 ;) |
| 18:39:54 | FromGitter | <kaushalmodi> :) |
| 18:40:01 | FromDiscord | <niv> oh, i'm not upgrading to 0.19 :p just having a play |
| 18:40:23 | FromGitter | <mratsim> but if there is a lot of breakage we can have a 0.18.2 in a week |
| 18:40:27 | FromGitter | <mratsim> 1) 19.2 |
| 18:41:01 | FromDiscord | <niv> im not too worried about the macro breakage. what worries me more is that suddenly my `write` proc isn't picked anymore. a problem in the generic system sounds more serious to my naive ears |
| 18:41:31 | FromGitter | <mratsim> Don’t worry, we have TheLemonMan, he eats generics for breakfast |
| 18:42:51 | FromGitter | <kaushalmodi> +1 for TheLemonMan/Boy |
| 18:48:50 | FromDiscord | <Shield> now how to inject the content of the nimNode in the body of a proc? |
| 18:49:03 | * | francisl quit (Quit: francisl) |
| 18:49:03 | FromGitter | <vivekimsit> cannot run the task, I am using this setting for tasks.json: ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bad261f41177e0bc7b76a61] |
| 18:49:10 | FromGitter | <vivekimsit> and I use fish shell btw |
| 18:49:52 | * | francisl joined #nim |
| 18:51:04 | FromGitter | <vivekimsit> anyways can someone help me with my query please? |
| 18:51:19 | FromGitter | <vivekimsit> I posted before |
| 18:51:26 | FromGitter | <vivekimsit> regarding default value |
| 18:52:15 | FromGitter | <mratsim> is your repo open source? |
| 18:52:42 | FromGitter | <vivekimsit> yes |
| 18:52:54 | FromGitter | <mratsim> avec a look here: https://github.com/status-im/nimbus/blob/master/.vscode/tasks.json |
| 18:52:58 | FromGitter | <vivekimsit> but changes are local right now |
| 18:53:00 | FromGitter | <mratsim> have a look* |
| 18:53:39 | * | cbaines8 joined #nim |
| 18:53:59 | FromGitter | <mratsim> make sure to have actual tests to run in tests/t* by default or override the test tasts in your project.nimble |
| 18:54:24 | * | Calinou quit (Quit: http://quassel-irc.org - Discuter simplement. Partout.) |
| 18:54:38 | * | Calinou joined #nim |
| 18:55:32 | FromGitter | <vivekimsit> sure will check that |
| 18:56:03 | * | Calinou quit (Remote host closed the connection) |
| 18:56:18 | * | Calinou joined #nim |
| 18:57:00 | * | nsf joined #nim |
| 18:57:17 | FromGitter | <vivekimsit> sorry guys but posting a long traceback: ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bad280d7bd81c5b9dc2ce5c] |
| 18:57:42 | FromGitter | <vivekimsit> I am doing this: ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bad28258ddf5867dc831b8c] |
| 18:57:52 | * | cbaines8 quit (Remote host closed the connection) |
| 18:59:00 | * | solarnetone0 joined #nim |
| 18:59:19 | FromGitter | <mratsim> you’re passing an extra parameter |
| 18:59:37 | * | Ven`` joined #nim |
| 18:59:48 | FromGitter | <vivekimsit> but in the doc `proc getOrDefaultA, B (t: Table[A, B]; key: A; default: B): B` |
| 18:59:59 | FromGitter | <vivekimsit> I can pass a default value right? |
| 19:00:18 | FromGitter | <mratsim> I think the way to do the default parameter changed between 0.18 and 0.19, are you using the latest? |
| 19:00:34 | * | TheLemonMan quit (Quit: "It's now safe to turn off your computer.") |
| 19:00:37 | FromGitter | <mratsim> or the default param input was added in 0.19 |
| 19:01:27 | FromGitter | <vivekimsit> no 0.18 |
| 19:01:35 | FromGitter | <mratsim> update to 0.19 then |
| 19:01:53 | FromGitter | <vivekimsit> ok |
| 19:02:55 | * | solarnetone0 quit (Remote host closed the connection) |
| 19:03:27 | FromGitter | <vivekimsit> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bad297f53c31c111174541a] |
| 19:03:39 | FromGitter | <mratsim> choosenim update stable |
| 19:04:40 | * | javier24 joined #nim |
| 19:05:42 | FromDiscord | <Shield> How can I pass a NimNode to a macro and add things to it? I get "Error: cannot add to node kind: nkSym" |
| 19:05:49 | * | javier24 quit (Remote host closed the connection) |
| 19:07:01 | FromGitter | <mratsim> use a proc not a macro to recive NimNode |
| 19:07:06 | FromGitter | <mratsim> receive* |
| 19:07:39 | FromGitter | <mratsim> you can check this library, it has a lot of examples on working with NimNode: https://github.com/numforge/loop-fusion |
| 19:08:22 | FromDiscord | <Shield> okay, thanks |
| 19:08:24 | FromGitter | <mratsim> Or this one is even simpler: https://github.com/SimonDanisch/julia-challenge/pull/3/files |
| 19:09:03 | FromGitter | <Bennyelg> heheh |
| 19:10:04 | * | Jesin quit (Quit: Leaving) |
| 19:11:29 | FromDiscord | <phelps> Hey, just found Nim and have a few questions about working with some system libraries on OS X |
| 19:12:36 | FromDiscord | <phelps> I have a project that uses the mach system libraries on OS X, does anyone here have experience with these and Nim? |
| 19:12:54 | FromGitter | <Bennyelg> @kaushalmodi i guess mapIt is related to this change evalOnceAs(s2, s, compiles((let _ = s))) |
| 19:13:40 | FromGitter | <mratsim> @phelps you can compile Nim to C, C++ or Objective C, as long as those libraries have such an interface you can use them |
| 19:14:01 | Araq | phelps, we have little OSX specific wrapper/libraries but read posix.nim to see how to interface with Posix |
| 19:14:02 | FromGitter | <mratsim> compile to and also load from* |
| 19:14:26 | * | francisl quit (Quit: francisl) |
| 19:14:26 | FromDiscord | <phelps> the project is currently in C |
| 19:15:41 | * | francisl joined #nim |
| 19:16:30 | * | Ven`` quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 19:19:21 | FromGitter | <Bennyelg> Why I can't find anywhere anything about the usage of `{.inject.} pragma` |
| 19:22:22 | FromGitter | <kaushalmodi> @Bennyelg I don't understand most of the code in `mapIt` (not there yet).. so I just copied the authors based on git blame |
| 19:23:08 | FromGitter | <mratsim> @Bennyelg inject basically means replace the symbol there by the actual name taken from the parameter |
| 19:23:37 | FromGitter | <mratsim> look here: https://nim-lang.org/docs/manual.html#templates-identifier-construction |
| 19:24:18 | FromGitter | <mratsim> you replace “name” by myint |
| 19:24:46 | * | francisl quit (Quit: francisl) |
| 19:24:47 | FromGitter | <mratsim> note that if T and P were parameters they would also be replaced |
| 19:26:26 | FromGitter | <kaushalmodi> @mratsim based on experiments, looks like `mapIt` needs the op (outType) to be the same type as that of `it` |
| 19:26:47 | FromGitter | <kaushalmodi> mapIt code for ref: https://github.com/nim-lang/Nim/blob/f6c5c636bb1a1f4e1301ae0ba5a8afecef439132/lib/pure/collections/sequtils.nim#L663-L697 |
| 19:27:03 | Araq | just fyi, I'm fixing the documentation links |
| 19:27:13 | FromGitter | <kaushalmodi> woohoo! |
| 19:27:20 | Araq | or is there even more important things to do? |
| 19:27:34 | FromDiscord | <phelps> does anyone have experience compiling nim to a dylib? |
| 19:28:03 | FromGitter | <kaushalmodi> Araq: Will that fix the search locally too? |
| 19:28:27 | Araq | phelps: there is tests/dll and these tests are green |
| 19:28:36 | Araq | and show the required configuration |
| 19:28:41 | FromGitter | <mratsim> @kaushalmodi that doesn’t read like that. assume your op is (it + 1). ⏎ ⏎ the output type is type(it + 1). |
| 19:28:56 | FromGitter | <kaushalmodi> Araq: I can test using `htmltest` once you ping me |
| 19:29:11 | FromGitter | <mratsim> @phelps nim c —app:yourlib.dylib |
| 19:29:24 | Araq | kaushalmodi: what's wrong with the search? It requires ../dochack.js build from tools/dochack.nim |
| 19:30:30 | FromGitter | <kaushalmodi> Araq: I am doing `koch docs` |
| 19:30:41 | FromGitter | <mratsim> @phelps the main pifall is that after loading a nim dll, you need to do NimMain() to initialize the Nim runtime (even from another language). But I guess we need a blog post on that |
| 19:30:42 | FromGitter | <kaushalmodi> is there a manual step needed to get that .js? |
| 19:31:04 | FromGitter | <mratsim> @kaushalmodi ^ want to do some notes about Nim dynamic library? |
| 19:31:16 | FromGitter | <kaushalmodi> right now, locally, I get "Failed to load resource: net::ERR_FILE_NOT_FOUND dohack.js" |
| 19:31:45 | Araq | yeah |
| 19:31:57 | Araq | you need to do: |
| 19:32:04 | Araq | nim js tools/dochack.nim |
| 19:32:05 | FromGitter | <kaushalmodi> @mratsim I don't have notes, just one toy project.. but that creates .so |
| 19:32:09 | FromGitter | <kaushalmodi> what is .dylib |
| 19:32:16 | FromGitter | <mratsim> the .so ofOS |
| 19:32:23 | FromGitter | <mratsim> of MacOS |
| 19:32:27 | FromGitter | <kaushalmodi> Araq: Aha! then we are missing that in the nightly docs deploy too |
| 19:32:34 | FromGitter | <mratsim> you can name dynlib both .so or .dylib on Mac |
| 19:32:37 | FromGitter | <kaushalmodi> let me try |
| 19:32:37 | Araq | cp tools/nimcache/dochack.js docs/ |
| 19:32:47 | Araq | er, doc/ I guess |
| 19:34:09 | FromGitter | <kaushalmodi> @mratsim actually I have only used `--app:lib`: https://github.com/kaushalmodi/nim-emacs-module/blob/master/test/Makefile#L14 |
| 19:34:16 | * | PrimHelios joined #nim |
| 19:34:25 | FromGitter | <kaushalmodi> *time to update that Makefile to config.nims*.. I did not know NimScript then |
| 19:35:23 | FromGitter | <mratsim> time to actually do something productive :P |
| 19:35:32 | FromGitter | <Bennyelg> @mratsim thanks. |
| 19:36:07 | * | yonkapin joined #nim |
| 19:36:47 | FromGitter | <kaushalmodi> Araq: I'll work through those js generation steps in few mins |
| 19:36:55 | FromGitter | <kaushalmodi> but.. should that be part of `koch docs`? |
| 19:37:31 | Araq | yeah |
| 19:37:32 | * | yonkapin quit (Remote host closed the connection) |
| 19:38:28 | * | tcsiwula joined #nim |
| 19:39:56 | FromGitter | <kaushalmodi> I'll send a PR to fix nightly docs soon |
| 19:40:16 | * | Nolan[m] joined #nim |
| 19:40:20 | FromGitter | <kaushalmodi> I had to add these 2 lines after koch nim: ⏎ ⏎ ```nim js ./tools/dochack/dochack.nim ⏎ cp -f ./tools/dochack/nimcache/dochack.js ./doc/.``` [https://gitter.im/nim-lang/Nim?at=5bad3224fe3778111009433d] |
| 19:40:40 | FromGitter | <kaushalmodi> s/koch nim/koch docs |
| 19:41:04 | * | tcsiwula quit (Read error: Connection reset by peer) |
| 19:43:57 | * | tay_ joined #nim |
| 19:43:57 | Nolan[m] | Hey, trying to evaluate Nim for a project I've been fighting with Rust over--namely, building a library usable via Wasm that needs to interface with some native C libraries. I'm trying to rapidly figure out whether this will work before investing lots of time in a rewrite. I grabbed a nim.cfg from the forums that claims to compile via Emscripten, but how do I pass compiler args (I.e. `-d:emscripten`) to `nimble build`? |
| 19:44:56 | Nolan[m] | Also, is there anything like a REPL that works with 0.19? I found inim but it doesn't seem to build. |
| 19:45:03 | FromGitter | <kayabaNerve> For the Nimble build task you should be able to use Nimscript. |
| 19:45:06 | FromGitter | <kayabaNerve> `nim secret` |
| 19:46:07 | * | PrimHelios quit (Ping timeout: 240 seconds) |
| 19:46:52 | Nolan[m] | Oh cool. |
| 19:47:02 | Araq | Nolan[m], most projects need to be updated to work with 0.19 |
| 19:47:08 | Nolan[m] | Why the odd name? |
| 19:47:26 | * | tay_ quit (Client Quit) |
| 19:48:46 | Araq | because it's unsupported |
| 19:48:52 | * | RifRaf joined #nim |
| 19:48:52 | Nolan[m] | Now to figure out how to do something like `nim build -d:emscripten`. I have a nim.cfg that claims to set things up correctly, I just need to test that. |
| 19:48:59 | Nolan[m] | Er, `nimble build`. |
| 19:49:31 | FromGitter | <kaushalmodi> Nolan[m]: nim.cfg is old way |
| 19:49:37 | FromGitter | <kaushalmodi> NimScript is the new way :) |
| 19:49:50 | FromDiscord | <exelotl> "The {.this.} pragma has been deprecated." oh no :( |
| 19:50:00 | FromGitter | <kaushalmodi> Using config.nims (NimScript), you can have `nim emscripten foo.nim` do what you want |
| 19:50:15 | FromGitter | <kaushalmodi> you just create a NimScript task named `emscripten` in the config.nims |
| 19:50:56 | FromGitter | <kaushalmodi> About NimScript task, here's an example: https://github.com/kaushalmodi/hello_musl/blob/master/config.nims |
| 19:51:00 | Nolan[m] | Hmm, wait, so I ran `nimble init <myprojname>` and got myprojname.nimble. Is that not how I should start a new package? |
| 19:51:15 | FromGitter | <kaushalmodi> yes, that's how you start a new package |
| 19:51:39 | * | RifRaf quit (Remote host closed the connection) |
| 19:51:46 | FromGitter | <kaushalmodi> .. but you don't have to use just `nimble` commands |
| 19:52:21 | Nolan[m] | Got it. Then I visited https://forum.nim-lang.org/t/2838 and grabbed that nim.cfg, placing it in the top-level directory. |
| 19:52:49 | FromGitter | <mratsim> @Nolan for emscripten, check out nimx or nimes (Nes emulator) or zengine, they all have emscripten support. |
| 19:52:56 | Nolan[m] | So that config is what I'd replace with nimscript? |
| 19:53:15 | FromGitter | <kaushalmodi> give me few minutes.. can give a quick untested sample |
| 19:53:35 | FromGitter | <mratsim> nim.cfg is fine tbh. |
| 19:53:39 | Nolan[m] | OK, cool, so folks are actually doing this and not just experimenting? :) |
| 19:54:03 | FromGitter | <mratsim> Nimx is used in production for the Reel Valley Fb + Android + iOS 5 stars game |
| 19:54:45 | FromGitter | <mratsim> https://yglukhov.github.io/Making-ReelValley-Overview/ |
| 19:55:08 | FromGitter | <mratsim> Also we have a lot of interest at Status (Nim main sponsor) due to Ethereum moving to WASM |
| 19:56:21 | FromGitter | <kaushalmodi> Nolan[m]: Put this in `config.nims` in your project dir |
| 19:56:30 | FromGitter | <kaushalmodi> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bad35ee56877c463a941383] |
| 19:56:49 | * | francisl joined #nim |
| 19:56:58 | FromGitter | <kaushalmodi> docs for NimScript: https://nim-lang.org/docs/nims.html |
| 19:59:42 | FromGitter | <vivekimsit> @mratsim I also want to learn about blockchain, how to get started? |
| 20:05:07 | Nolan[m] | HmmHmm, so how do I compile with the `em` task? If I run `nim em src/waynav.nim` I see nothing but hints about which config file is being used. If I run `nim em c src/waynav.nim` I get `Error: arguments can only be given if the '--run' option is selected`. |
| 20:06:27 | * | booto15 joined #nim |
| 20:07:12 | * | booto15 quit (K-Lined) |
| 20:07:15 | FromGitter | <kaushalmodi> Nolan[m]: Try putting `setCommand("c")` at the very end, *inside* that task |
| 20:09:26 | FromGitter | <kayabaNerve> Araq not if you reinstall devel every 3 days. |
| 20:09:38 | FromGitter | <kayabaNerve> Or every commit... |
| 20:09:50 | FromGitter | <kayabaNerve> I don't. Just commenting about updating to 0.19. |
| 20:10:00 | Nolan[m] | Ah, thanks, that did it! One additional question, it seems `c` doesn't place its output in the current directory. Guessing I need to explicitly specify that? Seems a bit odd if so. |
| 20:10:35 | FromGitter | <kayabaNerve> What output? The nimcache or the bin? |
| 20:11:05 | FromGitter | <kaushalmodi> Nolan[m]: it places the binary in the same dir as the source .nim |
| 20:11:18 | FromGitter | <kaushalmodi> now this is a .nims so you can use the awesome procs from the nimscript module |
| 20:11:44 | FromGitter | <kaushalmodi> see `mvFile` etc from https://nim-lang.org/docs/nimscript.html#mvFile,string,string |
| 20:11:58 | Nolan[m] | Ah, so it does. Thanks! |
| 20:12:30 | * | tzui joined #nim |
| 20:12:46 | * | tzui quit (Client Quit) |
| 20:12:47 | FromGitter | <kaushalmodi> though.. you might not be able to do mvFile etc *after* setCommand |
| 20:13:05 | * | scott_gonzalez11 joined #nim |
| 20:13:09 | FromGitter | <kaushalmodi> you you would need another task that calls `nim em ..` and then you do `mvFile` etc in that other task |
| 20:13:17 | FromGitter | <kaushalmodi> s/you you/though you |
| 20:13:45 | Nolan[m] | Cool, thanks for helping me get started! |
| 20:13:46 | * | scott_gonzalez11 quit (Remote host closed the connection) |
| 20:13:59 | FromGitter | <kaushalmodi> sure, play around with .nims, it's pretty flexible |
| 20:16:30 | Nolan[m] | And there we go, Hello world in wasm. :) Now to see if I can get the same thing running in JS packages. |
| 20:16:42 | FromGitter | <mratsim> @vivekimsit it depends at which level you want to learn: do you want to learn to program on the blockchain, to program a blockchain or learn something else? |
| 20:17:06 | FromGitter | <kaushalmodi> Nolan[m]: Make sure you write about such things as you learn and then publish somewhere |
| 20:17:11 | FromGitter | <kaushalmodi> like Nim forum or your blog :) |
| 20:19:32 | Nolan[m] | Will certainly do if I can get this working. I've had so many false starts on this that I hope Nim gets me somewhere. Rust was working out well until I tried exporting my Rust interfaces to JS, at which point it threw up because the community seems to be abandoning the Emscripten target which I need. |
| 20:19:35 | Nolan[m] | So back to the drawing board... |
| 20:20:19 | * | kipari18 joined #nim |
| 20:21:40 | * | skelett quit (Quit: WeeChat 2.1) |
| 20:22:16 | FromGitter | <zacharycarter> Nolan[m] - what are you trying to do? I have quite a bit of experience with A) the Nim JS target B) Nim and emscripten C) Nim and wasm via emscripten |
| 20:23:09 | * | kipari18 quit (Remote host closed the connection) |
| 20:24:37 | FromGitter | <zacharycarter> Rust is abandoning emscripten - because they have a LLVM target - which Nim does not |
| 20:24:41 | FromGitter | <zacharycarter> thus Nim needs another compilation step |
| 20:25:02 | FromGitter | <zacharycarter> but if you're goal is WASM w/ Rust - you shouldn't need emscripten |
| 20:25:10 | FromGitter | <zacharycarter> again - not sure what you're trying to do exactly |
| 20:26:35 | Nolan[m] | I'm using Spatialite, a C library, and creating a nicer interface around that to do some custom data processing. I then want to export that interface to JS, which gets wrapped in either a browser or mobile UI for geospatial analysis. |
| 20:27:11 | FromDiscord | <Shield> I'm still failing to pass the name of a NimNode and make it contain the Ast of a body |
| 20:27:16 | FromGitter | <zacharycarter> well - you won't be able to combine the JS and the C backend |
| 20:27:23 | FromGitter | <zacharycarter> they're not 1:1 |
| 20:27:25 | * | trijntje26 joined #nim |
| 20:27:29 | Nolan[m] | Basically, I want to access Spatialite from either a browser or something like React Native to create an offline-capable mapping app. But I don't know the capabilities of my platforms yet, so I'm trying to keep my interface/development options open. |
| 20:27:39 | FromGitter | <zacharycarter> gotcha |
| 20:28:21 | Nolan[m] | I.e. I recently thought desktop browsers could interface with connected GPS hardware. Apparently they can't. Neither can browsers on ChromeOS. So I'm looking at wasm for mobile targets, but maybe native for desktop apps. |
| 20:28:24 | FromGitter | <mratsim> sounds fun. be sure to check Vega as well @Nolan: http://vega.github.io/ and what mapbox does, and maybe Superset: https://github.com/apache/incubator-superset |
| 20:29:48 | Nolan[m] | So what I need is a wasm library that creates a nicer interface around Spatialite. Rust is a non-starter, unless I want to port the entire Spatialite stack to Rust and use wasm32-unknown-unknown. :/ |
| 20:30:07 | * | francisl quit (Quit: francisl) |
| 20:31:16 | FromGitter | <mratsim> @Nolan you should look into @Vindaar Neural network demo: https://github.com/Vindaar/NeuralNetworkLiveDemo, it combines C backend Arraymancer https://github.com/mratsim/Arraymancer with JS backend Plotly: https://github.com/brentp/nim-plotly |
| 20:32:14 | Nolan[m] | Thanks, will check all those projects out. |
| 20:32:18 | FromGitter | <zacharycarter> thoughts on this color palette for a personal website - http://colorpeek.com/#f4f7f4,778994,f0c743,b63b50,231926 ? |
| 20:32:30 | FromGitter | <zacharycarter> I want to kind of keep with a Nim(ish) theme |
| 20:32:43 | FromGitter | <zacharycarter> but still have some uniqueness |
| 20:34:22 | * | trijntje26 quit (Ping timeout: 244 seconds) |
| 20:37:59 | FromGitter | <bung87> why not developing base one electron |
| 20:38:03 | FromGitter | <bung87> on |
| 20:38:23 | * | miran quit (Ping timeout: 245 seconds) |
| 20:39:56 | Nolan[m] | Two reasons. First, I'll want to run this on mobile devices, and to my knowledge Electron doesn't work on mobile. Second, this is GPS navigation software for blind users, and Electron is inaccessible under Linux. |
| 20:43:45 | FromDiscord | <Shield> "static: var node = newStmtList()" makes it possible to add NimNodes to node when referred to it directly but I can't pass node as a variable for a macro or a proc to add an untyped body to it |
| 20:44:28 | FromGitter | <bung87> hmm have no idea about application for blind users, seems this is the major problem.. |
| 20:45:04 | Araq | <kaushalmodi>: fixed and uploaded new docs |
| 20:45:34 | FromGitter | <kaushalmodi> thanks! And I have that PR for fixing dochack for devel docs |
| 20:46:00 | FromGitter | <kaushalmodi> the PR is passing (there's nothing it's doing to fail the tests) |
| 20:46:07 | FromGitter | <kaushalmodi> affect only before_deploy stage |
| 20:46:27 | FromGitter | <kaushalmodi> https://github.com/nim-lang/Nim/pull/9096 |
| 20:46:40 | copygirl | Aww, no `from module import thing as custom` D: |
| 20:46:57 | * | Vladar quit (Remote host closed the connection) |
| 20:47:33 | copygirl | I guess `as` is module renaming, not identifier. |
| 20:48:14 | FromGitter | <bung87> there’s a issue for this ,not implemented. |
| 20:48:47 | FromGitter | <kaushalmodi> Araq: Tested the Source/Edit fix to work perfectly in local build. Thanks! |
| 20:49:31 | FromGitter | <bung87> I tried once ,very new to nim at that time :) |
| 20:54:06 | FromDiscord | <Shield> help! |
| 20:55:47 | Nolan[m] | Are exporting enums to C still unsupported? |
| 20:56:02 | * | krako left #nim (#nim) |
| 20:56:20 | Araq | Nolan[m], yeah and in fact header file generation is hardly supported at all |
| 20:57:48 | Nolan[m] | Not sure I'd need header generation, I just need symbols to be exported so Emscripten can see them. |
| 20:58:35 | Nolan[m] | So is there another data structure that might work in place of enums if I need C interop? Something like, say, the classic Direction enum with nort, northNortheast, etc.? |
| 20:59:28 | * | BlackBishop15 joined #nim |
| 21:00:06 | Araq | I think your best option is an {.emit: """...""".} in this case |
| 21:00:41 | Nolan[m] | Thanks, I'll read up on that. |
| 21:01:02 | * | BlackBishop15 quit (Remote host closed the connection) |
| 21:01:35 | PMunch | https://github.com/drujensen/fib ah, that's better. Now Nim is the fastest :) |
| 21:01:51 | * | jjido joined #nim |
| 21:02:50 | FromGitter | <mratsim> @Nolan, Electron works on Linux |
| 21:03:30 | Nolan[m] | mratsim: Not with my screen reader it doesn't. :) |
| 21:03:42 | FromGitter | <mratsim> @PMunch is Nim cpp backend so much faster than the C backend? Native exceptions? |
| 21:03:47 | FromGitter | <mratsim> Why* |
| 21:04:03 | PMunch | I have no idea, but when I tested it it was 16x faster than the C back-end |
| 21:04:13 | PMunch | He apparently didn't see the same speed-up though |
| 21:05:29 | PMunch | http://ix.io/1nIG/Nim <- these are the results I get |
| 21:06:22 | FromGitter | <mratsim> mmm, I think there is some constant folding going on, that doesn’t make sense |
| 21:07:35 | FromGitter | <mratsim> why is crystal using uint32 :? |
| 21:10:45 | federico3 | nim 0.19.0 is looking for /etc/nim/nim.cfg instead of /etc/nim.cfg |
| 21:12:16 | FromGitter | <tim-st> c uses long, is that an alias of int32? |
| 21:13:29 | FromGitter | <mratsim> not sure, it’s always confusing |
| 21:13:33 | FromGitter | <tim-st> btw why should nim be faster than cpp? |
| 21:13:43 | FromGitter | <tim-st> I mean nim with cpp backend |
| 21:14:05 | * | Kristine17 joined #nim |
| 21:14:21 | FromGitter | <tim-st> the only thing I can imagine is that there is a huge memory difference |
| 21:14:29 | * | Kristine17 quit (Remote host closed the connection) |
| 21:14:54 | FromGitter | <mratsim> Actually it might be because everyone else is using uint32 instead of uint64 |
| 21:15:25 | FromGitter | <tim-st> wikipedia says long is signed and atleast of size 32 |
| 21:15:29 | FromGitter | <mratsim> before all computations they would need to zero-extend the result number which adds additional cost |
| 21:15:31 | FromGitter | <tim-st> (in c) |
| 21:15:45 | FromGitter | <mratsim> the input not result |
| 21:15:51 | FromDiscord | <Shield> how to convert an untyped body to a NimNode so I can pass it to a proc? |
| 21:17:21 | FromGitter | <Bennyelg> @PMunch The mem section is annoying |
| 21:17:23 | * | mrshu quit (Remote host closed the connection) |
| 21:17:29 | FromGitter | <Bennyelg> this is not a test, mem is a mem |
| 21:18:34 | PMunch | Bennyelg, mem section? |
| 21:18:35 | * | gangstacat quit (Quit: Ĝis!) |
| 21:18:52 | FromGitter | <ephja> do buffered sockets result in received data being copied twice? |
| 21:18:57 | FromGitter | <Bennyelg> (https://files.gitter.im/nim-lang/Nim/xavX/image.png) |
| 21:19:20 | Araq | ephja, good question, I don't think we try to hard to avoid copies |
| 21:19:28 | PMunch | Oh yeah, I was considering to add a Nim mem example |
| 21:19:38 | PMunch | It's super easy to write and just as fast as the others |
| 21:19:42 | FromGitter | <Bennyelg> memoization shoud be enouge |
| 21:19:45 | FromGitter | <Bennyelg> should be* |
| 21:19:49 | Araq | keep in mind that every caching layer in your CPU produces "copies" |
| 21:23:55 | PMunch | http://ix.io/1nIL/Nim <- there you go |
| 21:24:06 | PMunch | With memorization |
| 21:25:02 | FromDiscord | <Shield> is there really no way to store an untyped body in a nimNode |
| 21:25:30 | FromGitter | <mratsim> @Shield I already gave you an example |
| 21:26:18 | FromDiscord | <Shield> whenever I try to store the result it gives me an error |
| 21:26:32 | FromGitter | <mratsim> @PMunch, it’s a bench on recursion :P I prefer the Haskell one though: ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5bad4b0898245f4ae093b3d0] |
| 21:26:43 | FromGitter | <mratsim> @Shield can you post a gist |
| 21:29:47 | FromGitter | <mratsim> ```code paste, see link``` ⏎ ⏎ ^ @Shield [https://gitter.im/nim-lang/Nim?at=5bad4bcaeba8e60bc6560c66] |
| 21:34:35 | FromGitter | <zacharycarter> would anyone be interested in a live stream of Nim web-dev? |
| 21:34:55 | FromGitter | <zacharycarter> at this point I have a hot-reloading JS / CSS site that I'm building as my personal website |
| 21:35:03 | * | PMunch quit (Remote host closed the connection) |
| 21:35:05 | Araq | live streams are always fun :-) |
| 21:35:21 | Araq | I would watch it, if I can |
| 21:35:37 | FromGitter | <zacharycarter> okay - I think I'm going to spend some time at my parent's house this weekend - so I'll see what I can do as far as setting up a camera and recording |
| 21:36:06 | FromGitter | <zacharycarter> thanks for the encouragement :D |
| 21:36:35 | * | SenasOzys quit (Quit: Leaving) |
| 21:36:53 | * | SenasOzys joined #nim |
| 21:37:54 | FromGitter | <zacharycarter> I will try to start super early on Saturday morning - so that all Nimmers might have a chance to attend |
| 21:38:43 | * | bozaloshtsh left #nim ("Killed buffer") |
| 21:38:46 | * | bozaloshtsh joined #nim |
| 21:40:24 | Araq | what's that in UTC? |
| 21:40:54 | FromGitter | <zacharycarter> I just spoke with my rents, and they're fine with the idea - so let me set up a time and I'll send that info out and put out a tweet as well |
| 21:40:56 | FromDiscord | <Shield> @mratsim https://pastebin.com/XGaygUX3 |
| 21:42:22 | FromDiscord | <Shield> I'm trying to make a nimNode which I can push blocks separately and get the whole result in the end |
| 21:42:59 | * | nsf quit (Quit: WeeChat 2.2) |
| 21:43:07 | Araq | Shield, that's entirely possilbe (and also bad style, but ok) |
| 21:43:31 | * | guillom3 joined #nim |
| 21:43:44 | * | guillom3 quit (Remote host closed the connection) |
| 21:45:16 | FromDiscord | <Shield> can I get an example please? it's hard to translate a body to a nimNode that you can pass around to other nimNodes, "macro foo(body:untyped):untyped" seem to inject its code directly without a way to capture it, passing a nimNode to a macro doesn't work |
| 21:46:12 | FromGitter | <zacharycarter> https://calendar.google.com/event?action=TEMPLATE&tmeid=NDc1aGIyZGY3ZW8yM3ZtYnExN2oydTA2bzQgY2FydGVyemFAbQ&tmsrc=carterza%40gmail.com |
| 21:47:03 | * | abm quit (Read error: Connection reset by peer) |
| 21:47:34 | FromDiscord | <Shield> i'd really appreciate if there's a template where you can pass a block and will returns a nimNode |
| 21:48:09 | FromGitter | <zacharycarter> I need a better way to share this w/o giving out my email |
| 21:48:15 | FromGitter | <zacharycarter> over twitter |
| 21:48:46 | * | inhahe1 joined #nim |
| 21:49:08 | FromGitter | <zacharycarter> I guess there's not a better way |
| 21:49:09 | FromGitter | <zacharycarter> oh well |
| 21:49:36 | FromGitter | <zacharycarter> what I get for using google |
| 21:50:09 | FromGitter | <tim-st> when I call `len` on `UncheckedArray[char]` it uses the len of `cstring`? is it expected? |
| 21:50:24 | FromGitter | <tim-st> not every UncheckedArray is null terminated |
| 21:50:27 | * | abm joined #nim |
| 21:51:21 | * | gangstacat joined #nim |
| 21:53:02 | FromGitter | <tim-st> ah, I see, it does it because the underlying type is just an array and array is null terminated, so this "works" but is highly dangerous |
| 21:54:09 | FromGitter | <zacharycarter> https://twitter.com/Hahaitsfunny/status/1045431375476805632 |
| 21:54:11 | * | inhahe1 quit (Ping timeout: 260 seconds) |
| 21:54:12 | FromGitter | <zacharycarter> in case anyone wants to RT |
| 21:57:06 | * | berendt_5 joined #nim |
| 21:58:33 | FromGitter | <zacharycarter> would someone be so kind as to check out the tweet - and see if the event link works for them? If not I'll work with something other than google calendar |
| 21:58:55 | * | berendt_5 quit (Remote host closed the connection) |
| 21:59:53 | FromGitter | <zetashift> the link works, but that page is confusing like hell |
| 21:59:54 | FromGitter | <mratsim> @Shield macro cannot receive NimNode, use proc for that |
| 22:00:26 | FromGitter | <mratsim> proc that have a Nimnode arg or return value have the same semantics as macros |
| 22:01:17 | * | flokli23 joined #nim |
| 22:01:32 | * | flokli23 quit (Remote host closed the connection) |
| 22:02:21 | FromGitter | <tim-st> is `cstring.len` O(n)? |
| 22:02:45 | FromGitter | <tim-st> (or is there some extra hidden field for length) |
| 22:02:46 | FromDiscord | <Shield> the problem is getting a macro to returna nimNode from an untyped body |
| 22:03:16 | FromGitter | <ephja> O(n) |
| 22:03:36 | FromGitter | <tim-st> thanks |
| 22:04:43 | FromDiscord | <Shield> something like var node = treeFromBlock(block) |
| 22:06:18 | * | inhahe27 joined #nim |
| 22:07:41 | * | inhahe27 quit (Remote host closed the connection) |
| 22:14:12 | * | lebarondemerde9 joined #nim |
| 22:16:53 | * | lebarondemerde9 quit (Remote host closed the connection) |
| 22:17:52 | * | elrood quit (Quit: Leaving) |
| 22:19:24 | * | wgrant25 joined #nim |
| 22:20:57 | * | twan_ joined #nim |
| 22:21:39 | * | krux02 quit (Remote host closed the connection) |
| 22:21:45 | * | Calinou quit (Quit: http://quassel-irc.org - Discuter simplement. Partout.) |
| 22:23:35 | * | wgrant25 quit (Remote host closed the connection) |
| 22:25:10 | * | twan_ quit (Remote host closed the connection) |
| 22:28:32 | FromDiscord | <Shield> solved it, but it's kinda convoluted that you need a temp nimNode and a macro to transform an untyped body to a nimNode |
| 22:31:22 | Nolan[m] | Can I get a variable's type in `nim secret`? |
| 22:34:47 | Nolan[m] | What I'm actually trying to do is figuring out the return type of a function that returns a sequence of tuples of the form (string, string, int). |
| 22:35:11 | Nolan[m] | But seq[tuple] doesn't seem to be it, unless I'm doing something wrong at the call site. |
| 22:35:55 | cavariux | Nolan maybe this could help you https://nim-lang.org/docs/tut1.html#advanced-types-tuples |
| 22:37:02 | Nolan[m] | RIght, I've seen that. I guess what I'm unclear on is whether or not I have to define function return types in a `type` block? |
| 22:42:50 | FromGitter | <kayabaNerve> Named tuples have their fields as part of the type. |
| 22:44:01 | FromGitter | <kayabaNerve> I think it's stupid. IMO, the idea of a named tuple is the names are for ease. Else, you would just just use objects. |
| 22:44:25 | FromGitter | <kayabaNerve> I think dom96 agreed with that, but Araq disagreed and said it fixed a lot of issues when you forced naming conventions. |
| 22:45:28 | FromGitter | <kayabaNerve> I forget dom's final stance... |
| 22:47:16 | FromGitter | <zacharycarter> @zetashift I don't have twitch / don't think I'll have time to troubleshoot / set it up before Saturday AM :/ |
| 22:48:25 | FromGitter | <kaushalmodi> Nolan[m]: haven't tried.. don't typetraits work in nim secret? |
| 22:50:21 | FromDiscord | <Shield> so far it seems impossible to access a NimNode variable unless it's in the same static body, even then attempting to apply what's inside it yields an error "attempt to access a nil address" |
| 22:50:58 | FromDiscord | <Shield> I guess you can't make a global nimNode, fill it then use its content? |
| 22:56:30 | * | armdale9 joined #nim |
| 22:58:02 | FromGitter | <codenoid> 05:57am still coding with nim <3 |
| 22:59:18 | FromGitter | <bung87> what project you work on? |
| 22:59:45 | * | kirkland15 joined #nim |
| 23:00:16 | * | armdale9 quit (Remote host closed the connection) |
| 23:00:56 | * | kirkland15 quit (Remote host closed the connection) |
| 23:00:57 | * | SenasOzys quit (Remote host closed the connection) |
| 23:01:14 | * | SenasOzys joined #nim |
| 23:03:52 | * | pips2 joined #nim |
| 23:05:07 | * | zama quit (Quit: leaving) |
| 23:05:39 | * | pips2 quit (Remote host closed the connection) |
| 23:06:56 | FromGitter | <dm1try> > What I'm actually trying to do is figuring out the return type of a function that returns a sequence of tuples of the form (string, string, int). ⏎ > But seq[tuple] doesn't seem to be it, unless I'm doing something wrong at the call site. ⏎ ⏎ Nolan, seq[(string, string, int)] ? [https://gitter.im/nim-lang/Nim?at=5bad6290cdc50131726383b3] |
| 23:07:02 | * | finnzi22 joined #nim |
| 23:09:13 | * | zama joined #nim |
| 23:09:53 | FromDiscord | <exelotl> uhhh I don't understand why this simple template usage isn't working https://gist.github.com/geckojsc/3d6a4a4c7b98b2e7d44ffb181429dc5c |
| 23:10:37 | FromDiscord | <exelotl> gives "Error: undeclared identifier: 'derp'" |
| 23:10:48 | FromGitter | <codenoid> social media bot @bung87 |
| 23:11:09 | * | finnzi22 quit (Remote host closed the connection) |
| 23:11:10 | FromDiscord | <exelotl> but the template takes an untyped parameter so why would it check whether derp is defined? |
| 23:15:27 | AlexMax | Does Nim come with a build in way to embed a windows manifest.xml? |
| 23:15:37 | FromGitter | <bung87> ah I have done something in that domain |
| 23:15:53 | FromGitter | <zacharycarter> going to use twitch instead of hangouts for the broadcast Saturday morning - https://www.twitch.tv/zachary_carter |
| 23:18:28 | * | thomasross joined #nim |
| 23:18:41 | AlexMax | Hrm |
| 23:18:57 | AlexMax | Koch just links in a .res/.o file directly...somehow\ |
| 23:19:12 | AlexMax | in order to apply an icon to an executable |
| 23:32:21 | * | Wessie16 joined #nim |
| 23:37:02 | FromGitter | <gogolxdong> cannot find reel valley in google play. |
| 23:37:47 | * | Wessie16 quit (Ping timeout: 240 seconds) |
| 23:41:24 | FromGitter | <zacharycarter> link seems dead |
| 23:42:48 | FromGitter | <zacharycarter> all I know is - I wish I had the city-build graphics that reel valley has - except sci-fi military and alien themed :P |
| 23:44:11 | FromGitter | <kdheepak> Some pages in the documentation are broken? |
| 23:44:19 | FromGitter | <kdheepak> https://nim-lang.org/docs/parseopt2.html |
| 23:44:30 | FromGitter | <kdheepak> How do I find the current page? |
| 23:47:54 | * | noonien quit (Quit: Connection closed for inactivity) |
| 23:50:09 | * | gangstacat quit (Quit: Ĝis!) |
| 23:56:06 | * | Apocalypse15 joined #nim |
| 23:57:34 | FromGitter | <kaushalmodi> Can you open an issue for that? |
| 23:57:45 | FromGitter | <kaushalmodi> Btw parseopt2 is deprecated |
| 23:57:51 | FromGitter | <kaushalmodi> Have a look at https://nim-lang.org/docs/parseopt.html |
| 23:57:57 | FromGitter | <kaushalmodi> That link works too. |
| 23:58:12 | * | Apocalypse15 quit (Remote host closed the connection) |
){kind=link}
){kind=link}
){kind=link}
){kind=link}