| 00:00:07 | * | minus joined #nim |
| 00:00:30 | * | Flox42 quit (Remote host closed the connection) |
| 00:00:31 | * | M-max quit (Read error: Connection reset by peer) |
| 00:00:31 | * | M-Quora quit (Remote host closed the connection) |
| 00:01:08 | * | niv joined #nim |
| 00:04:41 | * | SianaGearz joined #nim |
| 00:04:41 | * | SianaGearz quit (Changing host) |
| 00:04:41 | * | SianaGearz joined #nim |
| 00:04:54 | * | bjz joined #nim |
| 00:06:08 | * | Flox42 joined #nim |
| 00:11:48 | * | gokr joined #nim |
| 00:16:07 | * | bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 00:16:44 | * | yglukhov joined #nim |
| 00:20:55 | * | couven92 joined #nim |
| 00:21:03 | * | yglukhov quit (Ping timeout: 240 seconds) |
| 00:24:02 | * | fredrik92 quit (Ping timeout: 258 seconds) |
| 00:25:03 | * | libman quit (Ping timeout: 265 seconds) |
| 00:35:48 | * | mcc joined #nim |
| 00:39:51 | * | libman joined #nim |
| 00:58:52 | * | gokr quit (Ping timeout: 240 seconds) |
| 01:03:35 | * | gokr joined #nim |
| 01:08:39 | * | elrood quit (Quit: Leaving) |
| 01:14:02 | * | brson quit (Quit: leaving) |
| 01:18:57 | * | couven92 quit (Quit: Shutting down...) |
| 01:25:34 | * | irrequietus quit (Read error: Connection reset by peer) |
| 01:25:53 | * | irrequietus joined #nim |
| 01:31:14 | * | endragor joined #nim |
| 01:33:11 | * | irrequietus quit () |
| 01:35:42 | * | endragor quit (Ping timeout: 265 seconds) |
| 01:37:55 | * | bjz joined #nim |
| 01:42:09 | * | chemist69 quit (Disconnected by services) |
| 01:42:09 | * | chemist69_ joined #nim |
| 01:46:53 | * | der-landgraf quit (Quit: WeeChat 1.5) |
| 01:47:04 | * | bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 01:48:36 | * | bjz joined #nim |
| 02:05:10 | * | ftsf_ joined #nim |
| 02:07:22 | mcc | so i write a function like proc model() = return [1,2,3,4,5,6] ... and it says "no return type declared" is that not something nim can figure out the type of? |
| 02:08:06 | mcc | like, i feel like it does no, because if i give it the "wrong" type it says app.nim(25, 12) Error: type mismatch: got (Array constructor[0..23, float]) but expected 'array[int, float]' |
| 02:11:33 | ftsf_ | you need to specify the return type with : |
| 02:11:52 | ftsf_ | proc model(): array[6,int] = ... |
| 02:13:33 | ftsf_ | maybe it should be able to figure it out itself, but i don't think it can so far |
| 02:14:12 | * | gokr quit (Ping timeout: 250 seconds) |
| 02:14:50 | * | Goba_ joined #nim |
| 02:17:10 | mcc | ok |
| 02:18:27 | * | hooo quit (Ping timeout: 264 seconds) |
| 02:18:48 | * | yglukhov joined #nim |
| 02:23:28 | * | yglukhov quit (Ping timeout: 265 seconds) |
| 02:25:03 | libman | Can I read the nim compiler's --out argument from my code at compile time? |
| 02:26:17 | libman | I want my program to work slightly differently when evoked under different names (which is quite common in Unix, with different names `ln`ed to the same executable). |
| 02:27:00 | libman | getAppFilename() resolves symlinks, so that approach didn't work. |
| 02:30:39 | libman | Err, ignore the last line, paramStr(0) works fine. |
| 02:30:44 | * | M-Quora joined #nim |
| 02:30:52 | * | M-max joined #nim |
| 02:34:40 | ftsf_ | libman, did you solve your problem? |
| 02:35:25 | libman | Still curious about the approach of generating different code based on nim compiler's --out: value. |
| 02:36:20 | ftsf_ | hmm sounds like you want it to behave the same regardless of the --out but behave differently depending on the filename it's executed by? |
| 02:36:25 | ftsf_ | from what you said |
| 02:37:51 | mcc | Hmm |
| 02:37:55 | mcc | Behold, a webgl app written in Nim https://data.runhello.com/j/dry/2/ |
| 02:38:23 | libman | Well, there are two options: *1* different behavior at run-time based on symlink used to execute, which is the no-brainer method I went with, or *2* figure out if/how to access nim compiler's settings at compile time, which I'm still curious about. |
| 02:38:59 | libman | Like how C has special preprocessor constants like __FILE__ (for the source file, not the output executable, which is what I'm looking for) |
| 02:39:22 | * | beatmox quit (Ping timeout: 244 seconds) |
| 02:40:54 | * | beatmox joined #nim |
| 02:42:51 | ftsf_ | libman, what about just using a define and when? i know it's not the output file but should do the trick |
| 02:43:05 | libman | Yup, I just found that a second ago. |
| 02:43:07 | ftsf_ | mcc, \o/ |
| 02:43:25 | ftsf_ | mcc, easy to do? |
| 02:43:27 | mcc | ftsf_ :D i will probably post something on the nim forum this weekend about how i did it |
| 02:43:45 | ftsf_ | i'd be keen to rewrite my webgl app in nim, js is a mess |
| 02:43:56 | mcc | ftsf_: Uh... well... I spent a month writing the binding script to expose the WebGL API to Nim... |
| 02:44:03 | ftsf_ | heh, i see |
| 02:44:05 | mcc | ftsf_: But once I did that it was pretty easy! ^_^; |
| 02:44:07 | ftsf_ | =) |
| 02:44:53 | ftsf_ | gonna put the bindings on github/nimble? |
| 02:45:49 | mcc | My goal is to reach a point where I can write a single nim app and compile it to both sdl+opengl and js+webgl |
| 02:46:24 | ftsf_ | mcc, sounds nice =) |
| 02:46:28 | mcc | ftsf_: I'm planning to put the binding script on npm, maybe as soon as this weekend |
| 02:46:33 | libman | Nope, `when defined` is just for booleans... |
| 02:46:49 | libman | mcc: very cool! :D |
| 02:47:03 | mcc | ftsf_: if people would actually use it that would be an incentive to do it sooner rather than later :) |
| 02:48:26 | ftsf_ | mcc, i'd be keen |
| 02:48:41 | mcc | if whatever opengl bindings i write wind up being useful i'll likely put them on nimble |
| 02:49:19 | mcc | although i don't know if that would mean winding up with a nimble module that requires you to set up a build script which invokes an npm script... |
| 02:49:57 | mcc | ftsf_: OK, i'll try to remember to ping you when I have something together! |
| 02:50:08 | ftsf_ | cheers =) |
| 02:50:18 | ftsf_ | what does the npm do? |
| 02:53:06 | mcc | ftsf_: Here is the source I have so far, but it isn't in a very useful state (the binding script is intertwined with the webgl code) https://bitbucket.org/runhello/nim-webgl-test/src |
| 02:53:27 | libman | Hmm, found compileOption... |
| 02:54:06 | mcc | ftsf_: The way I created bindings was to generate them from typescript... this used the typescript compiler API, so it is itself written in typescript... so if i wanted to distribute the bindings tool, the most sensible way to do so would be through the npm package manager for javascript |
| 02:55:11 | * | ftsf_ looks up typescript |
| 02:55:52 | ftsf_ | ahh nice |
| 02:55:53 | mcc | typescript is very nice actually |
| 02:56:02 | mcc | it fixes most of the problems i have with javascript |
| 02:56:32 | ftsf_ | yeah, typelessness is hard |
| 02:57:08 | mcc | i do not mind typelessness a lot of the time. but javascript is such a sloppy language |
| 02:57:21 | mcc | i feel like i need types in js a lot more than i need them in python or lua |
| 02:58:17 | mcc | i would actually probably just write my webgl stuff in that |
| 02:58:17 | mcc | but |
| 02:58:21 | mcc | like i said, i want to run on desktop too |
| 02:58:27 | ftsf_ | mmm, the bigger the project gets the more you need types |
| 02:58:56 | mcc | but i could imagine a project which is a mix of nim and typescript working really well |
| 03:17:45 | * | User quit (Quit: Leaving) |
| 03:17:54 | * | ftsf_ quit (Ping timeout: 250 seconds) |
| 03:20:58 | * | yglukhov joined #nim |
| 03:25:24 | * | yglukhov quit (Ping timeout: 258 seconds) |
| 03:31:53 | * | bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 03:32:56 | * | bjz joined #nim |
| 03:46:29 | * | chemist69_ quit (Ping timeout: 258 seconds) |
| 03:50:52 | * | chemist69 joined #nim |
| 03:52:50 | libman | I always thought Nim was everything that TypeScript was intended to be and more... |
| 03:53:37 | libman | Didn't figure out how to access (into a const) compiler --out option (no big deal, just curious). |
| 03:57:56 | * | libman quit (Remote host closed the connection) |
| 05:22:42 | * | ftsf_ joined #nim |
| 05:22:51 | * | yglukhov joined #nim |
| 05:27:42 | * | yglukhov quit (Ping timeout: 265 seconds) |
| 05:51:56 | * | Demon_Fox quit (Ping timeout: 244 seconds) |
| 05:55:17 | * | chemist69 quit (Ping timeout: 258 seconds) |
| 05:57:36 | * | chemist69 joined #nim |
| 05:58:20 | * | GustavoLapasta joined #nim |
| 06:05:38 | * | bjz quit (Ping timeout: 258 seconds) |
| 06:10:47 | * | bjz joined #nim |
| 06:17:39 | * | fastrom joined #nim |
| 06:22:18 | * | xet7 joined #nim |
| 06:44:42 | * | yglukhov joined #nim |
| 06:48:12 | * | Senketsu_ joined #nim |
| 06:49:42 | * | yglukhov quit (Ping timeout: 276 seconds) |
| 06:50:05 | * | asdf__ joined #nim |
| 06:52:07 | * | euantor_ joined #nim |
| 06:53:15 | * | Sergio965_ joined #nim |
| 06:53:53 | * | Flox421 joined #nim |
| 06:54:37 | * | Flox42 quit (*.net *.split) |
| 06:54:37 | * | xet7_ quit (*.net *.split) |
| 06:54:37 | * | euantor quit (*.net *.split) |
| 06:54:38 | * | Senketsu quit (*.net *.split) |
| 06:54:38 | * | Gonzih quit (*.net *.split) |
| 06:54:39 | * | Sergio965 quit (*.net *.split) |
| 06:54:39 | * | reactormonk quit (*.net *.split) |
| 06:54:39 | * | asdf quit (*.net *.split) |
| 06:54:43 | * | euantor_ is now known as euantor |
| 06:56:57 | * | yglukhov joined #nim |
| 07:02:03 | * | yglukhov quit (Ping timeout: 276 seconds) |
| 07:08:33 | * | themagician quit (Ping timeout: 276 seconds) |
| 07:14:09 | * | themagician joined #nim |
| 07:17:08 | * | desophos quit (Read error: Connection reset by peer) |
| 07:22:13 | * | Demon_Fox joined #nim |
| 07:32:08 | * | reactormonk joined #nim |
| 07:32:22 | * | Gonzih joined #nim |
| 07:32:48 | * | yglukhov joined #nim |
| 07:35:23 | * | GustavoLapasta quit (Read error: Connection reset by peer) |
| 07:37:24 | * | mcc quit (Quit: Connection closed for inactivity) |
| 08:11:42 | * | themagician quit (Ping timeout: 258 seconds) |
| 08:16:06 | * | rbergmair joined #nim |
| 08:16:54 | * | themagician joined #nim |
| 08:18:42 | chemist69 | mcc: concerning your earlier question of automatic return type detection: there is the `auto` keyword: proc model(): auto = return [1,2,3,4,5,6] |
| 08:23:50 | * | gokr joined #nim |
| 08:31:04 | * | SianaGearz quit (Ping timeout: 244 seconds) |
| 08:31:46 | * | SianaGearz joined #nim |
| 08:31:46 | * | SianaGearz quit (Changing host) |
| 08:31:46 | * | SianaGearz joined #nim |
| 08:38:58 | * | beatmox quit (Ping timeout: 244 seconds) |
| 08:47:30 | * | beatmox joined #nim |
| 09:00:24 | * | xet7 quit (Quit: Leaving) |
| 09:04:51 | * | namnguyen quit (Ping timeout: 250 seconds) |
| 09:16:59 | * | namnguyen joined #nim |
| 09:19:18 | * | kingofoz quit (Ping timeout: 250 seconds) |
| 09:19:33 | * | kingofoz joined #nim |
| 09:23:42 | * | Demon_Fox quit (Quit: Leaving) |
| 09:38:03 | * | Matthias247 joined #nim |
| 09:40:09 | * | xet7b joined #nim |
| 09:43:08 | * | elrood joined #nim |
| 09:58:50 | dom96 | Calinou: Go for it. I think you're right, but it might just be a case of replacing the logo in the white background version with this one: https://raw.githubusercontent.com/nim-lang/assets/master/Art/logo-crown.png |
| 10:01:38 | * | nsf quit (Quit: WeeChat 1.5) |
| 10:09:45 | Calinou | ok :) |
| 10:13:37 | dom96 | Also, if you're going to create a new design it would be great if you did the same for the NimForum |
| 10:20:13 | dom96 | they don't have to be the same, in fact, it might actually be better if they aren't |
| 10:24:07 | Calinou | dom96: yeah, I started doing a Jekyll prototype (for experimenting with HTML/SCSS) |
| 10:29:07 | * | rbergmair quit (Quit: Leaving.) |
| 10:31:40 | * | asdf__ is now known as asdf |
| 10:40:27 | * | Matthias247 quit (Read error: Connection reset by peer) |
| 10:41:05 | * | Trustable joined #nim |
| 10:50:01 | * | fredrik92 joined #nim |
| 11:06:00 | * | arnetheduck quit (Remote host closed the connection) |
| 11:13:23 | * | Mat4 joined #nim |
| 11:14:36 | * | xet7 joined #nim |
| 11:20:23 | Mat4 | hello, can someone please explain to me what 'typelessness' may be and why it is hard to handle ? |
| 11:48:09 | * | arnetheduck joined #nim |
| 11:49:31 | * | stisa joined #nim |
| 11:54:02 | elrood | not really nim-related, and probably not better than you could explain it to yourself after a bit of googling and common sense. depending on context, most likeyl a language without a type system. think assembly, you never know if data at a memory address is a byte, word, some kind of object, list, etc. |
| 11:54:43 | elrood | you'll be able to figure out why that's hard to handle on your own, i bet |
| 11:56:50 | * | stisa quit (Ping timeout: 264 seconds) |
| 11:58:14 | elrood | in case that's some kind of homework question you're too lazy to do yourself, shame on you ;) |
| 12:01:00 | * | kunev quit (Ping timeout: 244 seconds) |
| 12:02:22 | * | kunev joined #nim |
| 12:06:34 | Calinou | dom96: here's my first mockup so far (not finished, only top is present): https://lut.im/a19CprN6L9/1wgCDQ00xUhVX5ej.png |
| 12:07:10 | dom96 | nice, so you are going for the dark design again :) |
| 12:07:20 | Calinou | we can have parts of the site as dark, other parts as light |
| 12:07:32 | Calinou | kind of like https://www.gnu.org/software/guix/ |
| 12:07:53 | dom96 | sure, I love it :D |
| 12:08:45 | dom96 | Calinou: Want me to create a new repo in nim-lang/website on github and add you as a contributor? |
| 12:09:09 | Calinou | you mean, create https://github.com/nim-lang/website ? |
| 12:09:14 | dom96 | yeah |
| 12:09:36 | * | stisa joined #nim |
| 12:09:50 | Calinou | right… note that I've barely ever programmed in Nim (I have basic C++ experience, and intermediate Python/PHP knowledge), so to make something based on existing code will require more work |
| 12:10:17 | dom96 | no worries, I can help with that. |
| 12:10:26 | dom96 | Feel free to commit HTML/CSS/images |
| 12:10:33 | Calinou | yeah, it's a Jekyll site right now |
| 12:10:35 | dom96 | or even your full jekyll project |
| 12:10:40 | Calinou | yeah I can upload t hat |
| 12:10:41 | Calinou | that* |
| 12:10:44 | dom96 | we can convert it to the infamous `nimweb` later |
| 12:12:09 | dom96 | Calinou: what's your github nick? same as your IRC? |
| 12:12:14 | Calinou | yes |
| 12:12:52 | * | yglukhov quit (Remote host closed the connection) |
| 12:13:54 | dom96 | Invited you to join |
| 12:14:57 | Calinou | dom96: ok, should I take take of initial commit? |
| 12:15:16 | dom96 | Calinou: sure |
| 12:15:23 | Calinou | I can push the Jekyll prototype to the gh-pages branch, so that people can visit http://nim.github.io/website/ perhaps |
| 12:15:31 | Calinou | (with "noindex, nofollow" in HTML so that robots don't index it) |
| 12:15:41 | dom96 | I really like what you wrote in your current design as well. |
| 12:15:45 | Calinou | so that people can get a preview, but it doesn't impact SEO until it's ready |
| 12:15:48 | dom96 | Sells the language pretty well :) |
| 12:16:22 | dom96 | hrm, I wouldn't do that. Somebody might submit the link to Reddit/HN with "Look at the new Nim website" |
| 12:16:26 | dom96 | before it's actually ready |
| 12:16:30 | Calinou | ah |
| 12:16:47 | Calinou | ok, I'll just push it in a jekyll_prototype folder in master branch |
| 12:16:54 | Calinou | I'll work on it a bit more first |
| 12:16:57 | dom96 | sure |
| 12:17:13 | dom96 | cloning it shouldn't be too difficult if someone wants a preview |
| 12:20:08 | * | bjz_ joined #nim |
| 12:21:13 | dom96 | Calinou: I'm not sure what your plans are but I think it would be cool to have those sorts of guix panels of differing colours, one such panel might include information about our bountysource campaign for example (which I would love to have on the front page, even if it's "below the fold") |
| 12:21:34 | * | bjz quit (Ping timeout: 250 seconds) |
| 12:23:04 | Calinou | dom96: I thought about putting a quick overview of the language below the fold, Bountysource campaign could be mentioned somewhere at the top (need to see how) |
| 12:23:27 | dom96 | okay |
| 12:23:27 | Calinou | need to find a place to put news posts (excerpts) too, ideally they should be visible without scrolling |
| 12:23:33 | Calinou | like on https://godotengine.org/ |
| 12:23:50 | * | nsf joined #nim |
| 12:24:12 | dom96 | sounds good. |
| 12:24:37 | dom96 | Although the news posts on godot's site are barely above the fold on my MBP :) |
| 12:24:52 | Calinou | also, what kind of browser support should we target? is IE11+ fine? note that I don't have any IE9 or IE10 machines to test, only IE11 and Edge |
| 12:25:17 | * | dom96 checks analytics |
| 12:25:48 | Calinou | I have no old PCs, thus no IE10 :P |
| 12:27:21 | dom96 | According to analytics ~1.5% of our users are on IE, of which 78% are using IE11 |
| 12:27:26 | dom96 | so IE11+ sounds fine |
| 12:27:51 | elrood | once you're at it anyways, you could make the site theme user-selectable, so that we don't run into light/dark-controversies |
| 12:28:16 | Calinou | elrood: yeah, like http://urho3d.github.io does |
| 12:28:28 | Calinou | we can maintain a base stylesheet, plus a dark and light one, created using Sass |
| 12:30:36 | elrood | great. personally i'd vote for a non-js reliant site, but it's your choice |
| 12:31:00 | * | HogynCymraeg joined #nim |
| 12:31:14 | dom96 | that sounds like a good idea, but it should get a lower priority than the main site. |
| 12:32:20 | dom96 | so that we don't end up in a situation where we only have 50% of the website designed in a dark&light theme and no time to do the rest. |
| 12:32:25 | Calinou | elrood: theme chooser requires JavaScript IIRC |
| 12:32:44 | dom96 | The site should at least be usable without JS |
| 12:32:50 | Calinou | yeah, that should be possible |
| 12:33:01 | Calinou | I'm using the Pure CSS framework, which is pretty lightweight, and works without any JavaScript or jQuery |
| 12:33:05 | HogynCymraeg | Hi everyone. I'm learning nim and am trying to figure out the best way to run a webserver in a different process so that it minimally impacts the main process. I will need some kind of IPC mechanism between them. Thoughts? |
| 12:33:14 | dom96 | And if you want to be super leet, ensure it works in Lynx/elinks ;) |
| 12:33:48 | dom96 | HogynCymraeg: maybe this could help? https://github.com/dom96/nim-jsonrpc |
| 12:35:02 | HogynCymraeg | dom96: thanks. I'll check it out now. BTW: Bought NIA last week. Excellent job! |
| 12:35:14 | dom96 | HogynCymraeg: Yay, thanks! :) |
| 12:35:30 | dom96 | I should really get off IRC and get back to work on it. |
| 12:35:38 | elrood | Calinou, check out link rel="alternate stylesheet" type="text/css" if you like. and have fun designing |
| 12:35:40 | dom96 | bbl :) |
| 12:36:16 | Calinou | elrood: thanks :) I'll check that |
| 12:38:26 | HogynCymraeg | dom96: the rpc server isn't quite what I'm thinking. I'm basically trying to run a webserver in parallel to an SDL program to inspect it. Using asyncdispatch.poll seems too slow. In node, i'd use child_process and send/on for ipc. |
| 12:40:37 | euantor | Calinou: that design draft looks great, nice work! |
| 12:40:45 | Calinou | thanks :D |
| 12:43:38 | Calinou | dom96: also, how can we integrate Sass (SCSS) into a Nim website? do we call a task runner like Gulp separately? |
| 12:44:32 | euantor | I think that's the best way, yeah. |
| 12:51:19 | * | xet7 quit (Quit: Leaving) |
| 12:51:52 | * | xet7 joined #nim |
| 12:54:20 | * | brechtm joined #nim |
| 12:57:10 | cheatfate | HogynCymraeg, and what is slow in asyncdispatch.poll() ? |
| 12:57:23 | * | Mat4 quit (Quit: Leaving.) |
| 12:59:11 | * | brechtm quit (Remote host closed the connection) |
| 13:02:25 | HogynCymraeg | cheatfate: turns out the default 500ms parameter ;-) Tried with a low value and it's much faster |
| 13:02:56 | cheatfate | HogynCymraeg, if you got 500ms then you doing something wrong... |
| 13:03:41 | HogynCymraeg | I thought that was the default value for poll? |
| 13:04:05 | cheatfate | HogynCymraeg, yep, but if you fall into 500ms timeouts then your approach is looks wrong |
| 13:04:34 | HogynCymraeg | hmmm interesting. |
| 13:05:11 | cheatfate | HogynCymraeg, how you transfer your data between processes? |
| 13:05:15 | cheatfate | what mechanism you are using? |
| 13:05:30 | HogynCymraeg | I had an infinite loop with a poll and incrementing a variable. That variable went up very slowly. Set poll to a low timeout value and it went really fast |
| 13:05:54 | HogynCymraeg | In this code, it's all in main process. I was enquiring how best to move it out of main |
| 13:06:34 | cheatfate | HogynCymraeg, if you decrease timeout you will get more CPU extensive loop... and this is not the way you are looking for |
| 13:07:42 | cheatfate | try to use pipes/unix sockets/sockets/files to transfer data |
| 13:07:46 | cheatfate | it will be faster |
| 13:08:25 | HogynCymraeg | ok thanks :-) |
| 13:08:50 | Calinou | https://lut.im/SP57HSNX06/dd3Z4uUrj8idhUaG.png - some progress |
| 13:09:04 | Calinou | not entirely sure on how to present news |
| 13:09:14 | Calinou | (I should copy-paste existing news so that it feels "more" real for testing, I know) |
| 13:10:23 | stisa | HogynCymr: maybe you could use osproc http://nim-lang.org/docs/osproc.html ? |
| 13:11:06 | * | SianaGearz quit (Ping timeout: 244 seconds) |
| 13:11:16 | cheatfate | Calinou, nice, i like it |
| 13:12:32 | * | mcc joined #nim |
| 13:12:44 | HogynCymraeg | stisa: Cool. That looks close to what I want. How do you send messages between procs with that? |
| 13:13:20 | * | yglukhov joined #nim |
| 13:13:46 | HogynCymraeg | stisa: ah, looks like stdin/stdout streaming. Cool |
| 13:14:31 | stisa | Haven't used it, but I guess inputStream can use write procs from http://nim-lang.org/docs/streams.html , and you can check for pending data with hasData |
| 13:16:33 | * | kingofoz quit (Read error: Connection reset by peer) |
| 13:17:38 | * | yglukhov quit (Ping timeout: 250 seconds) |
| 13:18:12 | * | kingofoz joined #nim |
| 13:19:09 | * | SianaGearz joined #nim |
| 13:21:22 | * | HogynCymraeg quit (Ping timeout: 250 seconds) |
| 13:22:24 | * | Goba_ left #nim ("Leaving") |
| 13:26:43 | mcc | Is there some way for a nim function or a function signature to take "something iterable"? |
| 13:26:57 | mcc | For example, a function that could take either an array[5, int] or a seq[int]? |
| 13:27:09 | euantor | openarray[int] |
| 13:27:21 | mcc | Cool |
| 13:27:36 | euantor | http://nim-lang.org/docs/manual.html#types-open-arrays |
| 13:31:59 | chemist69 | mcc: concerning your earlier question of automatic return type detection: there is the `auto` keyword: `proc model(): auto = return [1,2,3,4,5,6]` |
| 13:32:17 | mcc | oh, i see |
| 13:38:07 | dom96 | Calinou: Starting to look quite similar to https://haskell-lang.org/ |
| 13:38:21 | dom96 | not that that's a bad thing |
| 13:39:33 | * | xet7b quit (Quit: AndroIRC - Android IRC Client ( http://www.androirc.com )) |
| 13:46:49 | * | HogynCymraeg joined #nim |
| 13:46:49 | * | HogynCymraeg quit (Client Quit) |
| 13:59:47 | * | GustavoLapasta joined #nim |
| 14:12:21 | * | bjz joined #nim |
| 14:13:37 | * | bjz_ quit (Ping timeout: 244 seconds) |
| 14:18:52 | mcc | Does Nim have anything like Java "interfaces"? Or some sort of structural type that corresponds to "any type that supports these operations"? |
| 14:22:46 | mcc | I'm trying to write FFI declarations and at some point I need to write an {.importc.} class that takes as one argument, for example, "any type that has the method index()". |
| 14:23:27 | dom96 | look up concepts |
| 14:23:43 | mcc | thanks |
| 14:24:03 | * | rbergmair joined #nim |
| 14:25:52 | Araq | hey, I'm back |
| 14:25:55 | Araq | what did I miss? |
| 14:26:26 | dom96 | Araq: Calinou is working on a redesign: https://lut.im/SP57HSNX06/dd3Z4uUrj8idhUaG.png |
| 14:26:42 | * | yglukhov joined #nim |
| 14:29:33 | Araq | yummy, I like it |
| 14:30:14 | ftsf | \o/ |
| 14:30:50 | mcc | So when I look at a generated .js file from Nim , I notice it declares Int8Array and a bunch of stuff even if the Nim file is empty... does Nim ever use Int8Array etc internally? |
| 14:31:14 | Araq | mcc: yes the codegen sometimes emits it |
| 14:31:16 | mcc | Because I am now writing WebGL with Nim, and I had to go to a bunch of awkward effort to create Int8Arrays, Float32Arrays etc |
| 14:31:24 | Araq | ask yglukhov about the details |
| 14:31:29 | * | chemist69 quit (Ping timeout: 260 seconds) |
| 14:32:15 | mcc | Araq: OK. is there a way to force it to be emitted? Like, if I would like to have something that is typed as an openArray[float] but is being codegened as a Float32Array. |
| 14:32:31 | mcc | So that I could just pass this stuff around instead of using base Nim types and then converting them to TypedArrays |
| 14:32:31 | mcc | OK |
| 14:33:01 | dom96 | mcc: don't think so |
| 14:33:28 | * | chemist69 joined #nim |
| 14:35:06 | mcc | On a related note, when I generate js from nim, the code seems to be dominated by functions like rawwritestacktrace() and something that looks like stashing and clearing stack information at the start and end of each function, like : e_13410 = {m_type: NTI3440, parent: null, name: null, message: null, trace: null |
| 14:35:20 | mcc | What is that for? Is that just to recapture Nim's exception semantics.../ |
| 14:38:04 | Araq | it's just debugging convenience. |
| 14:38:07 | Araq | or used to be. |
| 14:38:21 | Araq | Nim produced its own stack traces that map to Nim source code. |
| 14:39:01 | Araq | that stuff predates source maps and source maps are kind of hard to retrofit into the code generator |
| 14:39:09 | Araq | it's all gone with -d:release |
| 14:39:18 | Araq | or stackTrace:off |
| 14:40:16 | mcc | ok cool |
| 14:46:50 | yglukhov | mcc: typed arrays are emitted for nim's arrays. |
| 14:47:28 | mcc | Oh! Interesting. Just fixed sized arrays, or also openArrays and seqs? |
| 14:47:32 | yglukhov | e.g. var myVertexData: array[120, GLfloat] |
| 14:47:48 | yglukhov | only fixed size for now. |
| 14:47:51 | * | Mat4 joined #nim |
| 14:50:31 | mcc | yglukhov: So like I said, I've been trying to write WebGL in Nim-- I made this last night: https://data.runhello.com/j/dry/2/ I'm doing this in a slightly odd way, I didn't find the WebGL stuff in the DOM module so I made a script that generates nim bindings from TypeScript and I'm basically writing against the typings in the Nim standard library |
| 14:50:43 | mcc | i'm sorry i mean the TypeScript standard library |
| 14:51:18 | mcc | Are there existing projects that do WebGL with Nim, am I reinventing any wheels? |
| 14:53:58 | * | Mat4 left #nim (#nim) |
| 14:54:48 | stisa | Well I have one, but it's really rough: https://github.com/stisa/webgl |
| 14:56:09 | yglukhov | mcc, you might want to have a look at nimx and rod. |
| 14:56:09 | mcc | stisa: neat :O i *think* the script generated bindings can do everything that can... except the matrix stuff. hm. |
| 14:56:35 | yglukhov | https://github.com/yglukhov/nimx, https://github.com/yglukhov/rod |
| 14:57:08 | stisa | mcc: btw I ported over your example to see if it would work, hope you don't mind: http://stisa.space/webgl/ex3.html |
| 14:57:30 | Araq | let a and b be positive integers such that a*b+1 divides a² + b² |
| 14:57:55 | mcc | stisa: heh, cool |
| 14:57:56 | Araq | show that (a² + b²) / (a*b + 1) is the square of an integer. |
| 14:58:37 | mcc | nimx is the gui framework, right? i guess it had to solve this problem internally? |
| 14:59:47 | mcc | Here's a question... is it possible to create a Nim fixed-length array of not-known-at-compile-time size? like the equivalent of char *x = new char[someVariable]. seq is different because that's an actual dynamically growable array |
| 15:03:46 | Araq | no, you have to use 'seq' for that. |
| 15:04:03 | Araq | which is a single indirection just like your new char example is, so nothing to worry about. |
| 15:12:21 | mcc | Sure... so what I'm thinking about is, I'm imagining writing a patch to nim where openArray *also* uses TypedArray/ArrayBuffer when generating for javascript. Like yglukhov described happens for array[20, float]. |
| 15:12:47 | mcc | However, that doesn't really work openArray also encompasses seq, and a seq cannot be represented as a TypedArray because js typed arrays are of fixed size and seq is growable. |
| 15:16:48 | * | Ven joined #nim |
| 15:20:23 | yglukhov | mcc: openArray is not a type. you can't "instanciate" it. and openarray is perfectly compatible with js typed arrays as it is now. instead we could extend typed array usage to seqs. |
| 15:24:07 | mcc | yglukhov: OK... how would you represent a seq as a js typed array? Since js typed arrays are fixed size and seq has add(). Sorry, I'm only just recently learning about WebGL :) I'm trying to figure out if you could implement something using DataViews. |
| 15:40:49 | yglukhov | mcc: dunno. maybe seq should have a typed array+capacity and realloc+copy typed array when len reaches capacity. basically just like seq behave in c. |
| 15:42:10 | mcc | yeah, that would work well. you'd have to figure out some way to create an object that correctly reports its .length property however when you pass it in to a pure js function. But maybe DataView gives you that. |
| 15:47:14 | * | xet7 quit (Quit: Leaving) |
| 15:51:11 | cheatfate | Araq, welcome back :) |
| 15:53:54 | * | errerr joined #nim |
| 15:54:04 | Araq | thanks. any troll attacks that I missed? :P |
| 15:55:43 | * | xet7 joined #nim |
| 15:56:37 | * | xet7 quit (Remote host closed the connection) |
| 15:57:41 | * | errerr left #nim (#nim) |
| 15:58:02 | elrood | occasionally trying to keep us entertained, never with much lasting effect though |
| 15:58:05 | elrood | welcome back |
| 16:00:50 | * | minus left #nim ("WeeChat 1.5") |
| 16:01:53 | Araq | "Closure iterators are not supported by js backend." yglukhov: that's your favourite issue now isn't it? |
| 16:03:36 | yglukhov | Araq: wow, you let me choose the next fixed issue? ;) |
| 16:04:20 | yglukhov | Araq: no, thats not my favourite. |
| 16:09:37 | mcc | araq: by the way, i was meaning to ask-- what does $ in an importc do? as described in https://github.com/nim-lang/Nim/issues/4579 it seems to have SOME special behavior |
| 16:16:51 | * | Salewski joined #nim |
| 16:19:20 | Salewski | Araq, I have just copied my IRC remarks from 12-AUG about nimsuggest problems with system.nim module to nimsuggest issue tracker. |
| 16:20:28 | Salewski | My GTK editor supports nimsuggest very good now, but the system.nim bug is a restriction. Do you have an idea, may it be very hard to fix? |
| 16:22:08 | Salewski | I have looked for about two hours on the nimsuggest code, but have no idea. |
| 16:23:37 | Salewski | There are some more, less serious errors in nimsuggest, but I think reporting or fixing them makes no sense as long as the system modules bug persists. |
| 16:26:16 | * | nsf quit (Quit: WeeChat 1.5) |
| 16:32:12 | * | Vladar joined #nim |
| 16:33:35 | * | Matthias247 joined #nim |
| 16:33:52 | Araq | Salewski: what does nimsuggest report when you use --debug? |
| 16:34:03 | Araq | and no, I don't think it's hard :-) |
| 16:34:53 | Araq | mcc: proc foo() {.importc: "prefix_$1".} produces "prefix_foo" |
| 16:35:11 | Araq | so that you can importc within a .push environment |
| 16:35:20 | Araq | pretty obvious, isn't it? |
| 16:35:22 | Salewski | Araq, for the reported issues at issue tracker I have used --debug option. |
| 16:35:38 | Araq | sorry I'm not there yet |
| 16:35:50 | Araq | still 50 emails to read, I'm sure it's there somewhere |
| 16:36:27 | * | kulelu88 joined #nim |
| 16:36:59 | Salewski | When it it not to hard, it may be a good idea if you inspect the reported issues. Because that is a serious restriction for all nimsuggest users. |
| 16:38:55 | mcc | araq: ok, i see. does $2, for example, do something? |
| 16:39:02 | mcc | is that a param maybe? |
| 16:39:18 | Araq | no, it crashes the compiler |
| 16:39:32 | Araq | or rather produces an unhelpful error message |
| 16:40:15 | mcc | araq: ok thanks |
| 16:40:45 | Araq | note that .codegenDecl uses it differently |
| 16:41:57 | Salewski | Bye... |
| 16:42:00 | * | Salewski left #nim (#nim) |
| 16:50:26 | * | stisa quit (Quit: quitting...) |
| 16:52:34 | * | fredrik92 quit (Ping timeout: 250 seconds) |
| 16:56:59 | * | GustavoLapasta quit (Quit: Leaving) |
| 16:58:52 | * | Sentreen quit (Ping timeout: 240 seconds) |
| 17:08:46 | * | Trustable quit (Quit: Leaving) |
| 17:12:40 | * | Sentreen joined #nim |
| 17:14:26 | * | Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 17:24:38 | * | chemist69 quit (Ping timeout: 250 seconds) |
| 17:26:08 | Calinou | ok, here's how it looks so far: https://lut.im/8ono9e3LjW/EPEaml4Cbaclsasg.png - and scrolling down: https://lut.im/zKkVeBDopE/e9VyyEpYTfyR2sy1.png |
| 17:26:17 | Calinou | there is a dedicated news archive page too: https://lut.im/qgu7Z07TxN/QuIzU6S15qtyplh8.png |
| 17:26:36 | Calinou | footer is not finished, should probably have some more content, or social media icons, or something like that |
| 17:29:53 | dom96 | nice |
| 17:29:56 | dom96 | I love it |
| 17:30:07 | * | nairobi joined #nim |
| 17:30:16 | Calinou | dom96: I can commit the prototype to the repository, doing some cleanup before :) |
| 17:30:36 | dom96 | Maybe a "See more..." link under "Documentation" and "Community" in the footer leading to a dedicated page for both? |
| 17:31:16 | Calinou | makes sense |
| 17:31:18 | dom96 | Calinou: feel free to commit even unfinished things |
| 17:31:26 | dom96 | It's nice to see a history of how it developed :) |
| 17:31:45 | Calinou | yeah, I should do that… sadly I don't have the habit of using Git locally :P |
| 17:32:49 | dom96 | Cool. So judging by the archive page, the main content will always be shown on a white background, right? |
| 17:36:42 | dom96 | Some more ideas: Should we have a "Learn More" button beside the Download and Documentation buttons leading to a more in-depth look at Nim's features? |
| 17:36:55 | dom96 | or should we extend the front page to include more information about Nim? |
| 17:37:17 | elrood | easily accessible links to at least the repo, wiki and available libraries / nimble packages would be nice |
| 17:38:34 | * | fredrik92 joined #nim |
| 17:38:38 | * | fredrik92 quit (Read error: Connection reset by peer) |
| 17:39:17 | dom96 | yeah, but I'm not sure where they could go |
| 17:40:22 | dom96 | we need to think carefully about what pages should be linked in the nav bar. |
| 17:40:27 | dom96 | and what those pages should contain |
| 17:42:26 | * | chemist69 joined #nim |
| 17:42:29 | * | nairobi quit (Remote host closed the connection) |
| 17:42:33 | dom96 | Calinou: Have you tested what your design looks like on mobile yet? |
| 17:42:49 | Calinou | sure, I make regular tests using Chromium's responsive mode |
| 17:43:01 | Calinou | I have a real smartphone too, I'll make sure to test it a bit on it too |
| 17:43:11 | dom96 | awesome |
| 17:43:17 | elrood | the ones mentioned could preferrably be positioned as prominently as possible, especially the wiki, imho. maybe add a general resources and about link, and you can place pretty much everything else below |
| 17:43:59 | elrood | the front page itself could be kept fairly simple. in its current form it's quite overburdened with nim's feature list |
| 17:46:25 | elrood | Calinou, firefox's developer tools come with a very useful preview for different screen sizes. chrome most likely has something equivalent |
| 17:48:48 | Calinou | yes, that's what I use |
| 17:49:04 | elrood | as long as you make sure navigation works equally well on touch- and point-and-click-oriented devices, you should be fine |
| 17:49:06 | Calinou | Chromium (and Chrome) have very extensive DevTools, sometimes even overkill (you can listen and replay for transitions!) |
| 17:50:45 | * | nairobi joined #nim |
| 17:53:54 | Calinou | heh, the site's news pages are quite friendly to Firefox's reader mode |
| 17:54:47 | dom96 | Good :) |
| 17:55:33 | dom96 | Sooo, anybody got experience analysing surveys? |
| 17:55:42 | * | dom96 is about to start looking at our Nim community survey |
| 17:55:52 | dom96 | 790 responses :D |
| 17:57:09 | euantor | You going to make some nice graphs and stuff @dom96? |
| 17:57:25 | dom96 | yep |
| 17:57:38 | euantor | Well, I don't envy the task ;) |
| 17:59:03 | Calinou | dom96: https://github.com/nim-lang/website - committed |
| 17:59:50 | dom96 | Calinou: brilliant, thanks. |
| 18:01:37 | dom96 | euantor: yeah... just downloaded the huge CSV file :\ |
| 18:02:02 | * | nairobi quit (Ping timeout: 258 seconds) |
| 18:02:51 | * | nairobi joined #nim |
| 18:03:47 | dom96 | Haven't used Excel this extensively since high school. |
| 18:07:04 | * | Sergio965_ quit (Max SendQ exceeded) |
| 18:07:20 | * | Sergio965 joined #nim |
| 18:08:14 | euantor | I'd need to first install Excel |
| 18:08:20 | euantor | Or dig out the work laptop |
| 18:08:42 | euantor | How big is the file with that many answers? |
| 18:08:46 | Calinou | heh, here we're taught Excel at uni, I got 20/20 as grade :P |
| 18:08:51 | Calinou | in 1st semester |
| 18:08:57 | euantor | Seems like a pretty good turnout |
| 18:09:01 | euantor | Oh, weird |
| 18:09:04 | * | Sergio965 quit (Max SendQ exceeded) |
| 18:09:06 | dom96 | Looks like Google's Spreadsheets are doing a better job |
| 18:09:11 | euantor | We never did anything like that at Uni |
| 18:09:17 | * | Sergio965 joined #nim |
| 18:09:27 | euantor | You were just expected to know how to use things like Excel and Word |
| 18:09:35 | euantor | And if you didn't, then tough |
| 18:09:38 | dom96 | Calinou: interesting, if you don't mind me asking, what Uni do you attend? |
| 18:10:04 | Calinou | I'm French, I attend "Métiers du Multimédia et de l'Internet" (= "Jobs of multimedia and the Internet", if I translate) at Troyes |
| 18:10:21 | Calinou | we do web development, web design, graphics (Photoshop/Illusator), video editing |
| 18:10:23 | Calinou | a lot of things :) |
| 18:12:11 | euantor | Cool. I always find it interesting how these things differ between countries |
| 18:12:23 | euantor | They're all similar, but there are always little differences |
| 18:12:37 | Calinou | for example we don't do system programming, only PHP/JavaScript |
| 18:13:24 | euantor | Our first year was dedicated to Java and Common Lisp, systems design and networking |
| 18:13:48 | Calinou | we do some networking too, really easy though… I typically get 17/20 or 18/20 grades :) |
| 18:14:03 | Calinou | I tend to be more interested in system programming than other students here |
| 18:16:05 | euantor | A lot of the other students on my course hadn't done any programming before University, so the first 3 months were the very basics |
| 18:16:23 | euantor | Introducing Java, what variables, functions, classes etc. are |
| 18:16:32 | Calinou | our PHP/JS teaching is pretty basic too, in 2nd semester we were taught about jQuery, but they don't really teach us how to write it, more how to read it… |
| 18:16:43 | Calinou | and what we had to do for evaluation in 2nd semester of PHP was: Fibonacci sequence |
| 18:16:50 | Calinou | (it was part of it) |
| 18:17:07 | euantor | If that was the whole evaluation, that's a very easy pass :P |
| 18:17:19 | Calinou | there were a few other things too, still easy |
| 18:17:55 | * | stisa joined #nim |
| 18:18:27 | Calinou | our 2nd semester networking class is about setting up an Apache web server |
| 18:18:31 | Calinou | sudo apt-get install apache2, done! :P |
| 18:18:55 | euantor | Some versions of Ubuntu Server have it pre-installed, so that's even easier! |
| 18:19:30 | Calinou | I'm possibly the only guy at my uni using Linux most of the time :P |
| 18:19:42 | Calinou | some use macOS, but usually don't know the command line, the vast majority just uses Windows |
| 18:19:51 | euantor | There were about 4 of us using it when I was at Uni |
| 18:20:02 | euantor | Most just using Ubuntu, but I usually run Fedora |
| 18:20:23 | Calinou | I'm on Antergos which is an Arch Linux derivative, basically it's already setup, and you can choose your desktop environment at installation |
| 18:20:33 | Calinou | Arch is nice (recent packages) but I like saving an hour or two of setup |
| 18:21:13 | Calinou | Fedora is pretty cool too, though it has some issues with font rendering out of the box (due to patents) |
| 18:21:44 | * | desophos joined #nim |
| 18:22:20 | euantor | Yeah, so long as it can render a monospace font, I'm usually fine |
| 18:22:31 | euantor | I use a Mac now as I was gifted one |
| 18:22:48 | Calinou | heh, I know someone at my uni who *won* a 15" MBP |
| 18:22:50 | Calinou | so lucky |
| 18:22:52 | euantor | I can still get done most of what I need to |
| 18:23:01 | euantor | That's pretty lucky |
| 18:23:13 | Calinou | here a 15" MBP starts at €2,249… |
| 18:23:13 | euantor | Anyway, I'm off to eat some dinner |
| 18:23:21 | euantor | Yeah, it's not that bad here |
| 18:23:24 | Calinou | my next laptop will probably be a ThinkPad/Dell XPS or something like that (I don't play games on it anyway) |
| 18:23:38 | euantor | Mine is a top of the line 15", was just under £2000 |
| 18:23:39 | Calinou | something with decent battery life is a must in 2016 |
| 18:23:50 | euantor | My work laptop is a ThinkPad. It's excellen |
| 18:23:54 | euantor | *excellent |
| 18:24:07 | euantor | It's got two batteries, one which is built in and the other can be replaced |
| 18:24:18 | Calinou | yeah, I saw a few ThinkPads which had that, pretty handy |
| 18:24:25 | Calinou | apparently one of them could achieve like, 11 hours of battery life in basic use |
| 18:24:25 | euantor | Meaning I can remove the main battery without it turning off and replace it with a pre-charged one |
| 18:24:37 | Calinou | going to eat dinner too, bye :p |
| 18:24:38 | euantor | It lasts about 12 hours on one battery for me |
| 18:24:53 | euantor | So long as I don't have the brightness on full and I'm not doing anything too strenuous |
| 18:25:18 | Calinou | maybe I'll wait for Kaby Lake CPUs, they should bring quite better performance according to recent leaks :P |
| 18:25:30 | Calinou | need more budget anyway first |
| 18:25:34 | * | nairobi quit (Remote host closed the connection) |
| 18:25:43 | * | nsf joined #nim |
| 18:26:40 | * | Matthias247 quit (Read error: Connection reset by peer) |
| 18:29:14 | * | nairobi joined #nim |
| 18:29:50 | * | nairobi quit (Remote host closed the connection) |
| 18:30:15 | * | nairobi joined #nim |
| 18:30:36 | * | Mat4 joined #nim |
| 18:34:57 | Mat4 | Araq: Ping |
| 18:35:10 | * | arnetheduck quit (Ping timeout: 252 seconds) |
| 18:35:15 | Araq | Mat4: pong |
| 18:36:11 | Mat4 | good afternoon, I'm able to start working on these Rim project again |
| 18:36:49 | * | xet7 joined #nim |
| 18:37:07 | * | yglukhov quit (Remote host closed the connection) |
| 18:37:49 | Mat4 | and want to ask about a recent syntax notation (BNF or something similar would be nice for example) |
| 18:38:52 | Araq | the manual uses a notation similar to BNF |
| 18:39:09 | Araq | it's documented and the grammar is available |
| 18:42:12 | Mat4 | I'm sorry, the documentation seem to have changed and I'm lost at current |
| 18:43:49 | Mat4 | ok, the notation is given for each chapter |
| 18:44:18 | Araq | http://nim-lang.org/docs/manual.html#lexical-analysis |
| 18:45:00 | Araq | http://nim-lang.org/docs/manual.html#syntax-grammar |
| 18:45:05 | Mat4 | thNKA |
| 18:45:09 | Mat4 | thanks |
| 18:47:11 | Araq | though it's way easier to use Nim's existing parser and call it a day :-) |
| 18:48:04 | * | yglukhov joined #nim |
| 18:56:11 | * | libman joined #nim |
| 18:59:02 | Mat4 | will future Nim versions support the 'do' notation ? |
| 19:00:44 | libman | Oh god, is Nim turning into Haskell now? |
| 19:01:38 | stisa | Mat4: Do you mean this http://nim-lang.org/docs/manual.html#procedures-do-notation ? |
| 19:01:45 | Mat4 | yes |
| 19:21:12 | * | elrood quit (Ping timeout: 244 seconds) |
| 19:23:52 | Araq | Mat4: the do notation is here to stay |
| 19:24:40 | Mat4 | that is good |
| 19:32:21 | Mat4 | ciao |
| 19:32:24 | * | Mat4 left #nim (#nim) |
| 19:36:50 | libman | Araq: can a program get a compile-time const of the --out argument passed to the Nim compiler? |
| 19:37:17 | * | irrequietus joined #nim |
| 19:37:19 | Araq | hmm I don't think so. why would that be useful? |
| 19:38:07 | libman | For The Greater Glory Of Nim! :D |
| 19:41:18 | libman | It is common in the Unix world to have symlinks of a single executable, which behaves differently depending on paramStr(0). |
| 19:42:38 | mcc | libman: the --out argument couldn't get you that, though, could it? |
| 19:42:41 | libman | And of course you can just have multiple different programs. But I was just curious if I can have a program compile differently depending on the target executable name. |
| 19:43:34 | mcc | libman: could you set your build script to also pass in the --out as a --define option? |
| 19:47:25 | libman | I'm just surprised all compiler options aren't available to the program as constants. Seems like a very simple feature. |
| 19:55:06 | * | nairobi quit (Remote host closed the connection) |
| 19:57:04 | * | libman hacks something together using staticExec |
| 20:02:34 | * | brechtm joined #nim |
| 20:03:20 | * | cyraxjoe quit (Remote host closed the connection) |
| 20:04:13 | * | nairobi joined #nim |
| 20:04:13 | * | brechtm quit (Read error: Connection reset by peer) |
| 20:04:31 | * | brechtm joined #nim |
| 20:04:44 | * | cyraxjoe joined #nim |
| 20:08:49 | * | nairobi quit (Remote host closed the connection) |
| 20:14:53 | * | Mat4 joined #nim |
| 20:16:03 | Mat4 | lines like these compile without problems: |
| 20:16:22 | Mat4 | let test = |
| 20:16:23 | Mat4 | proc (s: string): string = s & " ." |
| 20:17:55 | * | GangstaCat joined #nim |
| 20:19:13 | * | cyraxjoe quit (Remote host closed the connection) |
| 20:21:31 | Mat4 | alternatively I can exchange 'let' with 'var' |
| 20:22:12 | Mat4 | and as it seems both statements are equivalent. Is this a bug or a feature ? |
| 20:24:46 | euantor | Feature |
| 20:25:10 | euantor | Let and var are similar, but let us immutable |
| 20:29:51 | Mat4 | does it make any sense to you defining a variable function of static (which should be by definition immutable) behaviour ? |
| 20:30:55 | euantor | Doesn't make sense, but that doesn't mean that the compiler is wrong. You could reassign it to a different function later on. |
| 20:35:38 | Mat4 | sorry I does not see in which situation that should be more useful than just defining a second function and pass a pointer at demand. |
| 20:35:59 | euantor | So don't use it then? |
| 20:36:10 | * | gokr quit (Ping timeout: 250 seconds) |
| 20:36:44 | * | cyraxjoe joined #nim |
| 20:39:41 | Mat4 | I think it should be better to handle the second variant as error |
| 20:43:07 | * | xet7 quit (Quit: Leaving) |
| 20:45:09 | Mat4 | ciao |
| 20:45:12 | * | Mat4 left #nim (#nim) |
| 20:48:46 | * | Mat4 joined #nim |
| 20:50:31 | Mat4 | here is the reason: https://gist.github.com/anonymous/1983fe58aadac20cc6a68ad089060ba8 |
| 20:50:54 | Mat4 | this compiles but lead to a stack overflow at runtime |
| 20:51:40 | euantor | Surely any recursive function with no terminal state would do? |
| 20:51:46 | * | gokr joined #nim |
| 20:53:59 | * | libman quit (Quit: Leaving.) |
| 20:54:58 | * | Ven joined #nim |
| 20:55:18 | Mat4 | that's because test (s) is compiled as recursive call whereby it also interpreted as invocation of the previous declaration. I can envision that such ambiguity can occur frequently, in particular in larger sources |
| 20:56:02 | * | filwit joined #nim |
| 20:56:19 | Mat4 | ^interpreted=interpretable, invocation = call |
| 21:00:14 | Mat4 | so it is better to forbid such possibility (similar to why it makes sense to avoid nil pointers or references) |
| 21:00:53 | * | Mat4 left #nim (#nim) |
| 21:06:36 | * | brechtm_ joined #nim |
| 21:07:23 | * | brechtm quit (Read error: Connection reset by peer) |
| 21:09:17 | * | nairobi joined #nim |
| 21:09:57 | * | nairobi quit (Read error: Connection reset by peer) |
| 21:18:46 | * | shodan45 quit (Ping timeout: 276 seconds) |
| 21:32:53 | * | Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 21:41:17 | * | Trustable joined #nim |
| 21:43:42 | * | Matthias247 joined #nim |
| 21:45:20 | * | irrequietus quit () |
| 21:45:36 | * | stisa quit (Quit: quitting...) |
| 21:45:39 | * | irrequietus joined #nim |
| 21:48:18 | * | Vladar quit (Remote host closed the connection) |
| 21:50:33 | * | nsf quit (Quit: WeeChat 1.5) |
| 22:00:06 | * | cheatfate_ joined #nim |
| 22:01:08 | * | MyMind joined #nim |
| 22:03:20 | * | rbergmair quit (Ping timeout: 252 seconds) |
| 22:03:21 | * | delian66 quit (Ping timeout: 252 seconds) |
| 22:03:26 | * | rbergmair1 joined #nim |
| 22:03:37 | * | Sembei quit (Ping timeout: 252 seconds) |
| 22:03:37 | * | cheatfate quit (Ping timeout: 252 seconds) |
| 22:03:38 | * | Calinou quit (Ping timeout: 252 seconds) |
| 22:03:39 | * | ldlework quit (Ping timeout: 252 seconds) |
| 22:04:50 | * | delian66 joined #nim |
| 22:04:57 | * | Calinou joined #nim |
| 22:08:33 | * | ldlework joined #nim |
| 22:23:25 | * | SianaGearz quit (Ping timeout: 244 seconds) |
| 22:25:20 | * | SianaGearz joined #nim |
| 22:27:07 | * | Trustable quit (Remote host closed the connection) |
| 22:28:13 | * | gokr quit (Ping timeout: 244 seconds) |
| 22:39:37 | * | rbergmair1 quit (Quit: Leaving.) |
| 22:57:47 | * | irrequietus quit () |
| 23:00:28 | * | Flox421 quit (Remote host closed the connection) |
| 23:00:29 | * | M-max quit (Remote host closed the connection) |
| 23:00:29 | * | M-Quora quit (Remote host closed the connection) |
| 23:04:46 | * | Flox42 joined #nim |
| 23:13:11 | * | Matthias247 quit (Read error: Connection reset by peer) |
| 23:13:28 | * | GustavoLapasta joined #nim |
| 23:15:36 | * | nsf joined #nim |
| 23:28:10 | * | brechtm_ quit (Remote host closed the connection) |
| 23:51:49 | * | M-max joined #nim |
| 23:51:58 | * | M-Quora joined #nim |
