| 00:33:30 | * | bjz joined #nim |
| 00:45:10 | * | yglukhov quit (Remote host closed the connection) |
| 00:46:04 | * | yglukhov joined #nim |
| 00:50:14 | * | yglukhov quit (Ping timeout: 244 seconds) |
| 00:52:35 | * | PMunch quit (Quit: leaving) |
| 01:07:50 | * | desophos joined #nim |
| 01:16:16 | * | mtj_ left #nim ("http://quassel-irc.org - Chat comfortably. Anywhere.") |
| 01:26:35 | * | FuntDobra quit (Quit: leaving) |
| 01:26:44 | * | derka joined #nim |
| 01:26:55 | * | FuntDobra joined #nim |
| 01:36:59 | * | chemist69 quit (Ping timeout: 258 seconds) |
| 01:47:04 | * | derka quit (Ping timeout: 244 seconds) |
| 01:47:27 | * | yglukhov joined #nim |
| 01:50:50 | * | chemist69 joined #nim |
| 01:52:09 | * | yglukhov quit (Ping timeout: 260 seconds) |
| 01:56:04 | * | Jesin quit (Quit: Leaving) |
| 02:21:22 | * | ARCADIVS joined #nim |
| 02:57:16 | * | bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 03:12:00 | * | Demon_Fox joined #nim |
| 03:23:10 | * | bjz joined #nim |
| 03:37:10 | * | chemist69 quit (Ping timeout: 250 seconds) |
| 03:41:20 | * | chemist69 joined #nim |
| 03:45:41 | * | bjz quit (Ping timeout: 248 seconds) |
| 03:47:11 | * | bjz joined #nim |
| 03:48:44 | * | yglukhov joined #nim |
| 03:53:12 | * | yglukhov quit (Ping timeout: 260 seconds) |
| 04:39:25 | * | desophos quit (Read error: Connection reset by peer) |
| 05:04:04 | * | bjz quit (Ping timeout: 268 seconds) |
| 05:04:13 | * | bjz joined #nim |
| 05:21:32 | * | Dankrad1 joined #nim |
| 05:24:38 | * | Dankrad quit (Ping timeout: 250 seconds) |
| 05:44:29 | * | ARCADIVS quit (Quit: ARCADIVS) |
| 05:50:31 | * | yglukhov joined #nim |
| 05:54:37 | * | yglukhov quit (Ping timeout: 240 seconds) |
| 07:45:53 | * | chemist69 quit (Ping timeout: 260 seconds) |
| 07:49:55 | * | bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 07:50:01 | Araq | dom96: the Win installer is seriously broken -.- |
| 07:50:01 | * | chemist69 joined #nim |
| 07:50:14 | Araq | no error message when installing to C:\program files |
| 07:50:22 | Araq | and doesn't ask for permissions |
| 08:05:21 | * | bjz joined #nim |
| 08:24:19 | FromGitter | <gogolxdong> how can I change the value of an object field with type int32? like object.x.inc() |
| 08:24:58 | Araq | @gogolxdong use a 'var' parameter |
| 08:26:01 | * | bjz quit (Ping timeout: 260 seconds) |
| 08:26:47 | * | bjz joined #nim |
| 08:28:42 | FromGitter | <gogolxdong> declare the field as var int32 or else? |
| 08:29:01 | FromGitter | <gogolxdong> ''' |
| 08:29:51 | FromGitter | <gogolxdong> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=57f8ae7f0aa72e3c5be48e0c] |
| 08:30:20 | * | vl409 joined #nim |
| 08:31:23 | * | jh32 quit (Remote host closed the connection) |
| 08:32:28 | * | Matthias247 joined #nim |
| 08:34:05 | * | jh32 joined #nim |
| 08:36:06 | Araq | your 'i: Iterator' is a parameter |
| 08:36:22 | Araq | use 'i: var Iterator' instead |
| 08:36:29 | * | Matthias247 quit (Read error: Connection reset by peer) |
| 08:36:37 | * | bjz quit (Ping timeout: 240 seconds) |
| 08:36:54 | * | bjz_ joined #nim |
| 08:37:49 | * | Matthias247 joined #nim |
| 08:38:38 | * | cheatfate joined #nim |
| 08:39:39 | vl409 | installing from github is a mystery to me |
| 08:40:42 | vl409 | compilation creates only bin/nim ... how do I install tools? |
| 08:41:16 | Araq | you compile the tools with the bin/nim |
| 08:41:37 | vl409 | github readme says: nim e install_tools.nims # but that file executes somthing from "dist" directory |
| 08:41:52 | Araq | where does it say that? |
| 08:41:55 | vl409 | and "dist" is only in the prepackaged .tar.xz |
| 08:42:38 | Araq | ah, my bad, I forgot that readme.md is mirrored on githbu |
| 08:44:13 | vl409 | yeah readme.md line 58 |
| 08:47:01 | * | Demon_Fox quit (Ping timeout: 248 seconds) |
| 08:48:44 | * | Demon_Fox joined #nim |
| 08:48:49 | dom96 | Araq: *sigh* |
| 08:49:32 | dom96 | I knew I shouldn't have removed the requirement for admin rights |
| 08:49:35 | Araq | dom96: yeah plus the installer really doesn't install whatever it is that nimble needs in bin/ |
| 08:50:09 | Araq | why is that in bin/ anyway, can't it be a directory in the stdlib instead? |
| 08:50:41 | dom96 | how should it find the stdlib? |
| 08:51:17 | dom96 | probably best to just statically compile this file into nimble |
| 08:52:03 | Araq | it already does find the stdlib |
| 08:52:14 | Araq | and uses system.nim from it |
| 08:52:25 | Araq | and others for nimscript support |
| 08:53:31 | dom96 | yes, but it's not really part of the stdlib |
| 08:53:59 | dom96 | it should be installed into the system nimble package directory, but Nimble doesn't support that yet |
| 08:54:31 | Araq | lib/packages/nimble would be nice |
| 08:54:38 | Araq | we only have 'docutils' in there so far |
| 08:55:09 | dom96 | well, that's a separate issue |
| 08:55:21 | dom96 | and I won't get into it now because we'll end up bikeshedding. |
| 08:55:44 | Araq | everything is better than in bin/ really |
| 08:55:53 | Araq | no bikeshedding required |
| 08:56:15 | dom96 | I'll just modify nimble to statically compile it |
| 08:56:31 | dom96 | kill 2 birds with 1 stone |
| 08:56:46 | Araq | ok, we need to get 0.15.2 out |
| 08:56:59 | Araq | couple of horrible regressions to fix -.- |
| 08:57:38 | dom96 | this one seems difficult https://github.com/nim-lang/Nim/issues/4856 |
| 08:58:03 | Araq | why? |
| 08:58:11 | dom96 | I dunno |
| 08:58:23 | dom96 | generics are always tricky, no? |
| 08:58:37 | Araq | it's not generics, it's macros |
| 08:58:48 | dom96 | is it something I need to fix? |
| 08:58:58 | dom96 | actually, what do I need to fix? |
| 08:59:10 | Araq | the Windows installer |
| 09:02:36 | vl409 | I didn't want to interrupt ... but now, it's my turn |
| 09:03:08 | vl409 | Then there is this 'nim e install_nimble.nims' |
| 09:04:30 | vl409 | If I use packaged .tar.xz, then I don't use this file at all, when I clone from github it warns, that it will give me screwed-up nimble (that I will not be able to install binary packages) |
| 09:05:24 | Araq | nim e install_nimble.nims is a source based build of Nimble |
| 09:05:27 | vl409 | So what is the use of it. And how should I install nimble into github clone - in the first place. |
| 09:05:33 | Araq | which fits your source based build of Nim. |
| 09:05:49 | Araq | I dunno the warning you talk about, what does it say? |
| 09:06:30 | vl409 | :-) readme.md line 71 |
| 09:06:46 | dom96 | Warning: If you install Nimble this way, you will not be able to use binary Nimble packages or update Nimble easily. The Nimble readme provides thorough instructions on how to install Nimble, so that this isn't a problem. |
| 09:06:54 | dom96 | IIRC I wrote that |
| 09:07:06 | dom96 | Perhaps I exaggerated the issue a bit. |
| 09:07:26 | dom96 | I think you will be able to install binary Nimble packages, but you won't be able to update Nimble easily. |
| 09:08:34 | vl409 | I haven't met any binary package yet, so I can't even test it .... :-) |
| 09:08:49 | dom96 | indeed, they are rare |
| 09:09:44 | Araq | dom96: can I remove this warning? |
| 09:10:40 | dom96 | yeah |
| 09:10:42 | vl409 | So the nimble/readme.md has 2 sections about installation |
| 09:11:36 | vl409 | First talks about linux (recommends using nim e install_nimble.nims) and second about windows (recomends clone nimble from github) |
| 09:12:43 | vl409 | When I install nim by git clone, maybe i should install nimble this way too. Isn't that better. Or does it work? Or how do you do it? |
| 09:13:47 | dom96 | install_nimble.nims performs a git clone anyway |
| 09:13:49 | dom96 | so it's the same |
| 09:13:54 | dom96 | the only difference is where it is installed |
| 09:14:05 | dom96 | install_nimble.nims installs it into nim/bin |
| 09:14:23 | euantor | The preferably way to install Nimble on Windows should be the windows installer (when it works) really |
| 09:14:28 | dom96 | whereas if you install it using Nimble itself it is installed into ~/.nimble/pkgs/nimble-ver |
| 09:14:35 | dom96 | yeah |
| 09:14:43 | euantor | The only reason you should need a different version than the stable version is if you're working on developing Nimble |
| 09:15:04 | euantor | And eventually we should get to that state for Mac/Linux too |
| 09:15:19 | vl409 | But I can't install nimble using nimble. I don't have any nimble at that time. |
| 09:15:56 | vl409 | I have bin/nim and comandline. which should be ok |
| 09:16:26 | dom96 | vl409: in order to install like this, you need to clone the Nimble repo, compile Nimble, then install Nimble by using the Nimble you just compiled. |
| 09:16:38 | dom96 | You are better off just using nimble_install.nims |
| 09:17:07 | vl409 | aha. Now I'm starting getting it. |
| 09:18:06 | * | elrood joined #nim |
| 09:18:20 | dom96 | Araq: Maybe it would make sense for the Nimble repo to contain this install.nims file? |
| 09:18:29 | dom96 | or maybe it should also contain a similar one |
| 09:20:50 | vl409 | just to reiterate. Your remark about 'easy update' means: If I install nimble using nimble, I'm cool. |
| 09:21:20 | vl409 | Any other way is not 'easy' (hence more typing on commandline). |
| 09:22:11 | vl409 | Or when nimble updates nimble, does it perform some additional checks about installed packages? |
| 09:22:55 | dom96 | I don't think you should worry too much about this to be honest. |
| 09:23:09 | dom96 | Re-running install_nimble.nims to update should work |
| 09:23:43 | vl409 | I thought that too. |
| 09:24:48 | vl409 | Then there is this nimscript. |
| 09:25:36 | vl409 | When I installed nim, there was only bin/nim. So I went on to install_tools.nim and there was no nimscript. |
| 09:26:00 | vl409 | So I started greping that nimble list | grep ... |
| 09:26:22 | vl409 | And I still could not figure out how to install this nimscript :-) |
| 09:28:00 | vl409 | BTW neither 'nim -h' nor 'nim --advanced' still doesn't list 'e' as one of its commands. |
| 09:29:30 | * | Kingsquee quit (Quit: https://i.imgur.com/qicT3GK.gif) |
| 09:30:13 | Araq | dom96: I think the readme says it's part of Nim now |
| 09:30:22 | Araq | the repo is just for issues and its code |
| 09:30:33 | Araq | not for anything else, as far as I'm concerned |
| 09:31:02 | Araq | vl409: nimscript is 'nim e ' |
| 09:31:17 | Araq | good point about that it's not documented |
| 09:31:34 | vl409 | I know. Now. But the initial surprise... |
| 09:31:37 | dom96 | okay, I don't think the Nimble readme says that yet |
| 09:32:05 | Araq | I remember updating it, but I don't remember either |
| 09:32:51 | elrood | or, to put it bluntly, nim is in a messy state right now. you guys really need to clean up, streamline your release process, automate as much as possible, and make sure that there's one easy way to get things up and running, regardless of whether one got nim through tarball, installer, git, and that this way is documented prominently and identically on the homepage and a readme |
| 09:33:09 | arnetheduck | so nimscript is nim run by the internal compiler vm? |
| 09:33:14 | elrood | the way you're currently handling things is just digging holes for yourself to get lost in |
| 09:33:25 | vl409 | nimscript is a module |
| 09:33:59 | vl409 | I wouldn't calling it as digging holes. |
| 09:34:34 | vl409 | They just can't see by the eyes of a newbee anymore. So they miss these moments. |
| 09:36:39 | elrood | keeping the newcomer's experience in mind is absolutely crucial for nim to succeed, though |
| 09:36:53 | vl409 | Okey. One last point. You should give list of recommended nimble packages to install. Recommended means either as a library or as a clean code. |
| 09:37:37 | vl409 | Reading clean code is important if you're new. |
| 09:37:54 | vl409 | So I installed jester. Because I didn't know any better. |
| 09:37:55 | Araq | elrood: no, things have been sorted out, but the docs lag a bit behind |
| 09:38:25 | Araq | and a single way to install things doesn't work, so there are 2. |
| 09:38:31 | vl409 | And there are those immediate pragmas all over the place... |
| 09:40:25 | Araq | elrood: you either install via github or the tarball. the tarball is for stable releases. |
| 09:41:32 | elrood | you really don't need to tell me, i have been using nim for a while, am comfortably using git, and don't see those problems myself |
| 09:44:02 | elrood | but reading about all the little things that need patching right after a release which could and should have been automated, contradicting docs and different ways to get your installation set up, and undocumented but essential commands is just a mess and will drive many interested newcomers away, which is just sad and unnecessary |
| 09:45:18 | vl409 | Or maybe it should be kept that way :-) |
| 09:46:55 | elrood | well, that pretty much depends on whether you like to see nim used or just become a footnote in proglang history, i guess |
| 09:47:20 | vl409 | Or whether you sell a book :-) |
| 09:47:33 | vl409 | No that was ugly joke. I'm sorry |
| 09:48:23 | Araq | how about you add even more pathos to an issue you don't even experience on your own? |
| 09:48:32 | Araq | ;-) |
| 09:48:39 | elrood | nah, we're not that thin-skinned in here, no worries, vl409. but to sell a book you still need an audience ;) |
| 09:49:29 | flyx | seems like directive parameters are not supported by Nim's rST implementation, is that correct? |
| 09:50:08 | elrood | there's no pathos here, you just need to open your eyes and imagine the effect on all those people that don't make it to this channel, Araq. it's your project, handle it whichever way you like |
| 09:50:40 | Araq | elrood: all I can do is fix reported issues one after another. claiming "it's all a big mess" is not helpful. |
| 09:51:08 | vl409 | Ther is pathos. It is mine. But 'elrood' maybe did experienced it, when he came first time. |
| 09:52:57 | * | chemist69 quit (Ping timeout: 240 seconds) |
| 09:55:39 | * | chemist69 joined #nim |
| 09:57:14 | elrood | it's not all a big mess, and i've not said that. but certain aspects of nim are undeniably in a messy state, and if those affect the first experience newcomers make is just about the worst possible thing for an aspiring project which is in desperate need to attract interest |
| 09:59:28 | cheatfate | dom96, and what is problem happend with admin rights? |
| 10:00:02 | * | nsf joined #nim |
| 10:01:52 | dom96 | yeah, I guess in some ways it was worse. |
| 10:02:10 | dom96 | We need to figure out if we can get NSIS to prompt for admin rights when required |
| 10:04:52 | Araq | I would remove the NSIS installers |
| 10:05:09 | Araq | they are hard to maintain and keep producing problems |
| 10:05:28 | Araq | and they don't even solve the most common problem: having a wrong gcc.exe in the path |
| 10:12:59 | cheatfate | so we are going to move to msi installers? |
| 10:39:29 | elrood | one wild guess is that most common problems could be solved or at least mitigated if the homepage had a visibly placed direct link to a wiki with faq and common solutions |
| 10:41:32 | euantor | Well, the website is getting a redesign |
| 10:42:05 | elrood | i know, it's been getting a redesign for quite some time now |
| 10:43:33 | elrood | my gripe is that the redesign might a little too focussed on the design aspect instead of getting something more usable and navigatable up and running soon |
| 10:43:58 | euantor | From what I've see it seems to focus on both the design and the usability |
| 10:44:42 | elrood | that'd be great, and if it was ready soon-ish that'd be even better |
| 10:44:54 | euantor | I don't think it's too far off |
| 10:45:27 | euantor | But I'm not 100%. The person who's done all the work on it so far is busy with university and nobody else seems to be trying to contribute to it |
| 10:47:38 | Araq | cheatfate: hell no. I would simply provide a .zip file |
| 10:47:54 | * | gangstacat quit (Ping timeout: 250 seconds) |
| 10:50:20 | * | yglukhov joined #nim |
| 10:51:43 | * | yglukhov quit (Remote host closed the connection) |
| 10:54:15 | * | gangstacat joined #nim |
| 10:56:34 | Araq | but WiX seems to have come a long way |
| 10:57:56 | elrood | makes you wonder if any german-speaking developers have been involved in that toolkit |
| 11:01:59 | * | Snircle joined #nim |
| 11:19:46 | dom96 | we should provide a zip anyway |
| 11:19:50 | dom96 | a portable version of Nim |
| 11:29:05 | Araq | we could also provide nim32.exe and nim64.exe |
| 11:29:15 | Araq | and have a single zip |
| 11:30:33 | dom96 | no |
| 11:30:39 | dom96 | What is the issue with gcc installation? |
| 11:31:24 | Araq | gcc and nim disagree on the pointer size |
| 11:32:21 | * | xet7 joined #nim |
| 11:34:37 | Araq | hmm can I remove the "update" command from koch? |
| 11:34:57 | Araq | nobody builds with -d:withUpdate anyway |
| 11:35:05 | Araq | and I don't know what it does |
| 11:35:53 | Araq | and it contains stuff like https://github.com/Araq/Nim/zipball/master |
| 11:35:59 | Araq | no idea what that is |
| 11:39:17 | elrood | i'd take the radical approach on stuff like that. if you don't know what it's for and it's not documented anywhere, get rid of it. if something breaks, revert and make the effort to properly document it |
| 11:40:41 | vl409 | I want to know the name of the movie you was watching that night .. zipball master .. heh |
| 11:41:02 | * | pafmaf joined #nim |
| 11:42:53 | Araq | elrood: alright, it's gone |
| 11:43:25 | Araq | I'm patching koch.exe so that can add stuff to the PATH on Windows and detect a gcc |
| 11:43:41 | Araq | so people should unzip and run 'koch finish' |
| 11:44:12 | Araq | shitty+working wins over fancy+broken now |
| 11:44:27 | Araq | any better name for 'finish'? |
| 11:44:58 | elrood | well, in the context of unzip.. :P |
| 11:45:32 | vl409 | deploy |
| 11:45:57 | Araq | vl409: isn't that more confusing? |
| 11:46:05 | euantor | yes, finish is better |
| 11:46:27 | euantor | Ideally it would just be a case of running an exe without a command, but Koch already does other things |
| 11:46:43 | euantor | (e.g. `./install.exe`) |
| 11:47:26 | Araq | euantor: any exe with 'install' in it is subject to UAC |
| 11:47:36 | Araq | took me hours to figure out ... |
| 11:47:40 | euantor | damn, of course |
| 11:47:49 | euantor | Damn Microsoft |
| 11:48:05 | Araq | I want to add a 'finish.bat' that delegates to 'koch finish' |
| 11:48:15 | euantor | Yes, that would make sense |
| 11:48:27 | euantor | That would definitely make it clear I reckon |
| 11:48:49 | Araq | and since it reads something from stdin it doesn't just disappear on double click |
| 11:48:49 | euantor | Especially if the root of the ZIP is getting cleared up like dom96 talked about a couple of days ago |
| 11:49:38 | elrood | koch setup / install , but those are only marginally clearer, if at all |
| 11:50:11 | Araq | it doesn't install. it simply changes the environment |
| 11:50:30 | dom96 | ./koch setup-path |
| 11:50:41 | Araq | adds things to PATH via the registry and creates a start menu entry |
| 11:50:42 | elrood | that's all the installation that nim needs, isn't it? but true, probably not the perfect choice |
| 11:50:46 | dom96 | best to make it as descriptive as possible |
| 11:50:50 | dom96 | people will be copying the commands anyway |
| 11:51:02 | dom96 | screw the start menu entry |
| 11:51:09 | dom96 | PATH is the only thing that's important |
| 11:51:11 | Araq | it's 3 lines of code |
| 11:51:16 | Araq | and it's convenient |
| 11:51:18 | Araq | for many. |
| 11:51:26 | dom96 | it's also annoying for many |
| 11:51:28 | dom96 | including myself |
| 11:51:44 | dom96 | make it interactive at least |
| 11:51:47 | Araq | you don't have to press 'Y' for this question... |
| 11:51:49 | dom96 | and ask the user what they want |
| 11:51:53 | Araq | of course |
| 11:51:57 | dom96 | ok good |
| 11:52:32 | vl409 | is this the thing that checks version of installed gcc? |
| 11:53:14 | Araq | vl409: it checks for gcc and then asks about the path and then asks about start menu entry |
| 11:54:02 | elrood | no clue how far along filesystem features have come on windows, but perhaps even placing a link in one directory that is commonly included in PATH would suffice? |
| 11:54:16 | elrood | at least that's what i do on linux |
| 11:54:41 | Araq | I don't understand your question. |
| 11:55:27 | Araq | so ... setup-path vs finish? |
| 11:55:41 | euantor | He's asking if putting a shortcut to a program in somewhere like `C:\Windows\System32` makes the program available in the path |
| 11:55:47 | euantor | The answer, to my knowledge, is no |
| 11:55:48 | * | yglukhov joined #nim |
| 11:56:52 | elrood | basically, yep. no clue if that'd be more convenient on windows than changing the PATH, though |
| 11:57:26 | * | Salewski joined #nim |
| 11:57:34 | * | yglukhov quit (Remote host closed the connection) |
| 11:58:17 | Araq | elrood: that is possible but requires admin rights and sucks |
| 11:58:45 | elrood | probably installing something in windows' system directory is even more invasive, but perhaps there's another directory that's commonly on the path, equivalent to /usr/local/bin |
| 11:59:16 | * | dddddd joined #nim |
| 11:59:17 | Araq | no, the convention is to modify the path |
| 11:59:55 | elrood | alright, was just curious whether coventions on windows changed in the last few years. thanks |
| 12:00:17 | Salewski | Araq, thanks for fixing #4844 for tables nim. Has that never worked before, or was it working before Nim 0.15? |
| 12:00:43 | Araq | never worked afaict |
| 12:00:56 | Salewski | Ok, thanks. |
| 12:01:15 | Salewski | Bye. |
| 12:01:19 | * | Salewski left #nim (#nim) |
| 12:01:22 | Araq | you're welcome |
| 12:01:31 | elrood | btw, any recommendations for a new keyboard? this one is swallowing keypresses and sometimes even whole words occasionally? O;-) sorry |
| 12:01:37 | elrood | *conventions |
| 12:05:29 | Araq | sorry, I'm not a hardware guy |
| 12:07:08 | hohlerde | elrood: overall I am very happy with Logitech keyboards |
| 12:07:39 | elrood | thanks, just kidding and offtopically trying to find an apology for my horrible typing skills, ignore me.. i'll see myself out ;) |
| 12:38:33 | hohlerde | that hookrace.net blog from def, what blog software/theme is it based on? |
| 12:45:37 | * | enthus1ast quit (Ping timeout: 240 seconds) |
| 12:47:42 | * | Demon_Fox quit (Quit: Leaving) |
| 12:49:59 | arnetheduck | nlvm doesn't need gcc ;) |
| 12:50:07 | * | yglukhov joined #nim |
| 12:51:09 | arnetheduck | it needs a developer with windows however |
| 12:51:41 | Araq | I tried to build LLDB for windows once. |
| 12:52:22 | Araq | I don't intend to try again. ever. |
| 12:52:33 | * | yglukhov quit (Remote host closed the connection) |
| 12:52:52 | Araq | fortunately LLVM is easier to build |
| 12:55:17 | * | brechtm joined #nim |
| 12:55:50 | * | brechtm quit (Remote host closed the connection) |
| 12:56:29 | * | brechtm joined #nim |
| 13:23:28 | * | pafmaf quit (Ping timeout: 250 seconds) |
| 13:33:26 | * | nsf quit (Quit: WeeChat 1.5) |
| 13:40:13 | BlaXpirit | please advise on this issue with identifier construction: https://glot.io/snippets/ej6apcm021 |
| 13:42:31 | * | ARCADIVS joined #nim |
| 13:42:39 | arnetheduck | thought lldb and llvm used the same cmake build? |
| 13:45:58 | * | Matthias247 quit (Read error: Connection reset by peer) |
| 13:50:44 | * | libman joined #nim |
| 13:52:40 | * | irrequietus quit () |
| 13:52:56 | * | irrequietus joined #nim |
| 13:53:24 | baabelfish | Araq: static: assert(not(type(1f) is float)) # this is the single most wtf thing I know |
| 13:54:42 | baabelfish | one would assume 1f32 is float32, 1f64 is float64 and 1f is float |
| 13:55:01 | * | chemist69 quit (Ping timeout: 260 seconds) |
| 13:55:44 | * | brechtm quit (Remote host closed the connection) |
| 13:57:31 | baabelfish | sorry about the aggressive wording, didn't mean to be |
| 13:58:56 | * | irrequietus quit (Remote host closed the connection) |
| 13:59:24 | * | chemist69 joined #nim |
| 14:02:07 | * | irrequietus joined #nim |
| 14:04:32 | * | cheatfate quit (Ping timeout: 260 seconds) |
| 14:30:15 | flyx | is there anything like a `staticWrite` complementing `staticRead`? |
| 14:31:38 | libman | You can hack together anything with staticExec... |
| 14:32:30 | flyx | yeah well, I can just "echo foo >> somefile" repeatedly to write into it, but that wouldn't be nice, would it |
| 14:51:26 | BlaXpirit | flyx, have you tried simply writing any code in a static: block? |
| 14:52:00 | flyx | BlaXpirit: afaik normal file operations do not work in a static block, that's why staticRead exists |
| 14:53:08 | BlaXpirit | you're probably right |
| 15:13:51 | * | Trustable joined #nim |
| 15:17:31 | * | Demon_Fox joined #nim |
| 15:23:53 | * | Dankrad1 quit (Read error: Connection reset by peer) |
| 15:24:25 | * | enthus1ast joined #nim |
| 15:29:30 | * | yglukhov joined #nim |
| 15:40:17 | * | brechtm joined #nim |
| 15:43:34 | * | bjz_ quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 15:48:15 | Araq | flyx: writeFile works in a static block |
| 15:52:41 | enthus1ast | vl409 fyi ntfs can create symlinks (like ln -s) and junctions (like ln) |
| 15:52:54 | flyx | okay, thanks |
| 15:53:27 | enthus1ast | but afaik for junctions you need a third party tool or call the winapi |
| 15:53:28 | def- | hohlerde: jekyll, poole, monokai: https://github.com/HookRace/hookrace-web/ |
| 15:54:45 | hohlerde | thx def- |
| 15:58:06 | * | brechtm_ joined #nim |
| 16:01:38 | * | brechtm quit (Ping timeout: 250 seconds) |
| 16:07:08 | dom96 | Araq: Does readFile? |
| 16:07:25 | Araq | dunno |
| 16:14:48 | * | brechtm_ quit (Remote host closed the connection) |
| 16:15:25 | * | kulelu88 joined #nim |
| 16:17:25 | flyx | so, `nim rst2html` now dumps some random JavaScript includes into the output |
| 16:17:45 | flyx | apparently this is meant to be used for the nim source doc generation, but utterly useless for rst2html |
| 16:18:51 | * | brechtm joined #nim |
| 16:19:27 | flyx | also, why does `nim rst2html` not load a nimdoc.cfg in the parent folder? doesn't it search in parent folder like for nim.cfg? |
| 16:24:16 | * | brechtm quit (Remote host closed the connection) |
| 16:26:43 | dom96 | federico3: you there? Any issues with packaging 0.15.0 we should know about? |
| 16:33:55 | dom96 | Araq: I think that in these release notes we need to make it clear that Nimble should be packaged together with Nim |
| 16:37:07 | flyx | ah, rst2html *does* search the parent paths, but not of current working dir, but of the rst file |
| 16:40:22 | federico3 | dom96: you are aware of the special handling of nimscriptapi; nim doc2 failing on coro.nim; and currently using koch install? |
| 16:41:17 | dom96 | The first one I will fix soon |
| 16:41:32 | dom96 | I am aware of others but thank you for reminding me |
| 16:41:54 | dom96 | Araq: Any ideas how we can fix coro.nim? |
| 16:42:07 | dom96 | federico3: as for the last one, what should we do instead? |
| 16:43:32 | * | pafmaf joined #nim |
| 16:49:44 | * | pafmaf quit (Ping timeout: 250 seconds) |
| 16:52:33 | flyx | and I can get rid of the JavaScript stuff with my own nimdoc.cfg |
| 16:54:50 | dom96 | flyx: create an issue for this stuff |
| 16:56:11 | flyx | dom96: I am not sure if there is actually a bug. |
| 16:56:53 | dom96 | if it adds doc-related things to the output of rst2html then that sounds like a bug |
| 16:57:03 | dom96 | Doesn't hurt to report it |
| 16:57:08 | flyx | okay, will do |
| 17:00:19 | dom96 | Araq: Nim installation should include the compiler source, right? |
| 17:00:27 | dom96 | Arch package doesn't seem to: https://www.archlinux.org/packages/community/x86_64/nim/files/ |
| 17:00:32 | Araq | yes |
| 17:00:49 | Araq | I though about moving compiler to lib/ |
| 17:00:59 | * | xet7 quit (Quit: Leaving) |
| 17:01:02 | Araq | but it's tedious to move everything |
| 17:01:18 | dom96 | Where does ./koch install install it to? |
| 17:01:20 | Araq | I'm too used to my stuff being in compiler/ |
| 17:01:47 | Araq | oh yeah, I think our install.sh doesn't put it there either |
| 17:02:03 | dom96 | omg lol |
| 17:02:04 | Araq | meh |
| 17:02:12 | * | xet7 joined #nim |
| 17:02:19 | dom96 | It needs to. |
| 17:02:24 | dom96 | That's a showstopper, please fix it |
| 17:02:36 | dom96 | Gonna report it |
| 17:03:19 | Araq | well it's not like our config adds lib/compiler to the --path |
| 17:04:52 | dom96 | install it into lib/packages/compiler I guess |
| 17:05:21 | dom96 | In the future this will become the global Nimble package directory I guess |
| 17:05:28 | dom96 | similar to Python's `site-packages` |
| 17:07:17 | * | cheatfate joined #nim |
| 17:18:36 | * | dddddd quit (Remote host closed the connection) |
| 17:25:20 | federico3 | dom96: sorry I'm afk |
| 17:33:51 | * | zama quit (Quit: leaving) |
| 17:36:19 | * | zama joined #nim |
| 17:41:04 | Araq | dom96: yeah well it needs some thinking |
| 17:41:13 | Araq | as it's already a nimble package anyway |
| 17:44:13 | * | Matthias247 joined #nim |
| 17:47:29 | * | vl409 quit (Quit: Leaving) |
| 17:51:29 | dom96 | Araq: well, let's discuss it |
| 17:53:27 | * | pafmaf joined #nim |
| 17:56:43 | * | Snircle quit (Quit: Textual IRC Client: www.textualapp.com) |
| 17:57:43 | * | Snircle joined #nim |
| 18:10:29 | dom96 | btw the other issue with windows is the fact that you renamed the ssl DLLs |
| 18:10:42 | dom96 | libssl64 looks for libeay32.dll |
| 18:11:00 | dom96 | AFAIK there is no way to change this without recompiling |
| 18:11:43 | Araq | hmm? I patched this DLL |
| 18:11:47 | Araq | months ago |
| 18:12:04 | dom96 | then why did the forum user have trouble? |
| 18:12:08 | dom96 | I also had trouble with this |
| 18:14:44 | * | yglukhov quit (Remote host closed the connection) |
| 18:16:05 | * | Demon_Fox quit (Quit: Leaving) |
| 18:17:27 | Araq | I will look into it and fix it again |
| 18:17:32 | flyx | do I see it correctly that there is currently no proc in stdlib to walk over items in a directory sorted by name? |
| 18:18:01 | flyx | walkDir doc contains this: „but not necessarily in this order!“ |
| 18:18:05 | flyx | so, the order is undefined? |
| 18:18:06 | Araq | usually walkDir et al are by name |
| 18:18:10 | * | brechtm joined #nim |
| 18:18:12 | Araq | yes |
| 18:18:23 | Araq | but the OS usually gets it right |
| 18:18:29 | flyx | ah well |
| 18:18:32 | Araq | depends on how picky you wanna be |
| 18:18:54 | flyx | if it works under common OSes (Mac/Win/Linux) I'm fine |
| 18:19:11 | flyx | if it works under BSDs, it's a plus |
| 18:19:57 | flyx | since this is only important for NimYAML doc generation and that will usually be done by myself, it does not need to be super portable |
| 18:20:10 | dom96 | Araq: thanks |
| 18:20:36 | * | brechtm quit (Remote host closed the connection) |
| 18:20:51 | dom96 | flyx: you can convert the result of the iterator to a seq and then sort |
| 18:20:53 | Araq | flyx: why does the order matter for NimYaml doc generation? o.O |
| 18:22:04 | flyx | Araq: I use a fancy directory history for storing the snippets used in the documentation, so that I can automatically test them. and they should have a certain order when generating the documentation, so I named the folders `01`, `02`, … |
| 18:22:09 | flyx | *hierarchy |
| 18:22:22 | * | yglukhov joined #nim |
| 18:23:25 | flyx | dom96: yeah but I am allergic to O(n log n) ;) |
| 18:29:24 | dom96 | Maybe you could patch the stdlib to give the iterators an extra parameter for this? :) |
| 18:29:24 | * | chemist69 quit (Ping timeout: 250 seconds) |
| 18:30:51 | flyx | maybe I could |
| 18:31:07 | flyx | let's see if I am motivated when I'm done with my own code |
| 18:33:54 | * | chemist69 joined #nim |
| 18:43:27 | * | brechtm joined #nim |
| 18:53:06 | flyx | when having `var (a, b, c) = path.splitFile()`, is it possible to leave out values I won't use? |
| 18:53:25 | flyx | the compiler emits warning when I only use the third value |
| 18:56:22 | dom96 | _ |
| 18:58:52 | flyx | ah cool, I have only tried `(,,c)` |
| 18:59:03 | flyx | any reason for using `_` instead of just leaving out? |
| 18:59:17 | dom96 | Why is the default $ for objects showing "..." for some fields? |
| 18:59:21 | dom96 | it's becoming infuriating |
| 19:01:19 | Araq | it indicates there are fields that are not visible for $ |
| 19:01:27 | Araq | it's called "information hiding" |
| 19:02:02 | dom96 | that's silly |
| 19:02:45 | dom96 | What procedure in the compiler sources reads the nim.cfg files? |
| 19:07:10 | * | FuntDobra quit (Quit: leaving) |
| 19:08:22 | enthus1ast | dom96 looks like loadConfigs(DefaultConfig) # load all config files |
| 19:08:26 | enthus1ast | proc loadConfigs*(cfg: string) = |
| 19:08:35 | enthus1ast | in nimconf.nim |
| 19:09:43 | dom96 | nice, thanks! |
| 19:12:22 | enthus1ast | np |
| 19:15:30 | dom96 | cool, that worked |
| 19:37:00 | * | pafmaf quit (Ping timeout: 250 seconds) |
| 19:50:38 | * | yglukhov quit (Remote host closed the connection) |
| 19:56:47 | flyx | hmm, what can be the problem if fileExists() returns false on some path which actually exists? |
| 19:57:06 | flyx | permissions? I can cat the file from the command line |
| 19:58:10 | flyx | I also do not change the working directory |
| 20:00:29 | * | yglukhov joined #nim |
| 20:01:24 | flyx | it seems like calling startProcess with a workingDir changes the current working dir o.O |
| 20:02:57 | flyx | at least when it breaks |
| 20:03:09 | flyx | I guess startProcess needs an absolute path to some executable? |
| 20:03:33 | flyx | is there a way to get the path to the nim executable within a nim source file? at compile time? |
| 20:14:13 | dom96 | staticExec("which nim")? |
| 20:16:49 | * | FuntDobra joined #nim |
| 20:16:56 | * | brechtm quit (Remote host closed the connection) |
| 20:17:39 | * | nsf joined #nim |
| 20:19:55 | flyx | dom96: that only works if the nim executable is in the PATH. |
| 20:20:09 | flyx | dom96: I mean yeah, it would work in most of the cases |
| 20:21:25 | flyx | dom96: I just thought, Nim *could* propagate the path to its executable as that is the first argument of the process (on most OSes) and make it available at compile time |
| 20:21:30 | flyx | anyway, this seems to be a pretty serious issue: https://github.com/nim-lang/Nim/issues/4867 |
| 20:22:17 | dom96 | sure, it could |
| 20:31:09 | * | yglukhov quit (Remote host closed the connection) |
| 20:32:21 | * | cheatfate_ joined #nim |
| 20:35:02 | * | brechtm joined #nim |
| 20:35:32 | * | cheatfate quit (Ping timeout: 260 seconds) |
| 20:36:27 | * | brechtm quit (Remote host closed the connection) |
| 20:36:56 | * | cheatfate_ quit (Ping timeout: 260 seconds) |
| 20:37:46 | * | cheatfate joined #nim |
| 20:40:41 | * | brechtm joined #nim |
| 20:44:09 | dom96 | hrm, it is actually strangely surprising that there is not much better than NSIS/Inno/WiX |
| 20:45:34 | * | bjz joined #nim |
| 20:53:04 | dom96 | But then again, so many big names are using NSIS. We must be doing something wrong. |
| 20:53:07 | * | libman left #nim (#nim) |
| 20:56:30 | Calinou | dom96: Node.js adopted .msi a while ago |
| 20:56:48 | Calinou | but you are right, nobody seems to have written a modern Windows installer framework |
| 20:57:17 | Calinou | NSIS is pretty broken IMO, I tried it once, couldn't change header/finish images and had no idea why (they were .bmp) |
| 21:00:19 | * | Kaini quit (Ping timeout: 258 seconds) |
| 21:09:05 | * | brechtm quit (Remote host closed the connection) |
| 21:09:43 | * | brechtm joined #nim |
| 21:10:13 | * | Ven_ joined #nim |
| 21:31:41 | * | yglukhov joined #nim |
| 21:35:55 | * | yglukhov quit (Ping timeout: 244 seconds) |
| 21:36:53 | * | ARCADIVS quit (Quit: ARCADIVS) |
| 21:44:19 | * | gangstacat quit (Quit: Ĝis) |
| 21:47:23 | * | pie__ quit (Ping timeout: 268 seconds) |
| 21:48:47 | * | gangstacat joined #nim |
| 22:23:46 | * | cheatfate quit (Ping timeout: 268 seconds) |
| 22:33:04 | * | yglukhov joined #nim |
| 22:37:17 | * | yglukhov quit (Ping timeout: 240 seconds) |
| 22:52:09 | * | bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 22:53:39 | * | bjz joined #nim |
| 23:02:35 | * | elrood quit (Quit: Leaving) |
| 23:03:25 | * | bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 23:08:13 | * | Ven_ quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 23:12:51 | * | desophos joined #nim |
| 23:16:49 | * | brechtm quit (Remote host closed the connection) |
| 23:18:01 | * | brechtm joined #nim |
| 23:20:02 | * | bjz joined #nim |
| 23:37:02 | FromGitter | <gogolxdong> build nimx example with raised `could not import: SDL_CaptureMouse` |
| 23:39:30 | * | irrequietus quit () |
| 23:50:32 | BlaXpirit | gogolxdong, > This function is available since SDL 2.0.4. |
| 23:50:43 | BlaXpirit | check out that you have the required version |
| 23:51:36 | BlaXpirit | and maybe the library info page should mention this |
| 23:54:18 | dom96 | Awesome. One of Nim in Action's reviewers is from Ulster University (A uni in Northern Ireland!) |
| 23:58:30 | * | brechtm quit (Remote host closed the connection) |