| 00:09:40 | * | Matthias247 quit (Read error: Connection reset by peer) |
| 00:24:49 | * | brson_ quit (Read error: Connection reset by peer) |
| 00:25:58 | * | brson joined #nim |
| 00:27:12 | * | yglukhov joined #nim |
| 00:31:30 | * | yglukhov quit (Ping timeout: 240 seconds) |
| 00:31:30 | * | synshroud quit (Quit: ZNC 1.6.4 - http://znc.in) |
| 00:33:02 | * | synshroud joined #nim |
| 01:29:34 | * | bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 01:31:48 | * | bjz joined #nim |
| 01:38:14 | * | brson quit (Quit: leaving) |
| 01:50:24 | demi- | ok, i submitted a PR, i've tested it locally and was able to also submit a PR for the library i wrote today to the packages repo. |
| 01:53:56 | * | vlad1777d quit (Quit: Leaving) |
| 01:58:14 | * | bjz quit (Ping timeout: 255 seconds) |
| 02:01:11 | * | bjz joined #nim |
| 02:03:59 | * | Jesin quit (Ping timeout: 264 seconds) |
| 02:11:51 | * | bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 02:20:06 | * | chemist69 quit (Ping timeout: 258 seconds) |
| 02:20:59 | * | MyMind joined #nim |
| 02:21:10 | * | Sembei quit (Ping timeout: 240 seconds) |
| 02:26:40 | * | Jesin joined #nim |
| 02:33:13 | * | Kingsquee quit (Read error: Connection reset by peer) |
| 02:33:28 | * | Kingsquee joined #nim |
| 02:33:47 | * | chemist69 joined #nim |
| 02:46:42 | * | bjz joined #nim |
| 04:12:24 | * | libman quit (Ping timeout: 240 seconds) |
| 04:38:04 | * | eizua joined #nim |
| 04:43:42 | * | dv- quit (Read error: Connection reset by peer) |
| 04:52:17 | * | dv- joined #nim |
| 05:12:30 | * | pregressive joined #nim |
| 05:22:09 | * | pregressive quit (Remote host closed the connection) |
| 05:22:41 | * | pregressive joined #nim |
| 05:26:51 | * | yglukhov joined #nim |
| 05:27:29 | * | pregressive quit (Ping timeout: 255 seconds) |
| 05:29:07 | * | bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 05:31:34 | * | yglukhov quit (Ping timeout: 256 seconds) |
| 05:47:44 | * | bjz joined #nim |
| 06:15:44 | * | libman joined #nim |
| 06:23:04 | * | chemist69 quit (Ping timeout: 240 seconds) |
| 06:28:11 | * | chemist69 joined #nim |
| 06:29:43 | * | eizua quit (Quit: Leaving) |
| 07:15:23 | * | libman quit (Ping timeout: 260 seconds) |
| 07:19:17 | * | nsf joined #nim |
| 07:19:36 | * | libman joined #nim |
| 07:28:36 | * | libman quit (Ping timeout: 245 seconds) |
| 07:32:14 | * | bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 07:34:10 | * | libman joined #nim |
| 07:35:07 | * | bjz joined #nim |
| 07:37:32 | * | bjz quit (Client Quit) |
| 07:52:18 | * | GustavoLapasta joined #nim |
| 07:54:08 | * | Vladar joined #nim |
| 08:10:06 | * | bjz joined #nim |
| 08:19:01 | * | bjz_ joined #nim |
| 08:19:45 | * | bjz quit (Ping timeout: 240 seconds) |
| 08:39:50 | * | bjz joined #nim |
| 08:40:47 | * | yglukhov joined #nim |
| 08:41:43 | * | bjz_ quit (Ping timeout: 260 seconds) |
| 08:43:04 | * | libman quit (Ping timeout: 240 seconds) |
| 08:43:15 | * | peted quit (Quit: WeeChat 1.4) |
| 08:43:28 | * | peted joined #nim |
| 08:48:12 | * | libman joined #nim |
| 08:52:35 | * | arnetheduck quit (Remote host closed the connection) |
| 08:54:02 | * | xet7 joined #nim |
| 09:07:14 | * | yglukhov quit (Remote host closed the connection) |
| 09:07:48 | * | yglukhov joined #nim |
| 09:10:48 | * | yglukhov quit (Remote host closed the connection) |
| 09:11:02 | * | yglukhov joined #nim |
| 09:19:22 | * | yglukhov quit (Remote host closed the connection) |
| 09:20:07 | * | yglukhov joined #nim |
| 09:48:18 | * | Ven joined #nim |
| 09:56:00 | * | arnetheduck joined #nim |
| 10:02:50 | * | yglukhov quit (Remote host closed the connection) |
| 10:04:13 | * | Ven quit (Read error: Connection reset by peer) |
| 10:06:47 | * | Ven joined #nim |
| 10:10:51 | * | yglukhov joined #nim |
| 10:50:06 | * | Matthias247 joined #nim |
| 10:53:17 | * | Matthias247 quit (Read error: Connection reset by peer) |
| 10:54:38 | * | Matthias247 joined #nim |
| 11:22:16 | * | gokr joined #nim |
| 11:42:09 | * | yglukhov quit (Remote host closed the connection) |
| 11:47:34 | * | ARCADIVS joined #nim |
| 11:47:57 | * | Ven quit (Ping timeout: 276 seconds) |
| 11:50:25 | * | Ven joined #nim |
| 11:51:49 | dom96 | Araq: I noticed that you want different forum sections. I think it may be a good idea to support tags like Discourse does |
| 11:57:43 | dom96 | But then they have this problem... https://users.rust-lang.org/t/welcome-to-the-new-rust-forum/21/3 |
| 11:57:57 | dom96 | So maybe different boards/sections are a better idea |
| 11:59:52 | * | xet7 quit (Quit: Leaving) |
| 12:01:44 | * | xet7 joined #nim |
| 12:04:38 | * | Trustable joined #nim |
| 12:12:52 | * | yglukhov joined #nim |
| 12:17:20 | * | yglukhov quit (Ping timeout: 255 seconds) |
| 12:19:59 | * | GustavoLapasta quit (Quit: Leaving) |
| 12:49:11 | * | Ven quit (Ping timeout: 240 seconds) |
| 12:55:50 | * | Ven joined #nim |
| 13:08:34 | * | stisa joined #nim |
| 13:08:57 | * | adeohluwa joined #nim |
| 13:09:04 | * | yglukhov joined #nim |
| 13:12:34 | * | nsf quit (Quit: WeeChat 1.7) |
| 13:22:38 | * | Kingsquee quit (Quit: https://i.imgur.com/qicT3GK.gif) |
| 13:30:22 | * | bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 13:40:10 | * | vlad1777d joined #nim |
| 13:52:01 | Araq | dom96: I would just copy how the D forum does it. didn't spend much time thinking about it though |
| 14:09:55 | * | adeohluwa quit (Ping timeout: 255 seconds) |
| 14:10:01 | * | adeohluwa joined #nim |
| 14:11:43 | * | adeohluwa quit (Remote host closed the connection) |
| 14:13:12 | * | PMunch joined #nim |
| 14:26:34 | * | chemist69 quit (Ping timeout: 255 seconds) |
| 14:26:57 | * | chemist69 joined #nim |
| 14:35:14 | * | Ven quit (Read error: Connection reset by peer) |
| 14:37:37 | * | Ven joined #nim |
| 14:39:59 | demi- | how do I represent opaque C types (typedef'd structs) in a nim wrapper? |
| 14:42:38 | Araq | type Foo* = ptr object |
| 14:44:27 | demi- | neat, i have to say this is an extremely good FFI, one of the better ones i've worked with |
| 14:46:03 | * | Ven quit (Read error: Connection reset by peer) |
| 14:46:24 | * | Ven joined #nim |
| 14:55:29 | * | gangstacat_ joined #nim |
| 14:55:36 | Araq | you need to checkout c2nim :-) |
| 14:56:15 | * | arnetheduck quit (Ping timeout: 240 seconds) |
| 14:56:38 | Araq | gangstacat: use 'nim devel' and try ui master |
| 14:56:48 | Araq | with --cc:vcc |
| 14:56:58 | Araq | libui onyl builds with vcc for me |
| 14:57:32 | * | gangstacat quit (Ping timeout: 255 seconds) |
| 14:58:03 | * | gangstacat_ is now known as gangstacat |
| 14:58:29 | gangstacat | Araq, I will try, thank you :) |
| 14:59:35 | Araq | gangstacat: it doesn't work yet because of some manifest issue, but I can guide you, perhaps. |
| 15:01:29 | gangstacat | Arak, I'm still installing some components of Visual Studio 2017 RC because I haven't the lib folder in the Windows Kits folder so.. |
| 15:02:12 | gangstacat | Araq* |
| 15:06:06 | demi- | Araq: I have already made my own homebrew formula for it and using it now :) |
| 15:07:08 | Araq | alright. what did you end up creating for your 'cmd' package? |
| 15:13:33 | demi- | https://github.com/samdmarshall/cmd.nim/blob/master/src/cmd.nim it got approved and is in the packages repo no |
| 15:13:35 | demi- | now* |
| 15:19:57 | * | yglukhov quit (Remote host closed the connection) |
| 15:22:47 | * | yglukhov joined #nim |
| 15:27:11 | * | yglukhov quit (Ping timeout: 240 seconds) |
| 15:30:03 | demi- | what are the rules around submitting wrappers of C libraries to the package index? Seems like unless you package the source code of the library with it then it is apt to be incompatible in minor ways (header path) -- but if you ship the C code with the package then you are responsible for maintaining it and making sure it is up to date. doesn't seem very ideal for library maintainers. |
| 15:31:26 | federico3 | demi-: usually C files or headers are not shipped |
| 15:31:50 | federico3 | demi-: are you talking dynamic linking? |
| 15:33:06 | cheatfate | demi-, usually we make wrappers which loads wrapped library dynamically |
| 15:33:17 | demi- | mmm, also the location of the headers. like I just made a wrapper for libmagic, i could contribute it to the packages repo for everyone else but it is based on the location of my "magic.h" being in `/usr/local/include` and the library (be it static or dynamic) in `/usr/local/lib` |
| 15:33:58 | demi- | which might not always be the case for everyone -- wondering how to handle that gracefully |
| 15:33:58 | Araq | demi-: use #dynlib, not #header |
| 15:34:05 | demi- | ah |
| 15:34:12 | Araq | this way it also works with the LLVM backend |
| 15:35:28 | demi- | neat, thanks for the tip |
| 15:36:56 | * | yglukhov joined #nim |
| 15:36:57 | * | yglukhov quit (Remote host closed the connection) |
| 15:37:01 | dom96 | it's nice to support static linking too though, this is done via --dynlibOverride usually |
| 15:37:22 | dom96 | Calinou: hey, I merged https://github.com/nim-lang/website/pull/11, hope you don't mind. |
| 15:37:34 | * | yglukhov joined #nim |
| 15:40:09 | Calinou | hi |
| 15:41:41 | * | yglukhov quit (Ping timeout: 240 seconds) |
| 15:41:43 | demi- | dom96: i've got some bad news, it seems like i forgot half of my patch that you already merged locally without pushing it up >.< |
| 15:41:54 | Calinou | looks good, will need some fixes eg. for Sponsors page |
| 15:42:07 | dom96 | Calinou: sure :) |
| 15:42:08 | * | Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 15:42:11 | dom96 | demi-: no worries |
| 15:42:28 | demi- | ok, then i'll just make a quick follow-up one now |
| 15:50:28 | * | couven92 joined #nim |
| 15:59:10 | gangstacat | Araq, alright, 1- I have only Windows 10 Windows Kit with a Lib folder so I used it (changed everything in libui.raw) 2- nim asked for vccexe.exe so I directly compiled vccexe.nim and moved the binary in the bin Nim devel folder 3- I launch the Nim compilation of my app from VS 2017 RC dev command prompt which knows where to find cl.exe 4- I got error C2118: nimbase.h(443): negative index 5- vccexe.exe complains because it doesn't find any *.obj |
| 15:59:10 | gangstacat | file in the nimcache (there are only *.c files) |
| 15:59:37 | gangstacat | rawui.nim in fact |
| 16:03:19 | Araq | gangstacat: vccexe's job is that you don't have to launch from the dev command prompt |
| 16:07:44 | * | yglukhov joined #nim |
| 16:08:43 | * | ARCADIVS quit (Quit: ARCADIVS) |
| 16:12:28 | * | yglukhov quit (Remote host closed the connection) |
| 16:27:21 | * | Trustable quit (Remote host closed the connection) |
| 16:36:06 | * | stisa quit (Ping timeout: 240 seconds) |
| 16:40:23 | * | stisa joined #nim |
| 16:47:01 | * | yglukhov joined #nim |
| 16:50:40 | * | rokups joined #nim |
| 16:51:15 | * | yglukhov quit (Ping timeout: 240 seconds) |
| 16:53:02 | * | vlad1777d_ joined #nim |
| 16:54:24 | * | vlad1777d quit (Ping timeout: 252 seconds) |
| 17:12:23 | dom96 | that's weird, www.nim-lang.org points to a different IP than nim-lang.org |
| 17:13:07 | * | stisa quit (Ping timeout: 256 seconds) |
| 17:15:12 | * | stisa joined #nim |
| 17:22:30 | gangstacat | Araq, for ui_ui.obj I have a ton of errors LNK2019 unresolved external symbols, that seems very complicated |
| 17:34:36 | * | libman quit (Ping timeout: 240 seconds) |
| 17:37:39 | * | yglukhov joined #nim |
| 17:40:34 | * | stisa2 joined #nim |
| 17:41:30 | * | stisa2 quit (Client Quit) |
| 17:41:34 | Araq | gangstacat: yeah I'll likely fork libui for this to work reliably |
| 17:41:51 | gangstacat | Araq, you encouter the same problem? |
| 17:42:17 | Araq | no, because I patched my libui |
| 17:42:36 | gangstacat | can you push your fork? |
| 17:42:49 | * | stisa quit (Ping timeout: 240 seconds) |
| 17:43:33 | * | yglukhov quit (Remote host closed the connection) |
| 17:43:53 | demi- | is there a way to define the `-d:ssl` in the nim code file through a pragma? |
| 17:44:50 | * | yglukhov joined #nim |
| 17:47:40 | * | yglukhov quit (Remote host closed the connection) |
| 17:48:44 | dom96 | I'm moving nim-lang.org over to cloudflare so there may be some disruption |
| 17:58:20 | * | Matthias247 quit (Read error: Connection reset by peer) |
| 18:11:37 | * | Jesin quit (Ping timeout: 260 seconds) |
| 18:17:27 | * | PMunch quit (Quit: leaving) |
| 18:31:59 | * | libman joined #nim |
| 18:34:40 | * | Ven joined #nim |
| 18:35:51 | * | kulelu88 joined #nim |
| 18:56:15 | * | Snircle is now known as mr_teak |
| 18:56:55 | * | Jesin joined #nim |
| 19:12:29 | Araq | gangstacat: https://github.com/Araq/libui |
| 19:13:12 | Araq | demi-: no that needs to be in your config.nims or project.nims or project.nim.cfg .... |
| 19:13:34 | Araq | (Nim's config system is a bit overwhelming ... :-/ ) |
| 19:16:04 | demi- | ah i see |
| 19:20:07 | * | GustavoLapasta joined #nim |
| 19:22:51 | gangstacat | Araq, traceback ui.nim(14) init - error: unhandled exception: error initializing libui: initializing Common Controls: code 0 (0x00000000) operation successfull [ValueError] |
| 19:23:23 | Araq | yes, that's what I'm getting too |
| 19:23:34 | Araq | cheatfate said it's due to a missing resource file |
| 19:23:59 | cheatfate | its due to missing manifest, but it must be embedded as resource |
| 19:29:03 | Araq | ah and my compiler has a new bug |
| 19:29:06 | Araq | :-) |
| 19:29:18 | Araq | very easy to fix though |
| 19:32:34 | * | nsf joined #nim |
| 19:33:29 | * | mr_teak is now known as sliceofcheese |
| 19:33:58 | * | sliceofcheese is now known as Snircle |
| 19:38:11 | * | libman quit (Ping timeout: 240 seconds) |
| 19:39:55 | Araq | cheatfate: was the missing resource file :-) now it works |
| 19:41:01 | dom96 | why is this manifest file not necessary for GTK apps? |
| 19:41:18 | Araq | because GKT doesn't use extneded COM controls |
| 19:41:23 | Araq | *extended |
| 19:41:26 | Araq | *GTK |
| 19:41:37 | Araq | libui uses the native controls, GTK doesn't |
| 19:42:23 | Araq | ok, so new Windows "installer" is almost ready |
| 19:45:02 | dom96 | Will you get the UI library to add the manifest automatically? |
| 19:45:05 | dom96 | That would be nice :) |
| 19:45:28 | Araq | I just did |
| 19:46:57 | dom96 | https://forum.nim-lang.org/ tada |
| 19:47:17 | dom96 | Still running off the old VPS though |
| 19:48:47 | Araq | what does that mean? is it https or not? |
| 19:49:24 | Araq | https://nim-lang.org/ ah points to the http forum |
| 19:50:18 | dom96 | yes, it does. |
| 19:50:32 | dom96 | But you can still access it via http |
| 19:50:55 | dom96 | Once everything is set up I will set up a redirect from http to https |
| 19:51:22 | BlaXpirit | dom96, why not fix the links. just remove "http:" everywhere |
| 19:52:06 | dom96 | Because search engines |
| 19:52:35 | BlaXpirit | eh not sure if it affects them but maybe valid point |
| 19:53:43 | * | vlad1777d__ joined #nim |
| 19:54:02 | dom96 | There is hundreds of websites out there that link to http://nim-lang.org |
| 19:54:16 | Araq | hmm does https imply I can remove the hash signatures on our download page? |
| 19:54:22 | dom96 | no |
| 19:54:26 | Araq | bah |
| 19:54:36 | * | vlad1777d_ quit (Ping timeout: 240 seconds) |
| 19:55:06 | * | yglukhov joined #nim |
| 19:56:45 | BlaXpirit | dom96, redirecting is fine but i mean why have http:// links on the website itself just to redirect it to https |
| 19:57:04 | dom96 | They will be fixed of course |
| 19:57:16 | BlaXpirit | ok |
| 19:57:22 | dom96 | What gave you the impression they wouldn't be? |
| 19:57:23 | * | libman joined #nim |
| 19:58:19 | BlaXpirit | well when this was mentioned you replied that you would setup redirects from http to https. can be understood as being the solution for having http links |
| 19:59:34 | dom96 | yes, because it is. But not links on nim-lang.org :) |
| 20:00:00 | * | yglukhov quit (Ping timeout: 276 seconds) |
| 20:14:05 | gangstacat | the Quit button (with the closure) hangs here |
| 20:14:50 | Araq | gangstacat: what program are you testing? |
| 20:15:49 | gangstacat | Araq, the basic controlgallery with the new "ui" module |
| 20:16:07 | gangstacat | File > Quit hangs for say 5 seconds then quit |
| 20:16:31 | Araq | fun |
| 20:16:35 | * | libman quit (Ping timeout: 255 seconds) |
| 20:16:55 | gangstacat | (and on Linux there is a nice error message) |
| 20:17:10 | gangstacat | CRITICAL **: [libui] /builddir/libui-master/unix/menu.c:357:uninitMenus() POSSIBLE IMPLEMENTATION BUG; CONTACT ANDLABS: |
| 20:17:11 | gangstacat | menu item 0x19458a0 ((null)) still has uiWindows attached; did you forget to destroy some windows? |
| 20:17:11 | gangstacat | Trace/breakpoint trap |
| 20:22:04 | * | libman joined #nim |
| 20:22:26 | cheatfate | Araq, heh i told you, you are trying to make zombie from corpse :) |
| 20:24:11 | Araq | already forked it, we can adapt it to our needs |
| 20:24:37 | cheatfate | we need to rewrite it :) |
| 20:24:56 | Araq | we lack the know-how. |
| 20:29:09 | cheatfate | ? |
| 20:30:14 | * | GustavoLapasta quit (Quit: Leaving) |
| 20:31:06 | * | libman quit (Ping timeout: 240 seconds) |
| 20:31:07 | Araq | cheatfate: well at least I cannot write a UI library that abstracts over Cocoa, WinAPI and GTK in a reasonable timeframe |
| 20:31:47 | * | Trustable joined #nim |
| 20:32:26 | cheatfate | Araq, you will not be alone in this endeavor, i will help you with winapi and gtk :) |
| 20:33:02 | Araq | too bad then that I don't know Cocoa |
| 20:33:27 | Araq | I am also not interested *at all* to learn it |
| 20:34:03 | kulelu88 | start with the low-hanging fruit. GTK only first |
| 20:34:14 | kulelu88 | or QT if you're feeling brave :D |
| 20:34:54 | Araq | watched a colleague once trying to draw a button onto form. was an absurd masochistic endeavor |
| 20:35:37 | Araq | I thought "Delphi did that better in the 90ies" |
| 20:35:49 | cheatfate | kulelu88, i even dont want to make gtk :) |
| 20:36:28 | cheatfate | Delphi and PHP as the most worst technologies ever happened it generates very bad coders/ui designers |
| 20:36:39 | * | libman joined #nim |
| 20:36:40 | kulelu88 | wouldn't QT be a better option, as it is cross-platform? |
| 20:37:01 | dom96 | Why not just use QT then? |
| 20:37:03 | demi- | a UI library over cocoa seems like it would be challenging at best and a complete torture to maintain longterm |
| 20:38:22 | cheatfate | demi-, apple changes it all the time? |
| 20:38:58 | kulelu88 | what is currently the most supported & easiest method to develop GUI apps on the desktop? |
| 20:39:18 | flyx | cheatfate: quite the contrary, I think. the basic API has changed very little since 10.0 |
| 20:39:21 | cheatfate | kulelu88, Electron, libCEF |
| 20:39:24 | demi- | not so much that they change it, but it has bin-compat requirements -- i think it would time better spent making a good FFI to cocoa's primary libraries instead. |
| 20:39:40 | demi- | instead of trying to map over the massive bulk that is cocoa |
| 20:39:46 | kulelu88 | does electron count? |
| 20:40:05 | cheatfate | kulelu88, photoshop uses electron :) |
| 20:40:08 | cheatfate | so it count |
| 20:40:22 | flyx | if you don't need a „native“ UI, Electron counts |
| 20:41:01 | flyx | I use VS Code myself and its usefulness outweights the flaw of not being native |
| 20:41:26 | demi- | electron is a bad option imo, would rather use a good curses based UI library than that |
| 20:41:34 | dom96 | Does Photoshop really use Electron? |
| 20:42:37 | cheatfate | Photoshop CC, Adobe Illustrator CC |
| 20:42:40 | flyx | the problem with Elentron is that for Nim, we would need to compile to JS to use it. |
| 20:42:50 | cheatfate | flyx, libCEF |
| 20:43:29 | cheatfate | but anyway 100mb for hello world application is too bad |
| 20:43:41 | kulelu88 | Wouldn't the QT API (and the fact that it is C++) be the most approachable? |
| 20:43:44 | flyx | doesn't libCEF just embed a HTML renderer into an application and thus still needs JS |
| 20:43:58 | dom96 | cheatfate: source? |
| 20:44:08 | demi- | election is good... if you like shipping ffmpeg in whatever you make |
| 20:44:09 | flyx | the QT API is horrible. at least it was when I last had a look at it |
| 20:44:32 | kulelu88 | demi-: what is election? |
| 20:44:37 | demi- | er |
| 20:44:38 | Araq | the QT API looked superb on a superficial look |
| 20:44:39 | demi- | electron |
| 20:44:44 | kulelu88 | 0.o |
| 20:45:20 | Araq | kulelu88: nothing wrong with Qt but it's not really native and libui proved very easy to wrap |
| 20:45:23 | dom96 | Electron is designed to be used with JS. You could probably write something similar for Nim much easier than you could something like wxwidgets or libui. |
| 20:45:25 | Araq | and is rather slim. |
| 20:45:32 | kulelu88 | How many people actually need support for such an endeavour? perhaps Araq would rather spend his time on the more important compiler |
| 20:45:53 | demi- | electron ships a version of chromium to do the rendering for the ui run by the nodejs server or w/e; chromium ships with ffmpeg in it, you can check atom or vs code, they both have the ffmpeg library in them |
| 20:46:45 | flyx | QT has this signal/slot concept which is basically implemented as DSL on top of C++ and I would imagine it is rather hard to wrap. |
| 20:47:45 | demi- | iirc, Qt also has a slew of bugs on all platforms, but especially on os x. if you are looking for a "write once, use everywhere" that doesn't exist. |
| 20:48:18 | cheatfate | dom96, https://forums.adobe.com/thread/2172201 |
| 20:48:48 | cheatfate | ok, then how sublime text2/3 is cross platform and works perfectly :) |
| 20:48:56 | cheatfate | without GTK/QT and all other shit |
| 20:49:27 | flyx | Sublime afaik has its own GUI stuff. just look at it. it doesn't look native at all anywhere |
| 20:49:37 | dom96 | cheatfate: As far as I can see this refers to an Adobe plugin and mentions libcef (doesn't Electron use something else?) |
| 20:49:43 | demi- | heh |
| 20:49:58 | Araq | guys, this heavily depends on the application. |
| 20:50:33 | Araq | for a simple Windows installer UI, a Windows specific library would suffice and the size matters |
| 20:50:36 | dom96 | Sublime's GUI works in a similar way to nimx AFAIK |
| 20:50:45 | cheatfate | dom96, libCEF is almost equal to Electron, two ways of implementation but uses same chrome core |
| 20:51:00 | Araq | for most other apps portability is more important than "native" look and feel. |
| 20:51:30 | Araq | and here "portability" means Windows and OSX because commercial software doesn't care about Linux. ymmv. |
| 20:51:36 | cheatfate | flyx, Sublime Text has native menu, native dialogs, native font rendering |
| 20:51:56 | cheatfate | and much faster then gtk application on windows |
| 20:52:06 | Araq | most apps are websites today. they are alien on every system. |
| 20:52:25 | dom96 | cheatfate: yes, but libCEF != Electron |
| 20:52:37 | dom96 | Steam and Spotify also use libCEF. |
| 20:52:51 | cheatfate | dom96, libCEF is 90% equal to Electron |
| 20:52:53 | flyx | cheatfate: calling the native dialogs is much less work than using the native toolkit in your own window |
| 20:53:07 | cheatfate | flyx, sublime text uses native windows too |
| 20:53:31 | flyx | what's „native windows“? |
| 20:53:32 | Araq | sublime text is indeed native, there is a developer blog about how he does it |
| 20:54:17 | flyx | a single look at Sublime's main window makes it obvious that the native toolkit is not used. for example, the tabs |
| 20:54:30 | flyx | of course it uses some native drawing library |
| 20:54:31 | cheatfate | Araq, i saw some reverse engineering blog about sublimetext, but not developer itself |
| 20:55:12 | demi- | i'm pretty sure that anyone building commercial software on os x (with a dedicated team) is going to use native APIs, which seem to already work extremely well via the ffi interface that nim has. dunno if the same holds true for windows, but you cannot get market-share reach and quality at the same time, I don't think it matters about what the UI is implemented in, electron or another html based rendering option |
| 20:55:14 | demi- | might be the best compromise these days tbh |
| 20:55:40 | dom96 | cheatfate: When you said "photoshop uses electron" I assumed you meant that the main application is using it. |
| 20:58:14 | flyx | anyway, I am not planning to code UI stuff for the time being, so it does not concern me all that much |
| 21:05:13 | cheatfate | demi-, sublime text has same professional look on all systems |
| 21:05:31 | demi- | i disagree, but i've been using OS X for too long :) |
| 21:06:38 | cheatfate | of course it can't be so good on linux because of freetype, but if i install proprietary windows fonts to linux box, everything becomes equal |
| 21:07:34 | demi- | not sharing the same default keybindings on OS X is another thing it falls short on |
| 21:09:53 | cheatfate | i dont have mac, so maybe you are right, but anyway its much easier fix keybindings then font rendering on QHD desktops |
| 21:11:30 | demi- | but back to my point, i think extremely good hooks into native platform interfaces would make it way more portable than trying to craft generic platform agnostic UI APIs |
| 21:11:47 | * | Jesin quit (Quit: Leaving) |
| 21:12:06 | * | rokups quit (Quit: Connection closed for inactivity) |
| 21:12:59 | demi- | unless you are talking about a terminal-based UI, in which case that would be able to be platform agnostic for the most part. the underlying problem is with design pattern of the system and interaction. every system has their own (conflicting) architecture patterns for their UI APIs. |
| 21:14:14 | cheatfate | i dont know what you mean `agnostic UI` because you always need hooks at least on event loop |
| 21:14:37 | cheatfate | and also keybindings is always very big for every platform |
| 21:15:56 | demi- | I have never seen a good cross-platform ui implmenentation or API, that is my opinion on this topic; :) |
| 21:19:11 | cheatfate | +1 |
| 21:22:39 | kulelu88 | demi-: perhaps you should write your own then? |
| 21:25:33 | dom96 | I would say that the web is a good cross-platform UI API |
| 21:26:00 | dom96 | By no means perfect, but it sure is flexible and extremely portable. |
| 21:34:35 | * | bjz joined #nim |
| 21:49:37 | * | Jesin joined #nim |
| 21:54:19 | * | gokr quit (Quit: Leaving.) |
| 21:54:26 | demi- | kulelu88: lol, with what time?? |
| 21:54:53 | kulelu88 | I find it strange when people complain about not having the perfect desktop API but then aren't willing to build their own |
| 21:55:14 | demi- | well I have one that suites me well enough? |
| 21:56:55 | demi- | and i am not one to reinvent the wheel if it already exists, |
| 21:57:33 | kulelu88 | ok |
| 22:00:56 | demi- | oh, cheatfate, something i forgot to mention about the CFRunLoop API on OS X, it is reentrant. callbacks placed on it can be called more than once in the same cycle of the loop. |
| 22:01:07 | * | chemist69 quit (Ping timeout: 245 seconds) |
| 22:05:42 | * | libman quit (Ping timeout: 252 seconds) |
| 22:11:19 | * | libman joined #nim |
| 22:14:35 | gangstacat | Araq, are you interested by a libui bug with a closure on a button (visibly) which hang an app on Windows but not on Linux? or do you give up and try a custom solution in the future? |
| 22:15:56 | Araq | what? no. please fix it. |
| 22:16:20 | Araq | it's close to being stable and useful. |
| 22:16:25 | Araq | (I hope) |
| 22:16:41 | gangstacat | I can try to make a minimal code in order to reproduce it |
| 22:17:01 | Araq | no, fix it. |
| 22:20:13 | gangstacat | I'm not good enough in Nim for that |
| 22:24:42 | * | synshroud quit (Quit: ZNC 1.6.4 - http://znc.in) |
| 22:24:46 | * | gokr joined #nim |
| 22:25:16 | libman | gangstacat: How much Bitcoin are you willing to pay a Nim freelancer who is competent with this and available (not me)? |
| 22:26:36 | gangstacat | libman, I don't think every people who open an issue in Github are willing to pay bitcoins |
| 22:26:46 | * | synshroud joined #nim |
| 22:26:50 | gangstacat | I didn't said I want that to be fixed in the minute |
| 22:27:01 | * | Jesin quit (Quit: Leaving) |
| 22:27:30 | gangstacat | just that I felt the libui solution was not the best so I asked if there is still a future for it and if yes, then I may open an issue and that's all |
| 22:28:18 | gangstacat | (due to the discussion above) |
| 22:28:19 | * | chemist69 joined #nim |
| 22:29:06 | gangstacat | if I could fix it I would but if I can't then that would be bad to don't report something if there is a confirmed bug |
| 22:30:21 | * | Jesin joined #nim |
| 22:33:30 | Araq | gangstacat: controlgallery2 fails for me with the same error |
| 22:33:53 | libman | I am a big fan of people doing things for free too. But sometimes there's a gap, and paying bounties is the solution. |
| 22:34:06 | gangstacat | libman, I don't especially need it |
| 22:34:28 | gangstacat | I just play with the library |
| 22:40:18 | Araq | ugh, devel doesn't have version 0.16.1 :-( |
| 22:43:48 | * | Matthias247 joined #nim |
| 22:56:59 | * | Vladar quit (Quit: Leaving) |
| 23:04:17 | * | zhalla joined #nim |
| 23:04:33 | * | zhalla left #nim (#nim) |
| 23:05:47 | Araq | gangstacat: fixed |
| 23:06:52 | gangstacat | Araq, the one from the quit closure? |
| 23:07:56 | Araq | yes, check it out |
| 23:08:05 | Araq | windows and linux versions actually had the same bug |
| 23:08:17 | gangstacat | oh, nice |
| 23:11:39 | Araq | libui is very picky about its Window destructions |
| 23:11:51 | Araq | even though the rules for that are entirely mysterious. |
| 23:13:44 | cheatfate | demi-, how do you think? how long it takes to get your nim color syntax to micro nightly builds? because i can't use source distro on freebsd10 (it uses 1.4go) |
| 23:17:22 | * | nsf quit (Quit: WeeChat 1.7) |
| 23:24:21 | * | bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) |
| 23:29:20 | * | perturbation joined #nim |
| 23:32:15 | perturbation | Just wanted to say that I just used c2nim for the first time in a while- and it's gotten *really* good :) |
| 23:32:41 | perturbation | going to wrap the snowballstemmer library as part of a weekend project |
| 23:32:54 | * | vlad1777d__ quit (Quit: Leaving) |
| 23:35:27 | Araq | perturbation: thank you. I think we might need to change its name to cpp2nim though |
| 23:35:36 | Araq | few know it can parse C++ |
| 23:35:48 | Araq | and really MUCH C++, that is |
| 23:35:52 | perturbation | :) |
| 23:37:09 | Araq | was also interesting to figure out that parsing C++ is really not much harder than C, IMO. C is already fucked up, C++ doesn't add much on top. |
| 23:44:07 | * | flyx quit (Excess Flood) |
| 23:44:17 | * | flyx joined #nim |
| 23:44:18 | libman | Am I going blind, or did Araq switch from 2 spaces to 3?? http://forum.nim-lang.org/t/2765 |
| 23:45:12 | Araq | libman: I'm not consistent when typing snippets in a browser editor |
| 23:45:27 | Araq | but no, real code is still 2 spaces. |
| 23:46:38 | libman | You're the one who, after 20 years, switched me from tabs (rendered as 4 spaces) to 2 spaces. :P |
| 23:48:02 | Araq | good. |
| 23:48:45 | * | brson joined #nim |
| 23:50:45 | * | libman quit (Ping timeout: 240 seconds) |
| 23:53:05 | * | brson quit (Client Quit) |
| 23:55:43 | * | libman joined #nim |
| 23:58:06 | * | gokr quit (Quit: Leaving.) |
| 23:59:35 | * | Trustable quit (Remote host closed the connection) |