<< 20-12-2016 >>
00:17:42*yglukhov joined #nim
00:21:59*yglukhov quit (Ping timeout: 252 seconds)
00:25:45*gokr quit (Ping timeout: 260 seconds)
00:59:12vlad1777dIs redefinition totally disallowed ? https://pp.vk.me/c638428/v638428447/151f3/itDIp4pET9g.jpg
01:01:48*abeaumont quit (Ping timeout: 268 seconds)
01:01:56federico3vlad1777d: that's odd
01:03:28vlad1777dfederico3, I learn Nim, I don't understand some things yet )
01:12:34ftsf_vlad1777d, you're defining the same thing on the child object, you only need to define it on the parent object
01:12:45ftsf_you should give it another name on the child object if it should be different
01:13:04vlad1777dftsf_, thank you
01:15:13*abeaumont joined #nim
02:13:21*chemist69 quit (Ping timeout: 251 seconds)
02:19:02*vlad1777d quit (Quit: Leaving)
02:20:04*yglukhov joined #nim
02:22:29*space-wizard joined #nim
02:24:27*yglukhov quit (Ping timeout: 252 seconds)
02:27:04*chemist69 joined #nim
02:47:33*space-wizard quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
02:48:39*space-wizard joined #nim
02:49:06*space-wizard quit (Client Quit)
02:49:27*space-wizard joined #nim
02:49:54*space-wizard quit (Client Quit)
02:50:15*space-wizard joined #nim
02:50:42*space-wizard quit (Client Quit)
02:51:03*space-wizard joined #nim
02:51:30*space-wizard quit (Client Quit)
02:51:51*space-wizard joined #nim
02:52:18*space-wizard quit (Client Quit)
02:52:38*space-wizard joined #nim
02:53:06*space-wizard quit (Client Quit)
02:53:27*space-wizard joined #nim
02:53:54*space-wizard quit (Client Quit)
02:54:15*space-wizard joined #nim
02:54:42*space-wizard quit (Client Quit)
03:28:06*brson quit (Quit: leaving)
04:14:05*chemist69 quit (Ping timeout: 246 seconds)
04:18:47*chemist69 joined #nim
04:21:34*yglukhov joined #nim
04:26:18*yglukhov quit (Ping timeout: 256 seconds)
05:32:01*space-wizard joined #nim
06:17:58*gokr joined #nim
06:18:30*trinarybee quit (Quit: Leaving)
06:18:45*bjz joined #nim
06:23:31*yglukhov joined #nim
06:26:46*bjz quit (Ping timeout: 250 seconds)
06:28:04*yglukhov quit (Ping timeout: 250 seconds)
06:28:40*bjz joined #nim
06:30:22*bjz quit (Read error: Connection reset by peer)
06:30:25*bjz_ joined #nim
06:32:44*bjz_ quit (Read error: Connection reset by peer)
06:40:17*nsf joined #nim
06:46:39*subsetpark quit (Quit: Connection closed for inactivity)
06:46:40*bjz joined #nim
06:50:38*Snircle quit (Read error: Connection reset by peer)
06:51:44*Snircle joined #nim
07:04:22*rokups joined #nim
07:43:15*yglukhov joined #nim
07:47:32*yglukhov quit (Ping timeout: 250 seconds)
08:00:40*Arrrr joined #nim
08:09:45*space-wizard quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
08:10:05*space-wizard joined #nim
08:10:30*space-wizard quit (Client Quit)
08:10:53*space-wizard joined #nim
08:11:18*space-wizard quit (Client Quit)
08:11:40*space-wizard joined #nim
08:12:06*space-wizard quit (Client Quit)
08:12:28*space-wizard joined #nim
08:12:54*space-wizard quit (Client Quit)
08:13:16*space-wizard joined #nim
08:13:42*space-wizard quit (Client Quit)
08:14:04*space-wizard joined #nim
08:14:11*byte512 joined #nim
08:14:30*space-wizard quit (Client Quit)
08:14:51*space-wizard joined #nim
08:15:18*space-wizard quit (Client Quit)
08:15:39*space-wizard joined #nim
08:16:06*space-wizard quit (Client Quit)
08:16:27*space-wizard joined #nim
08:16:54*space-wizard quit (Client Quit)
08:17:15*space-wizard joined #nim
08:17:42*space-wizard quit (Client Quit)
08:18:03*space-wizard joined #nim
08:18:30*space-wizard quit (Client Quit)
08:19:01*space-wizard joined #nim
08:19:18*space-wizard quit (Client Quit)
08:32:48*yglukhov_ joined #nim
08:34:13*yglukhov_ quit (Remote host closed the connection)
08:34:13*yglukho__ joined #nim
08:34:15*yglukho__ quit (Remote host closed the connection)
08:34:49*yglukhov joined #nim
08:36:13*Andris_zbx joined #nim
08:40:56*pregressive quit (Remote host closed the connection)
08:41:31*pregressive joined #nim
08:46:00*pregressive quit (Ping timeout: 265 seconds)
09:00:37*yglukhov_ joined #nim
09:04:39*yglukhov quit (Ping timeout: 268 seconds)
09:04:58euantorMarket Nim as a better Crystal ;)
09:05:52euantorIIRC, it still doesn't support parallelisation, just concurrency (eg: it only ever uses one core)
09:06:19euantorAnd it still doesn't have Windows support, which is a deal breaker for many
09:16:38*yglukhov_ quit (Remote host closed the connection)
09:31:05ArrrrI think people using Crystal enjoy the freedom it provides. Nim is more coarsed.
09:37:08FromGitter<andreaferretti> IMO, if anything needs to be done on the marketing side, is just to give the site (and forum) a more modern look
09:37:57FromGitter<andreaferretti> and simplify the message (just look at the few bullet points describing the aims of Crystal vs. the long text + many bullet points describing the aims of Nim)
09:39:08FromGitter<andreaferretti> It may seem silly, but you would not believe how many people discard the language based on the first appearance of the site
09:39:23*[ui] joined #nim
09:43:27ArrrrI think the situation of nim is at stake. It might need some changes, but those changes won't happen without anyone helping, which at the same time won't happen without help. Like a Game of Life after several turns, cells are unable to move further.
09:45:11ArrrrRight now development is slow, i wish i could help, but im not useful, except to signal certain bugs.
09:46:27Araqdevelopment is slow because I keep getting paid for unforeseen developments, most recent example: deterministic C code generation.
09:47:45FromGitter<andreaferretti> @Araq certainly nobody is complaining about your work :-)
09:48:22FromGitter<andreaferretti> but still it would be nice if more people were able to help
09:48:41FromGitter<andreaferretti> I have been trying to give a shot at solving some type-related bugs
09:49:06FromGitter<andreaferretti> but it is just very slow at first to be able to do something in the compiler
09:49:09Araqimproves compile times for incremental developments tremendously but for most people compile times are already fast
09:49:49FromGitter<andreaferretti> still, I think that mostly what has to do is just a little better marketing and stdlib cleanup
09:50:19ArrrrMost projects are small, they won't bring nim eternal glory.
09:50:44Araq@andrea I can guide you further
09:51:09FromGitter<andreaferretti> thank you, I will try to give it a shot in the Xmas holidays :-)
09:51:17Araqsigmatch.nim line 100
09:51:26Araqevery type binding op goes through 'proc put'
09:51:31FromGitter<andreaferretti> wow, that is really precise :-D
09:51:57FromGitter<andreaferretti> thank you!
09:52:03FromGitter<andreaferretti> I do not have time to look just now
09:52:09FromGitter<andreaferretti> but I will asap
09:52:22Araqdude wait, I've not even started
09:52:35FromGitter<andreaferretti> :-D
09:53:06Araqadd something like
09:53:11Araq if c.callee != nil and (c.callee ?? "temp.nim"):
09:53:13Araq writeStackTrace()
09:53:14Araq echo "binding ", typeToString(val), " to ", typeToString(key)
09:53:31Araqassuming the test file is "temp.nim" (the name I use for everything)
09:54:18Araqtest via 'koch temp c temp.nim'
09:54:37Araqoh it's c.callee.info
09:54:44*vktec left #nim ("So long and thanks for all the fish")
09:55:50*djellemah quit (Ping timeout: 268 seconds)
09:56:03Araqtype related bugs the hardest though :-/
09:56:50FromGitter<andreaferretti> thanks again!
09:57:35*djellemah joined #nim
10:12:30Araqhmm can I summon Stefan Salewski this way?
10:12:35*byte512 quit (Quit: WeeChat 1.6)
10:15:25*bjz quit (Ping timeout: 265 seconds)
10:15:26*bjz_ joined #nim
10:18:04*byte512 joined #nim
10:20:53FromGitter<timeyyy> FWIW i like the website. I found this an interesting read (http://marktarver.com/open.html), about finding your own definition of success. No point to compare and contrast popularity unless that is what you deem to be the most important thing...
10:53:44Araqtimeyyy a bit more popularity is always nice though :-)
10:57:37FromGitter<timeyyy> yes but don't let yourself get to down about it, otherwise we would all be using windows...
10:57:52FromGitter<timeyyy> or java etc
11:05:13*Salewski joined #nim
11:05:43SalewskiAraq, have I done something bad?
11:06:21Araqwell I'm trying to make your nimchess compile on my ubuntu
11:07:36SalewskiShould be no problem on Linux. But OSX may be difficult.
11:07:50Araqwell the first problem is here:
11:08:10Araqhttp://forum.nim-lang.org/t/2198
11:08:21AraqI go through the instructions, build fails
11:08:33AraqI continue reading and find updated instructiongs, build fails
11:08:42AraqI continue and find instructions here:
11:08:51Araqhttps://github.com/ngtk3/nim-chess2
11:08:58Araqthat actually work :-)
11:09:15Araqbut PLEASE create nimble packages for this
11:09:31AraqI expect 'nimble install nimchess' to work out of the box these days
11:09:42Araq$nimble init
11:09:46Araq$nimble publish
11:09:58Araqit doesn't take dom96's book for these two commands
11:10:28SalewskiAraq, problem is nimble package for GTK3, as long as GTK2 is still available.
11:10:36*irrequietus joined #nim
11:10:45Araqwhy? one is named GTK2, the other GTK3
11:10:46*irrequietus quit (Changing host)
11:10:46*irrequietus joined #nim
11:10:48SalewskiI have no idea how to avoid conflicts.
11:10:51Araqshouldn't produce conflicts
11:11:27SalewskiIs is not, GTK3 is 10 packages, pango, cairo, glib2, much more.
11:11:51FromGitter<andreaferretti> (even if name was the same, one can ask for a given version of a dependency in .nimble file)
11:12:40SalewskiDom did not gave me a hint to solve it brst. Of course I can give all a salewski prefix, but that is ugly.
11:12:59Araqgive them all a '3' suffix for now
11:13:16SalewskiBut I want to use "import cairo", not import salewski-cairo.
11:13:21Araqwill push dom96 about vendor namespacing support
11:13:28Araq"import cairo3" ?
11:15:01SalewskiYes. a 3 suffix is what I considerest the best indeed. but it is confuling, because glib is still glib2, so it will become glib2_3. And Cairo will be cairo_2, while there is no real cairo2.
11:15:16Araqas a first step can you update your forum posts that contain outdated instructions?
11:15:19SalewskiCairo3.
11:15:22euantorI seem to remember a forum topic about this kind of issue that I started a while back
11:16:14AraqSalewski: how do I compile your editor?
11:16:58SalewskiYes can do. I was not too much intersted in doing it, seems most people hate GTK3.
11:17:05*yglukhov joined #nim
11:17:45Araqwell it's a self fullfillng prophecy right now
11:18:17Araqyou invested time in it and it's good work afaict
11:18:18SalewskiEditor NEd has asciidoc document manual with instructions. But you have to install files for syntax highlight also.
11:18:41Araqbut then you make it hard to install and discourage people from using it since "GTK3 is moribund"
11:19:07AraqI don't mind it being Linux specific
11:19:42Araqbut now I already have a 'ntgk3' directory and your instructions for NEd start with 'mkdir ngtk3'
11:19:59Araqwhich means I need to think and compare what to do
11:20:48SalewskiI can not really advertise GTK3 in these days unfortunately. Community is not really active, and it is hard to learn coding in GTK3 for newcommers.
11:21:37*yglukhov quit (Ping timeout: 252 seconds)
11:23:22SalewskiAraq -- we Nim users are generally not afraid of some "thinking". Real "Deppen" will not be happy with Nim and GTK3 currently.
11:23:48Araqno, you're wrong about that.
11:24:12SalewskiBut OK, when there is some minimal interest in Nim-Chess and NEd editor I will improve instructions.
11:24:36Araqwell right now I installed it to test my new compiler against it
11:25:24SalewskiGreat, I am really happy that you test it!
11:25:32Araqand while I can think about how to do things, I generally prefer to think about more pressing things
11:26:08Araqand NEd is interesting enough to tinker with it in its own right
11:27:20SalewskiYes I really like NEd, but I know that you have your own editor nimed, and manny people seems to like VSCode or similar.
11:28:16Araqwell if tinkering with it would be easier I could port some of my NimEdit features to it
11:29:28SalewskiFor NEd the fact is that it is GTK3, and people who hate GTK3 will never be happy with it.
11:30:46Araqyes, but many more people care about working software than what technologies it uses
11:31:12Araqand most new desktop apps are hidden fullblown webbrowsers
11:31:21Araqgtk3 is better than that at least.
11:31:40SalewskiNEd is a minimal editor, so people willing to do some coding can fix bugs or add features. But NEd can not compare to tools like vim or VSCode.
11:32:46Araqrm: das Entfernen von 'resources.o' ist nicht möglich: Datei oder Verzeichnis nicht gefunden
11:32:48AraqXMLLINT not set and xmllint not found in path; skipping xml preprocessing.
11:32:48SalewskiAll what I read in Nim forum about GTK3 was very negative.
11:32:49AraqPackage gtk+-3.0 was not found in the pkg-config search path.
11:32:51AraqPerhaps you should add the directory containing `gtk+-3.0.pc'
11:32:52Araqto the PKG_CONFIG_PATH environment variable
11:32:54AraqNo package 'gtk+-3.0' found
11:32:55*vlad1777d joined #nim
11:32:55Araqresources.c:1:21: fatal error: gio/gio.h: Datei oder Verzeichnis nicht gefunden
11:32:57Araqcompilation terminated.
11:33:07*yglukhov joined #nim
11:33:11*yglukhov quit (Remote host closed the connection)
11:34:09*yglukhov joined #nim
11:34:33Araqand yet you spend time on it. Don't do what other people think. Do what I tell you to do.
11:34:40Araq;-)
11:35:05SalewskiYes, will do.
11:35:20Araqwhat's the above error message?
11:35:26Araqhow do I fix it?
11:35:29*stisa joined #nim
11:36:31SalewskiHave to investigate it. So nim-chess compiles fine for you, but NEd not? Strange.
11:37:48SalewskiOf course for NEd you needs gtksourceview, which may be separate package from other GTK3. But message seems not to indicate that problem.
11:39:03SalewskiMr Krux has installed Ned successfully, so it should work.
11:40:28*irrequietus quit (Ping timeout: 245 seconds)
11:40:44SalewskiAraq, to install NEd you have to run the script "bash make_it"!
11:41:09Salewskinim c ned.nim will not wotk!
11:41:17Araqthat is what is producing this error
11:41:18*gokr quit (Ping timeout: 245 seconds)
11:42:04SalewskiWhich GTK3 version do you have installed?
11:43:04SalewskiAnd you can compile and install other GTK3 software?
11:43:15*pregressive joined #nim
11:43:30vlad1777dI had errors with NED too, but with gtksourceview.
11:43:55vlad1777dYou can correct the filename in gio.nim
11:44:05SalewskiI am using Gentoo, which compiles all from source, so missing developer files are no problem. For Ubuntu it may be a bit different.
11:44:07Araqgcc -O2 -c `pkg-config --libs --cflags gtk+-3.0` resources.c
11:44:14Araqfails for me
11:44:33Araqvlad1777d: doesn't seem to be my issue
11:46:22*[ui] quit (Quit: Connection closed for inactivity)
11:47:59AraqSalewski: 3.18.9-1ubuntu3.1
11:48:29SalewskiAraq, please see script make_it, there is a line commented out with gmodule-export-2.0, please try that.
11:48:52*pregressive quit (Ping timeout: 256 seconds)
11:49:13SalewskiIndeed I had a very hard time to get that script working, did a lot google search...
11:50:10vlad1777dAraq, header files are in -dev package, have you installed it?
11:50:31Araq"Perhaps you should add the directory containing `gtk+-3.0.pc'
11:50:33Araqto the PKG_CONFIG_PATH environment variable"
11:50:37AraqSalewski: same problem
11:50:59vlad1777dI use linux mint with the same version of GTK from Ubuntu Xenial, it compiled well.
11:51:42AraqI don't know, synaptic know no libgtk3 package to begin with
11:52:05SalewskiAraq, I am really sorry that I am wasting your time. I will try to find a solution, it may take some time...
11:52:36*gangstacat quit (Quit: Ĝis)
11:53:28vlad1777dAraq, apt install libgtk-3-dev
11:55:15Araqvlad1777d: that's it, thanks :-)
11:55:26SalewskiGreat!
11:55:56vlad1777dThe one problem I see - libgtk-3-dev has no gio.h inside
11:56:02vlad1777dlibglib2.0-dev has
11:56:13vlad1777dBut all that compiled for me today
11:57:41vlad1777dAnd Ned does not highlight syntax with my version of gtksourceview. Or I cannot turn it on.
11:58:01vlad1777dI use Atom now.
11:58:27FromGitter<nigredo-tori> @Araq, does documentation mention the fact that instantiations of the same generic type (not an alias) are compared structurally, rather than by type argument lists? E.g. ⏎ ⏎ ```type F[A] = object ⏎ echo F[int] is F[string] # This is true``` ⏎ ⏎ Manual section on type equality doesn't mention generics at all. [https://gitter.im/nim-lang/Nim?at=58591ce37a3f79ef5d7c139e]
11:58:45SalewskiYes, Kruxx told us. We have to do:
11:59:06Salewskicp nimdark1.xml /usr/share/gtksourceview-3.0/styles/
11:59:27Salewskicp nim.lang /usr/share/gtksourceview-3.0/language-specs/
12:00:08vlad1777dSalewski, thank you =)
12:00:16SalewskiAs root. I am not sure about destination directories, you have to see where the other files live.
12:00:24Araqnigredo-tori: that is what the compiler does but with the 'sighashes' branch we can do it differently
12:00:58AraqI think we shouldn't document this, but instead change it and then document it
12:01:05vlad1777dOk, I'll find, I thought that it does not work because of version incompatibility, but if it's missing style file - I'll correct it.
12:01:15SalewskiI have already asked at gtk list for better solutions and got a few answers, will update instructions soon.
12:01:18FromGitter<nigredo-tori> @Araq, thanks for the tip, I'll take a look at sighashes.
12:01:38Araqno need to look into it, I'm merging it now
12:01:48Araqsince all my tests work with it, including GTK3
12:02:41vlad1777dSalewski, I cannot use Ned now because there are no tabs, I need to edit several files, but when there is lack of free space on screen - I cannot do this.
12:03:08*stisa quit (Remote host closed the connection)
12:03:10vlad1777dSo for now I'll use Atom (Aporia highlights tabs with red color, it's very disturbing)
12:03:18SalewskiThere are tabs of course!
12:03:32vlad1777dMaybe I'll get used with Nim and will correct all this
12:04:15vlad1777dSalewski, but how to open it?
12:04:45SalewskiJust open a new file, it will show up inn new tab. And tiled view horizontal and vertical is great, including viewing same file in two views!
12:05:29SalewskiYou see all that in the picture in the manual.
12:05:52vlad1777dSalewski, thank you =) I had not thought about it )
12:06:39vlad1777dI got used to Xed's (old Gedit's) style, there was a button for this. Here it was absent - I thought that it's impossible to use tabs
12:08:11SalewskiAnd open the preferences menu -- you can hide tabs, hide scrollbars, show line numbers and much more. All what I need. Well not really, for some stuff I may use vim.
12:10:24vlad1777dStrange, but there is no preferences menu with this... https://pp.vk.me/c638428/v638428447/15465/6yzkzG2KPR4.jpg
12:12:28SalewskiPreferences Menu is not in NEd window, but on top of the screen!
12:13:00SalewskiThat location is used by all GTK3 programs for preferences!
12:14:26SalewskiAt least when you are using Gnome shell windows manager. It may be different for other window managers or KDE desktop.
12:15:09SalewskiSee where other GTK3 programs have preferences menu.
12:15:53*bjz joined #nim
12:16:48Araqtop of the screen menus are bullshit, thanks OSX.
12:17:46*bjz_ quit (Ping timeout: 256 seconds)
12:17:52*gangstacat joined #nim
12:17:55Araqit makes no sense whatsoever, why even have overlapping windows then.
12:18:38vlad1777dI did not understood nothing =)
12:18:51vlad1777dOn which top of screen =)
12:19:07SalewskiYes, I do not like the whole App style too much. But the preferences Menu is not that bad, it is not needed often.
12:19:15vlad1777dAll gtk applications have preferences or in Edit > preferences
12:19:31vlad1777dOr in some other button
12:19:32SalewskiWait a moment, I will give you a link vlad
12:19:42vlad1777dthank you, ok
12:21:28*circ-user-AmuXL joined #nim
12:21:50Salewskihttps://developer.gnome.org/gtk3/stable/ch01s04.html#id-1.2.3.12.5
12:22:42*circ-user-AmuXL is now known as _alexr_
12:22:52SalewskiBelow the middle of that page there is a picture showing the top on screen menu. At the bottom there is a picture for fallback.
12:23:31vlad1777dI understood
12:24:26*chemist69 quit (Ping timeout: 246 seconds)
12:24:37vlad1777dSeems like Cinnamon does not have such function =) https://pp.vk.me/c638428/v638428447/1546e/AwjtwV0DhYw.jpg
12:25:05*chemist69 joined #nim
12:25:05*irrequietus joined #nim
12:25:27vlad1777dIn popup menu in window list there are only actions with window (move it to other area, close, maximize)
12:25:46vlad1777detc
12:25:57vlad1777dBut I understood )
12:26:05SalewskiVLad, you have to investigate where other GTK3 tools like gedit have preferences menu.
12:26:46*_alexr_ quit (Remote host closed the connection)
12:27:46Araqbbl
12:29:39vlad1777dI understood what you mean. Cinnamon does not support this. All other apps have preferences available from buttons (mnemonics): https://pp.vk.me/c638428/v638428447/1547b/iJ5tf9bOVfE.jpg https://pp.vk.me/c638428/v638428447/15484/qS2zldxrfUE.jpg https://pp.vk.me/c638428/v638428447/1548d/_QJZAh08yRA.jpg
12:30:21vlad1777dCinnamon provides more "traditional" way for working with apps
12:37:50SalewskiVLad, try gedit. It must have app menu with preferences somewhere!
12:39:21vlad1777dok, I'll install it
12:41:31vlad1777dIt has preferences in properties > preferences: https://pp.vk.me/c638428/v638428447/15496/5vcXngWxNO4.jpg
12:42:53*ftsf_ quit (Quit: :q!)
12:43:07vlad1777d=) I understood. When I'll be on Gnome, I'll try Ned with preferences
12:43:28vlad1777dCinnamon does not support this menus.
12:45:50SalewskiYour gedit screenshot looks very different from real gedit of gnome3. May your cinnamon still using GTK2?
12:46:45vlad1777dheh, no, My Linux Mint 18 uses Gtk 3.18, in next Linux Mint 18.1, as I understood, will be Gtk 3.20
12:46:48SalewskiThere should really be a way to activate the app menu in Cinnamon, but I can not find it yet.
12:46:58vlad1777d=)
12:47:44SalewskiMaybe Cinnamon has patched all the GTK3 apps. But that would be strange.
12:47:58vlad1777dMaybe they'll make windowlist.js applet with support of such preferences soon )
12:48:29vlad1777dMaybe they patched some of them. I know that they patched terminal
12:48:41vlad1777dBut not system monitor and gedit
12:51:06SalewskiOf course I can provide a keyboard shortcut to open preferences, when it is really a problem for Cinnamon. I have to investigate that.
12:51:35*Arrrr quit (Quit: WeeChat 1.5)
12:51:42SalewskiHave to do some work, bye...
12:51:48*Salewski left #nim (#nim)
12:52:52vlad1777dbye )
12:58:57*stisa joined #nim
13:00:16FromGitter<dom96> > but it is just very slow at first to be able to do something in the compiler ⏎ ⏎ ⏎ @andreaferretti While the compiler is very important, it's not the only thing people could be helping out with. Nimble, NimForum, a website for Nimble packages are just some ideas. [https://gitter.im/nim-lang/Nim?at=58592b5fe7bdfe4e2978f549]
13:00:35vlad1777dWhere can I read about stack and heap difference ?
13:02:49flyxvlad1777d: in general, or specific to Nim?
13:03:16vlad1777dflyx, maybe both =)
13:03:24vlad1777dbetter specific to Nim
13:05:29flyxwell the concept is the same in all programming languages
13:05:40flyxwell, unless they abstract it away
13:06:21vlad1777dI want to know, how to make objects create in the heap, how on stack
13:06:40vlad1777dbecause seems I created all objects in stack, they're untraced
13:07:00flyxthere is no need for tracing if objects are on the stack
13:08:21vlad1777dBut will they be automatically removed with garbage collector?
13:08:44flyxno, there's no need. they will vanish when their stack frame is left
13:09:23vlad1777dand can I create reference to them ?
13:09:35vlad1777dI read that references point to objects in heap
13:09:53flyxyes, they do, and no, you cannot create a reference to stack objects
13:10:27vlad1777dthank you
13:11:18*stisa quit (Ping timeout: 245 seconds)
13:11:26flyxhere's a good SO on the general topic: http://stackoverflow.com/questions/79923/what-and-where-are-the-stack-and-heap
13:13:44vlad1777dthank you very much )
13:15:26*gokr joined #nim
13:18:37*mjanssen joined #nim
13:20:26mjanssenI am having problems building nim from source using Mingw64. http://pastebin.com/6k1c1E4q any ideas?
13:30:08*stisa joined #nim
13:55:26Araqmjanssen: never saw that error before but it looks like header files disagree about select?
13:58:43mjanssenAraq: looks like it. Seems to be an issue with 64 bit only. Will try with a clean mingw install
14:03:49mjanssenok false alarm, was using 32 bit headers with 64 bit gcc. Mingw installation error
14:13:05federico3are non-comments ignored by the syntax parser? Any way around it?
14:14:24Araqnon-comments ignored, what?
14:18:45federico3ahem, sorry, non-documentation comments: # foo (instead of ## foo)
14:25:26*Salewski joined #nim
14:26:50SalewskiVLad, your problem with missing application menu in Cinnamon seems to be confirmed, see https://mail.gnome.org/archives/gtk-list/2016-December/msg00048.html
14:27:18SalewskiI will see how I can better support Cinnamon.
14:27:59SalewskiBye.
14:30:53*Salewski left #nim (#nim)
14:40:14*chemist69 quit (Ping timeout: 246 seconds)
14:43:06*gokr quit (Read error: Connection reset by peer)
14:44:48*chemist69 joined #nim
14:47:25Araqfederico3: there is no way around it. they used to be part of the AST but people complained about the rigid commenting rules
14:50:15federico3pity, I was using parser+generator as a linter. It works with everything else but it's removing comments
14:51:45Araqwhat is parser+generator?
14:51:56Araqwhat kind of linter? what about 'nim check'?
14:53:36Araqah you mean parser+renderer
14:54:18Araqcheckout the 'nimfix' tool please
14:54:20federico3it currently uses the compiler's parser and the renderer but I could switch the latter to a different implementation to do more mangling
14:54:40federico3nimfix?
14:55:02federico3found it
14:56:13federico3thanks Araq
14:57:53Araqit needs to be patched to use the latest compiler version
15:00:18Araqthere you go.
15:27:49*stisa quit (Ping timeout: 260 seconds)
15:28:17*marchelzo joined #nim
15:28:22*stisa joined #nim
15:31:44*subsetpark joined #nim
15:32:36*stisa quit (Ping timeout: 240 seconds)
15:39:51*pregressive joined #nim
15:44:14vlad1777dI have some questions. 1. I declared variable: "var x: myTypeObjec", it's placed in Stack with "nil" values, right?
15:45:05vlad1777d2. If I'll make "x = new(val1 = 's', val2 = 323)", than will it be removed from stack and placed to heap ?
15:50:04euantorIs myTypeObjec an object or an object ref?
15:50:06FromGitter<andreaferretti> no, it doesn't work like that
15:50:25FromGitter<andreaferretti> the fact that a value lives in the heap is tracked in the type
15:50:43FromGitter<andreaferretti> if type myTypeObject = ref something
15:50:51FromGitter<andreaferretti> then myTypeObject is a reference type
15:50:59FromGitter<andreaferretti> and x lives in the heap
15:51:07FromGitter<andreaferretti> otherwise, on the stack
15:51:35FromGitter<andreaferretti> the `new` operator, for reference types, allocates the relevant space in the heap
15:52:29FromGitter<andreaferretti> If you see here
15:52:30FromGitter<andreaferretti> http://nim-lang.org/docs/system.html#new,ref.T
15:52:41FromGitter<andreaferretti> `new` only exists for reference types
15:53:33FromGitter<andreaferretti> for a reference type, when you write `x: myTypeObject`, x is initialized to a nil pointer
15:54:09FromGitter<andreaferretti> when you use `new`, x points to an area of the heap where there is room to write the content of x
15:59:59*stisa joined #nim
16:09:32*stisa quit (Ping timeout: 268 seconds)
16:13:19*dddddd joined #nim
16:17:17vlad1777dauantor, object
16:19:12*PMunch joined #nim
16:20:31vlad1777doh, I errored, euantor =)
16:30:04*pregressive quit (Remote host closed the connection)
16:30:40*pregressive joined #nim
16:35:02*pregressive quit (Ping timeout: 256 seconds)
16:40:44*yglukhov_ joined #nim
16:41:12*Andris_zbx quit (Remote host closed the connection)
16:44:17*yglukhov quit (Ping timeout: 240 seconds)
16:44:50*yglukhov_ quit (Ping timeout: 246 seconds)
16:50:47*yglukhov joined #nim
16:55:10*yglukhov quit (Ping timeout: 268 seconds)
16:58:52*yglukhov joined #nim
17:01:45*pregressive joined #nim
17:02:04*brson joined #nim
17:03:05*yglukhov quit (Ping timeout: 258 seconds)
17:08:27*Trustable joined #nim
17:11:03*space-wizard joined #nim
17:11:57*rokups quit (Quit: Connection closed for inactivity)
17:33:13*yeeve quit (Remote host closed the connection)
17:38:38*pie_ joined #nim
17:49:25*space-wizard quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
17:49:46*space-wizard joined #nim
17:50:12*space-wizard quit (Client Quit)
17:50:34*space-wizard joined #nim
17:51:00*space-wizard quit (Client Quit)
17:51:23*space-wizard joined #nim
17:51:48*space-wizard quit (Client Quit)
17:52:11*space-wizard joined #nim
17:52:36*space-wizard quit (Client Quit)
17:52:58*space-wizard joined #nim
17:53:24*space-wizard quit (Client Quit)
17:53:47*space-wizard joined #nim
17:54:12*space-wizard quit (Client Quit)
17:54:35*space-wizard joined #nim
17:55:00*space-wizard quit (Client Quit)
17:55:23*space-wizard joined #nim
17:55:48*space-wizard quit (Client Quit)
17:56:10*space-wizard joined #nim
17:56:36*space-wizard quit (Client Quit)
17:56:58*space-wizard joined #nim
17:57:24*space-wizard quit (Client Quit)
17:57:46*space-wizard joined #nim
17:57:49*yglukhov joined #nim
17:58:12*space-wizard quit (Client Quit)
17:58:35*space-wizard joined #nim
17:59:00*space-wizard quit (Client Quit)
17:59:10dom96Araq: why was 'map' renamed to 'apply'?
17:59:52FromGitter<andreaferretti> uh? when did it happen?
18:02:08*yglukhov quit (Ping timeout: 258 seconds)
18:03:06dom96oh wait
18:03:11dom96'apply' is in-place
18:03:15dom96map returns a new seq
18:03:24dom96that's cool
18:03:39Araqgood
18:03:49*Trustable quit (Remote host closed the connection)
18:09:31dom96andreaferretti: is andreaTP your colleague?
18:12:37Araqnow there is some action here, can we decide on an import syntax quickly?
18:12:47Araqimport compiler/foo, compiler/bar, compiler/baz is annoying
18:12:53dom96indeed
18:13:02Araqimport compiler[foo, bar, baz] # proposal
18:13:19dom96import compiler/[foo, bar, baz]
18:13:35dom96or import compiler/(foo, bar, baz)
18:13:42dom96or import compiler/{foo, bar, baz}
18:13:46dom96or import compiler.[foo, bar, baz]
18:13:47dom96etc
18:15:28dom96I wouldn't be surprised if there is an issue for this, I suggested this since Nimble's inception :P
18:20:11AraqI still want foo.bar to be different from foo/bar
18:20:53Araqbut that ship has sailed I guess -.-
18:21:22Araqit suggests we like to keep the '/' in there though
18:21:50Araqimport compiler/{foo, bar, baz}
18:23:05*confundus joined #nim
18:24:12Araqimport compiler/[foo, bar, baz]
18:24:20dom96different in what way?
18:24:42Araqdifferent semantics: foo/bar # import bar from foo directory
18:24:52Araqfoo.bar # import bar from foo *package*
18:25:38Araqplus some mechanism to tell the compiler how to find package foo
18:25:54vlad1777dmaybe it would be good to make: "from compiler import foo, bar, baz"
18:26:06Araqthat already means something else
18:27:07vlad1777dI mean that to use some word to specify directory, than to write import and names with comma
18:27:24vlad1777dBecause that slash\scopes looks not very clear and implicit
18:28:33Araqslashes give you the right idea though, it's about paths
18:29:42dom96Araq: directories and packages are already conflated
18:30:08dom96plus, you can't write: compiler/foo/bar.func()
18:32:09*Matthias247 joined #nim
18:32:20*rokups joined #nim
18:32:58Araqyeah
18:33:31Araqlet's just go with import compiler/[foo, bar, baz]
18:34:32dom96and import compiler.[foo, bar, baz]
18:34:58Araqwhy?
18:35:08dom96also, wouldn't {} make more sense, since the directories have to be unique? :)
18:35:30Araqthe dot is much harder to support, that would require a grammar/parser change
18:35:30dom96because I can write either 'import compiler/foo' or 'import compiler.foo' right now AFAIK
18:35:42vlad1777dthat's Perl-way =)
18:35:47Araqyes but 'import compiler.foo' should go away
18:36:50*space-wizard joined #nim
18:37:27vlad1777dmaybe to think about word to specify directory from where import modules ?) It's your task and thought, but for me, as for Python user it would be much better to understand )
18:37:44dom96Araq: But then it's not consistent with writing 'compiler.foo.blah()' in code
18:38:10dom96but maybe that's okay?
18:41:04Araqyou cannot write compiler.foo.blah()
18:42:00Araqthere is no notion of <directory>.<module>.name in the compiler, it's just a special case in 'import'
18:42:00dom96vlad1777d: what syntax would you suggest? :)
18:42:38Araqdom96: I'd argue that a list of modules follows, not a 'set'
18:42:42dom96Araq: hrm, you can write that in Python though
18:42:47vlad1777ddom96, some_word_instead_of_from_which_is_busy compiler import foo, bar, baz
18:43:11AraqPython maps directories to Python identifiers via some mechanism that escapes me
18:43:14Araqwe don't.
18:43:18dom96vlad1777d: finding that word is the difficult part :)
18:43:35dom96Araq: why don't we?
18:43:45Araqexistance of an __init__.py file within the dir or something.
18:43:51vlad1777ddom96, but it would be better by result..
18:44:05Araqbecause I find Python's import mechanism an abomination
18:44:22dom96Araq: and we instead have existence of a .nimble file :P
18:44:26vlad1777d+ for Python's import mechanism is bad
18:44:33dom96and yet it's even more subtle for us
18:44:41vlad1777dmaybe to change from to using
18:44:42Araqnever worked for me and the "easy" packages choose a single Python file distribution
18:44:58dom96because this .nimble file isn't required, until our compilation starts failing because there are two modules with the same name
18:45:18vlad1777d"using" keyword is used in many languages for that purposes, for which it's used in Nim and Python now
18:45:33vlad1777d(for example, in C#)
18:45:43Araqthat's a nice feature, dom96. package.module.symbol SHOULD be unique
18:46:21AraqI don't want to live in your world of "everything is more elegant when it has the same name"
18:47:16dom96shouldn't every directory be a package?
18:47:36Araqwhat kind of package is 'source'?
18:48:27Araqplus it causes other problems: /Users/araq/projects/nim/compiler/foo.nim
18:48:45Araqwhat is part of the relative path that we're interested in?
18:49:10Araqor should we base everything on absolute paths causing the builds to be completely path dependent?
18:50:27*confundus quit (Quit: leaving)
18:51:53dom96I don't understand that problem unfortunately.
18:52:43dom96The compiler has a list of paths, and you are always importing something from them.
18:52:56dom96So let's say "/Users/araq/projects/nim/" is in your path
18:53:07dom96You try to import "compiler/foo" and it works
18:55:02*chemist69 quit (Ping timeout: 258 seconds)
18:59:25vlad1777das for me, if "using compiler import foo" will specify what to use without explict full qualification, and "from compiler import foo, bar, other" will import foo, bar, other from compiler
18:59:47vlad1777dwill be better )
19:00:14vlad1777dbut I agree with your opinion - you have more experience
19:03:55Araqdom96: we need to generate code. something like modulenameInit()
19:04:17Araqbut oh, that's not unique so we use package_modulenameInit()
19:04:46Araqnow if package is not unique either the name needs to be full_path_to_moduleInit()
19:05:11Araqbut that's an absolute path encoded here, bad for all kind of reasons
19:05:30Araqfor example: we have tests that check the produced C code
19:05:48Araqnow that C code depends on the full path. which can differ from test machine to test machine...
19:12:58*[ui] joined #nim
19:13:18dom96so the only reason is because of tests?
19:13:20*chemist69 joined #nim
19:13:37dom96surely you can improve the tests to take this into account
19:15:36Araqthat's just one example.
19:15:54Araqyou can have your 'nimcache' on a shared server
19:16:55Araqgenerated code should not depend on absolute paths.
19:20:18dom96okay, so why is an absolute path necessary?
19:20:43Araqwell with the current setup, it isn't.
19:20:43dom96You've got `import compiler/foo` in the source code, `compiler/foo` is unique
19:20:52Araqit is with your proposal.
19:21:04Araqno, 'compiler/foo' is ambiguous.
19:21:24Araqfirst interpretation is relative to the file it was written in.
19:22:08*space-wizard quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
19:22:21Araq'compiler/foo' can exist in lots of different Nimble packages
19:26:26dom96Okay, I get it.
19:27:22dom96The question is, how do other languages solve it?
19:30:08*pie_ quit (Ping timeout: 256 seconds)
19:31:20AraqI don't know what's so bad about "unique name per nimble package"
19:32:45dom96the need to create a .nimble file is not ideal
19:33:33dom96The problem is that it's not immediately obvious that it's necessary
19:34:59dom96Nimble enforcing these files does solve it
19:37:52Araqthe compiler gives a nice error message for this nowadays
19:38:18Araq(ignoring the reported module rename bug that the compiler doesn't catch yet)
19:39:09Araqin Python you need to create an __init__.py file instead.
20:02:15*yglukhov joined #nim
20:04:02*djellemah_ joined #nim
20:06:16*djellemah quit (Disconnected by services)
20:06:19*djellemah_ is now known as djellemah
20:17:14*yglukhov quit (Remote host closed the connection)
20:17:50*yglukhov joined #nim
20:18:37*abeaumont quit (Ping timeout: 265 seconds)
20:21:56*yglukhov quit (Ping timeout: 240 seconds)
20:32:08*derlafff quit (Remote host closed the connection)
20:38:55*brson quit (Ping timeout: 258 seconds)
20:42:00*derlafff joined #nim
20:50:16*yglukhov joined #nim
20:51:57*rokups quit (Quit: Connection closed for inactivity)
20:54:27*brson joined #nim
20:55:51renesac_when I use --passL:"-static" to compile my program, it will crash with SIGSEGV immediatedly uppon launching in my system, but will run fine in a remote one with a different glib
20:56:20renesac_if I don't use that flag, it will run perfectly in my system, but not run in the remote one
20:57:57renesac_when compiling with that flag I get: ~/nimcache/stdlib_system.o: In function `nimLoadLibrary':
20:57:58renesac_stdlib_system.c:(.text+0x732a): warning: Using 'dlopen' in statically linked applications requires at runtime the shared libraries from the glibc version used for linking
20:58:16renesac_any idea?
21:04:23*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
21:14:12Araqrenesac_: I don't know about -static but I would try muslimC for static linking
21:14:32Araqmusl, or whatever, can never remember its name
21:14:51renesac_ok, I read that just by passing that it would link the glibc, and I didn't care about the binary size
21:15:07renesac_but I may use one of those mini libcs
21:16:16*irrequietus quit ()
21:16:23*[ui] quit (Quit: Connection closed for inactivity)
21:16:26marchelzomuslimC haha
21:16:51marchelzosomeone should make a libc called muslim
21:18:27marchelzoAraq: it's pronounced like "muscle" or "mussel". That should help you remember.
21:21:24AraqI don't want to remember. I want my life back. :P
21:25:15*shodan45 joined #nim
21:25:50*bjz joined #nim
21:31:04Araqping yglukhov
21:31:12yglukhovзong
21:31:17yglukhov*pong
21:31:50yglukhovAraq
21:31:53Araqabout https://github.com/nim-lang/Nim/issues/5131
21:32:24Araqit's hard to ensure backwards compatibility with getType()
21:33:04Araqthe API is fundamentally flawed, can't you use the new API instead?
21:33:39yglukhovAraq: I'm fine if you say i misused the API and there should be a way to achieve what i want.
21:34:25yglukhovi mean, in this particular case i dont mind if its a breaking change for the sake of correctness or smth.
21:35:19yglukhovwhat do you mean by the "new api"? did i miss anything?
21:35:59AraqgetTypeImpl, getTypeInst
21:37:37yglukhovoh. interesting...
21:38:10*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
21:38:42Araqoh well, I think I fixed it.
21:38:53Araqstill: a pita to maintain this strange thing
21:39:47yglukhovnp, i'll rewrite my code to whatever api you suggest. just would be nice to have a couple of days to do that ;)
21:39:48*ryshys joined #nim
21:39:54*ryshys left #nim (#nim)
21:40:23yglukhovso if you can fix it easily - would be cool. if not - oh well...
21:40:34Araqthe output of your example is just "vec2", right?
21:44:00Araqfixed. Someday I will rewrite it though. Now that we have a better type system internally we can do that :-)
21:45:14yglukhovyep, the output should be "vec2"
21:45:40yglukhovthanks, let me try it
21:46:43*byte512 quit (Ping timeout: 245 seconds)
21:48:37*yglukhov quit ()
21:51:11*yglukhov joined #nim
21:52:11vlad1777dI have some questions. 1. I declared variable: "var x: myTypeObjec", it's placed in Stack with "nil" values, right?
21:52:29vlad1777d2. If I'll make "x = new(val1 = 's', val2 = 323)", than will it be removed from stack and placed to heap ?
21:52:41vlad1777dI would be grateful for the answer on them )
21:53:34vlad1777dAh, I forgot. 3. In C++ data created on the heap will be pointed to by pointers and allocated with new or malloc. Is it right for Nim ?
21:56:17*Trustable joined #nim
22:03:33renesac_vlad1777d: I'm not 100% sure but:
22:03:50renesac_1. you will have a nil pointer on the stack
22:04:25renesac_2. that don't seems valid nim code, but a constructor for such ref object would allocate the space on the heap and populate it
22:05:07renesac_3. something like that, but I'm not sure what is your doubt
22:06:23renesac_the first answer assumes it is a ref object
22:06:44vlad1777drenesac_, thanks you, the first was the most important
22:07:13vlad1777dYes, in second I errored, I meant "x = myType1(val1 = 's')"
22:09:18yglukhovAraq, well the getType thing might be fix (it compiles), but now the program crashes, so there's another regression i guess..
22:10:43yglukhovthe crash is in SDL_GL_CreateContext, which is foreign. any clues?
22:11:00*nsf quit (Quit: WeeChat 1.6)
22:11:24yglukhovI'm trying to diff the C sources, but I see you've merged your new mangling rules.....
22:16:16yglukhovapart from mangling, the relevant chunks of code dont differ much.
22:30:57Araqyglukhov: type info generation changed
22:31:56yglukhovhrm.. any clues how that should matter in this case?
22:32:25AraqI have a tool for diffing btw
22:33:51Araqhttps://gist.github.com/Araq/7ae88f36e09c3d2b0c0bda3fdc254e30
22:34:10Araqthat's what I used to debug the new codegen :-|
22:34:27yglukhovcool!
22:34:45yglukhovso i use it on the new code to get the old code, right?
22:35:29Araqit strips the suffixes, you pass a backup filename extension
22:35:42Araqnimdiff .devel
22:35:46Araqnimdiff .sighashes
22:36:04Araqand then you use WinMerge for a nice UI
22:36:12Araqor some inferior tool on OSX
22:36:14Araq:P
22:36:22yglukhovah ok
22:36:27yglukhovlets see
22:37:02Araqif you can give me access to the project I can look into it
22:37:20Araqit's a pita to debug if you don't know what to look for
22:38:36yglukhovnimx =)
22:39:58Araqwhat test to run?
22:43:44yglukhovAraq: cd test && nake
22:44:18yglukhovAraq: i tried to launch it with nim, and it didnt crash
22:45:08yglukhovmaybe i should do some more experiments. but the behavior definitely depends on nim version.
22:52:38AraqCompiling nakefile...
22:52:40AraqNo task specified, running default task defined by nakefile.
22:52:42AraqWarning: icon was not found: build\windows\res\MyGame.ico
22:52:43AraqTraceback (most recent call last)
22:52:45Araqnake.nim(90) moduleHook
22:52:46Araqnakelib.nim(260) runTask
22:52:47Araqnaketools.nim(874) :anonymous
22:52:49Araqnaketools.nim(753) build
22:52:51Araqnaketools.nim(527) nimbleOverrideFlags
22:52:52Araqnaketools.nim(255) nimblePath
22:52:53Araqosproc.nim(1250) execCmdEx
22:52:55Araqosproc.nim(566) startProcess
22:52:56Araqos.nim(153) raiseOSError
22:54:33yglukhovAraq: reduced to nim only: nim c -r --threads:on --noMain -d:SDLStatic --passL:-lSDL2 main.nim
22:54:40*krux02 joined #nim
22:55:19yglukhovthe key is that we're using "native" linkage with SDL2 instead of Nim's dylinker
22:56:24yglukhovcompiling with Nim's dylinker makes this problem disappear.
22:57:20yglukhovand soon after that i see the next problem =) damn its a long chain...
23:00:58krux02yglukhov: programming is hard, it's a marathon
23:01:39*nsf joined #nim
23:08:10yglukhovAraq: so another regression is that drawing is that some elements (fonts, images, smth that uses paths along the way?) are not drawing. if you run "nim c -r --threads:on --noMain main.nim" you will be able to compile and run nimx with both nim devel and 9099fdd8f02b83670ae0ca8a5e434499c56c9904, but devel introduces this drawing bug
23:08:32yglukhovahh, sorry for the typos...
23:09:15yglukhovi feel like i need to better track down those bugs for you, do i?
23:10:16Araqwell I get some other stack traces on linux
23:10:39AraqException caught: over- or underflow.
23:10:49Araq...
23:10:51Araqfont.nim(144) bakeChars
23:10:53Araqstb_ttf_glyph_provider.nim(70) bakeChars
23:10:54Araqsystem.nim(2562) sysFatal
23:11:16Araqa smaller example would surely be nice
23:15:35yglukhovAraq: thats interesting because stb_ttf_glyph_provider.nim(70) is as follows: let gh = y1 - y0
23:15:47yglukhovwhere y1 and y0 are cint
23:16:00yglukhovdo we have range checks for cints?
23:18:17yglukhovAraq: also i think SDL_GL_CreateContext could be pretty platform specific. got a mac nearby?
23:23:06Araqsure
23:27:01yglukhovAraq: ok sorry gotta go, will do some more testing tomorrow.
23:27:08yglukhovsee you
23:27:11Araqbye
23:27:11*ftsf_ joined #nim
23:27:19*Matthias247 quit (Read error: Connection reset by peer)
23:27:28*yglukhov quit (Remote host closed the connection)
23:30:05FromGitter<zetashift> @dom96, will the Nim in Action book go on sale this holiday season I want to buy it but kinda poor at the moment
23:31:16Araq.nimble/pkgs/nimx-0.1/nimx/cursor.nim(70, 42) Error: undeclared identifier: 'SystemCursor'
23:31:22*Araq sighs
23:31:43dom96zetashift: no idea i'm afraid
23:33:08FromGitter<zetashift> ah okay, I'll just save up some more thanks!
23:42:35*nsf quit (Quit: WeeChat 1.6)
23:43:48*yglukhov joined #nim
23:48:21*yglukhov quit (Ping timeout: 248 seconds)
23:51:38*Trustable quit (Remote host closed the connection)