<< 20-04-2019 >>

00:00:44shashlickWow wonder how this works
00:05:17shashlickOk the tests help
00:11:54*rnrwashere quit (Remote host closed the connection)
00:29:40*smitop joined #nim
00:29:49smitopCan I update my Nim on Windows by running the newer installer on the Nim website?
00:48:22leorizethere's an installer?
00:49:03leorizewhat I typically do is that I run `choosenim` and forget about it :p
00:49:35*Tyresc quit (Quit: WeeChat 2.5-dev)
00:51:03*rnrwashere joined #nim
00:55:07*rnrwashere quit (Ping timeout: 240 seconds)
00:58:20disruptekwhy, typically, does `nim doc|doc2` refuse to document all procs in my module?
00:59:09leorizedid you export them?
00:59:27leorizeif you don't the docs won't be generated, there's an issue opened to request for it
00:59:43disruptek:-D thank you, sir!
01:02:43*rnrwashere joined #nim
01:19:19leorizeso far --newruntime is slower than the GC at binary tree benchmark :p
01:23:52*juliusdeane joined #nim
01:29:14leorizeAraq: this should not compile on newruntime I believe http://ix.io/1GGU
01:30:42leorizesince there's no owned ref to capture `new()` output, it is destroyed immediately, right?
02:07:11*banc quit (Quit: Bye)
02:15:58*dddddd quit (Remote host closed the connection)
02:21:48*rnrwashere quit (Remote host closed the connection)
02:22:55*juliusdeane quit (Remote host closed the connection)
02:28:31*banc joined #nim
02:42:28FromGitter<Varriount> Araq: Where can I see the changes to koch that were required for the new runtime?
02:56:09*rnrwashere joined #nim
03:08:52FromGitter<jrfondren> it seems like I have this exact problem 100% of the time I touch a Ruby project, so the problem may really be with me
03:09:59FromGitter<jrfondren> but even after an rbenv install of Nim/website's .ruby-version following https://jekyllrb.com/docs/installation/macos/ , and install bundler jekyll , and making sure I've got the right PATH and resolved binary
03:10:21FromGitter<jrfondren> jekyll build fails as it can't find a version of a gem that's not available anymore
03:11:20FromGitter<jrfondren> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5cba8dd7990feb4518cf5b36]
03:13:47FromGitter<jrfondren> ah. that's the wrong syntax to select a version, and "bundler install" grabs that and the hundred other specific versions for you.
03:21:34*smitop quit (Quit: Connection closed for inactivity)
03:23:21*arecacea1 quit (Remote host closed the connection)
03:23:40*arecacea1 joined #nim
03:25:26*rnrwashere quit (Remote host closed the connection)
03:38:36*rnrwashere joined #nim
03:41:35*rnrwashere quit (Remote host closed the connection)
03:43:13*rnrwashere joined #nim
04:14:41*rnrwashere quit (Remote host closed the connection)
04:23:20*arecacea1 quit (Remote host closed the connection)
04:23:39*arecacea1 joined #nim
04:29:28FromGitter<arnetheduck> why is otherwise trivial stuff like `min` and `max` compiler magic? just for the folding? if yes, does it *need* to be folded by magics? why not let ctfe deal with it? Araq?
04:39:00*rnrwashere joined #nim
04:50:08*i7sDream_ quit (Ping timeout: 246 seconds)
04:51:59*i7sDream_ joined #nim
04:54:31*kapil____ joined #nim
05:26:57Araqleorize: correct, bug report please
05:27:08*rnrwashere quit (Remote host closed the connection)
05:27:15Araqarnetheduck: pure legacy
05:28:00Araqthe ctfe engine can deal with generic/inline min/max just fine
05:28:31*rnrwashere joined #nim
05:29:00AraqVarriount: there weren't any changes required for koch.nim itself, the stdlib got some 'owned' annotations.
05:37:37*rnrwashere quit (Remote host closed the connection)
06:19:01*solitudesf joined #nim
06:38:01*Trustable joined #nim
06:41:56*disruptek quit (Quit: Bye)
06:45:38*disruptek joined #nim
07:00:00*gmpreussner quit (Quit: kthxbye)
07:00:02*nsf joined #nim
07:05:02*gmpreussner joined #nim
07:21:15*ua quit (Ping timeout: 252 seconds)
07:23:00*ua joined #nim
07:29:36*chickendan joined #nim
07:33:27*rauss quit (Ping timeout: 250 seconds)
07:35:33*rauss joined #nim
07:35:42Zevvwhich is pretty cool
07:37:02*kobi7 joined #nim
07:45:18*shadowbane quit (Ping timeout: 255 seconds)
07:45:40*solitudesf quit (Ping timeout: 268 seconds)
07:46:13*shadowbane joined #nim
07:59:57*literal quit (Remote host closed the connection)
08:07:37*vlad1777d quit (Remote host closed the connection)
08:09:50*vlad1777d joined #nim
08:10:24*vlad1777d quit (Remote host closed the connection)
08:11:21*vlad1777d joined #nim
08:24:53*neceve joined #nim
08:46:25AraqZevv, it was expected, koch.nim doesn't use 'ref'. In fact, you can write everything without 'ref', Nim is Turing complete without it :-)
08:55:29*shadowbane quit (Ping timeout: 246 seconds)
08:55:43*shadowbane joined #nim
09:06:19ZevvI find myself reffing very little these days indeed. It took some time letting go of the C habits and trusting the compiler on this.
09:13:13Zevvbut if all I'd need was Turing-completeness I'd use brainf*ck
09:13:42Araqsure, Turing is always a terrible argument, I know.
09:13:55ZevvI guessed that much :)
09:13:57Araqanyway, does 'gcc' on linux support @linkingfile like on Windows?
09:14:08AraqI'm running out of command line space on Linux
09:14:24AraqI think. Hard to say since the process simply dies with no good error message
09:15:07Zevvyou're hitting ARG_MAX?
09:15:35Zevvthat's a few hundred kb's I believe
09:15:46AraqI think I do, yes
09:15:51Araqit's a long linking command :P
09:16:44FromGitter<mratsim> If all I needed was Turing completeness, I would use PowerPoint/Magic cards/C++ templates :P
09:18:45Araqcan I increase ARG_MAX?
09:18:49ZevvAraq: my arg_max is 2MB - not sure if this is your problem, what do you get on 'getconf ARG_MAX'
09:19:41Zevvafaik there is no knob to tweak that though
09:20:29FromGitter<mratsim> @PMunch, apparently in your protobuf implementation you only support proto3, I'd like to build a pure Nim Starcraft2 client but I need proto2 for the following schema: https://github.com/Blizzard/s2client-proto/blob/master/s2clientprotocol/sc2api.proto, what's the difference between proto2 and proto3?
09:21:14Araqhmm the command line is 141kb only
09:21:27FromGitter<mratsim> "only"
09:21:40Araqand my argmax is 2 MB too
09:21:51Zevvlet me dig into that
09:21:57Araqmust be something else...
09:22:00Araqbbl
09:23:24*shadowbane quit (Ping timeout: 255 seconds)
09:24:31ZevvI can pass 1 MB without problems, so something else is your problem probably
09:34:34FromGitter<jrfondren> yeah you can increase ARG_MAX. not the constant, but the actual kernel limitation is set by sysctl
09:36:00FromGitter<jrfondren> er... that's macOS actually.
09:36:45FromGitter<jrfondren> posix_spawn can return an E2BIG there. nothing like that for linux
09:41:33FromGitter<jrfondren> after reading on the forum that nim-lang should have more code examples, I came up with https://gist.github.com/jrfondren/c2bc601f28a5c9349bdab52aa3270354 . and it's after that I got liquid more less figured out that I realize I don't want to propose to more code examples to the front page anymore
09:42:12FromGitter<jrfondren> but I thought all language homepages had such things but looking around now, it looks like that fad's over
09:42:41FromGitter<jrfondren> dlang.org has a drop-down list and a bunch of examples. it's the only one like that that I found.
09:43:46*Vladar joined #nim
09:43:56FromGitter<jrfondren> python.org has some small examples that aren't really there to be code examples. they're more like illustrations of the accompanying text.
09:45:01FromGitter<jrfondren> I much preferred the normal webpages, "here's a bunch of separate code examples and a bunch of prose talking about each of them in turn. scroll to read instead of fiddling with HTML forms"
09:46:09*nsf quit (Quit: WeeChat 2.4)
10:10:08*dddddd joined #nim
10:11:00*tefter joined #nim
10:17:08leorizeAraq: IIRC the command line size is 1/4 of the stack size
10:17:20leorizeuse `ulimit` to increase the stack size
10:19:36leorizealthough I think you might be better off generating a ld script to link your files
10:25:14*smitop joined #nim
10:25:18*xace quit (Ping timeout: 244 seconds)
10:26:26*xace joined #nim
10:26:51leorizeoops I was wrong, the arg size is fixed. so creating a linker script is the better option now
10:36:17Araqhow can I create such a linker script?
10:41:38ZevvThat'll require some fiddling I guess, it's not a common thing to do. Gnu ld scripts are usually only used for mapping sections etc.
10:42:09Araqon Windows I can pass @file to gcc and be done with it :P
10:42:18Zevvyeah :)
10:43:11Zevvhm wait but that doesn't work on linux?
10:43:30AraqI don't know, I assume it's a windows specific extension
10:43:42ZevvI see its in the man page - I never new it existed
10:44:00ZevvI was looking into putting this all in linker scripts, but thats something totally different
10:45:14Araqmaybe the command line should be a "file". lol I hate Unix.
10:45:15Zevvit's also for gcc
10:45:33Zevvthe command line is just '@file', so the same as your windows case
10:45:40Araqyup, it works :-)
10:45:42AraqI tested it
10:45:46Zevvsame here
10:46:00Araqcan anybody confirm clang also works with it?
10:46:01Zevvwell, good for you. another reason to love unix
10:46:10Zevvconfirmed
10:46:18Araqwow that was quick, thanks
10:47:02Zevvwell, I didn't need to use my mouse to open an explorer window and click through 13 folders first
10:47:22ZevvI could just type 'clang @cmd' and see the result. God, I love unix
10:48:00Zevvit even works with tcc :)
10:50:15Araqgood good
10:50:23Araqtime to go outside, bbl
10:50:27leorizein case anyone wonders about linker script. You just need to create a text file full of `INPUT(obj.o)` separated by new lines
10:50:33Zevvenjoy the blazing april sun
10:51:06Zevvleorize: yeah, that's a traditional linker script. But @file just reads regular command line arguments from a file
10:51:22ZevvI'm not sure if you can express all normal cmdline business in a linker script
10:51:22leorizedamn, never knew that exists
10:51:33Zevvsame here. and it's just in the man page
10:52:07leorizeaccording to ld manual the command line flags are compiled into linker script for linking :p
10:53:02leorizemore info here: https://sourceware.org/binutils/docs/ld/Scripts.html
10:53:42*rokups joined #nim
10:56:03leorize!eval echo "Hello"
10:56:04NimBotCompile failed: <no output>
10:58:08*ng0 joined #nim
11:02:27*nsf joined #nim
11:06:12*Perkol joined #nim
11:19:29*matti joined #nim
11:22:24*Perkol quit (Remote host closed the connection)
11:36:47*zyklon quit (Ping timeout: 240 seconds)
11:43:19*zyklon joined #nim
11:45:17*rnrwashere joined #nim
11:56:11*kobi7 quit (Quit: Leaving)
11:58:09*Snircle joined #nim
11:59:44Zevvstill dead
12:06:27*lritter joined #nim
12:10:35*stefanos82 joined #nim
12:16:44dom96The server is still free if anyone wants to try setting up the Nim playground on it
12:24:00*nsf quit (Quit: WeeChat 2.4)
12:28:14*rnrwashere quit (Remote host closed the connection)
12:31:17*rnrwashere joined #nim
13:02:09*ng0 quit (Ping timeout: 256 seconds)
13:06:44*ng0 joined #nim
13:26:29*rnrwashere quit (Remote host closed the connection)
13:54:40*solitudesf joined #nim
13:58:55*i7sDream_ is now known as i7sDream
13:59:59*zyklon quit (Ping timeout: 268 seconds)
14:03:30*zyklon joined #nim
14:41:37*couven92 joined #nim
14:47:51*Perkol joined #nim
14:56:54*nsf joined #nim
15:09:17*Azitrex joined #nim
15:17:58dom96shashlick, reviewed your diff, I did my best to state the changes that were just nit-picks
15:18:05dom96you don't have to fix them but it would be nice
15:18:18dom96also, for choosenim I'd like to get 0.4.0 building cleanly on travis
15:18:39dom96Can you figure out for me where the osinfo conflict comes from?
15:19:18dom96oh, I see. It's in the analytics package
15:19:27dom96I'll fix that tonight
15:19:38dom96also, would be nice to get a more informative error for that from nimble :)
15:19:49*rokups quit (Quit: Connection closed for inactivity)
15:20:54*Perkol quit (Quit: Leaving)
15:34:53*xet7 quit (Ping timeout: 250 seconds)
15:38:04*xet7 joined #nim
15:54:35*literal joined #nim
16:01:06*smitop quit (Quit: Connection closed for inactivity)
16:48:31*neceve quit (Read error: Connection reset by peer)
16:59:13*couven92 quit (Quit: Client Disconnecting)
17:01:42FromDiscord<Zireael> seq[string] isn't equal to seq[cstring]? and I can't force a cstring to string by doing string(st)... btw using javascript target
17:02:55*Snircle quit (Read error: Connection reset by peer)
17:03:10FromDiscord<Zireael> I implemented json serialization on js side and I am trying to load back strings, which got saved as arrays of ints, that I demangled on JS side, and I can't really use them... when I converted all the seqs to cstring I see numbers on screen
17:03:56*Snircle joined #nim
17:05:08dom96You can convert cstring to string
17:05:13dom96Just need to go through each element in the seq
17:06:07FromDiscord<Zireael> how can I force convert a cstring to string? as I said, I tried string(whatever) and it doesn't work
17:06:20FromDiscord<Zireael> and afaik one can't cast in javascript target
17:08:22*jasper joined #nim
17:09:23jasperIf you want a cstring you should be doing cstring(whatever), no?
17:10:04*tefter quit (Remote host closed the connection)
17:20:37*jasper quit (Ping timeout: 256 seconds)
17:23:45shashlickThanks @dom96 I'll review tonight and make fixes
17:24:02shashlickI will remove the imports in the nimble file as part of this pr itself
17:24:23shashlickAlso, had all sorts of trouble building choosenim in osx so some work there too
17:24:29shashlickBut I'll do that separately
17:24:49shashlickThanks for taking the time @dom96 and @Araq, appreciate it
17:32:41*rnrwashere joined #nim
17:32:50*jasper joined #nim
17:33:09FromDiscord<Zireael> nope, I want a string out of a cstring
17:33:29Zevv$ it
17:33:39FromDiscord<Zireael> cstring(whatever) works for making a cstring out of a string, but the other way around, nope
17:33:57Zevvzireal: use the $ operator
17:34:39FromDiscord<Zireael> thanks mate
17:34:43FromDiscord<Zireael> didn't occur to me
17:34:45Zevvhttp://paste.debian.net/1078643/
17:34:52Zevvbeen there :)
17:36:11Zevvlets make toString an alias of `$`, and while we're at it, toJson an alias for `%*` :)
17:36:21Zevvyou can use whichever you want, but searching for it is easier
17:36:55*rnrwashere quit (Remote host closed the connection)
17:49:14*chickendan left #nim ("Left")
17:55:11*jasper quit (Ping timeout: 256 seconds)
18:02:51FromDiscord<demotomohiro> Anyone going to release demos created with Nim at Revision 2019?
18:02:51FromDiscord<demotomohiro> https://2019.revision-party.net/
18:04:32*zyklon quit (Quit: Konversation terminated!)
18:04:42*zyklon joined #nim
18:07:34FromDiscord<demotomohiro> Live stream:
18:07:34FromDiscord<demotomohiro> https://www.twitch.tv/revisionparty
18:10:15*clyybber joined #nim
18:12:33*SenasOzys joined #nim
18:22:21*luis_ joined #nim
18:26:16*Jesin quit (Quit: Leaving)
18:28:03*Jesin joined #nim
18:39:01AraqNah, read a basic tutorial instead
18:39:12Araqis there a tutorial that does NOT cover $ ?
18:41:48*luis_ quit (Ping timeout: 268 seconds)
18:46:28*beatmox joined #nim
18:53:17*kapil____ quit (Quit: Connection closed for inactivity)
18:57:19FromDiscord<!!Liam is Unlucky> hi, been trying to get a test program to compile, never done nim before, anyone know what I need to get this working
18:57:19FromDiscord<!!Liam is Unlucky> https://cdn.discordapp.com/attachments/371759389889003532/569235155416514561/Capture.PNG
18:59:27FromGitter<Grabli66> Hi! Is there a way to call proc by it's name? Something like a reflection.
19:08:19*SenasOzys quit (Ping timeout: 268 seconds)
19:10:36Araq!!Liam is Unlucky: use the mingw that ships with Nim.
19:11:04AraqGrabli66: no, the general approach is to write a macro that fills a name->proc lookup table
19:11:23Araqthere are libraries available that do that but it's also not hard to roll your own
19:16:35*clyybber quit (Quit: WeeChat 2.4)
19:19:02planetis[m]Araq: hello, I have a question, in your post about "quirky exceptions" you said every function call maps to a function call, is that true even if its inside a try block? Also its broken on devel.
19:20:45Araqit's a blog post.
19:21:15Araqit is all true. I write so little because my articles don't have mistakes.
19:21:50Araqwhat do you mean "broken on devel"? it's not supported.
19:22:01Araqit only got merged to keep it from bitrotting
19:22:04planetis[m]okay my bad :p
19:23:14planetis[m]i mean when trying out the nimquirky branch ;) I get: /home/markedone/build/Nim/lib/system/fatal.nim(23, 5) Error: undeclared identifier: 'cstderr' when compiling a program, used to work.
19:23:41*luis_ joined #nim
19:25:02*planetis[m] sent a long message: < https://matrix.org/_matrix/media/v1/download/matrix.org/jaZhkTaGvNVCRCifzVhCeJiD >
19:30:07Araqyeah, use 'devel' instead
19:33:06*Jesin quit (Quit: Leaving)
19:33:06*luis_ quit (Read error: Connection reset by peer)
19:36:38*Jesin joined #nim
19:39:00*rnrwashere joined #nim
19:39:08*rnrwashere quit (Client Quit)
19:40:28planetis[m]seems it's unsupported, good night to y'all, see ya
19:43:17Araqbut it is in devel
19:43:38Araqand has a chance of working
19:44:16*rockcavera quit (Remote host closed the connection)
19:47:19planetis[m]araq: sorry i made a mistake in my message. yes i'm on devel and thats the error I'm getting: Nim/lib/system/fatal.nim(23, 5) Error: undeclared identifier: 'cstderr'
19:48:52*jjido joined #nim
19:50:23planetis[m]anyway good night
19:51:59Araqso strange but I can reproduce
19:52:17*Jesin quit (Quit: Leaving)
19:56:31*Jesin joined #nim
20:09:52*uvegbot joined #nim
20:10:59*zyklon quit (Ping timeout: 246 seconds)
20:11:50*abm joined #nim
20:24:24*stefanos82 quit (Remote host closed the connection)
20:32:43*luis_ joined #nim
20:57:19*luis_ quit (Ping timeout: 246 seconds)
20:58:38*Azitrex quit (Quit: Bye for now)
21:13:43*vlad1777d quit (Ping timeout: 244 seconds)
21:16:40*jjido quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
21:30:08solitudesfon devel path separators in compiler output are instead `7`: `CC: _7lib7packages7docutils7rst`
21:30:27solitudesfanyone else has this problem?
21:30:48*Vladar quit (Remote host closed the connection)
21:31:43Zevvyeah haha
21:39:36*solitudesf quit (Ping timeout: 255 seconds)
21:44:32shashlick@leorize: just tried the nim command you mentioned and there's an fflush defined in libc.so.6
21:45:21shashlicknm -D | grep fflush exists but there's no fflush@GLIBC
21:46:16*jjido joined #nim
21:46:58Zevvsolitudesf: the 7s seem to be on purpose, Araq did this in 0121dda9
21:51:22*mal`` quit (Quit: Leaving)
22:00:52*mal`` joined #nim
22:02:36*abm quit (Read error: Connection reset by peer)
22:03:02*abm joined #nim
22:17:46*jjido quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
22:21:52*nsf quit (Quit: WeeChat 2.4)
22:52:50*vlad1777d joined #nim
23:37:55*lritter quit (Quit: Leaving)
23:43:35*Trustable quit (Remote host closed the connection)
23:52:40*vlad1777d quit (Ping timeout: 246 seconds)