<< 24-05-2019 >>
00:00:46FromDiscord_<nothing to no one> ah, completely missed it. thanks!
00:42:10*Snircle quit (Quit: Textual IRC Client: www.textualapp.com)
00:42:45*Snircle joined #nim
00:52:15FromDiscord_<nothing to no one> is the current working directory thread-local, or global to the program?
00:53:10*a_b_m quit (Quit: Leaving)
01:01:48*kapilp joined #nim
01:03:24FromDiscord_<nothing to no one> nvm, not using it anymore
01:16:34*couven92 quit (Read error: Connection reset by peer)
01:47:42*dddddd quit (Remote host closed the connection)
01:55:54*rnrwashere quit (Remote host closed the connection)
02:25:48*banc joined #nim
03:08:30*rnrwashere joined #nim
03:09:19*theelous3 quit (Ping timeout: 248 seconds)
03:16:37FromGitter<Varriount> The CWD is global
03:17:25*jasper quit (Ping timeout: 256 seconds)
04:26:34*nsf joined #nim
04:27:18FromGitter<kaushalmodi> Is anyone familiar with the Python Marshal format?
04:28:48FromGitter<kaushalmodi> I'm reading up on how to easy parse perforce output and seems like its "-G" switch is the best way, but that outputs data in Python's binary marshal format. Ref: https://nedbatchelder.com/blog/200402/another_cool_thing_about_perforce_python.html
04:30:00FromGitter<kaushalmodi> This is me doing some night time reading, to plan out how to tackle p4 (perforce)'s -G marshal output in Nim tomorrow
04:45:23FromGitter<Varriount> @kaushalmodi The Python documentation should have an explanation.
04:48:01FromGitter<Varriount> Failing that, I would bet on the source code of the marshal module
04:50:22*leorize quit (Remote host closed the connection)
04:50:48*leorize joined #nim
04:58:20*Gaboose joined #nim
05:01:42GabooseHello friends. Newbie here. I'm intrigued by the "Easy C, C++ and Objective C wrapping" feature of nim. Does it mean it can import C++ files and then compile them to a different target language?
05:01:44Gabooselike javascript?
05:03:49leorizeGaboose: nope :p
05:04:16leorizeif it's that magical we won't need emscripten
05:04:51Gaboosei see, ok, misunderstood a little bit :)
05:05:17Gaboosecompilation to several languages is cool though
05:19:53*kapilp quit (Quit: Connection closed for inactivity)
05:29:24*vlad1777d joined #nim
05:37:47*narimiran joined #nim
05:40:39*solitudesf joined #nim
06:04:40*pyotr joined #nim
06:05:30pyotr😁️
06:08:22pyotrSomebody there? Don't know how to use this?
06:08:48pyotrSome Nim advocate?
06:09:23pyotr😴️ I know I know is time to sleep.
06:12:08*pyotr quit (Quit: pyotr)
06:14:45Zevvtell me
06:14:56Zevvor tell the channel, actually
06:16:22*solitudesf quit (Ping timeout: 272 seconds)
06:34:46*enthus1ast quit (Ping timeout: 259 seconds)
06:36:24*PMunch joined #nim
06:47:51*enthus1ast joined #nim
07:00:00*gmpreussner quit (Quit: kthxbye)
07:03:11*rnrwashere quit (Remote host closed the connection)
07:03:47*rnrwashere joined #nim
07:04:51*gmpreussner joined #nim
07:05:34*rokups joined #nim
07:08:08*rnrwashere quit (Ping timeout: 252 seconds)
07:08:56*krux02 joined #nim
07:22:28*neceve joined #nim
07:23:04*Vladar joined #nim
07:34:03*noonien quit (Read error: Connection reset by peer)
07:34:19*dddddd joined #nim
07:38:23*LyndsySimon quit (Ping timeout: 252 seconds)
07:39:40PMunchHmm, I'm looking at the code for DES encryption. It has an encrypt procedure that takes a `var openArray[byte]` to store the result in. The way I use it is to run it three times with a seq[byte].len == 8 and concatenate the three results into a single seq[byte].len == 24. Is there a way to create one seq[byte].len == 24 and pass slices of that sequence to the openArray parameter?
07:39:41*rokups quit (Ping timeout: 250 seconds)
07:40:00PMunchSimilar to how you would pass a pointer to the middle of an array in C to have it populate only that slice
07:40:58leorizeseq[byte][0..2]?
07:42:44PMunchWell that would create a copy wouldn't it?
07:43:08*rokups joined #nim
07:43:32*LyndsySimon joined #nim
07:43:42*solitudesf joined #nim
07:44:18Araqyup
07:44:24PMunchYeah: http://ix.io/1JUU/
07:45:41*noonien joined #nim
07:58:02PMunchhttp://ix.io/1JUW/Nim
07:58:31PMunchHmm, tried something that I thought would work (albeit being a bit hacky), but I get strange SIGSEGVs
08:00:20PMunchWith the echos in place it prints an empty line and an h (by the way, would it be good if Nim printed escape characters instead of printing things like 0 as nothing?)
08:02:18*stefanos82 joined #nim
08:03:07*dvn quit (Ping timeout: 248 seconds)
08:06:33*noonien quit (Ping timeout: 250 seconds)
08:06:42*rnrwashere joined #nim
08:06:59*LyndsySimon quit (Ping timeout: 250 seconds)
08:07:33PMunchHmm, trying to get GCC to add debug symbols for the C code that Nim spits out -ggdb doesn't seem to work
08:07:33*noonien joined #nim
08:07:54Araqit's -g
08:07:54PMunchCompiled the Nim code with -d:release to avoid it adding it's own debugging information
08:08:01PMunchTried that as well
08:08:15Araq--linedir:off --debuginfo
08:08:25Araqfor when you want to debug the C code
08:08:50PMunchAha
08:08:51PMunchNice
08:09:03*LyndsySimon joined #nim
08:09:49Araqalso good: GC_disable() and run it through Valgrind
08:10:28PMunchI just wanted to see what actually SIGSEGVed
08:10:59PMunchBecause with the Nim code debugging enabled it said it occured in code that was before the echo statements that had been executed
08:11:21PMunchOh wait..
08:11:27*rnrwashere quit (Ping timeout: 264 seconds)
08:11:32PMunchI'm an idiot.. I call populate twice..
08:12:15narimiraneven just the second one sigsegv's
08:12:54narimiranwell, the first one was one that printed stuff
08:13:16*ShalokShalom joined #nim
08:13:53PMunchEureka: https://play.nim-lang.org/index.html?ix=1JV4
08:14:02*dvn joined #nim
08:14:35PMunchnarimiran, yeah that was the issue. I was seeing the echos from the first `populate` call, which confused me, I think I need more cofffee
08:14:52livcdPMunch: clicking on playground should redirect you to playground not nim-lang.org
08:15:27PMunchHmm, good point. It was originally just the logo
08:15:58narimiranPMunch: argh, `addr`, you're not the only one who hasn't even noticed it that it is missing ;)
08:16:33PMunchlivcd, In newfrontend it doesn't link to anything: play.nim-lang.org/newfrontend/
08:16:57PMunchnarimiran, yeah I forgot that array was a pointer for a sec
08:16:58narimiranlink would be nice/expected
08:17:21PMunchYeah I'll fix it once I actually deploy the newfrontend
08:17:44PMunchWhich will be once Karax builds on an actual release and not just devel
08:23:20PMunchWhich will be once Karax builds on an actual release and not just devel
08:23:23PMunchWoops
08:23:44PMunchHmm, is there a way to save a cast?
08:23:49PMunchI guess I could use a template
08:27:01*solitudesf quit (Ping timeout: 246 seconds)
08:31:16narimiranPMunch: how about `populate toOpenArray(test, 1, 3)`?
08:33:30PMunchPfft, that would be easy!
08:33:40PMunchBut yeah, definitely a better solution :P
08:35:37livcdbtw what is the point of the "stable" release ?
08:37:14PMunchWell, I didn't want to rely on a random point on the devel branch
08:37:18PMunchIt case it had some weird bug
08:37:57PMunchnarimiran, that works like a charm, thanks :) It felt too wrong to do it my way :P
08:38:06livcdi have a feeling that random point on the devel is more stable than stable
08:38:46PMunchHaha, might be
08:39:11PMunchBut I kinda wanted other people to be able to deploy the playground easily
08:39:40PMunchAnd make Nim appear more stable if I put an article on HN or Reddit
08:45:16*seni joined #nim
08:45:39*hoijui joined #nim
08:47:58*kapilp joined #nim
08:49:38*Tanger left #nim ("Leaving")
08:49:58poopBotyglukhov[m], oh master of masters can you pls assist me with issue with nim-jwt
08:50:48leorizenarimiran: I think I fixed that indent problem with pragmas :)
08:51:05narimiranleorize: thumbs up!
08:51:13leorizewait a sec for me to push this
08:51:48leorizenow that I got a test suite, this should not cause too much problem
08:53:14narimiran...famous last words
08:54:19leorizeone day I'll have to rewrite the indent script :p
08:54:30leorizeanw, got that pushed :)
08:54:38leorizeplease wait for travis
08:56:20leorizenarimiran: you can pull now :)
09:02:34poopBotyglukhov[m], stop playing fonline, i need you :), will give you sneekboy+waki for it
09:03:06leorizeI think he's on gitter more than matrix
09:03:09leorizeso @yglukhov
09:05:15poopBot@yglukhov : can you check your jwt-nim pls i duno if i am retarded or if its buged tryed on 2 pcs
09:20:13*hoijui quit (Ping timeout: 250 seconds)
09:22:00narimiranpoopBot: based on the previous experience with your bug reports here......... :P :D :D
09:22:28narimiranleorize: did you get my PM yesterday about object fields highlighting?
09:22:36poopBotnp i can alwies create a new account xD
09:23:07narimirani'm sure nobody would recognize you :D :D
09:23:52*al2 joined #nim
09:24:41poopBotanyway i am like 65% sure its not my foult this time
09:25:03narimiranhahaha, improving the odds, i like it :)
09:26:29leorizenarimiran: nope
09:27:04leorizecan you send it again? :)
09:27:24poopBothttp://ix.io/1JVm
09:31:43poopBothmm it wont even compile duno why i thinked it copiled yesterday
09:33:14poopBotso when i use this ix.io it owerites old cuz i got same link or am just blind
09:33:32narimiranPMunch: i didn't even know that `toOpenArray` exists and can do that, i found out about it when i couldn't implicitly convert UncheckedArray to OpenArray trying to make your example more general :)
09:36:46PMunchHaha, yeah I've never seen it either
09:36:49PMunchBut it's a nice trick
09:38:19*al2 quit (Ping timeout: 258 seconds)
09:38:35poopBotduno if i should try wrap https://github.com/pokowaka/jwt-cpp then
09:40:02Zevvhm wait I probably could not use openarray to cheaply slice a string, right?
09:40:53*hoijui joined #nim
09:44:09narimiranZevv: ummm, it seems to work?
09:45:16Zevvah its basically just passing the pointer, but safer then doing it with addr. nice
09:47:09narimirannow two questions remain: 1. why nobody knows about it? 2. why it isn't called `slice`? :)
09:49:03Zevvwell, because it isn't?
09:49:10Zevvif you slice a string, you should get a string
09:49:16Zevvnot an array of chars
09:53:21*hoijui quit (Quit: Leaving)
09:53:36AraqtoOpenArray is new in 0.19 and we don't make good use for it, yet
09:58:19Zevvwhat's the typical place it should be used then
09:59:08Zevvand I guess it's not for C backends only, so we can't just sprinkle it through the stdlib
09:59:28Zevv /it's for C backetds only/
10:00:44Zevvoooh, and also: "syntax sugar yet to come" \o/
10:01:54AraqPMunch, karax should work with latest stable, I'll insert some workaround
10:08:19skrylar[m]well. i nim-ized the appkit (though its not implemented on any platform yet)
10:13:17skrylar[m]https://pastebin.com/LDj097RJ a very rough dump
10:18:30*rnrwashere joined #nim
10:22:51*rnrwashere quit (Ping timeout: 248 seconds)
10:28:19*poopBot quit (Remote host closed the connection)
10:38:48*ng0 joined #nim
10:40:13Araqskrylar[m], any opinion on "immediate mode UIs"?
10:44:22PMunchZevv, I guess the use for it is exactly what I needed to use it for here. So basically pass a part of a sequence or array as an openArray for direct manipulation.
10:45:03federico3https://nim-lang.github.io/Nim/system.html#UncheckedArray is not documented: what's its use-case?
10:46:17Araqfederico3, C interop mostly, use it with a 'ptr' in front, 'ptr UncheckedArray[int]'
10:46:51federico3there are comments in system.nim but they are not being rendered in the manual it seems
10:50:42skrylar[m]<Araq "skrylar, any opinion on "immedia"> i've never bought in to them. they seem to be some nontrivial work to make efficient, difficult to integrate with other things (ex. accessibility services) .. they seem to be great for debugging systems
10:51:35skrylar[m]i mean the Be stuff lets you do it if you want; you can just have a window process draws/events manually and just not use widgets
10:55:40*al2 joined #nim
10:58:29Araqwell but that's not good enough, you can have immediate mode plus widgets that are written for immediate mode
11:00:14FromGitter<alehander42> Araq, is this easy to do
11:00:21FromGitter<alehander42> i have a OS todolist
11:00:30FromGitter<alehander42> and UI is the only missing thing
11:01:03FromGitter<alehander42> trust me, i want the most hacky thing, the os would run everything on physical memory in kernel mode
11:02:01PMunchalehander42, actual GUIs? Or a TUI?
11:03:55FromGitter<alehander42> hm i needed some special stuff to be able to bootstrap a gui
11:03:55FromGitter<alehander42> right
11:04:34FromGitter<alehander42> i need to set the video resolution, and real multitasking
11:04:47FromGitter<alehander42> so probably for the beginning a TUI indeed
11:04:54Zevvyou could do without the multitasking
11:05:15PMunchfederico3, Araq, is UncheckedArray any different from the old array[0..int.high, int] {.unchecked.}?
11:05:21FromGitter<alehander42> hm nice
11:05:30FromGitter<alehander42> but i do need multitasking overally
11:05:46FromGitter<alehander42> basically my idea is simple: define a very simple typed language
11:05:48PMunchBut writing schedulers are the fun part :P
11:05:51FromGitter<alehander42> run everything without paging
11:06:11FromGitter<alehander42> but ensure somehow from the types that each program touches only valid part of the memory
11:06:29FromGitter<alehander42> and yeah write a very basic memory manager & scheduler
11:06:36FromGitter<alehander42> preemptive scheduler!
11:07:01Zevveven that is hard enough, given that you need context switching
11:07:35FromGitter<alehander42> well i do: but i thought it's not so hard
11:07:41FromGitter<alehander42> i remember looking at one toy os
11:07:45FromGitter<alehander42> several years ago
11:07:53FromGitter<alehander42> which had a not very complicated impl of context switching
11:07:56Zevvon toy CPU's it is pretty simple
11:08:08FromGitter<alehander42> i need to preserve/copy some stuff and move some stuff on each switch i think
11:08:32Zevvand create some new stuff
11:08:38FromGitter<alehander42> yes
11:08:43FromGitter<alehander42> so what would be the simplest "real" semi-modern CPU
11:08:49FromGitter<alehander42> for this kind of thing
11:09:05Zevvsome arm cortex
11:09:23skrylar[m]theres those blaze-things for fpgas
11:09:34skrylar[m]those have variable complexity iirc
11:10:21FromGitter<yglukhov> poopBot: hey whats up
11:11:10FromGitter<alehander42> yeah but i probably would run it on qemu
11:11:15FromGitter<alehander42> and maybe on an old laptop or something
11:11:32FromGitter<kaushalmodi> federico3: I use UncheckedArray to map to array pointers in C: `let uncheckedArrPtr = castptr UncheckedArray[cint (alloc0(arrSize))`
11:11:46*seni quit (Quit: Leaving)
11:12:49Zevvalehander42: STM32 is a nice platform, and has full system qemu support
11:13:25FromGitter<alehander42> yeah, but i'd love to run it on a normalish machine
11:13:32FromGitter<alehander42> what i mean is
11:13:34FromGitter<alehander42> hm
11:13:38PMunchWas the settings block concept removed from Jester?
11:13:41FromGitter<alehander42> but i can connect stm32 to a monitor, right
11:13:51FromGitter<alehander42> so in this case i can use it for coding actually
11:14:17Zevvalehander42 -> #nim-offtopic
11:14:39federico3PMunch: use https://github.com/dom96/jester#custom-router , it's much more flexible
11:14:41FromGitter<alehander42> yes
11:14:43PMunchalehander42, we used bochs in our OS course in uni
11:15:02PMunchbochs and some random lab machines
11:15:28FromGitter<alehander42> afaik qemu is pretty good as well
11:15:33FromGitter<alehander42> for development
11:16:18PMunchfederico3, hmm
11:16:49PMunchMy old design for custom routes still supported the simple case of a settings block I think..
11:16:56PMunchI just want to change the public dir
11:17:02PMunchFor debugging
11:17:55*solitudesf joined #nim
11:19:09*lritter joined #nim
11:21:06*al2 quit (Quit: WeeChat 2.4)
11:21:57AraqPMunch, it's different in that it clearly communicates that you don't have bounds checking
11:22:18Araqand it can also be the last field of an object, then to be used without the 'ptr'
11:22:36Araqso that you can reimplement your own 'seq', for example
11:23:08PMunchOooh interesting
11:23:47Araqworks well with the custom destructors
11:24:13*laaron quit (Remote host closed the connection)
11:24:55PMunchCool, looking forward to see what kind of crazy stuff will come of this
11:25:14PMunchNative feeling custom types could be pretty neat :)
11:26:01skrylar[m]something interesting but probably not as relevant, i was thinking about how the bmessage system could be tied in to adapters... so you could concievably have stuff like "do this libcurl thing and then throw the http response back as a message"
11:26:09*laaron joined #nim
11:26:34skrylar[m]they were basically using json objects for system wide messages before that was a thing
11:26:43PMunchSomething completely different, bit annoying that you can't do "initSomething().doSomething()" where doSomething expects a "var Something". You have to assign initSomething to a var first, even though I obviously don't care about the value in this case.
11:27:32skrylar[m]i guess because the system doesn't recognize temporary storage as an actual variable
11:28:16AraqPMunch, you brought this up before and I disagree
11:28:46AraqdoSomething() mutates and the result of the mutation can be important and so shouldn't operate on throw-away data
11:33:37Araqtemplate combine(a, b) = (var x = a; b(x)); initSomething().combine(doSomething)
11:36:07PMunchWell in this case for example it was initJester(myRouter, settings = newSettings(...)).serve()
11:36:42PMunchServe in this case will run forever, so there is no possibility of me changing that variable anyways.
11:36:55PMunchWell, I guess it would be possible, but that's beside the point
11:42:17*laaron quit (Remote host closed the connection)
11:43:28*PMunch wonders, with IncludeOS, could you make a pure Nim docker container?
11:45:22*laaron joined #nim
11:47:55*leorize quit (Ping timeout: 256 seconds)
11:48:47*joshbaptiste quit (Ping timeout: 245 seconds)
11:49:58*leorize joined #nim
11:52:25*joshbaptiste joined #nim
11:53:35federico3http://paste.debian.net/1082880/ some help with this?
11:54:26federico3It's being called as a callback from a dynamic library in C
11:59:19PMunchI'm guessing x is being garbage collected?
11:59:38FromGitter<kaushalmodi> I haven't used a "var" ptr before
11:59:57FromGitter<kaushalmodi> I'd keep the return ptr of ptr immutable, and directly assign the resp
12:00:48FromGitter<kaushalmodi> Not at my computer, but try something like:
12:00:57*couven92 joined #nim
12:01:44federico3PMunch: yes, the error happens when the callback exits
12:02:00FromGitter<kaushalmodi> I'd rename the resp arg to respPtrPtr for clarity. Then,
12:02:30FromGitter<kaushalmodi> let respPtr = respPtrPtr[]
12:02:47PMunchI guess just resp = cast[ptr ptr pam_response](alloc0(sizeof(pointer))); resp[] = cast[ptr pam_response](alloc0(sizeof(pam_response)))
12:02:55FromGitter<kaushalmodi> Then assign directly to `respPtr[]`
12:03:14FromGitter<kaushalmodi> That way the respPtrPtr arg doesn't need to be a var
12:03:36PMunchx in your example probably lives on the stack, so when you get the addr to it and return that pointer will be to some unused part of the stack and cause an error
12:03:59federico3ah you first allocate a pointer and then build pam_response!
12:04:37PMunchYes, otherwise the pointer to the pointer to pam_response won't be on the heap, and therefore not available to the caller
12:05:00federico3resp = cast[ptr ptr pam_response](alloc0(sizeof(pointer))) alone triggers the invalid pointer error
12:06:32FromGitter<kaushalmodi> Something like this (https://github.com/kaushalmodi/nim-systemverilog-dpic/blob/master/amiq_dpi_c_examples/libdpi.nim#L105), but for the second ptr dereferencing too
12:08:15*poopBot joined #nim
12:09:02poopBotso treform helpd me out , i head dumbest issue --d:ssl--passL:"-lcrypto" dident pass this to compiler, thats why nim-jwt dident work
12:10:33FromGitter<yglukhov> poopBot: so.. is it solved now?
12:10:42federico3kaushalmodi resp is not initialized. I cannot set resp[] straight away
12:10:45poopBotyep :)
12:11:23poopBotshould put that on github page for newbs who dont know that need to do that :)
12:12:45leorizePMunch: the settings block for jester is still there
12:12:57leorizeyou just have to build the doc and read it :p
12:13:01FromGitter<kaushalmodi> federico3: ah, understood.. in my case the C side allocates stuff before sending me the pointer
12:13:24FromGitter<Vindaar> @poopBot good it's working now. But why do you manually have to pass `lcrypto` to the C compiler?
12:13:40poopBotthat part i duno
12:14:16poopBotits like asking a cleaning crew what is the molecular mass of detergent xD
12:14:27FromGitter<Vindaar> @yglukhov shouldn't that be done in the library?
12:14:45poopBotcleaning crew would be me in that elaboration
12:16:00FromGitter<yglukhov> does it work with only `-d:ssl` and without `-lcrypto`?
12:16:02PMunchleorize, it is? Couldn't find it..
12:16:18FromGitter<Vindaar> on my machine `-d:ssl` is not required, only `-lcrypto`
12:16:21poopBotcan try
12:16:49FromGitter<Vindaar> (and only `-d:ssl` does not work)
12:16:53poopBotsame on my
12:17:00poopBotjust tested
12:17:11federico3PMunch: any other hint?
12:17:22*laaron quit (Remote host closed the connection)
12:17:32leorizePMunch: https://github.com/dom96/jester/blob/master/jester.nim#L1358
12:17:36PMunchJust a sec, looking into it
12:17:43poopBoti tryed before tht LD_PREALODING tought tht dident work :)
12:17:46leorizejust put the block before you use `router`
12:18:32leorize-d:ssl is only usable for the httpclient module iirc
12:18:48leorizenowadays it won't pass -lcrypto randomly
12:19:33PMunchfederico3, this seems to work fine: http://ix.io/1JWd
12:20:11PMunchleorize, strange, I searched through that file for the word setting..
12:20:19PMunchMust've mis-typed it
12:20:52PMunchfederico3, what does the C library look like?
12:20:59PMunchOr specifically the part that calls your code
12:21:08*laaron joined #nim
12:21:43PMunchTypically when you get passed a ptr ptr it's because you are supposed to change the pointer that gets pointed to
12:21:48FromDiscord_<kiwi> @federico3, not sure if you saw my question yesterday ... but I was wondering if you can use CSS or something to change button colors. Original: ... can you change the background colour of the UI & the background colour of the buttons in ui.nim / libui? Like this .... https://i.imgur.com/4DHBmQn.png
12:22:13federico3kiwi: wrong person :)
12:22:34FromDiscord_<kiwi> oops. sorry.
12:22:50PMunchI'm assuming it was me you tried to talk to?
12:22:53PMunchAbout wxNim?
12:24:03PMunchkiwi^
12:24:05FromDiscord_<kiwi> Not really. After working through a lot of wxnim stuff & writing a basic gui I concluded that I couldn't do it.
12:24:48FromDiscord_<kiwi> I was wondering about nim's libui libary and whether it would pass css to control the button colour on gtk3 buttons or not. https://github.com/nim-lang/ui
12:25:48PMunchAh, I know the Gtk wrapper can do it, but I'm not all that familiar with libui
12:26:41FromDiscord_<kiwi> I understand wxnim deciding to stay true to platform colours ... but I need to sort of flexibility that you get with qt ... but don't really want to try to write an interface. I couldn't tell with libui either ... basically QT packages the CSS a little but its basically doing it like gtk.
12:27:38PMunchDoesn't look like libui supports it: https://github.com/andlabs/libui/issues/313
12:28:00PMunchwxWidgets is able to set button colours
12:28:27PMunchJust not with a stylesheet
12:28:31FromDiscord_<kiwi> That seemed to involve creating some new button objects not standard buttons.
12:28:58FromDiscord_<kiwi> Or wasn't it as difficult as I suspected?
12:29:18WilhelmVonWeinerif I run len() on an array is that done at runtime
12:29:44livcdkiwi: try sciter
12:29:47WilhelmVonWeinershould I just store obj.size in my object? the array size won't change
12:30:08FromDiscord_<kiwi> thanks livcd
12:31:31FromDiscord_<kiwi> funny thing about colour .. i used to buy vdus that were green on black & didn't believe in colours ... then out came windows & mac stole an interface from xerox ... and here we are in a world where colours provide contrast with lots of grey buttons. Bizarre circle.
12:32:27FromDiscord_<kiwi> however, i shall investigate sciter 😃
12:40:23PMunchkiwi, hmm I'm able to set a buttons foreground (ie. text) colour with setForegroundColour
12:40:33PMunchBut the background seems a bit trickier..
12:41:40leorizeWilhelmVonWeiner: array length are compile-time
12:42:11leorizeopenArray length are runtime, but they're stored with the array
12:42:16leorizesame goes for seq
12:44:55Araqgah, Nim's nim.cfg is complex stuff
12:48:42FromGitter<mratsim> array length is not stored at runtime
12:48:44*laaron quit (Remote host closed the connection)
12:48:55FromGitter<mratsim> it's inlined everywhere needed
12:50:48*laaron joined #nim
12:53:36PMunchkiwi, the thing with GUIs are that you have three things to balanec: platforms, customisability, and complexity. Gtk runs great on many platforms, but it's only really "native" to Linux, you can customize pretty much everything, and it's not too hard to work with. WxWidgets uses only native components, is fairly easy to work with (once you know what you're doing), but customization is difficult since it has to be possible to do with native components. Things
12:53:36PMunchlike Sciter use no native components, allows you to customize everything, and are often fairly simple to work with. To create something that has native UIs on all platforms, allows complete customization, and is easy to both maintain and work with, now that's a feat!
12:53:44*kapilp quit (Quit: Connection closed for inactivity)
12:55:58FromGitter<mratsim> @Araq: did you see this Github sponsor thing - https://help.github.com/en/articles/about-github-sponsors
13:05:52Araqyeah, will apply
13:06:09Araqanyhow ... this config stuff is terrible
13:06:37AraqI'll hardcode it into extccomp.nim
13:06:42Araqfor VCC
13:06:51Araqany objections? no? ok then.
13:07:29Araqseems to be a big hack around the lack of a decent builtin feature
13:07:53FromDiscord_<kiwi> Thanks PMunch
13:12:09federico3PMunch: this is the simplest code to reproduce the error http://paste.debian.net/1082896/
13:15:13*nsf quit (Quit: WeeChat 2.4)
13:18:32PMunchHmm, can you declare a C callback in a struct like that?
13:19:23*poopBot quit (Remote host closed the connection)
13:20:21*solitudesf- joined #nim
13:20:26FromGitter<mratsim> yes
13:21:19*laaron quit (Remote host closed the connection)
13:22:33*solitudesf quit (Ping timeout: 245 seconds)
13:23:43Araq conv*: proc (num_msg: cint; msg: ptr ptr pam_message; resp: var ptr ptr PAMResponse;
13:23:43Araq appdata_ptr: pointer): cint -->
13:23:48Araq conv*: proc (num_msg: cint; msg: ptr ptr pam_message; resp: var ptr ptr PAMResponse;
13:23:48Araq appdata_ptr: pointer): cint {.cdecl.}
13:23:52*laaron joined #nim
13:24:01Araqtold you to watch out about implicit .closure
13:25:15federico3thanks Araq
13:26:16federico3same behavior with cint {.cdecl.}
13:26:30FromGitter<mratsim> I feel like very often when working with higher order functions we need to force nimcall (or cdecl for C API): https://github.com/nim-lang/Nim/issues/7808
13:27:35*Araq shrugs
13:27:42Araqc2nim does it decently
13:27:58Araqalehander42: any idea what's going on here: https://ci.appveyor.com/project/dom96/nim/builds/24784172/job/jyae0gcn17ykgnyf/tests
13:29:34Araqmy PR affects multi methods, does 'gara' use these?
13:34:42PMunchfederico3, I'm looking into this. It seems there are a couple things going on here
13:36:02narimiran@everybody: i need some (relatively short) code example that nicely shows nim's strengths, any ideas?
13:37:15*rnrwashere joined #nim
13:39:28shashlick@PMunch did you get to try out the radcli wrapper?
13:40:15*Cthalupa- quit (Ping timeout: 264 seconds)
13:41:19*Cthalupa joined #nim
13:41:26FromGitter<alehander42> how can i see the whole
13:41:31FromGitter<alehander42> error? Araq
13:41:38FromGitter<alehander42> as i dont see actual errors in this box
13:41:42PMunchshashlick, there is a radcli wrapper?!
13:41:47*rnrwashere quit (Ping timeout: 248 seconds)
13:43:12*laaron quit (Remote host closed the connection)
13:43:37*laaron joined #nim
13:47:56Araqalehander42: I think you can't but does gara use 'method'?
13:48:15*solitudesf- quit (Ping timeout: 248 seconds)
13:49:49shashlickI made one for you remember
13:50:05FromGitter<mratsim> @narimiran: echo "Hello world"
13:50:14PMunchshashlick, you did?
13:50:16FromGitter<mratsim> ask Rust to do the same ;)
13:50:27PMunchI think I used the one I generated with c2nim
13:50:33FromGitter<mratsim> println!("Hello world") ;)
13:50:42PMunchAnyways it works great :) Got MsCHAPv1 working and everything
13:50:55narimiran@mratsim :P
13:51:45shashlick@PMunch https://irclogs.nim-lang.org/22-05-2019.html#14:46:26
13:51:56ZevvPMunch: welcome to 1998!
13:52:01narimiransomething like the current example at https://nim-lang.org/ but it can be longer and show something not-so-obvious
13:52:19Zevvtrumpet winsock calling, they're asking you to startup your mosaic browser
13:53:10PMunchnarimiran, https://uploads.peterme.net/backside-jester-fullres.png https://uploads.peterme.net/backside-gol-fullres.png
13:53:25PMunchMe and dom96 made those for FOSDEM last year. Ended up not using them, but eh
13:53:29PMunchMight be helpful
13:53:46PMunchnarimiran, I think throwing a template in there would be nice
13:53:51narimiranPMunch: yep, i'm looking for stuff like that
13:53:56PMunchAt least give them a taste of metaprogramming :)
13:53:59narimiranand a template example would be nice too
13:54:08FromGitter<kaushalmodi> looking at that screenshot, are we moving to using `std/` prefix for stdlib modules? I kinda like it.
13:54:22FromGitter<kaushalmodi> helps me quickly distinguish the stdlib vs other imports in my code
13:54:55narimirankaushalmodi: i just use two lines of imports - first for stdlib, second for the rest
13:55:15PMunchshashlick, oh woops, I had left work half an hour before you posted that :(
13:55:40FromGitter<kaushalmodi> narimiran: I think it helps a Nim newcomer see how much stuff is already coming from Nim stdlibs
13:56:09PMunchfederico3, I'm a bit stumped by this. Found some example that seems to suggest that it's not a pointer to a pointer but rather a pointer to an array
13:56:16FromGitter<kaushalmodi> I use 3 lines: ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ce7f800d22ba766a2f11b0f]
13:56:26Araqnarimiran, how about we give a nice summary about how Nim *really* works
13:56:31FromGitter<kaushalmodi> line 1 is pkg internal modules, 2 is stdlibs, 3 is nimble installed
13:56:33PMunchBut I'm out of time right now, have to get fetch my car from the repair shop
13:56:59PMunchkaushalmodi, I do something similar :)
13:57:01narimiranAraq: ok, i'll wait.... :P
13:57:09Araqoh nah, that will only scare potential users :P
13:57:13*PMunch quit (Remote host closed the connection)
13:57:55shashlick@PMunch well is an example, as long as you have a wrapper, I'm good
13:59:23FromGitter<kaushalmodi> Varriount: I looked at the marshal source code and my eyes glazed over.
13:59:53FromGitter<kaushalmodi> Everyone here who's great with Python and its internal marshal serialized format: https://irclogs.nim-lang.org/24-05-2019.html#04:27:18
14:00:10FromGitter<kaushalmodi> ^ anyone has tried to decode the marshalled binary blob in Nim?
14:00:34FromGitter<kaushalmodi> Plan B is to call python via exec from Nim
14:00:57Araqwhy not make Python output .json
14:01:26FromGitter<kaushalmodi> Perforce p4 somehow decided to provide just the Python marshal blob
14:01:30FromGitter<kaushalmodi> that's quite odd
14:02:36Araqpython blob2json.py blob
14:05:19*al_ joined #nim
14:06:23FromGitter<MortimerSnerd> Is it a bug that offsetof() won't work with an object that ends with an UncheckedArray, because the type can't be instantiated as a value?
14:08:56FromGitter<kaushalmodi> Araq: yeah, that's the plan.. call a little blob2json Python script from nim
14:11:22*sealmove joined #nim
14:14:24AraqMortimerSnerd: maybe, please report it
14:16:21FromGitter<mratsim> @kaushalmodi Are you talking about pickle?
14:17:52FromGitter<kaushalmodi> no, just marshal
14:18:25FromGitter<kaushalmodi> @mratsim See https://www.perforce.com/manuals/cmdref/Content/CmdRef/global.options.html and look for the "-G" switch in there
14:18:43FromGitter<kaushalmodi> I wonder why the smart Perforce engineers didn't provide a JSON out
14:20:38Araqbtw what is Perforce?
14:21:17FromGitter<kaushalmodi> it's a version control system
14:23:09*rnrwashere joined #nim
14:25:56*theelous3 joined #nim
14:31:13FromGitter<Vindaar> @kaushalmodi if it's really a normal python dictionary marshaled via python, you can just use `nimpy` to parse it: http://ix.io/1JWP/nim
14:32:06FromGitter<kaushalmodi> @Vindaar Let me try that!
14:32:31FromGitter<kaushalmodi> In the meanwhile, I discovered another way which removes Python dependency.. p4 gives out another weird output called "ztag"
14:32:40FromGitter<kaushalmodi> it's format is: ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ce80088d22ba766a2f15d6a]
14:34:28*nsf joined #nim
14:40:24FromGitter<kaushalmodi> how do I do binary read in Nim?
14:42:36FromGitter<kaushalmodi> this is the python script that works with reading the `p4 -G` marshal output.. but doesn't do JSON: http://ix.io/1JWQ
14:42:43FromGitter<kaushalmodi> I'll play with it later today
14:43:52*Gaboose quit (Quit: Page closed)
14:44:26*couven92 quit (Quit: Client disconnecting)
14:46:15FromGitter<Vindaar> I used json, because it's easy to work with in Nim. You could of course iterate over the `PyObject` in Nim and call the `decode` function, too and build the json or whatever you want manually from that too
14:47:09FromGitter<kaushalmodi> I meant that I need to work on fusing the working bit of reading marshal from that python code with the bit about converting to json from your code :)
14:47:53FromGitter<Vindaar> ah, I see
14:48:59FromGitter<kaushalmodi> ugh, why wouldn't the Perforce guys document a Formatting switch.. I just learn from their official blog post: ⏎ ⏎ > Sharp-eyed users may have noticed the addition of a new undoc option to the command line client
14:49:09FromGitter<kaushalmodi> > -F formatstring
14:52:08*rnrwashere quit (Remote host closed the connection)
14:52:40*rnrwashere joined #nim
14:52:59FromGitter<Vindaar> that's certainly one way to spin "sorry for not documenting this feature" :D
14:53:18*Trustable joined #nim
14:54:32*rnrwashere quit (Remote host closed the connection)
14:54:45*rnrwashere joined #nim
14:56:46*rnrwashere quit (Remote host closed the connection)
15:25:52*rnrwashere joined #nim
15:28:29*kapilp joined #nim
15:32:58*vlad1777d quit (Ping timeout: 258 seconds)
15:41:01*gangstacat joined #nim
15:43:05*rnrwashere quit (Remote host closed the connection)
15:50:07federico3TIL: the auto type
15:51:26*solitudesf- joined #nim
15:51:42*rnrwashere joined #nim
15:53:00*solitudesf- quit (Client Quit)
15:53:16*solitudesf joined #nim
15:53:29*nsf quit (Quit: WeeChat 2.4)
16:00:56*laaron quit (Remote host closed the connection)
16:03:43*laaron joined #nim
16:10:50*laaron quit (Remote host closed the connection)
16:12:47*laaron joined #nim
16:26:39*xace quit (Ping timeout: 248 seconds)
16:27:42*Snircle quit (Quit: Textual IRC Client: www.textualapp.com)
16:39:43*laaron quit (Remote host closed the connection)
16:40:09*Jesin quit (Quit: Leaving)
16:43:10*laaron joined #nim
16:45:50*Jesin joined #nim
16:48:09*al_ quit (Quit: al_)
16:48:55Zevvwow, new to me as well. I really feel I should hate this language for its complexity and pleora of things I didn't know and only find out after two years, because that is usually my argument against C++
16:49:22Zevvbut I don't - then what is it?
16:49:43FromGitter<kaushalmodi> may be the community and the direct feedback to the devs here?
16:50:21FromGitter<kaushalmodi> .. and to be able to hack the language in the same language
16:51:26Zevvyeah, true. Bjarne just hangs up on me these days, he's not the guy I once new now he got his superstar status
16:52:17Zevvwe just used to hang out together you know. Chill and code a bit, smoke some weed, have a pizza. But no, mr big no does conferences and gets all the girls. And I'm just here at home, on the couch
16:52:25FromGitter<Vindaar> well, you can skim Nim's manual in less than a day, which covers every feature aside from some experimental ones. How long does it take to read C++'s manual. Oh wait, there isn't even one?
16:53:02Zevvhehe
16:55:06narimiran"i dident read it" :D
16:55:28*rnrwashere quit (Remote host closed the connection)
16:55:46Zevvhoooo
16:57:15*ShalokShalom quit (Remote host closed the connection)
17:01:11*rnrwashere joined #nim
17:03:19*xace joined #nim
17:07:32*laaron quit (Remote host closed the connection)
17:11:08*laaron joined #nim
17:14:26*rnrwashere quit (Remote host closed the connection)
17:15:11*rnrwashere joined #nim
17:20:08*rnrwashere quit (Remote host closed the connection)
17:28:17*rnrwashere joined #nim
17:39:16AraqZevv: do you really know Bjarne?
17:43:10*Vladar quit (Remote host closed the connection)
17:46:51*rnrwashere quit (Remote host closed the connection)
17:53:54*rnrwashere joined #nim
17:54:53Zevvnah :)
17:55:59Araqwhat's complex about the 'auto' type?
17:56:06*abm joined #nim
17:56:25FromGitter<kaushalmodi> Araq: Sorry for bugging with the `--out:` and nim doc issues.. but the URLs with index.html are really much cleaner
17:57:30*rnrwashere quit (Remote host closed the connection)
17:58:21Araqtbh I think I'll remove --out for 'nim doc', --outdir makes sense but everything else is simply to expensive to maintain and support
17:59:02Araqcross references cannot work reliably when you get to chose the .html name
17:59:08Araq*choose
17:59:39Araqyou can have an index.nim to produce an index.html
17:59:45Araqor an index.rst
18:02:20FromGitter<kaushalmodi> cross references uses the .idx .. so you need to just pass the --out: value to the index generation
18:02:47FromGitter<kaushalmodi> index.nim would generate a module named "index" .. for all my modules
18:03:37FromGitter<kaushalmodi> Please don't remove `--out:`.
18:05:48*couven92 joined #nim
18:06:18FromGitter<kaushalmodi> If you remove `--out`, may be the doc HTML default should be to generate `index.html` :)
18:06:55FromGitter<kaushalmodi> then using `--outdir:foo`, I can have `foo/index.html` if I liked
18:07:03Araqand what should be in this index.html? seems like a terrible idea
18:07:16*laaron quit (Remote host closed the connection)
18:07:29FromGitter<kaushalmodi> let me rephrase the example
18:10:06FromGitter<kaushalmodi> now: `nim doc foo.nim` -> `foo.html` -> need to visit this URL as `foo.html` ⏎ after (changing default to `index.html`): `nim doc foo.nim` -> `foo/index.html` -> can visit this URL as `foo/index.html` or `foo/` [outdir defaults to package name for `nim doc`] ⏎ after (changing default to `index.html`) + `--outdir:.`: `nim doc --outdir:. foo.nim` -> `index.html` -> can visit this URL as `/index.html` or `/`
18:10:48*laaron joined #nim
18:11:57FromGitter<kaushalmodi> .. actually that might be a more involved change, as you cannot have `foo/` dir and `foo` binary in the same dir
18:11:58FromGitter<kaushalmodi> oh well
18:16:40FromDiscord_<nothing to no one> How can I check if an element in a sequence is nil or not?
18:17:50*ShalokShalom joined #nim
18:17:56FromDiscord_<nothing to no one> I've tried `thr[0] != nil` but that gives a type error on `!=`
18:19:26FromGitter<kaushalmodi> what's the seq type?
18:19:45narimiran`isnil`
18:20:55FromDiscord_<nothing to no one> seq is of type `seq[Thread[tuple[config: Config, path: string]]]`. isNil throws type checking errors
18:21:45*leorize quit (Ping timeout: 256 seconds)
18:21:52FromDiscord_<nothing to no one> might be an x y problem. I have an array of threads that contains `n` elements, but some elements might be nil, so calling `joinThread` on them causes segfaults
18:23:12*ShalokShalom quit (Remote host closed the connection)
18:23:24Araqa Thread can never be nil
18:23:36Araqwhich is what the compiler is trying to tell you.
18:24:15FromDiscord_<nothing to no one> oh no, i'm gonna have to do this the proper way, aren't i?
18:24:59rayman22201I vote for simplifying doc generation fwiw @kuashalmodi and @araq
18:25:22Araq--project and --outdir go a long way
18:28:44rayman22201+1
18:30:18FromDiscord_<nothing to no one> line 17 here (https://hastebin.com/abotumizij.pl) is segfaulting. I can't really see what I'm doing wrong. am i misunderstanding how threads work?
18:30:37*rnrwashere joined #nim
18:34:47Mister_Magisterhow do you compile in release mode
18:35:02narimiranMister_Magister: nim c -d:release filename.nim
18:35:07*rnrwashere quit (Ping timeout: 248 seconds)
18:35:08Mister_Magisterthanks
18:39:32FromGitter<kaushalmodi> Araq: may be I am not seeing the full picture, but you already set the destFile in Pdoc correctly here: https://github.com/nim-lang/Nim/blob/d67a9f024eeeb2bc26fb38a98be9a53956003290/compiler/docgen.nim#L173-L174
18:39:56FromGitter<kaushalmodi> but to generate the index you use `d.filename` instead of `d.destFile`: let dest = dir / changeFileExt(relativeTo(AbsoluteFile d.filename,
18:40:11FromGitter<kaushalmodi> oops, meant to paste link: https://github.com/nim-lang/Nim/blob/d67a9f024eeeb2bc26fb38a98be9a53956003290/compiler/docgen.nim#L1016
18:43:40FromDiscord_<nothing to no one> oh, I can just check fi the thread is running before joining it, duh
18:48:52narimirankaushalmodi: PR?
18:49:49*leorize joined #nim
18:59:33FromGitter<kaushalmodi> I am trying it out locally but replacing d.filename with d.destFile made no difference :/
18:59:55FromGitter<kaushalmodi> I'll certainly submit a PR if I can fix this
19:00:09Mister_Magisterokay i can compile nim to cpp using nim cpp and compile that in into my cpp but how exactly can i use functions i wrote in nim?
19:01:02*rokups quit (Quit: Connection closed for inactivity)
19:01:13*rnrwashere joined #nim
19:03:50FromGitter<kaushalmodi> Mister_Magister: You should be able to use them if you added {.exportc.} to the Nim proc identifiers
19:04:03Mister_Magisterooh
19:06:36narimiranMister_Magister: https://nim-lang.github.io/Nim/manual.html#foreign-function-interface
19:06:43Mister_Magisterthanks
19:07:40Mister_Magister proc sendMessage(s: string) {.async.} {.exportc.} =
19:07:48Mister_Magisteri guess it's not a proper way of doing it?
19:07:58Mister_Magisterhow do i put 2 of
19:07:59*rnrwashere quit (Remote host closed the connection)
19:08:02Mister_Magister*those*
19:10:08FromGitter<kaushalmodi> `{.async, exportc.}`
19:11:11*laaron quit (Quit: ZNC 1.7.1 - https://znc.in)
19:11:25Mister_Magisterthanks
19:11:44*laaron joined #nim
19:15:20*NimBot joined #nim
19:15:22*a_b_m joined #nim
19:17:55Mister_Magister@kaushalmodi when i add exportc shouldn't nim generate some .h file to include?
19:18:43*abm quit (Ping timeout: 246 seconds)
19:20:05Mister_Magisterexportc doesn't seem to work :/
19:21:49*nsf joined #nim
19:23:26FromGitter<kaushalmodi> afaik Nim does not generate a .h
19:23:47FromGitter<kaushalmodi> how I use is that I compile Nim file via cpp backend to a .so
19:24:32Mister_Magisteroh --header:bot.h is the magic
19:24:46Mister_Magisterand -app:staticLib
19:25:33Mister_Magistertho this header doesn't make much sense either
19:33:49Zevvfederico3: https://github.com/zevv/nim-kernel-module
19:34:35Mister_Magisterno this is completely not working
19:34:40Mister_Magisternot even examples
19:34:53*natrys joined #nim
19:35:44federico3Zevv: I found my attempt at the same and it's pretty similar
19:35:51*sealmove quit (Quit: WeeChat 2.4)
19:36:20Zevvthat's because that's all there is to it
19:36:34Zevvalthough it would be a nice excersise to get the GC working on kmalloc and friends
19:40:08AraqMister_Magister, you cannot export an .async proc, C(++) has no async procs
19:40:54Mister_Magisternot even non async exist. and i can't link it
19:45:31Mister_Magisterye i can't compile it
19:48:41Mister_Magisterlol wait i added some -ldl -ld and it worked
19:49:51Mister_MagisterQuestion. If my nim function returns nim object how do i make variable from it in c
19:50:51Mister_Magisteroh it's int*
20:02:15*demond joined #nim
20:02:29*ShalokShalom joined #nim
20:02:30*demond left #nim ("++++++")
20:04:00*ShalokShalom quit (Remote host closed the connection)
20:04:32*demond joined #nim
20:05:02*demond left #nim ("~~~~~~~~~~~~~~~~~~~~~~~~~~~")
20:05:07Araqmost certainly not.
20:08:19*Jesin quit (Quit: Leaving)
20:09:43*couven92 quit (Read error: Connection reset by peer)
20:11:42FromGitter<kaushalmodi> Araq: woot! https://github.com/nim-lang/Nim/pull/11326/files
20:13:46*Jesin joined #nim
20:16:44Araqand now here is what you need to do:
20:17:00Araq- write a test case that uses your set of options and checks the produced HTML code
20:18:11Araqthat's what most developers seem to not understand. These switches cause on-going costs unless you don't care all that much about quality. Which is kinda what we did in the past, before I really understood the problem.
20:18:11FromGitter<kaushalmodi> The Nim testing infrastructure is too scary to me; I don'
20:18:43FromGitter<kaushalmodi> don't know how to create a test for this; I can provide the reference generated .html and .idx, but where do I commit them ..
20:19:04FromGitter<kaushalmodi> .. let me try grepping at the current test suite for tests similar to the one this needs
20:19:23Araqnimdoc/testproject has a test for the docgen
20:19:45Araqincluding how to setup stuff.
20:20:21Araqand don't get me wrong, I will accept the PR as is as long as the existing tests are still green. But if you want --out to stay, write a decent test for --out :-)
20:20:43Araqsame for the "read from stdin" junk
20:20:50FromGitter<kaushalmodi> found it
20:21:07Araqif it's not covered by tests, it's broken or it'll bitrot until it's broken
20:21:29Araqsame for our lovely Visual C++ support. :-)
20:21:36FromGitter<kaushalmodi> thanks; should I just add another execShellCmd in `if execShellCmd("nim doc --project --index:on -o:$1/htmldocs $1/testproject.nim" % dir) != 0:`?
20:21:46Araqthat's one way
20:21:47FromGitter<kaushalmodi> or just add `--out:` there?
20:22:33FromGitter<kaushalmodi> hmm, `-o` is already there ..
20:23:17FromGitter<kaushalmodi> let me just add another little testproject in there
20:23:27FromGitter<kaushalmodi> that way I won't have to touch the --project testing
20:26:49*kapilp quit (Quit: Connection closed for inactivity)
20:31:58Araqsounds good
20:34:04*a_b_m quit (Quit: Leaving)
20:35:00*nsf quit (Quit: WeeChat 2.4)
20:41:43*lritter quit (Ping timeout: 245 seconds)
20:48:10*pyotr joined #nim
20:51:52*pyotr quit (Client Quit)
20:52:07*luis_ joined #nim
20:53:50*rnrwashere joined #nim
20:55:45*rnrwashere quit (Remote host closed the connection)
20:57:14*rnrwashere joined #nim
21:01:21luis_Hi everyone, is there language server protocol support for nim currently?
21:01:53narimiranluis_: https://github.com/PMunch/nimlsp
21:02:49luis_this is awesome! Thank you!!
21:05:45FromGitter<kaushalmodi> Araq: Turns out that that PR of mine causes a regression :( But on the positive side, I refactored the nim doc tester to make it extensible
21:13:08*arecaceae quit (Remote host closed the connection)
21:13:27*arecaceae joined #nim
21:15:22FromGitter<Vindaar> Just played around calling Nim from C++. Sort of expected that handing a variant kind to C++ wouldn't work, but it does
21:21:24*luis_ quit (Ping timeout: 272 seconds)
21:29:47*narimiran quit (Ping timeout: 268 seconds)
21:39:54FromGitter<iffy> What's the koch command I can use to build a version of nim without stomping on my already-built version?
21:40:03FromGitter<alehander42> temp?
21:40:39FromGitter<iffy> aha, yes
21:40:41FromGitter<iffy> thank you!
21:40:46FromGitter<alehander42> not sure i dont use it
21:40:50FromGitter<alehander42> but hope its this one!
21:55:10*solitudesf quit (Ping timeout: 246 seconds)
21:58:33*rnrwashere quit (Remote host closed the connection)
22:12:31*natrys quit (Quit: natrys)
22:25:14*Trustable quit (Remote host closed the connection)
22:28:53*oscar___ joined #nim
22:33:25*oscar___ quit (Ping timeout: 256 seconds)
22:33:31Araqkaushalmodi: meh, please don't "refactor" stuff, I already can program and the programs I write are written exactly as I like them to be written
22:33:37*stefanos82 quit (Remote host closed the connection)
22:34:48FromGitter<kaushalmodi> Araq: the refactoring was to allow the expected vs created docs check extend to the new test
22:35:03FromGitter<kaushalmodi> else I would have needed to copy paste most of the existing code
22:36:10FromGitter<kaushalmodi> The "refactor" term was not meant to offend anyone
22:39:41*mosORadi joined #nim
22:41:23Araqto prevent the copy&paste write a const array and have the tester iterate over it, "data driven design" or something
22:43:03FromGitter<Vindaar> @kaushalmodi I just pushed the ggplotnim (https://github.com/Vindaar/ggplotnim) and the package I build that on top (https://github.com/Vindaar/ginger). But it'll be a few weeks probably before I'd recommend anyone cloning this :D
22:54:10Araqwow my gensym-refactoring works surprisingly well. better than the old implementation...
22:58:32*krux02 quit (Remote host closed the connection)
23:02:48FromGitter<kaushalmodi> Araq: I am doing about the same what you said; just that there are more variables involved. See just the last commit that add the new test: https://github.com/nim-lang/Nim/commit/80440328293ef026e89346c5450be954fd94e42c#diff-4f7c0cc51009eb64f446379068396ad1
23:04:19Araqhmm ok
23:04:21FromGitter<kaushalmodi> Things that change between the 2 tests: dir name, Nim prj name, Nim doc switches
23:05:02xaceVindaar: That is awesome, I'm keeping my eyes on the project. looking forward to render my data in a decent manner with nim :)
23:14:57FromGitter<kaushalmodi> @Vindaar awesome! Can you also commit the example code to generate the plots you have in README.org?
23:16:59*rnrwashere joined #nim
23:21:31*rnrwashere quit (Ping timeout: 248 seconds)
23:21:52FromGitter<Vindaar> the code to generate the plots is also still at the bottom of the `ggplotnim.nim` file: https://github.com/Vindaar/ggplotnim/blob/master/ggplotnim.nim#L358-L370 ⏎ Unless you mean the actual drawing part, which is currently still handled in `ggsave` above that
23:28:32*jasper joined #nim
23:48:45*rnrwashere joined #nim