<< 25-06-2018 >>
00:13:19shashlickWhat's your feature list goal? I'm wondering about a Nim shell that makes dev and test easy but am unsure what it should do
00:14:19*bozhe joined #nim
00:14:55bozhehello, is there possibility to capitalize name in importc
00:15:33bozhelike if i have nim field abcd and c field Abcd, can i it be shorter than {.importc: "Abcd"}
00:18:38shashlickC is case sensitive right, so you have to use the right name there
00:18:52shashlickYou can map it to abcd
00:23:15*jhorwitz_ joined #nim
00:23:40FromGitter<honewatson> Anyone using the https://github.com/krux02/ast-pattern-matching and https://github.com/alehander42/breeze combo?
00:24:08FromGitter<honewatson> ast-pattern-matching -> IN
00:25:40*bozhe quit (Quit: Page closed)
00:27:57*jhorwitz_ quit (Ping timeout: 264 seconds)
00:29:34*jhorwitz_ joined #nim
00:35:09*jhorwitz_ quit (Ping timeout: 264 seconds)
00:35:26*cspar_ joined #nim
00:43:57*dddddd quit (Remote host closed the connection)
00:53:07FromGitter<ephja> the equivalent of the Ballmer Peak for sleepiness is very relevant when trying to read code. if only there was a way to induce just the right amount of sleepiness for a certain amount of time
01:02:10*jhorwitz_ joined #nim
01:07:05*jhorwitz_ quit (Ping timeout: 260 seconds)
01:18:29*jhorwitz_ joined #nim
01:23:20*jhorwitz_ quit (Ping timeout: 256 seconds)
01:33:43*jhorwitz_ joined #nim
01:38:27*jhorwitz_ quit (Ping timeout: 240 seconds)
01:41:07*find0x90 quit (Quit: find0x90)
01:50:26*jhorwitz_ joined #nim
01:54:45*jhorwitz_ quit (Ping timeout: 256 seconds)
02:05:44*jhorwitz_ joined #nim
02:07:35*hutfolk joined #nim
02:10:09*jhorwitz_ quit (Ping timeout: 248 seconds)
02:10:27*hutfolk quit (Client Quit)
02:32:54FromGitter<honewatson> @captainbland I have started doing some work with my own version of dts2nim for nimlsp and also for some React stuff but haven't had a chance to work on it again lately. dts2nim doesn't work for me
02:35:56FromGitter<honewatson> Here is a discussin
02:42:40*endragor joined #nim
02:48:27*xet7 quit (Quit: Leaving)
03:54:40*kinkinkijkin joined #nim
04:02:15*cspar_ quit (Ping timeout: 256 seconds)
04:07:28FromGitter<Varriount> @honewatson I've used ast-pattern-matching. It's a nice library, albeit a bit limited for my use cases (I generally need to match ASTs of unknown length, which reduces how much I can use it).
04:13:03*lompik joined #nim
04:16:11FromGitter<honewatson> @Varriount but you can use it within a for loop and just cover all cases
04:17:48*user1101 quit (Quit: user1101)
04:18:28FromGitter<Varriount> @honewatson Could you give me an example?
04:23:00FromGitter<honewatson> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5b306e24ce3b0f268d403cdc]
04:24:12FromGitter<honewatson> thats a simple example I suppose
04:39:45FromGitter<Varriount> @honewatson What are you developing?
05:06:07*captainbland__ joined #nim
05:07:40*xkapastel quit (Quit: Connection closed for inactivity)
05:08:49*captainbland_ quit (Ping timeout: 265 seconds)
05:33:53FromGitter<Grabli66> Hi! Is it possible to create dynamic libraries(DLL) in nim and load it in nim? I want to use asyncdispatch and threads in that DLL-s.
05:34:18*nsf joined #nim
06:36:53FromGitter<Varriount> @Grabli66 Yes, it is. Could you explain your use-case a bit more?
06:39:45FromGitter<Grabli66> I want to have a core application in nim, that will be loads plugins(dynamic libraries). And these libraries must have possibility to use asyncdispatch module.
06:41:31FromGitter<Varriount> Well, dlls are explained somewhat here: https://nim-lang.org/docs/nimc.html#dll-generation
06:43:50FromGitter<Grabli66> What about that https://github.com/nim-lang/Nim/issues/6855 ?
06:44:29FromGitter<Varriount> @Grabli66 Your main program will still need to provide
06:45:34FromGitter<Varriount> Blah, stupid gitter mobile client
06:46:31FromGitter<Varriount> @Grabli66 Could you provide a abstracted interface?
06:47:12FromGitter<Varriount> One that doesn't directly expose async stuff?
06:47:40FromGitter<Varriount> Or you could try investigating why that issue occurs.
06:49:12FromGitter<Grabli66> Sorry, i don't understand. What interface and where?
06:49:49FromGitter<Grabli66> :)
06:53:38*Vladar joined #nim
07:14:54FromGitter<Varriount> @Grabli66 You still there?
07:15:43FromGitter<honewatson> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5b30969ed2abe46688879537]
07:24:08FromGitter<honewatson> I'm thinking I must just go with GraphQL though
07:30:51FromGitter<Grabli66> @Varriount , yes
07:58:08FromGitter<Varriount> @honewatson Oh, is there a GraphQL library for Nim?
07:59:03FromGitter<Grabli66> I think it's imposible to use dynamic libraries, written in nim, right now, cause of that https://github.com/nim-lang/Nim/issues/6886 . It's sad 😟
08:00:16*hoijui joined #nim
08:00:25hoijui#go
08:00:32FromGitter<Varriount> @Grabli66 Hm, but there are passing tests that use dynamic libraries.
08:01:31FromGitter<Grabli66> Then, i'm confused. :)
08:08:22*floppydh joined #nim
08:08:30*dddddd joined #nim
09:01:51*rauss quit (Read error: Connection reset by peer)
09:03:48*rauss joined #nim
09:15:04*yglukhov[i] joined #nim
09:20:44*Vladar quit (Quit: Leaving)
09:43:47*Zevv left #nim (#nim)
09:54:25*hoijui quit (Ping timeout: 260 seconds)
09:54:26*krux02 joined #nim
09:54:45krux02Araq: ping
09:57:00krux02Araq: I wanted to export the sym id have a simple way to resolve collisions, but resolving collisions is exactly what idOrSig from sighashes does not do for me.
09:57:59krux02for example when I compile ``for i in 0 ..< 10:`` I get a symbol collision for the symbol `i`
09:58:59krux02one symbol i is from the inline of the iterator and the other i is the local `i`
09:59:10krux02I wanted to have a simple way to resolve that.
10:02:53*data-man joined #nim
10:11:08*hoijui joined #nim
10:19:52Araqhmmm
10:22:54*data-man quit (Ping timeout: 268 seconds)
10:24:10krux02ah ok, you are online. You are right, the symbol id is probably not what should be exposed.
10:25:32krux02the only idea I have right now is to make the hash string with it's resolved conflicts property of the symbol. And this property will then be filled by the typechecker.
10:41:00Araqwhy not use gensym()?
10:41:17Araqwhy is (file, line, col) not good enough to disambiguate?
10:42:11krux02To be honest I haven't thought about file line col.
10:42:40krux02but gensym i can't use because all the sybols come from written code, so it is the nim compiler that generates the symbols, not me.
10:42:49*elrood joined #nim
10:45:45*noonien joined #nim
11:19:44*data-man joined #nim
11:23:17krux02Araq: I can't take file line col, because then I get the lineinfo of the use of the symbol not the declaration of the symbol.
11:45:45Araqso add some accessor to get that info
11:45:49Araqbut IDs suck.
11:46:11*user1101 joined #nim
12:02:05*fvs joined #nim
12:13:03*leorize joined #nim
12:32:06*Vladar joined #nim
12:32:40*cspar_ joined #nim
12:45:05krux02Araq: well currently I use id's for local variables only.
12:46:05FromGitter<codenoid> how i can screenshot my screen with only nim
12:46:44krux02codenoid: you can't
12:47:10krux02at least you call into some screen grabbing routine of the os that is not Nim.
12:48:11krux02codenoid: I think you can press the print screen that puts a screenshot in the clipboard and then you can access the clipboard somehow
12:48:38FromGitter<codenoid> what
12:48:56FromGitter<codenoid> i don't understand
12:49:02Yardanicokrux02, well, it depends on the OS and on DE :)
12:50:33livcdcodenoid: what OS ?
12:51:08krux02codenoid: I made a screenshot button for my opengl applicaiton, and in opengl I can capture the pixels of the current frame and save it to an image using sdl2
12:51:19krux02that is completely os independent.
12:51:45krux02but it doesn't capture anything that is not the opengl window
12:51:51krux02so it depends if that is what you want
12:52:34FromGitter<codenoid> ubuntu16.04
12:52:45*hoijui quit (Ping timeout: 268 seconds)
12:53:19YardanicoIt would probably be easier for you to just use something like maim
13:02:25*endragor quit (Remote host closed the connection)
13:04:15*endragor joined #nim
13:09:24*endragor quit (Ping timeout: 268 seconds)
13:23:00*fvs left #nim ("ERC (IRC client for Emacs 25.3.1)")
13:24:08*nsf quit (Quit: WeeChat 2.1)
13:51:37*Zevv joined #nim
13:53:43ZevvIs it possible to limit the scope of a type definition to a block?
13:57:45ZevvI'm trying to declare object types inside procs to aid json deserialization, but that requires unique type names for each function - if not Nim fails with "Error: execution of an external compiler program"
13:59:27dom96sounds like a bug
13:59:35dom96if you're using a macro then you can use ``genSym``
13:59:46ZevvNo, no macros here
13:59:53ZevvI'll reduce to a minimal example
14:01:36AraqZevv, that was fixed in devel afaik
14:01:53ZevvI just noticed that, because I'm not able to reproduce on my other machine
14:02:09ZevvSo types are supposed to be scoped like anything else, right
14:05:40ZevvIs there support for anonymous objects, similar to anonymous structs in C?
14:05:56Araqvia tuples, maybe
14:06:46*leorize quit (Ping timeout: 265 seconds)
14:11:10*TheLemonMan joined #nim
14:11:47TheLemonManAraq, did you see my question about being able to specify types for the unpacked tuples?
14:15:46shashlick@andreaferretti: thanks for the kudos on snip, glad to see it is helping with your dev. workflow and looking forward to hearing your feedback on improvements
14:18:22*cspar_ quit (Ping timeout: 264 seconds)
14:25:30*cspar_ joined #nim
14:44:00AraqTheLemonMan, no...
14:45:58TheLemonManAraq, cool, well I wanted to ask if allowing the user to specify the type for a nkVarTuple is something that wasn't implemented for a reason
14:47:30*hoijui joined #nim
14:50:36*craigger is now known as ChickenDad
14:55:13AraqTheLemonMan, not sure. :-)
15:00:35*hoijui quit (Ping timeout: 240 seconds)
15:00:57TheLemonManroger that, it _shouldn't_ be that hard to implement :)
15:02:11*ChickenDad quit (Quit: bye)
15:02:21*craigger joined #nim
15:07:40*endragor joined #nim
15:09:10TheLemonManand don't forget to drop your two cents in #7699 if you have a few seconds to spare
15:12:30*endragor quit (Ping timeout: 256 seconds)
15:12:48*TheLemonMan quit (Quit: "It's now safe to turn off your computer.")
15:14:13*miran joined #nim
15:14:48FromGitter<ephja> were channels sped up at some point or was "...is slow..." removed from the documentation anyway?
15:15:38*xkapastel joined #nim
15:16:48Araqephja: it was removed because "slow" was misunderstood
15:17:16Yardanicowtf https://github.com/nim-lang/packages/pull/770#issuecomment-399987898
15:23:45miranYardanico: is that because github is now M$? :P :D
15:25:50Yardanicomiran, IDK :D
15:26:03Yardanicobut notTito removed his repo with Nim package and then his account on github
15:28:46miranhardcore :D
15:30:07AraqYardanico, hmm that actually sounds plausible
15:31:18shashlickdom96: travis checks failing for https://github.com/nim-lang/packages/pull/772 - not sure what to do...
15:32:20*Trustable joined #nim
15:35:52dom96shashlick: Just be patient, leave your PR instead of re-merging master into it
15:38:24dom96I'm surprised your packages are installing fine
15:38:28dom96Isn't there a bug that should prevent this?
15:38:58dom96installation of nimpcre failed with: File doesn't exist: pcre.h [Exception]
15:39:03dom96Maybe I spoke too soon :)
15:42:00FromGitter<ephja> did someone else remove his account or what? and how did he respond after the fact?
15:42:17*nsf joined #nim
15:44:15FromGitter<ephja> oh so only the repo was removed first
15:48:11shashlickdom96: where do you see that? some of these are nimgen packages
15:48:58dom96I install all packages manually to verify them
15:49:57shashlickI see, ya these are nimgen packages, let me check nimpcre
15:51:04shashlickdo you have the latest nimgen installed?
15:52:54*themagician_k quit ()
15:56:46dom96just tried installing nimgen@#head and same issues
15:57:40shashlickweird
15:58:40shashlickokay let me get back to you - can you please paste the entire output?
15:59:05shashlickdo nim7z and nimarchive work for you?
16:04:37dom96yeah, they do
16:04:42dom96I pasted it in the PR
16:07:17shashlickokay cool, will review thanks
16:08:48*yglukhov[i] quit (Remote host closed the connection)
16:18:28*yglukhov[i] joined #nim
16:20:12*yglukhov_ joined #nim
16:20:13*yglukhov[i] quit (Read error: Connection reset by peer)
16:21:42*yglukhov[i] joined #nim
16:21:42*yglukhov_ quit (Read error: Connection reset by peer)
16:21:45*yglukhov[i] quit (Remote host closed the connection)
16:24:34shashlickyeah, you are running this on Mac, which I never tested on
16:26:48shashlickdoes `when defined(windows)` work in parsecfg?
16:28:10*miran quit (Ping timeout: 260 seconds)
16:31:07dom96no
16:31:25dom96you mean in a file that is parsed by parsecfg?
16:32:04shashlickyep, doesn't look like - just tried it
16:33:00dom96it's an ini parser
16:33:02dom96nothing more
16:34:37shashlickdom96: just pushed a fix - can you verify with nimpcre@#head? i'll set a tag once you confirm it works for you on mac
16:34:54Yardanicowell, parsecfg format is more powerful than ini :)
16:35:26dom96barely
16:35:47shashlickugh, another bug, a sec
16:38:52Araqcan we secretly move parsecfg to implement the TOML spec?
16:38:54shashlickokay hope it works now
16:39:48dom96Araq: Perhaps, but we already have a Nimble package for that
16:40:17dom96shashlick: works :)
16:40:27shashlickwhew
16:40:34shashlickyet another nimgen feature request
16:40:45YardanicoAraq, yes
16:40:54shashlickis there a place where I can SSH into a Mac and test my nim stuff?
16:40:55YardanicoI would 100% support that because TOML has more advanced features
16:41:27YardanicoandA FAIK it can be represented in the same structure (I mean in the memory) as JSON
16:42:13shashlickdom96: okay I've tagged as 0.1.1
16:42:19shashlickthanks for testing!
16:48:18shashlickdom96: so you don't want me to remerge master for package updates?
16:48:35shashlickI just figured it would make it easier to merge for you
16:58:21FromGitter<Varriount> Yardanico: Can TOML support arbitrarily nested maps?
16:58:59Yardanicowell, you can check it here - https://github.com/toml-lang/toml
16:59:27dom96shashlick: Sometimes it helps, but I'm happy to do it myself too
17:03:47shashlickokay cool
17:06:12*metasyn joined #nim
17:10:35*metasyn left #nim ("WeeChat 2.1")
17:11:56*riidom quit (Ping timeout: 256 seconds)
17:12:01*lompik quit (Ping timeout: 248 seconds)
17:12:45*riidom joined #nim
17:13:30*yglukhov[i] joined #nim
17:14:10*krux02 quit (Read error: Connection reset by peer)
17:15:34*metasyn joined #nim
17:17:51*yglukhov[i] quit (Ping timeout: 240 seconds)
17:22:37*krux02 joined #nim
17:23:15krux02Araq: I think the quest to remove the empty seq for good is not complete :(
17:25:14FromGitter<kaushalmodi> My +1 for inbuilt TOML support! :)
17:25:37FromGitter<kaushalmodi> Yardanico: Yes, TOML can support nested maps
17:26:55FromGitter<kaushalmodi> I meant to ping @Varriount for above reply..
17:27:35*natrys joined #nim
17:30:34krux02I have a case where the gc infinitely grows an object
17:30:38krux02I still have to find out why
17:31:49Araqkrux02, I know and that's why the "new runtime" was preponed
17:31:53*hoijui joined #nim
17:32:05Araqdon't want to fix what I consider a deadend anyway
17:40:12*dorelix quit (Ping timeout: 245 seconds)
17:40:18FromGitter<rayman22201> lol, "preponed" is a great word
17:48:42krux02Araq: so what is the state of the "new runtime"
17:50:11FromGitter<Varriount> TheLemonMan: I love all the help you've given for the compiler.
17:50:16krux02how do I enable it?
17:52:55Araq--newruntime but currently the strings and seqs are not affected by it
17:53:14krux02well I don't use ref types
17:53:24krux02only strings and seq
17:53:31krux02and I have bug that cause lots of allocations
17:57:45*hoijui quit (Ping timeout: 245 seconds)
17:58:00*dorelix joined #nim
18:02:23krux02Araq: sorry it was my fault all the time. I wrote a broken iterator that never terminated and I blamend seq.add for it, because the debugger always was in growObj
18:04:46*Ven`` joined #nim
18:15:21*yglukhov[i] joined #nim
18:16:19*yglukhov[i] quit (Remote host closed the connection)
18:24:46shodan45dom96: just a quick nitpick/suggestion on the survey: "How difficult did you find learning Nim?" was right after questions about various nim resources - I thought "learning Nim" was another book/resource at first
18:25:15federico3dom96: any preview on the outputs?
18:25:22shodan45the lowercase l on "learning" clued me in, eventually :)
18:25:52shodan45dom96: or maybe I'm just brain dead right now... I didn't sleep well :/
18:28:16dom96federico3: 285 responses so far :)
18:28:26dom96shodan45: I think you just need some sleep :)
18:28:45shodan45dom96: :P
18:28:47kinkinkijkinhow new is now() ?
18:28:59kinkinkijkinI'm not having any luck using it
18:29:04kinkinkijkinin 0.17
18:29:06kinkinkijkin.2
18:29:08dom960.18.0
18:29:11kinkinkijkindang
18:29:20kinkinkijkinhow about getTime.local()?
18:29:29dom96probably same
18:29:33dom96Why not just upgrade?
18:29:58kinkinkijkinusing an unsupported distro on an unsupported hardware platform (ubuntu on armv7)
18:30:08Yardanicokinkinkijkin, just compile nim from source
18:30:11Yardanicoeither stable or devel
18:30:15dom96yeah
18:30:19Yardanicoit's very easy and you already have C compiler
18:30:32dom96I guess if you're on an RPi it might take a long time though
18:30:40Yardanicoah, yes
18:30:56Yardanicohttps://nim-lang.org/install_unix.html manual installation from source
18:31:43kinkinkijkinnah I'm on an odroid xu4
18:31:53kinkinkijkinmany times faster than rpi
18:32:45Yardanicowow " Samsung Exynos5422 Cortexâ„¢-A15 2Ghz and Cortexâ„¢-A7 Octa core CPUs"
18:32:53dom96Based on the results so far choosenim seems to be the #1 install method :D
18:33:03Yardanicodom96, how many people submitted?
18:33:09shodan45kinkinkijkin: ooh nice
18:33:13dom96Yardanico: 285
18:33:19dom96287 now lol
18:33:20Yardanicook
18:33:22Yardanicolol
18:33:30Yardanicokinkinkijkin, wait, it's only $60, wtf
18:33:39kinkinkijkin$60 USD
18:33:43kinkinkijkinwas $120 for me in canada
18:33:46shodan45kinkinkijkin: the fan is half the size of the board, lol
18:33:56kinkinkijkinstill cheaper than a new x86 computer of the power level I need
18:34:19Yardanicoyeah, ARM processors are very energy-efficient
18:34:28Yardanicobecause they're used in mobile devices a lot :)
18:35:27kinkinkijkinbuying arm is about buying high compute density to price at lower price ranges
18:35:45kinkinkijkinbuying x86 is about having better system support imo
18:35:54dom96Interesting. More people saying that Nim 1.0 shouldn't be released yet
18:36:10kinkinkijkinthe lack of ARM support in most software is really hitting me hard tho
18:36:22Yardanicowell, it's pretty good with Linux distros
18:37:03kinkinkijkinI do pro audio and while most of my software works, there's one big flaw: the HMP of my processor causes incredible latency jitter and kills preempt_rt support completely as of right now
18:37:09Yardanicoyou can use things like libreoffice, firefox, etc
18:37:19kinkinkijkinand my interface really does NOT like latency jitter
18:37:40dom96kinkinkijkin: interesting, is your software written in Nim?
18:37:59kinkinkijkinnot the stuff I use but the stuff I write is yes
18:38:39kinkinkijkinbut I haven't compiled anything successfully on this computer yet because I'm following the standard library docs published online
18:38:43*cspar__ joined #nim
18:38:50kinkinkijkinand I have the wrong version
18:39:15dom96This might help: https://nim-lang.org/0.17.2/lib.html :)
18:39:21kinkinkijkinoh thanks !!!!
18:40:52kinkinkijkinoh the fix was really simple, and I get the feeling I probably used it last time I wrote a variant of what I'm writing right now (a feeder for my lemonbar to keep track of some stuff)
18:41:10kinkinkijkingetLocalTime() instead of getTime.local()
18:41:27*cspar_ quit (Ping timeout: 245 seconds)
18:47:56shashlickdom96: i feel 1.0 should be out already - in fact 0.19.0 should just be called 1.0
18:49:04shodan45dom96: you left Kotlin off the list of other langs, maybe?
18:49:26dom96I left many
18:49:35shashlickset a goal and ship - as dev guys, we will never be satisfied with anything as 100% but as project managers, you have to ship sometime
18:49:40dom96That's what the "other" is for :)
18:50:41shodan45I "know" most of these languages... but "most comfortable" is a higher bar, I think
18:51:28dom96I looked at last year's results and added the ones that were missing
18:52:25FromGitter<tim-st> Last time I tested the dev version (~20days ago) nimsuggest wasnt usable anymore, I think a 1.0 version cannot have this bug
18:52:49dom961.0 isn't about "no bugs"
18:52:58dom96it's about "from now on there will be no breaking changes"
18:53:18FromGitter<tim-st> Sure, but it's about: you should be able to use it without destroying your whole pc
18:53:33*krux02 quit (Remote host closed the connection)
18:53:54shodan45will there be a party when 1.0 is released?
18:54:12shodan45if so, nim is ready now :D
18:54:12FromGitter<ephja> at my house, yes. bring your own beer
18:54:56dom96tim-st: also, nimsuggest isn't "nim"
18:55:08FromGitter<tim-st> it's shipped with nim
18:55:31FromGitter<tim-st> and developers need development tools
18:56:06FromGitter<tim-st> So you would say a stable nim is ok together with nimsuggest at 100% cpu?
18:56:24FromGitter<Vindaar> @tim-st yeah, I'm suffering from the nimsuggest 100% bug a lot recently too (not sure I have an opinion about that being related to 1.0)
18:56:26*arecaceae quit (Read error: Connection reset by peer)
18:56:26shashlicki don't agree with "1.0 = no more breaking changes"
18:56:32shashlickthat's impossible to guarantee
18:56:43kinkinkijkin1.0 = no more breaking changes until 2.0
18:56:43FromGitter<GULPF> I agree that nimsuggest is pretty broken, but it shouldn't affect the version number of nim the language
18:57:20FromGitter<GULPF> 1) 0 is more like "breaking changes will be avoided to the extent that it's possible" imo
18:57:41elrooda 1.0-release implies a stability-promise though which is something araq or rather the nim community and ecosystem in general might not be comfortable with and ready for just yet
18:57:42FromGitter<tim-st> yes, but it's also about the release information. People will write: this is the stable version, and they ship with unstable dev tools, companies which read this will keep hands away
18:57:50*arecaceae joined #nim
18:58:20FromGitter<tim-st> I'm not telling nimsuggest should have many more features, just usable like in 0.18.0 at least
18:58:31shashlickI still strongly believe that anyone - individual or company - who wants stability will stick to the same version as long as possible
18:58:53Yardaniconimsuggest has all needed features, but it's not very stable
18:59:00shashlickif you keep looking at devel and desire to update every two days, you aren't looking for stability
18:59:02YardanicoI hope this will magically improve with the new runtime :)
18:59:04FromGitter<tim-st> it cannot suggest imports
18:59:38kinkinkijkinhmm, nim is taking twice as long so far to compile with -mfloat-abi=hard and -mtune=cortex-a15.cortex-a7
18:59:42kinkinkijkininteresting
18:59:50shashlickyou only update your build chain if you *have* to
18:59:56Yardanicokinkinkijkin, why are you doing that? :)
19:00:07kinkinkijkinwho knew, forcing the compiler to do more work makes it take longer :p
19:00:15Yardanicobut yeah, optimized builds always take longer to complete
19:00:35FromGitter<tim-st> shashlick: No, I also update if I know these strutils bugs are fixed one day, these are security issues
19:00:51kinkinkijkinyardanico, -mfloat-abi=hard is required on systems which are not running -mfloat-abi=soft software
19:00:58kinkinkijkinin the ARM world
19:01:06Yardanicodoesn't nim work without that?
19:01:19kinkinkijkinif it accesses any libraries or ever uses floats, no
19:01:28elroodkinkinkijkin, why -mtune and not -march?
19:01:37dom96shashlick: Then what should 1.0 mean according to you?
19:02:05kinkinkijkin-march is architecture overall, I have to compile it with specific optimizations for my cpus to get the optimizations for my specific cpu
19:02:05shashlicktim-st: security issues and bug fixes will be done in 1.0.x
19:02:19shashlicknew non-breaking features can be done in 1.x.x
19:02:29kinkinkijkinelrood it's not like x86 where everything compiles and runs happily with -march=native
19:02:32FromGitter<tim-st> ok
19:02:36shashlickwe already discussed this versioning strategy
19:02:54shashlicki don't see why 1.0 still holds us back
19:03:09FromGitter<tim-st> Are there high prio issues left?
19:03:30shashlickhttps://github.com/nim-lang/Nim/issues/7527
19:03:34Yardanicoofc, and issues required for 1.0
19:04:16shashlicksee https://gist.github.com/genotrance/4e4158909972617082dc223b2bb0b2f0 for details on the versioning strategy
19:04:54Yardanicoshashlick, what did we choose?
19:05:05shashlickat the very least, we need to take the time to stop new feature dev, fix whatever we think is high-priority fixes and release 1.0
19:05:26shashlickas long as new dev keeps going on, we will always have broken stuff and desire to fix something before 1.0
19:05:40shashlickAraq strongly wanted to do trunk based dev
19:05:54FromGitter<tim-st> the nimsuggest 100% thing should be fixed, if you dont believe it, try to use it
19:06:25Yardanico@tim-st I use it almost daily
19:06:32Yardanicoit's very good but not very stable
19:06:34FromGitter<tim-st> dev version?
19:06:37Yardanicoofc
19:06:42kinkinkijkinI agree that tools distributed with nim should be working before declaring nim to be shipping-ready
19:06:49Yardanico(if compiler crashes on some code - nimsuggest will crash on that code too)
19:07:12kinkinkijkinworking and working at least almost as well as nim itself
19:07:19FromGitter<tim-st> and of course a setup with gcc is needed for windows, that high prio
19:07:34kinkinkijkinbut msvc is evil
19:07:52shashlickall that is well and good - decide what's the priority for 1.0 and just focus on that, no more new features
19:08:15Yardanico@tim-st why is that bad?
19:08:20Yardanicochoosenim handles C compiler and all that stuff
19:08:31elroodkinkinkijkin, if you don't expect your code to run on another cpu, -march should still be preferrable, on arm you might even want to pass -mcpu, afair
19:08:36FromGitter<tim-st> also I think it's better this time to release a 1.0alpha and the final two weeks later
19:08:50Yardanicoit will be like this, yes
19:08:56Yardanicothere will be a 1.0-rcX
19:09:11*jhorwitz_ joined #nim
19:09:38FromGitter<tim-st> yardanico: maybe choosenim does it but windows users are windows users because they like gui and click on buttons (me too)
19:09:41kinkinkijkinelrood gcc optimizations page for arm prefers mtune over mcpu
19:09:52dom96Nimble ships with Nim too and I have big doubts that I'll release 1.0 of it before Nim 1.0 is released
19:09:59kinkinkijkin-march doesn't include big.LITTLE optimizations
19:10:00shashlicktim-st: that's on my todo list - all installer related stuff is of interest to me
19:10:09Yardanico@tim-st well, are there any gui installers for rust? :P
19:10:12FromGitter<tim-st> maybe a nsis installer like go
19:10:30Yardanicowell, I don't think it's good for a developer to be scared to use command-line
19:10:55kinkinkijkinwindows command line is a terrible thing, and windows dev environment shoos you away from it usually
19:11:16*hoijui joined #nim
19:11:16shashlicki need to put together an RFC for installer ideas - snap, yum, apt, GUI, merging choosenim with finish, linux binary releases, etc. etc
19:11:18FromGitter<tim-st> a gui is needed, imagine python userbase without gui installer for windows
19:11:19kinkinkijkinso it's understandable to not want to have a small piece of commandline in your big graphical dev experience
19:12:05Yardanico@tim-st well, that's python, a lot of people new to programming use python as a first language
19:12:30shashlickwe can certainly get there, GUI installer isn't a difficult problem
19:12:44FromGitter<tim-st> shashlick: yes, think so too
19:12:45kinkinkijkinI actually recommend nim to people as their second, and it would be nice to have a GUI installer on windows for them
19:13:18kinkinkijkinGUI is not so hard until you start trying to get into compiler control with GUI
19:13:46shashlickchoosenim is very capable, few more features and it can do everything we need - we just need a GUI wrapper for it, nothing exotic
19:14:25Yardanicowell, also we should think about nimble package index
19:14:46Yardanicohmm, I use a lot of "well"s, that's not good :D
19:15:27FromGitter<tim-st> nimble.directory should list new packages
19:15:45FromGitter<tim-st> I can not read this
19:15:46dom96We used to have a GUI installer
19:16:02shashlicktim-st: it already lists new packages
19:16:21FromGitter<tim-st> you mean this: https://nimble.directory/packages.xml
19:16:36shashlickno go to the home page - it lists latest packages
19:16:46FromGitter<tim-st> it lists one
19:16:55Yardanicobecause there was only one package added recently :)
19:17:10FromGitter<tim-st> but I want latest 20
19:17:12Yardanico(4 hours ago)
19:17:20Yardanicowell, make a feature request in nimble.directory repo
19:17:32shashlickfederico3: ^^^
19:18:00Yardanico@tim-st https://github.com/FedericoCeratto/nim-package-directory
19:19:07*yglukhov[i] joined #nim
19:20:09*miran joined #nim
19:21:00elroodkinkinkijkin, got a reference to your gcc optimization recommendations? they apparently don't match with what i got
19:21:24FromGitter<tim-st> yardanico: ok, I have done it
19:22:20kinkinkijkinhttps://gcc.gnu.org/onlinedocs/gcc/ARM-Options.html
19:22:45*jhorwitz_ quit (Ping timeout: 245 seconds)
19:23:43FromGitter<ephja> don't forget about -funrolls!
19:24:14*Zevv left #nim (#nim)
19:24:34elroodkinkinkijkin, thanks. fair enough, if you're sure you're doing the right thing for your system
19:34:12FromGitter<tim-st> and before 1.0 will be released there really needs a new code example on the start page (maybe categories from beginner -> c expert), maybe people can collect some code and then a voting can be made
19:35:02FromGitter<tim-st> which is the best code for the start page
19:35:59*jhorwitz_ joined #nim
19:40:49*jhorwitz_ quit (Ping timeout: 248 seconds)
19:48:12*Jesin joined #nim
19:54:51*skirkpatrick7551 joined #nim
19:57:01*cspar_ joined #nim
20:00:05*cspar__ quit (Ping timeout: 265 seconds)
20:04:51*metasyn quit (Ping timeout: 240 seconds)
20:05:16*noonien quit (Quit: Connection closed for inactivity)
20:11:01*metasyn joined #nim
20:12:58*kinkinkijkin quit (Quit: Leaving)
20:13:57*miran quit (Ping timeout: 240 seconds)
20:15:04*jhorwitz_ joined #nim
20:16:26*Ven`` quit (Read error: Connection reset by peer)
20:20:10*jhorwitz_ quit (Ping timeout: 264 seconds)
20:20:16*Trustable quit (Remote host closed the connection)
20:20:58*Ven`` joined #nim
20:34:22*xet7 joined #nim
20:37:12*Vladar quit (Quit: Leaving)
20:40:09*hoijui quit (Ping timeout: 256 seconds)
20:41:16*jhorwitz_ joined #nim
20:45:20*yglukhov[i] quit (Read error: Connection reset by peer)
20:45:49*jhorwitz_ quit (Ping timeout: 256 seconds)
20:45:58*yglukhov[i] joined #nim
20:48:32*jhorwitz_ joined #nim
20:53:07*jhorwitz_ quit (Ping timeout: 245 seconds)
20:53:33*Sentreen joined #nim
21:00:46*jhorwitz_ joined #nim
21:02:38*Ven`` quit (Remote host closed the connection)
21:02:48*Ven`` joined #nim
21:03:52*mal``` joined #nim
21:04:57*jhorwitz_ quit (Ping timeout: 240 seconds)
21:05:00*Death916_ joined #nim
21:05:42*jhorwitz_ joined #nim
21:06:05*mal`` quit (Quit: Leaving)
21:06:05*Death916 quit (Ping timeout: 256 seconds)
21:10:39*jhorwitz_ quit (Ping timeout: 265 seconds)
21:10:46*jhorwitz_ joined #nim
21:11:12*Ven`` quit (Quit: q+)
21:15:05*jhorwitz_ quit (Ping timeout: 240 seconds)
21:25:53*FromGitter * Varriount wishes vtable types would be released for 1.0
21:28:22*nsf quit (Quit: WeeChat 2.1)
21:32:57*skirkpatrick7551 quit (Quit: Page closed)
21:44:17FromGitter<kindlychung> why did i get an indexerror here? https://gist.github.com/kindlychung/4d64d87b41861f17a20140b92bc6a266
21:45:33Araqbecause only on message is returned by findMessages?
21:45:39Araq*only one
21:48:23FromGitter<kindlychung> @Araq The users were not inserted at all.
22:07:11dom96https://github.com/dom96/nim-in-action-code/blob/master/Chapter7/Tweeter/tests/database_test.nim
22:08:10*brainproxy quit (Ping timeout: 265 seconds)
22:09:23dom96That might help
22:13:54FromGitter<kindlychung> Thanks.
22:14:17FromGitter<kindlychung> I'll figure out what I did wrong.
22:16:32dom96I just ran it locally and it seems to be working fine
22:18:41*natrys quit (Quit: natrys)
22:21:12*yglukhov[i] quit (Remote host closed the connection)
22:31:46*jhorwitz_ joined #nim
22:33:35*elrood quit (Remote host closed the connection)
22:34:10federico3tim-st: if you use a RSS feed reader it will scrape new packages as they come up and store the full history: each entry is an RSS item
22:36:19*jhorwitz_ quit (Ping timeout: 256 seconds)
22:39:36*CcxWrk quit (Ping timeout: 256 seconds)
22:41:34*CcxWrk joined #nim
22:46:50*metasyn quit (Ping timeout: 276 seconds)
22:48:07*brainproxy joined #nim
22:50:38*data-man quit (Quit: Konversation terminated!)
22:59:37*yglukhov[i] joined #nim
22:59:54*metasyn joined #nim
23:03:00*kinkinkijkin joined #nim
23:04:05*yglukhov[i] quit (Ping timeout: 256 seconds)
23:17:20*captainbland_ joined #nim
23:17:40*captainbland_ quit (Remote host closed the connection)
23:19:01*captainbland__ quit (Ping timeout: 256 seconds)
23:22:25*captainbland_ joined #nim
23:26:23*krux02 joined #nim
23:26:39shashlickdoes this work? `when NimMajor == 0 and NimMinor < 18: {.experimental.}`
23:27:42krux02why the fuck does `debug` now need a `ConfigRef`
23:27:49krux02and where do I get it from?
23:28:03krux02and why does it crash with a segmentation fault when I just pass nil to it?
23:28:23krux02this is really annoying and prevents debugging the Nim compiler
23:29:56*jhorwitz_ joined #nim
23:35:07*xkapastel quit (Quit: Connection closed for inactivity)
23:35:09*jhorwitz_ quit (Ping timeout: 264 seconds)
23:48:37*xet7 quit (Quit: Leaving)
23:50:34*user1101 quit (Quit: user1101)
23:53:52FromGitter<Varriount> krux02: Probably because Araq moved globals out of the compiler.
23:55:42krux02Varriount: yea and it makes debugging nim even more complicated than it already is.
23:55:45FromGitter<ephja> I think I had to use a global when there was no context parameter
23:56:10krux02ephja: exacty that is horrible
23:56:27krux02you want to be able to just dump the debug proc everywhere
23:56:32krux02now you can't
23:56:42krux02you have to introduce global variables
23:57:18Araqor: you use the .config field of whatever context you happen to have. they are everywhere.
23:57:37Araqit's not like I don't work with this code base myself ffs
23:58:10krux02Araq: explain everywhere
23:58:51AraqCtx, PContext, PCtx, Parser, Lexer, BModule, ... sounds familiar?
23:59:18krux02how do I get it from PContext?