<< 03-09-2019 >>
00:16:30*Cthalupa quit (Ping timeout: 268 seconds)
00:21:21*Cthalupa joined #nim
00:23:48*gh0st[m] quit (Remote host closed the connection)
00:23:50*planetis[m] quit (Read error: Connection reset by peer)
00:23:56*Connor[m] quit (Remote host closed the connection)
00:23:56*k0mpjut0r quit (Write error: Connection reset by peer)
00:23:56*pigmej[m] quit (Write error: Broken pipe)
00:23:58*leorize[m] quit (Remote host closed the connection)
00:23:58*nergal[m] quit (Write error: Connection reset by peer)
00:24:00*meff[m] quit (Write error: Connection reset by peer)
00:24:02*narimiran[m] quit (Write error: Connection reset by peer)
00:24:02*xomachine[m] quit (Remote host closed the connection)
00:24:03*spymasterd[m] quit (Remote host closed the connection)
00:24:06*GitterIntegratio quit (Write error: Connection reset by peer)
00:24:07*Miguelngel[m] quit (Remote host closed the connection)
00:24:07*lqdev[m] quit (Remote host closed the connection)
00:24:10*disruptek o7
00:24:11*LEdoian[m] quit (Write error: Connection reset by peer)
00:24:12*BitPuffin quit (Remote host closed the connection)
00:24:13*macsek1911[m] quit (Remote host closed the connection)
00:24:13*zielmicha[m]1 quit (Remote host closed the connection)
00:24:13*encryptize[m] quit (Remote host closed the connection)
00:24:14*TheManiac[m] quit (Read error: Connection reset by peer)
00:24:16*jaens[m] quit (Remote host closed the connection)
00:24:16*isaac[m]1 quit (Remote host closed the connection)
00:24:16*yglukhov[m] quit (Remote host closed the connection)
00:24:17*Demos[m] quit (Remote host closed the connection)
00:24:17*skrylar[m] quit (Remote host closed the connection)
00:32:12*Connor[m] joined #nim
00:53:09*Demos[m] joined #nim
00:53:09*LEdoian[m] joined #nim
00:53:09*BitPuffin joined #nim
00:53:09*k0mpjut0r joined #nim
00:53:09*isaac[m]1 joined #nim
00:53:09*GitterIntegratio joined #nim
00:53:09*planetis[m] joined #nim
00:53:10*gh0st[m] joined #nim
00:53:10*jaens[m] joined #nim
00:53:10*nergal[m] joined #nim
00:53:10*leorize[m] joined #nim
00:53:10*pigmej[m] joined #nim
00:53:16*yglukhov[m] joined #nim
00:53:16*spymasterd[m] joined #nim
00:53:16*narimiran[m] joined #nim
00:53:16*macsek1911[m] joined #nim
00:53:16*TheManiac[m] joined #nim
00:53:17*Miguelngel[m] joined #nim
00:53:17*skrylar[m] joined #nim
00:53:17*encryptize[m] joined #nim
00:53:17*meff[m] joined #nim
00:53:18*lqdev[m] joined #nim
00:53:18*xomachine[m] joined #nim
00:53:18*zielmicha[m]1 joined #nim
01:21:45*endragor joined #nim
01:29:08*endragor_ joined #nim
01:32:08*endragor quit (Ping timeout: 245 seconds)
01:56:18*endragor_ quit (Remote host closed the connection)
02:18:17*rayman22201 quit (Quit: Connection closed for inactivity)
02:34:52*actuallybatman quit (Ping timeout: 246 seconds)
02:37:01*actuallybatman joined #nim
03:12:34shashlick@mratsim this is only for custom tasks
03:15:12*fjellfras joined #nim
03:15:38*fjellfras quit (Max SendQ exceeded)
03:16:07*fjellfras joined #nim
03:16:18*fjellfras quit (Max SendQ exceeded)
03:16:47*fjellfras joined #nim
03:17:07*fjellfras quit (Max SendQ exceeded)
03:17:30*endragor joined #nim
03:17:34*fjellfras joined #nim
03:17:54*fjellfras quit (Max SendQ exceeded)
03:24:09*dddddd quit (Remote host closed the connection)
03:25:53*pbb quit (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.)
03:26:33*pbb joined #nim
03:38:49*chemist69 quit (Ping timeout: 250 seconds)
03:40:51*chemist69 joined #nim
03:50:05*owl_000 quit (Ping timeout: 250 seconds)
04:24:04*alexander92 quit (Ping timeout: 246 seconds)
04:37:41*rockcavera quit (Remote host closed the connection)
04:49:43*nsf joined #nim
05:09:40*narimiran joined #nim
05:15:21*vlad1777d joined #nim
05:17:17*solitudesf joined #nim
06:31:02*PMunch joined #nim
06:44:02*krux02 joined #nim
07:00:00*gmpreussner quit (Quit: kthxbye)
07:04:48*gmpreussner joined #nim
07:27:34*Hideki_ joined #nim
07:28:36*ng0 joined #nim
07:32:08Hideki_I placed three files: /test.nim, /model/foo.nim, /api/foo.nim. Let's assume /model/foo.nim and /api/foo.nim are empty files. I imported the two files in /test.nim like import "model/foo.nim" and import "api/foo.nim". Then, I got a compile error Error: redefinition of 'foo': previouds declaration here: ... My question is: Should I use globally unique file names in a Nim project?
07:35:23*Hideki_ quit (Remote host closed the connection)
07:35:51*Hideki_ joined #nim
07:36:57solitudesfyou can do `import model/foo as foo1` or something
07:38:02*Hideki_ quit (Remote host closed the connection)
07:38:41*Hideki_ joined #nim
07:40:33Hideki_@solitudesf Thanks. It worked. I didn't know a filename is an identifier like proc name and type name are.
07:42:50*Hideki_ quit (Remote host closed the connection)
07:43:06*Hideki_ joined #nim
07:46:52*Hideki_ quit (Remote host closed the connection)
07:50:51*Hideki_ joined #nim
07:57:32*absolutejam joined #nim
08:05:05*noonien joined #nim
08:06:38noonienhello folks!
08:07:07noonienhow can i assign a string to a byte array?
08:07:59narimirannoonien: `toOpenArrayByte` ?
08:09:43noonienthanks
08:13:59*floppydh joined #nim
08:14:55*laaron quit (Quit: ZNC 1.7.1 - https://znc.in)
08:17:04*laaron joined #nim
08:57:06*Vladar joined #nim
09:10:39*ehmry joined #nim
09:17:14*skelett1 quit (Quit: WeeChat 2.4)
09:17:40*skelett joined #nim
09:25:17*absolutejam quit (Ping timeout: 244 seconds)
09:25:25*chemist69 quit (Quit: WeeChat 2.3)
09:29:39*laaron quit (Quit: ZNC 1.7.1 - https://znc.in)
09:30:28*laaron joined #nim
09:37:12*laaron quit (Quit: ZNC 1.7.1 - https://znc.in)
09:37:33*laaron joined #nim
09:39:08*absolutejam joined #nim
10:01:32*Vladar quit (Remote host closed the connection)
10:04:59*Vladar joined #nim
10:09:34*Hideki_ quit (Remote host closed the connection)
10:19:19*actuallybatman quit (Ping timeout: 246 seconds)
10:19:34*ZORR0W joined #nim
10:27:32*alexander92 joined #nim
10:29:34FromGitter<alehander42> i have
10:29:41FromGitter<alehander42> this weird error required type: Lang ⏎ but expression 'LangNim' is of type: Lang
10:32:17PMunchThat does indeed look weird
10:34:30alexander92i can workaround by changing the signature to int, and using arg.Lang in the functio
10:34:35alexander92but it doesnt make sense to me
10:34:52FromGitter<arnetheduck> what was the reason for not going with sqlite amalgamation? it's got all kinds of reasons to use it: faster, easier to deploy, no version incompatibilities etc etc... it's literally made to be deployed this way, specially since nim already heavily relies on `C~
10:36:03alexander92is sqlite going to be a long-term solution? i felt it's only a temporary one
10:36:49alexander92oh man, now i have it again with another type
10:36:54alexander92i am doing something wrong!
10:37:42FromGitter<arnetheduck> you'll see, soon we'll have an ORM in the compiler as well :)
10:39:10alexander92no, i remember ideas for a memory-mapped almost-no-deserialization needed format
10:41:53*Tongir joined #nim
10:42:06FromGitter<arnetheduck> regardless, that also sounds like a reason to go with the amalgamation - it's painful and causes all kinds of problems to have a separate dll, and you don't want to waste time on teaching people to deal with those problems specially if it's temporary..
10:42:40alexander92no, i just speculate, i dont think araq thinks it temp
10:43:09AraqI don't know, at this point I'm concerned to get the logic and algorithms right
10:46:40AraqI don't use much of sqlite though
10:47:08Araqsince it's way too slow :-)
10:48:22*absolutejam quit (Ping timeout: 245 seconds)
10:51:44alexander92exactly ; )
10:52:46FromGitter<arnetheduck> so fwiw, even though we're probably one of the bigger nim projects out there, compile times and thus incremental compilation are quite low on our prio list, compared to correctness and other issues around tooling - library, package and module management being on the forefront of tooling and stdlib curation/quality/obsolecense on the prime spot for nim right now.. we've mostly stopped using any advanced nim compiler
10:52:46FromGitter... features beyond 0.19.6 which is a motivating reason why we haven't made the effort to move to 0.20 as well - there'
10:53:41alexander92is it possible that `auto` generates weird type syms
10:54:11alexander92because when i get a "QueryNode" `query` from awaiting a `auto` await template .. it is not recognized as QueryNode
10:54:41*Hideki_ joined #nim
10:55:28alexander92a cast works as a workaround, but it's a totally confusing error
10:56:18Araqouch, I thought IC is feature number one
10:56:29Araqthat everybody needs
10:56:38Araqor at least would benefit from
10:57:57Araqesp Status
11:04:56*absolutejam joined #nim
11:12:58*Hideki_ quit (Ping timeout: 245 seconds)
11:18:32*Tongir quit (Read error: No route to host)
11:19:27FromGitter<arnetheduck> not really - our ability to cooperate efficiently is a lot more important - between each other and with other libraries/projects really - that's why package, module and library management is such a big deal
11:22:07*Araq wispers Monorepo
11:22:21FromGitter<arnetheduck> IC is something that lends the individual developer a little convenience, but by and large it can be worked around through other means (ie buy a better PC)..
11:24:57FromGitter<arnetheduck> well yeah, monorepo solves some problems but the tooling is not ready for that either
11:25:37FromGitter<arnetheduck> and it brings others - access control / ownership of the repo to begin with
11:28:09livcdarnetheduck: what features did you stop using ?
11:28:15*absolutejam quit (Ping timeout: 244 seconds)
11:29:16FromGitter<arnetheduck> everything `--experimental`, large parts of the std lib
11:31:00FromGitter<arnetheduck> another way to look at it: there's nothing in 0.20 that we really *need* and plenty of risks like newruntime and hcr as well as real breakage
11:32:34FromGitter<alehander42> i thought they are well hidden behind flags/gates
11:32:43FromGitter<alehander42> so they shouldn't really matter if you dont use them?
11:33:16FromGitter<alehander42> i honestly guess the discussions about it take longer then actually upgrading
11:34:24FromGitter<arnetheduck> the flags/gates leak and bring complexity and bugs
11:35:01FromGitter<arnetheduck> I mean, 3-4 of our devs have tried unsucessfully to upgrade the codebase to 0.20 so I trust that it's not trivial
11:39:08livcdwell good to know..if you guys are building a lot of libs that people might use it's a warning to consider
11:40:11*absolutejam joined #nim
11:40:56FromGitter<arnetheduck> mostly I think it represents churn that we're unwilling to prioritise at the moment - getting to a stable working solution to our own problems is more important
11:41:29AraqI'm looking at what it takes to build Nimbus with v1 as we speak
11:41:34*alexander92 quit (Ping timeout: 246 seconds)
11:42:03livcdhmm like chronicles-0.6.0\chronicles\topics_registry.nim(38, 5) Error: unguarded access: gActiveLogLevel
11:43:33FromGitter<arnetheduck> in general, I believe that it's better for nim itself to take the chance to cut out all obsolete stuff even if it means breakage - the sooner, the cheaper - it's really really easy to add stuff, and much harder to remove.. from status' perspective, we just don't want to follow every incremental step of that process necessarily
11:47:05FromGitter<arnetheduck> with proper package management, those kinds of decisions become easy to deal with as well - the units of breakage are smaller and come with less collateral damage
11:47:58Araqwell yeah, you believe in "Proper package management", I have not found that anywhere
11:48:36Araqon this very machine if I type 'python' I'm getting 2.4.4
11:48:46Araqstopped trying to update it
11:49:37Araqsomewhere I also have the Python 2.7 that I needed
11:50:38Araqif the one OS that lacks package management has the most amount of commercial production software that doesn't break (Windows) that could tell us a lot
11:52:04*absolutejam quit (Ping timeout: 246 seconds)
11:56:53*owl_000 joined #nim
11:59:51FromGitter<arnetheduck> yes, and the banks still use cobol
12:13:13*absolutejam joined #nim
12:17:13Araqone year ago I played with Elm, its package manager told me about some incompatible version numbers, I googled the problem and made it work. Here is what happens without a PM: the Elm compiler gives me some error, I google it and can make it work.
12:21:24*Kaivo joined #nim
12:23:22*rockcavera joined #nim
12:25:56owl_000are include and import same under the hood. i have a module contains two seperate proc. in `importer.nim` i used `from module import proc1` and use only `proc1`, in `includer.nim` `include module` and then used only proc1. i find no difference. even number of line compiled is exactly same.
12:27:05PMunchNo they are not
12:27:06solitudesfinclude acts like literally copypasting module, so all procs types and fields are visible
12:27:17Araqhttps://nim-lang.org/docs/manual.html#modules-include-statement
12:30:38owl_000i read that, i thought `from .. import ..` will import only that proc, on the other hand include will include all. but why the number of lines compiled is same.
12:31:12Araqbecause you don't 'include' it twice
12:32:24owl_000let me think
12:34:40owl_000module contains two proc, i imported only one proc in one file. in other file i included everything by include module. is it true?
12:35:53owl_000if it is true, shouldn't i see a difference in number of lines compiled?
12:36:51Araqno, 'import' doesn't work this way, it still compiles the full module
12:37:29*vlad1777d quit (Ping timeout: 244 seconds)
12:37:36narimiranowl_000: i think the best way to see the difference between `import` and `include` is to have some not-exported proc, and to try to use it in a file where you include/import it
12:38:47narimiranand here is my rule of thumb: don't use `include`
12:41:16owl_000the qustions arise because, after compeleting tut1, i am exploring github for nim packages. in `wnim` the compilation time is very low. i explore the source code and found that the author included everything in `wnim.nim`, i removed some of includes like `wRadioButton` and it increases compilation time.
12:41:40owl_000---
12:42:25owl_000and i am testing what is the difference and found that number of compiled line is same.
12:42:54owl_000narimiran ok i am doing it
12:44:28*Hideki_ joined #nim
12:45:24*dddddd joined #nim
12:46:22owl_000narimiran, there is no differences in number of compiled line. i am not sure whether i followed you correctly or !. module one i remove asterisk from one proc and keep other one.
12:47:04owl_000as Araq said it actually compiles whole module
12:47:21Araqyou can remove 'include' to improve compile-times but you can also remove 'import' and would also see improved compile-times
12:48:54narimiranowl_000: from my comment "...and to try to use it..." — the difference is in the usage, not in compiled lines or time
12:51:16owl_000`from module include proc` will include that proc and the proc and types it is using will be a good idea.
12:52:13*metadave_ joined #nim
12:52:16narimirani had no idea that `from foo include bar` is a valid syntax
12:52:31*metadave_ quit (Client Quit)
12:52:39owl_000it will be a good idea. not a valid syntax. lol
12:52:44narimiranah
12:52:58*ZORR0W quit (Ping timeout: 246 seconds)
12:56:43*laaron quit (Remote host closed the connection)
12:59:31*sacredfrog joined #nim
12:59:43*chemist69 joined #nim
13:00:44*nsf quit (Quit: WeeChat 2.5)
13:04:06*endragor quit (Remote host closed the connection)
13:06:42*laaron joined #nim
13:13:39*absolutejam quit (Ping timeout: 244 seconds)
13:27:22FromGitter<awr1> i guess its kinda unfortunate that the import+import fully qualified syntax in nim are the other way around from python
13:28:11narimiranah, that again. https://narimiran.github.io/2019/07/01/nim-import.html
13:28:16Araqoh not this again.
13:28:26Araqa module in Nim is a class in Python.
13:29:27*rockcavera quit (Remote host closed the connection)
13:29:47*rockcavera joined #nim
13:32:42FromGitter<awr1> well i know the reason nim did this was to discourage people from doing fully-qualified imports everywhere
13:33:18Araqno, the reason was we don't have 'class'
13:33:28FromGitter<awr1> hm.
13:33:36Araqand lacking classes is still hip, so ... *shrug*
13:35:36FromGitter<awr1> i mean you are right, i get the feeling if that nim encouraged OOP more i think much less people would be interested in it
13:37:21FromGitter<awr1> myself probably included
13:37:57FromGitter<awr1> (or maybe excluded? i don't know what the proper phrasing would be)
13:38:29lqdev[m]although it's minimal, the OOP paradigm used in Nim is sufficient for 99% cases. you really don't need much more than that
13:39:27FromGitter<awr1> yeah i agree
13:39:56FromGitter<awr1> in other languages OOP mentality is just forced into things that don't need it
13:40:27FromGitter<awr1> and of course, in java/C#/etc you have no choice anyway
13:43:36*absolutejam joined #nim
13:53:36PMunchHmm, is it just me or is the parseUri procedure in the uri module not doing what it should
13:54:33PMunchLooking at this: en.wikipedia.org/wiki/File_URI_scheme it seems like a Windows path like "file:///C:/Users/moigagoo/Projects/norm/src/norm.nim" should be an empty hostname and the path C:/Users/moigagoo/Projects/norm/src/norm.nim
13:55:06PMunchBut it has a leading slash on it like this: /C:/Users/moigagoo/Projects/norm/src/norm.nim
13:55:59PMunchSame for paths on Linux, paths given this way aren't actually supposed to have a leading /, they are supposed to be relative to the host, and in the case that hostname is "" then that means relative to the local file system.
13:56:55Cadeydom96: how would I define nim routers in separate modules?
13:58:50Cadeyjester*
13:58:56Cadeyi can't seem to figure it out
13:59:17PMunchrouter myrouter: <your route>
13:59:21PMunchexport myrouter
13:59:25PMunchI think that might work
14:00:43Cadeynope
14:00:52*laaron quit (Remote host closed the connection)
14:01:43PMunchWorks for me..
14:02:02PMunchhttp://ix.io/1Ua3/
14:02:30Cadeyah, i'm using the routes macro for my main router
14:02:44PMunchRunning "nim c -r jestertest 5000"
14:02:58*laaron joined #nim
14:03:34PMunchThen you'd probably have to use this: https://github.com/dom96/jester#manual-routing
14:04:34Cadeyi'll just change it over to routers :)
14:05:35PMunchYeah, but how do you use two routers together?
14:06:20PMunchAh, with extend: https://github.com/dom96/jester/blob/master/tests/alltest.nim#L29-L32
14:06:29PMunchThat should probably me mentioned somewhere..
14:09:02PMunchThis works, and uses the routes macro for the main route: http://ix.io/1Ua6/
14:09:05PMunchCadey ^
14:09:12Cadeyokay, i'll take a look
14:09:51*PMunch quit (Remote host closed the connection)
14:13:52*absolutejam quit (Ping timeout: 264 seconds)
14:15:15*absolutejam joined #nim
14:31:49*krux02 quit (Remote host closed the connection)
14:45:07owl_000simple question when will we get nim lang v 1.0
14:45:43Zevv"soon"
14:48:23*absolutejam quit (Ping timeout: 245 seconds)
14:49:40disruptekstatus can't afford to upgrade from 0.19.6 to 0.20.0 and future design changes are already difficult... i'd say nim has already passed 1.0.
14:50:15Araqhttp://lists.llvm.org/pipermail/llvm-dev/2016-October/106291.html interesting read
14:50:35Araqlooks like tracing GC wins for LLVM too, just like it does for the Nim compiler
14:51:22Araqowl_000: fwiw https://github.com/nim-lang/Nim/milestone/2
14:52:44disruptekthat llvm thread is pretty funny.
14:52:52Araqwhy?
14:52:56owl_000greeeat :O
14:53:19disrupteki love the idea that we throw out machinery.
14:53:40Araqno, what's funny about the LLVM thread?
14:53:50*absolutejam joined #nim
14:54:23disruptekthe load of alloc bs.
14:56:08Araqcompilers eat RAM
14:57:41disrupteki doubt the ram it ate changed to the downside.
15:03:35*ng0 quit (Remote host closed the connection)
15:04:40*ng0 joined #nim
15:08:46*absolutejam quit (Ping timeout: 246 seconds)
15:11:57*Hideki_ quit (Remote host closed the connection)
15:16:59*laaron quit (Remote host closed the connection)
15:19:52*laaron joined #nim
15:23:13*laaron quit (Client Quit)
15:23:35*laaron joined #nim
15:26:58euantorThe above code results in `Mode: 511`, which is a little different to the commented `777` from https://github.com/nim-lang/Nim/blob/devel/tests/async/tioselectors.nim#L324 - shall I open an issue? https://www.irccloud.com/pastebin/DvRsYQnK/
15:27:01*tiorock joined #nim
15:27:01*rockcavera is now known as Guest15382
15:27:01*tiorock is now known as rockcavera
15:27:01*rockcavera quit (Changing host)
15:27:01*rockcavera joined #nim
15:28:37Araqeuantor: just create a PR that removes the comment
15:29:28euantorAraq: CHMOD 511 results in `Error: unhandled exception: value out of range: 511 [RangeError]` on openBSD, so there's another issue with that block of code in that case
15:29:32*Guest15382 quit (Ping timeout: 244 seconds)
15:31:57euantorShall I instead just pass `777` to match the comment
15:39:27euantorPR created, hopefully CI passes: https://github.com/nim-lang/Nim/pull/12120
15:48:40*adeohluwa joined #nim
15:54:27*Trustable joined #nim
15:56:51*floppydh quit (Quit: WeeChat 2.5)
15:59:12*actuallybatman joined #nim
16:00:44*laaron quit (Quit: ZNC 1.7.1 - https://znc.in)
16:03:08*laaron joined #nim
16:19:53*Hideki_ joined #nim
16:20:05*Hideki_ quit (Remote host closed the connection)
16:27:51*adeohluwa quit (Remote host closed the connection)
16:32:25*adeohluwa joined #nim
16:42:04*matlock joined #nim
16:43:58*adeohluwa quit (Ping timeout: 246 seconds)
17:02:00*adeohluwa joined #nim
17:19:16*adeohluwa quit (Ping timeout: 264 seconds)
17:34:17*owl_000 quit (Ping timeout: 250 seconds)
17:34:18*metadave_ joined #nim
17:43:29*wildtrees joined #nim
18:20:22*nsf joined #nim
18:29:23*Trustable quit (Remote host closed the connection)
18:30:01*absolutejam joined #nim
18:30:57*krux02 joined #nim
18:31:50Zevvbwah, I was hoping hat execShellCmd("nimble install ...") would work in the playground :)
18:54:28*solitudesf- joined #nim
18:56:43*solitudesf quit (Ping timeout: 244 seconds)
19:08:56*ofelas quit (Remote host closed the connection)
19:08:59*drewr joined #nim
19:28:14*Tyresc joined #nim
19:33:22*Tyresc quit (Ping timeout: 245 seconds)
19:33:40*juturnas joined #nim
19:34:49juturnasWhat is the correct syntax for / is it possible to have a function pointer in a tuple or object? I've tried variations of `score: ((string) -> float)` with no success
19:36:53solitudesf-score: proc(s: string): float
19:37:47solitudesf-for arrow notation there is https://nim-lang.github.io/Nim/sugar.html#-%3E.m%2Cuntyped%2Cuntyped
19:38:39juturnasAwesome, thank you solitudesf-
19:40:59*metadave_ quit (Quit: Connection closed for inactivity)
19:42:27*Ven`` joined #nim
19:45:17CadeyI have a non-ref Snake, how do I convert it to a ref Snake?
19:47:31*Ven`` quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
19:48:57narimiranCadey: RefSnake = ref Snake ?
19:55:24solitudesf-proc toRef[T](a: T): ref T =
19:55:25solitudesf- new result
19:55:25solitudesf- copyMem addr result[], unsafeAddr a, sizeof T
19:55:35solitudesf-this works, but i dont know if thats completely retarded
19:57:08Cadeyi figured out a workaround, i just marked the entire Snake type as a ref object
20:04:12*narimiran quit (Ping timeout: 245 seconds)
20:10:00Araqdon't use copyMem for this
20:10:01Araquse
20:10:06Araq result[] = a
20:10:12*alexander92 joined #nim
20:12:07alexander92Cadey, did you find out
20:12:10alexander92the route thing
20:12:28solitudesf-ah, thats nice
20:13:05alexander92hm, after all what does `auto` do
20:13:10alexander92in template return-s
20:18:44FromGitter<awr1> infers the return type
20:18:52juturnas`applyInterpreter' is not GC-safe as it performs an indirect call here` <- I wasn't expecting that. Is the problem simply that I'm doing dynamic dispatch in the non-main thread?
20:23:27alexander92awr1 ok
20:23:29alexander92thanks!
20:23:52alexander92but the problem i see is that i dont understand where does it resolve it
20:23:55alexander92on the callsite
20:23:59alexander92or in the expansion
20:24:14alexander92because i get some expected Lang but got Lang
20:24:17alexander92errors as a result
20:24:53FromGitter<awr1> so in templates you don't get `return` or `result` like procs, you only implicit returns. so return type is inferred from the implicit return
20:25:10FromGitter<awr1> *only get
20:25:37FromGitter<awr1> because templates are essentially a "stronger form" of copy pasting
20:26:11FromGitter<awr1> i would assume they are inferred at expansion? i think?
20:30:15*nsf quit (Quit: WeeChat 2.5)
20:36:04alexander92yes, exactly
20:36:22alexander92but that's what i wonder, why do i actually need to specify auto instead of untyped
20:36:35alexander92because, they get expanded anyway
20:36:59alexander92does it skip re-semchecking the expansion?
20:37:28alexander92because when i do that, the resulting type is just not the same as the "same" type in the callsite
20:44:24*Ven`` joined #nim
20:49:55*Vladar quit (Remote host closed the connection)
20:53:51juturnasI've resolved the `is not GC-safe as it performs an indirect call here` error by marking the function that is called indirectly with the `{.gcsafe.}` pragma. Is that pragma checked by the compiler, or am I setting myself up for some strange GC bugs down the road?
20:54:53Araqunless you use the ugly '{.gcsafe.}:' block (note the colon), it's all checked
20:55:23juturnasAraq great, thanks
21:03:19FromGitter<awr1> so `untyped` may actually kind of be the same thing as `auto`
21:03:21FromGitter<awr1> https://play.nim-lang.org/#ix=1UbW
21:03:42FromGitter<awr1> for templates
21:09:37*juturnas quit (Remote host closed the connection)
21:09:56*juturnas joined #nim
21:20:46*cyraxjoe quit (Quit: I'm out!)
21:29:31*Ven`` quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
21:41:28*solitudesf- quit (Ping timeout: 246 seconds)
22:12:57*zedeus quit (Ping timeout: 245 seconds)
22:13:17*zedeus joined #nim
22:32:50FromGitter<brentp> other than iteration, is there a way to get the 2nd most frequently seen key from a CountTable?
22:36:54FromGitter<brentp> hmm. i didn't realize that `largest` does a full scan: https://github.com/nim-lang/Nim/blob/devel/lib/pure/collections/tables.nim#L2278-L2288
22:43:03FromGitter<awr1> you implicitly need iteration
22:43:25*laaron- joined #nim
22:43:29*laaron quit (Remote host closed the connection)
22:44:55FromGitter<awr1> the stdlib `tables` module is missing some things i think would be useful
22:44:59FromGitter<awr1> e.g. bimaps
22:49:59*dddddd quit (Remote host closed the connection)
23:00:50shashlick@brentp - can we close https://github.com/nim-lang/nimble/issues/696
23:06:16*absolutejam quit (Ping timeout: 244 seconds)
23:10:20*absolutejam joined #nim
23:21:34FromGitter<brentp> @shashlick, it's still broken for me. I have to manually install nimble after updating nim.
23:22:30FromGitter<brentp> i'll run `git pull`, `sh build_all.sh` now and see if it's still a problem
23:27:24FromGitter<mratsim> `tables` modules is missing a BTree-backed SortedTable
23:27:36*absolutejam1 joined #nim
23:27:58FromGitter<mratsim> Araq has promised a BTree implementation in the stdlib for months :p
23:29:41*absolutejam quit (Ping timeout: 258 seconds)
23:33:45*krux02 quit (Remote host closed the connection)
23:49:23*ng0 quit (Quit: Alexa, when is the end of world?)
23:52:45*cyraxjoe joined #nim
23:53:03*jken quit (Disconnected by services)
23:53:50*jken_ joined #nim
23:59:09*jken_ is now known as jken