<< 17-12-2018 >>
00:09:02*vlad1777d quit (Ping timeout: 250 seconds)
00:12:33*onionhammer joined #nim
00:16:33*wildlander quit (Quit: Konversation terminated!)
00:19:25*lritter joined #nim
00:53:10*ng0 quit (Quit: Alexa, when is the end of world?)
01:16:40*endragor joined #nim
01:18:35*endragor quit (Remote host closed the connection)
01:24:50*theelous3 quit (Ping timeout: 246 seconds)
01:32:17*ftsf joined #nim
01:52:13FromGitter<zacharycarter> wasn't aware of - http://webserver2.tecgraf.puc-rio.br/iup/
01:52:24FromGitter<zacharycarter> but it's good to know of its existence
01:57:55FromGitter<zetashift> @shashlick you there by any chance?
02:00:30FromGitter<zacharycarter> I think it's late for him
02:00:49FromGitter<zetashift> Yeah wanted to ask a quick question, I'll just go ahead and tinker around for a bit
02:05:03FromGitter<zacharycarter> gotcha - just wanted to make sure you weren't hanging on for a reply from him
02:05:25FromGitter<zacharycarter> I need to play with scopes more...
02:07:20FromGitter<gogolxdong> How to invoke async timer and its callback in Jester?
02:07:21edcraggtoday's aoc was glorious
02:28:13*kapil____ joined #nim
02:39:52*Tyresc quit (Quit: WeeChat 2.3-dev)
03:08:49*banc quit (Quit: Bye)
03:12:19shashlickHey
03:27:17*banc joined #nim
03:33:44FromGitter<zacharycarter> @gogolxdong - do you mean how do you start the event loop? I'm not sure what timer you're referring to
03:33:49FromGitter<zacharycarter> shashlick: hi bud
03:34:22FromGitter<zacharycarter> I believe @zetashift was trying to get a hold of you earlier
03:45:40*dddddd quit (Remote host closed the connection)
04:04:39FromGitter<gogolxdong> timing for send email , async timer or async event something.
04:06:02FromGitter<zacharycarter> @gogolxdong - not sure... https://nim-lang.org/docs/asyncdispatch.html#addTimer%2Cint%2Cbool%2CCallback - maybe?
04:06:34FromGitter<zacharycarter> I'm not aware of any timer functionality built into Jester - but I've been wrong plenty of times before :)
04:06:53FromGitter<gogolxdong> frontend need responding asynchronously , backend take over the timing, thread is an option , but may cost more overhead , want to try selectors.
04:06:54*Snircle quit (Quit: Textual IRC Client: www.textualapp.com)
04:08:55FromGitter<zacharycarter> nim does have a selectors module - I have no experience with it
04:09:12FromGitter<zacharycarter> I think httpbeast uses it
04:09:30FromGitter<gogolxdong> I'm trying to combine them
04:09:46*nsf joined #nim
04:10:32FromGitter<zacharycarter> out of my wheel house - I've only ever used selectors with kqueues
04:11:13shashlickLooks like @zetashift is out - ping when you can, I'll check in the morning
04:11:39FromGitter<gogolxdong> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5c1721fb039551387fad4db0]
04:22:31FromGitter<gogolxdong> I am wondering it might be not quite proper to use IO multiplexing, timing duration is unknow and also usually long, it wastes a socket of system within the duration if the fd keeps hanging.
04:31:07FromGitter<gogolxdong> which raise a question is the file descriptor could be created the same with threads ?Is so there is no big difference between them.
04:35:40FromGitter<gogolxdong> file descriptors could be created is usually more than thread, does it mean uses selector for timing is more efficient?
04:46:22*lritter quit (Ping timeout: 250 seconds)
04:47:04*lritter joined #nim
05:03:05*endragor joined #nim
05:10:44*Notkea quit (Ping timeout: 268 seconds)
05:24:55*Notkea joined #nim
05:55:45*Notkea quit (Ping timeout: 268 seconds)
06:08:38*Notkea joined #nim
06:15:26*kinkinkijkin quit (Remote host closed the connection)
06:19:50*vlad1777d joined #nim
06:55:05*kinkinkijkin joined #nim
07:00:02kinkinkijkinhmm, can I make a seq of tuples containing a name and a proc reference ?
07:00:59Araqkinkinkijkin, yeah but it can indicate you don't use a macro where you should
07:02:07*krux02 joined #nim
07:02:10Araqall procs need to have the same signature and that's usually more burdensome than generating a 'case' statement or nnkStmtList
07:03:43kinkinkijkinthe idea is to match command names to procs using as little and as readable code as I can
07:05:43kinkinkijkinplus to be able to make named thing
07:05:49kinkinkijkineuh words are hard
07:08:02FromGitter<timotheecour> @araq please take a look at https://github.com/nim-lang/Nim/pull/10018
07:09:42kinkinkijkininstead of writing a multistage parser for this language, writing a simple parser that integrates its own expansion without increasing code complexity
07:10:30kinkinkijkinim spewing
07:10:44kinkinkijkinim having an extremely hard time explaining what i mean
07:11:13kinkinkijkini know how it would work in code but not the words to describe the code that would work for it
07:12:02Zevvstory of my life
07:18:44kinkinkijkinwhat i want to do is compare a piece of code against a large seq of all 50 or so possible commands, multiple of which are executional duplicates with only differing data, without writing a 50-entry case spanning more than a screen's worth of lines for what is ultimately a small part of the program, but instead writing a slightly more readable huge seq, which can also be formatted more before affecting the actual compiled code
07:20:44FromGitter<gogolxdong> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5c174e4c987052387efa9f55]
07:21:21FromGitter<gogolxdong> It's something like this, but needs to be correct.
07:32:19*ftsf quit (Quit: Leaving)
07:48:24*tdc joined #nim
07:51:01*tdc quit (Client Quit)
08:36:53*kapil____ quit (Quit: Connection closed for inactivity)
08:43:08FromGitter<gogolxdong> What does ⏎ ⏎ ```if fdi in s:``` ⏎ ⏎ mean ... [https://gitter.im/nim-lang/Nim?at=5c17619c0a145b0e24b27a77]
08:44:11*ng0 joined #nim
08:47:04*ftsf joined #nim
08:48:23*dddddd joined #nim
08:48:58*floppydh joined #nim
08:57:03*vegax87 quit (Ping timeout: 264 seconds)
09:01:19FromGitter<gogolxdong> ｀｀｀ ⏎ proc setData*T (s: Selector[T], fd: SocketHandle|int, data: T): bool = ⏎ let fdi = int(fd) ⏎ s.checkFd(fdi) ⏎ if fdi in s: ... [https://gitter.im/nim-lang/Nim?at=5c1765df987052387efb3999]
09:01:50FromGitter<gogolxdong> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5c1765feb4c74555ccde9717]
09:18:40FromGitter<gogolxdong> I think minimal selector usage is as follows: ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5c1769f0539aca60b22208dd]
09:23:09FromGitter<alehander42> kinkinkijkin, so you want to e.g. map commands to `call(data)` ?
09:23:31FromGitter<alehander42> how do you signify what is `data`
09:26:30FromGitter<gogolxdong> can selector register any file descriptor ?
09:27:18FromGitter<gogolxdong> Won't they overlay with system file descriptor in use?
09:27:36*ftsf quit (Ping timeout: 250 seconds)
09:40:51*PMunch joined #nim
09:42:18*lritter quit (Quit: Leaving)
09:48:08*vegax87 joined #nim
09:54:31*pbodev1 joined #nim
10:00:22*Vladar joined #nim
10:00:57*azuri5 joined #nim
10:05:52*ng0 quit (Quit: Alexa, when is the end of world?)
10:06:07PMunchHow can you call a procedure that's generic and takes a concept?
10:06:09PMunchhttp://ix.io/1wcF/Nim
10:06:22PMunchThis fails with test.nim(12, 5) Error: cannot instantiate: 'test[int]'; got 1 type(s) but expected 2
10:07:04PMunchSo I'm guessing the concept adds a generic type somehow, and that doesn't combine with the one explicitly defined one?
10:11:28FromGitter<alehander42> hm, if you instead have `[int, seq[string]]` it works indeed
10:12:10PMunchOh you're right
10:12:44*Perkol joined #nim
10:12:51PMunchSo my suspicion was right, the use of concept overloads the generic mechanism
10:12:58FromGitter<alehander42> and even if you don't have T, you can actually pass `[seq[string]]` and it works
10:13:00FromGitter<alehander42> yeah probably
10:13:45PMunchYeah but without T it doesn't require [seq[string]]
10:14:09PMunchBut when you explicitly declare one generic type it can't infer the other
10:15:43PMunchSo this fails in a similar way: http://ix.io/1wcG/Nim
10:16:12PMunchBecause while U can be inferred you can't mix inferred with explicit
10:17:35FromGitter<alehander42> one workaround is to use type params instead
10:17:36FromGitter<alehander42> e.g.
10:17:44FromGitter<alehander42> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5c1777c8039551387faf5e9d]
10:17:58FromGitter<alehander42> but it's a good question anyway
10:18:05FromGitter<alehander42> open an issue I guess (or ping @zah)
10:18:24Araqit adds an implicit generic parameter indeed
10:18:42Araqbest is to avoid explicit type instantiations via []
10:18:52PMunchYeah I can work around it like that, but that would look super strange with what I was actually using it for
10:19:38PMunchAraq, but how would you otherwise do the thing I showed. As in create an empty data structure that has a certain type?
10:20:21FromGitter<alehander42> var t: T; t
10:20:53PMunchHuh?
10:21:16PMunchSo what I was trying to do is parse some data into a data structure
10:21:42PMunchSpecifically I'm trying to load a texture atlas
10:22:48PMunchI need to get some string data, prefferably not by reading the entire string and passing it in, and return an atlas with numbers parsed into a numeric data type (this is the generic parameter)
10:24:28PMunchThe current signature: proc loadAtlas*[T, Y](atlasIterable: IterableConcept, textureLoader: TextureLoader[Y] = defaultTextureLoader): TextureAtlas[T, Y]
10:24:51PMunchSo T would be something like int, cint, or whatever integer kind you need for your graphics framework
10:25:27PMunchTextureLoader[Y] is proc (x: string): Y
10:25:48PMunchSo a way to load the texture into the kind used by your graphics framework
10:26:46PMunchAnd the IterableConcept is similar to the concept from my first snippet (just with the .lines iterator) that will be something that can be read as input
10:27:11FromGitter<alehander42> cool
10:27:57PMunchYeah I'm trying to convert my SDLGamelib into a generic gamelib that can be used with any engine
10:27:58*Perkol quit (Ping timeout: 245 seconds)
10:28:15PMunchJust something fun I tried during the weekend :P
10:34:38PMunchHmm, can I get the type of whatever got passed in as the concept?
10:37:15*abm joined #nim
10:40:01*azuri5 quit (Ping timeout: 250 seconds)
10:47:46FromGitter<mratsim> type?
10:48:21AraqPMunch, the usual workaround is a 'typedesc' parameter for what cannot be inferred
10:49:28PMunchYeah I would need to do the same call to a different procedure, so if I do the explicit generic with added concept type I would need to get that type somehow..
10:49:47PMunchBut since it's added automatically I don't have a name that I can use to pass it on with..
10:51:11*dddddd quit (Remote host closed the connection)
10:51:16FromGitter<mratsim> macro and getTypeInst might help you. I will pray for you while you move in this labyrinth :P
10:57:28PMunchHmm, I've been down that rabbit hole before. I think I'll just go with typedescs :P
10:57:50PMunchAnd create some wrappers that makes it nicer to use for specific libraries
11:07:01*Vladar quit (Remote host closed the connection)
11:30:50*ng0 joined #nim
11:58:17*stefanos82 joined #nim
12:00:46*ng0 quit (Quit: Alexa, when is the end of world?)
12:11:27PMunchnimlsp is actually suprisingly helpful
12:14:15FromGitter<alehander42> does the sublime version work well now
12:17:25*Snircle joined #nim
12:44:33PMunchI think so
12:44:40PMunchThe Vim versions at least works fine
12:56:45*jjido joined #nim
13:16:25*Tyresc joined #nim
13:24:30*azuri5 joined #nim
13:52:25*azuri5 quit (Quit: azuri5)
14:12:28*Redfoxmoon2 joined #nim
14:13:52*Redfoxmoon2 left #nim (#nim)
14:29:25*endragor quit (Remote host closed the connection)
14:30:47*narimiran joined #nim
14:32:37*leorize joined #nim
14:35:29*kapil____ joined #nim
14:36:16*nsf quit (Quit: WeeChat 2.3)
14:39:43*Vladar joined #nim
14:41:56*dddddd joined #nim
14:44:37*endragor joined #nim
14:49:08*endragor quit (Ping timeout: 250 seconds)
15:08:34*nsf joined #nim
15:11:53*azuri5 joined #nim
15:21:46*narimiran quit (Remote host closed the connection)
15:32:53*nif quit (Read error: Connection reset by peer)
15:34:55*nif joined #nim
15:37:23*azuri5 quit (Quit: azuri5)
15:39:35*azuri5 joined #nim
15:43:57*azuri5 quit (Quit: azuri5)
15:43:58l1xhey guys, what is the difference between? proc start_server()* = ; proc start_server* = ; proc start_server():* = ?
15:44:24l1xI am not sure how to export a function from a module and what is the schematic meaning between these
15:45:54*regtools quit (Ping timeout: 252 seconds)
15:46:07FromGitter<zacharycarter> ```proc start_server*() = ⏎ echo "foo"``` [https://gitter.im/nim-lang/Nim?at=5c17c4be539aca60b224589a]
15:46:13*azuri5 joined #nim
15:46:15FromGitter<zacharycarter> that will expose start_server / make it publich
15:47:20Araqyou export the name so the star always comes after the name
15:47:31l1xthanks!
15:47:35Araqand there are no variations so it's pretty easy to remember
15:47:48l1xand what is the difference between blah blah() and blah():
15:48:24FromGitter<dandevelo> Is there anyone here that studied the karax dsl code? I've noticed it doesn't have any comments but I am trying to understand the buildHtml macro: `macro buildHtml*(tag, children: untyped): VNode = ⏎ let kids = newProc(procType=nnkDo, body=children) ⏎ expectKind kids, nnkDo ⏎ var call: NimNode ⏎ if tag.kind in nnkCallKinds: ... [https://gitter.im/nim-lang/Nim?at=5c17c548c428e9539d526084]
15:48:42FromGitter<mratsim> blah is the proc, blah() is a proc call, blah(): is a proc call with an argument equal to the expression or statement in the following block
15:49:23FromGitter<dandevelo> A link might be better. I am trying to understand what this line does: https://github.com/pragmagic/karax/blob/bd69de9f146b850ca5ba61397d59a18782af0366/karax/karaxdsl.nim#L181
15:49:38l1xthanks mratsim!
15:49:42FromGitter<mratsim> it steals the body of your kids
15:50:18FromGitter<dandevelo> > it steals the body of your kids haha
15:50:54FromGitter<alehander42> body is a helper in macros.nim
15:51:07FromGitter<alehander42> which returns the body node of a proc node iirc
15:53:18FromGitter<dandevelo> Thanks @alehander42
16:06:31FromGitter<brentp> hi all. I depend on my fork of an apparently abandoned nim project. is there a way to point to my fork in my $project.nimble file?
16:07:48FromGitter<brentp> here's the project: https://github.com/manguluka/duktape-nim/
16:07:55FromGitter<alehander42> you can point to the github link
16:07:58FromGitter<Vindaar> @brentp hey! you can just write e.g. `requires "https://github.com/brentp/duktape-nim"`
16:08:37FromGitter<brentp> oh. well, that's delightfully straight-forward.
16:08:40FromGitter<alehander42> and iirc add `#head` or `#<tag>` etc
16:08:46FromGitter<alehander42> if you need
16:08:58FromGitter<brentp> cheers
16:10:30shashlickHow do you know the author abandoned it?
16:11:04FromGitter<brentp> I don't, but he has no activity on github since august and doesn't respond to PR.
16:11:18FromGitter<brentp> he/she (/shrug)
16:13:11PMunchWhich project?
16:13:35FromGitter<brentp> project and PR here: https://github.com/manguluka/duktape-nim/pull/2
16:15:34PMunchHmm, last edited 5 months ago..
16:17:20PMunchHm, I think it should be possible
16:17:25PMunchBut no idea how or if
16:18:55l1xhow do I use the $ thing with logging?
16:19:43l1xinfo("$datetime Starting web server...") does not work
16:23:12PMunchWhat are you trying to do?
16:23:34l1xtrying to log date + time + message like you would normally do with logging?
16:24:21PMunchWell you could do `info($datetime & " Starting web server...")` assuming info is declared somewhere
16:24:28l1xthanks
16:24:33l1xso you need to do &
16:24:41PMunchOr you could use the strformat module: nim-lang.org/docs/strformat.html
16:24:43l1xit would be nice addition here: https://nim-lang.org/docs/logging.html
16:25:08PMunch$ in nim is "convert to string", & is concatenate strings
16:25:38l1xthis does not work though -> info($datetime & "Starting web server...")
16:25:50l1xError: undeclared identifier: 'datetime'
16:26:34PMunchOh wait
16:26:49PMunchSorry, I didn't know you were using the logging module
16:27:39*endragor joined #nim
16:29:38*regtools joined #nim
16:29:43*azuri5 quit (Quit: azuri5)
16:30:39shashlick@brentp: well let me know if you need any help - that's a nimgen based project
16:31:22FromGitter<brentp> @shashlick will do. is your new project doing to replace nimgen?
16:32:08*endragor quit (Ping timeout: 245 seconds)
16:33:30PMunchl1x, http://ix.io/1wef/Nim
16:33:33*pbodev1 left #nim (#nim)
16:33:47PMunchIt wasn't particularily clear from the documentation, but format strings have to be set for the logger
16:35:03PMunchSo everything that's logged will follow the same log pattern
16:42:08*vlad1777d quit (Ping timeout: 245 seconds)
16:47:12*PMunch quit (Remote host closed the connection)
16:49:14*floppydh quit (Quit: WeeChat 2.3)
16:51:44*tdc joined #nim
16:52:50*whaletechho joined #nim
16:54:06*azuri5 joined #nim
16:55:07*Trustable joined #nim
16:58:24*zahary quit (Quit: Connection closed for inactivity)
16:59:46*endragor joined #nim
17:03:18*endragor quit (Remote host closed the connection)
17:11:00*krux02 quit (Remote host closed the connection)
17:29:45*azuri5 quit (Quit: azuri5)
17:34:15FromGitter<zetashift> @shashlick I had had a weird error installing your x64 windows patch for choosenim, but it worked out thanks!
17:36:07*jjido quit (Ping timeout: 240 seconds)
17:37:49*azuri5 joined #nim
17:37:55*quipa joined #nim
17:41:00*endragor joined #nim
17:42:20*endragor_ joined #nim
17:45:03*endrago__ joined #nim
17:45:56*endragor quit (Ping timeout: 272 seconds)
17:46:43*endragor_ quit (Ping timeout: 245 seconds)
17:53:25shashlick@brentp: nimgen will live on for the near term
17:53:42shashlickTo maintain existing wrappers
17:54:07shashlickAnd anyone wanting to use c2nim
17:54:30shashlickNimterop will absorb more of the functionality over time though
17:55:31FromGitter<arnetheduck> l1x check out https://github.com/status-im/nim-chronicles if you want a more structured approach to logging
17:55:34shashlick@zetashift: still have to fix up per dom96 feedback - please let me know what you found
17:58:11*nsf quit (Quit: WeeChat 2.3)
18:02:13FromGitter<zetashift> so far smooth sailing, the error was on my part
18:04:57*quipa quit (Read error: Connection reset by peer)
18:07:59*endrago__ quit (Remote host closed the connection)
18:10:34shashlickI hope we get to a point where you can simply download a binary build for Linux with musl and osx as well
18:10:57*craigger_ quit (Quit: bye)
18:10:59shashlickPerhaps we can get there with @kaushalmodi's help
18:13:17FromGitter<kaushalmodi> shasklick: You make it sound like I am good at Linux binary packaging
18:13:35*craigger joined #nim
18:13:37shashlickYou're good at musl :)
18:13:46FromGitter<kaushalmodi> All I have is https://github.com/kaushalmodi/hello_musl, which is a distillation of everything that works(TM), after a lot of failed experiments
18:14:17FromGitter<kaushalmodi> I got a lot of help from folks here to pick the right switches :)
18:16:13shashlickDo you think it's possible to have all Linux Nim binaries built with musl?
18:16:43FromGitter<kaushalmodi> I haven't attempted that
18:17:03FromGitter<kaushalmodi> before I even venture into that, need to tackle the Travis failures when building Nim nightlies
18:17:36FromGitter<kaushalmodi> *just thinking of that gives me creeps* - I hate Travis
18:17:45l1xthanks!
18:17:55*azuri5 quit (Quit: azuri5)
18:18:40FromGitter<kaushalmodi> shashlick: before we get to static binaries for Nim, would like to get the static binary of choosenim
18:21:31shashlickAgreed
18:21:53shashlickAnd nimble which should eventually become independent of Nim
18:25:25FromGitter<arnetheduck> just pushed an update to `nlvm` to do `llvm` 7.0 which technically includes a linker - once `nlvm` uses that, nim could be used without a c compiler as well
18:25:53FromGitter<arnetheduck> (right now, nlvm calls gcc to do linking because it's easier to get the linking options right that way)
18:26:09FromGitter<kaushalmodi> shashlick: yes, nimble too
18:36:40*druonysus joined #nim
18:36:40*druonysus quit (Changing host)
18:36:40*druonysus joined #nim
18:42:21*azuri5 joined #nim
18:47:02*dddddd quit (Ping timeout: 250 seconds)
18:47:04FromGitter<Varriount> @arnetheduck Nice!
18:50:27FromGitter<zacharycarter> @arnetheduck - sweeeeet
18:51:06kinkinkijkinsorry, I went to bed shortly after sending my last thing, @alehander42 yeah call(data)
19:27:17*azuri5 quit (Quit: azuri5)
19:27:17*Cthalupa quit (Ping timeout: 268 seconds)
19:28:02*Cthalupa joined #nim
19:48:48Araqso ... if I say --import: "module" in a config.nims file does that mean other config.nims files should try to import this module implicitly?
19:49:34Araqthat's a switch for Nim, not for NimScript, right? and yet that's what Nimble uses it for, albeit via the compiler API
19:49:43*Araq hates --import
20:02:00FromGitter<arnetheduck> declarative config files don't have such issues :)
20:07:18Araqwell they are simpler for sure, on the other hand, they also had their own bugs
20:08:07Araqphase ordering problems, if file A says --debugger:on and file B says --debugger:off who wins
20:08:14FromGitter<arnetheduck> I'm not smart enough for anything else than simple
20:08:46Araqmy favourite solution is a Nim program that constructs the command line for the Nim compiler. ;-)
20:09:15FromGitter<arnetheduck> command likes are very declarative, works for me :)
20:09:26Araqsimpler than NimScript, more powerful too
20:09:37Araqand in the end you can look at the command line to see where the shit happens
20:10:28FromGitter<arnetheduck> and you have to fit everything in 65k characters which is a bonus on its own.. good incentive that you're doing something wrong, once you hit it
20:10:59Araqthe limit is 8K on windows but fortunately "response files" are a very common workaround
20:13:18FromGitter<arnetheduck> btw, a feature req for testament if you happen to be working on it: a response file with a test blacklist, so I can disable a bunch of hopeless tests in nlvm (those that emit c code and the like).. if I do that, nlvm is probably at some 98% test success right now, meaning I could actually do a release
20:16:14*kapil____ quit (Quit: Connection closed for inactivity)
20:26:22l1xhmm how is osproc.execProcess supposed to be working?
20:26:45l1xlet outp = execProcess(command = "echo 1") this returns a TaintedString
20:26:52l1xaccording to the doc
20:27:51FromGitter<zacharycarter> Araq: what you're describing (I think) is what Zig does
20:28:14FromGitter<zacharycarter> https://github.com/zacharycarter/zeal_zig/blob/master/build.zig
20:28:23FromGitter<zacharycarter> in fact build is a module in the stdlib
20:33:29*dddddd joined #nim
20:37:01*Trustable quit (Remote host closed the connection)
20:42:47Araqzacharycarter: ok, well, good to see others learn from my mistakes
20:43:18*Jesin quit (Quit: Leaving)
20:43:57Araqthough for your example it's probably wrong because then it's tied to the package and clients of your code need to copy&paste your link commands
20:44:26FromGitter<zacharycarter> yeah - that's a good point
20:44:47Araqthat's why most of this stuff should be in pragmas
20:44:55Araqbecause then it composes well
20:45:50*Jesin joined #nim
20:46:45Araqarnetheduck: that should be covered by 'disabled: nlvm' in the spec section IMO
20:51:34FromGitter<arnetheduck> it could but that's kind of messy since we work in separate repos
20:54:52*zachk joined #nim
20:55:57*zachk quit (Changing host)
20:55:57*zachk joined #nim
21:05:31*theelous3 joined #nim
21:08:54*stefanos82 quit (Remote host closed the connection)
21:37:48Araqwell a blacklist doesn't offer much over copying the tests/ dir and leaving out certain files
21:41:03*Jesin quit (Quit: Leaving)
22:21:20*Vladar quit (Remote host closed the connection)
22:42:23FromGitter<arnetheduck> true, I could rsync the tests somewhere, it's just an extra step (right now I just run testament on the Nim folder but with nlvm instead of nim - very convenient)..
22:43:23*smitop joined #nim
22:43:47FromGitter<arnetheduck> hm wonder if I can make that depend on the git submodule version, then it could even be part of the makefile I use.. hm
22:44:37FromGitter<arnetheduck> ... but that won't work on windows ... sigh
23:07:26*xet7 quit (Ping timeout: 268 seconds)
23:07:33*abm quit (Quit: Leaving)
23:14:06*xet7 joined #nim
23:30:39*Jesin joined #nim
23:51:49shashlickI prefer pragmas too, but it will help to have per file vs global pragmas
23:52:02shashlickLike a passC that only applies to the current file
23:52:22shashlickI really would also like a define and undefine pragma