<< 19-01-2018 >>
00:00:32*endragor joined #nim
00:03:28*yglukhov quit (Remote host closed the connection)
00:09:10*gokr quit (Ping timeout: 260 seconds)
00:16:03*devdri quit ()
00:17:00*fredrik92 quit (Quit: Client Disconnecting)
01:03:07*MJCaley quit (Quit: MJCaley)
01:05:58*vivus joined #nim
01:16:47*xet7 quit (Ping timeout: 248 seconds)
01:30:14FromGitter<honewatson> Do concepts work in the js backend?
01:31:38*MJCaley joined #nim
01:32:54*xet7 joined #nim
01:33:20dom96honewatson: They're just generics, so probably
02:07:42*vivus quit (Quit: Leaving)
02:23:53*Snircle quit (Quit: Textual IRC Client: www.textualapp.com)
02:25:07*vlad1777d quit (Ping timeout: 268 seconds)
03:00:26*chemist69 quit (Ping timeout: 255 seconds)
03:14:08*chemist69 joined #nim
03:21:05*demi- quit (Quit: Server shutdown in 3... 2... 1...)
03:54:39*wurui joined #nim
03:55:10*marenz_ joined #nim
03:59:09*marenz__ quit (Ping timeout: 264 seconds)
04:00:37*wurui quit (Remote host closed the connection)
04:01:37*endragor quit (Remote host closed the connection)
04:02:06*endragor joined #nim
04:10:44*MJCaley quit (Quit: MJCaley)
05:40:38*yglukhov joined #nim
05:45:08*yglukhov quit (Ping timeout: 255 seconds)
05:57:11FromGitter<data-man> @dom96: Print on a flyer an example from recently merged #6936 (Thanks, @Araq!) ⏎ I can be wrong, but I suppose that there is no such feature in any language (in a standard library) :)
05:59:28FromGitter<data-man> @PMunch: Should drawille.nim also be printed? ;)
06:00:40*Amrykid joined #nim
06:09:14*Amrykid quit (Ping timeout: 276 seconds)
06:09:27*kier quit (Ping timeout: 240 seconds)
06:11:50*dddddd quit (Remote host closed the connection)
06:25:51*nsf joined #nim
06:49:02*xet7 quit (Ping timeout: 255 seconds)
06:50:11*marenz_ quit (Ping timeout: 276 seconds)
07:01:58*xet7 joined #nim
07:22:00*xet7 quit (Ping timeout: 260 seconds)
07:23:00*gokr joined #nim
07:33:50*devdri joined #nim
07:36:42*xet7 joined #nim
07:39:34*gokr quit (Ping timeout: 256 seconds)
07:50:23*arnetheduck quit (Ping timeout: 248 seconds)
07:52:05*yglukhov joined #nim
07:54:13*arnetheduck joined #nim
07:55:25*gokr joined #nim
07:57:14*yglukhov quit (Ping timeout: 276 seconds)
07:57:48*devdri quit ()
08:10:18*Vladar joined #nim
08:12:30*devdri joined #nim
08:28:07*Jipok[m] quit (Ping timeout: 248 seconds)
08:28:47*trincyolo quit (Ping timeout: 265 seconds)
08:35:56*devdri quit ()
08:39:32*xkapastel quit (Quit: Connection closed for inactivity)
08:39:35*floppydh joined #nim
08:42:09*Jipok[m] joined #nim
08:43:09*trincyolo joined #nim
08:48:41*sendell joined #nim
09:05:20*PMunch joined #nim
09:08:21*yglukhov joined #nim
09:08:41*yglukhov quit (Remote host closed the connection)
09:08:57*yglukhov joined #nim
09:10:58PMunchdata-man, didn't you see my proposed game of life implementation? It includes Drawille :)
09:12:58FromGitter<data-man> Which a link is last?
09:14:08PMunchhttp://ix.io/EkC/
09:15:17FromGitter<data-man> Yes, I saw. But Drawille not included. :-)
09:15:27PMunchWhat do you mean not included?
09:16:26PMunchThe more I thought about it though the more I like @alehander42's idea of having multiple versions. Nim just has too much to show off :) I'm thinking maybe one example showing macro usage (Simple jester server with JSON perhaps?), a JS-backend example creating a website (maybe pulling data from the Jester example, or a deployed sample server implementing the same thing), and the game of life example. It would be cool to also get a macro implementation but I
09:16:26PMunchcan't think of anything you'd be able to fit in that size with an example..
09:16:33FromGitter<data-man> Drawille.nim has a 284 lines
09:17:48PMunchYou want to print out the entire thing? :P
09:19:42*vlad1777d joined #nim
09:20:00FromGitter<data-man> No, just don't use.
09:20:01*GaveUp quit (Ping timeout: 248 seconds)
09:20:36PMunchOh you want me to not use Drawille?
09:21:26PMunchYou could implement this like dom96 did with a regular ASCII character, but with much lower resolution
09:24:26FromGitter<data-man> Yes, to not use on a flyers, because the implementation details are not visible.
09:30:27*kier joined #nim
09:31:50*GaveUp joined #nim
10:09:44PMunchWell this works: http://ix.io/Emh/
10:14:24*yglukhov quit (Ping timeout: 268 seconds)
10:14:33FromGitter<data-man> πŸ’― :)
10:19:18PMunchOr if you want to get really crazy about it: http://ix.io/Emi/
10:24:26FromGitter<data-man> Maybe to put "Nim" to a board?
10:26:32FromGitter<data-man> os not needed
10:26:54livcdwhat are you guys trying to do ?
10:27:24PMunchdata-man, yeah I had os there to use sleep. But I removed it for those precious lines :P
10:27:46PMunchlivcd, dom96 wanted a sample to put on the back of the Nim flyers for FOSDEM
10:28:10PMunchSo we're miniaturizing a game of life example to fit
10:28:49livcdah :D
10:28:51livcdnice
10:29:55PMunchBut I don't think it can get much smaller now and still be readable :P
10:37:00Araqyou need useful helpers
10:38:31Araqgame of life can written shorter and more precise with "for all board fields" count("probe positions environment without bound errors")
10:41:43PMunchWell that is pretty much what this is doing
10:42:25PMunchBut you want to split it off into helper procedures? Not sure how that would be shorter..
10:42:33PMunchUnless you pull a go and hide everything in a libaryr
10:42:35PMunch:P
10:45:13FromGitter<data-man> @Araq: I implemented some useful procs for lists: isEmpty, len, count, peekFirst, peekLast, popFirst, popLast and append for SinglyLinkedList (popXXX/peekXXX returns a value). ⏎ To create a PR?
10:46:52AraqPMunch: yep, hide it in another library
10:46:57Araq:P
10:47:10Araqthese primitives might be generally useful though
10:47:36Araqso it's not a fake library. however, we need a DSL for that to show off :P
10:48:00Araqdata-man: lists? ugh
10:48:40FromGitter<data-man> lists.nim in collections :)
10:49:01PMunchYeah it's hard to show off all the cool parts in such a short example
10:49:14PMunchThat's why I'm starting to like alehander42's suggestion more and more
10:50:26PMunchJust create 3-4 different versions of the flyer with different code samples. For example a web-dev might be more interested in the JS backend and HTML stuff, while a server person might be interested in seeing Jester (plus it's a good example of DSLs), and then a general game of life example just to show some "regular" syntax.
10:50:28dom96I'm starting to wonder whether it would be better to have a larger example
10:51:13dom96So that it's easier for the reader to understand the code
10:51:38dom96and as a side effect it would show off procedures/types as wlel
10:51:39dom96*well
10:51:48dom96Something along these lines? https://gist.github.com/dom96/31f17b06874569a00108311a4c7a9139
10:52:50Araqprocs and types are boring though
10:53:32dom96not to people unfamilliar with the language
10:55:03FromGitter<andreaferretti> Yeah, looks boring
10:55:12FromGitter<andreaferretti> Does not really show the potentiality of the language
10:56:28PMunchWell it shows how types can be pretty complex
10:57:07PMunchBut I agree that the increase in size doesn't really lead to a large increase in wow-factor
10:57:16PMunchJust looks like you need more boilerplate :P
10:58:11PMunchIf we did multiple flyers it would be possible to have one that was like something you would find on https://learnxinyminutes.com/docs/Nim just showing off some cool features without any real purpose
11:01:20*yglukhov joined #nim
11:01:50Araqthen people need to collect flyers
11:01:59Araqto see all examples.
11:02:14Araqwe need more flyers then and should start to sell them.
11:02:54Araqand every flyer should be in an envelope so people can't see what version they are buying
11:03:18AraqNim Pokemon.
11:03:18FromGitter<data-man> Just to print https://github.com/def-/nim-units/blob/master/units.nim :)
11:03:32PMunchHaha, great idea Araq!
11:03:52PMunchThen we can design the examples to work together and "win" over each other :P
11:05:24Araq"my allocation free game of life implementation beats your implementation that uses 'seqs'"
11:05:42PMunchdata-man, what is this sorcery?!
11:06:40FromGitter<data-man> This is Nim's magic
11:06:56*arnetheduck quit (Ping timeout: 276 seconds)
11:07:33PMunchI know, it's just really cool!
11:07:41PMunchNever seen that file before
11:08:07dom96A Nim trading card game could actually be a pretty innovative way to teach people Nim :P
11:08:22dom96I think we should think about that separately
11:09:12dom96I still like the initial idea I had: a quick start guide at the back (Step 1. Install Nim via choosenim. Step 2. Compile hello world. Step 3. Learn Nim)
11:09:27dom96Step 1 would have the beautiful choosenim one-liner
11:09:36dom96step 2 would have `nim c -r hello.nim`
11:09:42dom96and step 3 would have links to the tutorial and docs
11:11:13Araqdom96: why not cheat sheet instead?
11:11:33Araq"keep this next to your computer"
11:11:53dom96We can do that too
11:11:59dom96Although my book already contains a cheat sheet :P
11:13:11FromGitter<data-man> @PMunch: https://github.com/jlp765/units
11:14:31Araqdom96: the better, then we don't have to come up with one
11:14:33PMunchdom96, but you won't be handing out your book at FOSDEM :P
11:14:47Araqand not everybody who gets a flyer also buys your book
11:24:32*skelett quit (Ping timeout: 256 seconds)
12:00:58*vlad1777d quit (Ping timeout: 265 seconds)
12:13:16*Yardanico joined #nim
12:13:45Yardanicogood day/night/afternoon/morning everyone
12:14:54PMunchGood mid-day to you too Yardanico
12:15:13Yardanicoyeah, I have mid-day here too, forgot about it :P
12:15:16mirangood afternoon Yardanico
12:24:40Yardanicodom96, about game of life for a flyer - we can probably change four lines after setCursorPos(stdout, x, y) into one - stdout.write(if board[x][y]: "●" else: "") ? I don't think that this removes readability
12:24:49Yardanico(I took the code from your deepCopy issue)
12:25:00*PMunch_ joined #nim
12:25:38Yardanicostdout.write if board[x][y]: "●" else: "" will also work, but it looks worse IMO :)
12:26:08*earenndil quit (Ping timeout: 248 seconds)
12:26:40*kunev quit (Ping timeout: 248 seconds)
12:26:40*oprypin quit (Ping timeout: 248 seconds)
12:27:25*oprypin joined #nim
12:27:28Yardanicoin fact if board[x][y]: x else: y is already used on line 12
12:27:38*earenndil joined #nim
12:27:43*PMunch quit (Ping timeout: 248 seconds)
12:27:44*gmpreussner quit (Ping timeout: 248 seconds)
12:28:54*Ven`` joined #nim
12:28:56*gokr quit (Ping timeout: 276 seconds)
12:29:03*kunev joined #nim
12:29:05*gmpreussner joined #nim
12:31:01livcddom96: where is the cheat sheet ? :-|
12:32:58dom96livcd: the inside covers
12:33:40dom96we need to decide what to put on the back cover soon...
12:33:44dom96Not much time left guys
12:33:49*xet7 quit (Remote host closed the connection)
12:41:37Yardanicodom96, so the idea with game of life is still here?
12:41:52dom96yeah
12:41:56livcdah lol the mini cheatsheet
12:41:56livcdok
12:42:01dom96but many other ideas have been put forward
12:43:59*PMunch_ is now known as PMunch
12:44:34*gokr joined #nim
12:44:51PMunchI like the idea of multiple examples. But if that is not feasible then either GOL or cheatsheet is good
12:47:44FromGitter<data-man> https://github.com/nim-lang/Nim/blob/devel/tests/generics/tthread_generic.nim is a good example
12:52:02*xet7 joined #nim
13:00:56*skelett joined #nim
13:12:05FromGitter<tim-st> Does anyone know if it's possible to save a `RuneEnumeration` used in `case ... of RuneEnumeration` to use it later again in a deeper case? (Where RuneEnumeration is something like `Rune('a')..Rune('z'), Rune('.')`
13:14:27*svennidal joined #nim
13:19:26FromGitter<tim-st> Ok, I got it working, I have to use brackets in the declaration
13:20:30*dddddd joined #nim
13:21:46FromGitter<tim-st> Hm, seems compiler doesnt accept const "Slices of Rune" for case check
13:25:42FromGitter<alehander42> @Araq might be kidding but this "collecting flyers" thing is exactly what I had in mind :D
13:25:57*gokr quit (Quit: Leaving.)
13:26:31FromGitter<alehander42> anyway, just a GOL example seems bad to me, as @andreaferretti said, it doesn't really demonstrate interesting stuff
13:26:36FromGitter<alehander42> enough*
13:27:15FromGitter<alehander42> (it demonstrates syntax & cool libs as drawille, but there is much more)
13:31:23*marenz_ joined #nim
13:35:55Yardanicobut how you would but everything on a small flyer? :)
13:36:29dom96Nim Flyers! Gotta catch 'em all!
13:36:50dom96we can't demonstrate cool stuff effectively on a flyer
13:37:11FromGitter<alehander42> that's why you have 3-4 variants of flyers, c'mon, *diversity*
13:37:17FromGitter<alehander42> :D
13:37:42FromGitter<alehander42> yeah, btw, do people do workshops or something like that?
13:38:33FromGitter<alehander42> something like introducing groups of people for 30 minutes would be pretty cool
13:39:04*Ven`` quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
13:39:22dom96They probably do
13:39:27dom96but we just have a stand at this one
13:40:12dom96and we do plan to have laptops with something on them
13:40:29*gangstacat quit (Quit: Ĝis!)
13:40:51dom96and maybe some embedded devices (any new ideas on that federico3?)
13:41:06dom96People in general will be able to talk to us for as long as they want
13:42:16federico3something that blinks, something with a display and if I could find the time a little sensor + oled
13:43:20*Vladar quit (Quit: Leaving)
13:44:04*arecaceae quit (Remote host closed the connection)
13:44:23*arecaceae joined #nim
13:45:01dom96awesome
13:45:40dom96alehander42: if you want 3-4 variants then please come up with what should be on those variants :)
13:46:20FromGitter<alehander42> that's fair, I'll try to think of something this week
13:46:48dom96okay, I'll probably print some sort of flyers ahead of that
13:46:59FromGitter<alehander42> ofc
13:52:07FromGitter<Quelklef> How bad is it to render JSON manually rather than by the json module?
13:52:42FromGitter<Quelklef> I feel like it's not a big deal but OTOH maybe it's not as simple as I thnk
13:53:32dom96why not just render it using the json module?
13:54:02FromGitter<Quelklef> Adds a dependency and seems like a pain
13:54:20dom96it's the very opposite of a pain
13:54:21*Snircle joined #nim
13:54:24FromGitter<Quelklef> But mostly feels like overkill, maybe? Which is why I'm asking
13:54:28dom96formatting a JSON string yourself is a pain
13:54:44FromGitter<RedBeard0531> JSON is pretty simple. The only real annoying bit is having to do the utf16-style surrogate pare escaping in strings.
13:55:16FromGitter<Quelklef> Ah, there we go. Thanks @RedBeard0531, hadn't thought of that
13:55:17dom96The only dependencies that are worth worrying about is C library dependencies
13:55:38FromGitter<Quelklef> Really?
13:56:01*gangstacat joined #nim
13:56:12FromGitter<Quelklef> Oh right only procs that are actually called are compiled, right?
13:56:28dom96yes
13:57:15*rokups joined #nim
14:01:57*Snircle quit (Ping timeout: 246 seconds)
14:02:06*Vladar joined #nim
14:05:11*Snircle joined #nim
14:05:21*Jesin quit (Quit: Leaving)
14:07:03FromGitter<RedBeard0531> Are closure iterators supposed to work on the js backend?
14:07:42Araqno, they are not implemented
14:08:42FromGitter<RedBeard0531> Ok, I was getting a weird error "jsiter.nim(5, 9) Error: internal error: symbol has no generated name: iter" and I couldn't tell if I was doing something wrong or not
14:09:12*Yardanico quit (Remote host closed the connection)
14:10:54*Yardanico joined #nim
14:12:09*Amrykid joined #nim
14:18:47*Yardanico quit (Remote host closed the connection)
14:25:26*Yardanico joined #nim
14:26:58*Snircle_ joined #nim
14:28:52*Snircle quit (Ping timeout: 265 seconds)
14:33:16FromGitter<data-man> ```code paste, see link``` ⏎ ⏎ How get a high index for T? [https://gitter.im/nim-lang/Nim?at=5a6201acce68c3bc74da087c]
14:33:43*juanfra joined #nim
14:34:55FromGitter<Yardanico> ??
14:34:58FromGitter<Yardanico> maybe for "v1" ?
14:35:22YardanicoDid you try v1[^1] ? :)
14:35:30Yardanicoand v[^1]
14:39:06FromGitter<data-man> I'm trying to add support for arrays and tuples to scanf macro
14:51:13FromGitter<data-man> Current implementation https://gist.github.com/data-man/5b80bdec995d01164beccd80ddf0bcfe
14:52:14FromGitter<data-man> ```code paste, see link``` ⏎ ⏎ A test code [https://gitter.im/nim-lang/Nim?at=5a62061e517037a212c9218c]
14:56:04*miran left #nim (#nim)
14:56:33FromGitter<data-man> Is a bad idea? :)
15:02:43*nsf quit (Quit: WeeChat 2.0.1)
15:07:51*xkapastel joined #nim
15:08:07*MJCaley joined #nim
15:14:56FromGitter<krux02> I have non deterministic crashes in the compiler
15:14:58FromGitter<krux02> yay
15:15:15FromGitter<krux02> it crashes reproducably, but not reproducable at the same position
15:17:05dom96interesting, macro problems?
15:17:10FromGitter<krux02> yes
15:17:26FromGitter<krux02> the vm is hell in Nim
15:17:41FromGitter<krux02> undebuggable and a mess in the codebase
15:18:15FromGitter<krux02> SIGSEGV: Illegal storage access. (Attempt to read from nil?)
15:18:41FromGitter<krux02> that is the error when I compile with 'koch test run ...'
15:18:53FromGitter<krux02> but when I compile with nim_temp, then the error message is much different
15:19:22FromGitter<krux02> first of all it fails much earlier
15:20:13FromGitter<krux02> and then it has the error message: ```
15:20:15FromGitter<krux02> ####bar #### ⏎ ⏎ nim.nim(121) nim ⏎ nim.nim(77) handleCmdLine ⏎ main.nim(172) mainCommand ... [https://gitter.im/nim-lang/Nim?at=5a620cafce68c3bc74da5082]
15:21:28FromGitter<krux02> system.nim 2833 is the sysFaltal proc
15:21:45FromGitter<krux02> and vm.nim(1257) is a local variable of type int
15:21:50FromGitter<krux02> not a node or anything
15:22:37*vivus joined #nim
15:28:28FromGitter<krux02> nope it is of type TRegister
15:28:40Araqwhat's messy about it?
15:29:35FromGitter<krux02> why do I always have to asskick the compiler before it tells me anything useful about what went wrong?
15:30:35FromGitter<naelt> Hello, i try nim's ffi for a c library function crypt_newhash on openbsd. The signature of the c function is:
15:31:07Araqkrux02: so you get a stack trace pointing to the bug. I wouldn't call this "undebuggable"
15:33:52FromGitter<krux02> I am stuck and there not a single instance of `intVal` in the stack
15:34:00FromGitter<krux02> at least I did not find it yeat
15:34:52FromGitter<krux02> currently i fail at implementing `resolveAlias`
15:35:21FromGitter<krux02> Araq: do you understand the problem with `getTypeImpl` and `getTypeInst`?
15:35:49AraqI guess, but it's stupid
15:35:57FromGitter<krux02> why is it stupid?
15:36:13Araqthere are multiple ways to spell a type and you only accept a single spelling
15:36:33FromGitter<krux02> what do you mean?
15:36:40Araqyou're after some canonical form that is not specified anywhere
15:37:13FromGitter<krux02> is there a problem with that?
15:38:14FromGitter<krux02> I specify the type that I want to accept
15:38:56Araqwell since it's not specified, of course it's problematic
15:39:31Araqyou want the Vec4[float] from 'type Vec4f = Vec4[float]' iirc
15:39:51*miran joined #nim
15:40:18Araqbut the compiler doesn't even have an obligation to keep this information, nowhere does the spec say the compiler is obliged to keep temporary instantiations around for getType()
15:40:22FromGitter<naelt> Hello, i try nim's ffi for a c library function crypt_newhash on openbsd. The signature of the c function is: ⏎ ⏎ ```code paste, see link``` ⏎ ⏎ I get a bus error when i try read the hash variable, which must be the password in hashed form. The return value of my wrapper is 0, which indicate a successful return. Has sombody a experience with it? Thanks [https://gitter.i
15:40:22FromGitter... m/nim-lang/Nim?at=5a6211655a9ebe4f75aa4062]
15:40:24FromGitter<krux02> no I want the Vec[4,float32], but good enough I guess
15:41:30FromGitter<krux02> then what should I do? pattern match the object implementation?
15:42:14Araqthat's one solution, I would match against the Vec4f alias, but that has the same problem I'm complaining about ;-)
15:42:59FromGitter<dom96> @naelt: char * == cstring, not 'ptr cstring'
15:43:19FromGitter<krux02> I guess you know what I try to do
15:44:15FromGitter<krux02> I want to translate a block of nim code `let x = vec4(1,2,3,4)` to glsl `ivec4 x = ivec4(1,2,3,4)` in a macro
15:44:29FromGitter<krux02> but for that I need to know the type of each node and map it to glsl
15:45:00FromGitter<krux02> and there are a lot of names for the same type as you already
15:45:02FromGitter<krux02> said
15:45:19FromGitter<krux02> meaning I need to normalize the types to one form
15:45:31FromGitter<krux02> the only form that I get, is getTypeImpl
15:45:39FromGitter<krux02> but then I loose the name of the type
15:45:55FromGitter<krux02> meaning any type with the same implementation would match, too
15:45:57FromGitter<krux02> I don't like that
15:50:26FromGitter<krux02> Araq: If you could tell me an alternative how I can do it, I can agree that what I want is stupid. But as long as I don't get any idea on how to do it otherwise I try my best at punching `resolveAlias` through
15:50:40FromGitter<krux02> because I don't see an alternative
15:51:18FromGitter<krux02> as I see it, the type is some sort of list internally
15:51:20FromGitter<naelt> @dom96 thank you, i removed ptr and addr. now i receive SIGSEGV: Illegal storage access
15:52:10dom96check what crypto_newhash returns
15:52:14dom96maybe it gives an error?
15:52:25FromGitter<krux02> @naelt: when you made changes show them. Otherwise we can't help.
15:53:39*MJCaley quit (Quit: MJCaley)
15:53:39euantor@dom96 He said it returned 0 which is success
15:53:42Araqnaelt: pass it a valid buffer
15:54:02euantorI think `size_t` should be `cint` rather than `uint16` too, right?
15:54:21Araqreturn value of 0 is irrelevant when you have a corruption
15:54:36euantorand do `var hash: cstring = cstring(newString(127))`
15:54:48FromGitter<krux02> Araq: what was the argument to koch, so it builds the compiler in debug mode
15:54:59Araqkoch temp c foo.nim
15:55:08FromGitter<krux02> when I run koch temp it misses debug information
15:56:17FromGitter<krux02> so how do I have a debug build that allows me to use a debugger on it?
15:58:12FromGitter<naelt> many thanks `var hash: cstring = cstring(newString(127))` do it :)
15:58:33*MJCaley joined #nim
16:00:24FromGitter<krux02> Araq: how do I add the debug symbols to the compiler?
16:03:41Araq--debugger:native
16:03:50Araqkoch temp --debugger:native c foo.nim
16:04:00Araqsee you in a couple of days
16:04:10FromGitter<krux02> then what is--debuginfo?
16:04:53Araq--debugger:native is --debuginfo --lineDir:on
16:04:56Araqbye
16:06:59*Amrykid quit (Ping timeout: 265 seconds)
16:10:08*MJCaley quit (Quit: MJCaley)
16:13:48*MJCaley joined #nim
16:19:28*azur_kind joined #nim
16:24:02*azur_kind quit (Client Quit)
16:36:28*endragor quit (Remote host closed the connection)
16:36:38*rokups quit (Quit: Connection closed for inactivity)
16:36:57*endragor joined #nim
16:39:24*endragor_ joined #nim
16:40:54*Snircle_ quit (Read error: Connection reset by peer)
16:41:27*endragor quit (Ping timeout: 240 seconds)
16:41:58*Snircle joined #nim
16:43:53*endragor_ quit (Ping timeout: 256 seconds)
16:51:37*Snircle quit (Read error: Connection reset by peer)
16:51:39*floppydh quit (Quit: WeeChat 2.0.1)
16:52:20*Snircle joined #nim
16:54:51*Jesin joined #nim
16:57:51FromGitter<dandevelo> Anyone else getting unhandled exception: Unknown IO Error [IOError] when trying to compile?
17:01:01FromGitter<dandevelo> It seems this started happening after updating my Windows 10 VM to the latest and greatest update. Curios if anyone else experienced this and how did you solve it
17:01:13FromGitter<dandevelo> It seems like it works eventually if I try it multiple times
17:02:00FromGitter<dandevelo> I found a similar report here: https://forum.nim-lang.org/t/3367
17:05:21FromGitter<zetashift> Do you also use VSCode?
17:05:41FromGitter<dandevelo> Yes
17:07:05*PMunch quit (Ping timeout: 240 seconds)
17:08:51FromGitter<Bennyelg> x_session = requests.Session() ⏎ x_session.verify = '/Applications/abc.crt' ⏎ auth={'auth':requests.auth.HTTPBasicAuth(okta_user,okta_password)} ⏎ conn = presto.connect(host="xxx.com", port=8443, protocol='https', ⏎ requests_kwargs=auth, requests_session=x_session) [https://gitter.im/nim-lang/Nim?at=5a6226236117191e61886b4e]
17:09:04FromGitter<Bennyelg> Any way to do it with nim httpclient ?
17:11:54FromGitter<Yardanico> yes but IDK what is presto
17:12:23FromGitter<Yardanico> httpclient supports basic auth and you can use custom ssl certs: https://nim-lang.org/docs/ssl.html
17:12:33FromGitter<Yardanico> well not with this module
17:12:46FromGitter<Bennyelg> just a connector not much,
17:12:53FromGitter<Yardanico> https://nim-lang.org/docs/net.html
17:12:54FromGitter<Bennyelg> it's python sample
17:13:04dom96httpclient doesn't handle session management for you
17:13:15dom96You'll have to store the handle the HTTP headers yourself
17:13:36FromGitter<Bennyelg> I c
17:14:39FromGitter<Bennyelg> Thanks.
17:15:40Yardanicodoes I c means "I see"?
17:15:50FromGitter<Bennyelg> Yes :D
17:18:34*svennidal quit (Quit: leaving)
17:19:06*yglukhov quit (Remote host closed the connection)
17:19:41*yglukhov joined #nim
17:25:19*yglukhov quit (Ping timeout: 248 seconds)
17:29:35*devdri joined #nim
17:29:41FromGitter<dandevelo> For anyone else experiencing that IO Error issue when using VS Code with the latest Windows 10 update, I wrote here about a workaround I just found out: https://forum.nim-lang.org/t/3367
17:33:19FromGitter<dandevelo> I guess the forum is moderated and the replies will only be visible after being approved
17:33:28*sendell quit (Remote host closed the connection)
17:33:56dom96yeah, approved you
17:34:44dom96This is odd
17:34:46dom96It should be an issue
17:36:04FromGitter<dandevelo> Thanks @dom96
17:36:07dom96https://github.com/nim-lang/Nim/issues/7111
17:36:23dom96I don't have Windows 10 sadly. It would be helpful if someone could investigate this.
17:38:54*zahary_ quit (Quit: Connection closed for inactivity)
17:42:04*Trustable joined #nim
17:43:27FromGitter<dandevelo> It seems like it only happens when using VS Code
17:43:40FromGitter<dandevelo> If I try compiling from the command line then it seems to work
17:43:45*couven92 quit (Quit: Client disconnecting)
17:44:10Yardanicoyes it was happening for me too
17:44:32YardanicoI needed to run the command like 5 times to actually run it fully :)
17:44:55*Arrrr joined #nim
17:44:59*Arrrr quit (Changing host)
17:44:59*Arrrr joined #nim
17:48:25*Jesin quit (Quit: Leaving)
17:51:30FromGitter<dandevelo> Seems like this is the root cause: https://github.com/Microsoft/vscode/issues/36630#issuecomment-350063413
17:56:32GitDisc<treeform> Hey Araq, remember we talked about packed json thing, were instead of using pointers to everything you use relative pointers or sizes in a single buffer? I have implemented some thing like this. I do not need full json, i just need Table[string, string], i benched my packed string table vs regular tables and regular tables win at smaller scales, but when you get 10000+ of these dictionaries my packed buffers start to win out, i suspect due to ma
17:56:32GitDisc<treeform> loc and pointer walking and cache misses becomes to much when you have this many objects.
17:58:08GitDisc<treeform> https://gist.github.com/treeform/4914bbbb7082e0e620085610ce22566a
17:58:28GitDisc<treeform> Its interesting that this idea does not win outright, but does win out at larger volumes.
17:59:11GitDisc<treeform> I called it kiwi, because it kind of sounds like "kv" which short key value store.
18:00:07*yglukhov joined #nim
18:02:18*wishi quit (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.)
18:04:43*yglukhov quit (Ping timeout: 256 seconds)
18:06:13*Vladar quit (Quit: Leaving)
18:36:52*couven92 joined #nim
19:14:23*skelett quit (Ping timeout: 255 seconds)
19:16:19*yglukhov joined #nim
19:20:27*yglukhov quit (Ping timeout: 240 seconds)
19:44:37*xkapastel quit (Quit: Connection closed for inactivity)
19:49:46*miran quit (Remote host closed the connection)
19:54:38*couven92 quit (Ping timeout: 265 seconds)
19:55:52*couven92 joined #nim
19:58:11*devdri quit ()
20:00:39*yglukhov_ joined #nim
20:02:45*wishi joined #nim
20:29:35*couven92 quit (Ping timeout: 240 seconds)
20:32:43*jxv joined #nim
20:35:54*Arrrr quit (Ping timeout: 256 seconds)
20:41:09*couven92 joined #nim
20:42:13*Yardanico quit (Remote host closed the connection)
20:55:10*jxv quit (Quit: zzz…)
20:55:39*Arrrr joined #nim
20:55:52*jxv joined #nim
21:00:25*jxv quit (Client Quit)
21:04:24FromGitter<RSDuck> is this behavior normal, or is it a bug?
21:04:37FromGitter<RSDuck> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5a625d659cdc721e4fa2fa2b]
21:06:15FromGitter<Quelklef> What's the behavior
21:06:28FromGitter<RSDuck> that it throws an error
21:06:37FromGitter<RSDuck> sorry, I should have made that more clear
21:06:47FromGitter<Quelklef> What's the error
21:06:54FromGitter<Quelklef> oh, i see
21:06:59FromGitter<Quelklef> that's your comments. My bad
21:08:19FromGitter<RSDuck> another noteworthy thing here is that after changing the name of close to something else(even to the name of another already defined proc) it works
21:08:32FromGitter<RSDuck> I only tested it on latest devel
21:11:39FromGitter<Quelklef> Dunno, but my guess is that close() is a proc/template/whatever defined in system.nim and that's messing with stuff
21:12:25FromGitter<RSDuck> yeah, that's what I thought too, eventhough close in system.nim looks like nothing special
21:15:23*vivus_ joined #nim
21:16:37FromGitter<Quelklef> See if there's already an issue created for this
21:16:50FromGitter<Quelklef> and if you can't find anything about it, probably create one.
21:17:28*vivus_ quit (Client Quit)
21:17:35FromGitter<RSDuck> I already did, there's currently no issue with this problem reported
22:02:51*xkapastel joined #nim
22:11:57Calinouhow do I use async procs in Nim?
22:12:04Calinouhttps://nim-lang.org/docs/httpclient.html
22:12:13Calinouan {.async.} proc causes an error here
22:13:22dom96import asyncdispatch?
22:18:00CalinouI see, thanks
22:19:02Calinou"You will need to run asynchronous examples in an async proc otherwise you will get an Undeclared identifier: 'await' error." -> does this mean, I need *two* async procs?
22:19:06Calinouone for progress reporting, one for HTTP request?
22:19:10Calinouor can I use one for both?
22:20:25dom96yes, you need two
22:23:10*Amrykid joined #nim
22:32:29Calinouhmm, I'm getting this when running `nimble build` now: https://paste.ee/p/lfMJQ
22:32:39Calinouthe last commit to NimYAML was in November, and it worked fine 3 days ago
22:32:45Calinou(I didn't update Nim in between)
22:34:44dom96that's a weird error
22:35:16Calinoucan I cleanup all dependencies in some way?
22:35:19Calinou(to force a rebuild or something)
22:35:47dom96You can try reinstalling yaml
22:35:49dom96I would open /home/hugo/.nimble/pkgs/yaml-#head/yaml/private/internal.nim
22:35:55dom96and see what's on line 9
22:36:02dom96and try to figure out why this error occurs
22:37:47dom96maybe you were compiling in release mode last time?
22:37:59dom96nimble c -d:release godotup?
22:38:50dom96although actually 'build' should be building in release mode implicitly
22:38:57Calinouthat command works
22:39:07CalinouI'm doing local development right now
22:39:12Calinouso, building in debug mode is advisable
22:39:20dom96so nimble build doesn't work but nimble c does?
22:39:25Calinouyeah
22:39:34dom96check what command each is running
22:39:35Calinou"nimble c godotup"
22:39:40Calinouoh wait
22:39:48Calinouit works *only* in release mode, not in debug mode
22:39:49*MJCaley quit (Quit: MJCaley)
22:40:11dom96That makes sense: https://github.com/flyx/NimYAML/blob/devel/yaml/private/internal.nim#L8
22:40:24*vlad1777d joined #nim
22:40:28Calinourelease: https://paste.ee/p/yLU3Z (disregard the error, it's on my side)
22:40:32Calinoudebug: https://paste.ee/p/N7ymv
22:41:58*Trustable quit (Remote host closed the connection)
22:42:09dom96It's also pointing here: https://github.com/flyx/NimYAML/blob/devel/yaml/serialization.nim#L1060
22:42:55*cspar__ joined #nim
22:43:01dom96seems like a yaml bug to me
22:43:05dom96flyx: ^
22:43:25dom96Maybe putting `untyped` as the return type to this https://github.com/flyx/NimYAML/blob/devel/yaml/private/internal.nim#L7
22:43:29dom96Might help
22:44:30dom96or even s/template/proc/
22:45:10dom96don't see why this is a template
22:45:59*cspar_ quit (Ping timeout: 256 seconds)
22:46:35dom96Calinou: can you try that?
22:49:11Calinoudom96: it works now :P
22:49:21CalinouI still get the "Hint" which doesn't prevent building
22:49:44Calinoucan I send a PR to flyx?
22:50:25dom96sure
22:54:11GitDisc<Lite5h4dow> i dont know what im looking for in the documentation, does nim have an equivalent to try in python?
22:54:20GitDisc<Lite5h4dow> if so what is it called?
22:54:57GitDisc<Lite5h4dow> if not how can i emulate it
22:55:43GitDisc<Lite5h4dow> nevermind, i found it πŸ˜„
22:55:57GitDisc<Lite5h4dow> hiding in the intro instead of the documentation
22:55:59GitDisc<Lite5h4dow> my bad
22:56:39CalinouI love https://hub.github.com/'s pull request feature
22:57:19Calinouhttps://github.com/flyx/NimYAML/pull/52
22:57:56GitDisc<Lite5h4dow> im actually wondering if the documentation is ever going to be made easier to use?
22:58:12GitDisc<Lite5h4dow> or is it pretty much finished as it is
22:58:24GitDisc<Lite5h4dow> since its a little difficult to find things sometimes
23:10:20*yglukhov_ quit (Remote host closed the connection)
23:10:54*yglukhov joined #nim
23:14:57*yglukhov quit (Ping timeout: 240 seconds)
23:15:42FromGitter<dom96> Lite5h4dow: definitely not finished
23:22:11*Arrrr quit (Read error: Connection reset by peer)
23:22:47FromGitter<ZarsBranchkin> Uhh, I'm having a hard time finding documentation on the basic File type. Wanted to see what procedures are there by default
23:27:35FromGitter<dom96> Did you try https://nim-lang.org/docs/theindex.html ?
23:29:27FromGitter<ZarsBranchkin> Yeah, hoped that there will be a place where all the file procs would be collected. Was bit lost on what would be a better way to read a file int by int, or at least word by word where space starts next word
23:30:42FromGitter<ZarsBranchkin> Ah okay, all the basic File procs are in system module, right
23:31:14FromGitter<dom96> Yeah. You can also sort by type in the system module docs.
23:32:13FromGitter<ZarsBranchkin> Oh wow, hadn't noticed that feature
23:34:51*MJCaley joined #nim
23:45:12*MJCaley quit (Quit: MJCaley)