<< 17-05-2017 >>
00:02:06*Matthias247 quit (Read error: Connection reset by peer)
00:19:21*lk000 joined #nim
00:19:31*lk000 left #nim (#nim)
00:40:45couven92Araq, regarding the used pragma. The compiler just complains and does not recognize it. https://github.com/nim-lang/Nim/issues/5810#issuecomment-301055221
01:20:08*chemist69 quit (Ping timeout: 240 seconds)
01:33:08*couven92 quit (Ping timeout: 240 seconds)
01:49:24*chemist69 joined #nim
01:55:11*pilne quit (Quit: Quitting!)
02:18:58*vlad1777d quit (Remote host closed the connection)
02:43:08*libman quit (Quit: Connection closed for inactivity)
02:46:05*def-pri-pub joined #nim
03:30:38*def-pri-pub quit (Quit: leaving)
03:35:37*Snircle quit (Quit: Textual IRC Client: www.textualapp.com)
04:11:43skrylarhmm.. nope. atomic still crashes on startup.
04:21:43*skrylar quit (Ping timeout: 246 seconds)
04:38:23*BennyElg quit (Remote host closed the connection)
05:00:39*smt_ joined #nim
05:01:01*smt_ quit (Remote host closed the connection)
05:03:43*smt quit (Ping timeout: 246 seconds)
05:16:08FromGitter<Varriount> skrylar: What's the exception? Is your compiler targeting the correct architecture?
05:32:29*skrylar joined #nim
05:53:34skrylarnow to just get nanovg drawing stuff
06:20:40*ludocode quit (Ping timeout: 240 seconds)
06:29:31*nsf joined #nim
06:29:53*ludocode joined #nim
06:30:53*BennyElg joined #nim
06:34:13*Ven joined #nim
06:34:36*Ven is now known as Guest75330
07:02:11*gokr joined #nim
07:12:34*Arrrr joined #nim
07:14:57*Guest75330 quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
07:18:31*Sembei joined #nim
07:20:13*yglukhov joined #nim
07:21:10*Pisuke quit (Ping timeout: 240 seconds)
07:23:05*Vladar joined #nim
07:23:55*yglukhov quit (Remote host closed the connection)
07:40:28*aedigix quit (Quit: disconnecting)
07:42:50*yglukhov joined #nim
07:54:23*Gonzih quit (Quit: WeeChat 1.7.1)
07:55:00*yglukhov quit (Remote host closed the connection)
08:38:33*yglukhov joined #nim
08:50:50*bjz joined #nim
08:57:24FromGitter<ephja> bleeding edge metaprogramming is always fun
09:27:18*Andris_zbx joined #nim
09:27:37*couven92 joined #nim
09:35:26skrylardoes that mean metaprogramming with a nubile compiler or the early stages of a macro? :)
09:46:54*aedigix joined #nim
09:52:40yglukhovcan i exclude effect tags from proc def?
09:53:23yglukhovlets say i've got some procs with an effect. and i want to write another proc that states that it doesnt have whis effect
09:53:44Araq.tags: []
09:53:46Araq?
09:54:07yglukhovtags [] tells that it has no effects at all, right?
09:54:27yglukhovbut i just want to make sure that i don't have one specific effect
09:54:42yglukhovand i dont care about the rest
09:56:00Araqyeah that's currently impossible. but usually it has RootEffect anyway
09:56:17skrylarAraq, it seems that nim gets highly confused when both a .nim and .c file exist of the same names
09:56:41skrylarex. if nanovg.nim contains a {.compile: "nanovg/src/nanovg.c".} it will build them both and then experience linker barf
09:58:49dom96that sounds like it should be possible (re: effects)
09:58:58dom96feature request time?
09:59:39Araqit's planned, but I don't see much point. your code is full of indirections you don't care about so RootEffect creeps in and RootEffect can be anything
10:12:13*krux02 joined #nim
10:27:08dom96Just released choosenim 0.2.2 (it fixes the exit code problem)
10:28:16ftsf\o/
10:29:17ftsfthanks dom96
10:34:24Araqdom96: IMO it should jump to 1.0 already. people use it, it works and there was an official announcement about its release
10:35:13dom96You're right, but I want more people to test it first
10:35:27ftsfmore people will test it if its 1.0 ;)
10:35:28dom96The release article for Nim 0.17.0 will include a link to it
10:36:05dom96bbl
10:39:46skrylari suppose nim-jack needs a nimble file of some sort
10:51:07ftsfnimble init
11:02:33*Snircle joined #nim
11:06:19FromGitter<zacharycarter> is there an equivalent to repr for JS?
11:06:30FromGitter<zacharycarter> something that dumps an object to console?
11:16:20ftsfconsole.log(foo) ?
11:16:28ftsfimport jsconsole
11:16:31FromGitter<zacharycarter> oh I forgot about jsconsole
11:16:32FromGitter<zacharycarter> thanks
11:18:22*sz0 quit (Quit: Connection closed for inactivity)
11:27:12ftsfvery handy!
11:49:55ftsfwhat order are command line options, nim.cfg options, pragma options applied? eg. for checks:on/off if nim c --checks:on and a proc has {.checks:off.} which gets applied?
11:50:46skrylarah feces. there's no quickcheck :(
11:50:53ftsfquickcheck?
11:51:25skrylarit's a haskell thing. some other languages have adopted equivalents (go has one). its basically a thing where you tell it what kind of data your interface uses and it makes random crap for unit tests
11:52:59dom96skrylar: create one for nim :)
11:54:52*adeohluwa joined #nim
11:56:51ftsfsounds cool
11:59:06*Tiberium joined #nim
12:00:57*BennyElg quit (Remote host closed the connection)
12:01:03skrylarthey're ok
12:01:30*BennyElg joined #nim
12:05:11skrylarhttp://www.migcalendar.com/miglayout/mavensite/docs/whitepaper.html ergh.
12:05:20skrylari remember implementing a chunk of that in go a year or so ago
12:09:27*Trioxin2 quit (Ping timeout: 272 seconds)
12:11:04*skrylar quit (Ping timeout: 246 seconds)
12:17:08*skrylar joined #nim
12:21:28*krux02 quit (Remote host closed the connection)
12:22:08skrylarhttps://github.com/Skrylar/nim-jack yay cruddy bindings
12:26:29*bjz quit (Ping timeout: 246 seconds)
12:26:56ftsf\o/
12:33:06*bjz joined #nim
12:42:43*yeeve quit (Quit: Leaving)
12:43:42*aedigix quit (Remote host closed the connection)
12:44:12*aedigix joined #nim
12:50:32*Snircle quit (Ping timeout: 260 seconds)
12:50:40ftsfwhat's the correct way to do conditional compilation for different backends?
12:53:33*Snircle joined #nim
12:56:45skrylarconst WearHats {.intdefine.} = 0; when WearHats > 0: ...
12:57:57ftsfyeah I know i need to use when, but when(backend == javascript) ?
12:58:00yglukhovftsf: when defined(js): js; elif defined(php): php; else: C/C++
12:58:11ftsfyglukhov, thanks
12:58:15skrylaroh that kind of backend
12:58:31*skrylar just dealt with nanovg which has backends for GL{2,3}/GLES
12:58:38ftsfahh, sorry
12:59:24skrylarintdefine and strdefine are neat pragmas
13:01:15*bjz_ joined #nim
13:03:15*bjz quit (Ping timeout: 240 seconds)
13:05:37*ftsf didn't realise there was a PHP backend
13:08:06skrylarftsf, there are masochists everywhere
13:09:04skrylari don't get why someone would willingly use php these days. if they inherited it and the company doesn't want to pay to port it to something good, i guess. but having to re-interpret all your framework crap on each page load just sounds horrible
13:09:26skrylarthere's all kinds of caching layers and technical debt paid just to get around things you can avoid by just being an application server instead x.x
13:10:31*yeeve joined #nim
13:15:20*skrylar wonders if its worth implementing an r-tree just for gui stuff
13:33:16Araqwhat's an r-tree?
13:36:45skrylarhttps://en.wikipedia.org/wiki/R-tree supposed to be an efficient way to find things spatially
13:38:20Araqomg, do it, I want it :-)
13:38:44skrylartrying to mentate how it works from their description
13:39:05skrylari was sorta sad that there was an IBM article a few years ago about how most professional programmers couldn't implement binary search from its description
13:43:49yglukhovskrylar: why would gui need r-tree? there's always a limited set of widgets to go through.
13:44:46skrylaryglukhov, depends how big the gui is ._.
13:45:03skrylarmusic software can get a few hundred on a single pane
13:45:48ftsfskrylar, what about a quadtree?
13:45:50yglukhovyeah, but then again the events are not propagated to all of the widgets. they are propagated if passing hittest of the superview.
13:46:10skrylarftsf, i dunno. it was mostly just a question
13:47:32FromGitter<ephja> FormatTypes = IntFormatTypes|MapFormatTypes|StringFormatTypes|openarray[tuple[k: FormatTypes, v: FormatTypes] *crosses fingers*
13:47:37skrylarset24 does this heinous thing where it O(n)'s the entire set of notes every time it seeks (which happens ~46 times per second or so)
13:48:08FromGitter<ephja> gotta use the CPU for something
13:54:28skrylari also thought about just keeping two lists for X and Ys
13:54:56*Matthias247 joined #nim
13:55:04skrylarsort the widgets by Y, then X, so it doesn't have to test a heap of rects
13:55:51*aedigix quit (Quit: disconnecting)
13:58:03*Matthias247 quit (Read error: Connection reset by peer)
13:58:41ftsfskrylar, is it an issue yet? ;)
14:01:06*Matthias247 joined #nim
14:02:57*adeohluwa quit (Quit: Connection closed for inactivity)
14:03:29FromGitter<ephja> no I have to use 'array' rather than 'openarray' for this recursive solution, and do FormatTypes[T=void] ...
14:04:10*libman joined #nim
14:11:04*rauss joined #nim
14:11:52*aedigix joined #nim
14:13:38skrylarftsf, but my jimmies. they rustle
14:13:45skrylarperceptive inefficiency!
14:13:55skrylarwell actually in seq24s case yes it performs poorly.
14:14:11skrylarbut that's because they're doing the silly inefficient stuff in a part of the code that actually does need to run in real time
14:22:52*PMunch joined #nim
14:23:10*aedigix quit (Quit: disconnecting)
14:24:40FromGitter<ephja> "Foo = int|array[???, tuple[k: Foo, v: Foo]]" I can't think of a way to do this without parameterizing Foo
14:29:43dom96I think you should use an object variant
14:35:53FromGitter<ephja> dom96: I'm simply trying to do it this way since all the necessary information is available at compile-time, but of course the object variant approach won't involve constructs that have dozens of bugs associated with them :-P
14:37:05dom96How can this work though? You've got an infinite recursion in there as far as I can see.
14:37:35*BennyElg quit (Remote host closed the connection)
14:38:59FromGitter<ephja> how come?
14:39:09FromGitter<ephja> the nesting can be terminated at any moment
14:40:42FromGitter<ephja> but I dunno if maybe it will result in an infinite recursion in the compiler because of some bug
14:41:04FromGitter<ephja> unless I'm missing something really obvious again
14:43:07Tiberiumdom96, new github avatar :)
14:43:30dom96:)
14:46:03PMunchHmm, when using "nim doc" is there a way to add code snippets?
14:46:15PMunchI wanted to add the input/output for codeRepr
14:47:44*planhths joined #nim
14:48:24dom96.. code-block::nim
14:48:43*nsf quit (Quit: WeeChat 1.7.1)
14:48:45PMunchOh, double :: -_-
14:49:23PMunchHmm
14:49:29PMunchThat didn't quite seem to do it
14:50:26PMunchIt made the main body line and the first line of what was supposed to be code bold..
14:51:38PMunchhttp://ix.io/u0R
14:53:47Tiberiumyou can check other stdlib procedures for examples
14:57:02PMunchAh, got it
15:01:33FromGitter<ephja> thou shalt not do at runtime what can be done at compile-time
15:02:08*Sentreen quit (Ping timeout: 260 seconds)
15:02:53PMunchephja, halleluja
15:06:55*GustavoLapasta joined #nim
15:16:08*Sentreen joined #nim
15:23:46ArrrrAmen Father Ephja
15:23:52Tiberiumlet's make some viruses in ( ͡° ͜ʖ ͡°)
15:23:58Tiberiumin Nim
15:24:48Tiberiumnevermind - https://github.com/Vikaton/nim-keylogger
15:25:14Tiberiumwait, using win api is SO easy?
15:28:14*vlad1777d joined #nim
15:29:24*GustavoLapasta quit (Quit: Leaving)
15:31:04skrylarTiberium, as easy as bashing one's face against the wall :>
15:32:38*PMunch quit (Quit: leaving)
15:40:05*sz0 joined #nim
15:49:59*bjz_ quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
15:52:43*bjz joined #nim
16:04:11*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
16:07:19*nsf joined #nim
16:07:41*Andris_zbx quit (Remote host closed the connection)
16:15:02*BennyElg joined #nim
16:24:01*Matthias247 quit (Read error: Connection reset by peer)
16:27:01Tiberiumwhy only 10 fires were changed? https://github.com/nim-lang/nim/commit/9c47bb9cc0fda686ca403f5abd0ae2f06d364c74
16:27:03Tiberium*files
16:30:10ArrrrHe got bored halfway
16:30:24TiberiumArrrr, auto-replace?
16:31:10ArrrrWhy not
16:35:13*Trioxin2 joined #nim
16:39:21FromGitter<mratsim> @skrylar What’s the difference in usage between R-Tree and Kd-Tree ? Most clustering/geoloc library in Python use Kd-Trees to store locations
16:41:33skrylar@mratsim idunno. i've not done geo stuff.
16:41:44skrylarkdtrees are probably more optimized
16:54:35FromGitter<mratsim> ok
16:57:31FromGitter<Varriount> Tiberium: You mean, as opposed to Linux, where you have to pick one of a myriad of interface methods (gui library, x11, etc) to get keyboard input?
16:58:29*yglukhov quit (Remote host closed the connection)
16:59:49TiberiumVarriount: I mean that interfacing with winapi is easy :)
17:00:05FromGitter<ephja> I suppose this is how you associate a typedesc with some type: "proc underlyingType(f: Foo): typedesc = int" and then a concept that includes such a signature
17:01:13FromGitter<ephja> it's slightly more verbose than defining a type or a const but it should get the job done
17:03:02*yglukhov_ joined #nim
17:03:56*libman quit (Quit: Connection closed for inactivity)
17:07:10*yglukhov_ quit (Ping timeout: 246 seconds)
17:31:17*libman joined #nim
17:40:24*pilne joined #nim
17:41:38*yglukhov joined #nim
17:46:18*yglukhov quit (Ping timeout: 260 seconds)
18:02:36*yglukhov joined #nim
18:04:47ftsfany recommended internationalisation lib for nim?
18:16:09Tiberiumftsf, i don't know, but 1 sec googling gave me this - https://github.com/Parashurama/nim-i18n
18:16:16Tiberiumalso https://github.com/Amrykid/nim-locale from 2014
18:17:38ftsfyeah both look kinda unused/unmaintained, but the first one might be ok
18:35:35*yglukhov quit (Remote host closed the connection)
18:40:12Araqftsf: I can tell you how to do it but hardly have the time to write it
18:41:16ftsfAhh well, not urgent =) Just asking in case someone is using something nice for it.
18:42:47*Arrrr quit (Read error: Connection reset by peer)
18:44:31Araqwhy not do my work?
18:45:37*yglukhov joined #nim
18:51:44*rauss quit (Ping timeout: 260 seconds)
18:53:00couven92Araq, how would we want it? Just call libgettext? work with .po files?
18:53:17FromGitter<Varriount> ftsf: tables.nim?
18:53:23*rauss joined #nim
19:03:44Araqcouven92: i18n("$1 takes $2", param1, param2)
19:04:38Araqi18n("$1 [items|item|item]$1 found", 13)
19:05:13Araqthe [] is a selector, 0, 1, many is mapped to this construct
19:06:01couven92Hmm...
19:06:06Araq [item|items] is also possible but some languages have a special plural form for zero
19:06:36Araqand i18n is of course a macro.
19:06:54couven92and it reads from a file?
19:07:17couven92what file? what should that look like? csv, json, yaml?
19:08:57Araqwell we also need an enum Language and setLanguage()
19:08:58Araqand an setI18nMapping() proc
19:09:53couven92maybe getting languages from locale?
19:10:12couven92I'd say all OS'es have a concept of language with API'
19:10:26Araqyeah, and we have a better concept, it's called an "enum"
19:11:18couven92Araq, maybe a macro that creates the macro by asking the OS for all languages it knows? :P
19:11:29skrylarmeh brain must be tired. took several hours for a basic merge sort o.O
19:11:35couven92creates the language enum, I mean
19:11:59skrylarftsf, well translation is inherently just a hash table really
19:12:11skrylarok so there's a minor exception to that; dates get weird.
19:12:26skrylaryou just need a special printf and a hash table
19:12:43couven92skrylar, NO!!! Look here: https://www.youtube.com/watch?v=0j74jcxSunY
19:13:24dom96Please do some research before creating a library. There are bound to be many well regarded i18n libraries out there.
19:13:36dom96(I'm sure there are many edge cases for these so it's worth researching)
19:14:33couven92dom96, like gettext in GNU for example?
19:14:57dom96Not sure if that's considered a "good" library but sure
19:15:01demi-lol
19:15:23yglukhovcouven92: please don't forget about the js target as well ;)
19:15:40demi-providing localizable strings is an entirely separate problem from i18n
19:15:44couven92problem is: gettext is GNU and mostly just Linux...
19:16:20couven92yglukhov, yeah... so ultimate goal: write it as pure Nim, right?
19:16:22demi-implementation of how to fetch and get localized content is very different from the entire subject matter of being able to support i18n
19:16:43yglukhovcouven92: sure =)
19:16:50demi-like, appending to a string based on read direction of the language
19:17:09couven92demi-, true... and I must say, Araq's suggestion actually look quite nice! :)
19:17:17FromGitter<ephja> I assume that the inability to call p(x: typedesc[Foo]) where Foo is a type class defined in a type section has been reported
19:17:47FromGitter<ephja> keeping track of too many issues
19:18:14Araqdom96: I did this research and the gist of it is the proper 'choice' for pluralization
19:18:30Araqof course, there are other things like units of measure, currencies etc etc
19:18:31skrylarok. i watched the video.
19:18:36skrylari was still mostly correct tho.
19:18:43skrylarmaybe overly brief with "special printf"
19:18:47Araqbut these are harder to solve
19:19:29FromGitter<ephja> wait. it fails when expanding the type class too. yeah, I located the bug report yesterday I think
19:19:39couven92skrylar, printf is always locale specific!
19:20:04Araqprintf cannot even do argument reorderings with a nice syntax
19:20:22Araqnim's $1, $2, $# is superior
19:20:34demi-skrylar: i think you are vastly underestimating how this works
19:20:38libmanRogue syntax highlighter doesn't handle type suffix apostrophes - https://github.com/Parashurama/nim-i18n/blob/master/i18n.nim#L263
19:20:52skrylardemi-, no, i'm just talking about string formatters like Inform has
19:21:18couven92but i18n is SOOO much more! :P
19:21:35skrylarinform's string formatter actually does deal with pronouns on objects and stuff.
19:21:43Araqcouven92: ok, so let's not provide the barebones that cover 90% of the cases
19:21:53couven92luckily, we at least have a capable string type that does UTF-8
19:22:03demi-so how do you handle cases were some languages have different pluralisation rules based on what you are talking about
19:22:04*Matthias247 joined #nim
19:22:09libmanOr is it https://github.com/github/linguist ...
19:22:23FromGitter<ephja> https://bitbucket.org/lyro/strfmt
19:22:31couven92solves quite a lot of the boring "I can't write æøå" problems
19:23:31skrylarI have $ducks$p:duck(20). ;p
19:23:35skrylarthat's not inform tho
19:24:10skrylargettext does have something like a getplural function although it's not very good
19:25:09Araqdemi-: I don't understand your problem. in the lookup table there is a mapping from "$1 takes [item|items]$2" to "$1 kriegt [Parameter]$2"
19:25:53skrylari could probably design one. although those things tend to be kinda.. eeh
19:25:53*Matthias247 quit (Read error: Connection reset by peer)
19:25:59*nsf quit (Quit: WeeChat 1.7.1)
19:26:07skrylari mean i did a whole unicode thing once. that's not worse than this
19:26:12demi-Araq: based on my naive understanding, something like that would work where you are applying it to segments of a whole string; but my question was aimed at what skrylar was saying
19:26:36Araqdifferent rules don't matter because the possible declinations are part of the language specific format string
19:27:00Araqas I said, it's a decent design.
19:27:17*Matthias247 joined #nim
19:27:31demi-yeah, i don't disagree
19:27:53skrylari would prolly just do it where you have like, translate(FOO, 20) where foo is an object instantiated from definitely not .po files
19:28:08skrylar.po files are standard and kind of ok but they don't deal with a lot of the plurality rules
19:28:43skrylarmaybe compile to bytecodes or something
19:29:32demi-i think most major platforms have an implemenation for doing this, i think it all trails back to xliffs
19:29:45skrylaryou would be very surprised
19:29:57couven92cazy idea: just create that as a block under a macro... Nim code with just const strings is not that hard to read or write
19:30:06yglukhovdom96: your regression fix looks wrong...
19:30:14skrylara lot of them have one but it sucks or is wrong
19:30:40dom96yglukhov: oh?
19:31:01yglukhovan i think the tests should fail...
19:31:39yglukhovcopyNimTree is intended to get the pragmas from the original proc and apply them to the closure iter
19:31:53yglukhovwhats the regression? do you have a sample?
19:32:05demi-skrylar: considering this is pretty industry standard: https://en.wikipedia.org/wiki/XLIFF
19:32:36dom96yglukhov: the test compiles for me.
19:33:06dom96yglukhov: https://gist.github.com/dom96/83b8ed5e5a1db2a26d949faee78c7962
19:35:02*Tiberium quit (Remote host closed the connection)
19:35:26couven92ah, my crzay idea doesn't work... because you have to be able to edit/add new translations at runtime...
19:36:45skrylaryes
19:37:59skrylarwell i mean you could just statically compile but the translators will be unhappy
19:38:05*aedigix joined #nim
19:38:49yglukhovdom96: well, your sample works in both cases, but differently. my way applies the pragma to not just the function but to the closure iter as well.
19:39:09yglukhovdom96: do you think my behavior is incorrect?
19:39:10*aedigix quit (Remote host closed the connection)
19:39:40Araqyglukhov: your behaviour breaks production code :P
19:39:43dom96I'm afraid so. In the sample I linked above the `print` macro is executed twice.
19:40:03*aedigix joined #nim
19:40:44libmanHope I reported this bug correctly - https://github.com/Varriount/NimLime/issues/82
19:41:22yglukhovAraq, dom96: but that does mean its incorrect?
19:43:20dom96what is?
19:43:45yglukhovmy behavior that I specifically introduced with my asunc macro fix
19:43:47*aedigix quit (Client Quit)
19:47:42dom96Looks like it. Do you need the pragmas to be applied to the closure iterator too?
19:49:51yglukhovdom96: well that was my intention, but now that i think of it, maybe its not that needed after all. gcsafe on closureIter (which there is) would be enough for me.
19:52:02AraqI think there is a problem with calling conventions
19:52:19Araq.cdecl for the wrapper proc might make sense, but the .closure iterator cannot be cdecl
19:52:48yglukhovok, makes sense
19:55:57Araqbut that is a fix for another time :-)
19:56:37Araqcan anybody confirm the android target is ok? seems to risky for 0.17
19:56:39FromGitter<Varriount> @lbmn >:D
19:57:14yglukhovAraq: android is fine with zah merged in =)
19:57:30Araqhow come android doesn't define 'linux'?
19:58:16couven92Araq, i'd say it still needs some polishing... and I need some input on stdlib redesign... I don't rellay know how to fiddle android in there in a good way...
19:58:42yglukhoverr... ok, i might have misunderstood the question. i compile for android with --os:linux -d:android
19:58:49AraqI can polish it later, but it won't make it to 0.17 then
19:58:51couven92my attempt to mess with lib/platforms.nim didn't go that well :P
19:58:58yglukhovdidnt try with --os:android
19:59:52couven92hmm... but yeah, --os:android should define both android and linux
20:00:52couven92Luckily discovering Android in build.sh is working on Termux, but won't work otherwise on Android...
20:01:14*yglukhov quit (Remote host closed the connection)
20:01:35couven92because uname on vanilla android does not tell anything about android
20:09:36FromGitter<Varriount> @lbmn @dom96 https://github.com/github/linguist/issues/3631
20:10:26FromGitter<nordborn> Looks like it's impossible to link mysql libs statically, especially with musl. Sadly:)
20:12:12*PMunch joined #nim
20:15:35*rupil joined #nim
20:17:06*Vladar quit (Quit: Leaving)
20:23:24*yglukhov joined #nim
20:53:42*rauss quit (Quit: WeeChat 1.8)
20:54:17*planhths quit (Quit: Konversation terminated!)
21:05:12*rupil quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
21:08:58*bjz joined #nim
21:14:48*nsf joined #nim
21:23:24skrylarhttps://images.discordapp.net/.eJwVyEsOhCAMANC7cABKwfq7DUGCRqWE1riYzN0n85bvY55-mdXsqk1WgO2QxH2zotxjybYwlyvHdohNfENUjWm_c1UBH5Z5nB15IjdhCANBwIEQ_-E8kl9GB089K7_VtlrM9wcFIiLE.Y4EoVJiYzTUmbOCQFhGSt9o2iak?width=736&height=587 neat.
21:28:31*cyraxjoe quit (Ping timeout: 246 seconds)
22:05:47*niv joined #nim
22:22:17FromGitter<zacharycarter> I've been going down this path of adding a html5 backend to frag with PIXI js
22:22:30FromGitter<zacharycarter> but I'm not going to be able to make the API consistent with the SDL backend
22:23:10FromGitter<zacharycarter> anyone have any thoughts on the web backend having a different API? And code reuse just not being a thing for thick / thin client?
22:23:34FromGitter<zacharycarter> I'm thinking I just ditch the html5 backend
22:27:57skrylari donno what you mean. but then i haven't done any html5
22:28:32FromGitter<zacharycarter> I have this game framework I've built for Nim
22:28:37skrylari mean the sdl renderer just either gives you a Gl context or it does some really basic shapes for you. there's the canvas apis that do similar things to that
22:28:42FromGitter<zacharycarter> http://fragworks.io/
22:29:31FromGitter<zacharycarter> it doesn't use SDL for rendering it uses a library called bgfx which abstracts away the differences between opengl, directx, gles, etc...
22:29:56*nsf quit (Quit: WeeChat 1.7.1)
22:30:52FromGitter<zacharycarter> I've been playing with / pondering adding a html5 backend to the project, but I don't want to write all the webgl code so I was going to use this library called pixijs, but unfortunately it has a radically different way of handling drawing submissions vs my library
22:31:26skrylarah. haven't done much raw gl stuff.
22:31:46skrylarsaw bgfx and a related thing where they talked about some clever draw call sorting tricks
22:31:53FromGitter<zacharycarter> aye
22:36:06*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
22:38:48dom96https://nim-lang.org/blog/2017/05/17/version-0170-released.html
22:39:14FromGitter<zacharycarter> :D
22:39:51*dom96 quit (Changing host)
22:39:51*dom96 joined #nim
22:40:29FromGitter<zacharycarter> can't wait to see how many things of mine break
22:43:22*yglukhov quit (Remote host closed the connection)
22:43:25dom96Awww, don't be so negative :)
22:43:37FromGitter<zacharycarter> hahaha
22:44:05couven92dom96, Nice! :)
22:44:07FromGitter<zacharycarter> actually things are looking very good, my examples are running :D
22:44:20FromGitter<zacharycarter> and thank you @dom96 and everyone else who contributed to the release
22:48:04PMunchOooh looking forward to this
22:49:04FromGitter<zacharycarter> I may have to write bindings to : https://github.com/libuv/libuv
22:49:31FromGitter<zacharycarter> even though nim already has really nice async
22:50:01dom96Posted to Reddit and HN too. Let's see how that goes
22:51:54FromGitter<zacharycarter> hopefully well
22:52:30FromGitter<zacharycarter> I guess I'm scratching this html5 backend for now
22:55:14nivcongratulations on .17 :)
22:56:05dom96Thanks :D
22:56:16niv"initialiazed"
22:56:26nivinit-aliased? :D
22:56:31nivinit-lazored?
22:57:15couven92dom96, did you actually make it on the 17th May? I see you updated the channel at 00:38
22:57:36couven92ah, wait, it's still the 17th in UTC, right?
22:57:40dom96indeed
22:57:43dom96BST here
22:57:54Araqalso
22:57:55couven92bst?
22:58:00dom96British Summer Time
22:58:03couven92ah
22:58:03AraqI like 17th and 21th
22:58:22couven92Araq, you mean 21st?
22:58:27Araqlol yes
22:58:30couven92:D
22:58:35nivreleasing nim 0.32 might get tricky
22:58:59couven92schedule 1.0 for 2018-01-01?
22:59:01couven92:P
23:00:59nivthat changelog is awesome. great stuff.
23:04:35Araqyeah, it's a decent release
23:04:55Araqbut as usual
23:05:11Araqif too many things come up, we'll release 0.17.2 within the next month
23:05:27nivwhat tickles me is EOFError for some reason. always hacked that in myself
23:05:39Araqdidn't work out for 0.16 unfortunately, but it worked before, it will work again
23:06:29PMunchOooh, the new try-except stuff is nice
23:06:34PMunchWas missing that
23:06:45nivyeah
23:06:58nivthe foo bar do: syntax is messing with me though
23:07:58PMunchOooh, and the JSON unmarshalling is cool
23:08:31nivi have a nimble package up to do that as a template, extensible to arbitary types
23:08:59dom96I implemented both :D
23:09:10nivhttps://github.com/niv/jser.nim
23:10:03PMunchThe colon block stuff has a type by the way
23:10:16*cyraxjoe joined #nim
23:10:22PMunchIt uses value as the paramater (and no type?) but uses val in the body
23:10:28dom96ahh, when Nim hits the HN front page. +20 visitors immediately on analytics.
23:10:56Xedom96: new website is great looking
23:11:33dom96Xe: thanks :)
23:11:40PMunchYeah the page is awesome
23:12:00Xedom96: if you are not the one responsible for it, please relay my gratitude to the people who are
23:12:00nivright, well, heading off. congrats again! will have to have a play soon and migrate some stuff.
23:12:00PMunchThe old one looked a bit 90s :P
23:12:40dom96Calinou deserves thanks too. And the many people that made various contributions to it as well.
23:13:02dom96Yeah, the new website feels modern.
23:15:57PMunchBy the way dom96, what do you think of the new doc entry for https://github.com/nim-lang/Nim/pull/5778#issuecomment-301775072
23:16:29dom96seems fine, the name of these procs still needs to be changed
23:17:18PMunchHmm, I was just trying to match the style og dumpList and dumpTree
23:17:24PMunchs/List/Lisp
23:24:03Xedom96: ooh i like the `to` operator
23:24:09Xeer macro
23:24:12Xesorry, long day
23:26:14*gokr quit (Quit: Leaving.)
23:32:37dom96yeah, it's a cool macro and many people have wanted something like it :)
23:37:52dom96Interesting, Nim's position on HN's front page dropped from about #9 to #23.
23:38:07dom96I wonder if HN's mods demoted it
23:38:22*sz0 quit (Quit: Connection closed for inactivity)
23:39:16*libman rips off his beard in anger at HackerNews.
23:39:32PMunchWhy would they do that?
23:39:54libmanIt's a conspiwwwacy.
23:43:39couven92Love, how the 0.17 nimsuggest automagically does so much better in VS Code immediately, without any additional steps!!!! :D
23:45:43couven92how do i actually refer to another proc in the doc comments? ``otherProc``?
23:45:53dom96yep
23:46:04couven92does that generate a link?
23:46:41couven92no it does not? Ho to create a link as well?
23:49:31enthus1astcool changelog!
23:50:00*Matthias247 quit (Read error: Connection reset by peer)
23:53:31dom96`blah <url>`_
23:53:38dom96You will need to find the anchor id though
23:54:11libman$searchEngine "hacker news censorship". They've been deceptively censoring me in the worst way possible for a very long time now.
23:54:43dom96libman: You've got negative karma, they probably auto kill all posts from accounts with negative karma
23:56:36libmanExplaining the specifics of the bureaucracy doesn't make it any better.
23:56:56*couven92 quit (Quit: Client Disconnecting)