<< 04-11-2013 >>

00:02:12fredmorcosemacs' c-mode supports #if 0
00:02:28fredmorcosbut in LaTeX's case, it was a \newcommand definition
00:02:59*freezerburnv quit (Quit: freezerburnv)
00:07:24fredmorcosgood night
00:07:25*fredmorcos quit (Quit: Leaving)
00:08:33*d34th quit (Quit: Deleted System32)
00:11:05*d34th joined #nimrod
00:21:25DemosI would enjoy a command that "commented" stuff out but still made the compiler try its best to typecheck it
00:44:46*d34th quit (Ping timeout: 245 seconds)
00:55:09*d34th joined #nimrod
00:56:18*filwit quit (Quit: Leaving)
01:01:30*Ricky_Ricardo joined #nimrod
01:03:00*freezerburnv joined #nimrod
01:21:16*hoverbear joined #nimrod
01:23:58*hoverbear quit (Client Quit)
01:31:16*hoverbear joined #nimrod
01:35:03*Demos quit (Read error: Connection reset by peer)
01:44:58*zezba9000 left #nimrod (#nimrod)
01:49:41*freezerburnv quit (Quit: freezerburnv)
01:59:41*eigenlicht quit (Ping timeout: 272 seconds)
02:00:57*eigenlicht joined #nimrod
02:20:27EXetoCNimrod is better than ponies
02:38:39*xenagi joined #nimrod
02:53:27*brson quit (Quit: leaving)
03:10:32*capisce quit (Ping timeout: 265 seconds)
03:13:53*capisce joined #nimrod
03:31:13*hoverbear quit (Quit: Textual IRC Client: www.textualapp.com)
04:34:57*hoverbear joined #nimrod
04:50:37*dyu_ joined #nimrod
04:53:14*Ricky_Ricardo left #nimrod (#nimrod)
05:10:33*hoverbear quit (Ping timeout: 245 seconds)
05:14:27*xenagi quit (Quit: Leaving)
05:23:24*VarriountPhone quit (Remote host closed the connection)
05:23:38*VarriountPhone joined #nimrod
05:28:47*gour joined #nimrod
05:56:23*OrionPK quit (Quit: Leaving)
06:08:19*mflamer quit (Ping timeout: 272 seconds)
06:50:07*Associat0r joined #nimrod
07:12:49*Jackneill joined #nimrod
07:20:40*tumak quit (Ping timeout: 260 seconds)
07:20:46*tumak joined #nimrod
07:34:39*Ywo joined #nimrod
07:39:00*Ywo quit (Client Quit)
07:39:04*vidot_j joined #nimrod
09:31:11*CarpNet joined #nimrod
11:08:53*vidot_j quit (Ping timeout: 245 seconds)
11:13:17*freezerburnv joined #nimrod
11:27:56*freezerburnv quit (Quit: freezerburnv)
11:49:46*Ywo joined #nimrod
12:28:52*Ywo quit (Remote host closed the connection)
12:30:00*vidot_j joined #nimrod
12:34:42*freezerburnv joined #nimrod
12:54:27*Ricky_Ricardo joined #nimrod
13:45:50*dymk quit (Ping timeout: 245 seconds)
13:55:32*dymk joined #nimrod
14:10:15*Ricky_Ricardo quit (Quit: Ricky_Ricardo)
14:26:39*markhend joined #nimrod
14:34:55*markhend left #nimrod (#nimrod)
15:03:15*mflamer joined #nimrod
15:04:37*faassen joined #nimrod
15:19:01*fowl joined #nimrod
15:22:13*Endy joined #nimrod
15:25:48*mflamer quit (Ping timeout: 240 seconds)
15:34:25*gour quit (Disconnected by services)
15:34:28*gour_ joined #nimrod
15:42:33*OrionPK joined #nimrod
16:33:50fowlhola
16:34:58*NimBot joined #nimrod
16:40:49EXetoCdðas?
16:40:59EXetoClooks like that on my UTF-8 setup
16:41:41EXetoCalso, hi :p
16:52:08*BitPuffin[Mobile joined #nimrod
16:54:38*BitPuffin[Mobile quit (Client Quit)
16:54:40*VarriountPhone quit (Remote host closed the connection)
16:54:53*VarriountPhone joined #nimrod
16:55:07*BitPuffin joined #nimrod
16:56:27BitPuffindom96!!
16:56:33BitPuffincoder radio
16:56:36BitPuffinjböuve
16:56:46BitPuffinjblive.tv
16:57:00BitPuffinI won't make it home on time
16:57:09BitPuffinI passed my driving test
16:57:20BitPuffinI am a qualified driver
16:57:36BitPuffinAraq maybr you should be in the live chatroom too
16:57:47BitPuffinbrb
16:57:52*BitPuffin quit (Client Quit)
17:01:43dyu_Araq: odersky's talk on strangeloop already on infoq. Your talk could already be available too
17:01:57*dyu_ quit (Quit: night!)
17:04:31*filwit joined #nimrod
17:10:15*dom96_and joined #nimrod
17:10:19dom96_andNoooo.
17:10:34filwit??
17:10:35dom96_andThe one time I'm not home...
17:11:04dom96_andBitPuffin: looks like I'll miss it sorry.
17:11:22dom96_andfilwit: jblive, read logs.
17:11:40filwitidk how to do that..
17:11:46dom96_andBitPuffin wrote an email to them about nimrod.
17:12:13dom96_andFilwit, join their chat if you can.
17:12:13filwitidk what you're talking about..
17:12:30filwitwho's chat?
17:12:53dom96_andCoder radio at jblive
17:13:05filwitk
17:14:21dom96_andAnyway. Bye.
17:14:27*dom96_and quit (Client Quit)
17:15:04*fowl quit (Quit: Leaving)
17:18:26OrionPKdom96 here's a fairly concise demonstration of the memory leak https://gist.github.com/onionhammer/7306011
17:21:12EXetoC"The public release of this presentation will be sometime in the next 6 months - date to be determined." :p
17:21:28AraqOrionPK: that's a very useful program. add it to your bug report please
17:21:52OrionPKAraq wasnt sure whether its an issue with asyncio or the GC
17:21:55EXetoCbut yeah some talks are already up
17:21:56OrionPKAraq but will do
17:23:23Araqalso try the different GCs please
17:24:42OrionPKhmm, could have sworn i submitted a bug
17:24:56OrionPKdont see it in there
17:25:05Araqyou did
17:25:13AraqI got an email
17:25:39OrionPKyou still see it?
17:26:11OrionPKoh there it is
17:26:11OrionPKjeez
17:29:35OrionPKAraq v2 isn't building
17:30:19EXetoChelper clone?
17:31:17OrionPKaraq where can I get boehmgc.dll?
17:31:27Araqmeh v2 is not important
17:31:39Araqoh you're on windows
17:31:46OrionPKmarkandsweep is worse, shoots up to 5mb right away
17:32:10Araqwell it's only a leak if it keeps growing
17:32:14filwitdisable GC and buy more ram - only solution
17:32:18OrionPKit does keep growing
17:32:24OrionPKgrows more slowly with refc
17:33:10*Demos joined #nimrod
17:34:39EXetoCfilwit: I bet rasmus lerdorf would be well happy with 128gb ram or so
17:35:15OrionPKjust restart the program every year or so
17:35:38Demoshey not having a GC is a form of GC as long as you keep adding memory!
17:35:52filwitEXetoC: i'm unfamiliar with his work
17:35:55OrionPKyeah you just take out the old ram and throw it in the trash when its full
17:36:37Demosfilwit: PHP
17:36:58filwitDemos: ahh.. EXetoC's joke makes sense now..
17:38:06Demosugh I need to write a sequencial solution for a super parellel problem on this assignment.....
17:39:48EXetoC:d
17:41:45filwitDemos: good schooling. that will be a very common software practice in the near future when we all have 32-64 core machines
17:42:08filwitDemos: or is right now if you're working on big server systems i'd imagine
17:42:09Demoswell they measure our timing so allowing the use of OpenGL would kinda ruin that
17:42:22Demos*OpenCL
17:42:37filwityeah, was confused as to how OpenGL was going to help you, lol
17:42:54Demoswell compute shaders I guess
17:43:02filwitsame thing
17:43:07AraqOrionPK: it's something to look into but it goes from : 135168 back to 94208
17:43:07Demoshell with nvidia thrust this would be like a three line function
17:43:27Araqthere is no leak
17:43:29OrionPKafter how long?
17:43:43Araqrunning it for a minute now
17:43:49Demoshey Araq did you see the disscussion of modules that went on a few nights ago, kinda a big feature but I am interested to hear your throughts?
17:44:02Araqsorry I missed that one
17:44:34OrionPKwhat OS?
17:44:41Araqlinux 64bit
17:44:48OrionPKchrome?
17:44:53Araqfirefox
17:44:57Demosessentially we found that it was annoying to have to say stuff like from table import TTable then do the same for ...every... function related to TTable
17:44:59OrionPKmmk, did you have the tab focused?
17:45:10*MFlamer joined #nimrod
17:45:13OrionPKI noticed if the tab goes in the background the timer slows down
17:45:41OrionPKI'll run it for 20 or 30 mins and see what happens
17:45:47AraqOrionPK: it doesn't matter if focussed or not
17:46:05Araqand it never reaches more than 135168
17:46:16Araqso ... why should I run it for 30 minutes?
17:46:25OrionPKI said I will
17:46:30dom96filwit: Have you been watching Coder Radio?
17:46:38filwitdom96: yeah
17:46:51AraqDemos: oh not that old import dicussion again
17:46:52filwitdom96: kinda confused as to why you wanted me to, though
17:47:06filwitdom96: thought they might have been talking about Nimrod or something
17:47:08dom96filwit: Did they mention Nimrod?
17:47:12Demosmy thought was to allow something like from table import TTable and fns and then import the functions in the module qualified (you need to access them by referring to the module). Essentially it came down to "we would need ADL to make this work"
17:47:19filwitdom96: not that i've heard so far
17:47:34filwitdom96: was going to suggest it, but didn't know if they had just got done talking about it or something
17:47:43Araqdelphi imports everything and doesn't even have "from import" as a feature and yet millions of lines of code have been written in it
17:47:54OrionPKAraq in the program it's showing 286720 for quite a while, but in task manager it keeps crawling up.
17:48:08AraqC doesn't even have a module system and yet billions of lines of code have been written in it
17:48:12dom96filwit: Well, have they looked at viewer emails yet?
17:48:27AraqNimrod supports a quite elaborate module system and yet everybody complains
17:48:31dom96It's coder radio, why the hell are they talking about MS Surface?
17:48:36DemosI was going to say that c++ essentially does "modules" in the maximally painful way and people use it
17:48:36filwitdom96: idk, i've never listened to this show before
17:49:19filwitDemos: honestly, i'm not a fan of adding a special select import feature.. sounds like you should just divide up your module
17:49:22Araqif your "from x import foo" list becomes too tedious maybe you should use "import" instead?
17:49:30OrionPKwhat does linux say it's using, memorywise, araq?
17:49:49filwitDemos: the "from x import y" feature is designed for when you specifically ONLY want certain things
17:50:06filwitDemos: otherwise just import and prefix everywhere with the module name
17:50:12Demosyeah, the "solution" is to add c++ style Koenig lookup, but I would be money that that causes problems elsewhere
17:50:19AraqOrionPK: 860KB and not growing
17:50:47Demos*bet
17:51:08filwitDemos: here's a quick solution to avoid conflicts: "from M import nil"
17:51:26filwitDemos: then you're forced to prefix 'M' everywhere in your code
17:51:49DemosOo OK, that works OK
17:52:10Demoscan I still do like M.TFoo.blarg?
17:52:20filwityes
17:52:38Araqfilwit: yeah but people want to write M.foo everywhere except when then they don't M.`[]=`(tab, key, val)
17:52:42Demosbut not blarg(M.TFoo)?
17:53:10filwitDemos: it would be M.blarg(M.TFoo)
17:53:28Demosnot bad
17:53:37OrionPKAraq maybe windows only then? mines at 8524k and growing
17:53:41filwitAraq: i understand (and i would just import), i was just suggesting it for a module he wants tight control over
17:53:53AraqOrionPK: yeah will check on win later
17:53:54OrionPKAraq even though the program is staying at 286720
17:54:07EXetoCsymbols can be qualified either way though, which is good
17:54:18OrionPKAraq I'll check OSX in a bit
17:54:19Demosbut it is a little strange that M.TFoo.blarg works when blarg(M.TFoo) does not. I am just sharing my thoughts, I do not have a huge problem with the current system
17:54:21Araqmy favourite is: import foo except clash fyi
17:54:47Araqblacklisting instead of whitelisting works much better
17:55:06filwitAraq: woah.. din't know you could use 'except' in a import statement :D
17:55:19Demosit is in the docs filwit
17:55:37filwitDemos: what specifically? the "from ..." thing?
17:55:47Demosimport foo except
17:55:57filwiti'd imagine it is
17:56:23filwitbut the docs aren't always the best, in my experience
17:56:41Araqthe docs are always lagging behind a bit
17:56:51filwityeah..
17:57:10filwitwhen Nimrod gets a bit more popular, docs will probably need an overhaul
17:57:23filwitbesides, i would like to make them look like the website ;)
17:58:03filwitbut i don't have time for that right now (and it'll require a lot of collab i think)
17:58:47filwitDemos: your best bet to find things in the doc is the use the Index + Ctrl-F
18:00:13Araqsee you later
18:00:28*Demos quit (Ping timeout: 240 seconds)
18:01:19filwitdom96: where they talking about Nimrod at some point (on coderRadio)?
18:02:17*brson joined #nimrod
18:02:42dom96filwit: Don't think so.
18:03:03filwitdom96: lol wut? why did you direct me there then?
18:03:09filwitlol
18:03:21dom96filwit: BitPuffin sent them an email telling them to talk about Nimrod.
18:03:34filwitahhh
18:03:52filwitwell they're talking about Java.. maybe they'll move onto Nimrod
18:04:09dom96They're actually kind of talking about Java positively.
18:04:27dom96Making me want to stop listening.
18:04:27filwitcause it's practical.. no one can argue with that
18:04:40filwitit has ALL the tools..
18:04:55filwitthat saves devs time, and that's the most expensive thing for most tech companies
18:04:56EXetoCso its infrastructure then
18:05:07*DAddYE joined #nimrod
18:05:19dom96I think they're done.
18:05:25filwityeah, too bad
18:07:05EXetoCBryan Lunduke was the only reason to watch Jupiter imo :p
18:07:07*mak2 joined #nimrod
18:07:12dom96"JBot> Darn, filwit beat you to "Nimrod programming language"."
18:07:16*dom96 wonders how to vote
18:07:45filwitdidn't know there was voting...
18:08:12dom96"* You must identify to a registered nick to private message JBot"
18:08:12dom96ffs
18:08:21filwitis all they talk about laptop hardware and tablets?
18:08:35filwitdom96: i'm a guest
18:09:17dom96quick
18:09:18filwitLOL
18:09:19filwiti know
18:09:22dom96ohh
18:09:29dom96he just skips it
18:09:53filwithe probably doesn't know anything about it
18:10:13filwitprobably didn't do any research and doesn't have anything to say, too bad
18:10:40filwiteveyone's talking about Java and C#... i should post the Raytracer comparison link we did
18:11:08filwitwe ported**
18:11:38dom96oh well.
18:14:31MFlamerfilwit: does that latest raytracer benchmark use SIMD?
18:14:33filwitdom96: good idea though. looks like that show gets a lot of listeners. if they would talk about Nimrod it would get a lot of people looking
18:14:53filwitMFlamer: not yet, but i will make it use SIMD in the future
18:15:20MFlamerand it's already several times faster than C#
18:15:39filwitMFlamer: GCC does auto-vectorize though, so i'm not sure of how much of a difference it will make (usually SIMD makes more of an impact for Matrices and larger data-structures than vectors)
18:15:56dom96filwit: it was BitPuffin's idea :)
18:16:02MFlamerI see....
18:16:50filwitMFlamer: we will see though, it might make a big difference. Can't really say till i try it.
18:17:22filwitMFlamer: like you said though, Nimrod is 3-10x faster in the tests than .NET/Mono depending on the OS/plat
18:17:56MFlamerThat is really good advertising
18:18:02filwitMFlamer: without hand-tuned optimization. so that should give a clear indication about real-world benefits in your code
18:18:04OrionPKAraq hmm i get "too many open files" error on OSX
18:18:13OrionPKnot properly closing the fds I guess
18:18:17filwitMFlamer: yup!
18:18:55filwitdom96: one thing i realize that Nimrod's website isn't doing, is distributing bin builds of Nimrod for different platforms
18:19:07dom96filwit: Yeah, I was going to fix that.
18:19:13dom96Not enough time :(
18:19:24filwitdom96: i'm surprised i actually wen to through the entire process of building Nimrod the first time
18:19:46filwitdom96: in fact, the reason I came back (now that i think about it) was because Nimrod was built for Arch repos
18:20:01filwitdom96: and i could just try it without having to build everything
18:20:24filwitdom96: it's not hard, but i'm guessing a lot of people come to the website, then just leave when they're no super-simple way to try Nimrod out
18:20:33dom96Well, you can technically still try it. But not latest from github.
18:20:42filwitdom96: how?
18:20:54dom96There are download links on the website for 0.9.2
18:21:11filwitoh pffttt..
18:21:17filwitnevermind.. i'm retarded
18:21:32filwitthere IS a download link.. idk how i assumed there wasn't
18:21:49filwitgod my memory is horrible..
18:25:32*mak2 left #nimrod ("WeeChat 0.4.2")
18:32:56*CarpNet quit (Quit: Leaving)
18:46:16dom96well, see you later guys.
18:46:24filwitbye
18:53:03*shodan45 quit (Quit: Konversation terminated!)
19:04:36*fowl joined #nimrod
19:08:11*NimBot joined #nimrod
19:11:45*Endy quit (Ping timeout: 272 seconds)
19:12:46*Demos joined #nimrod
19:14:29*wlhlm joined #nimrod
19:20:59EXetoCfowl: utf-8? dðas
19:21:16EXetoCor maybe my font sucks
19:21:55EXetoCyup :<
19:22:04fowlnot sure how to find out the encoding chromium is using
19:22:05*Demos quit (Ping timeout: 245 seconds)
19:22:24*BitPuffin joined #nimrod
19:23:19fowlwell my client says "IRC (Latin/Unicode Hybrid)" whatever that means
19:24:35EXetoCnevermind, it's probably fine. I'm using the proggy font
19:27:05*BitPuffin quit (Ping timeout: 245 seconds)
19:27:39filwitbtw, fowl, your SDL2 wrapper doesn't have SDL_Quit wrapped
19:27:50fowlwhats SDL_Quit
19:27:59filwitfowl: it shuts down SDL
19:28:38fowlit does
19:28:43fowlsdl2.quit()
19:28:47filwitfowl: idk exactly, but all the C SDL2 tuts i've found call it at the end (along with SLD_glDeleteContext, which is also missing)
19:29:15filwitfowl: your right, sorry, it was the SDL_glDeleteContext that was missing, not Quit()
19:29:27*wlhlm left #nimrod ("weechat 0.4.2")
19:30:17filwitfowl: i was making something with your wrapper, but because you didn't have that I made my own (didn't know if you actually needed to call SDL_glDeleteContext or not.. thought it might have been taken care of with SDL_Quit [like it should] but the tuts show to do it)
19:30:39EXetoCwasn't the previous hardware/software-agnostic API for SDL useful?
19:30:46filwitfowl: at least the tuts i was looking at
19:30:58filwitEXetoC: ?
19:31:26fowlfilwit, ill have all the sdl_gl_* functions in there in like 10 minutes
19:31:27filwitEXetoC: it's kinda impossible to make a _fully_ agnostic API unless you want to limit it
19:32:04filwitfowl: thanks, i'll look at using your wrapper again (to avoid duplication), but I just wanted to let you know
19:32:44filwitgot to run now though
19:32:45filwitbbl
19:32:51EXetoCdidn't SDL 1 work like that?
19:34:34EXetoCor does
19:34:53fowlwork like what
19:36:00EXetoCin a backend-agnostic manner
19:36:21EXetoCsoftware, OpenGL, etc
19:37:05fowlah yea
19:37:49fowlsdl1 is more geared toward using the software renderer though, where in sdl2 it more focused on using gl/directx
19:42:01EXetoCthere are APIs for both though, but I wonder how many people don't have access to neither gl nor directx these days
19:42:22EXetoCother than those using inadequate open source drivers in *nix
19:42:47*hoverbear joined #nimrod
19:54:41*Endy joined #nimrod
19:56:10*freezerburnv quit (Quit: freezerburnv)
19:59:48*faassen quit (Quit: Leaving.)
20:08:59filwitEXetoC: SDL is mostly about using a single API to easily create a a Window, GL/DX context, and handle user input (plus some other OS specific bits)
20:09:41filwitEXetoC: also, there's loading and saving image files in different formats (which DX does to a degree, but GL does not)
20:09:58fowlfilwit, allthe gl functions should be there, keeping the GL_ prefix
20:10:20filwitEXetoC: often, it's one or two SDL lines to accomplish what many lines of normal GL init code would do, and it's completely cross-platform (including Wayland, Mir, even Android/iOS now)
20:10:26fowlwhen i clean this up a bit more and remove the static linking stuff im going to add sdl2 to the sdl package
20:10:29filwitfowl: thanks, i'll take a look later
20:11:07filwitfowl: and great news to hear SDL2 support will be added to Nimrod
20:11:26fowlwhat do you mean? its already supported (with my wrapper)
20:11:53fowli mean sdl package as in when i pull sdl out of stdlib and make it a babel package
20:13:40filwitahh, so you're going to make an SDL & SDL2 part of babel?
20:14:11filwitwait, are all exist Nimrod wrappers going to be moved to babel eventually?
20:14:25*BitPuffin joined #nimrod
20:14:39BitPuffindom96: you failed me!!
20:14:44fowlfilwit, most are
20:15:10filwitfowl: how will you include them then? Nimrod.SDL?
20:15:19fowlfilwit, https://github.com/Araq/Nimrod/issues/623
20:17:00fowlfilwit, import sdl
20:17:06BitPuffindom96: I'll check tomorrow if they read it. The feedback segment is at the top of the show
20:17:19*ics quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
20:17:37filwitfowl: i haven't played with Babel much, you can added modules without a package prefix?
20:17:40*ics joined #nimrod
20:17:56fowlyea
20:18:00filwitfowl: in which case, will it be: import sdl2 ?
20:18:02fowlyea
20:18:07filwitgreat :)
20:18:22BitPuffindom96: if they didn't read it I'll write a post on their subreddit
20:18:24fowli only use the fowltek prefix because i want to spread my glory
20:18:36filwitlol
20:18:58filwityeah, i just usually don't like using anything "unofficial"
20:19:20fowlwhy not
20:19:23filwitit grinds my OCD to have: import GL, OpenGL, Fowltek.SDL2
20:19:30filwit:P
20:19:32fowlmy wrappers are choice
20:19:37fowlyou should not import gl and opengl
20:19:39fowljust use opengl
20:19:47filwitdoesn't link
20:20:08filwitbut i was thinking i might have been supposed to.. since they conflict everywhere
20:20:23filwitbut if i just include OpenGL, then i get link errors
20:20:31filwitwill report later i guess
20:20:33fowlweird
20:20:53filwityeah, idk, says can't find libGL.so or something like that
20:21:17filwitwas causing me a headache cause i was trying it like that for awhile
20:21:26filwitthen included GL again, and eveything worked
20:21:38filwitbut i have to do GL.glSomething and OpenGL.glSomething often
20:23:02fowlstrange because im looking at gl and opengl and they both use libGL.so.1
20:23:27filwitfowl: it's possible i was messing something up and read the error wrong
20:23:46filwitfowl: i'll try again later and let you know
20:23:51fowlok
20:26:30*Endy quit (Ping timeout: 272 seconds)
20:28:38*Demos joined #nimrod
20:51:51*vidot_j quit (Quit: Leaving)
20:52:08*vidot_j joined #nimrod
20:53:32vidot_jHello, I started with nimrod and I would like to know how countdown in a while please :)
20:54:41*Jackneill quit (Remote host closed the connection)
20:56:24*dymk quit (Ping timeout: 260 seconds)
20:56:39filwithi vidot_j, welcome
20:57:12vidot_jHi filwit, thanks =)
20:57:43filwitvidot_j: countdown doesn't really work with "while" statements i don't think (it doesn't make much sense to try and use a while with 'countdown')
20:57:53filwitvidot_j: unless i misunderstand your question
20:58:14filwitvidot_j: but you can just do: for i in countdown(10, 0): ...
20:59:17filwitvidot_j: which iterate 10 times, where i == 10, 9, 8, .. 1
20:59:44filwitvidot_j: haven't used countdown() in awhile, so i might be wrong (it might start at 9 and end at 0)
21:00:25vidot_jno, i know, i want just know the inverse command of inc(var) into a while
21:01:18vidot_ji'm french, sorry for my bad expression in english ^^
21:01:56filwitno, i understand you fine. just looking it up in the docs
21:02:02filwiti think it's something like 'dec'
21:02:07EXetoCI don't know what the difference would be, and inc(x) or inc(x, 1) is basically just x += 1. inc(x, 2) would be x += 2
21:02:20filwitvidot_j: yes it's 'dec'
21:02:21EXetoCyes dec
21:03:07filwitvidot_j: although i prefer not to use those (and almost wish they didn't exist).. i prefer to just do 'i += 1' or 'i -= 1'
21:03:09*gradha joined #nimrod
21:03:54EXetoCthe character length is the same though
21:04:19EXetoCand inc/dec is for enums as well, so they're slightly more generic
21:04:20filwithonestly.. why do those even exist?
21:04:27filwitahh.. i see
21:04:29filwitstill..
21:04:53filwitthere should just be "someEnum += 1"
21:05:00vidot_jIt work, thanks for your help filwit and EXetoC
21:05:09filwitvidot_j: np :)
21:06:36filwitEXetoC: one thing i do _not_ like about Nimrods system.nim is there are too many overly abbreviated "convenience" operators procs... at least it looks that way when i glance through it
21:07:32filwitEXetoC: not that's a big deal or anything, just I think things like 'inc/dec' are virtually pointless when you've got '+=/-='
21:08:25filwitEXetoC: also, i tried teaching a friend (physics student) some C# once.. and he was confused as to the '++/--' thing and just wanted to use '+= 1' cause it was simpler to understand what was going on
21:11:09gradhafilwit: isn't that like super advanced programming? surely you mean var = var + 1
21:11:12*shodan45 joined #nimrod
21:12:19filwitgradha: there's a trade off for sure, and the engineering student picked up on += pretty quickly
21:12:46gradhasure, my point is it wasn't natural either, he had to learn it
21:12:55filwitgradha: and i would like ++/-- myself if it where possible, but it's not, and inc/dec is confusing and uses common short-var names
21:13:27filwitgradha: good point.. but there is a balance to these things i think
21:13:44EXetoCyou need very little experience to figure out ++/--, so it doesn't matter imo
21:14:13filwiti agree, only not in the case of Nimrod
21:14:37filwit(well, EVEN in the case of Nimrod it's "easy enough" to understand... i just don't things it's "ideal")
21:14:56filwitbe cause learning "++" after you've learned "+=" is natural
21:15:04filwitlearning "inc' after "+=" is not..
21:15:34filwitso... just remove 'inc' altogether. Save everyone headache, IMO
21:15:50filwitnot that that's ever going happen
21:16:43gradhafilwit: yes, very natural http://this-plt-life.tumblr.com/post/36425242948/when-somebody-uses-the-word-natural
21:17:28filwitgradha: this isn't just about my "feeling", gradha
21:17:42filwitgradha: "+=" and "++" share symbols
21:17:51filwitgradha: "+=" and "inc" do not
21:18:20gradhaif you go pedantic on natural you should actually back using words instead of sigils for readability
21:18:26fowlall my future nimrod codes will use uppercased function names and keywords only
21:18:27EXetoCinc and dec might throw when T is enum, so I don't know if it's wise to use +, +=, - and -= in that case
21:18:33fowlgoing back to BASICS
21:18:37fowlVAR x = 1
21:18:42fowlFOR y in 0..2:
21:18:48fowlINC x
21:19:29filwitgradha: words is NOT how we learn math in math class, so no, i disagree
21:19:46filwitfowl: that's inconsistent once you start adding in other math symbols
21:20:09filwitfowl: and doesn't follow what everyone learns in school from a very early age
21:20:19gradhafilwit: play the dig, rewrite nimrod to use geometry instead of math
21:20:25fowlfilwit, me?
21:20:32*fredmorcos joined #nimrod
21:20:57dom96BitPuffin: I'm sorry :(
21:21:09gradhaactually, initial advances in mathematics didn't use symbols or numbers either, more like lengths of things, ropes, etc
21:21:18filwitgradha: you're over exaggerating my arguments. I'm not arguing for something so black-n-white, only that it makes _more_ sense (for the reasons i've given) in this situation
21:21:35gradhafilwit: ok
21:21:37filwitfowl: sorry, i didn't understand that. What where you asking?
21:21:39EXetoCI'm sure there are plenty of languages that are better entry points for newbies
21:21:56fowlfilwit, im not in the convo, why u pinging me
21:21:56filwitEXetoC: that's a poor excuse for "bad" language features :P
21:22:12filwitfowl: sorry, i thought you mentioned something, probably misread
21:22:45filwitfowl: no wait, you did join, you posted the "VAR x = 1; FOR ..." thing
21:22:52filwitfowl: maybe i misunderstood that?
21:22:53EXetoCso you're fine with those operators potentially throwing when T is enum?
21:23:07fowlyea but thats not really related to anything lol
21:23:12OrionPKdom96 i pruned down the leaking example from yesterday
21:23:21dom96OrionPK: cool
21:23:22filwitEXetoC: i think there's an easy solution to that
21:23:31OrionPKhttps://gist.github.com/onionhammer/7306011
21:23:57EXetoCactually, doesn't the language handle over/underflows already?
21:24:27EXetoCI can't remember in what situations. at least when truncating values
21:25:02filwitEXetoC: does 'high' work on enums?
21:25:25Araqfilwit: yes
21:25:51AraqEXetoC: it checks for under-/overflow except there are some bugs left ...
21:26:13Araqhi vidot_j welcome
21:26:16filwitokay, so i guess having an 'inc/dec' that works with enums (& just happens to support regular ints as well.. cause it can) makes sense
21:26:20fowlthese are not the bugs you are looking for
21:26:34filwiti guess i just don't like it being promoted over '+=' cause I think it's more confusing
21:26:42dom96OrionPK: Well, i'm not sure what I can help with. It's up to Araq.
21:26:50filwitbut that's a marketing thing, and a personal preference
21:26:53filwitso yeah..
21:26:53Araqfilwit: I've stopped caring about what you consider confusing.
21:27:08Araqbasically everything that's different from C# is confusing according to you
21:27:20OrionPKdom96 starting to look that way. my non-asyncio versoin doesnt leak though
21:27:28filwitAraq: well i stopped caring about weather you care about me caring or not ;P
21:27:46filwitAraq: and that's completely false, about me only knowing C#
21:27:52filwitAraq: for one, i know more D than C#
21:28:31*gradha wonders if NSA cares about these discussions
21:28:35filwitAraq: and i have _explained_ reason to my arguments.. feel free to disagree with them, but honest i don't care if it's just "well i disagreed with him in the past, so don't really care anymore about anything he says"
21:28:46dom96OrionPK: Perhaps asyncio does something silly, but I think I would have noticed that. Unless the memory leak is subtle?
21:28:56Araqa[i++] = ++i;
21:29:15filwitand that proves?
21:29:16dom96gradha: NimBot logs all, none shall escape the NimBot, all shall interest its multi-core CPU.
21:29:29fowldom96, i can kill nimbot
21:29:37filwitAraq: i think you may be confused as to what i _was_ arguing for
21:29:39OrionPKdom96 I think it's probably very subtle.. but no idea where. the httpserver's asyncio implementation doesnt seem to leak as far as I can tell
21:30:21*NimBot joined #nimrod
21:30:22filwitAraq: seriously dude, read my arguments
21:30:27OrionPKdom96 I'm also hitting the server ~50 times per second, and it crashes on OSX
21:30:48OrionPKit eats up all available fds on OSX
21:31:10dom96fowl: Looks like a bug in marshal: https://gist.github.com/dom96/2dbe83ad6f790d6f354a
21:31:26gradhafowl: you would kill only one of its cores, it would respawn two then
21:31:33dom96Unless...
21:31:37filwitAraq: no offense, but it's kinda insulting when you open with "i dont' care about what you have to say anymore, based on my personal opinion about your opinions in general" and then you dont even know what i was arguing for in the first place
21:32:24vidot_jA readline() can be an int ?
21:32:45Araqfilwit: my "insults" are the price you have to pay for absurd nitpicking
21:32:51vidot_jvar number: int = readLine(stdin) ?
21:33:20gradhavidot_j: there are two readlines, none return an int
21:33:27EXetoCvidot_j: I see no such function. You can verify that by browsing system.nim or checking the docs on the website
21:33:41filwitAraq: sorry, i'm not trying to get frustrated or mad at you. And i understand that i _do_ nitpick... but i do that because i _care_ about Nimrod
21:33:56EXetoCthe library documentation
21:34:10filwitAraq: so when i see something that i think is wrong, i'm not going to stay silent about it until *I* understand *why* it's that way first
21:34:14gradhavidot_j: readline returns a string (or modifies a string) and you have to parse it to an integer with either strutils or parseutils
21:34:28gradhavidot_j: you can use the documentation index to search for parseInt()
21:34:37Araqfilwit: inc/dec are natural for everybody who used Turbo Pascal or Modula 2,3 or x86 assembler ...
21:34:52gradhavidot_j: depending on the complexity of your parsing you will want one module or the other
21:35:02OrionPKwhy are mommy and daddy fighting?
21:35:10filwitAraq: which is a very small percent of programmers in comparison to all the C/C++/C#/Java folks
21:35:18*fowl holds OrionPK
21:35:31vidot_jok thanks ^^
21:35:32Araqfilwit: that matters very little though
21:35:33*OrionPK shakes gently
21:35:54filwitAraq: and i already said "I see the benifit of 'inc'/'dec' for enums.. so might has well have it for ints too"
21:36:13EXetoCbut they are all flawed in their own ways
21:36:15*Demos quit (Read error: Operation timed out)
21:36:26filwitAraq: (although, i think a "+=" might be better, but i havent' thought about or care enough to think about it)
21:37:18filwitAraq: and it *does* matter when you're talking about why or why-not a language feature should exist. My whole argument against 'inc' in the first place was because we don't learn that in math school (which we all go through at a young age)
21:37:27EXetoCmy guess would be that it's that way because it aids writing generic code
21:37:32*hoverbear quit (Quit: Textual IRC Client: www.textualapp.com)
21:37:48filwitAraq: it was about illuminating confusing from _the most_ potential new Nimrod coders
21:38:00Araqfilwit: but you don't learn += or ++ either in math...
21:38:21gradhanever seen += or ++ outside of programming
21:38:34gradhaand you see really funny sigils in math, not used in programming
21:38:36filwitAraq: YES. but like i said above, it's a _more natural_ (meaning uses the same symbols) as '+'
21:38:41Araqyou learn 'cos' and 'sin' though and these have 3 letters too ... so it's consistent with math too
21:38:43filwitAraq: which we DO learn in math
21:39:15filwitAraq: only cause you learn cos/sin in school, and they aren't "basic" algebra
21:39:35Araqwhatever man, it's too stupid to go on
21:39:39filwitAraq: besides, they're in the 'Math' module, not the 'system' one
21:39:45*gradha tries to remember when in school he learned about variable assignment
21:40:07filwitAraq: i agree.. i've ALREADY agreed that inc/dec are useful anyways..
21:41:33filwitgradha: try Junior High, when you learned algebra
21:41:38*dymk joined #nimrod
21:41:51gradhafilwit: I never learned anything in algebra, too alien for my little brain
21:42:11filwitgradha: sounds like a bad teacher :)
21:42:32filwitgradha: but what do i know, i taught myself everything anyways.
21:42:37dom96Anyone care to give a summary? I don't feel like reading all the logs.
21:42:52gradhadom96: everything is confusing
21:42:58fowldom96, we should have ++ and --, no we shouldnt, bicker bicker
21:43:15EXetoCif only the quality of the teachers was the only factor :p
21:43:19filwiti never said we should have "++/--"
21:43:21filwitthat's impossible
21:43:41filwiti said 'inc/dec' shouldn't be there, since we have '+=/-='
21:43:48gradhaI'd prefer +-+ and -+- for symmetry, more pleasing asthetically
21:43:54filwitand EXetoC pointed out inc/dec worked for enums
21:44:09fowloh
21:48:42fowlmy bad
21:49:28*NimBot joined #nimrod
21:50:25gradhaEXetoC: pass some urls
21:51:56filwitdom96: i understand you're playing peacekeeper, but that really is meaningless to mean when talking about a language. I hope you & Araq know that when i suggest something, it's usually cause i think it would help Nimrod's adoption.
21:53:33gradhanimrod 0.9.8 code named "nsa" will include a statistics module which will check the frequency of use of all the standard library, and just to be sure, your email too
21:53:35filwitdom96: i don't care if people disagree or get frustrated. Nimrod has an uphill battle to fight (like all new languages), but it actually has a chance to get very popular. If i didn't think that, i wouldn't be here.
21:54:33gradhaopen question: if I want to replace all "proc" for "ƒ" do I use a template or need a macro?
21:54:53Araqgradha: you need a source code filter
21:55:24gradhahuh, really? I thought maybe a macro could rewrite itself into a proc definition
21:55:37filwitdom96: i'm actually under-exaggerating Nimrods potential really. There really isn't any other language that's as fast as C and as safe as Java
21:56:07filwitdom96: besides maybe D, but it's implementations are buggy and doesn't have good cross-platform support.
21:57:37gradhaAraq: btw, the readme.txt is not in rst, it already has other markdown links
21:58:17Araqgradha: gah that's what happens if you don't write everything yourself
21:58:22Araqfine then I guess
21:58:34dom96is readme.txt /meant/ to be in rst?
21:58:35gradhaAraq: wait, I want to try with a symbolic link first
21:58:49Araqdom96: yeah
21:58:55dom96TIL
21:59:00gradhadom96: TIL?
21:59:01EXetoC*you* are a symbolic link
21:59:09dom96gradha: today I learn
21:59:16gradhaidem
21:59:20dom96gradha: You need to go on reddit more.
21:59:21dom96:P
21:59:24EXetoCbfzq
21:59:35gradhadom96: what was that?
21:59:41*io2 joined #nimrod
21:59:58gradhaoh, yeah, browser autocompletion rocks http://www.reddit.com/r/kpop/
22:00:03BitPuffindom96: you will not be forgiven!
22:00:50gradhaI need to convince Girl's Generation to start coding in nimrod
22:02:09*BitPuffin quit (Quit: WeeChat 0.4.2)
22:02:11dom96filwit: I know that you mean well. But remember that Araq has a limit to the energy he can spend on arguing about issues which are not of utmost priority, and AFAICS the issue you raise is pretty low priority and I mean this in the nicest way possible.
22:02:28gradha$ nimrod rst2html readme.txt -> readme.txt(29, 31) Error: '*' expected
22:02:37gradhaAraq: are you really sure it was meant to be rst?
22:02:40Araqgradha: yeah yeah yeah
22:02:53gradhano worries, I'm here for the rewritting and stuff
22:04:17filwitdom96: he joined the argument. i didn't provoke him. i just spoke my mind on a matter.
22:04:50Araqfilwit, dom96 no worries, everything is fine
22:05:48filwitdom96: and, if you read my original post, it wasn't me going "i think Nimrod should have this"... i said "i think it's pointless to have .." and then i retracted later (once i saw the benefit for enums... or at least that it wasn't confusing for enums)
22:05:57*BitPuffin joined #nimrod
22:06:30filwitAraq: yeah, i agree. no hard feelings.
22:06:41filwitAraq: i understand i can get annoying anyways
22:08:57filwitthough, didn't Berkley do a study which showed that brutal arguing is the best way to create good ideas?
22:09:07filwiti watched something where someone mentioned that
22:09:12gradhafilwit: more like brutal violence
22:09:22filwitthat was is, lol
22:09:26gradhasee all the military spending
22:09:53fredmorcosgradha, :)))
22:09:55filwitit's all so clear.. the US Military is only trying to make a better world for everyone!
22:10:38EXetoCinteresting theory. maybe those youtube comments serve a purpose after all :p
22:10:52*fredmorcos quit (Quit: Leaving)
22:11:11gradha"I was going to solve this hundreds century theory but I ragequitted due to some youtube comments"
22:11:38filwitEXetoC: it was this video: http://www.youtube.com/watch?v=7XTHdcmjenI
22:12:19filwithe mentions the study at the end.. haven't looked up up though
22:12:23EXetoCI watched about 5 seconds of that before
22:14:51*zahary1 quit (Ping timeout: 260 seconds)
22:15:24gradhais there any magical switch to "nimrod rst2html" which will make the output less spartan?
22:16:19Araqgradha: no; instead I learned the gotchas of RST ...
22:16:37*Associat0r quit (Quit: Associat0r)
22:17:55gradhaI guess my markdown previewer has better default CSS
22:19:21*dymk quit (Ping timeout: 245 seconds)
22:22:35gradhayay, https://github.com/Araq/Nimrod/pull/661 is closer to pr 666
22:26:05*io2 quit ()
22:36:48*mak3 joined #nimrod
22:37:08Araqhi mak3 welcome
22:40:15VarriountPhoneBlarg
22:40:45gradhaVarriountPhone: hey, how's your bouncer?
22:41:45VarriountPhoneWon't work
22:42:15gradhaVarriountPhone: maybe it's of the "no news is good news" camp
22:42:45VarriountPhoneToo sick to fiddle with it
22:43:55VarriountPhoneLast night I threw up 4 times in as many hours
22:44:30gradhaperpetuum vomit?
22:45:05gradhaVarriountPhone: get well, we need you for pushing nimrod to the infinite, and beyond
22:46:05VarriountPhoneAww, your too nice
22:48:05*gour_ quit (Quit: WeeChat 0.4.1)
23:07:16gradhaAraq: what is config/rename.rules.cfg used for? also, is config/nimdoc.cfg read for rst2html command?
23:08:26*dymk joined #nimrod
23:10:06Araqgradha: rename.rules.cfg is used for the upcoming "nimrod pretty"
23:10:16Araqand nimdoc is used for rst2html yeah
23:10:46gradhaI've heard rumors about this pretty, is it any good?
23:11:51gradhaaww, it does not overwrite the prettified file, too safe
23:12:31gradharaising Eio is awesome
23:12:41fowleieio
23:12:51gradhathere you go
23:13:01fowlTOldMcDonald
23:13:01Araqgradha: it's not ready I guess :P
23:13:51gradhafor a case insensitive language it mangless all my upper case vars into lowercase
23:15:11gradhameh, it all compiles anyway, that's no fun
23:16:11gradhait is interesting that it lowercases stack variables but not result
23:34:46*mak3 quit (Ping timeout: 245 seconds)
23:34:46*gradha goes to #rust, waits for Araq to welcome me there
23:38:26gradhagood night
23:38:36Araqsame here
23:38:36*gradha quit (Quit: bbl, need to watch https://www.youtube.com/watch?v=n7pXRdkdJxI again)
23:38:36Araqgood night
23:41:16dom96same :P bye
23:46:06*xenagi joined #nimrod
23:57:54*shevy joined #nimrod