<< 05-09-2013 >>
00:16:02*MFlamer joined #nimrod
00:16:30MFlamerAraq are you on?
00:19:02*Associ8or joined #nimrod
00:19:02*Associ8or quit (Changing host)
00:19:02*Associ8or joined #nimrod
00:19:26MFlamerI read a short forum post from awhile back where you mention utilizing tinySTM in nimrod. I'm curious if you still think this would be a good idea. If so I might do a little research...
00:20:04*DAddYE joined #nimrod
00:22:07*Associat0r quit (Ping timeout: 264 seconds)
00:23:37*Associat0r joined #nimrod
00:23:37*Associat0r quit (Changing host)
00:23:37*Associat0r joined #nimrod
00:25:02*Associ8or quit (Ping timeout: 240 seconds)
00:26:14*DAddYE quit (Ping timeout: 240 seconds)
00:27:55*Associ8or joined #nimrod
00:27:55*Associ8or quit (Changing host)
00:27:55*Associ8or joined #nimrod
00:28:45*Associat0r quit (Ping timeout: 245 seconds)
00:32:14*Associ8or quit (Ping timeout: 240 seconds)
00:36:47*MFlamer quit (Quit: Page closed)
01:23:25*DAddYE joined #nimrod
01:29:26*DAddYE quit (Ping timeout: 240 seconds)
02:18:08*q66 quit (Quit: Leaving)
02:26:29*DAddYE joined #nimrod
02:32:38*DAddYE quit (Ping timeout: 240 seconds)
02:50:44*ltbarcly__ left #nimrod ("Konversation terminated!")
02:54:19*Yeri joined #nimrod
03:14:30*brson quit (Ping timeout: 264 seconds)
03:16:02*brson joined #nimrod
03:56:19*MFlamer joined #nimrod
04:12:14*Yeri quit (Quit: Yeri)
04:20:29*OrionPK quit (Read error: Connection reset by peer)
04:48:26*brson quit (Quit: leaving)
05:02:56*DAddYE joined #nimrod
06:16:15*DAddYE quit (Ping timeout: 245 seconds)
06:28:40*MFlamer quit (Quit: Page closed)
06:31:12*schurig joined #nimrod
06:35:14schurigI'm reading the manual ... what would be a use case for a procedure with "Var return type" in real-life programming?
06:39:35*DAddYE joined #nimrod
07:15:00*zahary quit (Quit: Leaving.)
08:08:26*Araq_ joined #nimrod
08:09:57Araq_schurig: tables.mget requires 'var T' as a return type
08:20:55schurigso it returns an l-value?
08:22:13*DAddYE quit (Remote host closed the connection)
08:22:58*DAddYE joined #nimrod
08:26:15*mal`` quit (Ping timeout: 245 seconds)
08:27:51*DAddYE quit (Ping timeout: 268 seconds)
08:31:19*mal`` joined #nimrod
08:48:34Araq_yes
08:48:48*Araq_ quit (Quit: ChatZilla 0.9.90.1 [Firefox 23.0.1/20130814063812])
09:23:14*DAddYE joined #nimrod
09:30:14*DAddYE quit (Ping timeout: 240 seconds)
09:36:00NimBotAraq/Nimrod master b8bf499 Zahary Karadjov [+0 ±1 -0]: reverting the GC change
09:56:23*q66 joined #nimrod
10:26:36*DAddYE joined #nimrod
10:28:23schurighmm, does Nimrod has some variable-length array type? array is fixed, openarray is not generally usable
10:28:39schurigand the "tables" collection isn't a table at all, it's a dictionary ?!?!
10:29:05zahary_seq[T] is the dynamic array type
10:29:28zahary_var a = @[1, 2, 3] # notice the leading @
10:30:53schurigah, okay, didn't realize that this is a dynamic array, where i can add items at runtime
10:31:27*Associat0r joined #nimrod
10:33:02*DAddYE quit (Ping timeout: 240 seconds)
10:38:33*faassen joined #nimrod
10:44:17schurighmm, that doesn't work either: (nimrod 0.9.2 and 0.9.3):
10:44:17schurigvar a: seq[int]
10:44:17schuriga.add(1)
10:44:17schurigfor i in a: echo i
10:44:17schurigcog.nim(12) cog
10:44:17schurigSIGSEGV: Illegal storage access. (Attempt to read from nil?)
10:44:17schurig(I had comemnts in my file, it's actually in the a.add(1) line
10:44:50zahary_currently, sequences are initialized to nil
10:45:03zahary_you can either do var a: seq[int] = @[]
10:45:09zahary_or var a: seq[int]
10:45:17zahary_newSeq(a, 10)
10:45:23zahary_10 is an example size
10:45:50schurigurgs
10:46:13zahary_this is a common error and we plan to change this in the future
10:46:27zahary_var a: seq[int] will give you an initialized sequence
10:46:52schurigokay, I now do "var a: seq[int] = @[]"
10:47:02schurigThat's probably a FAQ entry :-)
10:47:12zahary_there is also var a = newSeq[int]()
10:47:36zahary_this is a helper proc from system.nim
10:53:26*d34th quit (Ping timeout: 240 seconds)
10:53:53schurigokay, I'm now using newSeq. In the manual it's mentioned, but as I was first looking into the examples is escaped my attention :-/
10:54:58*d34th joined #nimrod
10:58:03BitPuffinwhat 0.9.3? o.O that's not out
10:58:14BitPuffinAraq: you still haven't told me what you ment :(
10:58:34schurigBitPuffin: that was nimrod self-compiled from GIT, I have both versions here
10:59:02BitPuffinschurig: ah, for a while the version number wasn't bumped
10:59:12BitPuffinI think
10:59:38BitPuffinschurig: probably better to call it the git version though since I am not sure that it actually is 0.9.3 yet hehe
11:00:11schurigschurig@desktop:/usr/src/nimrod-git$ bin/nimrod | head -n 1
11:00:11schurigNimrod Compiler Version 0.9.3 (2013-09-05) [Linux: i386]
11:00:11schurig
11:00:20zahary_BitPuffin, btw I noticed your suggestion about vec.xz = [1, 2]; I'll try to come up with a way to support it
11:01:12BitPuffinzahary_: Cool!
11:01:32BitPuffinokay my phone is creepy
11:01:36BitPuffinit is out of control
11:02:58BitPuffinand no I am fairly sure it isn't a virus :P
11:29:29*DAddYE joined #nimrod
11:33:50*DAddYE quit (Ping timeout: 240 seconds)
11:39:41*shevy joined #nimrod
12:30:09*DAddYE joined #nimrod
12:36:38*DAddYE quit (Ping timeout: 240 seconds)
12:42:41*faassen left #nimrod (#nimrod)
12:46:12*EXetoC joined #nimrod
12:49:58*shevy quit (Read error: Connection reset by peer)
13:06:23*shevy joined #nimrod
13:33:19*DAddYE joined #nimrod
13:37:30*DAddYE quit (Ping timeout: 245 seconds)
14:20:58*tryzor joined #nimrod
14:33:57*DAddYE joined #nimrod
14:40:25*DAddYE quit (Ping timeout: 245 seconds)
15:18:13dom96BitPuffin: Odd versions mean that the nimrod compiler is unstable.
15:18:35dom960.9.2 -> release
15:18:41dom960.9.3 -> unstable
15:23:36*schurig quit (Remote host closed the connection)
15:27:26BitPuffindom96: I see! Didn't know that
15:37:05*DAddYE joined #nimrod
15:39:06BitPuffindom96: I really wonder how to get pacman to execute --depth 1. Upgrading the aur package always feels so ridiculously slow :/
15:39:23dom96BitPuffin: Ask in #archlinux :P
15:39:32BitPuffinHmm
15:39:34BitPuffinyeah
15:39:40BitPuffinThat might be a good idea
15:43:40*DAddYE quit (Ping timeout: 260 seconds)
16:02:59BitPuffinAraq: Has the repo been gc'd and repacked for a while?
16:03:58dom96the problem is the csources.zip that we have been pushing
16:04:10dom96only way to get rid of it is to rewrite the history
16:04:14dom96which will break all forks
16:06:45BitPuffindom96: have you gc'd that? :P
16:07:11BitPuffindom96: it looks like the one taking ages right now is just nimrod though
16:08:50BitPuffindom96: yep cloning csources took now time at all
16:22:19*BitPuffin quit (Quit: WeeChat 0.4.1)
16:37:17*brson joined #nimrod
16:38:53*BitPuffin joined #nimrod
16:54:23*DAddYE joined #nimrod
16:59:04*ltbarcly joined #nimrod
17:03:34*enurlyx joined #nimrod
17:03:46*fowl joined #nimrod
17:11:51*io2 joined #nimrod
17:30:52Araqhi tryzor, welcome
17:37:24BitPuffinYO!
17:37:35BitPuffinAraq: What did you mean by pretend I don't know the rules :(
17:37:42BitPuffinyou know I won't stop asiking you
17:39:33Araqwell obviously it's $ and not 'toString' and you know that
17:39:44Araqso stop pretending that you don't
17:41:01dom96Someone with actor experience should help this guy: http://forum.nimrod-code.org/t/225
17:42:35Araqdom96: yeah I know ...
17:57:00*Mat2 joined #nimrod
17:57:03Mat2hi @ all
17:57:16Araqservus Mat2
17:57:31Mat2hi Araq
18:03:07*EXetoC quit (Quit: WeeChat 0.4.1)
18:05:52*tryzor quit (Quit: Verlassend)
18:22:41Mat2continuation-passing style is a nice way to implement interpreters in functional languages, can I mark functions for tail-call optimization in Nimrod ?
18:30:35Araqyou can only rely on GCC's optimizer to do it for you
18:31:08Araqit seems quite reliable ... *cough*
18:33:58Mat2hmm, I don't know if GCC grant tail-call optimization
18:36:07Araq-foptimize-sibling-calls
18:36:09Araq Optimize sibling and tail recursive calls.
18:36:10*BitPuffin quit (Quit: WeeChat 0.4.1)
18:36:33Mat2thanks
18:44:05*BitPuffin joined #nimrod
18:47:51Mat2hi BitPuffin
19:20:35*enurlyx quit (Quit: Verlassend)
19:26:55reactormonkAraq, how well does C++ work? I have c++ api here.
19:27:40AraqI bootstrapped in C++ not too long ago
19:33:30reactormonkAraq, any c++2nim for APIs?
19:34:00Mat2hi reactormonk
19:35:00Araqsomebody could read c2nim's source code, determine it's a piece of cake once you grokked the parsing based on infinite lookahead and heuristics and add C++ support ...
19:35:20*Araq is dreaming again
19:36:20reactormonkAraq, so, nope? :-/
19:37:40Araqwell I will do it. But I don't know when. It's a nice challenge. :-)
19:38:50Mat2isn't it, that most C code is also valid in C++ ?
19:39:00Mat2if so, it should be possible to build up from the C version
19:39:20AraqMat2: that's what I'm talking about, yeah
19:40:20BitPuffinhey there Mat2!
19:40:20Araqreactormonk: you could try to use swig's XML output and to transform that to a Nimrod wrapper
19:40:55Mat2BitPuffin: Hi !
19:41:10BitPuffinMat2: whadduy?
19:41:10BitPuffinwhaddup?
19:41:10BitPuffin*
19:42:10Araqand the nice thing about this is that it doesn't require you to modify Swig's code which I wouldn't touch
19:42:20reactormonkAraq, that's an idea
19:42:20BitPuffinAraq: what did you mean by pretend I don't know the rules?
19:43:35AraqBitPuffin: read the logs. I won't answer anything about this topic anymore.
19:43:50BitPuffinoh so you did reply!
19:44:00BitPuffinI was getting used to that you weren't
19:44:15Araqand rightly so :P
19:44:40BitPuffinAraq: I wasn't pretending I didn't know it. Well I knew about $ but when I wrote the code I didn't know that it was illegal to have toString
19:45:15Mat2BitPuffin: I'm starting to implement a minimalist, memory efficient and embeddable Nimrod environment (mean' you can program interactive for embedded systems if you want)
19:45:35Mat2It's also intended as backend for Nimrod (because of the JIT compiler)
19:46:00BitPuffinMat2: so it compiles to bytecode?
19:46:15BitPuffinMat2: or is it native code with a smaller runtime you mean
19:46:25Mat2it compiles to native-code
19:46:35BitPuffinAraq: by the way is it possible to get a nimrod hello world as small as a C hello world?
19:47:35fowlBitPuffin, you tried to put toString() in std lib?
19:47:50BitPuffinfowl: no
19:47:50BitPuffinfowl: only in my library
19:48:20fowlyou're fired
19:48:20fowlget out
19:48:40BitPuffinnoooo
19:48:40BitPuffinnooo preasee
19:48:55fowlsrsly tho nobody wants to use tostring for your type and $ for everything else
19:49:10BitPuffinfowl: well $ would still be an alias
19:49:10fowlwhoop whoop juggalo brothers
19:49:40fowlBitPuffin, more liek proc tostring[t]*(some:t):string{.inline.}=$some amiright
19:49:55fowltostring*[t]
19:50:55Mat2Araq: can you please give the link to your blog ?
19:50:55AraqBitPuffin: yeah and in fact 'echo "hello world"' should do that.
19:51:20BitPuffinAraq: it doesn't
19:51:40BitPuffinAraq: even with --gc:off and -d:release it is larger
19:51:40*Hannibal_Smith joined #nimrod
19:51:51Araqhi Hannibal_Smith, welcome
19:51:51BitPuffinAraq: you maybe you interpreted it as source file size
19:51:51Hannibal_SmithHi Araq
19:52:01BitPuffinAraq: I mean executable size
19:52:11AraqBitPuffin: I got you. how much bigger is it?
19:52:21Mat2BitPuffin: you can compile the C sources with -Os
19:52:21BitPuffinAraq: forgot, gonna check
19:53:21Mat2on my system, this halves the code size compared to -d:release
19:53:36Mat2more or less
19:56:01fowltheres a nimrod switch for that too
19:56:31fowl--opt
19:57:01BitPuffinBeatriceLarsson% du -h helloc
19:57:11BitPuffin8.0K helloc
19:57:11BitPuffinBeatriceLarsson% du -h hellonim
19:57:11BitPuffin24K hellonim
19:57:11Mat2thanks
19:57:31BitPuffinhelloc was compiled with -O3 and hellonim was compiled with -d:release
19:57:31BitPuffinAraq: ^
19:58:01Araqok now disassemble and tell us where the bloat comes from :P
19:58:11fowli got 20k with -d:release --opt:size
19:58:41BitPuffinAraq: I am assuming system.nim
19:58:51BitPuffinAraq: but how am I supposed to disassemble?
19:59:51Araq--passC:"-S" iirc
19:59:51BitPuffinI am looking at the csources
20:01:16BitPuffinAraq: umm that failed the compilation
20:01:16BitPuffinnimrod c -o:hellonim -d:release --passC:"-S" hello.nim
20:01:51BitPuffin/usr/bin/ld:/home/isak/src/cvsnim/nimcache/system.o: file format not recognized; treating as linker script
20:01:51BitPuffin/usr/bin/ld:/home/isak/src/cvsnim/nimcache/system.o:1: syntax error
20:02:21BitPuffinbug?
20:02:36Araqnimrod c --noLinking --passC:"-S" hello.nim
20:03:01Araqand then you should open the .o files and see what's in there
20:03:16Mat2just compile from the nimcache directory with -S (only an idea)
20:03:36Araqor do what Mat2 says
20:03:46BitPuffinAraq: and you think I know how to read .o files? :P
20:05:11Mat2you get binaries and files with .s extension <- assembler output
20:05:41BitPuffindoesn't look like that no
20:06:51*fowl quit (Quit: Leaving)
20:07:41Araqah I see
20:07:41Mat2wait, I try it myself
20:07:51Araqthe compiler is to stupid to optimize away the memory allocator
20:08:11BitPuffinoh really?
20:09:21Araqyeah, it optimizes away the GC though :P
20:09:36BitPuffinyeah haha
20:09:36BitPuffinor actually no
20:09:51BitPuffinthe exectuable is smaller with noGC flag
20:10:01Mat2you can combine both
20:10:31BitPuffinAraq: then it is 12k
20:10:31BitPuffininstead of 24
20:11:21BitPuffinhttp://pastebin.com/YJxGmPV4
20:11:21BitPuffinAraq: ^
20:12:36Mat2BitPuffin: gcc -Os -fno-strict-aliasing <file.c> -S
20:12:36Mat2works here
20:13:11BitPuffinMat2: Well we shouldn't need to call the C compiler ourselves with nimrod. Then we'll be digging our graves :P
20:13:36Mat2(and don't forget to add -I<path-to-your-nimrod-includes>)
20:14:11BitPuffinMat2: xD
20:15:41Mat2uh, the GCC compiler seem need some additional flags for proper optimization, the assembler output is quite horrible
20:16:16BitPuffinMat2: GCC sucks :P
20:17:01BitPuffinMat2: so what makes your nimrod more suited for embedded than classic nimrod?
20:21:11Mat21.) it's an interpreter with (optional) JIT compiler, 2.) it will fit initial in ~16 kB 3.) you can develop on the board and do not need a PC
20:21:41Mat2you can do so without C (most important)
20:22:51BitPuffinMat2: but by cross compiling you should be able to get standard nimrod code even smaller?
20:25:31Mat2well, it's all about interactive development at run. For example, it's not a good idea to patch a print machine by crosscompiling because this will lead to time problems if in between a print cylinder breaks out with something about 3000 Rps
20:25:31AraqBitPuffin: it needs some slight modifications but I got it down to 8K here :P
20:26:41BitPuffinAraq: with just -d:release?
20:27:11Araqno with -d:release -d:useMalloc --gc:none -d:noSignalHandler ... XD
20:27:21BitPuffin_lol_ :)
20:28:31BitPuffinlib/nimrod/system/repr.nim(120, 12) Error: undeclared identifier: 'TCellSet'
20:28:56BitPuffinAraq: what would be optimal would be if the compiler could figure out from a simple program like that that it won't need those things
20:29:21Araqtrue but that's optimizing for a meaningless benchmark
20:29:36BitPuffinI know that it is meaningless
20:29:36Araqyou know, you can also just do: sh hello.nim
20:29:51Araqand then have no .exe at all
20:29:51BitPuffinBut people assume that it is bloated compared to C if it doesn't pass the test
20:30:16BitPuffinAraq: you mean if I add #!?
20:30:36BitPuffinbecause it didn't work for me
20:31:01Araqworks for me
20:33:11dom96Araq: You should just get the compiler to check whether the code being compiled only consists of a single simple echo, if so generate some super small asm. :P
20:33:31Araqdom96: no, generate a shell script instead :P
20:33:51BitPuffinwell that's not really that good either
20:34:01BitPuffinpeople might start comparing it to other nearly that level of C
20:34:11Mat2*lol*
20:34:11BitPuffinlike with var msg = "prutt"
20:34:36BitPuffinecho(msg, "I farted")
20:34:36BitPuffinetc
20:34:46BitPuffinthen that "optimizer" falls apart
20:34:56BitPuffinIt's a ridiculous thing to "have" to optimize
20:35:16BitPuffinbut having that does make nimrod look reaally good when people compare simple examples
20:36:16dom96Actually, I wouldn't be surprised if Nimrod optimises that into echo("prutt", "I farted).
20:36:31BitPuffindom96: well bad example then
20:36:41Araqdom96: it doesn't yet but it's do-able
20:36:51BitPuffindom96: var msg = readline \n echo("the user farted out", msg)
20:36:51dom96oops, missed a quote there.
20:37:16BitPuffinbasically some people are ridiculously picky
20:37:26BitPuffinif we can surprise them then maybe they feel confident in switching
20:37:36BitPuffinand telling their friends to do the same :P
20:38:01dom96Well you know, you can get an even smaller program by writing some ASM.
20:38:01dom96Why do these people use C?
20:38:21BitPuffindom96: because they are N00bs
20:38:21dom96In comparison to ASM it's pretty bloated no? :P
20:38:41Araqour resources are better spent caring about sane people, BitPuffin
20:38:51dom96Yes, so I don't care if we get them to switch or not.
20:38:51BitPuffinAraq: I suppose
20:38:51Mat2what kind of programmers are attracted by the executable size of an 'hello-world' example ?
20:40:51Mat2"wow, I trash the ELF header and get rid of 0,125 kB, now I must reboot after execution but that's ok"
20:41:21dom96lol
20:41:41Araq"in the mean time I can watch a MB sized youtube video on my other machine"
20:43:01*Associat0r quit (Quit: Associat0r)
20:44:11Mat2*g*
20:52:16*Hannibal_Smith quit (Quit: Sto andando via)
21:02:01*ltbarcly quit (Ping timeout: 264 seconds)
21:03:36*EXetoC joined #nimrod
21:04:51*ltbarcly joined #nimrod
21:11:01*MFlamer joined #nimrod
21:14:06Araqhi MFlamer
21:14:20Araqfigured out tinySTM already? ;-)
21:14:37Araqlooked at its docs ... they seem non-existent
21:14:58Mat2get some sleep, ciao
21:15:07*Mat2 quit (Quit: Verlassend)
21:15:08Araqbye Mat2
22:12:53*OrionPK joined #nimrod
22:16:33*Yeri joined #nimrod
22:16:44Araqhi Yeri welcome
22:16:55Yerihiya
22:21:35*ltbarcly quit (Ping timeout: 246 seconds)
22:27:23NimBotAraq/Nimrod master 69c3b82 Araq [+0 ±2 -0]: bugfix: --gc:none -d:useMalloc works again
22:27:23NimBotAraq/Nimrod master 84c89b9 Araq [+0 ±1 -0]: Merge branch 'master' of github.com:Araq/Nimrod
22:41:48*ltbarcly joined #nimrod
22:47:12BitPuffinwhy does lamps suck?
22:47:40Araqdunno. because linux, apache, mysql and php suck?
22:48:55BitPuffinAraq: what's that got to do with lamps? lol
22:49:18Araqthe famous LAMP stack?
22:49:27BitPuffinno silly
22:49:30BitPuffinI mean lamps
22:49:32BitPuffinin real life
22:50:04BitPuffinthey make paper too glary
22:50:18BitPuffincan't read properly xD
22:50:29BitPuffinAraq: linux sucks, but it is at least better than windows
22:50:44Araqdebatable
22:51:18BitPuffinperhaps
22:51:22BitPuffinBut I don't think so
22:51:26BitPuffinI'm not really a fan of either
22:51:54BitPuffinbut I can objectively say that the genjeuh sehlesh linjucks is greater
22:52:15BitPuffinjust because if it sucks I can at least replace what sucks with something that sucks less
22:52:54BitPuffinlike if I don't like the windows desktop I am just gonna have to live with it
22:53:19BitPuffinI mean it is probably POSSIBLE somehow to replace it. But it would be insane to even try
22:54:00BitPuffingranted haiku has the same problem as windows in that regard
22:54:10BitPuffinBut subjectively I like what haiku presents more
22:54:20BitPuffinso I kind of like haiku more than both linux and windows
22:54:54Araqwell Nimrod runs fine on Haiku now. right?
22:55:01BitPuffinand haiku at least really takes advantage of how integrated everything is
22:55:30BitPuffinAraq: I have only tested hello world and had failure with that bug that came with zaharys update but I am fairly sure it works just fine yes
22:55:41BitPuffinAraq: can't compile babel though :/
22:55:45Araqwell you bootstrapped
22:55:48BitPuffinbecause sockets
22:55:53BitPuffinAraq: yep that was successful
22:56:11BitPuffinAraq: just had to edit the cfg and add when not defined haiku in math.nim
22:56:16BitPuffinAraq: which should be added to master imo
22:56:36Araqso you compiled a 60K loc nimrod application ... ;-)
22:56:47Araqthey are already in master
22:56:48BitPuffinAraq: hm?
22:56:54BitPuffinoh they are? great!
22:57:07BitPuffinoh you mean nimrod itself?
22:57:10Araqwell the compiler is written in nimrod, you know ...
22:57:18BitPuffinyep :)
22:57:28BitPuffinThat's kind of a good test
22:57:32BitPuffinif you can compile the compiler
22:57:36BitPuffinyou are probably fine :P
22:58:04BitPuffinjust gotta fix those sockets
22:58:20BitPuffinthen we could probably even compile with caas
22:58:35BitPuffinwe could at least compile babel which I find important
22:58:35Araqwhen hostos == "solaris":
22:58:37Araq {.passl: "-lsocket -lnsl".}
22:58:50Araqmaybe we need the same for haiku?
22:59:03BitPuffinI wondel if haiku has nsl
22:59:14Araq when defined(solaris):
22:59:15Araq const MSG_NOSIGNAL = 0
23:00:47BitPuffinAraq: I am gonna ask if 0 is interpreted as MSG_NOSIGNAL
23:01:13Araqok but it doesn't matter much
23:01:27BitPuffinAraq: why not?
23:02:23Araqbecause then sockets fail in some edge cases I think, the main thing is to get it all through the C compiler without complains from it
23:03:33BitPuffinhmm
23:03:35BitPuffinwell
23:03:39BitPuffinI don't know much about sockets
23:03:57BitPuffinbut I kind of had the same idea when I was looking in sockets.nim the other day
23:04:19BitPuffinthat what if I added and defined(haiku) to the solaris clause
23:04:29BitPuffinbut for some reason I decided not to try (kind of stupid I know)
23:04:44Araq'or defined(haiku)'
23:04:57BitPuffinoops eah
23:04:59BitPuffinyeah*
23:05:11BitPuffinI was probably thinking or
23:05:20BitPuffinjust that my brain isn't working right now haha :)
23:06:37*BitPuffin has grand plans for nimrod if nimrod proves itself in a project or two
23:33:14*jd^p joined #nimrod
23:52:37*ltbarcly quit (Ping timeout: 248 seconds)