<< 30-10-2015 >>
00:21:47*strcmp1 quit (Ping timeout: 264 seconds)
00:42:51*yglukhov quit (Remote host closed the connection)
00:46:06*bjz joined #nim
00:50:44*bjz_ joined #nim
00:51:18*bjz quit (Ping timeout: 244 seconds)
00:55:57*strcmp1 joined #nim
00:58:13*bjz joined #nim
00:58:44*bjz_ quit (Ping timeout: 265 seconds)
00:59:18*zepolen quit (Remote host closed the connection)
01:02:40*brson quit (Quit: leaving)
01:43:19*yglukhov joined #nim
01:47:58*yglukhov quit (Ping timeout: 260 seconds)
01:59:27*lazypenguin quit (Remote host closed the connection)
02:00:03*zepolen joined #nim
02:04:38*zepolen quit (Ping timeout: 240 seconds)
02:07:00*Demos joined #nim
02:14:38*kulelu88 quit (Quit: Leaving)
02:29:49*desophos_ joined #nim
02:34:07*desophos_ quit (Ping timeout: 244 seconds)
02:37:42*pregressive joined #nim
02:42:28*pregressive quit (Ping timeout: 250 seconds)
03:05:24*DecoPerson quit (Read error: Connection reset by peer)
03:05:29*TylerE_ quit (Read error: Connection reset by peer)
03:05:59*n1ftyn8_ quit (Read error: Connection reset by peer)
03:06:03*pmbauer quit (Read error: Connection reset by peer)
03:07:33*DecoPerson joined #nim
03:07:47*TylerE_ joined #nim
03:08:33*n1ftyn8_ joined #nim
03:08:57*pmbauer joined #nim
03:34:03*strcmp1 quit (Ping timeout: 244 seconds)
03:44:35*ephja quit (Ping timeout: 264 seconds)
03:44:55*yglukhov joined #nim
03:49:05*yglukhov quit (Ping timeout: 240 seconds)
03:58:29*darkf joined #nim
04:01:15*desophos quit (Read error: Connection reset by peer)
04:01:22*strcmp1 joined #nim
04:19:39*darkf quit (Ping timeout: 240 seconds)
04:22:39*Demos quit (Read error: Connection reset by peer)
04:22:51*darkf_ joined #nim
04:30:07*darkf_ is now known as darkf
04:42:59*nande quit (Remote host closed the connection)
04:56:51*desophos joined #nim
05:01:44*zepolen joined #nim
05:01:48*desophos quit (Ping timeout: 272 seconds)
05:06:41*zepolen quit (Ping timeout: 265 seconds)
05:21:08*vendethiel joined #nim
05:29:38*bjz_ joined #nim
05:31:28*bjz quit (Ping timeout: 250 seconds)
05:44:54*vendethiel quit (Ping timeout: 250 seconds)
05:46:33*yglukhov joined #nim
05:51:54*yglukhov quit (Ping timeout: 260 seconds)
06:46:07*zepolen joined #nim
06:48:13*zepolen quit (Remote host closed the connection)
06:48:19*zepolen joined #nim
06:48:49*linkedinyou quit (Read error: Connection reset by peer)
07:13:40*strcmp1 quit (Read error: Connection reset by peer)
07:14:17*strcmp1 joined #nim
07:23:52*desophos joined #nim
07:28:18*desophos quit (Ping timeout: 255 seconds)
07:43:38*strcmp1 quit (Read error: Connection reset by peer)
07:44:15*strcmp1 joined #nim
09:11:46*coffeepot joined #nim
09:19:15*Demon_Fox quit (Quit: Leaving)
09:20:11*yglukhov joined #nim
09:47:07*bjz joined #nim
09:49:20*bjz_ quit (Ping timeout: 272 seconds)
09:49:27*darkf_ joined #nim
09:50:59*bjz_ joined #nim
09:52:04*darkf quit (Ping timeout: 246 seconds)
09:53:34*gokr_ joined #nim
09:53:38*bjz quit (Ping timeout: 268 seconds)
09:55:30*thotypous quit (Ping timeout: 260 seconds)
10:00:16*darkf_ is now known as darkf
10:16:09*Kingsquee quit (Quit: https://i.imgur.com/qicT3GK.gif)
10:25:41*makoLine quit (Ping timeout: 265 seconds)
10:37:55*jakesyl quit (Ping timeout: 246 seconds)
10:47:29OnOargh... I think my fix is right and it was a typo, moreover there are wrong comments on in semTemplBody elif s.owner == c.owner and sfGenSym in s.flags case is for locals not for template arguments
10:48:11OnObut the comments gives example of: template tmp[T](x: var seq[T]) = var yz: T, as if it was about T
10:48:15OnOAraq: you there?
10:48:40Araqalways
10:49:22Araqthe comments are likely correct
10:51:25*jakesyl joined #nim
10:51:29Araqit was not a typo
10:53:39OnOokay, then where were template args symbols to be processed in semTemplBody
10:53:49OnObecause they do not have sfGenSym
10:54:09OnOflag, only template body locals symbols have sfGenSym
10:54:31OnOtemplate generic args symbols are created semGenericParamList
10:55:04OnOvia symG either as skGenericParam for statics or skType for others (type arguments)
10:55:38Araqtemplate parameters are not sfGenSym so why would the type parameters be?
10:56:08OnOthat's why s.owner == c.owner and sfGenSym in s.flags is NOT for generic template args
10:56:12OnOso the comment is wrong
10:56:19OnOthey get into semTemplSymbol
10:56:41Araqwell tbh I have no idea what the comment means
10:56:42OnOif (s.typ != nil) and (s.typ.kind == tyGenericParam) #<- my fix correctly replaces them in the body
10:56:55OnObefore: if (s.typ != nil) and (s.typ.kind != tyGenericParam) was completely meaning less
10:57:11Araqwhy have the 'if' at all?
10:57:44OnOyou tell me, what was the purpose of this if
10:58:09Araqto not replace them in the body ;-)
10:58:23Araqbecause they should have stayed nkIdent
10:58:45Araqbecause they are then correctly replaced by the generic instantiation process
10:59:02Araqwhich is a hacky way we also should get rid of.
10:59:11Araqbut that was the intention.
10:59:20OnOhehe, but we want exactly the opposite, generic argument needs to be replaced
10:59:43OnOeverything else should stay as is
10:59:44Araqbut this was like "an exception, don't replace these"
10:59:51Araqso you should get rid of the 'if'
11:00:11Araqeverything should be replaced then, right?
11:00:49OnOfirst of all I doubt there will be any skType that is s.typ.kind != tyGenericParam
11:01:10OnOlike where it should came from?
11:02:08Araqtype Foo = object
11:02:17Araqtemplate t(): untyped = Foo
11:02:23Araqlike so?
11:02:59OnOokay so we can just get rid of (s.typ.kind == tyGenericParam)
11:03:07OnOand leave s.typ != nil check
11:03:16Araqwhat? why?
11:03:33Araqthe s.typ != nil check doesn't mean anything without the s.typ.kind test
11:04:02OnOokay I am removing the whole if then and checking if it's gonna blow
11:04:17OnOsomehow all tests passed with previous != --> ==
11:04:29OnOso probably this nkIdent was replaced somewhere else
11:05:44OnOsometimes I wonder if having generics on templates or macros make sense at all
11:06:09OnOalso there is some major shortcoming in current implementation, because all normal args and generic args are put in single list
11:06:25OnOtemplate evaluation cannot really tell if we miss normal arguments or generic arguments
11:06:33OnObecause there's no way to distinguish them
11:06:38Araqyeah I told zahary it sucks, but he woudn't listen
11:07:04AraqI agree this should be some nkPair node instead with 2 lists
11:07:16OnOor at least have empty node as a separator
11:07:28Araqugh, nah please don't
11:07:43OnOnah right now I am not going to touch it anymore
11:07:59Araqwe have nkArgList for this, with good unclear semantics
11:08:16Araqwhich we can use for this. or maybe not.
11:08:36OnObut I think it I will be able to blow up compiler with template tmp[A, B](c, d) calling tmp[int](1)
11:08:50OnOeven with my both 3496 & 3498 fixes
11:09:31Araqhow about tmp(A, B; c, d) as an unambiguous shortcut for tmp[A, B](c, d)
11:09:38AraqI'm beginning to like this
11:10:26Araqwe could also make tmp(A, B;) to mean tmp[A, B]
11:10:40OnOthis isn't a problem of syntax, but this single list, tmp[int](1) misses 1 generic arg and 1 normal arg, but evaluator think it misses 2 normal args
11:10:55OnObecause it assumes that generic args are always complete
11:10:56AraqI know.
11:11:07Araqbut the syntax is also flawed.
11:12:45OnOI think newSeq(string, string;) is less readable than newSeq[string, string]()
11:14:39*strcmp1 quit (Ping timeout: 255 seconds)
11:15:44Araqwell it surely is shorter though
11:15:50Araqand easier to type
11:17:11OnOisn't it ambigious someProc(1, let x = 1; x) ??
11:17:23Araqthat already requires ()
11:17:32OnOI see
11:17:33AraqsomeProc(1, (let x 1; x))
11:17:41Araqbut yeah ... hrm
11:18:04Araqtmp[.A, B](x, y)
11:18:24Araqwe could decide {. .} is stupid and {. } should be used instead
11:19:04Araqwhich is ofc allowed since forever.
11:19:13*strcmp1 joined #nim
11:20:16*darkf_ joined #nim
11:21:01OnOhow about tmp(:A, B: a, b)
11:22:01Araqmeh, ruby-like colon sex
11:22:19OnOok, thinking about something readable
11:23:39*darkf quit (Ping timeout: 240 seconds)
11:23:51Araqwell the current language kind of pushes us to foo[.A, B.] or foo[.A, B]
11:24:27Araqsince [. .] and (. .) always were valid tokens with no meaning.
11:25:38Araqwe can also then finally have named type parameters, foo[.T = int.] # use defaults for everything else
11:29:59OnOyeah, but it would be cool to leave them to users, btw does {{ }} have any meaning?
11:31:41*ephja joined #nim
11:35:56Araqsorry bbl
11:36:53OnOno pbm
11:39:31*elrood joined #nim
11:52:10*gokr_ quit (Ping timeout: 240 seconds)
11:53:12*octoploid joined #nim
11:54:20*gokr_ joined #nim
12:01:12*darkf_ is now known as darkf
12:01:43*boopsiesisaway is now known as boopsies
12:08:15*linkedinyou joined #nim
12:18:53Araq{{}} is the set including the empty set
12:19:23Araqand it's ugly to have mean something else
12:21:33Araqdom96: can I tell people to install nimsuggest via a nimscript rather than via nimble?
12:43:28*gokr_ quit (Read error: Connection reset by peer)
12:45:50OnOyeah, I think there's no other way than using [. or (.
12:46:13OnOmy only pbm it is quite hard to type ;P
12:47:21OnOalso \ backslash is not used, easy to type but I dunno if: newSeq \int, string\ looks nice
12:49:17*egor joined #nim
12:54:31OnOor maybe using :: eg: newTable::string::int or newSeq::int(10)
12:58:43egorHi, I have nim 0.12 and gcc 4.4.7. Got an "error: redefinition of typedef ‘TNimType’" while compiling trivial program. Am I right that gcc 4.4 cannot be used with nim 0.12 ?
13:05:48OnOAraq: here is some copilation of alternative syntaxes for templates https://gist.github.com/nanoant/b80990aa0cd864944705
13:06:21reactormonkegor, bootstrapping devel works?
13:07:57OnOegor: are you using C++ or C compiler?
13:09:14*Arrrr joined #nim
13:09:28egorOnO: c compiler
13:10:18OnOegor what OS?
13:10:59egorOnO: redhat 6.7
13:13:29egorreactormonk: You mean, does build.sh from nim-0.12.0 work? No, it doesn't work: c_code/2_2/compiler_nim.c:13: error: redefinition of typedef ‘TNimType’
13:13:50reactormonkegor, from git?
13:14:42egorreactormonk: didn't try it yet. just downloaded from http://nim-lang.org/download/nim-0.12.0.tar.xz
13:26:42Araqegor: your problem sounds familiar, I'm afraid gcc indeed it not compatible. we should be able to fix it though.
13:28:15egorAraq: ok, thanks for clarification
13:32:07OnOI just tried boostraping Nim with GCC 4.4.1 on OSX, works fine
13:33:16OnOare you sure that gcc command is gcc 4.4? or maybe you need explicit alias eg gcc-4.4 to run it?
13:33:34OnOand gcc w/o suffix is some old version
13:34:04AraqOnO: not sure which GCC version is affected but Jehan complained about this issue too and he knows what he is doing.
13:34:12egorOnO: gcc -v gave me: gcc version 4.4.7 20120313 (Red Hat 4.4.7-16) (GCC)
13:34:36AraqRed Hat has its own GCC version, so it's quite possible
13:36:49OnOAraq: btw. anyone who made this build.sh generator never heard of "set -x" :)
13:39:02egorok, looks like I've found acceptable workaround -- install clang (yum install clang) and use clang instead of gcc
13:39:15Araqindeed, I have no idea what 'set -x' means nor do I care.
13:39:32Araqmaybe they could have used full words for uncommon features.
13:40:22OnOegor: just tested on Ubuntu 4.4.7 everything compiles well
13:40:56OnOgcc version 4.4.7 (Ubuntu/Linaro 4.4.7-8ubuntu1)
13:41:09Araqfor that they all masturbate over their all-ascii-based solutions they surely don't even use ascii well.
13:42:33OnOahh... yes, I downloaded http://nim-lang.org/download/nim-0.12.0.tar.xz and it doesn't compile :)
13:43:00OnOworks on 4.6 though
13:43:02OnO:(
13:43:35OnOso I guess someone has to write that GCC 4.6 is bare minimum for Nim
13:44:20AraqI intend to release 0.12.2 shortly after 0.12.0 fixing the most problematic regressions and incompatibilities
13:45:33AraqI think this tick-tock model works rather well.
13:46:15OnOcsources repo is what version?
13:46:29Araquh oh
13:46:47Araqnobody updated csources?
13:48:29OnOMon May 25 22:01:56 2015 +0200
13:49:00Araqwell 0.11.2 should be able to compile 0.12.0
13:52:47Araqhttps://github.com/nim-lang/Nim/wiki/Creating-a-relase
14:26:31*vqrs quit (Ping timeout: 256 seconds)
14:27:55OnOAraq: can you mark it a RFC please https://github.com/nim-lang/Nim/issues/3502 ? I couldn't find an existing issue about that
14:28:26*vqrs joined #nim
14:34:17AraqOnO: interesting the barbaric old school pascal defined (.a.) to mean [a] since keyboards might not have [] readily available
14:36:49OnOyeah, something similar to C/C++ trigraphs
14:37:19Araqyup except it didn't rewrite (. .) to [] in string literals ... ;-)
14:37:20OnO??(a??) =:= [a] :)
14:37:50reactormonkOnO, how about getting rid of the array syntax instead?
14:37:59Araqreactormonk: nooooooooo!!!!!
14:38:03OnOwooot?
14:38:22AraqNim loves arrays, arrays need special syntax.
14:38:43reactormonkso do types. Which one do you use more often?
14:38:50AraqThe array is the essence of computing.
14:39:36reactormonkwould it be possible to do arr(1) ?
14:39:39AraqI use a[i] for array-like access much more often than generic types
14:39:41OnOI concur
14:40:22Araqreactormonk: yeah, technically it would be quite possible.
14:40:42OnOI just gave some ideas of new syntax, let's choose one and deprecate the old ambiguous shit
14:41:05OnOI prefer a[..] to be substript op, since it is close to math and many other languages
14:41:16OnOand feels natural
14:41:18Araqwell ... "deprecate" in non-type-contexts.
14:41:36reactormonkAraq, proc `()` ?
14:41:41Araqand even then, "deprecate" within one year.
14:41:58Araqreactormonk: read the logs, Arrrr showed how to do it.
14:42:16OnOas long we can find some decent non-eyestraining syntax
14:43:33OnOnewArray[:10](1) & newArray::10(1)
14:43:40OnOthese are my favorites
14:44:05OnO[: is meaningless today, right?
14:44:25Araqnot sure, I guess so since ':' is not a valid operator
14:44:51Araqbut then why [: ] instead of [. ] which is consistent with pragma syntax?
14:45:02OnO: is close to [ on keyboard and both require shift to be pressed
14:45:07OnOso ergonomics
14:45:28reactormonkOnO, depends on the layout as usual
14:45:36Araq[ requires alt-gr to be pressed ;-)
14:45:39reactormonkand code is more often read than written...
14:46:32Araq:: is too confusing for rubyists and c++nistas
14:46:39OnOreactormonk: agree, that's why it has to be readable too
14:47:12Araqfoo.bar[.x](a, b,c )
14:47:26Araq--> bar[.x](foo, a, b, c)
14:47:37reactormonkI'd prefer it to be symmetric
14:47:44Araqis another advantage of a non-ambiguous syntax btw
14:48:31*BitPuffin|osx joined #nim
14:48:37Araqreactormonk: both [. ] and [. .] would be possible.
14:49:13OnOthe one dot version looks really bizzare
14:50:37OnOalso if we could make it obvious even to these who don't know the language
14:51:53OnOnewSeq{:int:}(10) ?
14:52:09OnOor newSeq{:int}(10)
14:52:29OnOthen it is similar to pragma ;P
14:54:11reactormonkOnO, I rest my case.
14:59:25ephjanewSeq(int, 10) :p
15:01:43ephjawho was it that wanted to use this syntax instead in most cases? because you need to be able to make it work with inference somehow
15:02:57*pregressive joined #nim
15:05:12*gokr quit (Read error: Connection reset by peer)
15:05:36*askatasuna joined #nim
15:15:59AraqnewSeq(int, 10) is possible but different and it's hard to unify these two different ideas.
15:20:29OnOhow about new "of" keyword: newSeq(10) of int
15:20:44OnOlet s = newSeq(10) of int
15:20:54OnOlet t = newTable of (string, string)
15:21:09OnOlet d = default of int
15:21:31OnOproc newSeq(size: int) of T = ...
15:23:33Araqthat's horrible.
15:32:46*desophos joined #nim
15:38:10Arrrri prefer newSeq(int, 10)
15:41:08ephjayes but inference might be necessary in other cases
15:41:14Arrrrhonestly, i'd rather use arr(2) for arrays than some of the ideas i have read, that's my opinion
15:41:55Arrrris not that strange, one should look at arrays access like some sort of function
15:43:31Arrrror maybe for generics '<>' like in c#: newSeq<int>(123)
15:45:28*octoploid quit (Quit: WeeChat 1.3-rc1)
15:53:15ephjayou don't think it would have any negative effects on the readability? one should know what type said symbol is referring to though
15:55:54Arrrrare you referring to 'newSeq<int>(123)' ?
15:57:26ephjasorry, lack of context. I was referring to the array syntax
15:59:22ArrrrnewSeq(int, 123) ?
15:59:43ephjaarray indexing :p
15:59:53Arrrrarr(2) ?
16:00:00*egor quit (Ping timeout: 246 seconds)
16:00:09Arrrrwhat is the difference with arr[2] ?
16:07:14ephjaone uses syntax that is more generic (less specific), basically
16:09:36*filcuc joined #nim
16:10:39ephjaI dunno if it matters, and the "functor" approach is equally concise
16:11:03Arrrrhow is that?
16:21:19reactormonkArrrr, https://github.com/nim-lang/Nim/issues/3502#issuecomment-152544658 :-)
16:26:32Arrrrheh, i'm with you, i think it is the lesser evil. People are gonna freak if they have to write '[.T]' for generics.
16:28:41ephjayeah but aren't type parameters deduced by inference in most cases?
16:28:55reactormonkmost cases.
16:30:06ArrrrnewSeq isnt one of them, i dont know why
16:34:07ephjaI don't think inference is relevant in that case. anyway, foo(int) seems like a fairly recent idiom, but I don't know if it's due to an old compiler bug/limitation or the lack of momentum
16:39:53ephjabinary literals are borked?
16:40:18ephjaI don't think anyone has reported that, but it's probably not commonly used
16:42:51ephjawait a second
16:44:07ephjaFixarrayValMask = 0b00001111u8 .. FixstrValMask = 00011111u8
16:47:15ephjasyntax highlighting for these literals would be helpful
16:50:13*darkf quit (Quit: Leaving)
16:52:40ephjavertical alignment would be useful too. time to augment my vim
16:55:34*filcuc quit (Ping timeout: 246 seconds)
16:58:46federico3question about introspection: can I lookup a proc by name at runtime?
16:59:15reactormonkfederico3, unlikely. Not stored.
17:01:26*vendethiel joined #nim
17:02:45*brson joined #nim
17:02:53federico3I could probably use a template at build a lookup table at compile time tho
17:03:25*Trustable joined #nim
17:03:29*Trustable quit (Read error: Connection reset by peer)
17:07:57OnOanother idea, annotating generic arguments with dot, eg: let y = newSeq int., 10 var v: seq int. proc newSeq(A., size: int) seq A. = ...
17:08:35AraqOnO: these are all worse than simply requiring [. .] if (and only if) overly ambiguous
17:09:12Araqinsert the usual "real editors can render [. .] as Unicode brackets" argument here.
17:09:33OnOproc newSeq(size: int) of A: seq of A = ... IMHO is not ambigious and much more readable
17:09:57OnOvar v: seq of int looks really good
17:12:03Araqx.y(a, b, c) of (int, float)
17:12:28Araqif obj of SomeClass: # 'of' already exists
17:13:06Araqalso there is nothing wrong with var v: seq[int], it's the non-type context that is ambiguous
17:13:41Araqfor consistency we would allow var v: seq[.int.] but it's not like we encourage it everywhere
17:14:18*coffeepot quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
17:14:32*strcmp1 quit (Read error: Connection reset by peer)
17:14:42OnOyeah seq is builtin but Table Map or many other isn't
17:14:54Araqnot the point.
17:15:06Araqvar x: <type context>
17:15:44Araqthe type context arises from the colon, not from the builtin nature of the term 'seq'
17:15:59Araq(which is not a keyword anyway, it's just system.seq)
17:15:59*strcmp1 joined #nim
17:16:30OnOvar x: genericMacro ?
17:16:35Araqhowever, this does mean your [: ] suggestion makes sense
17:16:48Araqsince the ':' introduces a type context here then too
17:17:10Arrrrisnt better '.'? less invasive
17:17:23*linkedinyou quit (Quit: linkedinyou)
17:18:30Arrrrand slices uses [:]
17:18:34AraqArrrr: [:T, S] is perhaps more beautiful than [.T, S]
17:18:44Araqslices in Nim don't use [:]
17:18:45OnOImmutable[.Table[.string, string.].] looks like shit
17:19:11AraqImmutable[T] already looks like shit. come up with a better example :P
17:19:30OnOstatic[.Map[.string, string.].]
17:20:01Arrrrit remainds me of pragmas too much
17:20:18OnOstatic[:Map[:string, string]] is bit better
17:20:29Araqand now please consider where static[.Map[string, string]] (note the minimal usage of the '.') actually would come up.
17:21:04OnOvs static of Map of (string, string) quite very readable
17:21:15OnObut verbose too
17:21:25Araq1. 'of' is already used.
17:21:43Araq2. precedences of binary keyword operators in general is problematic
17:21:45OnOisn't of used in case, and must be 1st atom in expr?
17:23:26ArrrrIn my opinion, the problem with [./[: is that people will freak out. Generics is a basic thing in the language i think, we will see a lot of [.int], [.T]
17:23:55OnOyes this is why I am not feeling comfortable with [.
17:24:04OnOcoz it will look like spaghetti
17:24:12Arrrrhahaha
17:24:20ephjaexplicit instantiation doesn't seem that common as I said
17:24:21Arrrrbut you are actually, in my opinion, right
17:24:58Arrrrdoesnt it affecet procs/templates declarations too?
17:25:07OnOexcept all newSeq[int] and s: seq[string]
17:25:17OnOwhich there are plenty
17:26:21*vendethiel quit (Ping timeout: 255 seconds)
17:26:33Arrrrmake arrays work with () would be easier, "esthetic" and ... visual basic uses it
17:36:46*Trustable joined #nim
17:37:33Arrrrand scala http://www.scala-lang.org/api/current/index.html#scala.Array
17:44:40*UberLambda joined #nim
17:44:42*Trustable quit (Remote host closed the connection)
17:44:51*askatasuna quit (Read error: Connection reset by peer)
17:45:02*brson quit (Ping timeout: 265 seconds)
17:45:40*Trustable joined #nim
17:51:17*brson joined #nim
17:55:23*OnO gets crazy on MS Excel 2015 for Mac garbase
18:17:30AraqArrrr: and Ada. and I hated it.
18:25:59*pwzoii left #nim ("Leaving...")
18:26:03*elrood quit (Quit: Leaving)
18:29:22*Arrrrr joined #nim
18:29:33*Arrrr quit (Ping timeout: 252 seconds)
18:32:23reactormonkAraq, sometimes you'll have to bite the bullet.
18:32:26*tdc joined #nim
18:34:13Araqbecause "somebody will see [. .] everywhere in nim and freak out"?
18:34:39*Matthias247 joined #nim
18:35:06Araqsome opinions detached from reality won't make me "bite the bullet"
18:41:38*BlaXpirit joined #nim
18:53:39*Matthias247 quit (Read error: Connection reset by peer)
18:56:56reactormonkbecause the syntax is simpler with one-char operators - what occupies {} btw?
18:57:31Araqa{i} is available as a general accessor with no builtin semantics.
18:57:48Araqfor example, it can introduce by-row access for matrixes
18:58:20Araqit's another lovely feature of the language you'll gladly butcher since "Scala does not have it" ...
18:58:21*thotypous joined #nim
19:01:37ArrrrrWhy would i get this exception 'only' in release mode?: Error: unhandled exception: Can't obtain a value from a `none` [UnpackError:ObjectType]
19:03:05AraqArrrrr: now that is a good question.
19:03:20Araqthis shouldn't happen.
19:04:09BlaXpiritI made an article http://blaxpirit.com/blog/22/advanced-uses-of-travis-ci-with-nim.html
19:05:08ephjaBlaXpirit: you play xonotic?
19:05:28ephjaexetoc here btw. I have a new nick now
19:06:02*yglukhov quit (Ping timeout: 272 seconds)
19:06:26BlaXpiritephja, hi. i played xonotic
19:06:51ephjaI saw your nick in some video. small world
19:08:03ephjait's fun, but man is it difficult
19:16:34BlaXpiritwhat's the place where all the cool kids link to their articles?
19:17:48BlaXpiritright, that was already done for me :|
19:19:35ephjareddit? 4chan?
19:22:47Arrrrrhello BlaXpirit my friend. WHy dont you write about doing a game in nim, desophos was interested in the subject
19:23:03BlaXpiritArrrrr, i gave up on nim and on nim-csfml
19:23:25BlaXpiritthe lib is kinda broken :/
19:23:43BlaXpiritmaybe there were advancements with finalizers, who knows, but i doubt it
19:23:52desophosyeah that's kind of what i figured :/ that's why i'm using SDL2
19:24:20*gXen joined #nim
19:24:21ArrrrrOk, i thought you were using it judging by your article
19:25:01BlaXpiritnah, i just had some time doing nothing at work
19:25:29BlaXpiritso i went ahead and changed testing to travis, while fixing the libs according to breaking changes in nim
19:27:32BlaXpiritand then it was just a matter of adding a few comments and i got an article
19:28:41ArrrrrMaybe you get bored again and decide to write a RogueNim
19:35:47ArrrrrDid you see nim has its own scripting lang ?
19:39:02BlaXpiritwhatever
19:46:36*avsej quit (Quit: Quit)
19:47:27*avsej joined #nim
19:47:27*avsej quit (Changing host)
19:47:28*avsej joined #nim
20:25:42*Demon_Fox joined #nim
20:26:25*gXen quit ()
20:30:44*yglukhov joined #nim
20:33:17*ayeganov joined #nim
20:33:43*renatofdds joined #nim
20:34:57*CryptoToad joined #nim
20:35:35*yglukhov quit (Ping timeout: 264 seconds)
20:38:20*renatofdds quit ()
20:40:59*mat4 joined #nim
20:41:03mat4hello
20:41:11Araqhi mat4
20:41:20mat4hi Araq
20:45:46*makoLine joined #nim
20:47:38*Trustable quit (Remote host closed the connection)
20:49:06*Arrrrr quit (Quit: WeeChat 1.2)
20:52:49*mat4 quit (Quit: Verlassend)
20:54:27*UberLambda quit (Quit: GTG)
20:55:26*UberLambda joined #nim
20:55:46*makoLine quit (Ping timeout: 250 seconds)
21:01:48ephjahow's mat4's compiler coming along?
21:02:23Araqha, too late, he's gone already
21:04:46*mat4 joined #nim
21:05:05Araqmat4: how's your compiler coming along?
21:07:44mat4the compiler is finished. I'm working on an assembler which then can be used as backend for Nim
21:08:33ephjanice
21:12:38*makoLine joined #nim
21:12:47*BitPuffin|osx quit (Ping timeout: 264 seconds)
21:16:01mat4however my time for working on it is somewhat limited and I don't have the opportunity to delegate work at current
21:21:52*gokr joined #nim
21:27:32mat4I'm compiling the sources right now. What is the future status of the 'unsigned' module ?
21:27:57Araqmat4: it was merged into system.nim so all you have to do is to remove the import.
21:28:22mat4good decision, thanks
21:32:12*yglukhov joined #nim
21:36:25*yglukhov quit (Ping timeout: 240 seconds)
21:42:06mat4hmm, is Nimscript independent from the compiler ?
21:46:41Araqno, it's just Nim with the VM as backend rather than C.
21:50:19Araqyou can import it into your project though the API still sucks a bit. ;-)
21:59:54*UberLambda quit (Quit: GTG)
22:02:40mat4is someone here using Vim ?
22:03:09*pregressive quit (Remote host closed the connection)
22:03:25*pregressive joined #nim
22:03:40*mat4 quit (Quit: leaving)
22:05:38*yglukhov joined #nim
22:06:04*mat4 joined #nim
22:06:12ephjamat4: yes
22:06:35mat4syntax hightlighting seem to be broken
22:08:13ephjaI do have the latest git version and it seems to work
22:10:30mat4eh, ok recompile vim (again)
22:19:46mat4ciao
22:19:57*mat4 quit (Quit: leaving)
22:34:10*BlaXpirit quit (Quit: Konversation)
22:37:14*Jesin joined #nim
22:49:41*strcmp2 joined #nim
22:53:44*strcmp1 quit (Ping timeout: 268 seconds)
22:58:09*elrood joined #nim
22:59:55*tdc quit (Ping timeout: 256 seconds)
23:09:37*Matthias247 joined #nim
23:18:42*pregressive quit (Remote host closed the connection)
23:20:49*BitPuffin|osx joined #nim
23:32:04*mrkishi quit (Quit: Bye!)
23:34:20*jefus left #nim ("Leaving")
23:40:45*BitPuffin|osx quit (Ping timeout: 240 seconds)
23:46:10*elrood quit (Quit: Leaving)
23:48:22*boopsies is now known as boopsiesisaway
23:50:39*mrkishi joined #nim