<< 24-11-2016 >>
00:08:09*aa__ joined #nim
00:10:38*yglukhov joined #nim
00:14:51*yglukhov quit (Ping timeout: 246 seconds)
00:16:50*shodan45 quit (Quit: Konversation terminated!)
00:18:48*ftsf_ joined #nim
00:52:33*chemist69 quit (Ping timeout: 256 seconds)
00:53:22*yglukhov joined #nim
00:57:37*yglukhov quit (Ping timeout: 240 seconds)
01:16:38*chemist69 joined #nim
01:35:38*yglukhov joined #nim
01:36:18*dddddd quit (Quit: Hasta otra..)
01:40:13*yglukhov quit (Ping timeout: 260 seconds)
01:58:30vktecIs it possible to import C macros defined in a header file as consts?
01:58:42*Jesin quit (Ping timeout: 250 seconds)
02:05:08*chemist69 quit (Ping timeout: 260 seconds)
02:18:17*yglukhov joined #nim
02:18:46*chemist69 joined #nim
02:22:54*yglukhov quit (Ping timeout: 268 seconds)
02:38:09*Jesin joined #nim
02:52:00*Jesin quit (Ping timeout: 246 seconds)
03:00:56*yglukhov joined #nim
03:02:53*chemist69 quit (Ping timeout: 256 seconds)
03:05:16*saml_ quit (Remote host closed the connection)
03:05:43*yglukhov quit (Ping timeout: 260 seconds)
03:12:33*PMunch quit (Quit: leaving)
03:14:02*brechtm joined #nim
03:16:47*chemist69 joined #nim
03:18:38*brechtm quit (Ping timeout: 256 seconds)
03:20:51*shodan45 joined #nim
03:31:56onionhammeryou should be able to
03:32:16onionhammerwith the {.importc.} pragma
03:43:20*yglukhov joined #nim
03:47:33*yglukhov quit (Ping timeout: 245 seconds)
03:51:21*mal`` quit (Quit: Leaving)
03:52:31*mal`` joined #nim
03:56:02*aa__ quit (Ping timeout: 265 seconds)
03:58:49cheatfateconstants must be calculated at compile time, but Nim compiler have not access to `c constant value` at the compile time, so it can be only `variable`.
04:46:48*yglukhov joined #nim
04:51:07*yglukhov quit (Ping timeout: 256 seconds)
05:40:10*nsf joined #nim
05:52:57*chemist69 quit (Ping timeout: 246 seconds)
05:55:48*chemist69 joined #nim
05:58:56*Jesin joined #nim
06:00:22*Jesin quit (Remote host closed the connection)
06:07:51*Jesin joined #nim
06:11:47*cheatfate quit (Read error: Connection reset by peer)
06:11:53*cheatfate_ joined #nim
06:22:07*yglukhov joined #nim
06:25:59*aa__ joined #nim
06:26:35*yglukhov quit (Ping timeout: 258 seconds)
06:27:17*kulelu88 quit (Quit: Leaving)
06:28:29*aa__ quit (Read error: Connection reset by peer)
06:30:04*space-wizard quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
06:36:57*bjz joined #nim
06:40:51*Dankrad quit (Ping timeout: 260 seconds)
06:42:50*space-wizard joined #nim
06:47:46*bjz quit (Read error: Connection reset by peer)
06:56:57*chemist69 quit (Ping timeout: 240 seconds)
07:01:42*bjz joined #nim
07:03:27*yglukhov joined #nim
07:07:40*yglukhov quit (Ping timeout: 250 seconds)
07:16:23*ftsf_ quit (Remote host closed the connection)
07:17:33*chemist69 joined #nim
07:28:04*chemist69 quit (Ping timeout: 260 seconds)
07:31:01*space-wizard quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
07:33:42*Arrrr joined #nim
07:45:26Arrrr"the compiler now detects and forbids recursive module dependencies" good grief, i think i won't update nim for a while.
07:45:59*chemist69 joined #nim
07:47:24*yglukhov joined #nim
07:51:55*yglukhov quit (Ping timeout: 252 seconds)
07:53:06FromGitter<ephja> hm :>
08:08:40Arrrrhttps://i.imgur.com/CCMSjAv.png
08:11:35AraqArrrr: does your project use this feature?
08:13:39ArrrrSometimes i have no other choice. But i can wait until you fully implement import with dependencies. It is not a commercial thing
08:13:48AraqI thought about making it a warning instead
08:16:04*bjz_ joined #nim
08:17:26flyxArrrr: *whom
08:18:13*bjz quit (Ping timeout: 248 seconds)
08:19:24*chemist69 quit (Ping timeout: 260 seconds)
08:20:55ArrrrI modified two things
08:23:01flyxAraq: would Ada's `limited with` solve the problem or is it similar to whatever D does?
08:25:14*brechtm joined #nim
08:29:37*yglukhov joined #nim
08:34:11*yglukhov quit (Ping timeout: 250 seconds)
08:34:20ArrrrCan i directly emit c define/includes?
08:34:59Araqflyx: whatever Ada does, doesn't help Nim
08:35:27Araqfor cyclic imports to work better, we need a different forward resolution
08:35:42Araqa different forward resolution means some kind of prepass in the compiler
08:36:09Araqexcept that the prepass should acknowledge Nim's flexibility.
08:36:17*brechtm quit (Remote host closed the connection)
08:36:33Araqtemplates and macros need to be resolved before you can even start to collect all the 'type' sections for the prepass
08:36:36*Jesin quit (Ping timeout: 268 seconds)
08:36:52Araqmacros themselves depend on symbol lookup
08:37:16Araq--> it's a big mess and the only viable solution is what D does in this case
08:38:10Araqthat pretty much means to rewrite large parts of the compiler.
08:38:41AraqAlternatively we could go the C++ route which allows ptrs/refs to unresolved structs/classes
08:39:00*brechtm joined #nim
08:39:29*chemist69 joined #nim
08:39:47Araqcurrently my preferred solution :-)
08:43:05*byte512 joined #nim
08:43:38flyxsomewhat similar to what Ada does. `limited with` imports a „limited view“ of the types, meaning that all types are considered limited, but you can have access types on them.
08:54:00*Andris_zbx joined #nim
09:25:56*Sembei quit (Ping timeout: 268 seconds)
09:31:22*vendethiel- joined #nim
09:31:28*Demos quit (Ping timeout: 265 seconds)
09:31:44*vendethiel quit (Ping timeout: 260 seconds)
09:31:45FromGitter<ephja> do we have an async keyboarding reading library?
09:33:26vktecwrt my earlier question about consts, C macros and FFI, this does not work: https://gist.github.com/vktec/1bb3dd15261bf6fa6383a3694f5269f1 (cc. onionhammer, cheatfate_)
09:34:16vktecFrom the compiler error, it seems consts and vars must have values, and cannot simply import them from a C header, but I could be wrong.
09:34:58vktecIf this _is_ the case, it'd be good to have a way around this, so people can write wrappers that won't go out of date if the C library changes its constants
09:40:00*shodan45 quit (Quit: Konversation terminated!)
09:41:03*andris_ joined #nim
09:41:49*Andris_zbx quit (Ping timeout: 260 seconds)
09:47:19flyxvktec: you are aware that C constants are simply preprocessor macros?
09:47:46vktecYes
09:48:34flyxa C library may never change its constants because the values are compiled into the object code importing the header. so if you update a constant in a C library, all previously compiled code using that library will fail
09:49:00flyxso a C library changing its constants is not really a use-case anywhere
09:50:22flyxvktec: and vars do *not* need to have values. but if not, you must provide a type. then it may work together with the header pragma
09:50:30vktecWell, if a C header is modified to use new values for its constants, then the Nim wrapper would have to be modified as well. I'd like to import them from the header, making the wrapper easier to maintain.
09:50:44vktecflyx: Ah, that's what I was missing. I'll try using a var with a type and importc
09:51:40flyxvktec: as I said, no-one in a sane state of mind would change the value of a constant in a C header
09:52:03flyxso it is generally fine copying the value
09:52:31vktecSo you say, but I've seen it happen :P
09:52:33*desophos joined #nim
09:54:12vktecI'll probably just copy the values for now, but I'm still interested in whether or not it's possible
09:54:21Araqthe var CNST {.importc: "CNST", header: "foo.h".}: cint hack is widely known
09:54:31Araqand a liability for the LLVM codegen
09:56:00vktecAraq: I'm assuming it's better just to copy the values and hope some idiot doesn't change them
09:57:40Araqit's better to use c2nim and keep the patches to the C files around
09:57:56Araqthis way you can regenerate your wrappers easily.
09:58:08Araqbut yeah, often that's not feasible
09:58:50vktecDamn. I completely forgot about c2nim
09:59:47vktecLucky I've not done that much work on this wrapper yet :D
10:04:44*yglukhov joined #nim
10:09:18*yglukhov quit (Ping timeout: 258 seconds)
10:13:37flyxAraq: I improved https://github.com/nim-lang/Nim/pull/4874 some days ago, could you have a look at it again?
10:15:46*yglukhov joined #nim
10:15:51*yglukhov quit (Remote host closed the connection)
10:16:50*yglukhov joined #nim
10:25:46Araqwhy does gorge take the ABCDE variant and gorgeEx ABCD?
10:26:35*chemist69 quit (Ping timeout: 256 seconds)
10:34:53*bjz_ quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
10:36:09*bjz joined #nim
10:36:40yglukhovAraq: hi, this "disallow recursive module dependencies" seems to be breaking a lot. why this change?
10:37:20Araqbecause it produced weird errors when you have recursive deps but still the wrong symbol order
10:37:47Araqbut as I said, if your project depends on it, I can make it a warning instead
10:38:02yglukhovpretty please do
10:39:47Araqmaybe I can even instead emit the error message for "undeclared identifier"?
10:39:48yglukhovshould not we move in the opposite way of allowing this things instead of prohibiting?
10:40:16Araqevery new code that I write considers recursive module deps
10:40:44yglukhovthats nice to hear of course... =)
10:43:03Araqwell we don't need a warning either, I think I can do instead:
10:43:30Araq"undeclared identifier: foo, this might be caused by this recursive module dependency ...."
10:45:45ArrrrThat's a good idea.
10:45:46*Demos joined #nim
10:47:02yglukhovAraq: im happy with anything as long as it doesnt contstraint existing recursive-deps rules even more that they are ;)
10:50:07Araqon the other hand, why not disallow it and introduce forward type decls?
10:53:41*chemist69 joined #nim
10:57:36*Demos quit (Ping timeout: 258 seconds)
11:14:18ArrrrThat can be useful too, it is a struggle to arrange imports as it is now
11:14:51Araqoh well, I stick to the improved error message for now
11:43:45*elrood joined #nim
11:52:44*vendethiel joined #nim
11:54:09*vendethiel- quit (Ping timeout: 246 seconds)
12:02:25*Snircle joined #nim
12:20:18*brechtm_ joined #nim
12:21:27*brechtm quit (Ping timeout: 246 seconds)
12:31:33*byte512 quit (Ping timeout: 268 seconds)
12:40:23*Trixarian joined #nim
12:40:53TrixarianWell, that works
12:43:25TrixarianHmmm, what happened to my ZNC
12:44:00*byte512 joined #nim
12:58:56*Arrrr quit (Quit: WeeChat 1.5)
13:05:04*Trixarian quit (Ping timeout: 252 seconds)
13:07:17*chemist69 quit (Ping timeout: 240 seconds)
13:10:53*brechtm joined #nim
13:12:28arnetheduckAraq, any issues with the branch?
13:14:35*brechtm_ quit (Ping timeout: 252 seconds)
13:14:44*arnetheduck quit (Remote host closed the connection)
13:18:47AraqI still haven't tested manually on Windows and OSX
13:28:41*chemist69 joined #nim
13:31:26*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
14:06:55*desophos quit (Read error: Connection reset by peer)
14:07:01*chemist69 quit (Ping timeout: 256 seconds)
14:08:48*vlad1777d joined #nim
14:15:54*MrAxilus[m] quit (Remote host closed the connection)
14:15:55*zielmicha[m] quit (Remote host closed the connection)
14:15:55*prose[m] quit (Write error: Connection reset by peer)
14:15:55*jivank[m] quit (Remote host closed the connection)
14:15:55*erwana[m] quit (Remote host closed the connection)
14:15:55*M-Quora quit (Remote host closed the connection)
14:15:56*terry[m] quit (Remote host closed the connection)
14:15:56*Guest73656[m] quit (Read error: Network is unreachable)
14:15:57*TheManiac quit (Read error: Connection reset by peer)
14:15:58*hohlerde quit (Read error: Connection reset by peer)
14:27:24*chemist69 joined #nim
14:34:49*zielmicha[m] joined #nim
14:41:27flyxAraq: because I added an optional onErrorBreak parameter to gorge
14:42:13flyxAraq: it's not really needed because you can just use gorgeEx and check the result, but I thought it would be a nice addition
14:42:31flyxsince I have the information about the result anyway
14:47:29*prose[m] joined #nim
14:47:30*hohlerde joined #nim
14:47:30*jivank[m] joined #nim
14:47:30*MrAxilus[m] joined #nim
14:47:30*M-Quora joined #nim
14:47:36*TheManiac joined #nim
14:47:37*erwana[m] joined #nim
14:47:37*Guest73656[m] joined #nim
14:47:38*terry[m] joined #nim
14:53:07*arnetheduck joined #nim
14:57:14arnetheduckmost interesting would be to run with -d:checkabi on osx to see how different the abi is
14:57:41arnetheduckwindows shouldn't be so much affected
15:00:17cheatfate_arnetheduck, most of windows api is depends on structures and its size, so some changes to types can affect everything :)
15:00:53cheatfate_it can even break asyncdispatch.nim code on windows, because we are using some hacks with OVERLAPPED structure
15:01:14arnetheduckcheatfate_, yeah, but most my changes were to the horribly named posix.nim (what's posix anyway).. and in practice it should keep reading the sizes from c headers anyway
15:01:54arnetheduckit's a funny feature anyway, specifying structs like that...
15:02:40cheatfate_arnetheduck, i dont think posix.nim has bad naming, maybe it has some part (like 20% or so) which is not very posix
15:02:47arnetheduckI noticed it's all-or-nothing though.. if you remove .header from a struct and the header is included because of some other header directive, you'll get a double definition error when compiling in c
15:03:17arnetheduckcheatfate_, posix has evolved over time - that file is a funny mix of a number of versions resulting in.. quite a mess
15:03:48*cheatfate_ is now known as cheatfate
15:06:35*Sembei joined #nim
15:06:38arnetheduckon that topic, I could use some help with windows and macosx for nlvm, feel free to drop by ;) it actually works pretty well, except some stupid optimization bug that breaks gc
15:07:30arnetheduckgetting the abi right in the std lib is probably the biggest issue.. on windows, it might also be useful to handle SEH
15:09:11flyxarnetheduck: I wanted to try out nlvm on OSX once, but failed at compiling llvm
15:10:08arnetheduckflyx, that I don't know about.. though they changed the build system in 3.9, might wanna try again.. osx is kind of the main platform for llvm given that the main dude works for apple
15:10:21flyxI know
15:10:51flyxI might try again sometime
15:11:12arnetheduckthat said, there's patching to be done to make it work, for sure
15:11:32arnetheducknlvm that is, llvm should be fine ;)
15:18:23*chemist69 quit (Ping timeout: 245 seconds)
15:25:08*arnetheduck quit (Ping timeout: 250 seconds)
15:27:30federico3is goldenreign around sometimes?
15:32:00FromGitter<nigredo-tori> Is there a good way to compare whether two untyped `NimNode`s refer to the same type? Or do I have to make a macro that returns a call to another macro? The whole thing feels more and more goldbergian.
15:37:02*chemist69 joined #nim
15:41:24*djellemah_ joined #nim
15:50:16*Varriount|Mobile quit (Ping timeout: 250 seconds)
15:51:55*Varriount|Mobile joined #nim
15:56:55federico3how to check in an async socket is still connected or needs to be restarted?
15:57:32*couven92 joined #nim
16:03:58*djellemah_ quit (Ping timeout: 250 seconds)
16:10:55FromGitter<ephja> the return value of recv will tell you whether or not the socket has disconnected
16:18:36FromGitter<ephja> sleepAsync and withTimeout can be used for defining timeouts
16:20:41FromGitter<ephja> federico3: I just discovered asyncdispatch.withTimeout actually. it'll come in handy
16:21:19federico3very nice
16:46:49*shodan45 joined #nim
17:20:52*djellemah_ joined #nim
17:27:44*andris_ quit (Remote host closed the connection)
17:31:02*space-wizard joined #nim
17:34:29*vendethiel quit (Ping timeout: 248 seconds)
17:35:05*vendethiel joined #nim
17:37:32*yglukhov_ joined #nim
17:41:15*yglukhov quit (Ping timeout: 258 seconds)
17:42:12*chemist69 quit (Ping timeout: 260 seconds)
17:42:21*yglukhov_ quit (Ping timeout: 268 seconds)
18:00:38*chemist69 joined #nim
18:05:04*Snircle quit (Ping timeout: 260 seconds)
18:05:22*Trustable joined #nim
18:05:51*Snircle joined #nim
18:23:02*Sembei quit (Ping timeout: 256 seconds)
18:23:37*yglukhov joined #nim
18:28:01*yglukhov quit (Ping timeout: 258 seconds)
18:38:03*Trustable quit (Remote host closed the connection)
18:38:57*Trustable joined #nim
18:39:11FromGitter<konqoro> @nigredo-tori there is a sameType proc, that it might be what you need
18:53:47*space-wizard quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
18:54:52*Snircle quit (Ping timeout: 250 seconds)
18:58:10*djellemah_ quit (Ping timeout: 250 seconds)
19:03:15*Snircle joined #nim
19:06:01*vendethiel- joined #nim
19:06:07*vendethiel quit (Ping timeout: 260 seconds)
19:10:19*brechtm quit (Read error: No route to host)
19:11:03*brechtm joined #nim
19:15:10*kulelu88 joined #nim
19:17:26*yglukhov joined #nim
19:35:53*djellemah_ joined #nim
19:46:39*djellemah_ quit (Quit: Leaving)
19:55:49*byte512 quit (Ping timeout: 248 seconds)
20:00:05*kulelu88 quit (Ping timeout: 248 seconds)
20:01:58*vendethiel- is now known as vendethiel
20:05:45*Snircle quit (Ping timeout: 260 seconds)
20:08:07*Ven joined #nim
20:08:56*Snircle joined #nim
20:10:37*Ven is now known as ven
20:22:06*ven quit (Read error: Connection reset by peer)
20:22:14*Ven joined #nim
20:25:43*space-wizard joined #nim
20:30:54*kulelu88 joined #nim
20:36:11*Matthias247 joined #nim
20:48:39*Sembei joined #nim
20:51:37*Snircle quit (Ping timeout: 256 seconds)
20:54:23*Snircle joined #nim
20:55:19*Jesin joined #nim
21:06:42vktecDespite this https://github.com/nim-lang/Nim/blob/v0.15.2/lib/system.nim#L2714 and this http://nim-lang.org/docs/system.html#open,string,FileMode,int I get the error "undeclared identifier: 'open'" when trying to use open to open a file. Any ideas what I'm doing wrong?
21:08:03*Ven quit (Ping timeout: 256 seconds)
21:09:53Araqit's not an iterator
21:10:15vktecI'm not using it like one.
21:10:30vktecvar fh = open("foo", fmWrite)
21:11:10*Ven joined #nim
21:11:11Araqnamed your file 'open'?
21:11:28vktecNope
21:11:50Araqgist it then
21:12:26vktec1 sec
21:14:36vktecWell, this is odd
21:14:49*Sembei quit (Quit: WeeChat 1.5-dev)
21:14:57vktecIt's working perfectly in a minimal example, just not in the file I want to use it in
21:15:01vktecLemme debug some more
21:15:10Araqjust gist your full module
21:16:08vktecSeems it was the filename
21:16:10vktecHow odd
21:16:43Araqthe filename is also used as the module name
21:17:06Araqthough Nim got much smarter at module name resolutions
21:17:13vktecDoes Nim not like modules ending with _nim?
21:17:44AraqNim tells you when it doesn't like something
21:17:50vktecNo, that's not it...
21:17:53vktecHow strange...
21:20:18*brechtm quit (Read error: Connection reset by peer)
21:21:02Araqwell it's definitely interesting
21:21:47vktecIt seems Nim doesn't like a nim file having the same name as a nims file
21:22:48vktecI originally had this in a nims file, but was getting "undeclared identifier: 'open'", so copied it to a nim file. Then I started running into the same issue, seemingly because of the conflicting filenames
21:23:00*Ven_ joined #nim
21:23:06vktecIt works fine with a different filename
21:23:14*Ven quit (Read error: Connection reset by peer)
21:23:27*brechtm joined #nim
21:23:33vktecNow, the root question: does NimScript have an equivalent to system.open, which doesn't seem to be supported?
21:25:21*Jesin quit (Ping timeout: 265 seconds)
21:34:26cheatfateAraq, does osproc.nim still has problems with stdin on windows?
21:36:10AraqI think so.
21:36:29*Ven_ quit (Read error: Connection reset by peer)
21:36:57*Ven joined #nim
21:36:57Araqvktec: staticRead
21:37:47vktecAnd for writing?
21:41:58*brechtm quit (Ping timeout: 250 seconds)
21:44:28vktecwriteFile worked for writing. Thanks for all the help
21:58:51cheatfateAraq, i have found one problem, not a problem of osproc.nim, but can be a problem of application you are running
21:59:07AraqI'm all ears
22:00:22cheatfatehttps://msdn.microsoft.com/en-us/library/ms682075.aspx
22:01:05cheatfatelooks like handles obtained via GetStdHandle() works only for console, but not for redirected console...
22:02:08Araqcheatfate: remember that we're looking for a regression
22:02:25Araqwell I still haven't tested NimEdit again, maybe it's not a regression
22:02:36Araqworked really well with nimedit though
22:03:36cheatfateso, no problem anymore?
22:05:02Araqno, but I'm not sure it's a regression or not
22:05:37*[CBR]Unspoken1 quit (Ping timeout: 240 seconds)
22:07:41*Ven quit (Read error: Connection reset by peer)
22:08:11*Ven joined #nim
22:13:39cheatfateok i was wrong... its better to use getstdhandle, then conin/conout
22:14:34*Trustable quit (Remote host closed the connection)
22:23:10*Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
22:24:34*nsf quit (Quit: WeeChat 1.6)
22:37:04*xet7 quit (Quit: Leaving)
22:38:55*vlad1777d quit (Ping timeout: 260 seconds)
22:38:56*xet7 joined #nim
22:53:54*krux02 joined #nim
22:54:41krux02wasn't something like this possible: proc `update=`(a: var A; index: int; value: string)
22:54:59krux02myVar.update(17) = "hello world"
22:55:06krux02seems not to work
22:55:36Araqnever was supported afaik
22:55:46Araqoverload `{}` instead
22:55:53*chemist69 quit (Ping timeout: 245 seconds)
22:55:56Araqit's sexier too
22:56:08AraqmyVar{17} = "hello world"
22:57:10dom96may as well use myVar[17] = "hello world" in that case
23:00:04vktecI've completely forgotten how object constructors are handled in Nim. Can anyone point me to a page describing it or give me a quick explanation?
23:00:06Araqyeah but every language can do this. Only Nim has the power of the curly accessors.
23:00:48Araqvktec: MyObj(field: value, fieldB: valueB) but since it's leaks every implementation detail there is, use a constructor proc
23:01:04*vlad1777d joined #nim
23:01:21vktecI was more wondering what the convention for constructor procs was
23:01:54AraqnewT or initT
23:02:03Araqdepending on ref vs object
23:03:05vktecSo newT for refs and initT for objects?
23:03:20Araqyes
23:04:10vktecOkay, thanks Araq
23:04:29vktecNo way to override the `new` operator then?
23:04:43vktecSeems like that would've been a nice thing to be able to do
23:05:02Araqno that's a horrible idea even if technically possible
23:05:05krux02is there a new operator?
23:05:20Araqno. it's a builtin proc.
23:05:45Araqyou can overload it, Nim applies the general overloading resolution rules for everything anyway
23:05:57Araqbut it's a horrible, horrible idea.
23:06:13krux02I've overridden new with a var parameter to initialize my values of my types
23:06:34krux02proc `new`(this: var MyType): void = ...
23:06:58krux02is that an ok-style or a bad idea?
23:07:31vktecFrom what Araq seems to be saying, "BADBADBADBADBAD!!!" :P
23:08:06krux02hmm I might have to change that
23:08:14krux02I never used the internal new proc
23:08:55Araqkrux02: well if MyType is not a 'ref' it's only bad style :-)
23:09:14krux02it's not a ref
23:09:27dom96Please stick to the NEP-1 whenever possible
23:09:27krux02I have not a single use case of a ref type in any of my projects yet
23:09:38dom96There are many projects out there that are not
23:09:56dom96The main culprit is putting `type` on the same line as the object definition
23:10:09*vlad1777d_ joined #nim
23:10:15dom96The `type` is a type section! Use it as such!
23:10:19*vktec quickly corrects that
23:10:20vktec:P
23:10:27Araqkrux02: sounds like your code is awesome :-)
23:10:42*vlad1777d quit (Ping timeout: 244 seconds)
23:11:12*Araq is serious.
23:12:35krux02not sure if my code is awesome, there as surely a lot of improvements possible
23:12:38dom96Because everything is allocated on the stack?
23:12:46krux02but honestly I am quite happy with what I can do with nim
23:14:52krux02nim is the best (statically typed language that I know) to transpose an algorithm
23:17:07krux02and as far as I know every statically typed language is bad at refactoring, so maybe it just the best language in existence to do that
23:19:31FromGitter<ephja> it beats COBOL easily
23:22:15*cheatfate quit (Quit: Leaving)
23:22:52*chemist69 joined #nim
23:23:36Araqwait ... what?
23:23:58Araqrefactoring is easy pease with static typing, the complier tells you what needs changes
23:24:34Araqinstead of hacking around and producing trivial bugs all the time which then are caught by testing. or not.
23:24:56Araqbut even if they are, compilers are faster than test suites
23:26:15dom96So true
23:27:25*Demos joined #nim
23:27:33krux02generally compilers are faster than test suits, but have heard stories of rust
23:30:19krux02Araq: what do you think of the idea, to add nil checks to both setLen procs in system?
23:31:08krux02then proc newSeq[T](s: var seq[T]; len: Natural) would become completely redundant
23:31:16Araqjust patch the compiler already to use small strings/vectors optimization
23:31:30Araqand get rid of 'nil' for these two types
23:32:29*yglukhov quit (Remote host closed the connection)
23:32:52krux02so that means when I have an object that contains seq and string as member values, they will become valid and empty?
23:33:04Araqaye
23:33:16krux02whay does "aye" mean?
23:33:27Araqyes
23:33:29dom96Aye Captain
23:33:33krux02ok
23:34:02dom96Araq is the captain of this ship, apparently.
23:34:26krux02that would truely be awesome
23:34:44krux02I really like that
23:35:24krux02dom96: but the captain does not say "aye"
23:35:36krux02or does he?
23:37:02dom96http://www.yarr.org.uk/talk/
23:37:25Araqyou never watched pirates of the caribbean, did you?
23:39:34krux02I did watch it 10 years ago or so
23:40:18krux02and played Monkey island more than 20 years ago
23:47:23*[CBR]Unspoken joined #nim
23:48:56*vlad1777d_ quit (Ping timeout: 252 seconds)
23:51:58FromGitter<Alan-FGR> (https://files.gitter.im/nim-lang/Nim/RXIr/capt.PNG)
23:52:10FromGitter<Alan-FGR> Hey guys, I'm trying to compile a code and I'm getting that error... any ideas?
23:52:24FromGitter<Alan-FGR> Just getting started really
23:53:07FromGitter<Alan-FGR> I've configured nim to use vcc and added the path to the tools to the system Path variable
23:53:26FromGitter<Alan-FGR> I can run for example cl.exe just fine, I don't think that's the problem really
23:54:21FromGitter<Alan-FGR> Apparently some includes could be the problem, no idea how the nim compilers is supposed to work
23:58:59*shodan45 quit (Quit: Konversation terminated!)
23:59:07FromGitter<Alan-FGR> What's that limits.h file?