<< 02-02-2016 >>
00:04:42*junw_ quit (Read error: Connection reset by peer)
00:05:33*junw joined #nim
00:14:29*vendethiel quit (Ping timeout: 247 seconds)
00:18:04*junw_ joined #nim
00:18:29*Matthias247 quit (Read error: Connection reset by peer)
00:21:26*junw quit (Ping timeout: 240 seconds)
00:27:11*lompik joined #nim
00:27:26*ephja quit (Ping timeout: 252 seconds)
00:36:02*desophos joined #nim
00:45:55*Egon_ joined #nim
00:46:21*desophos quit (Read error: Connection reset by peer)
00:46:57Egon_Guys I have a noob question (can't register to the forum
00:47:28*desophos joined #nim
00:49:29Egon_That is, what are type classes and concepts in nim and how do they compare to, say, C# interfaces or Haskell type classes?
00:50:43Araq_Egon_: a concept is a user-definable type-class
00:50:52Araq_and they are totally different from an interface
00:51:10Araq_it's just a fancy way of checking constraints for generic parameters, nothing more, nothing less
01:14:24*FreezerburnV joined #nim
01:24:42*Egon_ quit (Ping timeout: 252 seconds)
01:25:45*Jesin quit (Quit: Leaving)
01:28:18*FreezerburnV quit (Ping timeout: 252 seconds)
01:35:20*gokr quit (Quit: Leaving.)
01:43:44*lankanmon_ quit (Quit: I have to Go! Bye Guys!)
01:45:17*lankanmon joined #nim
02:06:36*brson quit (Quit: leaving)
02:20:44*junw_ quit (Read error: Connection reset by peer)
02:20:55*quark1 joined #nim
02:25:41*endragor joined #nim
02:29:29*desophos_ joined #nim
02:29:48*desophos quit (Disconnected by services)
02:29:50*desophos_ is now known as desophos
02:55:14*pregressive joined #nim
02:56:25*apotheon quit (Ping timeout: 260 seconds)
02:57:06*apotheon joined #nim
03:01:31*pregressive quit (Read error: Connection reset by peer)
03:02:02*pregressive joined #nim
03:04:14*junw joined #nim
03:05:16*quark1 quit (Ping timeout: 272 seconds)
03:23:51*cfkoch joined #nim
03:24:00*vendethiel joined #nim
03:31:14cfkochalright folks, i'm pretty close to building a pkgsrc Makefile for nim
03:32:05cfkochlooks like Nim already understands what NetBSD is, so that should be easy
03:33:41*gmpreussner quit (Ping timeout: 245 seconds)
03:34:42*gmpreussner joined #nim
03:38:02*Jesin joined #nim
03:42:39*Varriount_ joined #nim
03:42:39*Varriount quit (Disconnected by services)
03:44:38Varriount_cfkoch: Yay!
03:44:43*Varriount_ is now known as Varriount
03:47:10*vendethiel quit (Ping timeout: 260 seconds)
03:48:37def-cfkoch: cool, i tested Nim on NetBSD once and it seemed fine
03:48:38*pregressive quit (Read error: Connection reset by peer)
03:49:12*pregressive joined #nim
03:54:14*MyMind quit (Read error: Connection reset by peer)
03:55:22*MyMind joined #nim
04:02:48cfkochyeah it builds OK
04:02:52cfkochthe install script is weird though
04:03:45cfkochno big deal, as long as you can install anything wherever you want, it should be fine
04:11:24cfkochOK done
04:11:26cfkochhttps://wip.pkgsrc.org/cgi-bin/gitweb.cgi?p=pkgsrc-wip.git;a=commit;h=46dd63f421ee5bd30d086fc12ec99f9bbdb3e2d5
04:13:37*krux02 quit (Remote host closed the connection)
04:16:46*Varriount_ joined #nim
04:16:57*Varriount quit (Disconnected by services)
04:41:11*endragor quit ()
04:53:20*endragor joined #nim
05:06:20Varriount_cfkoch: Oh! This is like pkgbuild on OpenBSD?
05:28:37*Varriount joined #nim
05:30:18*Varriount_ quit (Ping timeout: 252 seconds)
05:32:13endragorWhen using --compileOnly flag, can I make Nim not recreate .cpp files for modules that already exist in nimcache folder and have not been modified? at least for stdlib modules
05:34:13*pregressive quit (Remote host closed the connection)
05:40:22*pregressive joined #nim
06:11:03cfkochVarriount: closer to OpenBSD or FreeBSD Ports
06:11:50cfkochit's NetBSD's package manager, but it's portable and works on other OSes too
06:12:07*bninja joined #nim
06:14:15*lompik quit (Ping timeout: 240 seconds)
06:45:12*pregressive quit (Remote host closed the connection)
06:47:18*cfkoch left #nim (#nim)
07:13:00*qazwsx_ joined #nim
07:15:01*qazwsx_ quit (Client Quit)
07:24:01*darkf joined #nim
07:41:50*gokr joined #nim
07:43:17*gokr1 joined #nim
07:46:23*gokr quit (Ping timeout: 264 seconds)
07:57:54*Demon_Fox quit (Quit: Leaving)
07:58:24*desophos quit (Read error: Connection reset by peer)
08:01:27*yglukhov joined #nim
08:12:26*Egon_ joined #nim
08:14:28*darkf quit (Ping timeout: 265 seconds)
08:16:16*darkf joined #nim
08:16:20*vendethiel joined #nim
08:23:15*Egon_ quit (Ping timeout: 252 seconds)
08:32:59*lg_ joined #nim
08:36:54*Trustable joined #nim
08:37:44*gour joined #nim
08:39:25*vendethiel quit (Ping timeout: 260 seconds)
08:52:35bbl_Araq_: could args with no type specified be implicitly auto? proc foo(a) = a.echo? would it be a bad language decision or hard implementation detail?
08:53:27*vendethiel joined #nim
08:58:26*bninja left #nim ("Textual IRC Client: www.textualapp.com")
08:59:47*coffeepot joined #nim
09:09:23Araq_bbl_: we did that and it confused too many people
09:12:23veganskbbl_, use ``(a) => a.echo`` from ``future`` module
09:16:26*vendethiel quit (Ping timeout: 240 seconds)
09:26:36bbl_Araq_: oh dear
09:31:42*vendethiel joined #nim
09:38:46*coffeepot quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
09:39:17*coffeepot joined #nim
09:40:31*bungoman_ joined #nim
09:40:56*Ven joined #nim
09:43:26*bungoman quit (Ping timeout: 240 seconds)
09:45:24*Ven_ joined #nim
09:48:46*Ven quit (Ping timeout: 240 seconds)
09:53:08*vendethiel quit (Ping timeout: 248 seconds)
09:57:02*coffeepot quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
09:57:54*coffeepot joined #nim
10:02:01*vqrs quit (Max SendQ exceeded)
10:02:52*vqrs joined #nim
10:04:53*yglukhov quit (Ping timeout: 250 seconds)
10:04:56*Ven_ quit (Ping timeout: 252 seconds)
10:06:53*yglukhov joined #nim
10:17:12*Ven joined #nim
10:42:56*yglukhov quit (Remote host closed the connection)
10:44:23*yglukhov joined #nim
10:48:41*yglukhov quit (Ping timeout: 245 seconds)
11:00:35*junw quit (Ping timeout: 240 seconds)
11:01:35*filcuc joined #nim
11:03:24filcuchow can i express the concept of a static member (as in C++) in Nim?
11:05:27reactormonkfilcuc, make it a global something?
11:06:05Araq_use a global variable like the rest of us. or better yet a thread local variable ...
11:08:40*Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
11:14:15filcucAraq_: is there an ordering of evaluation of global variables?
11:14:44filcucnot only between them, but for examples the execution of main
11:15:55*junw joined #nim
11:17:33Araq_there is an ordering but it doesn't work too well so zahary planned yet another pragma to influence it. IMO we need something different.
11:27:36*Dildosan joined #nim
11:29:04*Salewski joined #nim
11:29:10*junw quit (Ping timeout: 260 seconds)
11:34:07SalewskiAraq: It may sound silly, but following sentence in manual sounds strange for me: It should be or, not and. And, if there are only local and global variables, why not just " Var statements declare new variables and initialize them." Same for let.
11:34:21Salewski>Var statements declare new local and global variables and initialize them.
11:34:42*johnsoft quit (Quit: Leaving)
11:40:01nchambersbecause it does both of those things
11:41:13SalewskiBut not both at the same time, so or.
11:41:47nchambersit does both in the same context, ie the program
11:42:07nchambershence and
11:43:22SalewskiWell, english is not my native language. For me it sounded a bit strange when I read it one year ago, and I just found it again :-)
11:43:35*ephja joined #nim
11:45:12*Salewski quit ()
11:48:08*vendethiel joined #nim
11:51:48Araq_learn to create PRs in the same time it takes you to post here
11:52:00nchambershe left
11:52:00*jaco60 quit (Ping timeout: 252 seconds)
11:54:04Araq_I'm sure he knows about the logging
11:54:19nchambersoh right
12:02:09endragormy earlier question (haven't figured it out): when using --compileOnly flag, can I make Nim not recreate .cpp files for modules that already exist in nimcache folder and have not been modified?
12:08:13Araq_endragor: the compiler does that out of the box unless you use --forcebuild
12:08:43*Ven joined #nim
12:11:10*vendethiel quit (Ping timeout: 260 seconds)
12:12:16*vendethiel joined #nim
12:12:55endragorah, I see, the problem was in the build script...
12:24:18*derka joined #nim
12:27:53*vqrs quit (Max SendQ exceeded)
12:28:13*vqrs joined #nim
12:28:23*krux02 joined #nim
12:28:42krux02what does 'Error: illegal capture' mean?
12:33:52flyxkrux02: you try to access a variable from an anonymous iterator or proc which you're not allowed to access
12:34:28krux02yes I do that and I want to do that
12:34:44flyxit must not be a 'var' parameter
12:35:29krux02but doesn't it mean that when I remove the var-ness, I get copies all over the place?
12:35:59flyxusually, the compiler does not create copies if it does not need to
12:36:16flyxyou should only use 'var' if you actually want to modify the parameter
12:36:23*theduke_ joined #nim
12:36:26*clone1018 quit (Ping timeout: 240 seconds)
12:36:28*pigmej quit (Ping timeout: 240 seconds)
12:36:28*theduke quit (Ping timeout: 240 seconds)
12:36:53krux02that means I can pass big objects as value when defining the `[]` operatior?
12:36:57Araq_krux02: to transfer something to the heap, it has to be copied
12:38:03*pigmej joined #nim
12:38:08*clone1018 joined #nim
12:38:26krux02actually I just want to have a local function in a function that modifies a var parameter
12:38:34derkaHi everyone
12:38:42krux02I solved it by passing the var parameter manually to the local function
12:38:50*Dildosan quit (Remote host closed the connection)
12:38:52krux02so it's not a capture anymore, but an explicit parameter
12:39:02derkaAraq_ what is the best way to go through Nim source code
12:39:07krux02compiler is happy
12:39:07flyxanother possibility is to use a ref variable
12:39:19derkaAraq_ the best way to understand the inner working of Nim
12:39:23krux02flyx: I don't use ref at all
12:39:36krux02flyx: at least not yet
12:40:26krux02not until I find a situation where I think I need it, and in all my code I haven't found a single situation.
12:42:50Araq_derka: read the internal docs
12:43:07Araq_read ast.nim
12:43:31derkathe doc directory
12:43:32derkaok
12:44:21Araq_http://nim-lang.org/docs/intern.html
12:45:14derkathx Araq
12:55:14flyxtheduke_: didn't see you here for some time. I made progress on my YAML-Parser. have you plans for yours?
12:56:15*vendethiel quit (Ping timeout: 240 seconds)
13:00:14krux02is it possible to write an iterator based algorithm library like in c++, because all algorithms provided by sequtils only work on the seq type
13:00:36*vendethiel joined #nim
13:00:41*vqrs quit (Max SendQ exceeded)
13:01:10*vqrs joined #nim
13:03:21*Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
13:08:07flyxkrux02: sure it's possible, but be aware that Nim iterators are a completely different concept from C++ iterators
13:10:00krux02flyx: yes for some reason in nim iterators are language constructs, and there are special functions that provide iteration over sequences, for an algorithm library like in c++ it would be necessary to pass iterators like objects, and I haven't seen that possible
13:11:05flyxkrux02: yes, because in Nim, iterators are coroutines (at least if they're closures), but in C, iterators are, essentially, pointers
13:12:20krux02in c++ iterators are essentially anything, often objects
13:12:56flyxkrux02: it is possible to pass iterators as first-class values
13:13:37krux02flyx: I haven't seen any examles to that yet
13:13:57flyxkrux: http://nim-lang.org/docs/manual.html#iterators-and-the-for-statement-first-class-iterators
13:19:33Araq_yeah yeah yeah Nim does it wrong and C++ gets it right and that's why they're adding "resumable functions" to C++ ... hmmm
13:22:12*vendethiel quit (Ping timeout: 248 seconds)
13:23:21krux02iterators in c++ are horrible to implement and horrible to use, I have to implement the ++ operator and a bunch on other operators to imitate the behavior of a pointer, but when using them in generic code, I should not use the ++ operator I should use std::next so it does call the right operator depending on which operators are available.
13:23:23*vqrs quit (Max SendQ exceeded)
13:24:41*vqrs joined #nim
13:25:43krux02flyx: thanks for the link
13:26:27*Dildosan joined #nim
13:26:53*Egon_ joined #nim
13:33:17*StarBrilliant joined #nim
13:36:42*vendethiel joined #nim
13:37:56Egon_Hi guys... I know you have probably already heard this question a thousand of times, but anyway, wouldn't be nice to have Rust-like traits in Nim?
13:38:48flyxEgon_: Nim has concepts
13:39:16flyxI don't know Rust traits in-depth, but I believe they're similar
13:47:35StarBrilliantI found that c2nim is not converting C structs to {.pure.}, does it matter?
13:48:32StarBrilliantOfficial manual says that "This used to be necessary for binary compatibility with other compiled languages."
13:49:20StarBrilliantSo I think if a Nim object is not marked as {.pure.}, it is not ABI compatible with underlying C code.
13:51:40Araq_c2nim knows better than you. ;-)
13:54:31Egon_The only thing that disappoints (not really, I just would like t
13:55:06Egon_then this way) is that they can't be used ad interfaces
13:55:32Araq_Egon_: there has been work on this but it's sleeping
13:56:30Egon_Ah ok, are they scheduled for the (short or long) future?
13:58:27*vendethiel quit (Ping timeout: 250 seconds)
13:58:36*Dildosan left #nim ("Leaving")
13:59:26krux02does a seq in Nim duplicate it's underlying data each time on add like a c++ std::vector does on push_back?
14:03:09*Egon_ quit (Ping timeout: 252 seconds)
14:03:51StarBrilliantkrux02: not until it reaches its capacity and needs a relocation
14:04:39krux02StarBrilliant: yes of course that's how the c++ works, too. But haw do I set the capacity?
14:05:26krux02std::vector has a constructor for size and capacity
14:05:43*lompik joined #nim
14:06:19StarBrilliantkrux02: use newSeq to create a new seq with desired size?
14:06:39StarBrilliantkrux02: then change each element as you desire
14:07:12krux02StarBriliant: I would prefer to have as size of 0 and a capicity manually set
14:07:19StarBrilliantkrux02: or if you prefer, newSeq with a large capacity, and then setLen it to 0
14:07:39StarBrilliantI'm not sure whether it works
14:07:55StarBrilliantsee the compiled C code to make sure it's ok
14:17:01*derka quit (Quit: derka)
14:22:50krux02StarBrilliant: the C code calls setLengthSeq and incrSeqV2 but I can't find it's implementation
14:23:10krux02I think seq needs more documentation
14:23:20def-krux02: last time i benchmarked that worked fine
14:23:43StarBrilliantkrux02: If I can remember it, it is in the same C code as your main program
14:23:51StarBrilliantsearch setLengthSeq in the code
14:23:54krux02def-: I really would like to know what seq does, since I use it all the time
14:24:00StarBrilliantor in nimbase.c?
14:24:34krux02aproximately 10000 hits, because it's forward declaration is in every nimcache file
14:24:35def-krux02: lib/system/sysstr.nim
14:26:12StarBrilliantbtw, I just registered an account at Nim forum with Hotmail
14:26:23StarBrilliantI haven't received my activation mail for about half a hour
14:26:27StarBrilliantis there anything wrong?
14:26:55StarBrillianthttp://forum.nim-lang.org/profile/m13253 <--- I thought it was nearly an hour
14:27:10StarBrilliantNothing in junk folder
14:27:28def-StarBrilliant: would have to ask dom96, i think he is in charge of that. personally i can say that hotmail blocks many ips totally, no idea if it's related
14:28:04StarBrilliantI could register again with Gmail or such, but that means I must give up this username
14:28:35krux02ok I now try to understand setLengthSeq and stuff in sysstr.nim, but it's unsatisfying
14:28:36def-StarBrilliant: or just wait for dom96 and he might be able to find out what's going on
14:28:53krux02how do I read the capacity of a string or ar seq?
14:28:56*coffeepot quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
14:29:07StarBrilliantdef-: okay then. just by curious, is the mail server at DigitalOcean-amsterdam?
14:29:12StarBrilliantmaybe hotmail is blocking it
14:30:43def-krux02: they have a private "reserved" field, see TGenericSeq in system.nim
14:31:50krux02def-: so capacity is private?
14:32:02*tdc joined #nim
14:32:37def-krux02: yes
14:33:09*derka joined #nim
14:33:50krux02is there a way to tell ignore the fact that 'reserved' is private and just read the data?
14:33:55reactormonkkrux02, nope.
14:34:12def-krux02: sure, with modifying system.nim or without?
14:34:22krux02without
14:34:40def-then you have to be careful because the memory layout of seqs could change at any moment
14:34:54krux02reactormonk: this is for me a good reason not to like nim a lot
14:35:43*lompik quit (Ping timeout: 250 seconds)
14:35:44krux02def-: in my program I have to pass data to opengl, that means the memory layout needs to be sequential
14:36:00reactormonkkrux02, sounds like you should not use seq then.
14:36:08krux02any non-sequential layout of the data would break the program
14:36:27krux02great :-/
14:36:35krux02not very happy here
14:36:53reactormonkYou sure you don't want an array?
14:36:54def-krux02: the data in seqs is always sequential, i just mean that new data fields could be added
14:37:16krux02def-: unlikely
14:37:19reactormonkdef-, only when you use newSeq, if oyu add more fields the mmap could do whatever
14:38:06*vendethiel joined #nim
14:38:11*derka quit (Ping timeout: 276 seconds)
14:38:11def-reactormonk: oh, that's news to me
14:38:24krux02I don't care if the mmap put my data at random places as long as in virtual memory space it's still sequential
14:38:24def-krux02: https://gist.github.com/def-/3a65d2dc5e7738cfc180
14:38:45flyxOpenGL usually doesn't store pointers you pass to it, but rather copies them to VRAM if you use VBOs and the like
14:38:58flyxso you should be perfectly fine with passing a seq to OpenGL
14:39:04flyxeven if you change it afterwards
14:39:53reactormonkkrux02, read the source.
14:40:28*derka joined #nim
14:41:20*tdc quit (Ping timeout: 245 seconds)
14:43:31def-reactormonk: where do you read that the data in seqs can be non-sequential?
14:44:12reactormonkdef-, I wouldn't expect it to be.
14:44:23reactormonkRead the source, maybe it is.
14:44:29def-it is sequential
14:44:51flyxthis should be documented as a guarantee. just like it is with std::vector in C++
14:45:03reactormonkdef-, ok, then put it into the doc.
14:48:03krux02just found out the capacity goes from 0 to 4 on the first 'add' of an item, after that it doubles the capacity when necessary
14:48:23krux02def-: thanks a lot for the code
14:53:16*tdc joined #nim
14:55:14*coffeepot joined #nim
15:17:37*vqrs quit (Max SendQ exceeded)
15:17:55*vqrs joined #nim
15:18:06warkidhi all. I'm new to Nim and curious what's the current state of cpp backend? Just tried to to loadLib("kernel3.dll") and it doesn't compile... nim 0.13 with msvc 2015 gives ...\stdlib_dynlib.cpp(390): error C2664: 'BOOL FreeLibrary(HMODULE)': cannot convert argument 1 from 'void *' to 'HMODULE'
15:33:03Araq_warkid: it's pretty stable except that some modules (like dynlib?) have not been ported
15:35:32*coffeepot quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
15:40:16warkidAraq_: so it makes sense to expect cpp backend to work and report issues in case it doesn' t? or even propose a patch occasionally?
15:40:26Araq_yup.
15:40:51Araq_though Visual Studio != C++ backend
15:41:37*pregressive joined #nim
15:42:02VarriountAraq_: Ever consider making the compiler output templates in C++ mode? :3
15:42:09warkidI understand. I'm using msvc usually and Min seems to work fine with it(i.e. 'nim c')
15:42:35warkidit's the 'nim cpp' seems to have problems with type conversions
15:43:21Varriountwarkid: It's mainly because C++ has different type conversion rules than C
15:56:20gokr1warkid: The Urhonimo stuff uses C++ and works quite fine. https://github.com/3dicc/Urhonimo
15:58:31Araq_Varriount: no, that's completely unfeasible.
16:00:34*cyraxjoe quit (Ping timeout: 240 seconds)
16:00:42*cyraxjoe_ joined #nim
16:04:22*endragor quit (Remote host closed the connection)
16:06:11*gokr joined #nim
16:06:14*gokr1 quit (Quit: Leaving.)
16:24:29*MyMind quit (Read error: Connection reset by peer)
16:25:35*MyMind joined #nim
16:32:27*jaco60 joined #nim
16:36:50*aziz joined #nim
17:00:06*vendethiel quit (Ping timeout: 240 seconds)
17:12:16*filcuc quit (Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/)
17:17:06*vendethiel joined #nim
17:34:11*cyraxjoe joined #nim
17:34:15*cyraxjoe_ quit (Ping timeout: 245 seconds)
17:50:08*dthrvr quit (Ping timeout: 250 seconds)
17:51:21*brson joined #nim
17:53:32*derka quit (Ping timeout: 252 seconds)
17:53:37*dthrvr joined #nim
17:57:50*derka joined #nim
17:58:36*shodan45 joined #nim
18:02:54*derka_ joined #nim
18:03:00*derka quit (Ping timeout: 245 seconds)
18:03:00*derka_ is now known as derka
18:06:50*Demon_Fox joined #nim
18:10:15*Matthias247 joined #nim
18:33:10*Wildefyr quit (Killed (Sigyn (Spam is off topic on freenode.)))
18:44:41*warkid_ joined #nim
18:47:24*qazwsx_ joined #nim
18:52:30*derka quit (Quit: derka)
18:54:12*Wildefyr joined #nim
19:02:43*darkf_ joined #nim
19:05:21*Pisuke joined #nim
19:05:41*Matthias247 quit (Read error: Connection reset by peer)
19:07:10*rektide_ joined #nim
19:11:10*dom96_ joined #nim
19:11:12*OmlkRoNiXz joined #nim
19:12:00*MyMind quit (*.net *.split)
19:12:00*krux02 quit (*.net *.split)
19:12:00*darkf quit (*.net *.split)
19:12:01*vegansk quit (*.net *.split)
19:12:01*ldlework quit (*.net *.split)
19:12:02*thotypous quit (*.net *.split)
19:12:02*Roin quit (*.net *.split)
19:12:02*sarlalian quit (*.net *.split)
19:12:02*dom96 quit (*.net *.split)
19:12:03*rektide quit (*.net *.split)
19:12:03*OmIkRoNiXz quit (*.net *.split)
19:12:03*dom96_ is now known as dom96
19:15:41*krux02 joined #nim
19:15:57*thotypous joined #nim
19:16:05*Roin joined #nim
19:16:13*vegansk joined #nim
19:16:17*shodan45 quit (Quit: Konversation terminated!)
19:17:22*derka joined #nim
19:17:31*sarlalian joined #nim
19:17:44*ldlework joined #nim
19:23:48*derka quit (Ping timeout: 248 seconds)
19:27:40*darkf_ is now known as darkf
19:30:03*Demon_Fox quit (Quit: Leaving)
19:39:34*desophos joined #nim
19:51:09*sarlalian quit (Ping timeout: 240 seconds)
19:51:22*sarlalian joined #nim
20:02:12*qazwsx_ quit (Quit: qazwsx_)
20:07:30*pregressive quit (Read error: Connection reset by peer)
20:08:01*pregressive joined #nim
20:09:46*shodan45 joined #nim
20:13:59*silven_ quit (Ping timeout: 264 seconds)
20:14:33*silven joined #nim
20:21:18*Jesin quit (Quit: Leaving)
20:39:48*Jesin joined #nim
21:03:29*tdc quit (Quit: Bye bye)
21:11:56*Varriount_ joined #nim
21:11:56*Varriount quit (Disconnected by services)
21:46:30*saml quit (Remote host closed the connection)
21:47:14*gour quit (Quit: WeeChat 1.3)
21:51:28*saml joined #nim
21:54:48*vqrs quit (Max SendQ exceeded)
21:55:35*vqrs joined #nim
22:38:01gokrTrying to do the following: "--passL:-static" in a cfg file. How? Also... using dietlibc I am supposed to call gcc using "diet gcc" instead of "gcc". How can I do that with the cfg file? I just fail because of the space.
22:39:20def-gokr: passL:-static doesn't work?
22:39:40gokrPerhaps passL = "-static" works.
22:39:59gokrThe "-" in static seems to trip up the cfg parser if I use "--passL:-static"
22:40:17def-gokr: --gcc.exe:dietgcc --gcc.linkerexe:dietgcc maybe
22:40:43def-but i've only tried musl, not diet
22:41:06gokrmusl worked fine since they don't use the same odd trick.
22:46:45*warkid_ quit (Ping timeout: 252 seconds)
22:50:24*darkf quit (Quit: Leaving)
23:00:29Araq_--gcc.exe:"diet gcc" ?
23:00:50Araq_with quotes because of the space.
23:00:53gokrDidn't fly for me - it generates a command later that starts with 'diet gcc' -c
23:00:56gokrAnd that barfs.
23:01:38gokrI ended up making my own command /usr/bin/dietgcc containing "diet gcc $@"
23:02:03gokrThe passL thing worked with passL = "-static"
23:02:40gokrJust toying with making my Ni interpreter small.
23:02:45Araq_ok, so everything is fine.
23:03:11gokrWell, yeah.
23:03:38gokrAnd oh, I started reading up on how to rewrite my interpreter as stackless :)
23:07:28gokrI presume that will buy me ... better performance, better depth, ability to mess with my stack (coroutines, continuations etc). Right?
23:07:46*Matthias247 joined #nim
23:14:52*Demon_Fox joined #nim
23:15:35Araq_right. plus IMHO the result is just natural
23:15:55Araq_took me some time to even understand what the Python guys mean by "stackless"
23:16:19Araq_for me a "stack based" interpreter is a hack that's harder to come up with
23:20:38*vendethiel quit (Ping timeout: 272 seconds)
23:21:05Xethe new nimscript configuration system has made me realize how powerful nim templates actually are
23:26:01*Egon_ joined #nim
23:28:23*shodan45 quit (Quit: Konversation terminated!)
23:28:52gokrAraq_: I was kinda smitten with the "recursive eval" style from my early exposure to evaluators in Scheme etc, so... it turned out the same way in Nim. But... if you take a classic iterative approach - and utilize that procs etc can be held as variables, then... stackless should come naturally I guess.
23:29:42gokrOr even use a fat switch in the loop.
23:35:33gokrBtw, got myself some "not gcsafe" warnings - but... not sure exactly why.
23:38:24*vendethiel joined #nim
23:38:25*pregressive quit ()
23:44:13federico3dom96: http://hnapp.com/?q=Nim have you tried this?
23:47:17*lompik joined #nim
23:48:20Araq_gokr: gcsafety error messages are usually perfectly clear
23:48:52gokrMmmm, not to me - but let me get back to you when I am committed and clean. Perhaps not today :)
23:49:42gokrIt basically says this method x calls y and y calls z and z calls z. I have several overloaded z methods, so... not sure what.
23:50:43gokrBtw, threadpool has some old names there.
23:55:52*Matthias247 quit (Read error: Connection reset by peer)