<< 02-09-2015 >>
00:02:16*Demon_Fox joined #nim
00:08:33*yglukhov joined #nim
00:13:37*yglukhov quit (Ping timeout: 246 seconds)
00:22:17*^aurora^ quit (Quit: My Mac has gone to sleep. ZZZzzz…)
00:25:52*imapiekindaguy joined #nim
00:30:24*avsej quit (Ping timeout: 264 seconds)
01:02:32*CryptoToad quit (Quit: Leaving)
01:41:29*imapiekindaguy quit (Quit: WeeChat 1.4-dev)
02:09:52*pregressive quit (Remote host closed the connection)
02:14:24*avsej joined #nim
02:14:24*avsej quit (Changing host)
02:14:24*avsej joined #nim
02:23:35*bendiken quit (Ping timeout: 264 seconds)
02:30:19*BitPuffin|osx quit (Ping timeout: 244 seconds)
02:58:19*yglukhov joined #nim
03:02:36*yglukhov quit (Ping timeout: 250 seconds)
03:10:00*enquora quit (Quit: enquora)
03:13:05*bendiken joined #nim
03:18:50*darkf joined #nim
03:24:40*strcmp1 quit (Quit: Leaving)
03:29:43*NimBot joined #nim
03:44:48*pregressive joined #nim
03:51:47*NimBot joined #nim
04:13:13*NimBot joined #nim
04:27:03*strcmp1 joined #nim
04:28:21*endragor joined #nim
04:31:30*gokr1 joined #nim
05:04:37*NimBot joined #nim
05:13:01*NimBot joined #nim
05:14:57*pregressive quit (Remote host closed the connection)
05:26:12*NimBot joined #nim
05:34:44*NimBot joined #nim
05:46:24*strcmp1 quit (Quit: Leaving)
05:53:04*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
06:01:16*jszymanski joined #nim
06:06:13*NimBot joined #nim
06:06:42*yglukhov joined #nim
06:11:50*yglukhov quit (Ping timeout: 240 seconds)
06:22:00*bendiken joined #nim
06:22:15*Kingsqueee joined #nim
06:22:42*Kingsquee quit (Read error: Connection reset by peer)
06:26:06*bendiken quit (Ping timeout: 240 seconds)
06:27:07*NimBot joined #nim
06:48:04*kilon quit (Read error: Connection reset by peer)
06:56:45gokr1Good morning!
06:57:44Demon_FoxMornin
06:58:08Demon_FoxMorning*
06:58:45gokr1federico3: About to post to the forum and also personally email those 48 that got unsubbed.
06:59:04gokr1I can email the list to you
07:01:02*NimBot joined #nim
07:05:44*filcuc joined #nim
07:10:18*kilon quit ()
07:19:50*^aurora^ joined #nim
07:35:58*Trustable joined #nim
07:48:54*keyle joined #nim
07:51:31*Sembei joined #nim
07:56:50*Demon_Fox quit (Ping timeout: 250 seconds)
07:57:40*Demon_Fox joined #nim
08:00:32OnOAraq: what kind of videos, of your screen screen, or edit video cam footage?
08:09:25*irrequietus joined #nim
08:10:45*Sembei quit (Max SendQ exceeded)
08:11:59*Sembei joined #nim
08:13:12*darkf_ joined #nim
08:16:29*darkf quit (Ping timeout: 244 seconds)
08:21:01*xet7_ quit (Remote host closed the connection)
08:21:30*xet7_ joined #nim
08:26:16*jszymanski quit (Ping timeout: 246 seconds)
08:27:19*jszymanski joined #nim
08:28:56*Ven joined #nim
08:30:06*darkf_ is now known as darkf
08:42:38*^aurora^ quit (Quit: My Mac has gone to sleep. ZZZzzz…)
08:44:07*Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
08:44:38*Demon_Fox quit (Quit: Leaving)
08:51:39*yglukhov joined #nim
08:51:46*Ven joined #nim
08:51:55*yglukhov quit (Remote host closed the connection)
08:52:11*yglukhov joined #nim
08:52:15*yglukhov quit (Remote host closed the connection)
08:56:53*yglukhov joined #nim
08:59:01OnOAraq: anyways I recommend Camtasia
09:02:45*durm joined #nim
09:31:36*Ven quit (Ping timeout: 264 seconds)
09:36:44*jszymanski quit (Quit: computer sleeps...)
09:39:54*jaco60 joined #nim
09:42:10*Sembei quit (Ping timeout: 272 seconds)
09:48:16*bendiken joined #nim
09:49:47*kilon joined #nim
09:52:24*Sembei joined #nim
09:59:18*huonw quit (Ping timeout: 244 seconds)
10:01:12*huonw joined #nim
10:05:32*Miruya quit (Quit: Page closed)
10:08:06*Ven joined #nim
10:16:31*xcombelle joined #nim
10:32:45*Ven quit (Read error: Connection reset by peer)
10:53:53*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
10:54:54*bendiken joined #nim
11:05:17*NimBot joined #nim
11:05:35*bendiken joined #nim
11:12:08*golak joined #nim
11:15:18*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
11:16:22*bendiken joined #nim
11:26:50*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
11:27:35*bendiken joined #nim
11:34:38*BitPuffin joined #nim
11:36:10*Ven joined #nim
11:37:24*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
11:38:19*bendiken joined #nim
11:43:35*Ven quit (Ping timeout: 250 seconds)
11:48:49*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
11:48:49*dashed joined #nim
11:49:43*bendiken joined #nim
11:50:22*Sembei quit (Ping timeout: 244 seconds)
12:00:14*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
12:01:06*bendiken joined #nim
12:06:08*Ven joined #nim
12:08:27*Sembei joined #nim
12:11:27*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
12:11:49*NimBot joined #nim
12:12:19*bendiken joined #nim
12:23:24*bendiken quit (Read error: Connection reset by peer)
12:23:40*bendiken joined #nim
12:24:14*NimBot joined #nim
12:24:57*Sembei quit (Max SendQ exceeded)
12:24:58*matkuki joined #nim
12:26:07*Sembei joined #nim
12:29:35*infinity0 quit (Ping timeout: 264 seconds)
12:31:31*chemist69 joined #nim
12:33:41*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
12:34:33*bendiken joined #nim
12:35:37*matkuki quit (Quit: ChatZilla 0.9.92 [Firefox 40.0.3/20150826023504])
12:40:01*Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
12:42:17*infinity0 joined #nim
12:44:38*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
12:45:32*bendiken joined #nim
12:52:44*repax joined #nim
12:56:01*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
12:56:53*bendiken joined #nim
12:57:35*vendethiel joined #nim
13:02:47*raza joined #nim
13:06:50*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
13:07:37*jszymanski joined #nim
13:07:41*bendiken joined #nim
13:09:04*golak quit (Quit: leaving)
13:09:24*golak joined #nim
13:12:58*endragor_ joined #nim
13:13:56*kilon quit ()
13:16:36*endragor quit (Ping timeout: 255 seconds)
13:17:11*endragor_ quit (Ping timeout: 250 seconds)
13:19:43*NimBot joined #nim
13:35:39*joelmo quit (Quit: Connection closed for inactivity)
13:36:24*lokulin quit (Ping timeout: 264 seconds)
13:36:58*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
13:38:00*bendiken joined #nim
13:38:56*endragor joined #nim
13:45:24*enquora joined #nim
13:47:42*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
13:48:52*bendiken joined #nim
13:49:08*keks__ joined #nim
13:49:26*lokulin joined #nim
13:49:31keks__hi
13:51:19keks__little question, what is the prefered way to make ed25519 crypto with nim? Is there a usable binding to NaCl?
13:52:06federico3keks__: https://github.com/nim-lang/packages/blob/master/packages.json for a package list
13:59:07*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
13:59:59*bendiken joined #nim
14:00:28keks__federico3: thanks, i've searched the net/the package list already, it seems that the only thing that could work is the openssl binding... Or do i not see the wood for the trees? :)
14:06:17*golak quit (Ping timeout: 268 seconds)
14:10:05*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
14:10:56*bendiken joined #nim
14:11:01*pregressive joined #nim
14:18:41*zemm quit (Remote host closed the connection)
14:18:49*zemm joined #nim
14:21:02*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
14:21:58*bendiken joined #nim
14:23:12*Sembei quit (Ping timeout: 264 seconds)
14:29:15*^aurora^ joined #nim
14:29:55*Kingsqueee quit (Quit: http://i.imgur.com/EsXzoum.png)
14:32:00*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
14:32:56*bendiken joined #nim
14:33:43federico3keks__: nacl or libsodium?
14:36:08*keyle quit (Quit: http://twitter.com/keyle/)
14:40:50*infinity0 quit (Remote host closed the connection)
14:41:19*infinity0 joined #nim
14:43:21*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
14:44:14*bendiken joined #nim
14:44:25keks__federico3: libsodium looks nice, though. I was not aware of an NaCl fork.
14:49:09keks__Something to play with after work :)
14:50:34keks__federico3: Very nice! ( https://github.com/judofyr/sodium.nim ) thanks again for the hint!!
14:51:46*thotypous quit (Ping timeout: 240 seconds)
14:54:13*brson joined #nim
14:54:18*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
14:55:10*bendiken joined #nim
14:56:29*endragor quit (Remote host closed the connection)
15:05:15federico3keks__: and it seems to be quite popular
15:05:17*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
15:05:22*pregressive quit (Remote host closed the connection)
15:06:10*bendiken joined #nim
15:06:38*pregressive joined #nim
15:07:05*NimBot joined #nim
15:09:05*filcuc quit (Ping timeout: 240 seconds)
15:14:07*thotypous joined #nim
15:15:56*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
15:17:19*NimBot joined #nim
15:22:07*dashed quit (Quit: Connection closed for inactivity)
15:26:12*drewsrem joined #nim
15:26:41*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
15:27:32*bendiken joined #nim
15:27:58*boopsiesisaway is now known as boopsies
15:30:21drewsremIs it possible to construct an object "inplace" and cast into a "byte array"? - i.e. I have: "var data: array[32, char]" and an object-type (e.g. foo) which sizeof is 32, now I want to construct an instance of that type into that data var, I tried simply casting it by "data = cast[array[32, char]](foo(params))" but it doesn't work, atm I create a tempoprary var, construct foo there and movemem it into data. - Any pointers?
15:30:46*pregressive quit ()
15:31:31*pregressive joined #nim
15:35:17*jszymanski quit (Quit: computer sleeps...)
15:37:38*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
15:38:31*bendiken joined #nim
15:39:26*^aurora^ quit (Quit: My Mac has gone to sleep. ZZZzzz…)
15:41:18*ChrisMAN quit (Remote host closed the connection)
15:45:12*ChrisMAN joined #nim
15:48:09*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
15:49:03*bendiken joined #nim
15:59:06*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
16:00:05*bendiken joined #nim
16:09:39*keks__ quit (Ping timeout: 246 seconds)
16:09:54*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
16:10:55*bendiken joined #nim
16:16:10*enquora quit (Quit: enquora)
16:26:26*raza quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
16:27:48*raza joined #nim
16:28:42*Jesin quit (Quit: Leaving)
16:29:09*Jesin joined #nim
16:31:56*strcmp1 joined #nim
16:36:54*icebattle joined #nim
16:39:47*yglukhov quit (Ping timeout: 265 seconds)
16:46:57*X67r joined #nim
16:55:40*brson quit (Quit: leaving)
16:57:23*keks__ joined #nim
16:57:48*endragor joined #nim
17:01:16*icebattle quit (Quit: Page closed)
17:09:16*yglukhov joined #nim
17:11:23*yglukhov quit (Remote host closed the connection)
17:11:40*yglukhov joined #nim
17:12:40*raza quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
17:12:47*icebattle joined #nim
17:14:18*raza joined #nim
17:20:03*xcombelle quit (Ping timeout: 255 seconds)
17:25:54*gokr_ joined #nim
17:26:09*Demon_Fox joined #nim
17:27:43*Matthias247 joined #nim
17:34:43*darkf quit (Ping timeout: 246 seconds)
17:35:16*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
17:36:18*bendiken joined #nim
17:42:26*unclechu joined #nim
17:42:32*unclechu quit (Max SendQ exceeded)
17:43:33*unclechu joined #nim
17:46:33*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
17:47:40*bendiken joined #nim
17:52:26*golak joined #nim
17:54:14*matkuki joined #nim
17:56:03*keks__ quit (Ping timeout: 246 seconds)
17:57:55*matkuki quit (Client Quit)
17:58:01*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
17:59:03*bendiken joined #nim
18:03:02*^aurora^ joined #nim
18:08:48*xcombelle joined #nim
18:09:08*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
18:09:52*bendiken joined #nim
18:10:58yglukhovdrewsrem, have you tried through pointers?
18:11:37*irrequietus quit (Ping timeout: 265 seconds)
18:11:54drewsremyglukhov, isn't that what I'm doing now? - create a temporary var and copymem it over
18:11:58yglukhove.g. var myArray : array[32, char]; cast[ptr MyObj](addr myArray[0])[] = MyObj(myField:0)
18:12:31drewsremyglukhov, ah thanks, I'll try this
18:13:22yglukhovnp
18:13:25*xcombelle quit (Remote host closed the connection)
18:20:05*kilon joined #nim
18:20:06*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
18:20:10drewsremOn a different note: any idea how I could achieve the following: http://ix.io/kAL ?
18:21:10*bendiken joined #nim
18:21:26drewsremi.e. store a closure with a single typed-parameter in a "generic/untyped proc-type" variable and then retrieve it again
18:23:48*BitPuffin quit (Ping timeout: 264 seconds)
18:24:23*irrequietus joined #nim
18:26:45drewsremi.e. function pointers
18:31:01*endragor quit (Remote host closed the connection)
18:31:28*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
18:32:26*bendiken joined #nim
18:33:15*NimBot joined #nim
18:42:20*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
18:43:14*bendiken joined #nim
18:48:51*NimBot joined #nim
18:50:56*jszymanski joined #nim
18:52:32*NimBot joined #nim
18:52:57*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
18:53:47*bendiken joined #nim
18:56:58yglukhovi think you should store func ptr and ctx ptr separately.
18:57:13yglukhovbut your code looks arguably valid
18:57:36yglukhovmaybe that's a bug. why not report it! =)
19:00:13yglukhovoh, maybe you're doing the wrong cast. try adding {.closuse.} in your casts
19:00:42yglukhovyour gist works for me in any case
19:03:55*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
19:04:08*raza quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
19:04:51*bendiken joined #nim
19:14:54*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
19:15:54*bendiken joined #nim
19:20:39*jefus is now known as jefus_
19:25:59*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
19:26:53*bendiken joined #nim
19:37:13*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
19:37:47*victor-rene joined #nim
19:38:10*NimBot joined #nim
19:38:12*bendiken joined #nim
19:47:43*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
19:49:02*bendiken joined #nim
19:52:12*jefus_ is now known as jefus
19:58:14*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
19:59:16*bendiken joined #nim
20:02:10drewsremyglukhov, oh I wasn't on HEAD, my bad, thanks a lot for testing
20:02:32yglukhovanytime ;)
20:09:19*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
20:10:15*bendiken joined #nim
20:20:12*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
20:21:04*bendiken joined #nim
20:21:05*jefus is now known as jefus_
20:21:15*jefus_ is now known as jefus
20:25:10*brson joined #nim
20:30:19*jszymanski quit (Quit: computer sleeps...)
20:31:31*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
20:32:19drewsremyglukhov, oh I'm silly, the last line in my snippet is wrong, I cast "foobar" again, I should cast anyproc tho, if you swap that it stops working again
20:32:30*bendiken joined #nim
20:34:33Araqdrewsrem: what are you doing?
20:41:23yglukhovdrewsrem. Now that's definitely Nim bug, because the C code is invalid. PLease submit it.
20:43:14drewsremAraq, essentially a replacement for unions I have an object type that holds the type/kind and a payload as an untyped byte-array that holds the actualy event object, I then send this around in an event system, the id is supposed to encode an event hierarchy, maybe, anyway, eventually I need to look at the type/payload tuple, cast the payload to some object-type and send it to a callback the object-type specifies, so every
20:43:14drewsremobject should basically has a field where it adds (kind, callback) tuples, for what kind of event it listens to and some closure that gets executed, the signature/prototype/functionhead/whatever of the callback has to be generic tho because the callbacks should already take the event object as a parameter but there's no generics for proc-types for fields? - ... something along those lines
20:43:25drewsremyglukhov, sure
20:44:01drewsremGrammar much, it's late
20:59:50*renesac quit (Ping timeout: 272 seconds)
21:04:06jckwhy does this proc not work? proc bit_length(T: typedesc): int = sizeof(T)*8
21:04:35*renesac joined #nim
21:12:00*NimBot joined #nim
21:21:58*Jehan_ joined #nim
21:23:03Araqjck: dunno but you can just use a template instead
21:23:17*Mat4 joined #nim
21:27:32*NimBot joined #nim
21:28:08*Mat4 quit (Quit: Konversation terminated!)
21:28:38*Mat4 joined #nim
21:28:46Jehan_jck: It works for me (at least on devel)?
21:29:52jckI have this error: Error: cannot evalutate 'sizeof' because type is not defined completely
21:31:49jckThis is what I'm trying to do: http://ix.io/kAW
21:33:29Jehan_jck: The problem seems to be that T is a generic parameter, i.e. unknown.
21:33:58Jehan_So the compiler can't infer the range type.
21:37:36*victor-rene quit (Ping timeout: 246 seconds)
21:38:26*golak quit (Quit: leaving)
21:39:36*boopsies is now known as boopsiesisaway
21:40:17jckGiven that in my case, Reg s would only be declared at compile time, is there any way i could work around this?
21:44:51*Mat4 left #nim ("Konversation terminated!")
21:52:19*Matthias247 quit (Read error: Connection reset by peer)
21:52:32*pregressive quit (Remote host closed the connection)
21:55:29*gokr quit (Quit: Page closed)
21:55:52*gokr_ quit (Remote host closed the connection)
22:00:04*BitPuffin|osx joined #nim
22:03:38*drewsrem quit (Quit: Leaving)
22:04:16*Trustable quit (Remote host closed the connection)
22:14:17*NimBot joined #nim
22:15:31Araqr-ku: are you looking into the coroutine crashes? I suspect your version of setjmp is not yet good enough?
22:16:35*jaco60 quit (Ping timeout: 244 seconds)
22:18:40*bendiken quit (Quit: My Mac has gone to sleep. ZZZzzz…)
22:19:50*bendiken joined #nim
22:22:33*NimBot joined #nim
22:25:45Jehan_jck: You can use a template with the type as a typedesc parameter to generate the procedure.
22:27:54federico3Any way to put objects of different types in the same container and then call a method on each of them?
22:29:54Jehan_jck: Simplified example here: https://gist.github.com/rbehrends/7a69285a917a306c0660
22:30:27Jehan_federico3: If they all inherit from the same type, yes. Just make it a container of the inherited type.
22:31:15federico3Jehan_: that's what I've done. Unfortunately Nim runs the proc in the ancestor type
22:31:16Jehan_If some of the objects aren't like that (i.e. ints or strings), wrap them in an inheritable type or a variant type and make it a container with elements of that type.
22:31:53Jehan_federico3: There's currently a bug in method dispatch where if you have the ancestor method after the others, it'll intercept them.
22:32:03Jehan_The ancestor method must come first in the file.
22:32:28Jehan_I ran in that bug today because I was moving methods around and suddenly my code blew up. :)
22:32:55federico3urgh, huge bug. But anyhow, only the ancestor method is being executed (and it was at the top
22:33:29Jehan_federico3: Huh. That's weird.
22:33:49Jehan_Are you using `ref object` or `object` types?
22:34:09federico3ref object of RootObj
22:34:24Jehan_Hmm. Then that can't be it.
22:34:25federico3for the ancestor - and then ref of <ancestor>
22:34:38Jehan_For what it's worth, I can't seem to reproduce it.
22:34:56federico31 minute
22:35:10gokr1Araq: Did you hear Mike Pall is leaving LuaJIT?
22:35:12*renesac quit (Ping timeout: 250 seconds)
22:35:52*renesac joined #nim
22:36:30Araqgokr1: yep
22:36:47AraqJehan_: it's actually a really nasty bug to fix ...
22:36:55Jehan_Araq: I'm not surprised. :)
22:37:44Jehan_Araq: It's nothing that affects me personally, since (1) I know about it and (2) can avoid it. But it can create some really difficult to discover bugs if you are unaware.
22:38:03AraqI want some .dispatcher pragma annotation
22:38:17Araqso that at least I can report an error
22:38:34Araqplus it's really weird without it and Nim's effect system
22:38:48Jehan_federico3: I have this here (and it works for me): https://gist.github.com/rbehrends/35cf5bf5ffc824a556d8
22:38:58federico3Jehan_: https://paste.debian.net/310165/
22:39:20Araqactually I want to remove 'method' from the language but I know too much code now depends on it ...
22:39:52*brson quit (Quit: leaving)
22:39:59Jehan_federico3: You're using procs, not methods.
22:40:12*brson joined #nim
22:40:27Jehan_Araq: Any particular reason?
22:40:53AraqI never ever use them, when I need dynamic dispatch, I just use closures
22:40:59*wuehlmaus quit (Quit: Lost terminal)
22:41:02Jehan_Well … I use them.
22:41:12Jehan_Closures are just too much memory overhead for small objects.
22:41:50Jehan_And variant types take up as much memory as the largest branch, so they don't always work, either.
22:42:06federico3is there a solution that does not require methods?
22:42:10Araqyou can always put the closures into your own vtable
22:43:37Araqas Demos said, x.vtbl.f(x, y) is not sexy, but overloading of '.' should work for that, or perhaps it needs some other form of sugar
22:43:57Jehan_Araq: Not really. Closures have their environment.
22:44:29*NimBot joined #nim
22:44:31Jehan_I could use procvars, but that means I'd essentially writing a lot of extra machinery.
22:44:49Araqyeah but that's what we have macros for
22:44:57Jehan_For starters, I need a separate call operator like Lua in order to not explicitly pass the self argument again.
22:45:22Jehan_Araq: Well, I don't always want to write my own sublanguage for every little thing.
22:45:51Jehan_If it were a standard library, maybe, though I'm not sure how good compile times would be with macros everywhere.
22:47:42Araqfor me it's just cruft that I need to support and test for 3 different backends
22:48:00*yglukhov quit (Remote host closed the connection)
22:48:12Araqso a macro based solution is far superior ;-)
22:48:22Araqcause then I can simply not care
22:49:30Jehan_Araq: I'd be happy with a simpler OO system, to be honest, but then it also shouldn't be hard to support it in the compiler.
22:49:33*yglukhov joined #nim
22:49:51Jehan_That said, extensible polymorphism is a fairly important feature IMO.
22:53:40Araqthere are lots of problems with the current design
22:53:58Jehan_Araq: I think it's unnecessarily ambitious.
22:54:46Jehan_For example, multiple dispatch in particular is nice in theory, but has only so many practical applications over single dispatch.
22:55:01Araqfor example: for true multiple dispatching, I can use lookup[x.id + y.id*col](x, y)
22:55:28*X67r quit (Quit: leaving)
22:55:41Araqand that's far more efficient than what Nim currently does
22:56:01Araqalso the explicit Type IDs can be very handy for other serialization purposes
22:56:18Jehan_if you know the ids beforehand. I thought we had discussed this earlier and there were problems with DLLs?
22:57:12Araqyeah but a homegrown manual solution can come up with better workarounds for DLLs as well
22:57:45Araqmy point is: the homegrown solution beats the builtin solution.
22:58:10Araqand that's sad.
22:59:25Araqit's also what often happens in C++ frameworks, many roll their own object system, their own RTTI system, etc.
22:59:37Jehan_By homegrown you mean what? Macros?
23:00:15Araqmacros, qt's moc, going back to C
23:01:06Jehan_Hmm. Is abandoning the current approach even an option, given that a fair amount of code is already using it?
23:02:01Araqno, but at least I can make people annotation the .dispatcher so that's clear "this is the guy which needs all the effect annotations"
23:02:10Araq*people annotate
23:02:32Jehan_*nod*
23:04:53federico3ehm, should I switch to methods for my simple use case?
23:05:48Araqfederico3: yes
23:06:04Araqseems to be the solution you're most comfortable with
23:06:35federico3and suggest using methods in https://github.com/nim-lang/Nim/wiki/Nim-for-Python-Programmers ?
23:08:28federico3I mean, as long as they are not going to be deprecated
23:10:45AraqJehan_: different topic: I made the threadpool aware of its limitations, meaning a thread doesn't spawn when all workers are busy, and instead the task is run on the thread that tried to spawn
23:11:25Araqbut for the quicksort example it just stack overflows, because spawn directly calls 'f', f calls spawn and so on
23:11:34Jehan_Araq: Hmm, not sure if it helps. The problem is generally waiting within a task.
23:11:59Araqwell that's the obvious solution that C# also uses iirc
23:12:03Jehan_And there's really no fix for it (other than coroutines, and then you have to explicitly yield when you wait).
23:12:17Jehan_*nod*
23:13:11Araqif I manually use 'spawnX' instead of 'spawn' in the quicksort example, things work
23:13:41Araqtemplate spawnX(f) = if allBusy(): f() else: spawn f()
23:13:47Araqso this works
23:14:04Araqbut as soon as I move the f into the spawn I get an endless recursion
23:14:10*Kingsquee joined #nim
23:14:15Araqhrm difficult to explain
23:14:48Jehan_federico3: Methods are the best solution for a number of circumstances.
23:15:59*irrequietus quit ()
23:16:05federico3oh, there is contains(string, string) but not contains(seq, string)
23:16:26Jehan_Araq: Hmm. Not sure I'm following, but that maybe because I'm tired.
23:17:42*yglukhov quit (Remote host closed the connection)
23:18:05Jehan_Anyhow, I'm off for the night. Take care. :)
23:18:07*Jehan_ quit (Quit: Leaving)
23:28:03*keyle joined #nim
23:28:13Araqfederico3: pretty sure there is
23:33:31*vendethiel quit (Ping timeout: 265 seconds)
23:37:24federico3actually there's contains on a slice or set or openarray
23:38:24federico3I also really like: if <A> in <B>
23:40:36Sornaensiscan I use nim to replace C++
23:40:53federico3Sornaensis: you are authorized to do so.
23:40:59Sornaensisthanks
23:52:45*unclechu quit (Quit: Leaving.)