<< 18-09-2013 >>

00:00:35fowloh theres no doc for them
00:01:01Araqnot yet, nimbuild on x86 crashes now
00:01:09Araqso docgen doesn't work
00:01:19fowlo
00:01:23fowlwhat did you do to it!
00:01:29Araqdunno yet
00:01:34fowl:p
00:01:36Araqand I know how to hunt the bug
00:01:50fowlwhen is your talk, tonight?
00:01:55Araqand the result will be a --gc:debug mode that will kick ass :P
00:02:11Araqtomorrow at 11:40
00:02:21*brson quit (Remote host closed the connection)
00:03:26*brson joined #nimrod
00:04:03NimBotAraq/Nimrod master 460be96 Araq [+0 ±1 -0]: should fix newly introduced bugs wrt TR macros
00:04:27fowlAraq, which hotel are you at
00:04:31fowla nice one downtown hopefully
00:04:34*dyu joined #nimrod
00:04:42Araqunion station
00:04:51fowlah ok
00:04:56Araqit's crazy
00:05:07Araqyou can get lost there
00:05:29fowlinside it you mean
00:05:29fowl?
00:05:33Araqyeah
00:06:24fowlnah its not that big in there
00:09:33Araqthe room numbers have 4 digits ...
00:10:19fowlwell i didnt know there was a hotel inside union station i thought it was just a mall
00:11:07Araqit's hard to say where one ends and the other starts :P
00:22:00fowlhow do you like STL so far
00:24:16Araqeverything is so big here ... vast amounts of space if you get what I mean
00:24:41Araqyou think "that's right over there" and it's a walk of half an hour ...
00:28:57fowlno i dont quite follow lol
00:29:45Araqmeh come to europe and you'll see the difference :P
00:30:03fowli can judge distance fairly well lol
00:30:10*Yeri joined #nimrod
00:30:14fowlmaybe its because things are in miles here and not kilometers
00:31:27Araqno that's not what I mean
00:32:13Araqfor instance, the sidewalks are wide enough that cars could drive on it
00:33:00fowloh thats just downtown for all the foot traffic
00:35:02Araqyeah I suppose
00:35:16Araqwell so far I like it, it's quite like I imagined :P
00:35:58Araqbut it's hostile to people who prefer to walk ;-)
00:36:16fowltrue
00:36:35fowli always had a bike or rollerblades
00:36:53fowli just started skateboarding in cali
00:46:34Araqping zahary
00:48:26*Yeri left #nimrod (#nimrod)
00:52:38Araqargh fowl can you remember why .global variables are initialized *after* the other top level statements?
00:56:30Araqthe use case had to do with your crazy component stuff iirc
00:58:29NimBotAraq/Nimrod master d16ff3f Araq [+0 ±3 -0]: fixes #576
01:38:03*Associat0r quit (Quit: Associat0r)
01:40:18*fowl quit (Ping timeout: 264 seconds)
01:55:40*DAddYE quit (Remote host closed the connection)
01:56:28*q66 quit (Quit: Leaving)
02:01:00*fowl joined #nimrod
02:19:02*shodan45 joined #nimrod
02:42:27fowlAraq, um
02:44:16fowlAraq, they werent working right otherwise
02:44:34Araqnow that's helpful
02:45:11Araqbbl
02:45:44fowliirc, looking at entitty.nim, i think numMessages was being 0 outside the module but from inside entitty.nim it was right
02:46:23fowllike a.nim imports entitty, it was reporting numMessages and numComponents as 0 after they had been declared
02:46:49fowlthey work now as non-globals (but top-level, exported, so really they are global)
02:47:20fowlgotta go. gl talking tomorrow and enjoy STL
02:47:47Araqthanks
02:49:31fowland seriously
02:49:41fowleast stl is no good
02:49:49fowlcya
02:49:50*fowl quit (Quit: Leaving)
02:56:41*DAddYE joined #nimrod
03:03:06*DAddYE quit (Ping timeout: 264 seconds)
03:05:03*DAddYE joined #nimrod
03:11:51*DAddYE quit (Remote host closed the connection)
03:13:31*Arcanum_za joined #nimrod
03:14:33*Trixar_za quit (*.net *.split)
03:14:34*Zor quit (*.net *.split)
03:15:23*DAddYE_ joined #nimrod
03:15:59*Zor joined #nimrod
04:16:36*OrionPK quit (Quit: Leaving)
04:18:42*DAddYE_ quit (Read error: Connection reset by peer)
04:18:46*DAddYE joined #nimrod
04:20:45*DAddYE quit (Remote host closed the connection)
06:03:08*brson quit (Quit: leaving)
06:52:30*ltbarcly joined #nimrod
07:42:19*Amrykid quit (*.net *.split)
07:43:12*Amrykid joined #nimrod
07:46:07*Associat0r joined #nimrod
07:46:07*Associat0r quit (Changing host)
07:46:07*Associat0r joined #nimrod
08:00:48*rndbit quit (Excess Flood)
08:01:50*rndbit joined #nimrod
08:11:41*capisce quit (Ping timeout: 245 seconds)
08:13:02*capisce joined #nimrod
09:35:38*zahary_ joined #nimrod
09:36:07*ltbarcly quit (Ping timeout: 260 seconds)
09:42:41*zahary_ quit (Read error: Connection reset by peer)
09:42:59*zahary_ joined #nimrod
10:22:20*q66 joined #nimrod
11:03:25*gs joined #nimrod
11:03:49*gs is now known as Guest38704
11:05:11Guest38704How would I go about creating an array whose size is only known at runtime but doesn't change after that? It's kinda silly, but I
11:05:17Guest38704I'm stumped on that right now
11:07:07zahary_do you need the memory to be allocated on the stack (alloca style)? I can just use a sequence otherwise
11:16:08Guest38704nope, no need for it to be on the stack
11:17:06Guest38704Does seq have any guarantee on only taking up as much memory as needed?
11:18:04Guest38704(given that it's generally geared for sequences of variable length)
11:26:10*faassen joined #nimrod
11:27:23zahary_it doesn't provide strong guarantees. it's backed by a heuristic allocator just like dynamic arrays in C/C++
11:27:43zahary_the current allocator rounds allocation sizes to the nearest multiple of 8
11:28:48zahary_most allocators use a progressive scheme where the "wasted" memory is proportional to the allocation size.
11:30:40*shafire joined #nimrod
11:30:42shafirehi
11:31:48*Arcanum_za is now known as Trixar_za
11:37:41*shafire left #nimrod (#nimrod)
11:37:47*shafire joined #nimrod
11:38:06shafireHow was the talk
11:38:17shafirehi dyu
11:38:28dyuhey shafire
11:38:32dyuthe talk is tomorrow
11:40:54profmakxtalk?
11:52:57Guest38704zahary_: ok, that's what I assumed - so how would I actually allocate an array of some specific size at runtime? "let atoms = new(array[123, int])" segfaults, so that obviously isn't the way to go :-D
12:03:05dyuprofmakx: strangeloop conf
12:14:10AraqGuest38704: let atoms = newSeq[int](123)
12:16:16Guest38704Araq: Yeah, I realize I can do that with sequences, but I was specifically looking for a way of creating an array (or openarray?), as seq has some memory overhead I can't control
12:17:58Araqiirc it has close to 0 overhead if you give it a specific size at creation
12:19:02Araqbut you can also do:
12:20:16Araqvar x: ref array[FakeSize, int]; unsafeNew(x, realSize*sizeof(int))
12:20:34Araqbut beware there is a codegen bug with 'ref array'
12:24:40Guest38704Ah ok, and the memory allocator for seq's reasonably unlikely to change?
12:24:54Guest38704As for the second option: seems kinda hacky / codegen bug?
12:25:13Araqyeah it's definitely hacky
12:25:48Araqwell the memory allocator is unlikely to change unless I'm wrong and it doesn't do what I told you ;-)
12:27:20Guest38704That's reasonable enough for whatever contraptions I'm going to come up with in nimrod anytime soon :-)
12:28:30Guest38704On an unrelated note, has anyone done something like a prelude module already? (i.e. expanded upon the functional helpers in sequtils)
12:30:16Araqthere is lib/prelude.nim
12:30:33Araqadd sequtils to it and do a pull request
12:31:39AraqI gotta go, bye
12:32:22Guest38704thanks, bye!
12:51:13*Amrykid quit (Changing host)
12:51:13*Amrykid joined #nimrod
13:13:45*Guest38704 quit (Ping timeout: 245 seconds)
13:44:50*ltbarcly joined #nimrod
14:03:14*dyu quit (Ping timeout: 264 seconds)
14:06:02*dyu joined #nimrod
14:11:21*faassen left #nimrod (#nimrod)
14:14:00*ltbarcly quit (Quit: Computer has gone to sleep.)
14:52:54dom96Good luck Araq!!!
15:39:32*shafire quit (Quit: Page closed)
15:54:50*MFlamer quit (Quit: Page closed)
16:25:35*MFlamer joined #nimrod
16:28:54*brson joined #nimrod
16:51:43*DAddYE joined #nimrod
16:51:46*EXetoC joined #nimrod
16:57:27*Associat0r quit (Quit: Associat0r)
16:57:59MFlameranyone at the conference with Araq?
17:00:02*Guest38704 joined #nimrod
17:05:00*shodan45 quit (Quit: Konversation terminated!)
17:05:18*shodan45 joined #nimrod
17:05:18dom96nope :(
17:05:24dom96I think he's doing his talk right now
17:10:16MFlamerI wonder how soon he can post slides or a recording. Looking forward to it. I know some of you have been working with him for years on Nimrod. It's great for your work to get some exposure.
17:10:57dom96indeed. I really wish I could be there.
17:13:25Guest38704what talk?
17:23:15dom96emerging languages talk
17:34:48EXetoCit should be live
17:34:55dom96lol
17:34:55EXetoCcome on, it's 2013
17:34:58dom96true
18:18:26Guest38704heh
18:19:39Guest38704hey, how come tuple unpacking only works when the var/let is in a single line?
18:41:20EXetoCno idea
18:56:12EXetoCbeep
19:05:51*dyu quit (Quit: Leaving)
19:27:32*Guest38704 quit (Ping timeout: 260 seconds)
19:36:58dom96http://www.infoq.com/conferences/strangeloop2013/
19:37:16dom96Access is limited to people attending...
19:37:22*dom96 sighs
20:32:23reactormonkdom96, I'm sure Araq will ask for a recording
20:34:31*ltbarcly joined #nimrod
20:37:18dom96I think he will have access to the videos at the link I posted
20:48:38*ltbarcly quit (Ping timeout: 240 seconds)
20:52:03*io2 joined #nimrod
20:57:47MFlamerThat would be cool if he shared his password
20:58:05MFlameror, downloaded the recording for us
21:59:22*OrionPK joined #nimrod
22:06:30*Guest38704 joined #nimrod
22:33:24*Yeri joined #nimrod
23:07:10*q66 quit (Quit: Leaving)
23:17:28*Guest38704 quit (Ping timeout: 240 seconds)
23:30:36*Associat0r joined #nimrod
23:30:36*Associat0r quit (Changing host)
23:30:36*Associat0r joined #nimrod
23:53:47*io2 quit ()
23:57:48*ltbarcly joined #nimrod