<< 17-02-2019 >>
00:08:46*zachk joined #nim
00:09:57*noeontheend quit (Ping timeout: 255 seconds)
00:12:07*zachk quit (Changing host)
00:12:07*zachk joined #nim
01:23:17*zachk quit (Quit: Leaving)
01:54:18FromDiscord<exelotl> should boolVal work for ident nodes?
01:56:26FromDiscord<exelotl> it just tripped me up big time because I was traversing an untyped tree and assumed that an ident node with "true" inside it should be, well, true
01:56:47*rnrwashere quit (Remote host closed the connection)
01:57:23*rnrwashere joined #nim
02:01:52*rnrwashere quit (Ping timeout: 258 seconds)
02:15:45*rnrwashere joined #nim
02:22:35*rnrwashere quit (Remote host closed the connection)
02:27:51*ng0_ joined #nim
02:29:43*ng0 quit (Ping timeout: 256 seconds)
02:30:44*ng0_ quit (Client Quit)
02:42:29*Tyresc quit (Quit: WeeChat 2.4-dev)
03:01:59*rnrwashere joined #nim
03:02:30*banc quit (Quit: Bye)
03:10:45*rnrwashere quit (Ping timeout: 250 seconds)
03:23:40*banc joined #nim
03:30:45*zyklon quit (Ping timeout: 252 seconds)
03:38:32*nsf joined #nim
03:48:15*zyklon joined #nim
04:13:31FromDiscord<drewp> i welcome code review on https://github.com/drewp/sse-broadcast-benchmark/commit/23b69fba4efb2c1c618017b9ec3bc29500b3da62 especially lines 60,61,64,70
04:16:30leorizewhy are you using ref seq?
04:18:43*Ven`` joined #nim
04:32:12*rnrwashere joined #nim
04:37:48*Ven`` quit (Ping timeout: 272 seconds)
05:03:29FromDiscord<drewp> because it's the first thing that worked. What should i use there?
05:05:50*drazan left #nim ("The Lounge - https://thelounge.chat")
05:05:50FromDiscord<drewp> plain seq didn't work because i modify it in async procs, and even a 'var clients' param leads to invalid capture error
05:06:21leorizehmm, I'm pretty sure plain seq[] worked...
05:06:38leorizeclosure should be able to capture that
05:06:45leorizeI'll test that later
05:06:49FromDiscord<drewp> oh, and it was a global before. Now that i wrapped it into main() maybe it's ok
05:08:37*rnrwashere quit (Remote host closed the connection)
05:10:05FromDiscord<drewp> proc handleSSE(req: Request, clients: var seq[AsyncSocket]) {.async.} =
05:10:06FromDiscord<drewp> await req.client.send("HTTP/1.1 200 OK\c\L")
05:10:06FromDiscord<drewp> clients.add(req.client)
05:10:06FromDiscord<drewp>
05:10:06FromDiscord<drewp> ^ illegal capture 'clients' of type <var seq[AsyncSocket]>
05:10:17FromDiscord<drewp> so no, i don't have a solution other than ref seq
05:27:49*vlad1777d joined #nim
05:52:16Araqptr seq would also work
05:58:31*narimiran joined #nim
06:05:25*rnrwashere joined #nim
06:06:19*rnrwashere quit (Remote host closed the connection)
06:15:06*Summertime quit (Quit: Sunsetting.)
06:15:39*Summertime joined #nim
06:36:28*lwit31 joined #nim
06:41:23*lwit31 quit (Ping timeout: 256 seconds)
06:47:28*nsf quit (Quit: WeeChat 2.3)
06:51:04*NimBot joined #nim
07:01:25*NimBot joined #nim
07:10:49FromGitter<Varriount> Isn't that even worse, from a safety standpoint?
07:17:38*NimBot joined #nim
07:19:21*rockcavera joined #nim
07:34:54*aguspiza joined #nim
07:35:46*Trustable joined #nim
08:00:00*gmpreussner quit (Quit: kthxbye)
08:03:58*dddddd quit (Remote host closed the connection)
08:04:50*gmpreussner joined #nim
08:05:10*NimBot joined #nim
08:27:28*Xe quit (Ping timeout: 250 seconds)
08:32:08*kapil____ joined #nim
08:34:10*krux02 joined #nim
08:44:52*Trustable quit (Remote host closed the connection)
09:29:06*Vladar joined #nim
09:54:29*narimiran quit (Ping timeout: 255 seconds)
10:14:41Araqhey LemonBoy, are you here?
10:29:13*sealmove quit (Quit: WeeChat 2.3)
10:36:48*stefanos82 joined #nim
10:42:09*vlad1777d quit (Ping timeout: 246 seconds)
11:08:28*lritter joined #nim
11:52:48*nsf joined #nim
12:04:41Zevv"Mesh: Compacting Memory Management for C/C++ Applications" / https://arxiv.org/abs/1902.04738.
12:05:53Araqnot too bad
12:06:28FromGitter<mratsim> now, let's reduce the consumption of Electron ..
12:07:25Araqbut maybe fragmentation is an API problem, 'malloc' has too little information so that permanent objects are mixed with shortlived ones
12:07:40Zevvbasic algorithm is to merge pages which have non-overlapping data
12:08:59Zevvnot sure how portable this is, mmap()ing to move virtual pages around sounds unix-y
12:09:10AraqI also read plenty of papers that claimed fragmentation is a paper tiger and a non-issue for most systems
12:10:17Araqmost issues in Nim came from the "idiom", "here, have a multi MB string and let the GC eventually collect it"
12:10:19ZevvOf coure, for most systems it is no issue. I can run a single webkit based browser on my machine without caring about the memory. But at one of my current projects we want to run a 1000 independent instances on a single machine, and that is were every hole matters
12:14:30Araqhow can you move memory btw? some stack slot could point to it
12:15:58ZevvMemory is not moved, two physical pages are merged into one, which is then mapped to both the old and the new address
12:16:01Zevvfrom what I understand
12:16:09Zevvmmap allows such tricks
12:17:57Araqhmmmm
12:23:30FromGitter<Clyybber> I think Windows' posix subsystem has _mmap
12:24:35Zevvwith all true mmap semantics? because mmap is quite a versatile beast
12:25:21FromGitter<Clyybber> Hmm, I don't know. I can't find any documentation for it.
12:29:14*NimBot joined #nim
12:38:38*Senketsu quit (Quit: WeeChat 2.3)
12:39:40*Senketsu joined #nim
13:00:43*dddddd joined #nim
13:08:54*vlad1777d joined #nim
13:09:40FromGitter<mratsim> windows has VirtualAlloc
13:12:02*krux02 quit (Remote host closed the connection)
13:25:37*vlad1777d quit (Ping timeout: 244 seconds)
13:26:23*quipa joined #nim
13:26:32*ng0 joined #nim
13:33:59*Xe joined #nim
13:52:00*vlad1777d joined #nim
13:52:56*Trustable joined #nim
14:18:46*narimiran joined #nim
14:57:18FromGitter<mratsim> The compiler is calling quit() upon accessing a field of my object variant? is there a way to get a stack trace?
14:57:27FromGitter<mratsim> (https://files.gitter.im/nim-lang/Nim/naVL/DeepinScreenshot_select-area_20190217155509.png)
14:57:29FromGitter<mratsim> (in a macro)
15:45:17*Ven`` joined #nim
15:45:20*abm joined #nim
15:48:37*Tyresc joined #nim
15:51:02*quipa quit (Read error: Connection reset by peer)
15:52:46*rnrwashere joined #nim
15:54:43narimirani need a sanity check: does `nimprof` work for you on devel?
15:55:23narimiranjust put `import nimprof` in some file and compile it with `--profiler:on` option
15:59:55narimirannevermind, even i cannot reproduce it anymore
16:14:39*Ven`` quit (Ping timeout: 268 seconds)
16:20:11*arecaceae quit (Remote host closed the connection)
16:20:29*arecaceae joined #nim
16:26:51*rnrwashere quit (Remote host closed the connection)
16:29:53*rnrwashere joined #nim
16:30:01*rnrwashere quit (Remote host closed the connection)
16:32:47*rnrwashere joined #nim
16:37:16*rnrwashere quit (Ping timeout: 264 seconds)
16:42:59*fanta7531 joined #nim
16:57:02FromGitter<mratsim> So I have some PR for Complex in Arraymancer - re https://github.com/nim-lang/Nim/pull/9590 Complex32 and Complex64 are wrong for me, and not consistent with 32 and 64 being the type size in the lang
17:18:41*rnrwashere joined #nim
17:31:39*rnrwashere quit (Remote host closed the connection)
17:34:16*abm quit (Read error: Connection reset by peer)
17:36:13*rnrwashere joined #nim
17:39:36*Ven`` joined #nim
17:40:04*rnrwashere quit (Remote host closed the connection)
17:41:01*rnrwashere joined #nim
17:44:50FromGitter<deansher> I am struggling to get rid of "Error: cannot lift assignment operator to 'case' object". It happens in the ``let`` statement near the end of this code: ⏎ ⏎ ```code paste, see link``` ⏎ ⏎ Suggestions? [https://gitter.im/nim-lang/Nim?at=5c699d924003460b2d684383]
17:49:06*Vladar quit (Remote host closed the connection)
17:50:24*kapil____ quit (Quit: Connection closed for inactivity)
17:52:40FromDiscord<PusiteGA> how to hnadle post requests in jester?
17:53:17*vlad1777d quit (Ping timeout: 255 seconds)
18:01:40*narimiran quit (Remote host closed the connection)
18:04:16FromGitter<deansher> That was on 0.19.4. Works on 0.19.9.
18:21:36FromDiscord<treeform> Is there a way to disable seq range checking only in one spot? --boundChecks:off just for one function?
18:23:27*Trustable quit (Remote host closed the connection)
18:33:40*Perkol joined #nim
18:48:51*lritter quit (Quit: Leaving)
18:51:19*abm joined #nim
19:08:29*rnrwashere quit (Remote host closed the connection)
19:18:52*dewf joined #nim
19:36:28*rnrwashere joined #nim
19:39:46*nsf quit (Quit: WeeChat 2.3)
19:42:49*gangstacat quit (Quit: Ĝis!)
19:44:08*rnrwashere quit (Remote host closed the connection)
19:46:48*rnrwashere joined #nim
19:46:58xace_ry
19:47:17xace_my bad*
19:48:06*gangstacat joined #nim
19:48:09*Perkol quit (Quit: Leaving)
19:49:06*Ven`` quit (Ping timeout: 250 seconds)
20:02:55*Vladar joined #nim
20:04:57*druonysus quit (Quit: druonysus)
20:35:02*aguspiza quit (Ping timeout: 250 seconds)
20:48:04leorizetreeform: {.push boundchecks: off.} proc... {.pop.}
20:55:43*fanta7531 quit (Quit: fanta7531)
21:05:51*rnrwashere quit (Remote host closed the connection)
21:09:44*Snircle joined #nim
21:21:54*Vladar quit (Remote host closed the connection)
21:27:49*dewf quit (Quit: Leaving)
21:42:40*sealmove joined #nim
21:59:28*Ven`` joined #nim
22:09:37FromDiscord<jos> question
22:09:48FromDiscord<jos> i want to pass the address of a variable from nim to c
22:09:50FromDiscord<jos> so i can modify it in c
22:10:04FromDiscord<jos> used to work with "var int" as the type, and then passing it normally in nim
22:10:11FromDiscord<jos> var int in the importc signature, that is
22:10:46FromDiscord<jos> is there a better way to do that? i just tried ptr int in the importc signature and addr(var) when calling but that didn't work either
22:12:37FromDiscord<jos> nevermind it's working fine i'm low iq
23:09:03*Ven`` quit (Ping timeout: 246 seconds)
23:11:26*abm quit (Remote host closed the connection)
23:11:51*abm joined #nim
23:16:24*rnrwashere joined #nim
23:16:43*abm quit (Ping timeout: 245 seconds)
23:58:41*rnrwashere quit (Remote host closed the connection)
23:59:08*sealmove quit (Quit: WeeChat 2.3)