03:44:51  * saghuljoined
03:49:21  * saghulquit (Ping timeout: 248 seconds)
04:14:25  * vurtzquit (Ping timeout: 256 seconds)
05:35:00  * tumdedumquit (Ping timeout: 256 seconds)
05:46:15  * tumdedumjoined
05:52:46  * tumdedumquit (Ping timeout: 260 seconds)
05:54:21  * tumdedumjoined
06:00:33  * tumdedumquit (Ping timeout: 248 seconds)
06:01:30  * tumdedumjoined
07:10:14  * saghuljoined
10:13:12  * vurtzjoined
10:25:09  * mylesborinsquit (Quit: farewell for now)
10:25:40  * mylesborinsjoined
13:04:34  * Klownerquit (Ping timeout: 264 seconds)
14:36:03  * Klownerjoined
14:53:01  * saghulquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
15:13:00  * saghuljoined
15:27:11  * s4ghuljoined
15:30:09  * saghulquit (Ping timeout: 248 seconds)
16:01:21  * vurtzquit (Ping timeout: 240 seconds)
16:01:46  * s4ghulquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
16:35:14  * vurtzjoined
17:45:08  * Fishrock123joined
17:47:02  <Fishrock123>How do you convert uv_timespec_t (from uv_stat_t->st_mtim) to a double to send back to uv_fs_utimes?
17:47:07  <Fishrock123>Is that what this code does?
17:47:08  <Fishrock123>s->st_atim.tv_sec + (s->st_atim.tv_nsec / 1000000000.0)
17:47:25  <Fishrock123>(where 's' is a 'uv_stat_t')
17:54:21  * sgimenoquit (Ping timeout: 240 seconds)
17:54:40  * sgimenojoined
18:02:32  * Fishrock123quit (Remote host closed the connection)
18:10:01  <refack>For Fishrock123 or posterity, this is what we do in node - RE mtime and uv_stat_t, this is what we do in node ATM - https://github.com/nodejs/node/blob/59ace5752a13136eee7ae07ca173bc2addda2e9f/src/node_internals.h#L349-L361
18:11:02  <refack>https://www.irccloud.com/pastebin/M75pXHUG/
18:12:53  <refack>Where `fields` is an array of doubles that goes out to JS to be used as miliseconds
18:13:50  * Fishrock123joined
18:14:04  <refack>*sorry, array of `(unsigned long)`
18:14:48  <refack>**No sorry again it is doubles
18:15:48  <refack>We first cast the `(##.tv_sec)` to `(unsigned long)` to avoid Y2K38
18:17:59  <refack>Ohh your back Fishrock123. ^^^^^
18:18:09  <refack>*you're
18:29:23  <Fishrock123>yeah now I'm back
18:29:31  <Fishrock123>didn't think I'd dc but heh
18:29:56  <Fishrock123>oh yeah it does look like that code makes a double
18:30:02  <Fishrock123>or at least that's what vscode syas
18:30:46  <refack>So just one note. `uv_timespec_t` is `long[2]`, but since fs times can be assumed to be positive you should cast to `unsigned long` to avoid Y2K38
18:32:28  <Fishrock123>(what?)
18:32:47  <Fishrock123>uv's test-fs.calready does `s->st_atim.tv_sec + (s->st_atim.tv_nsec / 1000000000.0)`
18:32:52  <Fishrock123>is that wrong?
18:32:59  <Fishrock123>test-fs.c
18:33:31  <refack>tv_sec for "03:14:07 UTC on 19 January 2038" will be 2^^31
18:35:16  <refack>so I'd suggest to do `s->(unsigned long)st_atim.tv_sec + (s->st_atim.tv_nsec / 1000000000.0)`
19:23:10  <Fishrock123>hmmm ok
19:41:20  <Fishrock123>refack: ok so... this apparently fails to compile: https://ci.nodejs.org/job/libuv-test-commit-linux/960/nodes=debian9-64/console
19:41:30  <Fishrock123>guess my compile mode is more modern on my macbook
19:42:21  <Fishrock123>I know how to fix the first
19:42:27  <Fishrock123>but what is wrong with the second?
19:42:38  <Fishrock123>oh...
19:42:39  <Fishrock123>nvm
19:42:50  <genio>I always set the standard I'm going to use. -std=c89 -Wall -Wextra or whatnot
19:43:54  <Fishrock123>I'm surprised that the configure didn't automatically do that
19:44:02  <refack>I'm was also playing with adding a Y2K38 test...
19:44:06  * joocain2_quit (Write error: Connection reset by peer)
19:46:26  <refack>There's also a different issue with windows - https://github.com/nodejs/node/blob/28e4e43e513ae90e8a4236dbae7f3442ab8dbb4f/test/parallel/test-fs-utimes.js#L202-L210
19:46:43  <Fishrock123>I don't think I'm running into that
19:47:05  <Fishrock123>I think I'm running into an APFS bug on mac but I'm trying to narrow down the possibilities still
19:47:46  <Fishrock123>speaking of which, refack do you know is all out macos test machines are still 10.12 or something?
19:48:01  <refack>10.10
19:48:34  <Fishrock123>oh, right.
19:48:36  <refack>and some very old XCode/clang
19:48:48  <Fishrock123>well thats actually good for me
19:48:53  <Fishrock123>because this never failed before 10.13
19:49:58  * joocain2joined
19:50:37  <Fishrock123>nope, still not testing this correctly... hmm.
19:52:40  <refack>We have this repertoire https://github.com/nodejs/build/blob/master/ansible/inventory.yml#L138-L148 but only 10.10 are in the metrix
19:53:45  <Fishrock123>why
19:53:47  <Fishrock123>just... why
19:53:54  <refack>time
19:54:19  <refack>build-WG volunteer time
20:14:25  <Fishrock123>:(
20:38:08  * vurtzquit (Ping timeout: 265 seconds)
21:07:15  * Fishrock123quit (Remote host closed the connection)
22:01:12  * Fishrock123joined
22:14:31  * concept10joined
22:57:11  * joocain2quit (Remote host closed the connection)
22:58:29  * joocain2joined
23:10:34  * joocain2quit (Remote host closed the connection)
23:10:48  * Fishrock123quit (Quit: Leaving...)
23:11:06  * joocain2joined
23:14:49  * concept10quit (Quit: Textual IRC Client: www.textualapp.com)