00:00:00  * ircretaryquit (Remote host closed the connection)
00:00:08  * ircretaryjoined
00:06:52  * indexzerojoined
00:09:04  * amartensquit (Quit: Leaving.)
00:12:10  * dshaw_quit (Quit: Leaving.)
00:20:16  * hzjoined
00:20:39  * dshaw_joined
00:29:11  * kenperkinsjoined
00:32:28  * mikealquit (Quit: Leaving.)
00:33:55  * hzquit (Read error: Connection reset by peer)
00:34:25  * felixgequit (Quit: felixge)
00:34:44  * hzjoined
00:37:59  * TooTallNatequit (Quit: Computer has gone to sleep.)
00:43:25  * dapquit (Quit: Leaving.)
00:52:08  * bnoordhuisjoined
00:52:13  * bradleymeckjoined
00:56:09  * dapjoined
00:56:19  * dapquit (Client Quit)
00:56:52  * bnoordhuisquit (Ping timeout: 264 seconds)
01:00:09  * dapjoined
01:02:26  * indexzeroquit (Quit: indexzero)
01:02:43  * mikealjoined
01:05:04  * Damn3dquit (Ping timeout: 264 seconds)
01:06:30  * hzquit (Ping timeout: 264 seconds)
01:10:38  * Damn3djoined
01:12:20  * kenperkinsquit (Quit: Computer has gone to sleep.)
01:16:52  * dapquit (Quit: Leaving.)
01:18:11  * abraxasjoined
01:22:13  * abraxasquit (Ping timeout: 240 seconds)
01:24:57  * bradleymeckquit (Quit: bradleymeck)
01:28:44  * bradleymeckjoined
01:33:44  * mikealquit (Quit: Leaving.)
01:37:28  * abraxasjoined
01:48:58  * pachetjoined
01:55:44  * indexzerojoined
01:59:20  * indexzeroquit (Client Quit)
02:04:55  * pachetquit (Quit: leaving)
02:14:26  * mikealjoined
02:17:43  * dshaw_quit (Quit: Leaving.)
02:32:24  * indexzerojoined
03:03:27  * brsonquit (Ping timeout: 260 seconds)
03:05:10  * brsonjoined
03:05:11  * mikealquit (Quit: Leaving.)
03:08:08  * wolfeidaujoined
03:09:00  * dshaw_joined
03:13:29  * mikealjoined
03:14:30  * bradleymeck_joined
03:14:37  * bradleymeck_quit (Client Quit)
03:14:45  * bradleymeckquit (Ping timeout: 248 seconds)
03:16:21  * julianduquequit (Ping timeout: 248 seconds)
03:17:09  * st_lukequit (Remote host closed the connection)
03:18:44  * kenperkinsjoined
03:24:25  * wolfeidauquit (Ping timeout: 246 seconds)
03:31:11  * mikealquit (Quit: Leaving.)
03:57:46  * kazuponjoined
03:59:30  * st_lukejoined
04:02:54  * brson_joined
04:03:40  * wolfeidaujoined
04:06:16  * brsonquit (Read error: Operation timed out)
04:13:08  * TooTallNatejoined
04:13:41  * TooTallNatequit (Client Quit)
04:13:53  * c4miloquit (Remote host closed the connection)
04:13:56  * TooTallNatejoined
04:14:30  * mikealjoined
04:18:15  * brson_quit (Quit: leaving)
04:18:30  * brsonjoined
04:18:53  * mikealquit (Ping timeout: 256 seconds)
04:24:24  * hueniverse1joined
04:24:42  * hueniversequit (Ping timeout: 264 seconds)
04:26:57  * kellabytequit (Read error: Connection reset by peer)
04:37:07  * kenperkinsquit (Ping timeout: 264 seconds)
04:37:12  * kenperkins_joined
04:37:56  * mikealjoined
04:46:21  * stagasjoined
04:48:06  * TooTallNatequit (Quit: Computer has gone to sleep.)
05:01:04  * stagasquit (Remote host closed the connection)
05:01:36  * stagasjoined
05:04:47  * wolfeidauquit (Remote host closed the connection)
05:09:10  * kenperkins_quit (Quit: Computer has gone to sleep.)
05:12:25  * indexzeroquit (Quit: indexzero)
05:15:22  * indexzerojoined
05:16:00  * amartensjoined
05:17:38  * indexzeroquit (Client Quit)
05:28:01  * mikealquit (Quit: Leaving.)
05:29:55  * Damn3dquit (Ping timeout: 264 seconds)
05:31:47  * Damn3djoined
05:43:55  * csaohjoined
05:49:28  * amartensquit (Quit: Leaving.)
05:49:30  * csaohquit (Quit: csaoh)
05:58:11  * mikealjoined
06:00:11  * wolfeidaujoined
06:06:49  * mikealquit (Ping timeout: 246 seconds)
06:12:41  * st_lukequit (Remote host closed the connection)
06:21:48  * dominictarrjoined
06:25:13  * felixgejoined
06:34:38  * rendarjoined
06:34:44  * mikealjoined
06:37:21  * paddybyersjoined
06:38:49  * AvianFlujoined
06:39:55  * dshaw_quit (Quit: Leaving.)
06:48:14  * wolfeidauquit (Remote host closed the connection)
07:02:15  * dominictarrquit (Quit: dominictarr)
07:18:44  * paddybyersquit (Ping timeout: 240 seconds)
07:24:50  * dshaw_joined
07:30:31  * csaohjoined
07:32:53  * defunctzombiechanged nick to defunctzombie_zz
07:35:24  * rendarquit (Ping timeout: 276 seconds)
07:39:14  * mordy__quit (Ping timeout: 264 seconds)
07:41:24  * dshaw_quit (Quit: Leaving.)
07:42:29  * brsonquit (Quit: leaving)
07:44:38  * rendarjoined
07:51:45  * `3E|Zzzchanged nick to `3rdEden
08:08:40  * wolfeidaujoined
08:32:10  * bajtosjoined
08:38:43  * wolfeidauquit (Ping timeout: 246 seconds)
08:38:54  * hzjoined
08:48:42  * defunctzombie_zzchanged nick to defunctzombie
09:11:10  * skebcioquit (Read error: Connection reset by peer)
09:11:17  * skebciojoined
09:25:56  * dominictarrjoined
09:48:07  * bnoordhuisjoined
09:52:02  * kazuponquit (Remote host closed the connection)
10:08:11  * dominictarrquit (Read error: Connection reset by peer)
10:08:37  * dominictarrjoined
10:23:32  <indutny>hoya
10:24:13  <indutny>bnoordhuis: ping, https://github.com/joyent/node/pull/6009
10:38:30  * Somebody_joined
10:40:52  * defunctzombiechanged nick to defunctzombie_zz
10:47:09  * kazuponjoined
10:47:40  * mordy__joined
10:48:17  * csaohquit (Quit: csaoh)
10:55:41  <bnoordhuis>indutny: sorry, i'm afk today
11:06:23  * csaohjoined
11:08:20  * dominictarrquit (Quit: dominictarr)
11:09:32  * piscisaureus_joined
11:16:57  * bajtosquit (Quit: bajtos)
11:27:48  * Somebody_quit (Remote host closed the connection)
11:28:11  * bnoordhuisquit (Ping timeout: 240 seconds)
11:29:44  * dominictarrjoined
11:58:15  * piscisaureus_quit (Ping timeout: 260 seconds)
12:07:41  * julianduquejoined
12:10:39  * jmar777joined
12:11:54  * julianduquequit (Client Quit)
12:21:04  * pachetjoined
12:25:37  * kazuponquit (Remote host closed the connection)
12:29:58  * kenperkinsjoined
12:33:46  * bnoordhuisjoined
12:36:37  * abraxasquit (Remote host closed the connection)
12:38:52  * bnoordhuisquit (Ping timeout: 264 seconds)
12:44:48  * hzquit (Ping timeout: 276 seconds)
12:51:17  * kenperkinsquit (Quit: Computer has gone to sleep.)
13:05:29  * kazuponjoined
13:07:13  * c4milojoined
13:37:58  * bajtosjoined
13:41:43  * kazuponquit (Remote host closed the connection)
13:44:13  * c4miloquit (Remote host closed the connection)
13:55:58  * julianduquejoined
14:11:34  * kenperkinsjoined
14:17:24  * kenperkins_joined
14:19:40  * kenperkinsquit (Ping timeout: 264 seconds)
14:31:55  * c4milojoined
14:36:50  * abraxasjoined
14:37:36  * mikealquit (Quit: Leaving.)
14:41:09  * abraxasquit (Ping timeout: 248 seconds)
14:52:10  * kazuponjoined
14:56:40  * dapjoined
14:57:41  * kazuponquit (Ping timeout: 248 seconds)
14:58:23  * kazuponjoined
14:59:16  * kellabytejoined
14:59:40  * pfox__quit (Changing host)
14:59:41  * pfox__joined
15:00:31  * pfox__quit (Quit: leaving)
15:01:06  * amartensjoined
15:02:24  * bradleymeckjoined
15:03:49  * paulfryzeljoined
15:04:10  * pfox__joined
15:04:19  * pfox__part
15:07:11  * paulfryz_joined
15:09:50  * paulfryzelquit (Ping timeout: 240 seconds)
15:14:35  * AvianFlu_joined
15:28:11  * kenperkins_quit (Quit: Computer has gone to sleep.)
15:43:04  * kenperkinsjoined
15:43:04  * `3rdEdenchanged nick to `3E|CHEFCOOK
15:45:32  <isaacs>good morning
15:46:42  * dapquit (Quit: Leaving.)
15:54:54  * TooTallNatejoined
15:55:16  * TooTallNatequit (Client Quit)
15:58:04  * bradleymeckquit (Remote host closed the connection)
15:58:43  * bradleymeckjoined
16:02:29  * julianduquequit (Quit: leaving)
16:03:39  * hzjoined
16:18:13  * bajtosquit (Quit: bajtos)
16:18:58  * TooTallNatejoined
16:22:11  * kenperkinsquit (Quit: Computer has gone to sleep.)
16:22:40  * hzquit (Ping timeout: 240 seconds)
16:24:20  * dapjoined
16:39:46  * brsonjoined
16:43:27  * c4miloquit (Remote host closed the connection)
16:48:46  * TooTallNatequit (Quit: Computer has gone to sleep.)
16:50:10  * Somebody_joined
16:53:13  * spolujoined
16:56:04  * AvianFlu_quit (Remote host closed the connection)
16:56:11  * toothrotquit (Read error: Connection reset by peer)
16:56:27  * dominictarr_joined
16:58:21  * toothrjoined
16:59:36  * dominictarrquit (Ping timeout: 276 seconds)
16:59:37  * dominictarr_changed nick to dominictarr
16:59:43  * indexzerojoined
16:59:49  * csaohquit (Quit: csaoh)
17:03:14  * kazuponquit (Remote host closed the connection)
17:06:05  * indexzeroquit (Quit: indexzero)
17:07:19  * defunctzombie_zzchanged nick to defunctzombie
17:12:33  * TooTallNatejoined
17:23:33  <MI6>libuv-v0.10: #105 UNSTABLE windows (5/188) smartos (2/187) http://jenkins.nodejs.org/job/libuv-v0.10/105/
17:24:59  <MI6>libuv-node-integration: #139 FAILURE http://jenkins.nodejs.org/job/libuv-node-integration/139/
17:25:02  * spoluquit (Ping timeout: 240 seconds)
17:26:26  * AvianFlu_joined
17:27:18  * `3E|CHEFCOOKchanged nick to `3E
17:28:52  * hzjoined
17:29:12  * paddybyersjoined
17:33:20  * mikealjoined
17:34:39  * brsonquit (Quit: leaving)
17:34:44  * AvianFlu_quit (Ping timeout: 260 seconds)
17:34:55  * brsonjoined
17:35:01  * hueniverse1quit (Ping timeout: 248 seconds)
17:35:54  * hzquit (Disconnected by services)
17:35:57  * hzjoined
17:36:30  <MI6>libuv-v0.10-gyp: #69 UNSTABLE smartos-ia32 (2/187) windows-x64 (3/188) smartos-x64 (2/187) windows-ia32 (4/188) http://jenkins.nodejs.org/job/libuv-v0.10-gyp/69/
17:36:58  * kazuponjoined
17:37:28  * hueniversejoined
17:37:50  * st_lukejoined
17:38:09  <MI6>libuv-master-gyp: #91 UNSTABLE smartos-ia32 (2/191) smartos-x64 (2/191) windows-x64 (4/192) windows-ia32 (3/192) http://jenkins.nodejs.org/job/libuv-master-gyp/91/
17:40:39  * AvianFlu_joined
17:46:01  * AvianFluquit (Remote host closed the connection)
17:46:30  * kazuponquit (Remote host closed the connection)
17:46:35  * AvianFlujoined
17:46:58  * paddybyersquit (Ping timeout: 256 seconds)
17:51:34  * AvianFluquit (Disconnected by services)
17:51:38  * AvianFlu_changed nick to AvianFlu
17:52:04  * AvianFlu_joined
17:54:21  * bnoordhuisjoined
17:57:11  * piscisaureus_joined
17:57:27  * c4milojoined
18:02:23  * indexzerojoined
18:02:35  * julianduquejoined
18:02:50  * Somebody_quit (Remote host closed the connection)
18:07:51  * hzquit (Ping timeout: 260 seconds)
18:19:02  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
18:32:41  * dominictarrquit (Quit: dominictarr)
18:37:02  * c4miloquit (Remote host closed the connection)
18:37:37  * mikealquit (Quit: Leaving.)
18:39:21  * c4milojoined
18:40:56  * mikealjoined
18:42:31  * c4milo_joined
18:42:36  * c4miloquit (Remote host closed the connection)
18:43:10  * mikealquit (Client Quit)
18:57:11  * kazuponjoined
19:02:28  <bnoordhuis>http://jenkins.nodejs.org/job/libuv-node-integration/139/ <- jenkins trouble again or ?
19:03:06  * kazuponquit (Ping timeout: 264 seconds)
19:09:33  * kenperkinsjoined
19:10:58  * indexzeroquit (Quit: indexzero)
19:12:33  <MI6>joyent/libuv: Ben Noordhuis master * 6837e48 : unix: fix missing return value warning in stream.c - http://git.io/ywdvVg
19:15:30  <brson>i turned on the new libuv-based rust task scheduler yesterday. thanks for the help
19:17:03  <MI6>libuv-master: #153 FAILURE windows (3/192) http://jenkins.nodejs.org/job/libuv-master/153/
19:18:20  <MI6>libuv-master-gyp: #92 UNSTABLE smartos-ia32 (2/191) smartos-x64 (2/191) windows-x64 (4/192) windows-ia32 (4/192) http://jenkins.nodejs.org/job/libuv-master-gyp/92/
19:21:10  <bnoordhuis>configure.ac:19: option `serial-tests' not recognized <- wut?
19:21:22  <bnoordhuis>i added a check for that, didn't i?
19:21:54  <bnoordhuis>tjfontaine: ping
19:24:56  * c4milo_quit (Remote host closed the connection)
19:27:13  <bnoordhuis>... i'm getting more than a little annoyed by the massive amount of duplication between node_crypto.cc and tls_wrap.cc
19:27:29  <indutny>bnoordhuis: hm...
19:27:35  <trevnorris>bnoordhuis: not sure this will even makes sense, but any way to assign an anonymous struct to a memory pointer?
19:27:37  <indutny>bnoordhuis: better check something more interesting
19:27:41  <indutny>bnoordhuis: tcp_wrap.cc and pipe_wrap.cc
19:27:42  <indutny>;)
19:28:04  * jepickettjoined
19:28:12  <bnoordhuis>i mean, we're not talking conceptual duplication here - there's a lot of stuff in tls_wrap.cc that's a verbatim copy of node_crypto.cc
19:28:26  * c4milojoined
19:29:00  <bnoordhuis>trevnorris: well... struct { int x; } *p = (void*) 0xDEADBABE;
19:29:10  <bnoordhuis>but it's kind of pointless if you can't dereference it later on
19:29:29  <bnoordhuis>also, that's valid c but not valid c++
19:29:44  <trevnorris>poop, ok.
19:30:23  <trevnorris>basically i'm want to grab the external memory from an object and use it as a struct. sort of what you helped me do w/ the info box thing
19:34:40  * c4miloquit (Read error: Connection reset by peer)
19:39:11  <trevnorris>wow, can't believe I got that to work so quickly.
19:41:27  <trevnorris>bnoordhuis: when you have a moment kind sir: https://github.com/joyent/node/pull/6019
19:42:39  <trevnorris>isaacs: here's how i'm using the smalloc addition ^
19:43:00  <isaacs>trevnorris, bnoordhuis, tjfontaine: review, plz? https://github.com/joyent/node/pull/5916
19:43:04  * c4milojoined
19:43:41  <trevnorris>isaacs: reviewing, done any benchmarking yet?
19:44:03  * pachetquit (Quit: leaving)
19:46:00  * kenperkinsquit (Ping timeout: 276 seconds)
19:46:19  <isaacs>trevnorris: yes, quite a bit
19:46:34  * kenperkinsjoined
19:46:35  <isaacs>trevnorris: you lose speed initially, then gain it all back with the streams commit in that same pull req
19:47:26  <isaacs>trevnorris: basically, now that streams3 uses on('data') in pipe() again, you can short circuit the whole push(chunk)->buffer.push(chunk)->emit('readable')->read()->buffer.shift()
19:47:44  <isaacs>trevnorris: and instead, if we're flowing, you can just do push(chunk)->emit('data',chunk)
19:47:56  <trevnorris>hah, quite the sequence. ok cool. see that now.
19:49:04  <trevnorris>isaacs: seems straight forward enough. lgtm.
19:49:52  * stagas_joined
19:50:23  * stagasquit (Ping timeout: 246 seconds)
19:50:27  * stagas_changed nick to stagas
19:50:30  <tjfontaine>bnoordhuis: pong
19:51:48  <bnoordhuis>tjfontaine: what was the address of that jenkins smartos build slave again?
19:52:04  <tjfontaine>that one is actually smartos.nodejs.org
19:52:25  <trevnorris>othiym23: #6019 is another step to get that public api working
19:53:39  <bnoordhuis>https://github.com/bnoordhuis/node/compare/joyent:master...new-strings-api <- it admittedly doesn't make the code purdier
19:57:21  <bnoordhuis>tjfontaine: what franken-autoconf is on that machine that recognizes AM_SILENT_RULES but not serial-tests?
19:57:30  <bnoordhuis>franken-autoconf/franken-automake
19:58:00  <tjfontaine>pkg_info | grep automake
19:58:01  <tjfontaine>automake-1.11.1nb1 GNU Standards-compliant Makefile generator
19:58:34  <bnoordhuis>hm, looks like it's straight from pkgsrc. does joyent patch the pkgsrc tree?
19:58:57  <tjfontaine>ocassionally, but not normally
19:59:12  <bnoordhuis>okay
19:59:23  <bnoordhuis>well, bleh - you anger me, automake
19:59:49  <tjfontaine>https://github.com/joyent/pkgsrc/commits/trunk/devel/automake I don't see any related commits about it
20:00:28  * kenperkinsquit (Quit: Computer has gone to sleep.)
20:02:19  <trevnorris>bnoordhuis: lgtm, though have to admit visually not loving the jump between the macro and function call.
20:02:36  <MI6>joyent/node: isaacs master * c0e7035 : stream: Short-circuit buffer pushes when flowing (+1 more commits) - http://git.io/rTfO-A
20:04:01  <trevnorris>out for the doctor, someone please review #6019 :)
20:04:03  * trevnorris&
20:04:04  <LOUDBOT>I DIDN'T KNOW THAT SHADOWCAT DOESN'T COUNT AS CORPORATE SUPPORT
20:07:25  * EhevuTovjoined
20:11:25  <bnoordhuis>trevnorris: yeah. my initial attempt had a ONE_BYTE_STRING() macro that called OneByteString() instead
20:11:41  <bnoordhuis>but i figured having a macro that just calls a function is kind of pointless
20:12:03  <MI6>nodejs-master: #391 UNSTABLE smartos-ia32 (2/621) linux-x64 (1/621) smartos-x64 (9/621) http://jenkins.nodejs.org/job/nodejs-master/391/
20:19:03  * c4miloquit (Remote host closed the connection)
20:19:29  * c4milojoined
20:21:03  * rendarquit
20:23:19  * c4miloquit (Remote host closed the connection)
20:23:40  * c4milojoined
20:24:31  <MI6>nodejs-master-windows: #192 UNSTABLE windows-x64 (17/621) windows-ia32 (16/621) http://jenkins.nodejs.org/job/nodejs-master-windows/192/
20:24:46  * mikealjoined
20:27:35  * indexzerojoined
20:30:41  <MI6>joyent/libuv: Ben Noordhuis master * e4c1483 : build: serial-tests was added in automake v1.12 - http://git.io/gcypWw
20:32:33  <bnoordhuis>^ the amount of pain that fscking option has caused me
20:32:53  <tjfontaine>heh
20:32:53  <isaacs>bnoordhuis: new-string-api lgtm, if all tests pass.
20:32:55  <bnoordhuis>and if the automake had just taken a second to think throughs and ADD A GODDAMN MACRO FOR IT
20:32:57  <isaacs>bnoordhuis: nothign controversial in it.
20:33:23  <bnoordhuis>isaacs: no, indeed. it's just ugly
20:33:43  <bnoordhuis>i'm going to go over it one more time, just to check if there are places where utf-8 is more appropriate than latin-1
20:33:44  <isaacs>bnoordhuis: yeah.
20:33:55  <isaacs>bnoordhuis: tedious.
20:34:02  <isaacs>bnoordhuis: thanks for doing this cleanup
20:34:21  <bnoordhuis>quite so. and yeah, someone's got to do it :)
20:34:37  <isaacs>bnoordhuis: on the subject of ugliness, it also looks like there's some scope.Close(return_value) stuff still sticking around. I thought there was some more verbose way of doing that now?
20:35:09  <bnoordhuis>isaacs: you mean args.GetReturnValue().Set()? that's only when returning from the binding layer
20:35:20  <isaacs>args.GetReturnValue().Set(return_value)
20:35:21  <isaacs>right
20:35:42  <isaacs>oh, right, that's only if you were called from JS, and thus have an args obj
20:35:49  <bnoordhuis>everywhere else where you have a handlescope and return js values to the caller, you still need to call scope.Close(val)
20:35:57  <isaacs>right
20:36:02  <MI6>libuv-master: #154 FAILURE windows (3/192) http://jenkins.nodejs.org/job/libuv-master/154/
20:36:16  <isaacs>i'm clearly low on calories. be back in an hour or two
20:36:43  <bnoordhuis>tjfontaine: that one's for you -> dtrace: failed to link script src/unix/uv-dtrace.d: failed to open .libs/libuv_la-fs-poll.o: No such file or directory
20:37:09  <MI6>libuv-master-gyp: #93 UNSTABLE smartos-ia32 (2/191) smartos-x64 (2/191) windows-x64 (3/192) linux-ia32 (1/191) windows-ia32 (3/192) http://jenkins.nodejs.org/job/libuv-master-gyp/93/
20:37:23  <bnoordhuis>ah well, maybe i'll fix that one as well. i'm still logged in
20:38:34  <tjfontaine>bnoordhuis: I can get it if you don't want to
20:39:04  <bnoordhuis>nah, it's probably something trivial like updating the find command
20:40:11  <tjfontaine>nod
20:41:36  * jmar777quit (Remote host closed the connection)
20:46:59  * Benvie_joined
20:49:48  <bnoordhuis>hm, no dice
20:49:49  <bnoordhuis>Undefined first referenced symbol in file
20:49:52  <bnoordhuis>__dtrace_uv___tick__stop ./.libs/libuv.so
20:49:54  <bnoordhuis>__dtrace_uv___tick__start ./.libs/libuv.so
20:50:13  <bnoordhuis>that's after this:
20:50:13  <bnoordhuis>- `grep '^pic_object' $$(find ${top_builddir} -name "*.lo") | cut -f 2 -d\'`
20:50:17  <bnoordhuis>+ `find ${top_builddir} -name "*.lo") | sed 's/\.lo$$/.o/'`
20:50:27  <bnoordhuis>i know, i know - not a great fix
20:52:57  * wolfeidaujoined
20:59:17  * mikealquit (Quit: Leaving.)
21:00:22  * amartensquit (Quit: Leaving.)
21:06:14  * paulfryz_quit (Remote host closed the connection)
21:06:48  * paulfryzeljoined
21:08:49  * bradleymeckquit (Quit: bradleymeck)
21:10:13  * jmar777joined
21:11:16  <bnoordhuis>=== release test-next-tick-loops-quick ===
21:11:16  <bnoordhuis>Path: pummel/test-next-tick-loops-quick
21:11:17  <bnoordhuis>FATAL ERROR: JS Allocation failed - process out of memory
21:11:28  <bnoordhuis>^ wuuh?
21:11:48  * paulfryzelquit (Ping timeout: 276 seconds)
21:16:59  * AvianFluquit (Remote host closed the connection)
21:17:20  * bradleymeckjoined
21:27:03  * AvianFlujoined
21:30:52  <bnoordhuis>isaacs: you around?
21:33:39  * trevnorrisfg
21:33:54  <trevnorris>bnoordhuis: so i'm just missing the reason for needing to -1 on the data length
21:34:02  <trevnorris>but then again i'm looking that hard
21:34:21  <bnoordhuis>*that* hard? or *not* that hard?
21:35:01  <bnoordhuis>it's because v8::String::NewFromWhatever() takes an optional length arg that defaults to -1
21:35:45  <trevnorris>haha, not only correcting my code but also my english ;)
21:35:48  <trevnorris>ah ok
21:39:32  <trevnorris>bnoordhuis: and nice job btw. those massive api updates aren't exactly on the fun side.
21:39:54  * bnoordhuisquit (Read error: No route to host)
21:40:09  <othiym23>trevnorris: thanks, saw that, but haven't had time to review it yet
21:40:29  <othiym23>trying to get a new build of the New Relic out the door
21:40:39  <othiym23>the delta from the last beta is massive
21:42:25  * paddybyersjoined
21:43:13  * bnoordhuisjoined
21:43:27  <bnoordhuis>back. connection dropped
21:46:49  <trevnorris>othiym23: cool. main point is that now we can track the exact domain depth from c++, which allows to get around some of the perf hit from before.
21:46:54  * jepickettquit (Ping timeout: 264 seconds)
21:53:34  <bnoordhuis>so, simple/test-chdir fails after the NewFromOneByte() changes
21:53:49  <bnoordhuis>and it's because of that long-standing bug where we interpret file paths as utf-8
21:53:55  * paddybyersquit (Ping timeout: 264 seconds)
21:54:20  <bnoordhuis>not sure what to do about it. i could revert the changes in code that deals with file paths
21:54:36  <bnoordhuis>but that's arguably reintroducing a bug...
21:54:39  <bnoordhuis>opinions?
21:55:27  * c4miloquit (Remote host closed the connection)
21:55:42  <bnoordhuis>btw, the reason that test fails is that it creates a directory called 'weird \uc3a4\uc3ab\uc3af characters \u00e1\u00e2\u00e3'
21:55:43  <trevnorris>bnoordhuis: well, you can have utf8 paths, right?
21:55:53  * c4milojoined
21:55:54  <bnoordhuis>not on unices, paths are just byte strings
21:56:09  <bnoordhuis>windows is different but i don't know it well enough to comment
21:58:29  <bnoordhuis>trevnorris: btw, when i said 'paths are just byte strings', i mean that a unix kernel attaches no significance to the bytes in the path
21:58:52  * c4milo_joined
21:58:52  <bnoordhuis>it could be utf-8 but the kernel doesn't care. as long as it ends in a nul byte, it's all good
21:59:00  * c4miloquit (Read error: No route to host)
21:59:03  <trevnorris>ah, ok
21:59:06  <trevnorris>thanks
21:59:26  * c4milo_quit (Remote host closed the connection)
22:00:11  <bnoordhuis>this interpreting file paths as utf-8 is a bug people first complained about in v0.3 days, i think
22:00:22  <trevnorris>so then shouldn't strings be resolved as 'binary' ?
22:00:48  <bnoordhuis>yeah. i don't know why that never got fixed tbh
22:01:30  <trevnorris>well, if we're using one byte strings on both ends then there shouldn't be any noticable difference to users
22:01:34  <trevnorris>so i'd say go ahead and fix it
22:01:51  * c4milojoined
22:01:56  <bnoordhuis>well, the thing is if you pass a utf-8 string from js
22:02:09  <bnoordhuis>wait, let me rephrase that
22:02:14  <bnoordhuis>if you take a look at the test
22:02:26  <bnoordhuis>it first creates a string with unicode characters
22:02:34  * c4miloquit (Remote host closed the connection)
22:02:54  <bnoordhuis>then it mkdirs a directory with that string
22:03:06  <bnoordhuis>then it chdirs into that directory (so far so good)
22:03:20  <bnoordhuis>then it calls process.cwd() and compares the result to the original string
22:03:34  <bnoordhuis>that's where it breaks down because the return value from process.cwd() is a one-byte string
22:03:44  * c4milojoined
22:04:05  * kenperkinsjoined
22:04:18  <bnoordhuis>so what's one character in the utf-8 string, is two characters in the one-byte string
22:04:32  <trevnorris>just a thought experiment, Buffer(process.cwd(), 'binary').toString('utf'); would make that pass, right?
22:04:39  <trevnorris>*utf8
22:04:56  <bnoordhuis>yeah, converting the one-byte string to utf-8 works, that's not the issue
22:05:32  <bnoordhuis>it's that i'm kind of scared of breaking people's code
22:05:40  <bnoordhuis>even if their code was wrong to start with :-/
22:06:17  <trevnorris>well, from the command line if i mkdir Ȣ; cd Ȣ then that's how the folder name appears, in my terminal anyways.
22:06:36  <trevnorris>so i'll assume it's the terminals decision to interpret the string as utf8?
22:07:18  <bnoordhuis>yep
22:08:04  * mikealjoined
22:08:26  <trevnorris>well, bummer. so it appears there's ambiguity in implementation, i.e. pwd on my box gives me back utf8
22:08:39  * c4miloquit (Ping timeout: 260 seconds)
22:09:51  <bnoordhuis>trevnorris: i doubt that pwd does that, it's just your terminal interpreting the raw bytes as utf-8 :)
22:09:56  <trevnorris>bnoordhuis: also, what's interpreting process.env.PWD?
22:10:12  <bnoordhuis>in node, you mean?
22:10:15  <trevnorris>ah, ok.
22:10:17  <trevnorris>yeah.
22:10:27  <bnoordhuis>well, it used to be String::New()
22:10:48  <bnoordhuis>which is indistinguishable from String::NewFromOneByte() when it's just ascii
22:11:05  <bnoordhuis>probably the common case, that
22:13:54  * groundwaterquit (Ping timeout: 264 seconds)
22:14:06  * c4milojoined
22:15:21  * hzjoined
22:15:41  * groundwaterjoined
22:21:37  * bradleymeckquit (Quit: bradleymeck)
22:21:55  * Benvie_quit
22:26:54  * hzquit (Ping timeout: 264 seconds)
22:27:11  * AvianFlewjoined
22:27:58  * jmar777_joined
22:30:17  * hzjoined
22:31:06  * AvianFluquit (Ping timeout: 276 seconds)
22:31:26  * AvianFlewquit (Ping timeout: 240 seconds)
22:31:26  * jmar777quit (Ping timeout: 240 seconds)
22:38:35  * kenperkinsquit (Quit: Computer has gone to sleep.)
22:38:57  * sblomjoined
22:41:28  <sblom>tjfontaine: A few unit test (well, build) related questions.
22:41:48  <tjfontaine>sup sup
22:41:48  * c4miloquit (Remote host closed the connection)
22:41:56  <sblom>tjfontaine: 1) I swear I used to log into Jenkins using GitHub, but it doesn't seem to do that now.
22:42:27  <sblom>2) I'm trying to log in so that I might answer the following question myself, but I'll ask instead. What OS version does the windows build agent run?
22:42:41  <tjfontaine>sblom: you did, and I've disabled that because I'm trying to narrow down just which plugin I'm using that utlimately grinds jenkins frontend to a halt (I'm guessing TAP)
22:42:58  <tjfontaine>sblom: server 2k8r2
22:43:04  <sblom>Perfect--thanks.
22:43:12  <tjfontaine>sblom: I can create an account for you if you'd like
22:43:28  <sblom>Interestingly, I have an account, but I think it's passwordless.
22:43:34  <sblom>Don't worry about it.
22:43:40  <sblom>OS Version was most of what I needed for now.
22:43:51  <tjfontaine>nod, I can also give you access to the VM if/when that's necessary
22:43:54  <sblom>I'm trying to repro the few consistent build server failures that I can't repro locally.
22:43:56  <sblom>Okay--awesome.
22:43:57  <sblom>Thanks.
22:44:19  <sblom>jpickett and I are partying on unit tests today and tomorrow--this is exciting.
22:44:43  <tjfontaine>excellent, making progress then?
22:45:08  <sblom>tjfontaine: some
22:45:23  <sblom>tjfontaine: One fix even. A few investigations are in flight.
22:45:58  <tjfontaine>sblom: excellent, the test-os stuff bert said he wanted to actually adhere to the proper windows ordering, I odn't know if you spent anytime looking at that one
22:47:44  * Somebody_joined
22:48:39  <sblom>tjfontaine: No, didn't look at that one yet.
22:49:01  <tjfontaine>sblom: ok, it's not a very high impact one, the problem is mostly in the test I think
23:01:32  <brson>on mac (at least) the event loop creates another worker thread. what is it for? does it exist on all platforms?
23:01:49  <tjfontaine>brson: that's used for some of the tty checking code, iirc
23:02:00  <tjfontaine>brson: afaik it's a mac specific work around
23:02:53  <brson>tjfontaine: thanks. do you know if it can be disabled?
23:04:15  <tjfontaine>brson: looking at it, it doesn't seem so, https://github.com/joyent/libuv/blob/master/src/unix/tty.c#L40-L44 which calls https://github.com/joyent/libuv/blob/master/src/unix/stream.c#L284-L371
23:07:42  * AvianFlu_quit (Remote host closed the connection)
23:20:28  * stagasquit (Read error: Connection reset by peer)
23:27:42  * wolfeidauquit (Ping timeout: 264 seconds)
23:29:04  * AvianFlujoined
23:31:33  <bnoordhuis>brson: what tjfontaine said, it works around a deficiency in os x's kqueue implementation
23:31:48  <bnoordhuis>the thread itself spends most of its life sleeping in syscalls
23:32:02  * Somebody_quit (Remote host closed the connection)
23:33:15  * hzquit
23:33:18  <bnoordhuis>hm, that reminds me... we should probably limit its stack
23:33:38  <tjfontaine>ya probably doesn't need all that much
23:34:17  <brson>bnoordhuis: if we don't use libuv's tty code is it feasible to disable?
23:34:28  <brson>though certainly reducing the stack would help
23:35:11  <tjfontaine>right, so I guess the question is why you need/want to disable it :)
23:35:48  <bnoordhuis>brson: the thread is created lazily. if you don't have file descriptors that kqueue can't cope with, libuv won't spawn the thread
23:36:15  <brson>bnoordhuis: ah, was the laziness perhaps added in the last 6 months or so? we're pretty behind
23:37:50  <bnoordhuis>brson: i believe we initially added that about a year ago
23:37:59  * bnoordhuisdigs through git log
23:38:07  <tjfontaine>was it that long ago? man.
23:39:01  <bnoordhuis>2012-08-16 - i wasn't far off :)
23:39:16  <isaacs>bnoordhuis: pong
23:39:30  <tjfontaine>only a week away
23:39:35  <bnoordhuis>isaacs: did you see the discussion with trevor?
23:39:41  <brson>hm, we're using libuv from March and seeing an extra thread per loop
23:40:03  <trevnorris>about file paths?
23:40:20  <bnoordhuis>brson: that's right. it creates a thread per kqueue-incompatible fd currently
23:40:38  <bnoordhuis>brson: our thinking was that you normally will have only one or two of those in a program
23:40:54  <isaacs>bnoordhuis: re pummel/test-next-tick-loops-quick?
23:40:55  <bnoordhuis>hence not much point in trying to coordinate across event loops
23:41:09  <bnoordhuis>isaacs: no, simple/test-chdir. and file paths
23:42:02  <bnoordhuis>pummel/test-next-tick-loops-quick running out of memory is kind of worrisome btw. but that's something for another day
23:42:17  <isaacs>bnoordhuis: what's wrong wiht simple/test-chdir.?
23:42:23  <isaacs>bnoordhuis: seems to work for me
23:42:24  <bnoordhuis>brson: btw, if that's an issue for rust, we can probably address it. file an issue with rationale and i'll look into it
23:42:38  <bnoordhuis>isaacs: not with the NewFromOneByte() patch :)
23:42:44  <isaacs>ohhh...
23:42:46  <isaacs>ok
23:42:49  <brson>bnoordhuis: sorry, not trying to be dense, but this thread is created during platform_init, before creating any handles, here: https://github.com/brson/libuv/blob/master/src/unix/darwin.c#L73
23:42:57  <isaacs>bnoordhuis: 'weird \uc3a4\uc3ab\uc3af characters \u00e1\u00e2\u00e3'
23:43:06  <isaacs>seems like that could be a problem, indeed
23:43:07  <bnoordhuis>brson: oh, that's probably fedor's fsevents thread
23:43:22  <isaacs>bnoordhuis: i didn't test the NewFromOneByte patch
23:43:26  <isaacs>bnoordhuis: just read it
23:43:40  <tjfontaine>does it fail because of utf8 paths?
23:43:55  <bnoordhuis>isaacs: yeah. the test uses that string as a directory name, mkdirs and chdirs into, then expects process.cwd() == dir
23:43:55  <tjfontaine>right
23:44:28  <isaacs>hm. right
23:44:35  <isaacs>ah ohhhh.... yeah, i remember this
23:44:47  <bnoordhuis>we have a bug report from the 0.3 or 0.4 days about that somewhere
23:45:16  <isaacs>right
23:45:40  <isaacs>bnoordhuis: it seems like you and trevnorris are more or less agreed that chdir/mkdir should be using binary encoding rather than utf8. i think that's sensible.
23:45:54  <isaacs>actually, lemme check something first...
23:46:29  <bnoordhuis>brson: yes, it's used to drive the CFRunLoop that in turn powers fsevents
23:47:04  <bnoordhuis>kind of lame that it gets created unconditionally, i wonder why i didn't notice that during review...
23:47:30  * c4milojoined
23:48:23  <brson>lazy initialization of that thread would be sweet :)
23:49:34  <bnoordhuis>yeah. i'll file an issue
23:50:06  <isaacs>bnoordhuis: so... yeah. we should probably leave it as-is.
23:51:50  <bnoordhuis>isaacs: 'leave as-is' meaning?
23:52:02  <isaacs>bnoordhuis: keep as utf8
23:52:14  <isaacs>https://gist.github.com/isaacs/6189920
23:53:02  <bnoordhuis>right
23:53:52  <bnoordhuis>the flipside is that it makes it impossible to open some files or directories
23:54:09  <isaacs>if the bytes form invalid utf8, you mean?
23:54:12  <bnoordhuis>yep
23:54:25  <isaacs>we could allow you to pass in a Buffer as the path
23:54:42  * jmar777_quit (Read error: Connection reset by peer)
23:54:42  * jmar777joined
23:54:43  <isaacs>or, you know, just throw because fuck you and your unreadable paths.
23:54:54  <isaacs>seems like no one is really complaining about that right now
23:55:10  <bnoordhuis>yeah. adding support for buffers was my last comment on that old bug report :)
23:55:18  <isaacs>but i think that if Node doesn't support the same paths as Finder/Explorer/etc. then it's rather confusing and weird.
23:55:35  <bnoordhuis>it does
23:55:41  <bnoordhuis>wait, let's make sure we're on the same page
23:55:55  <isaacs>Keep behavior the same as v0.10
23:56:31  <bnoordhuis>right
23:56:48  <bnoordhuis>do finder and what have you use utf-8 internally?
23:58:32  <isaacs>not sure
23:58:41  <isaacs>but the current behavior is very non-wtf-inducing
23:58:48  <isaacs>so whatever v0.10 does, is probably right.
23:59:10  <bnoordhuis>well, that's somewhat debatable
23:59:13  * dapquit (Quit: Leaving.)
23:59:22  <bnoordhuis>but it probably works okay for most people