00:00:01  * ircretaryquit (Remote host closed the connection)
00:00:07  * ircretaryjoined
00:03:08  <indutny>ok, CI is green https://github.com/joyent/node/pull/5257
00:04:43  <indutny>have you guys seen this https://github.com/joyent/node/pulse/monthly ?
00:04:48  <indutny>github pulse
00:06:01  <trevnorris>indutny: ah darn. that's what I get for using sed.
00:07:35  <indutny>yeah
00:09:16  <trevnorris>indutny: so the second "lint" commit should be in that PR?
00:09:40  <indutny>well, I'm not sure if I bother about it
00:09:47  <trevnorris>ok.
00:10:07  <trevnorris>alright. fixed and removed.
00:10:13  * dominictarrquit (Quit: dominictarr)
00:10:18  <trevnorris>that pulse page is cool.
00:10:28  <indutny>haha
00:10:31  <indutny>good for our egoo
00:10:33  <indutny>s/egoo/ego/
00:11:43  <trevnorris>lol, yeah. though that issues/pr list is ridiculously long.
00:13:38  <indutny>that's how it is
00:14:05  <indutny>in reality :)
00:21:40  * trevnorrisquit (Quit: Leaving)
00:56:33  <kellabyte>C & git noob question, if I'm including libuv as part of another repo, should that go in /include or should I create like /lib
00:56:47  <kellabyte>I was going to put it in as a git submodule
00:57:06  <tjfontaine>node has a convention of them going into deps/
00:57:53  <kellabyte>ah cool. what typically goes into an include/ for a C project?
00:57:53  <mmalecki>kellabyte: I tend to use deps/ with git submodules
00:58:06  <tjfontaine>kellabyte: headers :)
00:58:08  <mmalecki>kellabyte: and I put my library's header files in include/
00:58:21  <kellabyte>ah ok gotcha
00:58:28  <kellabyte>thanks :)
00:59:26  <indutny>kellabyte: usually you do not put anything external in your include/
00:59:35  <indutny>just add -Ideps/uv/include to gcc args
00:59:40  <kellabyte>I noticed the libuv gyp files explicitly has each .c file, can you just do a directory? I probably won't have as complicated as structure as libuv
00:59:51  <indutny>nope, you can't
01:00:02  <kellabyte>damn
01:00:10  <indutny>how many files do you have?
01:00:43  <kellabyte>not many yet but I'm lazy and I hate having to do a bunch of stuff when I'm coding in an IDE, I just wanna code not manage crap
01:00:45  <kellabyte>lol
01:01:13  <mmalecki>someone has yet to come up with a perfect build system
01:01:37  <kellabyte>I hate friction, I just wanna code, add new files with ease, and code :P
01:01:54  * inolen1quit (Quit: Leaving.)
01:08:22  * AvianFluquit (Remote host closed the connection)
01:08:51  * kevinswiberquit (Remote host closed the connection)
01:13:35  * Benvie_quit
01:18:29  * c4milojoined
01:20:58  * defunctzombiechanged nick to defunctzombie_zz
01:21:26  <isaacs>kellabyte: that's why i love js :)
01:27:31  <indutny>isaacs: writev
01:27:42  <isaacs>indutny: so. i looked it over briefly.
01:28:05  <isaacs>indutny: stream.bulk(function() { ... }) is kinda weird.
01:28:31  <isaacs>indutny: (still focused on API, not really reviewing code too deeply)
01:28:35  <indutny>sure
01:28:39  <indutny>why is it weird?
01:28:41  <indutny>its explicit
01:28:42  <indutny>at least
01:28:58  <isaacs>i guess
01:29:22  <isaacs>oh, i gotta run.
01:29:26  <isaacs>i'll read more.
01:29:28  <isaacs>bbiab
01:32:41  <indutny>ok
01:43:37  * abraxasjoined
01:44:56  * inolenjoined
01:47:43  * `3Ejoined
01:52:40  * `3Equit (Ping timeout: 272 seconds)
01:55:01  * kazuponjoined
01:59:36  * pooyaquit (Quit: pooya)
02:00:31  * kazuponquit (Remote host closed the connection)
02:00:42  * kazuponjoined
02:10:22  * wavdedjoined
02:11:20  <isaacs>indutny: so, here's what's a bit strange about this...
02:11:33  <isaacs>indutny: stream.bulk(function() { ... call write() a bunch of times ... })
02:11:55  <isaacs>indutny: it's odd because you might want to do like, 10 writes over the span of a few ms, and THEN send them all out.
02:12:03  <isaacs>indutny: what about .cork() and .uncork()?
02:12:18  <isaacs>indutny: similar to the API that bnoordhuis was proposing to add to libuv
02:12:38  * kazuponquit (Remote host closed the connection)
02:22:56  * TooTallNatequit (Quit: ["Textual IRC Client: www.textualapp.com"])
02:25:26  <indutny>oook
02:25:30  <indutny>this might work as well
02:27:39  <indutny>btw, this https://github.com/joyent/node/pull/5335
02:27:53  <indutny>it seems that it comes with performance loss
02:28:00  <indutny>I'm working on measuring it properly
02:28:11  <indutny>because on osx numbers are varying drastically
02:30:36  <isaacs>indutny: yeah, that'll happen
02:30:47  <isaacs>indutny: i usually do NODE_BENCH_RUNS=10 or so when runnning the compare script
02:30:52  <indutny>yep
02:30:53  <isaacs>indutny: helps somewhat
02:30:54  <indutny>I do it
02:30:55  <indutny>in ubuntu vm
02:31:00  <indutny>s/do/doing/
02:31:07  <indutny>just needs some time to finish
02:31:10  <isaacs>indutny: this is more correct, behavior-wise, though.
02:31:14  <indutny>but tests are passing
02:31:16  <isaacs>at least, seems like it should be.
02:31:22  <indutny>yeah
02:31:33  <isaacs>so WS gets a stream that isn't already in flowing mode, which is a bit weird.
02:32:06  * sblomquit (Ping timeout: 245 seconds)
02:32:11  <indutny>I wonder what affects performance most
02:32:19  <indutny>I don't believe that its just buffer.slice()
02:32:26  <indutny>should be something more serious
02:33:43  <isaacs>eah
02:34:00  <isaacs>i've gotta run
02:34:18  <indutny>sure
02:34:19  <indutny>see ya
02:36:25  * indexzerojoined
02:37:57  * kazuponjoined
02:49:25  * kazuponquit (Remote host closed the connection)
02:53:05  * wavdedquit (Quit: Hasta la pasta)
02:59:15  * kazuponjoined
03:00:21  * piscisaureus_quit (Ping timeout: 256 seconds)
03:13:42  * c4miloquit (Remote host closed the connection)
03:16:24  * kazuponquit (Remote host closed the connection)
03:16:58  * brsonquit (Quit: leaving)
03:17:40  * kazuponjoined
03:24:57  * piscisaureus_joined
03:44:32  * piscisaureus_quit (Read error: Operation timed out)
03:47:46  * defunctzombie_zzchanged nick to defunctzombie
03:48:20  * `3Ejoined
03:48:22  * piscisaureus_joined
03:53:00  * `3Equit (Ping timeout: 272 seconds)
04:07:47  * brsonjoined
04:24:12  * piscisaureus_quit (Ping timeout: 260 seconds)
04:42:34  * kazuponquit (Remote host closed the connection)
04:48:31  * pooyajoined
04:51:48  * kazuponjoined
05:06:20  * defunctzombiechanged nick to defunctzombie_zz
05:18:48  * eris0xffquit (Ping timeout: 245 seconds)
05:19:02  * loladirojoined
05:28:50  * sblomjoined
05:29:59  * benoitcquit (Excess Flood)
05:33:11  * sblomquit (Ping timeout: 252 seconds)
05:38:34  * benoitcjoined
05:57:26  * pooyaquit (Quit: pooya)
06:33:25  * brsonquit (Ping timeout: 256 seconds)
06:33:41  * brsonjoined
06:35:39  * wolfeidauquit (Remote host closed the connection)
06:36:48  * dpemmonsquit (Ping timeout: 256 seconds)
06:36:59  * hij1nxquit (Ping timeout: 252 seconds)
06:41:03  * rphillipsquit (Ping timeout: 256 seconds)
06:41:33  * dpemmonsjoined
06:41:47  * indexzeroquit (Read error: Connection reset by peer)
06:45:56  * rphillipsjoined
06:46:46  * hij1nxjoined
06:49:14  * `3rdEdenjoined
07:18:06  * rendarjoined
07:18:25  <saghul>oh, is this github pulse thing new? https://github.com/joyent/libuv/pulse
07:21:54  * stagasjoined
07:25:08  * brsonquit (Quit: leaving)
07:27:02  * stagasquit (Ping timeout: 252 seconds)
07:38:17  * perezdjoined
07:39:48  * stagasjoined
08:10:27  * dsantiagoquit (Ping timeout: 245 seconds)
08:11:09  * loladiroquit (Quit: loladiro)
08:12:08  * dsantiagojoined
08:15:48  * loladirojoined
08:15:59  * loladiroquit (Client Quit)
08:21:36  * kazuponquit (Remote host closed the connection)
08:26:36  * hzjoined
08:28:25  * kazuponjoined
09:25:34  * wolfeidaujoined
09:58:55  <MI6>joyent/node: Sean Silva v0.10 * 63466e5 : doc: document value of `this` inside listeners Fixes #5326. - http://git.io/vzu9SQ
10:05:55  * `3rdEdenchanged nick to `3E|BRB
10:16:59  <MI6>nodejs-v0.10: #149 UNSTABLE windows-ia32 (7/578) smartos-x64 (1/578) smartos-ia32 (1/578) windows-x64 (7/578) http://jenkins.nodejs.org/job/nodejs-v0.10/149/
10:22:48  * abraxasquit (Remote host closed the connection)
10:32:08  * dominictarrjoined
10:45:43  * piscisaureus_joined
10:56:56  * wolfeidauquit (Read error: Connection reset by peer)
10:57:01  * wolfeida_joined
11:02:07  * perezdquit (Quit: perezd)
11:07:03  * `3E|BRBchanged nick to `3E
11:29:24  * bajtosjoined
11:40:00  * piscisaureus_quit (Ping timeout: 245 seconds)
11:41:05  * bnoordhuisjoined
11:50:07  * kevinswiberjoined
11:52:55  * kazuponquit (Remote host closed the connection)
12:05:28  * inolenquit (Ping timeout: 246 seconds)
12:08:13  * kevinswiberquit (Remote host closed the connection)
12:28:30  * kevinswiberjoined
12:33:10  * sgallaghjoined
12:38:34  * hzquit (Disconnected by services)
12:38:37  * hzjoined
12:44:50  * piscisaureus_joined
12:58:45  * hzquit
13:00:34  <MI6>joyent/node: Ben Noordhuis master * 223607c : build: fix arm build after v8 upgrade V8 was upgraded to 3.18 in commit - http://git.io/J02euw
13:03:35  * kazuponjoined
13:09:12  * kazuponquit (Ping timeout: 264 seconds)
13:17:55  <MI6>nodejs-master: #160 FAILURE windows-ia32 (9/582) smartos-x64 (1/582) smartos-ia32 (2/582) http://jenkins.nodejs.org/job/nodejs-master/160/
13:20:26  * piscisaureus_quit (Ping timeout: 245 seconds)
13:27:55  <bnoordhuis>bajtos: olla
13:28:05  <indutny>olla
13:28:18  <bnoordhuis>sup fedor?
13:28:29  <indutny>nothing just hanging around
13:28:35  <bnoordhuis>before you ask, i'm reviewing your ondata PR
13:28:40  <indutny>oh
13:28:47  <indutny>ok
13:28:54  <indutny>I wasn't really going to ask
13:28:55  <indutny>:)
13:29:03  <indutny>I decided to take a passive route
13:29:14  <indutny>and just watch how things are ignored for ages
13:29:35  <bnoordhuis>that's the way of the OSS world, fedor
13:29:38  <bnoordhuis>and not just OSS
13:30:47  <MI6>joyent/node: Trevor Norris master * 0a4ebc3 : src: replace Holder() with This() Switch to always use args.This() to re - http://git.io/ngglSw
13:31:07  <indutny>has he fixed it?
13:31:12  <bnoordhuis>yep
13:31:20  * piscisaureus_joined
13:31:30  <indutny>ok
13:31:47  <bnoordhuis>re OSS, at least once a month i get an update (usually "CLOSED WONTFIX") on an old bug report of mine
13:32:05  <bnoordhuis>my record is a CLOSED WONTFIX in 2010 for an apache httpd issue i filed in 2004 :)
13:32:40  <bnoordhuis>wontfix because apache 1.2 was by that time no longer supported
13:33:13  * defunctzombie_zzchanged nick to defunctzombie
13:33:43  <bnoordhuis>want to hear more anecdotes? i have tons of them
13:33:57  <indutny>not sure
13:34:01  <indutny>:)
13:34:19  <bnoordhuis>when i was young, we didn't have these newfangled bits
13:34:32  <bnoordhuis>all we had were zeroes and we were happy. we didn't know any better!
13:35:08  <bnoordhuis>anyway, how are the bahamas, fedor?
13:35:16  <bnoordhuis>nice weather, crappy internet?
13:35:24  <indutny>I found a spot where it works good
13:35:25  <indutny>:)
13:35:34  <indutny>so everything is perfect now
13:35:41  <indutny>but unfortunatelly there're a lot of raged kids around
13:35:46  <bnoordhuis>raged?
13:36:08  <indutny>yeah
13:36:20  <indutny>they've definite problems with behavior
13:36:40  <indutny>I'm installing another IRC client
13:36:41  <bnoordhuis>how so?
13:36:56  <indutny>bnoordhuis: well, they are very angry and noisy
13:36:57  <bnoordhuis>the behavior problem, that is, not the irc client
13:37:07  <indutny>even when parents are near them
13:37:17  <bnoordhuis>hm. they're not dutch kids, are they?
13:37:27  <indutny>and not russian, indeed
13:37:32  <indutny>american kids
13:37:32  <indutny>meh
13:37:35  <bnoordhuis>h
13:38:26  <indutny>brb
13:41:16  <indutny>ok good
13:41:20  <indutny>I'm finally back
13:41:31  <bajtos>bnoordhuis: hi
13:41:43  <bnoordhuis>bajtos: sup? what are you working on?
13:42:04  <bajtos>bnoordhuis: debugging debugger :-D
13:42:07  <bnoordhuis>hah
13:42:19  <indutny>yay
13:42:21  <indutny>history works now!
13:42:22  <indutny>kewl
13:42:32  <bnoordhuis>bajtos: need help/pointers/etc.?
13:42:40  <bajtos>bnoordhuis: not yet
13:43:07  <bnoordhuis>okay
13:43:57  <bajtos>bnoordhuis: but I might use your insights later on. I'll ping you when I have my ideas formulated enough for discussion
13:44:17  <bnoordhuis>okay, cool
13:44:41  <indutny>ok, time to reboot
13:44:44  <bnoordhuis>indutny: what was your beef with message.seq? calling it a new feature is stretching the definition of 'feature', isn't it?
13:44:51  <indutny>yep
13:44:58  <indutny>I mean
13:45:03  <indutny>its a new feature
13:45:06  <indutny>and new code
13:45:11  <indutny>but I'm ok with it :)
13:45:22  <bnoordhuis>it's a variation on the existing message passing code
13:45:30  <bnoordhuis>only less prone to confusion
13:46:13  <bnoordhuis>the existing code in cluster.js will get awfully confused if you send e.g. two 'online' messages
13:46:24  <bnoordhuis>i guess i should add that to the commit log
13:47:10  <indutny>ok
13:47:35  <indutny>brb
13:47:40  <indutny>returning to room
13:48:04  <MI6>nodejs-master: #161 FAILURE windows-ia32 (9/582) smartos-x64 (2/582) smartos-ia32 (1/582) http://jenkins.nodejs.org/job/nodejs-master/161/
13:48:10  * c4milojoined
14:02:30  * bnoordhuisquit (Ping timeout: 276 seconds)
14:07:36  <indutny>and I'm baaack
14:23:48  <isaacs>indutny: wb
14:24:38  <indutny>wb?
14:28:02  * AvianFlujoined
14:28:26  * bnoordhuisjoined
14:31:27  <isaacs>welcome back
14:37:07  * bnoordhuisquit (Ping timeout: 264 seconds)
14:39:56  <MI6>joyent/node: isaacs created branch v0.11.1-release - http://git.io/TPyJrA
14:43:26  <indutny>ah
14:43:27  <indutny>thanks
14:44:18  <indutny>I don't think your patch will improve it
14:44:43  <indutny>I think its just profiler getting weirdo :)
14:44:58  <indutny>let me try flamegraphing it
14:56:31  <isaacs>indutny: try that patch. it might help. if it actually is spending a bunch of time there, 3.18 supports inlining, and the early return will prevent it
14:56:47  <indutny>ah ,ok
14:56:49  <indutny>one sec
14:57:12  <isaacs>of course, if the profiler is just being a lying jerk, then who can say? :)
14:57:23  <indutny>ok, so it looks like its a bit different
14:57:27  <indutny>from flamegraph point of view
14:57:31  <indutny>wanna take a look?
14:58:17  <indutny>http://blog.indutny.com/f/http-no-ondata.svg
14:58:49  <indutny>need to benchmark master
14:58:51  <indutny>one sec
14:59:57  <isaacs>indutny: not super helpful, that flamegraph
15:00:06  <indutny>well
15:00:06  <isaacs>indutny: and probably yeah, my patch will do nothing.
15:00:12  <indutny>looks like so
15:00:15  <indutny>you can try it yourself
15:00:20  <indutny>while I'm creating another flamegraph
15:00:24  <isaacs>indutny: it's spending time in emit('readable') because that's where it does 100% of its work ;)
15:00:31  <indutny>yep
15:00:32  <indutny>:)
15:00:34  <isaacs>ie, in teh handler, not in the emitReadable *function* itself
15:00:42  <indutny>but obviously
15:00:50  <isaacs>yeah, not any big news there.
15:00:51  <isaacs>ok
15:00:55  <indutny>I think we should try short-circuiting .push() => emit('readable') path
15:01:06  <indutny>this might help
15:01:17  <indutny>I wonder how can we do it...
15:01:27  <indutny>so it was like .ondata(buf)
15:01:44  <indutny>and now it is .push() => internalBuffer.push() => emit('readable') => internalBuffer.pop()
15:01:50  <indutny>and some other checks and everything
15:02:54  <indutny>file:///Users/indutny/http-ondata.svg
15:02:57  <indutny>err
15:03:07  <indutny>http://blog.indutny.com/f/http-ondata.svg
15:03:15  <indutny>ah, its not there
15:03:41  <indutny>looks like github is having hard time
15:04:00  <indutny>ok, uploaded
15:04:01  <indutny>http://blog.indutny.com/f/http-ondata.svg
15:04:03  <indutny>vs
15:04:04  <indutny>http://blog.indutny.com/f/http-no-ondata.svg
15:05:19  <isaacs>indutny: lookks like http://blog.indutny.com/f/http-ondata.svg is borked?
15:05:22  <indutny>yep
15:05:24  <indutny>let me restart it
15:05:32  <indutny>it might be 0.9.6-pre bug :)
15:05:49  <isaacs>your blog runs on 0.9.6-pre??
15:05:56  <indutny>should be working now
15:06:04  <indutny>isaacs: yeah, I was too lazy to update it
15:06:15  <indutny>ok, so basically
15:06:27  <indutny>interesting numbers:
15:06:35  <indutny>socketOnReadable: 81.28%
15:06:45  <indutny>socketOnData: 87%
15:07:03  <isaacs>i suppose i should look at the code here, as welll
15:07:23  <indutny>code is the same
15:07:31  <indutny>it just changed a bit
15:07:35  <indutny>so instead of receiving data
15:07:36  <indutny>it reads it
15:07:48  <isaacs>right
15:08:04  <indutny>gosh
15:08:52  <indutny>so looks like I'm right and .push() => .read() path is just too long
15:11:54  <indutny>ok, time to grab food
15:11:55  <indutny>ttyl
15:11:58  * indutny&
15:11:58  <LOUDBOT>YOU'RE GONNA LOVE MY COCKBALLS
15:12:04  * indutnychanged nick to indutny_away
15:18:47  * bajtosquit (Ping timeout: 255 seconds)
15:19:25  * sgallaghquit (Remote host closed the connection)
15:21:25  * pooyajoined
15:22:26  * stagasquit (Read error: Connection reset by peer)
15:26:17  * bajtosjoined
15:26:53  * mikealjoined
15:27:13  * bajtos_joined
15:27:20  * bajtosquit (Read error: Connection reset by peer)
15:29:17  * bajtos_quit (Read error: Connection reset by peer)
15:30:43  * kazuponjoined
15:31:03  * bajtosjoined
15:31:34  * bajtosquit (Client Quit)
15:32:00  * bajtosjoined
15:32:42  <MI6>joyent/node: isaacs v0.11.1-release * 4babd2b : 2013.04.19, Version 0.11.1 (Unstable) * V8: upgrade to 3.18.0 * uv: Upg (+1 more commits) - http://git.io/xH1R7A
15:37:46  <isaacs>binaries building
15:44:01  * pooyaquit (Quit: pooya)
15:52:36  * jez0990_joined
15:53:36  * jez0990quit (Ping timeout: 252 seconds)
16:08:18  * piscisaureus_quit (Ping timeout: 272 seconds)
16:11:40  <tjfontaine>good day good sirs
16:13:27  <isaacs>morning!
16:13:27  <MI6>joyent/node: isaacs created tag v0.11.1 - http://git.io/ugQP_Q
16:13:57  <tjfontaine>did you cut a libuv 0.11.2 with that?
16:14:07  <indutny_away>finally
16:14:19  * indutny_awaychanged nick to indutny
16:14:35  <indutny>food time ended
16:14:53  <MI6>joyent/node: isaacs master * e4406b7 : Now working on 0.11.2 (+3 more commits) - http://git.io/CwuWxw
16:14:56  <isaacs>tjfontaine: no, i did not.
16:14:59  <tjfontaine>isaacs: btw you know that v0.11.1 windows-x64 won't work this release?
16:15:08  <isaacs>tjfontaine: ??
16:15:18  <isaacs>holy shit.
16:15:19  <isaacs>you're right.
16:15:23  <tjfontaine>the v8 upgrade to 3.18 is hosed on x64 windows
16:15:29  <isaacs>i tested on 32, not 64
16:15:30  <isaacs>crap
16:15:33  <indutny>isaacs: not that I care much
16:15:38  <indutny>but haven't BIO landed in 0.11
16:15:38  <isaacs>it builds fine, but then just... doesn't start.
16:15:55  <tjfontaine>isaacs: build would fail if you weren't doing nosnapshot
16:16:03  <isaacs>tjfontaine: i'm not doing snapshot
16:16:12  <isaacs>oh, if iw asn't doing nosnapshot
16:16:13  <isaacs>right
16:16:19  <tjfontaine>right
16:16:35  <isaacs>ok
16:16:42  <isaacs>well... sounds like we should fix that, then.
16:16:51  <isaacs>or pull in a v8 update if they fix it.
16:17:29  <tjfontaine>when bert and I looked there hadn't been a fix for it, but it also didn't have a branch yet at the time either
16:17:42  <isaacs>i'll add a note to the release docs that this won't work with x86 windows.
16:17:50  <tjfontaine>*x64
16:17:51  <isaacs>we can get it in the next unstable.
16:17:54  <isaacs>er, right
16:17:57  <isaacs>x64
16:22:24  <MI6>joyent/node: isaacs v0.10 * 0b04abc : blog: v0.11.1 does not work on windows x64 (+2 more commits) - http://git.io/gBFlSg
16:27:42  <indutny>great
16:32:50  <MI6>nodejs-master: #162 FAILURE windows-ia32 (7/582) smartos-x64 (1/582) smartos-ia32 (1/582) http://jenkins.nodejs.org/job/nodejs-master/162/
16:38:33  * loladirojoined
16:50:06  * perezdjoined
16:50:24  <MI6>nodejs-v0.10: #150 UNSTABLE windows-ia32 (7/578) smartos-x64 (1/578) smartos-ia32 (1/578) windows-x64 (7/578) linux-x64 (1/578) http://jenkins.nodejs.org/job/nodejs-v0.10/150/
16:51:13  * inolenjoined
16:55:28  * pooyajoined
17:01:06  * mikealquit (Quit: Leaving.)
17:07:29  * dominictarrquit (Quit: dominictarr)
17:10:19  <isaacs>indutny: ohhh.... right. that BIO stuff got removed from 0.10...
17:10:24  <indutny>yep
17:10:25  <indutny>np
17:10:33  <isaacs>i was thinking it was just a merge-in.
17:10:36  <isaacs>totally forgot about that
17:10:38  <isaacs>my bad.
17:14:24  <indutny>its ok
17:15:35  <MI6>joyent/node: Ryan Doenges master * 9026675 : path: add path.isAbsolute(path) An absolute path will always open the sa - http://git.io/Vgn1xg
17:19:35  * `3Equit (Remote host closed the connection)
17:19:44  * TooTallNatejoined
17:29:26  * kazuponquit (Remote host closed the connection)
17:29:53  * kazuponjoined
17:33:24  <MI6>nodejs-master: #163 FAILURE windows-ia32 (9/582) smartos-x64 (2/582) smartos-ia32 (2/582) http://jenkins.nodejs.org/job/nodejs-master/163/
17:34:14  * kazuponquit (Ping timeout: 252 seconds)
17:34:56  * mikealjoined
17:43:14  * mikealquit (Quit: Leaving.)
17:43:18  * dominictarrjoined
17:52:53  <isaacs>tjfontaine: oh, the pull req job doesn't build windows, does it?
17:53:13  <tjfontaine>no, but I will do it manually as I was starting to look into it now
17:55:54  <isaacs>k
17:56:06  <tjfontaine>it's in the process of building
18:00:15  * mikealjoined
18:01:09  * kazuponjoined
18:04:07  <tjfontaine>isaacs: it built, running tests now
18:09:13  * kazuponquit (Ping timeout: 240 seconds)
18:09:38  <tjfontaine>isaacs: tests done, nothing unexpected
18:10:20  <isaacs>tjfontaine: kewl. it builds on windows x64?
18:10:25  <tjfontaine>yes
18:10:28  <isaacs>\o/
18:10:30  <isaacs>problem solved!
18:10:45  <tjfontaine>those the good ones to solve :)
18:11:08  <MI6>joyent/node: Ben Noordhuis master * 2efb6cf : v8: reapply floating patches (+1 more commits) - http://git.io/Rsio5A
18:11:12  <isaacs>especially when the solution is just "pull in in the update"
18:11:21  <tjfontaine>indeed
18:14:08  * perezdquit (Quit: perezd)
18:18:45  * isaacsyoga &
18:18:50  * isaacs&
18:18:50  <LOUDBOT>NOBODY EXPECTS THE ICEBERG.
18:18:59  <isaacs>theerrre it is :)
18:19:04  <isaacs>LOUDBOT: i've missed you.
18:19:05  <LOUDBOT>isaacs: I'LL ALLOW IT, BUT YOU'D BETTER BE GOING SOMEWHERE WITH THIS
18:19:15  <isaacs>LOUDBOT: oh... trust me. we're going somewhere.
18:19:15  <LOUDBOT>isaacs: HODE UP FO DAT BURN
18:20:22  <tjfontaine>heh
18:22:00  * kazuponjoined
18:24:02  * abraxasjoined
18:25:45  * brsonjoined
18:28:24  * bajtosquit (Quit: bajtos)
18:28:49  * abraxasquit (Ping timeout: 256 seconds)
18:30:33  <MI6>nodejs-master: #164 UNSTABLE windows-ia32 (7/582) linux-ia32 (1/582) linux-x64 (2/582) smartos-x64 (2/582) smartos-ia32 (2/582) windows-x64 (8/582) http://jenkins.nodejs.org/job/nodejs-master/164/
18:31:05  <tjfontaine>yay for fixed builds
18:32:40  * sblomjoined
18:35:44  * piscisaureus_joined
18:37:09  <tjfontaine>fear not, the windows guys have arrived!
18:39:47  * `3rdEdenjoined
18:40:30  <MI6>libuv-master: #80 UNSTABLE smartos (3/188) windows (4/189) linux (1/188) osx (1/188) http://jenkins.nodejs.org/job/libuv-master/80/
18:40:53  <tjfontaine>piscisaureus_: do windows pipes have to be \\.\pipe\? I'm looking at being able to run multiple libuv test suites at the same time, trying to decide how to do the naming
18:41:28  <piscisaureus_>tjfontaine: yes. Or \\?\pipe\
18:41:38  <indutny>what's the difference?
18:42:07  <tjfontaine>hmm ok
18:42:25  <piscisaureus_>If you start with \\.\ windows will do more processing like removing . and .. and replacing forward slashes by backward ones
18:43:30  <tjfontaine>off hand, do you know if the pipes used in the test suites are used in child processes? or could I append the pid to the name?
18:51:59  * dominictarrquit (Quit: dominictarr)
18:53:14  * dominictarrjoined
18:59:35  <MI6>libuv-node-integration: #28 UNSTABLE smartos-ia32 (2/582) osx-x64 (1/582) smartos-x64 (2/582) osx-ia32 (1/582) windows-x64 (8/582) windows-ia32 (8/582) http://jenkins.nodejs.org/job/libuv-node-integration/28/
19:01:40  <sblom>tjfontaine: I don't know anything about the pipes used in the test suites off hand, but I can probably figure it out fairly quickly. Do you have a specific test case in mind?
19:03:13  * bnoordhuisjoined
19:10:46  <bnoordhuis>back
19:11:03  <bnoordhuis>does 3.18.1 fix windows x64?
19:29:53  * brsonquit (Ping timeout: 240 seconds)
19:31:00  * brsonjoined
19:31:40  * kevinswiberquit (Remote host closed the connection)
19:38:03  * mikealquit (Quit: Leaving.)
19:45:11  <tjfontaine>bnoordhuis: yes
19:45:40  <tjfontaine>sblom: well test-ping-pong is probably the easiest to check
19:46:55  <tjfontaine>or we could ask bnoordhuis about them
19:50:42  * inolenquit (Ping timeout: 256 seconds)
19:51:47  * inolenjoined
19:56:57  <piscisaureus_>tjfontaine: the pipes are anonymouse
19:57:01  <piscisaureus_>*anonymous
19:58:00  <tjfontaine>piscisaureus_: then what are the #define's for in task.h?
19:58:02  <piscisaureus_>tjfontaine: however I don't know about parallelizing. libuv often needs to bind a specific port etc
19:58:11  <piscisaureus_>est
19:58:21  <tjfontaine>well I can work around TEST_PORT pretty easily
19:58:53  <piscisaureus_>tjfontaine: oh - those are just for a particular test. I get it.
19:59:15  <piscisaureus_>tjfontaine: yeah you can add a pid but you have to sprintf then
19:59:48  <tjfontaine>piscisaureus_: right, but that doesn't necessarily work if the tests require the same name to exist in a child process (if any tests like that exist)
19:59:56  <piscisaureus_>tjfontaine: sorry, I was confused, I though that you were talking about the pipe between the test runner and the test implementation
20:00:14  <tjfontaine>right no, not that one
20:00:43  <piscisaureus_>tjfontaine: sure. I don't know about that off hand, you need to look inside the test then
20:01:13  <tjfontaine>I may re-use the TEST_PORT to define the name of the pipe
20:01:25  <tjfontaine>just to be safe if someone does it that way
20:11:06  * piscisaureus_quit (Ping timeout: 256 seconds)
20:12:40  * kevinswiberjoined
20:17:21  * piscisaureus_joined
20:17:27  * defunctzombiechanged nick to defunctzombie_zz
20:19:05  * kevinswi_joined
20:21:20  * kevinswiberquit (Ping timeout: 256 seconds)
20:21:45  * kazuponquit (Remote host closed the connection)
20:23:26  * defunctzombie_zzchanged nick to defunctzombie
20:24:01  * perezdjoined
20:33:54  <MI6>libuv-master: #81 FAILURE smartos-ia32 (3/188) smartos-x64 (3/188) windows-ia32 (4/189) osx-ia32 (1/188) osx-x64 (2/188) linux-x64 (1/188) linux-ia32 (1/188) http://jenkins.nodejs.org/job/libuv-master/81/
20:35:28  * mikealjoined
20:38:06  <indutny>whoa
20:38:28  <tjfontaine>I'm adding target builds for libuv fear not
21:01:38  * stagasjoined
21:11:37  * mikealquit (Quit: Leaving.)
21:17:21  * mikealjoined
21:19:16  <tjfontaine>piscisaureus_: have you done a `vcbuild.bat x64` for libuv recently?
21:19:40  <piscisaureus_>tjfontaine:: long time ago
21:20:04  <tjfontaine>ok, well it needs a couple tweaks to work, I'll submit a patch here shortly
21:20:56  * rendarquit
21:22:06  * kazuponjoined
21:22:08  <tjfontaine>piscisaureus_: https://github.com/tjfontaine/libuv/compare/build-win-64
21:22:49  <piscisaureus_>tjfontaine: great!
21:24:21  <tjfontaine>shall I do a full pr for it?
21:26:39  * kazuponquit (Ping timeout: 245 seconds)
21:28:47  * trevnorrisjoined
21:29:31  <trevnorris>hello
21:29:36  <tjfontaine>gday trevnorris
21:33:15  <piscisaureus_>tjfontaine: weird. I never noticed that bug actually. Which version of VS is on the build machines?
21:33:53  <tjfontaine>VS2010 is what it uses, but 2012 is there as well, but when I pull up uv.sln the config is x64 not amd64
21:34:26  <tjfontaine>the GYP_MSVS_ change is not entirely necessary, but if you're coming from the windowsdk build world VCINSTALLDIR may not be set
21:35:21  <piscisaureus_>tjfontaine: ok, the amd->x64 lgtm
21:35:41  <piscisaureus_>tjfontaine: VCINSTALLDIR not set? so windowsdk build works when that is not set?
21:35:46  * piscisaureus_has no clue
21:35:56  <tjfontaine>yes, it still works without that being set
21:41:19  * bnoordhuisquit (Ping timeout: 264 seconds)
21:59:30  <kellabyte>any idea what this means? I added libuv as a dependency in my gyp file - gyp: Undefined variable library in lib/libuv/uv.gyp while loading dependencies
22:05:44  <tjfontaine>kellabyte: gist your gyp file
22:06:08  <TooTallNate>kellabyte: i think you need to define "library" to either "static_library" or "shared_library"
22:06:18  <TooTallNate>depending on your preference of course…
22:06:49  <tjfontaine>TooTallNate: is there a -D variable that can be used to override the active configuration or do I have to use a config.gypi?
22:07:04  <TooTallNate>tjfontaine: you can use -D
22:07:18  <tjfontaine>which variable controls it?
22:07:28  <TooTallNate>controls what?
22:07:47  <tjfontaine>which configuration I'm building, i.e. Debug vs Release
22:07:56  <tjfontaine>./gyp_uv -Dtarget_arch=ia32 -Dconfiguration=Release
22:07:56  <tjfontaine>['-Dtarget_arch=ia32', '-Dconfiguration=Release', '/Users/tjfontaine/Development/libuv/uv.gyp', '-I', '/Users/tjfontaine/Development/libuv/common.gypi', '--depth=.', '-f', 'make', '-Goutput_dir=/Users/tjfontaine/Development/libuv/out', '--generator-output', '/Users/tjfontaine/Development/libuv/out', '-Dgcc_version=42', '-Dclang=1', '-Dhost_arch=x64', '-Dlibrary=static_library', '-Dcomponent=static_library']
22:08:02  <tjfontaine>localhost:libuv tjfontaine$ make -C out/
22:08:04  <tjfontaine> CC(target) /Users/tjfontaine/Development/libuv/out/Debug/obj.target/libuv/src/fs-poll.o
22:08:20  <TooTallNate>tjfontaine: oh for the Configuration, i'm not sure about that part
22:08:26  <tjfontaine>right ok
22:08:30  <TooTallNate>tjfontaine: you can override the Configuration when invoking make though
22:08:46  <TooTallNate>make BUILDTYPE=Debug
22:08:48  <TooTallNate>iirc
22:08:56  <kellabyte>tjfontaine: https://gist.github.com/kellabyte/5423553 I'm super noob right now so :P
22:09:04  <TooTallNate>kellabyte: https://github.com/TooTallNate/node-gyp/blob/master/lib/configure.js#L387
22:09:29  <tjfontaine>TooTallNate: we're in libuv land, in node land Makefile just runs configure and writes config.gypi :)
22:09:47  * loladiroquit (Quit: loladiro)
22:09:50  <TooTallNate>tjfontaine: i'm talking about gyp's generated makefile though :)
22:09:57  <tjfontaine>hm
22:09:57  <TooTallNate>not node's
22:10:14  <tjfontaine>oh I'll be damned
22:10:33  <tjfontaine>well why can't I tweak that at generation
22:10:36  <tjfontaine>whatever
22:10:45  <tjfontaine>thanks TooTallNate
22:11:26  <TooTallNate>tjfontaine: for gyp, try --build=Debug or --build=Release
22:11:50  * sblomquit (Ping timeout: 256 seconds)
22:12:01  <kellabyte>I tried -Dlibrary=shared_library on the gyp command line but same error
22:12:06  * sblomjoined
22:12:08  <tjfontaine>TooTallNate: well it would be nice if they documented that in --help
22:12:20  <TooTallNate>tjfontaine: that's where I found it :p
22:12:20  <tjfontaine>oh that triggers the full build there
22:12:44  <tjfontaine>you looked in the real gyp though, I was in the gyp_uv wrapper :P
22:12:51  <tjfontaine>--build=CONFIGS configuration for build after project generation
22:12:52  <tjfontaine>oh
22:12:52  <tjfontaine>fuck
22:12:54  <tjfontaine>I give up
22:13:08  <TooTallNate>tjfontaine: hahaha :p
22:15:41  <kellabyte>updated my gist with the gyp error and what I'm running on the command line: https://gist.github.com/kellabyte/5423553
22:16:02  <TooTallNate>kellabyte: try --depth=.
22:21:45  * c4miloquit (Remote host closed the connection)
22:22:12  * c4milojoined
22:22:44  * kazuponjoined
22:23:12  <kellabyte>TooTallNate: I think that worked, no errors, is that supposed to output my executable somewhere? :P
22:23:30  <TooTallNate>kellabyte: that generates build files… now you need to build the project
22:23:38  <kellabyte>ah!
22:23:41  <TooTallNate>kellabyte: assuming you're using the "make" target, you should just need to type `make`
22:24:21  <kellabyte>I didn't so it spit out xcode project but I can change it to make
22:24:34  <kellabyte>why did I have to define -Dlibrary=shared_library if in my gyp it says it's an executable?
22:24:50  <TooTallNate>kellabyte: the "library" variable is used by gyp
22:25:02  <TooTallNate>kellabyte: err sorry, i mean libuv
22:25:21  <TooTallNate>kellabyte: so you're saying "build libuv as a shared library, then build my executable and link to the shared libuv I built"
22:26:32  <kellabyte>ahh I get it, so -Dlibrary isn't a gyp thing telling it to compile my app to a library, its a libuv variable, gotcha
22:27:06  * c4miloquit (Ping timeout: 256 seconds)
22:29:30  * kazuponquit (Ping timeout: 276 seconds)
22:29:45  <kellabyte>TooTallNate: ok closer, looks to have compiled my executable and libuv dylib to an out directory but when I run my hello world app I get - dyld: Library not loaded: /usr/local/lib/libuv.dylib, the lib is sitting in the current dir though
22:29:54  <kellabyte>TooTallNate: sorry I know these are basic questions probably :)
22:30:12  <TooTallNate>kellabyte: you probably want static_library instead of shared_library ;)
22:31:28  <kellabyte>ah cool worked! although my hello world doesn't call anything in libuv yet :)
22:32:08  <kellabyte>but it will in a minute!
22:32:29  <TooTallNate>well… onward!
22:32:39  <tjfontaine>ah well, gyp --build is useless on windows without full VS
22:32:43  <tjfontaine>'yay'
22:32:50  <TooTallNate>really?
22:32:58  <TooTallNate>that seems weird…
22:33:01  <tjfontaine>yes, it usees devenv.com which isn't there
22:33:06  <tjfontaine>at least according to http://stackoverflow.com/questions/7818543/no-devenv-file-in-microsoft-visual-express-10
22:33:20  <TooTallNate>tjfontaine: but setting "default_configuration" in the gypi works?
22:33:35  <tjfontaine>TooTallNate: well I can msbuild myself, but if I want gyp to invoke I'm SOL
22:33:59  <tjfontaine>you know, the whole platform independent thing not really being all that independent
22:35:04  <isaacs>tjfontaine: platform independent is only for very high-level stuff
22:35:13  <isaacs>tjfontaine: mostly, you are lucky to get platform-competent
22:35:51  <tjfontaine>indeed
22:36:28  <tjfontaine>lets try some other flags
22:38:46  <tjfontaine>why on earth --build in gyp tries to invoke devenv instead of msbuild, though I guess it's because they don't want to do the vcvars jump
22:44:57  <sblom>tjfontaine: Using devenv to build a .sln pre-dates msbuild by 3 years or maybe more. I bet that part of gyp was written back then and never really updated.
22:45:50  <tjfontaine>sblom: right, but they have the info because we tell them what version of msvs we're using
22:46:09  <sblom>oic
22:46:26  <tjfontaine>if they know we're using 2010 then they know we have msbuild, there's no reason to act like it's vc6 :)
22:47:02  <sblom>sure, unless they just never updated.
22:47:04  <sblom>:p
22:47:21  <tjfontaine>heh
22:47:54  * bnoordhuisjoined
22:50:46  * bnoordhuisquit (Read error: Operation timed out)
23:00:27  * loladirojoined
23:02:49  <kellabyte>TooTallNate: am I supposed to add my .h files under the sources along with my .c files? guessing maybe not? it's giving me a linker that it can't find my http_parser.o
23:03:06  <TooTallNate>kellabyte: the .h files are not necessary
23:03:22  <TooTallNate>including them in "sources" does nothing to the generated build files AFAIK
23:07:46  <kellabyte>TooTallNate: holy shit! it worked! the http server compiles and runs on unix now :)
23:07:56  * kevinswi_quit (Remote host closed the connection)
23:08:03  <tjfontaine>*magic*
23:08:07  <TooTallNate>kellabyte: what did you change?
23:08:25  <kellabyte>TooTallNate: I removed the .h and I was trying some wild things like including the C file too lol
23:08:39  <kellabyte>err .c file
23:08:47  <TooTallNate>of the http_parser you mean?
23:08:58  <kellabyte>no, in my program.cs
23:08:59  <kellabyte>err .c
23:09:02  <kellabyte>damn C# habits lol
23:09:45  <kellabyte>basically I was an idiot
23:10:20  <tjfontaine>that's me in a nutshell today
23:25:25  <MI6>joyent/node: isaacs master * c773124 : lint (+1 more commits) - http://git.io/edZ9Dw
23:25:49  * kazuponjoined
23:28:46  * kevinswiberjoined
23:30:38  * kazuponquit (Ping timeout: 255 seconds)
23:35:20  * pooyaquit (Quit: pooya)
23:39:41  * inolenquit (Remote host closed the connection)
23:39:42  <kellabyte>is there a curl -v type equivilent for http pipelined requests? for some reason my pipelined requests are timing out but non-pipelined is fine :P
23:40:02  * inolenjoined
23:42:55  * perezdquit (Quit: perezd)
23:44:17  * `3rdEdenquit (Remote host closed the connection)
23:44:31  <MI6>nodejs-master: #165 UNSTABLE windows-ia32 (8/582) osx-x64 (1/582) smartos-x64 (7/582) smartos-ia32 (6/582) windows-x64 (7/582) http://jenkins.nodejs.org/job/nodejs-master/165/
23:49:35  <indutny>hey devs
23:49:46  <indutny>just FYI
23:50:10  <indutny>I'll be not online today (for the reset of the day) and tomorrow (until evening EST)
23:51:14  <indutny>not sure if you care
23:51:16  <indutny>but anyway
23:58:27  <TooTallNate>indutny: impossibru!?!!!
23:58:45  <indutny>yep
23:58:50  <indutny>but that's a price to pay :)
23:58:51  <indutny>I mean
23:58:55  <indutny>internet costs 20$/day here
23:59:04  <indutny>and I don't want to pay for it, just to check email at morning
23:59:08  <indutny>and fly away to USA
23:59:10  * kevinswiberquit (Remote host closed the connection)
23:59:15  <indutny>that seems to be stupid