00:00:20  <isaacs>TooTallNate: no, it waits until the end, then fails
00:00:35  <rphillips>looks like ev_default_loop and ev_loop isn't getting exported on linux
00:00:39  <rphillips>worked fine on OSX
00:02:41  <igorzi_>isaacs: i'm actually seeing similar v8 problems in master
00:02:43  <rphillips>this might be my fault
00:02:56  <isaacs>igorzi_: ok
00:03:19  <isaacs>well... now you can see those problems more easily by installing from the msi in the x64 folder. ;)
00:03:53  <igorzi_>nice :)
00:04:13  <igorzi_>piscisaureus_: https://github.com/joyent/node/issues/2807 can be closed?
00:05:01  <piscisaureus_>igorzi_: I think so. Sorry, I just landed a fix from brandon philips
00:05:09  <piscisaureus_>igorzi: it's the same as your patch :-)
00:05:27  * sh1mmerquit (Quit: sh1mmer)
00:05:39  <piscisaureus_>igorzi_: but we should upgrade libuv in node
00:06:11  <igorzi_>piscisaureus_: yep
00:08:08  <piscisaureus_>pfox___: you need to build with -lws2_32
00:10:02  <piscisaureus_>igorzi_: before we even start, is that pfx file format parsable at all?
00:11:17  <piscisaureus_>igorzi: http://www.cs.auckland.ac.nz/~pgut001/pubs/pfx.html kind of scares me
00:12:53  <piscisaureus_>igorzi_: (plus the observed experience of a colleague of mine, that spent a week getting azure pkcs-12 files to work)
00:15:03  <igorzi_>piscisaureus_: i think openssl knows how to deal with it
00:15:49  <piscisaureus_>igorzi: I think eventually we received some secret cantation from scott hanselman that made it all work
00:16:14  <piscisaureus_>igorzi_: if openssl can deal with it I think I am fine supporting pfx
00:16:55  <piscisaureus_>(I'm not saying that it would be out of the question otherwise - but if openssl supports it it's gonna be much easier)
00:17:12  * mikealjoined
00:24:43  * perezdquit (Ping timeout: 252 seconds)
00:25:36  <philips>igorzi_: piscisaureus_ : Can it land in master too?
00:26:01  <philips>How does v0.6 move into master?
00:26:08  <TooTallNate>philips: it'll get in there the next time 0.6 gets merged into master
00:26:15  <philips>oh, ok
00:26:20  <TooTallNate>they usually do that before a release I think
00:27:02  * perezdjoined
00:30:10  * xaqquit (Remote host closed the connection)
00:30:59  <CIA-99>libuv: Brandon Philips master * rfca18c3 / src/win/fs.c :
00:30:59  <CIA-99>libuv: win: fs: handle EOF in read
00:30:59  <CIA-99>libuv: in luvit after upgrade libuv from 243cfc to d3efef readSync started
00:30:59  <CIA-99>libuv: failing. It seems that the code cleanup stopped handling EOF
00:30:59  <CIA-99>libuv: Trivially reproduced with this
00:31:00  <CIA-99>libuv: local fs = require('fs')
00:31:00  <CIA-99>libuv: print(fs.readFileSync('foo.luvit')) - http://git.io/9aDiDQ
00:31:00  <CIA-99>libuv: Bert Belder master * rfbc2154 / (6 files in 2 dirs): Merge branch 'v0.6' - http://git.io/1yWpkA
00:31:17  <piscisaureus_>^-- philips, that's what you wanted right?
00:31:36  <philips>piscisaureus_: exactly, thanks!
00:31:40  * perezd_joined
00:31:52  <philips>piscisaureus_: we are tracking libuv master on luvit via submodules, thx
00:32:36  <piscisaureus_>philips: np
00:32:51  * travis-cijoined
00:32:51  <travis-ci>[travis-ci] joyent/libuv#104 (master - fbc2154 : Bert Belder): The build is still failing.
00:32:51  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/d3efefb...fbc2154
00:32:51  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/731451
00:32:51  * travis-cipart
00:33:16  * perezdquit (Ping timeout: 248 seconds)
00:33:16  * perezd_changed nick to perezd
00:35:43  * perezd_joined
00:37:44  * perezdquit (Ping timeout: 244 seconds)
00:39:01  * xaqjoined
00:39:14  * perezdjoined
00:40:12  * perezd_quit (Ping timeout: 255 seconds)
00:42:17  * perezd_joined
00:46:20  * perezdquit (Ping timeout: 276 seconds)
00:46:21  * perezd_changed nick to perezd
00:48:33  * AvianFlujoined
00:49:00  * perezd_joined
00:50:37  * perezdquit (Ping timeout: 240 seconds)
00:50:38  * perezd_changed nick to perezd
00:52:43  * brsonjoined
01:00:29  <isaacs>oh, man, master is in such a state with v8 on solaris right now
01:00:49  <isaacs>standalone v8 3.9.9 seems to work fine, so i wonder if maybe it's just us
01:12:45  * diogosnowsquit (Ping timeout: 252 seconds)
01:15:20  * paddybyersquit (Quit: paddybyers)
01:20:18  <mmalecki>hm. there might be something wrong with 0.7.5 release.
01:20:27  <mmalecki>or some regression was introduced earlier.
01:20:42  <TooTallNate>mmalecki: what's wrong?
01:20:52  <mmalecki>http://travis-ci.org/#!/flatiron/flatiron/jobs/731523/L138 <- "execvp(): No such file or directory"
01:21:08  <mmalecki>I haven't seen this on previous releases
01:21:26  <mmalecki>I tested this project on v0.7.5-pre and haven't seen this neither.
01:21:41  <mmalecki>just upgraded to v0.7.6-pre and I started seeing this.
01:21:57  <TooTallNate>seems strange. that error usually happens when a child process is spawned and the program is not found
01:22:22  <mmalecki>yeah!
01:22:38  <mmalecki>and vows doesn't do any funky stuff with subprocesses.
01:22:45  <mmalecki>OH WELL MAYBE A BIT.
01:23:23  <mmalecki>but this happens only with `vows --isolate`, which means spawning one child_process per each test file.
01:23:43  <isaacs>mmalecki: maybe there's a problem with doing that
01:23:52  <isaacs>mmalecki: er, a problem with *how* it's doing that
01:24:03  <isaacs>i didn't see this issue with my projects that use tap, which always does this
01:24:07  <mmalecki>isaacs: it worked on previous releases and I don't think we changed the API
01:24:12  <isaacs>there's no "don't isolate" option
01:24:14  <isaacs>nope
01:24:23  <isaacs>but i don't know what vows is doing
01:24:52  <TooTallNate>mmalecki: i'd bet more likely its something with the environment
01:24:59  <TooTallNate>some subtle difference
01:25:15  <TooTallNate>at least, try running vows tests
01:25:23  <mmalecki>I have a bad feeling about this...
01:25:35  <mmalecki>sec, I might've broken something in startup
01:26:09  <mmalecki>oh well, yeah, I did
01:26:50  * mraleph1quit (Quit: Leaving.)
01:26:57  <mmalecki>isaacs: can you revert https://github.com/joyent/node/commit/b0c15412270f32e00c268c578f07a1ed032323f5 ?
01:27:51  <indutny>morning
01:28:24  <isaacs>mmalecki: oh, really?
01:28:40  <isaacs>mmalecki: can you send me a pull req that reverts it, and adds a test so we don't make the same mistake again next time?
01:28:44  <indutny>don't want to be sceptic
01:28:46  <indutny>:)
01:28:50  <mmalecki>isaacs: yes
01:28:56  <isaacs>mmalecki: thanks, that'd be awesome
01:29:13  <indutny>isaacs: heya
01:29:24  <indutny>isaacs: I'm not sure about .connections stuff in Andreas PR
01:29:57  <indutny>isaacs: But it seems to be a way to a viable API
01:30:06  <indutny>isaacs: where everything works as user expects
01:30:23  <isaacs>indutny: yeah.
01:30:30  <isaacs>do we really need to expose a .connections anyway?
01:30:43  <indutny>that's it, I'm not sure
01:30:56  <indutny>for straight-forward minimal implementation - no
01:30:58  <isaacs>i'm ok with just removing it from the docs and moving to _connections, and whatever, it's not valid.
01:31:14  <isaacs>at least then if you want to look at it, you know you're in hazardous territory
01:31:17  <indutny>isaacs: why? it works fine with regular (non-cluster) serer
01:31:25  <indutny>s/serer/server
01:31:48  <indutny>isaacs: actually he's decrementing connections to determine when server can .close it's socket
01:31:57  <indutny>isaacs: to support .close([callback])
01:32:31  <indutny>removing connections from API sounds like an idea
01:32:53  <isaacs>yeah, he needs it for cluster.
01:33:04  <isaacs>or, i mean, it's *eventually* consistent, right?
01:33:27  <isaacs>we could just say, when you pass a handle to a child, or use cluster, it's not 100% consistent
01:33:38  <indutny>hm... I think no
01:34:13  <indutny>we can know max number of connections
01:34:24  * mikealquit (Quit: Leaving.)
01:34:28  <indutny>and that's ok for waiting when server will be free enough
01:34:49  <indutny>I think cluster should not count at all
01:35:23  <indutny>it may request workers to wait until all their connections will be dead and close socket on their responds
01:35:37  <indutny>counting looks like an overhead
01:35:48  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
01:36:26  <isaacs>yeah
01:36:46  <isaacs>it'd be better to either have an accurate (or, eventually-accurate, like on a heartbeat or something) count, or none at all
01:37:01  <indutny>+1 for latter :)
01:37:22  <indutny>so I think I'll merge everything except connections
01:37:25  <isaacs>with this change, it's already removed from the API when using cluster, which means effectively, you can't use it or you're giving up that functionality
01:37:36  <isaacs>indutny: did he fix up the spelling?
01:37:44  <indutny>isaacs: dunno
01:37:45  <isaacs>useing, fluch, etc.
01:37:47  <indutny>isaacs: opening
01:38:47  <indutny>isaacs: no
01:38:59  <indutny>ok, I'll wait until Andreas wiil be online
01:39:03  <isaacs>k
01:40:45  <mmalecki>isaacs: if I want to pass data before test and fixture, should I use fork or stdout?
01:41:04  <mmalecki>s/before/between
01:41:05  <indutny>gtg
01:41:09  <indutny>ttyl guys
01:41:14  * indutnychanged nick to indutny_away
01:41:42  <isaacs>mmalecki: i dunno. depends on what you wnat to test :)
01:42:01  <mmalecki>isaacs: process.argv[0] being fucked
01:42:01  <isaacs>mmalecki: i usually use stdout and stdin to talk to children.
01:42:08  <mmalecki>ok
01:42:27  <isaacs>mmalecki: oh, in node tests.
01:42:28  <AvianFlu>I usually use small words and a friendly tone of voice.
01:42:45  <isaacs>mmalecki: we usually pass arguments, that's always a nice way to do it
01:42:59  <mmalecki>isaacs: I need fixture to talk to the test
01:43:05  * perezd_joined
01:43:25  <isaacs>mmalecki: like, if (process.argv[2] !== "child") { child = spawn(process.execPath, [__filename, "child"]) } else { do child stuff }
01:43:54  <isaacs>mmalecki: then also the test logic is all visible in one place.
01:44:05  <mmalecki>isaacs: ok, sounds good
01:44:06  <isaacs>but if that's not possible for some reason, of course, a fixture is fine, too
01:44:11  * mjr_quit (Quit: mjr_)
01:44:15  * xaqquit (Remote host closed the connection)
01:45:32  * perezdquit (Ping timeout: 244 seconds)
01:45:32  * perezd_changed nick to perezd
01:58:24  <mmalecki>ah... this can only happen with global node :/
01:59:33  <isaacs>mmalecki: orly?
01:59:35  <isaacs>mmalecki: what happens?
02:00:01  <mmalecki>isaacs: so we've replaced this path mangling with path.resolve
02:00:18  <isaacs>mmalecki: also, you can test that by doing spawn('node', args, {env:{PATH:path.dirname(process.execPath)}})
02:00:30  <mmalecki>isaacs: hacky. I like this.
02:00:43  <isaacs>probably wanna copy the rest of process.env onto there, too
02:01:11  <isaacs>var childEnv = {PATH:...}; Object.keys(process.env).forEach(function(k) { childEnv[k] = process.env[k]; });
02:01:50  * AvianFluquit (Quit: Leaving)
02:02:20  <mmalecki>isaacs: util._extends :)
02:02:30  <mmalecki>_extend, even!
02:03:00  <mmalecki>ok, got the test, thanks
02:03:13  <isaacs>np
02:03:57  * TooTallNatequit (Quit: Linkinus - http://linkinus.com)
02:05:34  * orlandovftwquit (Ping timeout: 272 seconds)
02:13:10  * dapquit (Quit: Leaving.)
02:19:32  * xaqjoined
02:26:35  <mmalecki>isaacs: https://github.com/joyent/node/pull/2821
02:28:11  <mmalecki>feel free to rephrase commits, there are both pretty much the same
02:31:43  * pieternquit (Quit: pietern)
02:32:45  <mmalecki>isaacs: sorry about this.
03:04:29  * brsonquit (Quit: leaving)
03:07:13  <isaacs>mmalecki: it's cool
03:07:53  <isaacs>mmalecki: unstable branches ftw :) that's why we try not to do this kind of stuff in stable branches.
03:07:59  <isaacs>we got a test out of it
03:09:04  * AvianFlujoined
03:09:45  <mmalecki>isaacs: oh well. I feel silly anyway
03:09:49  <isaacs>:)
03:14:28  * sh1mmerjoined
03:16:55  * sh1mmerquit (Client Quit)
03:18:51  * orlandovftwjoined
03:20:00  * indexzerojoined
03:26:03  * dshaw_joined
03:27:05  * pfox___quit (Ping timeout: 245 seconds)
03:30:28  <isaacs>wow, so at least as far back as 0.7.1, node's been pretty completely busted on solaris
03:30:29  <isaacs>that sucks
03:31:44  * mmaleckichanged nick to mmalecki[zzz]
03:39:38  * orlandovftwquit (Ping timeout: 252 seconds)
03:55:35  * felixgejoined
03:55:35  * felixgequit (Changing host)
03:55:35  * felixgejoined
03:57:41  * CoverSlidequit (Ping timeout: 244 seconds)
04:01:46  * indexzeroquit (Quit: indexzero)
04:02:54  * felixgequit (Quit: felixge)
04:04:12  * orlandovftwjoined
04:04:12  * orlandovftwquit (Client Quit)
04:04:26  * orlandovftwjoined
04:13:14  * xaqquit (Remote host closed the connection)
04:15:39  * felixgejoined
04:15:39  * felixgequit (Changing host)
04:15:39  * felixgejoined
04:24:13  <CIA-99>node: Maciej Małecki master * rb73ec84 / test/simple/test-process-argv-0.js : (log message trimmed)
04:24:13  <CIA-99>node: test: test for `process.argv[0]` correctness
04:24:13  <CIA-99>node: joyent/node@b0c15412270f32e00c268c578f07a1ed032323f5 introduced a
04:24:13  <CIA-99>node: regression causing `process.argv[0]` to be invalid in node processes
04:24:13  <CIA-99>node: spawned from `PATH` (without explicit path to executable file - for
04:24:13  <CIA-99>node: example when using global node installation).
04:24:15  <CIA-99>node: Instead of finding a correct path to the executable, `process.cwd()`
04:24:15  <CIA-99>node: Maciej Małecki master * r27a937b / src/node.js : (log message trimmed)
04:24:16  <CIA-99>node: Revert "startup: use `path.resolve` instead of `path.join(cwd, ...)`"
04:24:16  <CIA-99>node: This reverts commit b0c15412270f32e00c268c578f07a1ed032323f5.
04:24:17  <CIA-99>node: Reverted commit introduced a regression causing `process.argv[0]` to be
04:24:17  <CIA-99>node: invalid in node processes spawned from `PATH` (without explicit path to
04:24:18  <CIA-99>node: executable file - for example when using global node installation).
04:24:18  <CIA-99>node: Instead of finding a correct path to the executable, `process.cwd()`
04:38:42  * travis-cijoined
04:38:42  <travis-ci>[travis-ci] joyent/node#491 (master - 27a937b : Maciej Małecki): The build is still failing.
04:38:42  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/a84adad...27a937b
04:38:42  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/732635
04:38:42  * travis-cipart
04:54:04  * mikealjoined
04:54:09  * igorziquit (Quit: Page closed)
04:55:04  * isaacsquit (Remote host closed the connection)
05:09:35  * mikealquit (Quit: Leaving.)
05:15:20  * AvianFluquit (Quit: Leaving)
05:15:35  * felixgequit (Quit: felixge)
05:15:47  * pfox___joined
05:18:38  * isaacsjoined
05:25:16  * AvianFlu_joined
05:26:40  * mikealjoined
05:38:38  * indexzerojoined
05:38:52  * indexzeroquit (Client Quit)
05:39:07  * indexzerojoined
05:42:16  * mikealquit (Quit: Leaving.)
06:18:10  * indexzeroquit (Quit: indexzero)
06:20:56  * perezdquit (Quit: perezd)
06:35:11  * paddybyersjoined
06:37:04  * isaacsquit (Remote host closed the connection)
06:52:13  * indexzerojoined
06:52:18  * indexzeroquit (Client Quit)
06:59:00  * mikealjoined
07:02:29  * mikeal1joined
07:03:10  * mikealquit (Ping timeout: 244 seconds)
07:25:03  * indutny_awaychanged nick to indutny
07:47:36  * stephankjoined
07:50:08  * AvianFlu_quit
07:53:23  * AvianFlujoined
07:58:58  * pfox___quit (Ping timeout: 272 seconds)
08:03:22  * mraleph1joined
08:07:45  * paddybyersquit (Quit: paddybyers)
08:41:46  * txdvjoined
08:46:06  * txdv_quit (Ping timeout: 260 seconds)
08:55:44  * mraleph1quit (Quit: Leaving.)
09:13:11  * paddybyersjoined
09:15:24  * paddybyers_joined
09:17:36  * paddybyersquit (Ping timeout: 260 seconds)
09:17:36  * paddybyers_changed nick to paddybyers
09:52:59  * orlandovftwquit (Ping timeout: 276 seconds)
10:47:17  * avsejjoined
10:52:42  * avsejpart ("Ex-Chat")
10:54:43  * avsejjoined
10:57:31  * avsejquit (Quit: Quit)
10:58:07  * avsejjoined
11:14:33  * dshaw_quit (Quit: Leaving.)
11:16:15  * bnoordhuisjoined
12:10:38  <CIA-99>libuv: Frank Denis master * rf6c8e78 / src/unix/darwin.c : Fix memory leak in uv_exepath() on OSX. - http://git.io/lLqUqw
12:12:33  * travis-cijoined
12:12:33  <travis-ci>[travis-ci] joyent/libuv#105 (master - f6c8e78 : Frank Denis): The build is still failing.
12:12:33  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/fbc2154...f6c8e78
12:12:33  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/734203
12:12:33  * travis-cipart
14:06:19  * mmalecki[zzz]changed nick to mmalecki
14:10:14  * elijah-mbpjoined
14:14:58  <bnoordhuis>mmalecki: were you involved with https://github.com/joyent/libuv/pull/322 ?
14:16:10  <mmalecki>bnoordhuis: a bit, testing and some random advices
14:16:29  <mmalecki>bnoordhuis: why?
14:16:33  <mmalecki>bnoordhuis: also, hello!
14:17:09  <bnoordhuis>mmalecki: hey :)
14:17:26  <bnoordhuis>i had a question about the use of mmap() but i see why it's done like that
14:17:47  <mmalecki>bnoordhuis: oh, yeah, to get PID of the second child
14:18:39  <bnoordhuis>the other thing is that i don't quite understand why it needs to fork twice
14:18:52  <bnoordhuis>it's not a big issue but it makes debugging with strace and gdb harder
14:19:10  <mmalecki>bnoordhuis: iirc, it's because after first fork setsid wouldn't work as it should work
14:19:24  <bnoordhuis>oh, did it fail with EPERM?
14:20:02  <mmalecki>bnoordhuis: no, I don't think it failed, it didn't set first child as process group leader at all
14:20:25  <mmalecki>AvianFlu: hey, you up?
14:20:26  <bnoordhuis>hm, that's odd
14:21:01  <mmalecki>bnoordhuis: a bit. but two forks matches the semantics, imo
14:22:09  <mmalecki>bnoordhuis: AvianFlu should be awake in few hours
14:22:20  <mmalecki>bnoordhuis: I have to run, throwing a birthday party today
14:22:44  <bnoordhuis>mmalecki: whose birthday is it?
14:22:55  <mmalecki>bnoordhuis: mine
14:23:05  <bnoordhuis>congrats! 16 already!
14:23:13  <mmalecki>bnoordhuis: lol
14:23:17  <bnoordhuis>my, how they grow
14:25:28  <indutny>bnoordhuis: kids
14:26:02  <mmalecki>bnoordhuis: ok, so if anything, shoot me an e-mail, maciej at notimplemented.org or maciej at nodejitsu.com
14:26:22  <mmalecki>talk to you guys tomorrow
14:26:45  * mmaleckichanged nick to mmalecki[party]
15:30:00  * pfox___joined
15:37:32  <AvianFlu>bnoordhuis, without the second fork(), the first child's setsid() succeeds, but the original parent stays up
15:37:37  <AvianFlu>until you make it exit
15:37:54  <AvianFlu>I'm sure there's an event-loop related way that could be fixed, but double-fork seemed simplest, aside from the mmap part
15:38:30  <AvianFlu>also, good morning!
15:55:57  <bnoordhuis>AvianFlu: hey
15:56:25  <bnoordhuis>the original parent stays up <- what does that mean? the parent doesn't exit?
16:02:11  <AvianFlu>not until the child does, or until it's made to exit
16:02:25  <AvianFlu>i.e. it's ref count or whatever ends up still including the child until it exits
16:03:45  <AvianFlu>there might be an obvious way to stop that from happening, but it's why I ended up sticking with the double fork()
16:09:47  <bnoordhuis>AvianFlu: ah, i see what you mean
16:09:59  <bnoordhuis>i can think of two reasons why that happens
16:10:09  <bnoordhuis>i'll see if i can fix it (tomorrow)
16:10:24  <AvianFlu>sounds good
16:10:35  <AvianFlu>there's no other real reason I need two forks, aside from the fact that it's a common practice
16:17:52  * isaacsjoined
16:27:48  * bnoordhuisquit (Read error: Operation timed out)
16:31:45  * dapjoined
16:47:50  * mikeal1quit (Quit: Leaving.)
16:52:32  * mikealjoined
16:53:38  * mikealquit (Client Quit)
17:24:02  * orlandovftwjoined
17:28:00  * mikealjoined
18:00:30  <igorzi_>isaacs: does npm support tunneling through http proxy?
18:12:35  <isaacs>igorzi_: not yet.
18:12:45  <isaacs>igorzi_: npm just uses request. we should add this feature to request.
18:12:59  <isaacs>then most node users doing http client stuff would just get it automagically by default
18:13:34  <isaacs>I mentioned it to mikeal, he'd take a pull req to do it, but it's not a use case he uses, so it's not a priority
18:14:18  <igorzi_>isaacs: koichik's 'tunnel' module does that.. we could either add it to npm or make request work with it
18:14:40  * TooTallNatejoined
18:14:43  <isaacs>igorzi_: yeah, i figured it'd be worth at least figuring out a way to have those two things talk to one anotehr
18:14:47  * orlandovftwquit (Ping timeout: 276 seconds)
18:15:24  <isaacs>igorzi_: it's sort of nice that request only depends on node, but maybe figuring out some api to pass in a reference to tunnel as the proxy instead of just a proxy url, i dunno.
18:15:32  <isaacs>that could work well.
18:16:02  * dshaw_joined
18:16:28  <igorzi_>isaacs: the way tunnel works - it it creates an agent that knows how to do CONNECT
18:18:50  <isaacs>ohh, ok.
18:18:51  <isaacs>neat
18:19:02  <isaacs>i think request can take an agent argument.
18:19:32  <isaacs>want to try an npm patch to add this?
18:20:03  <igorzi_>isaacs: yep!
18:21:27  <isaacs>also, there's a new request released today, i think. might be worth updating in npm as well.
18:22:22  <isaacs>gotta run, i'll be back online in a little bit.
18:22:23  * isaacsquit (Remote host closed the connection)
18:22:42  * AndreasMadsenjoined
18:26:28  <AndreasMadsen>AvianFlu: thanks for the help with my immortal debugging issue, it now works on mac and linux. The bug was so silly -> https://github.com/AndreasMadsen/immortal/commit/72305c68a49bd7be3d8351f10beeec67a708522e
18:26:43  * AndreasMadsenchanged nick to AndreasMadsen[aw
18:26:56  <AvianFlu>there you go :D
18:27:57  <mikeal>hrm....
18:28:19  <mikeal>once you have the agent working, tell me
18:28:38  <mikeal>i'll use that agent when a proxy is set
18:37:00  * AvianFluquit (Quit: Leaving)
18:42:01  * `3rdEdenjoined
19:00:32  * mikealquit (Quit: Leaving.)
19:03:16  * mikealjoined
19:07:52  * isaacsjoined
19:16:11  * orlandovftwjoined
19:31:17  * dshaw_quit (Quit: Leaving.)
19:31:41  <igorzi_>mikeal: https://github.com/koichik/node-tunnel/blob/master/lib/tunnel.js#L13.. this gets you the agent
19:35:46  <mikeal>i added an issue on that
19:35:54  <mikeal>i need to get this in to request
19:36:36  * paddybyersquit (Quit: paddybyers)
19:38:03  * mikealquit (Quit: Leaving.)
19:46:59  * sh1mmerjoined
19:50:46  * paddybyersjoined
19:53:25  * AvianFlujoined
19:53:45  <igorzi_>mikeal: when do you think you'll have it in request?
19:54:09  * CoverSlidejoined
20:03:20  <isaacs>igorzi_: missed him :)
20:04:01  <igorzi_>isaacs: how does node it-self get request?
20:04:07  * mikealjoined
20:09:38  <isaacs>igorzi_: node itself?
20:09:42  <isaacs>igorzi_: you mean npm itself?
20:10:03  <isaacs>igorzi_: npm checks its dependencies into git
20:12:00  <igorzi_>isaacs: ok got it.. i was just thinking that if npm needs tunneling to work to get 'request' or 'tunnel' first.. then that's a problem
20:12:20  <isaacs>yeah
20:12:29  <isaacs>that's why npm bundles its deps, because it needs them to install itself.
20:13:50  <igorzi_>isaacs: yep, makes sense
20:18:18  * sh1mmerquit (Remote host closed the connection)
20:18:37  * sh1mmerjoined
20:26:27  * AndreasMadsen[awquit (Remote host closed the connection)
20:35:04  * sh1mmer_joined
20:38:16  * sh1mmer_quit (Client Quit)
20:38:57  * sh1mmerquit (Ping timeout: 240 seconds)
20:43:23  * mjr_joined
20:45:00  * mraleph1joined
20:55:21  * mjr_quit (Quit: mjr_)
21:14:23  * sh1mmerjoined
21:17:38  * sh1mmerquit (Read error: Connection reset by peer)
21:33:08  * dshaw_joined
21:36:07  * AvianFluchanged nick to food|AvianFlu
21:41:14  * AndreasMadsenjoined
21:55:18  * paddybyersquit (Quit: paddybyers)
22:00:21  * `3rdEdenquit (Quit: Linkinus - http://linkinus.com)
22:07:02  * paddybyersjoined
22:09:48  * food|AvianFluchanged nick to AvianFlu
22:10:05  * AvianFluchanged nick to ESPIPE
22:10:18  * ESPIPEchanged nick to AvianFlu
22:14:33  * paddybyersquit (Quit: paddybyers)
22:17:23  * paddybyersjoined
22:30:30  * perezdjoined
22:34:24  * paddybyersquit (Quit: paddybyers)
22:37:58  * mralephquit (Ping timeout: 260 seconds)
22:42:27  * mralephjoined
22:57:27  * pfox___quit (Quit: leaving)
23:20:53  * AndreasMadsenquit (Remote host closed the connection)
23:52:37  * mikealquit (Quit: Leaving.)