00:05:49  <CIA-99>node: ssuda master * r249c3c1 / (12 files):
00:05:50  <CIA-99>node: Avoiding unnecessary ToString() calls
00:05:50  <CIA-99>node: String::Utf8Value and String::AsciiValue constructors take Handle<Value>
00:05:50  <CIA-99>node: So no need to convert to Handle<String> - http://git.io/a_cvJg
00:10:53  * mikealjoined
00:10:53  <CIA-99>node: Igor Zinkovsky v0.6 * r0dcc433 / (lib/net.js test/simple/test-net-write-after-close.js): don't crash when queued write fails - http://git.io/_ROWSA
00:10:59  <igorzi>isaacs: ^
00:11:36  <isaacs>igorzi: sweet.
00:11:46  <isaacs>igorzi: wanna see if npm 1.1.12 fixes the window AV issue?
00:12:08  <isaacs>igorzi: it doesn't do it the clean way, but it does the up-to-1-second retry thing in all cases where fs.rename is called on windows;
00:13:19  <igorzi>isaacs: yep.. is it in v0.6 branch?
00:13:25  <isaacs>igorzi: yes
00:17:31  <igorzi>isaacs: i'm not seeing any issues so far
00:17:37  <isaacs>sweet :)
00:17:51  <igorzi>isaacs: when do you think you'll have an rc?
00:18:18  <igorzi>isaacs: i'll get the azure team to do a quick smoke test as well
00:19:31  * travis-cijoined
00:19:31  <travis-ci>[travis-ci] joyent/node#636 (v0.6 - 0dcc433 : Igor Zinkovsky): The build was fixed.
00:19:31  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/d227084...0dcc433
00:19:31  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/926387
00:19:31  * travis-cipart
00:19:56  * travis-cijoined
00:19:56  <travis-ci>[travis-ci] DylanLukes/libuv#6 (ipv46agnostic - d86889d : Dylan Lukes): The build is still failing.
00:19:56  <travis-ci>[travis-ci] Change view : https://github.com/DylanLukes/libuv/compare/6575c89...d86889d
00:19:56  <travis-ci>[travis-ci] Build details : http://travis-ci.org/DylanLukes/libuv/builds/926440
00:19:56  * travis-cipart
00:20:31  <isaacs>igorzi: writing the changelog now
00:20:36  <isaacs>should have something very soon
00:21:03  * travis-cijoined
00:21:04  <travis-ci>[travis-ci] joyent/node#635 (master - 249c3c1 : ssuda): The build is still failing.
00:21:04  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/e1199fa...249c3c1
00:21:04  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/926362
00:21:04  * travis-cipart
00:24:14  <TooTallNate>i don't think i put anything in 0.6.x this time :(
00:24:20  <TooTallNate>there goes my streak :p
00:34:44  <isaacs>igorzi: http://nodejs.org/dist/v0.6.14/node-v0.6.14-RC0.tar.gz
00:35:01  <isaacs>TooTallNate: you've been busy on master. that's fine :)
00:38:37  * orlandovftwquit (Ping timeout: 244 seconds)
00:39:29  <isaacs>igorzi: or if you prefer to pull from a branch: https://github.com/isaacs/node/commits/v0.6.14-release
00:40:07  <TooTallNate>isaacs: :)
00:46:15  * mmaleckichanged nick to mmalecki[zzz]
01:01:40  * mikealquit (Quit: Leaving.)
01:24:52  * perezdquit (Quit: perezd)
01:25:10  * pieternquit (Quit: pietern)
01:30:55  * abraxasjoined
01:31:56  * isaacsquit (Remote host closed the connection)
01:47:01  * pieternjoined
01:48:59  * dapquit (Quit: Leaving.)
01:49:06  * bnoordhuisquit (Ping timeout: 240 seconds)
01:58:11  * TooTallNatequit (Ping timeout: 246 seconds)
02:01:08  * dshaw_joined
02:01:30  * luxigojoined
02:07:46  * pieternquit (Quit: pietern)
02:10:34  * brsonquit (Quit: leaving)
02:11:33  * hij1nxquit (Quit: hij1nx)
02:17:14  * sh1mmerjoined
02:38:15  * luxigoquit (Quit: Page closed)
02:47:06  * orlandovftwjoined
03:21:40  * mikealjoined
03:27:06  * orlandovftwquit (Ping timeout: 240 seconds)
03:32:31  * seebeesjoined
03:52:11  <seebees>hey, anyone around?
04:05:22  <dylukes>me!
04:05:29  <dylukes>brb
04:21:17  * mikealquit (Quit: Leaving.)
04:22:38  * mikealjoined
04:34:59  * mikealquit (Quit: Leaving.)
04:38:29  * mikealjoined
04:53:12  * mikealquit (Quit: Leaving.)
04:54:08  * orlandovftwjoined
05:12:58  * mikealjoined
05:21:05  * isaacsjoined
05:22:29  * isaacsquit (Remote host closed the connection)
05:32:02  * orlandovftwquit (Ping timeout: 260 seconds)
05:37:14  * perezdjoined
05:43:26  * mikealquit (Quit: Leaving.)
05:47:58  * mikealjoined
05:53:55  * hij1nxjoined
05:58:27  * mikealquit (Quit: Leaving.)
06:18:17  * mikealjoined
06:18:39  * dylukesquit (Quit: Computer has gone to sleep.)
06:20:24  * mikealquit (Client Quit)
06:32:50  * mikealjoined
06:35:35  * perezdquit (Quit: perezd)
06:45:37  * paddybyersjoined
06:48:36  * dshaw_quit (Ping timeout: 240 seconds)
07:09:43  * orlandovftwjoined
07:10:34  * orlandovftwquit (Client Quit)
07:10:50  * orlandovftwjoined
07:15:19  * sh1mmerquit (Quit: sh1mmer)
07:19:33  * paddybyersquit (Quit: paddybyers)
07:42:48  * paddybyersjoined
07:50:09  * paddybyersquit (Quit: paddybyers)
08:07:55  * mmalecki[zzz]changed nick to mmalecki
08:08:36  * piscisaureus_joined
08:08:52  * paddybyersjoined
08:17:33  * rendarjoined
08:18:33  * paddybyersquit (Quit: paddybyers)
08:47:50  * paddybyersjoined
08:50:51  * paddybyers_joined
08:52:27  * mmaleckichanged nick to mmalecki[away]
08:54:10  * paddybyersquit (Ping timeout: 260 seconds)
08:54:11  * paddybyers_changed nick to paddybyers
09:02:43  * dvvjoined
09:03:10  <dvv>Hi!
09:03:20  <piscisaureus_>hi
09:03:50  <dvv>Wonder what else but malloc() can be used for short-living memory allocations, say write_req etc.
09:05:19  <piscisaureus_>dvv: well, it depends
09:05:34  <piscisaureus_>dvv: if you only ever have one of them at the same time, you can use a static allocation
09:05:41  <piscisaureus_>(what many of the tests do)
09:06:41  <piscisaureus_>dvv: or you can use a stack var if you are sure the stack frame will never be left (but that almost never happens with libuv)
09:06:54  <dvv>yes. but let's say i need multiple.
09:07:04  <dvv>you mean alloca()?
09:07:08  <piscisaureus_>dvv: you could also use a freelist or something if you want to crack down on memory allocations, optionally combined with a stack allocator.
09:07:32  <piscisaureus_>dvv: I would not recomment alloca.
09:07:35  <piscisaureus_>*recommend
09:07:50  <dvv>i see. are you aware of such advanced allocation libs?
09:07:59  <piscisaureus_>Ok I misphrased that one.
09:08:00  <piscisaureus_>[10:06] <piscisaureus_> dvv: you could also use a freelist or something if you want to crack down on memory allocations, optionally combined with a stack allocator.
09:08:04  <piscisaureus_>should be slab allocator
09:08:17  <piscisaureus_>dvv: no, but a freelist is pretty trivial to implement
09:08:36  <dvv>not gonna be a part of libuv someday?
09:08:44  <piscisaureus_>no
09:09:14  <piscisaureus_>dvv: you should also be aware that cracking down on mallocs() is sort of an end of the line optimization
09:09:26  <piscisaureus_>not something you'd want to start with
09:09:52  <dvv>right
09:11:16  <piscisaureus_>dvv: a freelist allocator is here: https://github.com/joyent/libuv/blob/master/test/benchmark-pump.c#L294-326
09:11:30  <piscisaureus_>dvv: it is suitable for uv_write_t and uv_connect_t requests
09:11:38  <piscisaureus_>oh wait - for any req :-)
09:12:46  <dvv>great. thanks!
09:16:59  * paddybyersquit (Quit: paddybyers)
09:25:42  <indutny>piscisaureus_: is that for stubs?
09:25:51  <indutny>piscisaureus_: ah, no
09:27:45  * orlandovftwquit (Ping timeout: 248 seconds)
09:40:52  <dvv>wonder if suggested_size passed to on_alloc callback to uv_read_start() has practical limit?
09:41:35  <dvv>can't be > tcp rcvbuf length?
09:42:00  * paddybyersjoined
09:47:46  * txdvjoined
09:49:26  <piscisaureus_>dvv: I think we mostly request 65k
09:49:51  <piscisaureus_>dvv (which is the maximum size of a tcp packet)
09:50:09  <piscisaureus_>dvv: and if you supply it with less than requested that is usually fine
09:50:36  <piscisaureus_>dvv: (exceptions are udp - if the buffer is too small the packet might be truncated, and line-buffered stdin)
09:51:35  <dvv>right. grepped the sources. neat. with the help of freelists i should be fine.
10:06:08  <dvv>where can i look at POC code for https://github.com/joyent/libuv/blob/master/src/unix/stream.c#L185-187 , if ever
10:08:45  * mmalecki[away]changed nick to mmalecki
10:09:26  <piscisaureus_>dvv: POC?
10:09:40  <dvv>sorry, proof-of-concept
10:19:18  <piscisaureus_>dvv: https://github.com/joyent/node/blob/v0.4/lib/net.js#L928-935
10:21:31  <dvv>i see. thank you
10:26:42  * abraxasquit (Remote host closed the connection)
10:28:49  * rendarquit
10:52:34  * mmaleckichanged nick to mmalecki[away]
10:58:22  * rendarjoined
12:37:10  * skomskijoined
12:51:09  * skomskiquit (Quit: skomski)
13:37:24  * dvvquit (Remote host closed the connection)
13:56:06  * piscisaureus_quit (Ping timeout: 240 seconds)
14:00:23  * piscisaureus_joined
14:03:57  * piscisaureus__joined
14:04:00  * piscisaureus_quit (Read error: No route to host)
14:04:11  * piscisaureus__changed nick to piscisaureus_
14:07:48  * isaacsjoined
14:16:49  * piscisaureus_quit (Ping timeout: 248 seconds)
14:19:38  * piscisaureus_joined
14:27:17  * piscisaureus_quit (Ping timeout: 245 seconds)
14:31:05  * piscisaureus_joined
14:31:08  * seebees1joined
14:36:04  <seebees1>Does anyone have a second? Or is it too early for stupid questions?
14:38:56  * piscisaureus_quit (Read error: Connection reset by peer)
14:39:14  * piscisaureus_joined
14:39:25  * mmalecki[away]changed nick to mmalecki
14:48:25  <piscisaureus_>isaacs: rmdir <link>
14:48:43  <isaacs>piscisaureus_: that removes the contents.
14:48:50  <piscisaureus_>isaacs: it shouldn't
14:48:58  <piscisaureus_>isaacs: are you using windows xp ?
14:49:11  <isaacs>ha! no, it doesn't.
14:49:26  <isaacs>unlink and del traverse it
14:49:32  <isaacs>er, unlink gives you EPERM
14:49:50  <piscisaureus_>isaacs: yeah, I have known this for a long time :-)
14:50:04  <piscisaureus_>isaacs: on windows you use rmdir to remove directory symlinks and not unlink
14:50:19  <piscisaureus_>for great unix compatibility we should probably support unlink
14:50:22  <isaacs>can we make it work the unix way in node?
14:50:30  <isaacs>also, lstat needs to say that it's a symlink.
14:50:39  <piscisaureus_>isaacs: I think it does nowadays
14:50:48  <piscisaureus_>isaacs: I added that in 0.6.11 or so
14:51:38  <piscisaureus_>oh hmm
14:51:40  <piscisaureus_>it doesn't
14:51:54  <isaacs>nope
14:52:06  <isaacs>as of 0.7.6 it's still claiming to be a regular dir
14:52:22  <piscisaureus_>isaacs: I think lstat doesn't even lstat atm :(
14:52:49  <isaacs>well, considering that isSymbolicLink() is the only difference between stat and lstat, yeah ;)
14:53:22  <piscisaureus_>isaacs: no the problem is that we should open with CREATEFILE_OPEN_REPARSE_POINT (or whatever the thing is named)
14:53:26  <piscisaureus_>I think we don't atm
14:55:50  <piscisaureus_>isaacs: ah, crap
14:55:58  <piscisaureus_>isaacs: stat and lstat is the same code path on windows
14:56:24  <isaacs>piscisaureus_: how hard to fix? can we get this in 0.7 reasonably easily?
14:58:06  <piscisaureus_>isaacs: it's easy
14:58:46  <isaacs>k
14:58:51  * pfox___joined
14:59:02  <isaacs>you get a chance to test 0.6.14?
14:59:11  <isaacs>i forgot to update the AUTHORS, so there'll be an RC1
14:59:17  <isaacs>but i haven't found any other problems yet.
14:59:35  <piscisaureus_>isaacs: the major problem is that windows lacks _S_ISLNK
14:59:41  <piscisaureus_>(the constant :-()
14:59:47  <piscisaureus_>isaacs: yeah probably
14:59:53  <piscisaureus_>isaacs: you have an rc?
14:59:59  <piscisaureus_>(I prefer a tag btw)
15:00:04  <isaacs>http://nodejs.org/dist/v0.6.14/node-v0.6.14-RC0.tar.gz
15:00:28  <isaacs>oh, or http://github.com/isaacs/node/commits/v0.6.14-release
15:00:46  <isaacs>about to update authors and force-push there.
15:01:36  <piscisaureus_>ok, I'm already testing from the tarball
15:01:47  <seebees1>I'm trying to access the process object in C. But my C skills a weak. Can anyone point me to an example that might help me do something like this: https://github.com/seebees/node-1/commit/48a192e8180bb2045f3e78742e27755f581e5416#L0R57
15:04:13  <piscisaureus_>seebees1: I assume that doesn't compile?
15:04:22  <seebees1>yes
15:04:42  <seebees1>because process is not valid in handle_wrap.cc
15:05:02  <seebees1>the other stuff works fine
15:05:38  <piscisaureus_>seebees: You probably need to look it up via the global object
15:05:57  <piscisaureus_>or you would have to make sure that node.cc shares the process object
15:08:25  <piscisaureus_>seebees:
15:08:25  <piscisaureus_>node.cc: `Persistent<Object> process;` <-- remove the "static" part
15:08:25  <piscisaureus_>handle_wrap.cc: `extern Persistent<Object> process;` <-- add that
15:09:14  <seebees1>piscisaureus_: excellent! you are as fast a google
15:09:57  <seebees1>piscisaureus_: I'm assuming that would not fly in the long run, but it should allow me to find out if this is even a good idea...
15:13:28  <piscisaureus_>isaacs: 8 failures
15:13:34  <piscisaureus_>isaacs: questionable failures:
15:13:51  <piscisaureus_>test-http-get-pipeline-problem
15:13:57  <isaacs>piscisaureus_: 8 is an improvement over 0.6.13, i believe
15:14:01  <piscisaureus_>test-net-pipe-connect-errors
15:14:08  <piscisaureus_>test-net-write-slow
15:14:28  * hij1nxquit (Quit: hij1nx)
15:14:41  <piscisaureus_>test-dns also failed but that is because my ISP doesn't support ipv6 (we should probably disable that test)
15:15:22  <isaacs>piscisaureus_: i think the dns test may be disabled in master win32 already, i'm not sure
15:15:39  <piscisaureus_>isaacs: I think test-net-write-slow is really broken\
15:16:03  <piscisaureus_>I get
15:16:08  <piscisaureus_>timers.js:96
15:16:08  <piscisaureus_> if (!process.listeners('uncaughtException').length) throw e;
15:16:20  <isaacs>piscisaureus_: that's a throw in a timeout
15:17:22  <isaacs>piscisaureus_: what's the rest of the stack?
15:17:49  <piscisaureus_>isaacs:
15:17:50  <piscisaureus_>AssertionError: "flushed: 2, received: 181040/2000000" "undefined"
15:17:50  <piscisaureus_> at Socket.<anonymous> (D:\node-v0.6.14\test\simple\test-net-write-slow.js:37:12)
15:17:50  <piscisaureus_> at Socket.emit (events.js:64:17)
15:17:50  <piscisaureus_> at Socket._onTimeout (net.js:153:8)
15:17:50  <piscisaureus_> at Timer.ontimeout (timers.js:94:19)
15:18:47  <piscisaureus_>igorzi: yt?
15:20:21  <piscisaureus_>isaacs: actually, neither of these tests are actually regressions
15:20:36  <piscisaureus_>I tested with 0.6.13-pre and got the same errors
15:20:49  <isaacs>kewl
15:21:49  * bnoordhuisjoined
15:24:45  * pfox___quit (Remote host closed the connection)
15:24:57  <piscisaureus_>meh
15:25:22  <piscisaureus_>isaacs: it seems that we cannot make lstat work on windows because we have to break libuv's binary interface
15:25:33  <piscisaureus_>isaacs: so it can only go into 0.8
15:26:10  <isaacs>piscisaureus_: totally ok with that.
15:26:21  <isaacs>i'd prefer not to do it in 0.6 for philosophical reasons anyway
15:26:34  <isaacs>it's a change to semantics
15:28:58  * pfox___joined
15:39:59  * skomskijoined
15:41:24  * dylukesjoined
15:48:57  * stephankjoined
15:50:13  * k-s[AWAY]changed nick to k-s
16:00:18  * hij1nxjoined
16:00:50  <piscisaureus_>isaacs: it's not difficult but it turns out to be a little more complicated than I thought
16:01:05  <isaacs>k
16:01:17  <piscisaureus_>more work, mostly
16:01:37  <isaacs>piscisaureus_: can you post an issue with some thoughts or what needs to be done there? i mean, if you dont' feel like doing it right now, that is :)
16:01:48  <isaacs>piscisaureus_: then maybe someone else will fix it for you.
16:02:17  <isaacs>piscisaureus_: a bunch of people want npm link, so if i had some link to give them and say "go do what this says" then i bet they would.
16:03:23  * skomskiquit (Quit: skomski)
16:04:38  * AndreasMadsenjoined
16:07:14  <piscisaureus_>ok
16:09:37  * pieternjoined
16:13:08  <CIA-99>node: Shigeki Ohtsu master * r2cf5f04 / doc/api/tls.markdown : doc: add cleartextStream.getCipher() in tls - http://git.io/Jo70dA
16:13:08  <CIA-99>node: Shigeki Ohtsu master * r8727e5f / test/simple/test-tls-getcipher.js : test: add test of cleartextStream.getCipher() in tls - http://git.io/7LMoCg
16:13:09  <CIA-99>node: Shigeki Ohtsu master * r75face6 / doc/api/tls.markdown : doc: fix TLS cipher names - http://git.io/XBrQOw
16:17:29  <piscisaureus_>isaacs: https://github.com/joyent/libuv/issues/355
16:19:29  * orlandovftwjoined
16:19:34  * dapjoined
16:22:06  * orlandovftwquit (Client Quit)
16:24:55  <CIA-99>node: Shigeki Ohtsu master * r4515987 / doc/api/dgram.markdown : doc: fix default dgram multicast ttl to 1 - http://git.io/00rtRQ
16:29:51  * travis-cijoined
16:29:51  <travis-ci>[travis-ci] joyent/node#637 (master - 75face6 : Shigeki Ohtsu): The build is still failing.
16:29:51  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/249c3c1...75face6
16:29:51  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/931646
16:29:51  * travis-cipart
16:30:06  * AndreasMadsenquit (Remote host closed the connection)
16:38:40  <piscisaureus_>isaacs: there are broken links in the docs.
16:39:22  <piscisaureus_>isaacs: like there is a link to fs.Stats in the fs documentation but it does not work
16:39:41  <piscisaureus_>is this an issue with the generator or with the markdown files?
16:40:31  * travis-cijoined
16:40:31  <travis-ci>[travis-ci] joyent/node#638 (master - 4515987 : Shigeki Ohtsu): The build is still failing.
16:40:31  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/75face6...4515987
16:40:31  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/931742
16:40:31  * travis-cipart
16:42:10  <isaacs>piscisaureus_: with the markdown
16:42:14  <isaacs>piscisaureus_: i'll fix it, thanks
16:42:26  <piscisaureus_>isaacs: I can do it myself, but it's nice to know where to start looking
16:42:37  <isaacs>wll, i wanna get it in the release.
16:42:39  <isaacs>i'm on it
16:42:50  <piscisaureus_>isaacs: I also remember that there was something wrong with the http.get documentation
16:43:27  <isaacs>piscisaureus_: wait, which file has that broken link?
16:44:11  <isaacs>all the links i see point to http://nodejs.org/docs/v0.6.14/api/fs.html#fs_class_fs_stats
16:44:38  <piscisaureus_>isaacs: nodejs.org/docs/latest/api/fs.html#fs.Stat
16:44:53  <isaacs>piscisaureus_: where do you see a link to that?
16:44:57  <isaacs>piscisaureus_: that's the latest, not 0.6.14
16:45:01  <isaacs>i think it's already fixed, maybe
16:45:55  <isaacs>piscisaureus_: the only link i see is this one:
16:45:56  <isaacs>doc/api/fs.markdown:`stats` is a [fs.Stats](#fs_class_fs_stats) object. See the [fs.Stats](#fs_class_fs_stats)
16:46:22  <piscisaureus_>isaacs: http://nodejs.org/docs/latest/api/fs.html#fs_fs_stat_path_callback
16:46:33  <piscisaureus_>look in the description for the link to fs.stats
16:46:56  <isaacs>piscisaureus_: fixed here: http://nodejs.org/docs/v0.6.14/api/fs.html#fs_fs_stat_path_callback
16:47:04  <isaacs>piscisaureus_: those are the latest docs (ie, v0.6.13, not the new ones)
16:47:34  <isaacs>piscisaureus_: * 90b785c Rod Vagg doc: fix # links from (and within) api/fs (7 days ago)
16:47:37  <isaacs>fixed there ^
16:49:03  <piscisaureus_>ah, right
16:50:21  <dylukes>So, there's significant amount of failing tests, regardless of platform. On OS X 10.8, I'm getting 16 failing tests out of 130.
16:50:30  <dylukes>On Windows, there are less failures, but... still.
16:57:05  * seebeesquit (Quit: Leaving.)
16:57:09  <dylukes>One potential issue could be that OS X does not actually implement fdatasync...
17:02:50  * travis-cijoined
17:02:50  <travis-ci>[travis-ci] DylanLukes/libuv#7 (ipv46agnostic - 9b21617 : Dylan Lukes): The build is still failing.
17:02:50  <travis-ci>[travis-ci] Change view : https://github.com/DylanLukes/libuv/compare/d86889d...9b21617
17:02:50  <travis-ci>[travis-ci] Build details : http://travis-ci.org/DylanLukes/libuv/builds/932059
17:02:50  * travis-cipart
17:06:23  * skomskijoined
17:09:40  * mikealquit (Quit: Leaving.)
17:12:37  * mikealjoined
17:12:43  * travis-cijoined
17:12:43  <travis-ci>[travis-ci] DylanLukes/libuv#8 (ipv46agnostic - a8e6287 : Dylan Lukes): The build is still failing.
17:12:43  <travis-ci>[travis-ci] Change view : https://github.com/DylanLukes/libuv/compare/9b21617...a8e6287
17:12:43  <travis-ci>[travis-ci] Build details : http://travis-ci.org/DylanLukes/libuv/builds/932160
17:12:43  * travis-cipart
17:15:21  * skomskiquit (Read error: Connection reset by peer)
17:15:28  * skomskijoined
17:17:00  * travis-cijoined
17:17:00  <travis-ci>[travis-ci] DylanLukes/libuv#9 (ipv46agnostic - a59ba61 : Dylan Lukes): The build is still failing.
17:17:00  <travis-ci>[travis-ci] Change view : https://github.com/DylanLukes/libuv/compare/a8e6287...a59ba61
17:17:00  <travis-ci>[travis-ci] Build details : http://travis-ci.org/DylanLukes/libuv/builds/932192
17:17:00  * travis-cipart
17:22:22  <piscisaureus_>dylukes: can you turn off the irc notifications for your travis thing?
17:22:43  <dylukes>I'm trying to figure out how right now.
17:22:51  <dylukes>ah, derp. found it.
17:26:12  * skomskiquit (Quit: skomski)
17:29:25  * dshaw_joined
17:37:01  * TooTallNatejoined
17:46:22  * mikealquit (Quit: Leaving.)
17:48:54  <piscisaureus_>I am off. bye all.
17:49:07  <mmalecki>see you piscisaureus_
17:49:10  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
17:51:33  * dshaw_quit (Quit: Leaving.)
17:51:48  * dshaw_joined
17:51:51  <TooTallNate>damnit
17:52:01  <TooTallNate>i wanted him to try out that windows terminal patch
17:56:45  * dylukesquit (Quit: Computer has gone to sleep.)
17:58:14  * mikealjoined
18:08:35  * piscisaureus_joined
18:09:12  * indutnychanged nick to indutny_away
18:14:22  * orlandovftwjoined
18:14:22  * orlandovftwquit (Client Quit)
18:14:33  * orlandovftwjoined
18:16:34  <CIA-99>node: isaacs v0.6.14-release * r76f31fa / (6 files in 5 dirs): Remove hard-coded version number from docs - http://git.io/WAILzQ
18:16:34  <CIA-99>node: isaacs v0.6.14-release * rd497bf8 / (5 files in 4 dirs): doc: Remove extraneous index.html's from hyperlinks - http://git.io/1NXgYQ
18:16:36  <CIA-99>node: isaacs v0.6.14-release * re513ffe / (AUTHORS ChangeLog src/node_version.h): (log message trimmed)
18:16:36  <CIA-99>node: 2012.03.22 Version 0.6.14 (stable)
18:16:36  <CIA-99>node: * net: don't crash when queued write fails (Igor Zinkovsky)
18:16:36  <CIA-99>node: * sunos: fix EMFILE on process.memoryUsage() (Bryan Cantrill)
18:16:36  <CIA-99>node: * crypto: fix compile-time error with openssl 0.9.7e (Ben Noordhuis)
18:16:36  <CIA-99>node: * unix: ignore ECONNABORTED errors from accept() (Ben Noordhuis)
18:16:37  <CIA-99>node: * Add UV_ENOSPC and mappings to it (Bert Belder)
18:17:20  * dylukesjoined
18:17:52  <CIA-99>node: isaacs v0.6 * rd497bf8 / (5 files in 4 dirs): doc: Remove extraneous index.html's from hyperlinks - http://git.io/1NXgYQ
18:17:54  <CIA-99>node: isaacs v0.6 * ra7dce47 / src/node_version.h : Now working on v0.6.15 - http://git.io/xP80uA
18:17:56  <CIA-99>node: isaacs v0.6 * re513ffe / (AUTHORS ChangeLog src/node_version.h): (log message trimmed)
18:17:56  <CIA-99>node: 2012.03.22 Version 0.6.14 (stable)
18:17:56  <CIA-99>node: * net: don't crash when queued write fails (Igor Zinkovsky)
18:17:56  <CIA-99>node: * sunos: fix EMFILE on process.memoryUsage() (Bryan Cantrill)
18:17:56  <CIA-99>node: * crypto: fix compile-time error with openssl 0.9.7e (Ben Noordhuis)
18:17:56  <CIA-99>node: * unix: ignore ECONNABORTED errors from accept() (Ben Noordhuis)
18:17:57  <CIA-99>node: * Add UV_ENOSPC and mappings to it (Bert Belder)
18:17:57  <CIA-99>node: isaacs v0.6 * r76f31fa / (6 files in 5 dirs): Remove hard-coded version number from docs - http://git.io/WAILzQ
18:17:58  <CIA-99>node: isaacs v0.6 * rdbd0e2b / (9 files in 6 dirs): Merge branch 'v0.6.14-release' into v0.6 - http://git.io/riS8_A
18:18:30  * sh1mmerjoined
18:21:51  * piscisaureus_quit (Ping timeout: 246 seconds)
18:22:41  <igorzi>isaacs: rc0 is looking good so far
18:23:25  <isaacs>igorzi: i'm pushing rc1
18:23:28  <isaacs>it's good everywhere
18:24:03  <igorzi>isaacs: what's the delta?
18:25:27  <isaacs>igorzi: I'd forgotten to upate the AUTHORS file
18:25:29  <isaacs>nothign relevant
18:25:34  * travis-cijoined
18:25:34  <travis-ci>[travis-ci] joyent/node#639 (v0.6.14-release - e513ffe : isaacs): The build passed.
18:25:34  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/76f31fa^...e513ffe
18:25:34  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/932817
18:25:34  * travis-cipart
18:26:05  * travis-cijoined
18:26:05  <travis-ci>[travis-ci] joyent/node#640 (v0.6 - a7dce47 : isaacs): The build passed.
18:26:05  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/0dcc433...a7dce47
18:26:05  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/932824
18:26:05  * travis-cipart
18:26:46  <isaacs>oh, whoops, i left the date as 3/22
18:26:48  <isaacs>oh well.
18:27:36  <TooTallNate>igorzi: hey do you wanna try out this patch on cmd.exe? https://github.com/joyent/node/pull/2985
18:28:00  <TooTallNate>igorzi: i just tried on osx and windows xp's cmd.exe, it seems good
18:28:44  <dylukes>:\
18:28:53  <igorzi>TooTallNate: yep, will try it out
18:29:01  <dylukes>Is there any way to run a specific test?
18:29:13  <dylukes>That is, if I don't want to run the whole suite, and I just want to focus on one test.
18:29:34  <dylukes>Wait. isaacs.
18:29:39  <dylukes>the build *passed* o__o
18:29:42  <dylukes>did you fix all of the tests?
18:29:43  <isaacs>:)
18:29:49  <isaacs>dylukes: it passes occasionally on linux
18:29:52  <dylukes>Oh heh.
18:29:54  * AndreasMadsenjoined
18:29:54  <isaacs>dylukes: rarely on os x
18:29:56  <isaacs>never on windows
18:29:59  <dylukes>I'm getting +114 -16 on OS X
18:30:09  <dylukes>And +132 -4 on Win
18:30:13  <isaacs>dylukes: you're testing libuv or node?
18:30:20  <dylukes>Oh, libuv, herp.
18:30:31  <dylukes>I thought that was libuv for a moment hehe ^^.
18:30:54  <dylukes>I'd like to start fixing tests on OS X and Windows here but, I worry I'll end up cramming *too many changes* into my fork.
18:31:01  * sh1mmerquit (Quit: sh1mmer)
18:31:01  <dylukes>And then merging it back in later will be a bit of a pita.
18:31:37  <creationix>dylukes, just use a branch for each feature
18:31:41  <igorzi>dylukes: you can run specific tests with run-tests[.exe] test-name
18:31:46  <dylukes>igorzi: excellent.
18:31:52  <dylukes>creationix: indeed, but,
18:31:57  <dylukes>since I changed the API, it's a bit harder.
18:32:26  <dylukes>I could fix them in the existing master,
18:32:39  <dylukes>but then they'll have to be changed when I get my changes pulled it :\.
18:34:11  <dylukes>It seems I managed to break udp_send_and_recv in my changes though. So I'll approach that first...
18:34:50  * brsonjoined
18:38:42  * coderarityjoined
18:42:15  * dapquit (Quit: Leaving.)
18:42:33  * dapjoined
18:51:18  <dylukes>What exactly does uv__set_sys_error do?
18:51:46  <dylukes>In some places I see it being set to the negaitve of a requests status... seems odd?
19:08:22  <dylukes>Agh. What is this.
19:08:22  * `3rdEdenjoined
19:08:39  <dylukes>req->status is -22... but then you set the syserror to the negative of that (22)
19:08:56  <dylukes>so, it's EINVAL but... are you just keeping the status as a negative so it's easy to check for any error
19:08:58  <dylukes>?
19:09:04  <dylukes>Why not use != 0 :\?
19:17:50  * dapquit (Quit: Leaving.)
19:19:34  <dylukes>Hm... so I broke something with the server returning the client's message. Awesome >_>.
19:22:10  <dylukes>oh... great. So perhaps it's just the test was *wrong in the first place*. Whoo!
19:23:04  <dylukes>hahahaha... yeaaaah
19:23:07  <dylukes>It was...
19:27:31  <dylukes>So, sendto() does not like being given a sockaddr_storage.
19:27:44  <dylukes>But I'm wondering, would libuv be converting into a sockaddr_storage at some point?
19:29:42  * dshaw_quit (Quit: Leaving.)
19:32:21  * dshaw_joined
19:46:27  * sh1mmerjoined
19:48:36  * sh1mmerquit (Client Quit)
19:49:09  * sh1mmerjoined
19:49:34  * mikealquit (Quit: Leaving.)
20:14:30  * seebeesjoined
20:14:44  * AndreasMadsenquit (Remote host closed the connection)
20:16:27  * AndreasMadsenjoined
20:22:16  * mikealjoined
20:44:18  * mikealquit (Quit: Leaving.)
20:47:21  <bnoordhuis>isaacs: i don't quite understand how brian's patch fixes process.memoryUsage()
20:47:36  <bnoordhuis>is there some kind of bug / limitation in fopen / read on sunos?
20:51:46  * `3rdEdenquit (Read error: Connection reset by peer)
20:51:50  * V1joined
20:52:45  * isaacs_mobilejoined
20:54:19  * V1changed nick to `3rdEden
20:55:24  * sh1mmerquit (Quit: sh1mmer)
21:11:14  <ryah>mjr_: did you try my patch?
21:11:40  * pfox___quit (Quit: leaving)
21:14:17  * isaacs_mobilequit (Remote host closed the connection)
21:23:20  * `3rdEdenquit (Quit: Leaving...)
21:25:52  * AndreasMadsenquit (Remote host closed the connection)
21:27:31  * AndreasMadsenjoined
21:34:38  * piscisaureus_joined
21:47:15  * pfox___joined
21:57:22  * pfox___quit (Ping timeout: 244 seconds)
21:59:13  * pfox___joined
22:00:08  * mikealjoined
22:03:27  * pfox___quit (Ping timeout: 244 seconds)
22:04:32  * AndreasMadsenquit (Remote host closed the connection)
22:05:37  * pfox___joined
22:08:36  * mikealquit (Ping timeout: 240 seconds)
22:09:42  * sh1mmerjoined
22:10:52  * sh1mmerquit (Client Quit)
22:22:41  * rendarquit
22:27:17  * dylukesquit (Quit: Computer has gone to sleep.)
22:29:10  <isaacs>bnoordhuis: yes, there is.
22:29:34  <isaacs>bnoordhuis: fopen uses a uchar for the fd, apparently.
22:29:41  <bnoordhuis>hah, really?
22:29:51  <bnoordhuis>that's my wtf moment for the day right there
22:31:00  <isaacs>bnoordhuis: yep.
22:31:04  <isaacs>pretty wild, huh?
22:31:11  <isaacs>who'd ever need to open more than 256 files!?!?
22:31:17  <bnoordhuis>quite so :)
22:31:27  <isaacs>"there aren't even that many files on the whole machine!"
22:31:45  <isaacs>-- Some Forward Thinking SunOS Engineer, ca 1980-something
22:33:43  <bnoordhuis>the patch needs to be ported to libuv, i assume?
22:36:22  * dylukesjoined
22:39:23  <isaacs>bnoordhuis: IIUC, the issue is using fopen with a FILE*
22:39:49  <bnoordhuis>isaacs: yes, libuv does that too atm
22:39:57  <isaacs>bnoordhuis: there may be issues that need to be ported to libuv, i'd suspect, but i don't think we were hitting any
22:41:03  <bnoordhuis>not yet anyway but v0.8 will use libuv for platform specific stuff instead of the src/platform_*.cc files
22:41:29  <isaacs>yeah
22:41:34  <isaacs>so, yes, we need to do that there, then
22:49:20  <bnoordhuis>isaacs: https://github.com/bnoordhuis/libuv/commit/c6c8800
22:49:39  <pfox___>query: im binding uv_read_start() in rust..
22:49:51  <bnoordhuis>oh wait, let me redo that patch
22:49:58  <pfox___>and, in my tests.. my uv_read_cb is only ever called once.. it has valid data in it too, it seems
22:50:11  <pfox___>but im never getting that subsequent call where nread == -1, as in the docs...
22:50:47  <pfox___>well, i guess that meanst hat the other end of the stream isn't closing it, duh.
22:50:58  <pfox___>so i have to figure out, for myself, whether i have a complete response..?
22:51:18  <pfox___>this is in the context of a tcp request that is impl'ing a poor man's http request
22:52:29  <bnoordhuis>isaacs: https://github.com/bnoordhuis/libuv/commit/9804f32 <- that's the one
22:53:02  <bnoordhuis>pfox___: that's correct, libuv just passes on the data as it comes off the wire
22:53:19  <pfox___>yeah. this is my "learn the semantics of tcp" trail-by-fire
22:53:24  <bnoordhuis>:)
22:53:26  <pfox___>so forgive my thinking out loud :)
22:53:38  <bnoordhuis>look at joyent/http-parser if you want a neat http parser
22:53:57  <pfox___>yeah, me and another person in #rust have talked a bit about pulling it into the rust build
22:54:09  <pfox___>since it's, pretty much, the build-blocking of an http server
22:55:26  <pfox___>anywho. good times, good times.
22:56:04  <pfox___>enjoy your weekend
22:56:05  * pfox___quit (Quit: leaving)
22:57:14  <AvianFlu>bnoordhuis, I noticed something peculiar last night: https://gist.github.com/2175904
22:57:35  <AvianFlu>it looks like the errno doesn't get set somehow for the sync read on this, for some strange reason
22:58:01  <bnoordhuis>AvianFlu: plausible
22:58:11  <bnoordhuis>do you see the errno when you strace it?
22:58:30  <bnoordhuis>i can of course test that myself...
22:58:32  <AvianFlu>ha, I forgot to strace it
22:58:53  <AvianFlu>I can have a look, but I'd only gotten to the point of scratching my head about it and realizing it was already mapped properly
22:59:32  <bnoordhuis>i think know what it is
22:59:42  <bnoordhuis>read() is returning EIO and that's not mapped in libuv
23:01:04  <AvianFlu>bnoordhuis, very possible - but also, this seems like it's not the case outside the repl
23:01:16  <AvianFlu>I just made a script to strace, and the script throws ESRCH like I (sort of) expect
23:01:26  <bnoordhuis>AvianFlu: try this -> strace -e open,read out/Release/node -pe 'require("fs").readFileSync("/proc/self/mem")'
23:01:37  <AvianFlu>ok, just a sec
23:02:25  <AvianFlu>there we go, that hit the unknown, gisting
23:02:39  <AvianFlu>https://gist.github.com/2176136
23:02:42  <AvianFlu>you're right, EIO
23:02:57  <bnoordhuis>thanks
23:03:00  * `3rdEdenjoined
23:03:18  <bnoordhuis>yes... i told someone a couple of days ago that we're not going to handle that because you should never see EIO errors :/
23:05:41  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
23:06:05  <AvianFlu>funny how that works sometimes
23:06:45  <bnoordhuis>i wonder what the windows equivalent is, EIO is like the kernel throwing up its hands, saying "i don't know either"
23:07:14  <dylukes>What function are you getting it from?
23:07:27  <bnoordhuis>read()
23:07:46  <dylukes>EIO can show up there when there's legitimately an issue reading from say, a disk.
23:08:04  <bnoordhuis>scroll up - it's a special file :)
23:08:24  <AvianFlu>yeah, I was trying to be hacky and get the memory usage from the os - that part is a personal problem
23:08:33  <AvianFlu>but I figured unknown errors aren't intended behavior :D
23:08:44  <dylukes>my man pages give three possible causes
23:08:56  <dylukes>1) File system error...
23:09:01  <dylukes>2) Process group is orphaned.
23:09:05  <dylukes>3) The file is a regular file, nbyte is greater than 0, the starting position is before the end-of-file, and
23:09:05  <dylukes> the starting position is greater than or equal to the offset maximum established for the open file
23:09:05  <dylukes> descriptor associated with fildes.
23:09:40  <dylukes>But... it's not a regular file heh.
23:11:02  <bnoordhuis>AvianFlu: i've opened an issue - https://github.com/joyent/libuv/issues/356
23:12:18  <AvianFlu>bnoordhuis, cool, thanks
23:13:01  <bnoordhuis>dylukes: never trust man pages, always read the source
23:13:20  <bnoordhuis>the linux kernel raises EIO for a whole lot of edge cases
23:13:26  <dylukes>I'm on OS X mind you.
23:13:41  <bnoordhuis>oh. i'm sorry
23:14:10  <dylukes>But, I'd assume it might do something similar.
23:31:11  * `3rdEdenquit (Quit: Linkinus - http://linkinus.com)
23:44:23  * mmaleckiquit (Ping timeout: 260 seconds)
23:51:23  * hij1nxquit (*.net *.split)
23:51:23  * demarchiquit (*.net *.split)
23:51:23  * ryahquit (*.net *.split)
23:51:23  * coderarityquit (*.net *.split)
23:51:24  * Raynosquit (*.net *.split)
23:51:24  * mrb_bkquit (*.net *.split)
23:51:24  * sj26quit (*.net *.split)
23:51:24  * arlolraquit (*.net *.split)
23:51:24  * ircretaryquit (*.net *.split)
23:51:24  * toothrquit (*.net *.split)
23:51:24  * saghulquit (*.net *.split)
23:51:24  * dylukesquit (*.net *.split)
23:51:24  * seebeesquit (*.net *.split)
23:51:24  * AvianFluquit (*.net *.split)
23:51:24  * isaacsquit (*.net *.split)
23:51:25  * orlandovftwquit (*.net *.split)
23:51:25  * dshaw_quit (*.net *.split)
23:51:25  * DrPizzaquit (*.net *.split)
23:51:25  * elijah-mbpquit (*.net *.split)
23:51:25  * chiltsquit (*.net *.split)
23:51:25  * igorziquit (*.net *.split)
23:51:25  * avsejquit (*.net *.split)
23:51:26  * txdvquit (*.net *.split)
23:51:26  * mjr_quit (*.net *.split)
23:51:26  * jcequit (*.net *.split)
23:51:26  * mordy_quit (*.net *.split)
23:51:26  * pieternquit (*.net *.split)
23:51:26  * stephankquit (*.net *.split)
23:51:26  * CoverSlidequit (*.net *.split)
23:51:26  * brsonquit (*.net *.split)
23:51:26  * bnoordhuisquit (*.net *.split)
23:51:26  * pquernaquit (*.net *.split)
23:51:26  * k-squit (*.net *.split)
23:51:27  * einarosquit (*.net *.split)
23:51:27  * russell_hquit (*.net *.split)
23:51:27  * CIA-99quit (*.net *.split)
23:51:27  * tjfontainequit (*.net *.split)
23:51:27  * indutny_awayquit (*.net *.split)
23:51:27  * creationixquit (*.net *.split)
23:51:27  * ljacksonquit (*.net *.split)
23:51:27  * TooTallNatequit (*.net *.split)
23:51:27  * paddybyersquit (*.net *.split)
23:51:27  * benviequit (*.net *.split)
23:51:27  * rphillipsquit (*.net *.split)
23:51:27  * philipsquit (*.net *.split)
23:51:28  * seebees1quit (*.net *.split)
23:51:28  * kohaiquit (*.net *.split)