00:02:58  <bnoordhuis>isaacs: where did ircetrary go?
00:03:27  * ircretaryjoined
00:03:29  <isaacs>ircretary: hi
00:03:29  <ircretary>isaacs: Hello :)
00:03:35  <bnoordhuis>heh
00:03:39  <isaacs>bnoordhuis: doesn't handle netsplits
00:03:59  <bnoordhuis>ircretary: tell indutny https://gist.github.com/5f441b7eea192d3f7949 - environ is really _NSGetEnviron() and we don't control the pointer it returns
00:03:59  <ircretary>bnoordhuis: I'll be sure to tell indutny
00:04:26  <bnoordhuis>ircretary: tell indutny the fact that it didn't crash in non-root mode was purely accidental
00:04:27  <ircretary>bnoordhuis: I'll be sure to tell indutny
00:04:40  <bnoordhuis>isaacs: ^ also relevant for you
00:04:44  * perezdjoined
00:05:15  <isaacs>ah, nice
00:08:04  * chobi_echanged nick to chobi_e_
00:14:52  * chobi_e_changed nick to chobi_e
00:20:21  * beachdogquit (Remote host closed the connection)
00:20:31  * beachdogjoined
00:25:09  * chobi_echanged nick to chobi_e_
00:25:30  * dapjoined
00:30:49  * kerrickjoined
00:30:59  * EhevuTovquit (Quit: This computer has gone to sleep)
00:31:30  <kerrick>Is there any way to modify Chrome (e.g. using a plugin) so that you can call libuv routines from within it?
00:34:30  * benviejoined
00:34:48  <bnoordhuis>kerrick: you mean event loop integration?
00:37:42  <kerrick>bnoordhuis, I would like to call operating system routines (e.g. file manipulation) from within Chrome code. Actually, I'm not sure if libuv provides this.
00:38:14  <kerrick>Err, I mean, my understanding is that NodeJs is basically V8 + libuv
00:38:35  <kerrick>And libuv is the library that allows interaction with the OS
00:38:54  <kerrick>but I'm not sure that that understanding is correct; I'm just going off what's on Wikipedia
00:41:37  <bnoordhuis>kerrick: you mean you want to combine v8 + libuv to make a kind of node.js in the browser?
00:42:32  <kerrick>Yeah, more or less
00:43:05  <bnoordhuis>err, i kind of doubt chromium lets you do that
00:43:12  <kerrick>I would like to be able to write an app that uses Node's APIs to make OS calls, but renders its GUI in Chrome
00:43:15  <bnoordhuis>#chromium is probably a better place to ask that though
00:43:22  <kerrick>bnoordhuis, yeah, I asked there
00:43:40  * piscisaureus_joined
00:43:47  <kerrick>It looks like I could write a plugin to do it, but I'm just wondering if there's already a solution out there
00:43:52  <bnoordhuis>not that i know of
00:44:21  <kerrick>Building desktop apps that render their interface using a browser seems like it would be a fairly common idea
00:44:41  <kerrick>and I'd like to do it using Chrome (rather than just an embedded WebKit view) because that'll make debugging significantly easier
00:44:47  <kerrick>I'll look around; thanks
00:53:38  <mitsuhiko>anyone of you any ideas how to write a wrapper around something that takes a va_list with the intention of prefixing the format string and injecting one argument?
00:53:54  <mitsuhiko>foo("%d", 42) -> xfoo("foo %s %d", "injected", 42)
00:56:34  <bnoordhuis>mitsuhiko: vsnprintf(tmp, sizeof(tmp), fmt, ap) + snprintf(buf, sizeof(buf), "foo %s %s", "injected", tmp)
00:57:09  <mitsuhiko>bnoordhuis: well, it's actually a call into one of the redis functions and not vsnprint :)
00:57:12  <mitsuhiko>so i can't split the call up
00:57:45  <mitsuhiko>guess va_list is just unportable mess and avoid the problem
00:58:59  <bnoordhuis>i admit i don't quite follow but it sounds like you're boned
00:59:56  <mitsuhiko>bnoordhuis: be able to call foo(x, "%s:%s", a, b) and have it to redisAysncWhatever(x, "SUBSCRIBE %s:%s", a, b) basically
01:02:28  <bnoordhuis>mitsuhiko: oh, like that. vsnprintf(tmp, sizeof(tmp), fmt, ap); snprintf(buf, sizeof(buf), "prefix %s %s", "bla", tmp); redisAysncWhatever(x, "SUBSCRIBE %s", buf);
01:02:45  <bnoordhuis>that probably doesn't make much sense in the context of redis but you get the point
01:02:47  <mitsuhiko>bnoordhuis: jsut that redis %s does quoting :)
01:02:59  <mitsuhiko>but well, i will just ignore the problem
01:04:32  * kerrickquit (Ping timeout: 255 seconds)
01:04:32  * c4miloquit (Remote host closed the connection)
01:11:04  * mjr_quit (Ping timeout: 248 seconds)
01:12:40  * ericktquit (Ping timeout: 248 seconds)
01:16:29  * piscisaureus_quit (Ping timeout: 240 seconds)
01:21:36  * joeandaverdejoined
01:30:31  <isaacs>bnoordhuis: you still need access to my mac?
01:42:53  * abraxasjoined
01:48:19  <bnoordhuis>isaacs: no - already logged out
01:48:48  <ryah>building node with ninja :)
01:49:34  <ryah>it's super nice
01:49:36  <ryah>you should try it
01:50:22  <ryah>ryan@ryan-ThinkPad-X201:.../node/out/Release% touch ~/projects/node/src/node_buffer.cc
01:50:25  <ryah>ryan@ryan-ThinkPad-X201:.../node/out/Release% time ninja
01:50:28  <ryah>[2/2] LINK node
01:50:30  <ryah>ninja 0.89s user 0.14s system 97% cpu 1.052 total
01:56:43  * bnoordhuisquit (Ping timeout: 245 seconds)
02:00:37  <ryah>compare with this
02:00:39  <ryah>ryan@ryan-ThinkPad-X201:~/projects/node% touch src/node_buffer.cc
02:00:40  <ryah>ryan@ryan-ThinkPad-X201:~/projects/node% time make
02:00:40  <ryah>make -C out BUILDTYPE=Release
02:00:40  <ryah>make[1]: Entering directory `/home/ryan/projects/node/out' CXX(target) /home/ryan/projects/node/out/Release/obj.target/node/src/node_buffer.o LINK(target) /home/ryan/projects/node/out/Release/node LINK(target) /home/ryan/projects/node/out/Release/node: Finished
02:00:43  <ryah>make[1]: Leaving directory `/home/ryan/projects/node/out'
02:00:46  <ryah>ln -fs out/Release/node node
02:00:48  <ryah>make 1.29s user 0.22s system 95% cpu 1.574 total
02:00:57  <ryah>33% faster
02:01:00  <ryah>woo :)
02:09:42  * benviequit (Ping timeout: 244 seconds)
02:25:39  * ericktjoined
02:39:43  * ericktquit (Quit: erickt)
02:45:39  * kerrickjoined
03:02:31  * beachdogquit (Quit: beachdog)
03:21:18  * ericktjoined
03:23:49  * lohkeyjoined
03:25:44  * ericktquit (Client Quit)
03:38:36  * EhevuTovjoined
03:42:23  * EhevuTovquit (Client Quit)
03:43:42  <tjfontaine>ya ninja is useful
03:43:52  <tjfontaine>the chromium builds are much more tolerable
03:50:52  * TooTallNatequit (Quit: Computer has gone to sleep.)
04:06:18  * lohkeyquit (Quit: lohkey)
04:14:49  * ericktjoined
04:15:24  * dapquit (Quit: Leaving.)
04:29:29  * lohkeyjoined
05:02:55  * lohkeyquit (Quit: lohkey)
05:12:13  * beachdogjoined
05:14:24  * beachdog_joined
05:14:42  * beachdogquit (Read error: Connection reset by peer)
05:14:43  * beachdog_changed nick to beachdog
05:24:55  * ericktquit (Quit: erickt)
05:36:05  * joeandaverdequit (Quit: joeandaverde)
06:00:04  * indexzerojoined
06:00:20  * paddybyersjoined
06:11:57  * ericktjoined
06:19:43  * mjr_joined
06:32:54  * mjr_quit (Quit: mjr_)
06:40:48  * ericktquit (Quit: erickt)
07:01:22  * indexzeroquit (Read error: Connection reset by peer)
07:02:51  * indexzerojoined
07:05:50  * stephankquit (Quit: *Poof!*)
07:18:03  * EhevuTovjoined
07:30:37  * loladirojoined
07:36:20  * loladiroquit (Read error: Connection reset by peer)
07:36:24  * loladiro_joined
07:51:41  * EhevuTovquit (Quit: This computer has gone to sleep)
08:11:43  * loladiro_quit (Ping timeout: 245 seconds)
08:24:22  * theColequit (Quit: theCole)
08:25:58  * indexzeroquit (Quit: indexzero)
08:32:40  * bnoordhuisjoined
08:37:41  <indutny>morning
08:37:44  <indutny>bnoordhuis: morning
08:39:21  <indutny>bnoordhuis: _NSGetEnviron
08:39:25  <indutny>that's what I was afraid of
08:42:30  <indutny>but why are we shadowing it?
08:45:06  <indutny>bnoordhuis: it seems that it's only necessary when libuv is compiled as shared library
08:45:15  <indutny>bnoordhuis: http://developer.apple.com/library/mac/#documentation/Darwin/Reference/ManPages/man7/environ.7.html
09:03:09  <bnoordhuis>indutny: hey
09:03:51  <indutny>so
09:03:51  <bnoordhuis>environ is not always defined
09:04:05  <indutny>it's undefined only in cases where linker don't know it's address
09:04:11  <indutny>according to http://developer.apple.com/library/mac/#documentation/Darwin/Reference/ManPages/man7/environ.7.html
09:04:25  <indutny>Shared libraries and bundles don't have direct access to environ, which is only available to the loader
09:04:25  <indutny> ld(1) when a complete program is being linked.
09:04:25  <bnoordhuis>which was happening some time in the past because that's why it got added
09:04:44  <indutny>ok
09:04:54  <indutny>that's just process title
09:05:03  <indutny>why should we spend so much time on it :D
09:05:51  <indutny>I've a very important question
09:06:04  <indutny>why are we doing assertions instead of throwing js errors in node_zlib.cc
09:06:06  <indutny>oooh
09:06:12  <indutny>zlib's code is so retarded
09:25:37  * hzjoined
09:32:51  * rendarjoined
10:01:58  <indutny>bnoordhuis: what do you think about node-zlib.cc refactor?
10:02:01  <indutny>useless waste of time?
10:02:23  * AndreasMadsenjoined
10:08:59  <bnoordhuis>indutny: uhm, why?
10:09:21  <bnoordhuis>it seems compiling v8 with gdbjit=on is pretty broken
10:10:07  <indutny>for quite a long time
10:11:05  <bnoordhuis>yeah... so why don't they a) fix it, or b) remove it?
10:11:07  <indutny>bnoordhuis: basically because we're doing assertions here instead of throwing errors and allowing users to handle it
10:11:25  <indutny>bnoordhuis: well, they ain't removing it because it's useful
10:11:33  <indutny>they ain't fixing it because noone uses it atm
10:11:34  <bnoordhuis>a broken build is useful how?
10:11:34  <indutny>:D
10:11:45  <bnoordhuis>guess i'll give it a try
10:11:46  <indutny>well, it's useful in general
10:11:53  <indutny>good luck with that, man
10:12:04  <indutny>I've given up with implementing gdbjit for candor
10:12:15  <bnoordhuis>../src/gdb-jit.cc:350: error: ‘virtual bool v8::internal::ELFSection::WriteBody(v8::internal::Writer*)’ was hidden
10:12:15  <bnoordhuis>../src/gdb-jit.cc:465: error: by ‘virtual void v8::internal::StringTable::WriteBody(v8::internal::Writer::Slot<v8::internal::ELFSectionHeader>, v8::internal::Writer*)’
10:12:37  <bnoordhuis>^ that doesn't instill faith it's been updated recently...
10:13:01  <indutny>well you can run blame on it
10:13:03  <indutny>and see
10:13:18  <bnoordhuis>already did - last touched some time in 2011
10:13:40  <bnoordhuis>actually, that was a merge that broke it
10:13:41  <indutny>bnoordhuis: last commit https://github.com/v8/v8/commit/1daa81bcda15b0783a7654dd099989757157a996
10:30:32  <indutny>bnoordhuis: btw, where is Bert
10:30:43  <indutny>hadn't heard a word from him in last day
10:30:47  <indutny>s/day/days
10:30:56  <mmalecki>indutny: it's like 12:30 in our timezone
10:30:59  <mmalecki>let people sleep
10:31:22  <indutny>mmalecki: well, I'm up since 11:30 :)
10:31:26  <indutny>my TZ
10:31:33  <indutny>and I'm not calling him
10:31:57  <indutny>I hope mentioning bert on irc doesn't wake up him :)
10:32:00  <mmalecki>indutny: oh man, you're an early bird
10:32:10  <mmalecki>I got woken up by indexzero at 4 AM, lol
10:35:09  <mmalecki>indutny: btw, re that peers thing
10:35:28  <mmalecki>I was thinking about it and I want to make a connect-my-home-pcs example
10:35:44  <mmalecki>with playing music (as in, streaming and having various devices play it)
11:04:01  * kerrickquit (Read error: Connection reset by peer)
11:08:48  * theColejoined
11:12:59  * abraxasquit (Remote host closed the connection)
11:18:45  * theColequit (Quit: theCole)
11:19:14  * kerrickjoined
11:30:14  * beachdogquit (Quit: beachdog)
11:36:30  * beachdogjoined
11:56:20  * beachdogquit (Quit: beachdog)
12:21:07  <indutny>hahah
12:27:34  * beachdogjoined
12:29:17  * chobi_e_changed nick to chobi_e
12:49:31  * theColejoined
13:00:42  <AndreasMadsen>morning
13:06:30  * ikpart
13:07:03  <indutny>morning
13:07:08  <mmalecki>mornin'
13:08:34  * hzquit (Ping timeout: 272 seconds)
13:10:32  * hzjoined
13:16:22  * hzquit (Disconnected by services)
13:16:25  * hzjoined
13:22:04  <AndreasMadsen>oh, not cool, someone has registered #morning
13:22:51  <indutny>you should join #chilling
13:24:24  * hzquit (Ping timeout: 272 seconds)
13:25:33  <AndreasMadsen>Man, only one has joined. Of course that guy chill so much that he counts for at least 17 chillers.
13:32:51  * benviejoined
13:54:23  * c4milojoined
14:22:53  * AndreasMadsenquit (Remote host closed the connection)
14:30:56  * theColequit (Quit: theCole)
14:35:45  <CIA-108>libuv: Ben Noordhuis v0.8 * re3a2850 / src/unix/pipe.c : unix: fix errno reporting in uv_pipe_connect() - http://git.io/Lo2Taw
14:35:45  <CIA-108>libuv: Ben Noordhuis v0.8 * rff59525 / src/unix/pipe.c : unix: fix uv_pipe_connect() with existing fd - http://git.io/F5WT6g
14:35:45  <CIA-108>libuv: Ben Noordhuis v0.8 * r94355e4 / (src/unix/freebsd.c test/test-process-title.c): unix: fix format string vulnerability in freebsd.c - http://git.io/gsqglQ
14:38:02  * travis-cijoined
14:38:03  <travis-ci>[travis-ci] joyent/libuv#497 (v0.8 - 94355e4 : Ben Noordhuis): The build is still failing.
14:38:03  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/69a6afea63fa...94355e4718a7
14:38:03  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/1904753
14:38:03  * travis-cipart
14:57:03  * beachdogquit (Quit: beachdog)
14:57:47  * beachdogjoined
15:09:24  <CIA-108>node: isaacs v0.8.3-release * re7d5c8e / (AUTHORS ChangeLog src/node_version.h): 2012.07.17, Version 0.8.3 (Stable) - http://git.io/ky_6Cg
15:13:14  <indutny>try # 3
15:13:14  <indutny>?
15:13:18  <indutny>isaacs: morning
15:13:27  <isaacs>g'morning
15:15:37  <isaacs>bnoordhuis: was 98b1ce9 and 5a5e128 supposed to make the build work on smartos again?
15:23:58  * mjr_joined
15:29:36  * mikealjoined
15:31:08  <isaacs>indutny: what's up with proctitle.c on os x? i'm strongly considering just rolling back that change.
15:31:12  <isaacs>so we can get 0.8.3 out
15:31:24  <indutny>em... I thought we've already rolled it back
15:31:28  <indutny>at least in libuv
15:31:48  <indutny>isaacs: https://github.com/joyent/libuv/commit/69a6afea63faba510e83cefb53cf4cd77d8dcd6f
15:32:28  <indutny>I can merge uv@0.8 into node@0.8 if you want
15:32:32  <isaacs>ok, i'll pull that into node
15:32:33  <indutny>isaacs: ^
15:32:34  <bnoordhuis>isaacs: not supposed - does (at least for me)
15:32:36  <indutny>ok
15:33:02  <bnoordhuis>i managed to make gdbjit compile again but not actually work
15:33:11  <isaacs>bnoordhuis: if you log into 165.225.128.184 you can see it fail
15:33:15  <indutny>bnoordhuis: hahaha
15:33:20  <indutny>bnoordhuis: that was expected
15:33:27  <bnoordhuis>yeah, well
15:33:42  <bnoordhuis>the new ScopeInfo impl does a heap allocation at a time when it can't / shouldn't
15:33:46  <indutny>bnoordhuis: I'm quite sure that debugging descriptors has changed much
15:33:47  <bnoordhuis>kind of a bummer
15:33:58  <indutny>bnoordhuis: well, that's again was expected :P
15:34:11  <indutny>though, good try
15:34:39  <bnoordhuis>it's surprising though - v8 uses ScopeInfo a lot, presumably at times when it's not safe to alloc either
15:35:08  <bnoordhuis>isaacs: have you done a distclean?
15:35:14  <isaacs>bnoordhuis: yes
15:35:29  <isaacs>bnoordhuis: https://github.com/joyent/node/issues/3736#issuecomment-7101184
15:35:32  <bnoordhuis>then i don't know what's up - it's compiling on no.de
15:35:50  <bnoordhuis>oh, ustack
15:36:01  <bnoordhuis>that's dap's problem - compile with --without-dtrace :)
15:36:01  <isaacs>don't use no.de for this stuff.
15:36:21  <isaacs>bnoordhuis: we can't just say "no dtrace on smartos" for node
15:36:32  * ericktjoined
15:36:36  <mmalecki>are you guys seeing it on 0.8.2 too?
15:36:39  <bnoordhuis>well, i can - i didn't break it
15:36:53  <isaacs>bnoordhuis: it broke on https://github.com/joyent/node/commit/b40f813
15:36:59  <isaacs>bnoordhuis: before that it was fine
15:37:08  <isaacs>mmalecki: no, it started failing as of https://github.com/joyent/node/commit/b40f813
15:37:35  <mmalecki>okay, that'd make sense
15:37:49  <mmalecki>but no dtrace is quite a big blocker, at least for us
15:37:54  <isaacs>bnoordhuis: but yeah, no.de is not a suitable build box. the smartos drone machine has 32 cores and 8gb of ram
15:38:10  <bnoordhuis>isaacs: what happens when you compile with make CXXFLAGS+="-ffunction-sections -fdata-sections"?
15:38:52  <isaacs>trying that now
15:38:52  <bnoordhuis>i ask because i'm stil waiting for the key exchange to clear :)
15:38:57  * theColejoined
15:39:25  <isaacs>wow, why is that so slow for you? i don't get it
15:39:40  <isaacs>you and i both logged in to that machine last yesterday
15:39:51  <isaacs>bnoordhuis: you're logging in as root, right?
15:39:56  <bnoordhuis>yes
15:39:59  <isaacs>bnoordhuis: or did you create a bnoordhuis user?
15:40:04  <bnoordhuis>no
15:40:20  <isaacs>maybe try creating a non-root user and then adding a .ssh/authorized_keys so that it doesn't do the SmartLogin daemon dance
15:40:31  <isaacs>yep, with make CXXFLAGS+="-ffunction-sections -fdata-sections" it works
15:40:51  <isaacs>can we just set that on sunos alwyas?
15:41:01  <isaacs>or by default at least?
15:41:24  <bnoordhuis>i can say with some authority that function/data sections is horribly, horribly buggy
15:42:09  <bnoordhuis>isaacs: did it work?
15:43:08  <indutny>bnoordhuis: how is that related?
15:43:37  <bnoordhuis>indutny: sorry, what?
15:43:57  <indutny>V8DBG_OFF_FP_CONTEXT
15:44:01  <indutny>and ffunction-sections
15:44:32  <bnoordhuis>it's possible that the dtrace linker postprocessor expects some things in separate sections
15:45:07  <indutny>ooh
15:45:09  <indutny>that thing
15:45:09  <bnoordhuis>and apparently it does
15:45:22  <bnoordhuis>well, that's pretty crappy
15:45:47  <indutny>odd
15:46:04  <indutny>I can't find anything similar to off_fp_context in v8's code
15:46:19  <indutny>where is compiler getting this stuff from?
15:46:49  <indutny>oh
15:46:51  <indutny>postmortem
15:47:42  <indutny>ok
15:47:51  <indutny>it really looks like a compiler bug then
15:48:47  <bnoordhuis>tell me about it!
15:49:03  * stephankjoined
15:49:08  <bnoordhuis>my life recently has been nothing but chasing down compiler bugs
15:49:15  <bnoordhuis>maddening is what it is
15:49:33  <tjfontaine>next up you'll be working for a compiler company!
15:50:33  <bnoordhuis>$DEITY forbid
15:50:49  <indutny>lets create a compiler company
15:50:52  <indutny>what should it sell, compilers?
15:51:12  <mmalecki>well, obviously...
15:51:23  <mmalecki>compmart
15:51:53  * TooTallNatejoined
15:52:07  <indutny>compiler supermarket
15:52:09  <indutny>what could be better
15:52:22  <indutny>isaacs: call today?
15:52:29  <isaacs>yeah, sure
15:52:33  <isaacs>no bert, though
15:52:36  <indutny>oh
15:52:45  <indutny>I've almost forgot about it
15:53:03  <indutny>let me find some secure room for that
15:55:53  * philips_quit (Excess Flood)
15:56:50  * philips_joined
15:57:42  <isaacs>TooTallNate: ping
15:57:50  <TooTallNate>isaacs: pong
15:57:56  <isaacs>skype :)
15:58:05  <TooTallNate>oh wtf
15:58:07  <TooTallNate>i opened it
15:58:10  <TooTallNate>not logged in
15:58:44  <TooTallNate>ok im ready
15:59:03  <isaacs>bnoordhuis, indutny: ping
15:59:07  <indutny>pong
15:59:23  <bnoordhuis>isaacs: pong
15:59:24  <isaacs>skype?
16:04:38  * dapjoined
16:05:16  * loladirojoined
16:06:35  * chobi_echanged nick to chobi_e_
16:12:02  * TooTallNatequit (Quit: Computer has gone to sleep.)
16:17:50  * theColequit (Quit: theCole)
16:20:41  * AndreasMadsenjoined
16:22:16  <AndreasMadsen>isaacs: I think nodejs.org is down
16:22:39  <isaacs>AndreasMadsen: works for me
16:22:50  <indutny>for me too
16:22:52  <isaacs>http://www.downforeveryoneorjustme.com/nodejs.org/
16:23:04  <isaacs>AndreasMadsen: what url fails for you?
16:23:04  <indutny>http://www.downforeveryoneorjustme.com/nodejs.org
16:23:06  <indutny>hahaha
16:23:07  <indutny>isaacs: ++
16:23:09  <kohai>isaacs has 18 beers
16:23:26  <indutny>why do you think it's down?
16:23:51  <AndreasMadsen>indutny: because I can go to all sites except nodejs.org
16:24:04  <indutny>and what happens if you're going to nodejs.org?
16:24:09  <indutny>s/if/when
16:24:23  <AndreasMadsen>Chrome could not found nodejs.org
16:24:38  <mmalecki>try with `curl -v`, maybe?
16:24:47  <indutny>curl -v http://nodejs.org/
16:25:08  <AndreasMadsen>I will try
16:25:29  <indutny>AndreasMadsen: https://gist.github.com/4f2f91e68b29005cbd8c
16:25:51  <indutny>1778 requests worked
16:25:55  <indutny>and yours few has not :D
16:26:22  <AndreasMadsen>https://gist.github.com/3145094
16:26:45  <mmalecki>some dns issues. try `host nodejs.org`
16:26:46  <AndreasMadsen>ould not resolve host: nodejs.org; nodename nor servname provided, or not known
16:26:55  <indutny>dns problem
16:27:19  <mmalecki>I remember when all production servers started failing to resolve registry.npmjs.org, that was funny
16:27:32  <indutny> dig @8.8.8.8 nodejs.org
16:27:39  <indutny>AndreasMadsen: ^
16:27:50  <AndreasMadsen>host nodejs.org; connection timed out; no servers could be reached
16:27:56  <indutny>huh
16:27:59  <indutny>dig replied that?
16:28:06  <AndreasMadsen>not yet
16:28:33  <AndreasMadsen>dig works
16:28:53  <indutny>what IP it returns?
16:29:00  <AndreasMadsen>nodejs.org. 365 IN A 8.12.44.238
16:29:03  <indutny>8.12.44.238
16:29:04  <indutny>ok
16:29:09  <indutny>just `dig nodejs.org`
16:29:10  <indutny>?
16:29:45  <mmalecki>if that fails, your name servers are misconfigured/having problems atm
16:29:46  <AndreasMadsen>dig nodejs.org connection timed out; no servers could be reached
16:29:56  <indutny>ok
16:29:59  <indutny>your dns servers are down
16:30:00  <indutny>:D
16:30:03  <CIA-108>node: Ben Noordhuis v0.8 * rb264893 / common.gypi : build: unbreak ustack on smartos - http://git.io/uUg59A
16:30:10  <indutny>change them to 8.8.8.8 8.8.4.4
16:30:12  <mmalecki>bnoordhuis: ++ <3
16:30:13  <kohai>bnoordhuis has 20 beers
16:30:22  <indutny>++:++
16:30:25  <indutny>++: ++
16:30:29  <indutny>kohai: you're stupid
16:30:30  <mmalecki>indutny: nope, wrong regexp
16:30:31  <kohai>I am Kohai, semi-useful communications-facilitating pseudointelligence!
16:30:37  <mmalecki>indutny: that'd be me, actually :)
16:30:44  <indutny>bnoordhuis: !
16:30:47  <indutny>bbiab
16:30:47  <indutny>ttyl
16:34:20  <CIA-108>node: isaacs v0.8.3-release * r845b9e9 / (8 files in 3 dirs): uv: Upgrade to 94355e4 - http://git.io/LWdTig
16:34:20  <CIA-108>node: isaacs v0.8.3-release * rc130282 / (AUTHORS ChangeLog src/node_version.h): 2012.07.17, Version 0.8.3 (Stable) - http://git.io/9VS6dw
16:34:25  * madsonjoined
16:34:41  * AndreasMadsenquit (Read error: Connection reset by peer)
16:35:40  * madsonquit (Remote host closed the connection)
16:35:52  * AndreasMadsenjoined
16:36:17  <isaacs>testing now
16:36:23  <isaacs>smartos builds fine
16:37:13  <AndreasMadsen>manually set DNS to resolve using 8.8.8.8 works :D
16:39:15  <isaacs>tests pass, too
16:39:17  <isaacs>bnoordhuis ++
16:39:19  <kohai>bnoordhuis has 21 beers
16:40:40  * ericktquit (Ping timeout: 252 seconds)
16:41:53  <isaacs>bnoordhuis: i propose that we never touch the build system again
16:42:08  <tjfontaine>haha
16:42:47  <bnoordhuis>:)
16:43:11  <bnoordhuis>(gdb) bt
16:43:12  <bnoordhuis>#0 0x00000000004040d3 in xxx::Print(v8::Arguments const&) ()
16:43:12  <bnoordhuis>#1 0x000000000045334a in HandleApiCallHelper<false> (args=..., isolate=0xca1070) at ../src/builtins.cc:1145
16:43:12  <bnoordhuis>#2 0x000000000044e27a in Builtin_Impl_HandleApiCall (args=..., isolate=0xca1070) at ../src/builtins.cc:1162
16:43:12  <bnoordhuis>#3 0x000000000044e24b in Builtin_HandleApiCall (args=..., isolate=0xca1070) at ../src/builtins.cc:1161
16:43:12  <bnoordhuis>#4 0x00000b0a1c00618e in ?? ()
16:43:16  <bnoordhuis>#5 0x00000b0a1c0060e1 in ?? ()
16:43:18  <bnoordhuis>#6 0x00007fffffffd6d0 in ?? ()
16:43:21  <bnoordhuis>#7 0x00007fffffffd728 in ?? ()
16:43:22  <bnoordhuis>#8 0x00000b0a1c029d5e in hello () at tmp/hello.js:2
16:43:26  <bnoordhuis>^ notice that list line - hello() in tmp/hello.js:2
16:43:29  <bnoordhuis>it's a start
16:43:35  <isaacs>bnoordhuis: nice!
16:43:44  <isaacs>you're doing ::jsstack type stuff in gdb??
16:43:46  <bnoordhuis>yes
16:43:53  <isaacs>that's fantastic!
16:44:57  <isaacs>it's actually busted in v0.8 for some reason. i think V8 changed their API at some point, and the ustack helper isn't updated.
16:45:15  <bnoordhuis>yes, i've seen that
16:45:26  <bnoordhuis>i've prodded at it a little but i don't know dtrace well enough
16:47:14  <isaacs>sweet, pkg built
16:47:18  <isaacs>i think we might have a winner here
16:47:28  * dapquit (Quit: Leaving.)
16:50:55  * mjr_quit (Quit: mjr_)
16:52:54  * AndreasMadsenquit (Read error: Connection reset by peer)
16:54:24  * AndreasMadsenjoined
17:02:15  <isaacs>oh, crap, the date is wrong on the changelog. whoops
17:07:16  * theColejoined
17:10:18  * xaqjoined
17:11:44  * hzjoined
17:13:03  * AndreasMadsenquit (Remote host closed the connection)
17:16:07  * TooTallNatejoined
17:19:21  <CIA-108>node: isaacs v0.8 * r0c91b0e / (8 files in 3 dirs): uv: Upgrade to 94355e4 - http://git.io/-kcxzg
17:19:43  <CIA-108>node: isaacs v0.8.3-release * r60bf2d6 / (AUTHORS ChangeLog src/node_version.h): 2012.07.19, Version 0.8.3 (Stable) - http://git.io/QnZRAg
17:20:51  <CIA-108>node: isaacs v0.8 * r845b9e9 / (8 files in 3 dirs): uv: Upgrade to 94355e4 - http://git.io/LWdTig
17:20:51  <CIA-108>node: isaacs v0.8 * r60bf2d6 / (AUTHORS ChangeLog src/node_version.h): 2012.07.19, Version 0.8.3 (Stable) - http://git.io/QnZRAg
17:20:51  <CIA-108>node: isaacs v0.8 * r4439f7b / (AUTHORS ChangeLog src/node_version.h): Merge branch 'v0.8.3-release' into v0.8 - http://git.io/iLnLVw
17:20:52  <CIA-108>node: isaacs v0.8 * r02ff974 / src/node_version.h : Now working on 0.8.4 - http://git.io/rjwHEg
17:25:32  <isaacs>oh, whoops:
17:25:33  <isaacs>|\
17:25:33  <isaacs>| * 60bf2d6 isaacs (v0.8.3, ry-push/v0.8.3-release, v0.8.3-release) 2012.07.19
17:25:33  <isaacs>| * 845b9e9 isaacs uv: Upgrade to 94355e4 (52 minutes ago)
17:25:33  <isaacs>* | 0c91b0e isaacs uv: Upgrade to 94355e4 (51 minutes ago)
17:25:35  <isaacs>|/
17:25:41  <isaacs>that's a bit unsightly. oh well.
17:28:59  <CIA-108>node: isaacs v0.8 * rf648484 / doc/blog/release/v0.8.3.md : Blog post about v0.8.3 - http://git.io/jcsQ4w
17:34:20  <isaacs>mikeal: you know how we were talking about how nextTick should always come before I/O, so that http works?
17:34:38  <CIA-108>node: isaacs master * r4e5fe2d / (3 files in 2 dirs): nextTick: Handle tick callbacks after each tick - http://git.io/rKx3jQ
17:34:38  <CIA-108>node: isaacs master * r19ecc17 / (src/node.js test/simple/test-next-tick.js): nextTick: explicitly no-op when _exiting - http://git.io/l4ztCw
17:34:38  <CIA-108>node: isaacs master * ra52a44e / doc/api/process.markdown : doc: nextTick semantics and justification - http://git.io/zuMZtg
17:34:38  <CIA-108>node: isaacs master * r430d94e / (4 files in 2 dirs): nextTick: Preserve depth in error/reentry cases - http://git.io/-Bx9jA
17:35:43  <isaacs>^
17:36:25  <tjfontaine>nextTick of death!
17:38:33  * lohkeyjoined
17:41:42  * beachdogquit (Remote host closed the connection)
17:41:52  * beachdogjoined
17:45:10  * beachdogquit (Client Quit)
17:47:20  * mjr_joined
18:04:21  * indexzerojoined
18:13:12  * mjr__joined
18:13:21  * loladiroquit (Read error: Connection reset by peer)
18:13:28  * mjr_quit (Ping timeout: 248 seconds)
18:13:28  * mjr__changed nick to mjr_
18:14:28  * loladirojoined
18:14:44  * loladiroquit (Client Quit)
18:25:37  <indutny>bnoordhuis: it works?
18:25:38  <indutny>yay
18:32:08  * xaqquit (Read error: Connection reset by peer)
18:37:06  * xaqjoined
18:40:11  * AndreasMadsenjoined
18:42:47  * AndreasMadsenquit (Remote host closed the connection)
18:44:27  * beachdogjoined
18:46:27  * loladirojoined
18:46:48  * loladiroquit (Client Quit)
18:50:31  * EhevuTovjoined
18:50:34  * EhevuTovquit (Remote host closed the connection)
18:51:01  * EhevuTovjoined
18:53:39  * mikealquit (Quit: Leaving.)
18:54:36  * beachdogquit (Quit: beachdog)
18:55:19  <isaacs>indutny: i do like merging in the release branch with --no-ff, but it's silly to have the same exact commit on both spots
18:55:30  * mikealjoined
18:55:45  <isaacs>60bf2d6 should be a child of 0c91b0e
19:03:13  * mmaleckiquit (Quit: Lost terminal)
19:08:45  * indexzeroquit (Quit: indexzero)
19:12:07  <indutny>:)
19:12:20  <indutny>isaacs: I don't care
19:13:22  <isaacs>yeah, it wasn't worth fixing
19:15:26  <indutny>indeed
19:26:19  * dapjoined
19:28:13  * paddybyersquit (Quit: paddybyers)
19:37:58  * Sh4rKjoined
19:38:40  * dapquit (Quit: Leaving.)
19:38:50  * Sh4rKpart ("Leaving")
19:39:30  * dapjoined
19:39:40  * dapquit (Client Quit)
19:43:05  * arlolrajoined
19:49:11  * hzquit (Disconnected by services)
19:49:15  * hzjoined
19:54:36  <ryah>isaacs: did you figure out that sudo problem?
19:54:51  <isaacs>ryah: yeah, bnoordhuis fixed it
19:54:56  <isaacs>it was the proctitle chnages
19:54:57  <ryah>what was it?
19:54:59  <ryah>ah
19:55:10  <isaacs>he said it was just coincidence that it didn't fail without sudo
19:55:42  <ryah>so, v0.8 is look good
19:55:43  <isaacs>good thing that there's a `sudo make install` somewhere in the build process :)
19:55:48  <ryah>is voxer on v0.8 yet?
19:56:05  <isaacs>i think so, at least partly, but i'm not sure
19:56:08  <isaacs>mjr_: ^?
20:00:55  * mikealquit (Quit: Leaving.)
20:01:27  * paddybyersjoined
20:05:47  <indutny>isaacs: really?
20:05:50  <CIA-108>node: isaacs v0.8 * rd9057cc / (102 files in 7 dirs): npm: Upgrade to 1.1.44 - http://git.io/PKi70g
20:05:57  <indutny>I thought voxer is only planning migration
20:06:06  <isaacs>indutny: oh, maybe that's what i'm thinking of
20:06:10  <indutny>ryah: proctitle was relying on magic
20:06:16  <isaacs>i know that there was something they were trying out 0.7.12 on
20:06:17  <indutny>ryah: and that magic was borked in osx
20:06:24  <indutny>ryah: since it's using black magic
20:06:28  <indutny>it's like black vs white magic
20:06:34  <indutny>but black won on osx
20:06:35  <indutny>:D
20:06:50  * indexzerojoined
20:08:17  <CIA-108>node: isaacs master * r3c99894 / (205 files in 48 dirs): Merge remote-tracking branch 'ry/v0.8' into master (+19 more commits...) - http://git.io/ypdvfA
20:08:24  <ryah>ry/v0.8 ?
20:08:46  <indutny>yeah
20:08:49  <isaacs>ryah: yeah. i changed the remote urls way back when, but not the remote names.
20:08:49  <indutny>everyone likes it
20:08:57  <indutny>I like it!
20:09:14  <isaacs>"gp ry" was already too ingrained in the muscle memory
20:09:21  <ryah>can i op everyone in #node.js again?
20:09:34  <isaacs>lol
20:09:37  <isaacs>why do you do that?
20:09:41  <ryah>its fun :)
20:09:45  <isaacs>freaks people out
20:09:48  <ryah>people like it :)
20:09:51  <isaacs>but i guess it's sort of a slow news day
20:09:58  <isaacs>i kinda think it's funny
20:10:26  <isaacs>helps reduce the illusion that irc status is important :)
20:10:36  <tjfontaine>delusions?!
20:11:16  <TooTallNate>lol
20:11:28  * c4miloquit (Remote host closed the connection)
20:11:36  <indutny>lol
20:11:38  <indutny>it happens, again
20:12:02  * c4milojoined
20:14:07  <ryah>:)
20:14:09  <indutny>:)
20:14:09  <ryah>i think most people don't realize that on freenode we have ChanServ
20:14:11  <ryah>so you dont have takeover problems like on efnet
20:14:28  * AvianFluquit (Quit: Leaving)
20:14:44  <indutny>I think this is decreasing morale
20:15:14  * mmaleckijoined
20:21:26  <ryah>isaacs: anything new with the npm website?
20:21:52  * EhevuTovquit (Quit: This computer has gone to sleep)
20:21:54  <isaacs>ryah: not too much. it's progressing.
20:22:10  <isaacs>had to refactor some stuff out of npm to make it easier to actually do stuff with your packages from the website
20:22:14  <isaacs>starring, mostly
20:22:28  <isaacs>but i also want to make it easy to add/remove admins on a project
20:22:54  <ryah>would be nice to get module pages up before all the advanced features
20:25:45  <mjr_>ryah: mère moments away from 0.8
20:25:55  <indutny>isaacs: so, about tls verification stuff
20:25:58  <mjr_>Keep running into tiny little annoying issues.
20:25:58  * indexzeroquit (Quit: indexzero)
20:26:01  <indutny>what API do you really expect it to have?
20:26:47  <isaacs>ryah: there are module pages, but they're not super advanced
20:26:56  <isaacs>ryah: and there's not really a great way to search or explore them
20:26:59  <ryah>mjr_: nice - would love to hear about how it runs once you're up
20:28:57  <mjr_>ryah: the instant we roll this out, I'll joyfully report it here
20:29:31  * ryahquit (Quit: leaving)
20:29:38  * ryahjoined
20:32:29  <indutny>isaacs: so
20:32:34  <indutny>tls verify identity?
20:33:16  <isaacs>indutny: meh. whatever. that api is fine, i suppose.
20:33:24  <indutny>oh ok
20:33:26  <isaacs>we shouldn't expose it on 0.8 anyway, since that's an API addition
20:33:41  <indutny>yeah
20:33:46  <indutny>lets add API after this patch
20:33:48  <isaacs>i was smoking the crack
20:33:55  <isaacs>or someting
20:34:16  <indutny>I'm doing it right now :D
20:34:19  <indutny>hahah
20:34:25  <indutny>feeling sick all the day
20:35:23  <indutny>isaacs: https://github.com/joyent/node/pull/3717 ?
20:36:41  <isaacs>indutny: reading
20:36:48  <indutny>isaacs: TL;DR
20:36:59  <indutny>Guys are passing Infinity to the socket's .setTimeout
20:37:03  <isaacs>indutny: so, sock.setTimeout(null/Infinity/0, cb) just removes the timeout?
20:37:05  <indutny>that makes node crash in libuv internals
20:37:10  * EhevuTovjoined
20:37:14  <indutny>null/Infinity is ignored
20:37:18  <indutny>now
20:37:22  <indutny>but it may remove it
20:37:33  <indutny>if we want it to do so
20:37:39  <isaacs>oi
20:37:39  * leofseigejoined
20:37:49  <isaacs>hm.
20:37:57  <isaacs>i think ignoring it is fine.
20:38:13  <isaacs>sure, lgtm
20:38:30  <indutny>I wonder what will happen if they'll pass string
20:38:37  <indutny>.setTimeout('no-limit')
20:38:43  <indutny>hm...
20:38:50  <isaacs>is a string > 0?
20:38:52  <isaacs>i think so.
20:38:55  <indutny>yeah
20:39:08  <isaacs>should do a msecs = +msecs first
20:39:09  <indutny>but it's ->IntegerValue() should be zero
20:39:28  <indutny>though some other things may broke
20:39:32  <indutny>yeah, I'll refine patch
20:39:33  <isaacs>oh, no, nvm
20:39:39  <isaacs>isNaN('foo') == true
20:39:47  <indutny>isaacs: huh
20:39:48  <isaacs>because 'foo' isn't a number
20:39:49  <indutny>good
20:39:53  <indutny>no
20:39:54  <isaacs>and +'foo' is NaN
20:40:01  <indutny>because isNaN()'s argument is converted to number
20:40:02  <indutny>yes
20:40:12  <isaacs>so your patch covers that case
20:40:12  <indutny>ok, good
20:40:16  <isaacs>yeah, it's fine
20:40:26  <indutny>but I should test .setTimeout('123')
20:40:47  <isaacs>that should be the same as .setTimeout(123)
20:41:09  <isaacs>> msecs = '123'; (msecs > 0 && !isNaN(msecs) && isFinite(msecs))
20:41:10  <isaacs>true
20:41:24  <indutny>ok, it doesn't crash
20:41:35  <indutny>so I suppose it works :)
20:41:43  <isaacs>but yeah, it should probably cast to a number explicitly. maybe even msecs = parseInt(msecs, 10)
20:41:55  <isaacs>but i suppose we're doing that in C++ anyway
20:42:07  * isaacsand indutny are polishing doorhandles on the bikeshed
20:42:14  <indutny>hehe
20:42:18  <ryah>sounds like that uncaught parser error issue might be back
20:42:24  <ryah>https://github.com/joyent/node/issues/2997
20:42:28  <isaacs>land it quick before we waste any more time :)
20:43:02  <indutny>isaacs: ok
20:43:04  <indutny>:)
20:44:08  <isaacs>ryah: i don't think we ever really fixed it, we just made its window for occurrence a bit smaller.
20:44:51  <isaacs>ryah: the problem is that in http.js there are two places where we use process.nextTick() to let the caller assign event listeners before we do stuff, and IO can occur in that window, under load.
20:44:55  <CIA-108>node: Fedor Indutny master * rb0950cb / (lib/net.js test/simple/test-net-settimeout.js): net: ignore socket.setTimeout(Infinity) (and NaN) - http://git.io/vwqQFQ
20:45:07  <isaacs>ryah: that's the reason for the recent process.nextTick chnages in master.
20:45:09  <indutny>done ^
20:45:13  <isaacs>indutny: thanks!
20:45:17  <indutny>isaacs: no, thank you
20:45:19  * mjr_quit (Ping timeout: 255 seconds)
20:46:05  <isaacs>indutny: what's remaining on the tls identity thing? is that ready to go, or still need work to fix the bugs bnoordhuis found?
20:46:20  * mjr_joined
20:46:24  <indutny>no
20:46:36  <indutny>bnoordhuis has pointed out some flaws in check function
20:48:23  <indutny>ryah: yt?
20:48:24  <isaacs>yeah, i saw
20:48:42  <indutny>ryah: https://github.com/fferreira/ICFP2012
20:48:48  <indutny>Fragment from readme:
20:48:55  <indutny>We cannot submit our binaries because all the computation in our code
20:48:55  <indutny>happens at compile time, during template instatiation. In fact we use
20:48:55  <indutny>templates as a Pure Functional Programming Language! Long live the
20:48:55  <indutny>λ. That is, we use the C++ compiler as a functional interpreter.
20:49:01  <indutny>that's just fucking awesome
20:49:04  <indutny>I love the idea
20:50:21  <ryah>indutny: you mistake me for a nerd
20:50:33  <ryah>:)
20:50:35  <indutny>:)
20:50:57  * AlbireoX__quit (Quit: IRC crashed.)
20:51:03  <indutny>do you think I'm nerd?
20:51:43  <ryah>i think that's obvious
20:52:05  <indutny>I do not classify myself as nerd
20:52:07  <indutny>really
20:52:13  <indutny>I don't like gadgets and all the stuff
20:52:14  <ryah>anyone who has written a JIT...
20:52:17  <indutny>hahahaha
20:52:23  <indutny>ok, I'm a compiler nerd
20:52:42  <ryah>indutny: you dont like dart?
20:52:56  <indutny>ryah: if that makes me a nerd , then I don't
20:53:23  <indutny>:)
20:54:06  <indutny>bnoordhuis: yt, or sleeping?
20:54:47  <ryah>indutny: i like it
20:54:50  <ryah>but it's slow as fuck
20:55:00  <indutny>yes, I'm reading twitter time after time
20:55:14  <indutny>vyacheslav said that they're actively working on compiler
20:55:22  <ryah>yeah
20:55:24  <indutny>though timing is not as rapid as they initially expected
20:55:42  <ryah>normally i would discount the project entirely after this shit
20:55:55  <indutny>well
20:55:58  <ryah>but i trust those guys know what they are doing
20:56:01  <indutny>I think Dart is a step back
20:56:13  <ryah>it's pretty complex language
20:56:15  <indutny>it's like Lars has tired of Ecmascript simplictiy
20:56:24  <indutny>and wanted to return back to something Java-like
20:56:39  <ryah>i like the idea of templates
20:56:49  <ryah>but the only reason for that would be to have fucking fast code
20:57:02  <indutny>hahah
20:57:08  <indutny>v8 is fucking fast
20:57:13  <ryah>yes
20:57:17  * mikealjoined
20:57:18  <ryah>so i really hope they can do better
20:57:40  <indutny>well, as far as JVM is faster than everyone - they have a chance
20:57:59  <indutny>isaacs: https://github.com/joyent/node/pull/3689
20:58:08  <indutny>it's passing bnoordhuis' patches
20:58:24  <indutny>so, if no other objections - lgtm
21:03:12  * mikealquit (Quit: Leaving.)
21:04:03  * isaacsreading
21:05:03  <isaacs>indutny: so, DNS:... has to have at least 2 non-* items, right?
21:05:10  <indutny>yes
21:06:08  <isaacs>i thought there was a special exception for stuff like .co.uk or .co.au to count as one thing?
21:07:07  <indutny>huh?
21:07:11  <indutny>what do you mean?
21:07:19  <indutny>google.co.uk
21:07:22  <indutny>google.co.au
21:07:31  <isaacs>i mean, i thought that DNS:*.co.uk would not apply to a.co.uk
21:07:32  <creationix>indutny: I can't join #candor. do you need to invite me?
21:07:37  <isaacs>ie, .co.uk is like .com
21:07:49  * isaacscould be wrong
21:07:50  <indutny>creationix: huh, really?
21:07:56  <indutny>isaacs: let me reread RFC
21:08:02  <creationix>isaacs: supercookies!
21:09:25  <indutny>odd
21:09:31  <indutny>creationix: someone has changed mode to cint
21:10:04  <indutny>creationix: thanks for reporting
21:10:17  <creationix>lol, now I don't have voice in #candor
21:10:58  * hz_joined
21:11:27  <isaacs>indutny: otherwise lgtm
21:11:40  <isaacs>indutny: and if i'm wrong about that, then lgtm :)
21:11:46  <indutny>:)
21:20:32  * leofseigequit (Remote host closed the connection)
21:22:44  * EhevuTovquit (Quit: This computer has gone to sleep)
21:26:21  * philips_quit (Excess Flood)
21:27:21  * philips_joined
21:27:55  * c4milo_joined
21:27:59  * c4miloquit (Read error: Connection reset by peer)
21:28:42  <CIA-108>node: isaacs v0.9.0-release * r63eab37 / (AUTHORS ChangeLog): 2012.07.20, Version 0.9.0 (Unstable) - http://git.io/6XQrTw
21:28:44  <isaacs>ok, release branch started
21:29:11  <isaacs>not a very big departure from 0.8
21:29:35  <tjfontaine>for now
21:31:20  <indutny>isaacs: so, this stuff seems to be correct
21:31:23  <indutny>specification is really flacky
21:32:03  <isaacs>i guess it's better to err on the side of allowing too much
21:32:06  <isaacs>rather than too little
21:32:10  <isaacs>since we allow everything nwo
21:32:36  <indutny>yeah
21:32:38  <indutny>lets fix it after
21:32:42  <indutny>going to push it
21:32:56  <indutny>will it get to 0.9.0?
21:33:06  * xaqquit (Read error: Connection reset by peer)
21:33:13  <indutny>isaacs: ?
21:35:07  * xaqjoined
21:35:32  * theColequit (Quit: theCole)
21:36:06  <isaacs>indutny: i think this actually has to go in 0.8
21:36:12  <isaacs>maybe 0.6 as well.
21:36:19  <isaacs>it's a pretty serious security flaw.
21:42:00  * xaq_joined
21:43:56  * xaqquit (Ping timeout: 260 seconds)
21:46:15  <indutny>let me push it to 0.9
21:46:19  <indutny>then cherry-pick in 0.8
21:46:30  <indutny>and only after we'll release 0.8.4 - merge it into 0.6
21:46:42  <indutny>so we'll be sure that there won't be next hurry release of 0.6 :D
21:46:55  <CIA-108>node: Fedor Indutny master * r2b3ba3f / (lib/http.js lib/https.js): http/https: pass request to .createConnection() - http://git.io/GHMCeA
21:46:55  <CIA-108>node: Fedor Indutny master * r8ba189b / (4 files in 3 dirs): tls: veryify server's identity - http://git.io/Kxe6RQ
21:47:28  <indutny>isaacs: should b0950cbea282024a21c735ba20803ac8b05e0b48 also go to 0.8?
21:47:33  * hz_quit (Quit: Yaaic - Yet another Android IRC client - http://www.yaaic.org)
21:47:36  <isaacs>indutny: you can probably cherry-pick into 0.6 also
21:47:44  <indutny>what
21:47:47  <indutny>timer or tls?
21:47:52  <isaacs>tls
21:48:00  <indutny>what about timer?
21:48:06  <isaacs>indutny: what does sock.setTimeout(NaN) do on 0.8? crash?
21:48:11  <indutny>yes
21:48:15  <isaacs>k, that's stupid
21:48:19  <indutny>indeed
21:48:21  <isaacs>yes, cherrypick into there as well
21:48:38  <isaacs>doesn't change the API any, just makes it dtrt
21:48:41  <indutny>yes
21:49:18  <indutny>huh
21:49:23  <indutny>0.8 doesn't have tls slab buffer?
21:50:20  <isaacs>indutny: no, i think that's just in master
21:50:35  <CIA-108>node: Fedor Indutny v0.8 * r1fa0bca / (lib/net.js test/simple/test-net-settimeout.js): net: ignore socket.setTimeout(Infinity) (and NaN) - http://git.io/dnb8vA
21:50:35  <CIA-108>node: Fedor Indutny v0.8 * r53716eb / (lib/http.js lib/https.js): http/https: pass request to .createConnection() - http://git.io/QLgMCw
21:50:36  <CIA-108>node: Fedor Indutny v0.8 * reb2ca10 / (4 files in 3 dirs): tls: veryify server's identity - http://git.io/ZXBUOg
21:50:37  <indutny>yeah, really
21:50:47  <indutny>I think we should pull this too
21:50:52  <indutny>but lets postpone it until tomorrow
21:50:56  <isaacs>yeah, sounds good
21:50:58  <isaacs>i'm fading.
21:51:02  <indutny>yeah, me too
21:51:06  <isaacs>g'nite :)
21:51:13  <indutny>isaacs: you too, man
21:51:16  <indutny>get better
21:51:22  <isaacs>thanks
21:52:19  * mikealjoined
21:54:52  * theColejoined
21:56:25  * EhevuTovjoined
21:56:48  * mikealquit (Client Quit)
22:01:08  * loladirojoined
22:03:46  <TooTallNate>isaacs: aren't you supposed to be sleeping?
22:04:17  <indutny>TooTallNate: and you/
22:04:30  <indutny>looks like travis-ci is definitely sleeping
22:04:44  <TooTallNate>indutny: i mean since he's not feeling well
22:04:53  <TooTallNate>indutny: it's only 3pm over here, so not sleepy time yet :)
22:05:04  <indutny>well, I know
22:05:11  <indutny>but when you're sick - time passes in other way
22:05:23  <TooTallNate>like video games :p
22:05:34  <indutny>http://travis-ci.org/#!/joyent/node/builds
22:05:41  <indutny>huh
22:05:46  <indutny>3 months old logs?
22:06:02  <indutny>mmalecki: %
22:06:33  * rendarquit
22:07:21  * hzquit
22:07:23  <mmalecki>indutny: sup?
22:07:37  <mmalecki>wat...
22:07:41  <indutny>mmalecki: why travis build logs are so old here
22:07:54  <mmalecki>you guys are sure that the hook is still enabled?
22:08:01  <indutny>probably not
22:08:02  <indutny>idk
22:08:10  <indutny>TooTallNate: you know?
22:08:28  <mmalecki>ask someone with admin rights to the repo
22:08:33  <TooTallNate>indutny: no i don't have access to that page
22:08:49  <TooTallNate>isaacs and ryah are the only one's I know for sure that do
22:08:52  <indutny>looks like we have turned it off
22:09:03  <indutny>haven't seen anything in like months
22:09:11  <indutny>:)
22:10:01  * dapjoined
22:24:09  * theColequit (Quit: theCole)
22:28:28  * mikealjoined
22:29:31  * xaq_quit (Remote host closed the connection)
22:33:18  <indutny>mmalecki: budzik dzvonek zroano
22:33:26  <indutny>s/zroano/zrano
22:34:11  <indutny>oh
22:34:37  * blackorzarjoined
22:34:45  <indutny>budzik dzwonek
22:34:50  <indutny>mmalecki: yt?
22:35:10  <indutny>mmalecki: I just realized that Volen De Mort was speaking Polish
22:35:33  <mmalecki>indutny: wat lol
22:35:43  <mmalecki>indutny: oh, lulz
22:36:00  <indutny>you have a lot of s, sh sounds
22:36:04  <indutny>sz
22:36:04  <indutny>dz
22:36:05  <mmalecki>indutny: budzik dzwoni rano - I think this is what you mean
22:36:13  <indutny>haha
22:36:17  <indutny>right
22:36:27  <indutny>so polish is really like a snake language
22:36:28  <indutny>no offense
22:36:33  <indutny>:D
22:36:35  <mmalecki>lulz
22:36:44  <indutny>can you say "My precious!" on polish?
22:37:08  <mmalecki>sure
22:37:13  <mmalecki>"mój skarb!"
22:37:19  <mmalecki>or "moje skarby!", for plural
22:37:27  <indutny>chor rowniez ulega modernizacji
22:37:36  <indutny>I'm learning polish
22:37:45  <indutny>Innego rdzaju sztuka jest Orphee
22:37:59  <indutny>Dziwaczna postac konia nie opuszcza sceny
22:38:06  <indutny>lulz
22:38:21  * indutnytopic: Dziwaczna postac konia nie opuszcza sceny
22:38:25  <mmalecki>hahahahahahaha
22:38:51  <mmalecki>that's actually a valid polish sentence!
22:38:54  <indutny>hahaha
22:38:55  <indutny>I know
22:38:55  <mmalecki>congrats, indutny :)
22:38:59  <indutny>I've copied it from book
22:39:01  <indutny>lulz
22:39:01  <mmalecki>lol
22:39:09  <indutny>I'm not going to learn Polish, actually
22:39:18  <indutny>but my wife learnt it at University
22:39:26  <indutny>and that's all that she knows :P
22:39:30  <mmalecki>nice, what was she studying?
22:39:32  <mmalecki>lol
22:39:48  <indutny>Polish language :)
22:39:50  <indutny>hahaha
22:40:01  <indutny>well, her main speciality is Journalist
22:40:37  * indutnytopic: Свобода Равенство Скорость
22:41:01  <mmalecki>indutny: what's the phonetic version of the topic?
22:41:06  <mmalecki>I might be able to understand it
22:41:24  <indutny>Svoboda Ravenstvo brat·stvo
22:41:31  <indutny>oops
22:41:35  <indutny>last word is incorrect
22:41:36  <indutny>Skorost
22:41:42  <indutny>Svoboda = Liberty
22:41:47  <indutny>Ravenstvo = Unity
22:41:58  <indutny>Skorost = Velocity
22:42:03  <mmalecki>oh, nice
22:42:10  <mmalecki>also, that chatlog is full of lulz right now
22:42:16  <indutny>yeah
22:42:51  * blackorzar_joined
22:43:16  <indutny>ircretary: tell bnoordhuis that Так и будет, если дело продолжится, если сам народ не опомнится; а интеллигенция не поможет ему.
22:43:16  <ircretary>indutny: I'll be sure to tell bnoordhuis
22:43:29  * AlbireoX_joined
22:43:31  <indutny>lulz night
22:43:41  <mmalecki>indutny: svoboda is pretty much the same in polish
22:43:42  * arlolraquit (Quit: Linkinus - http://linkinus.com)
22:43:46  <mmalecki>"swoboda"
22:43:54  <indutny>yeah, polish is funny russian
22:43:56  <mmalecki>but rest is different
22:43:57  <indutny>or other way
22:44:01  <mmalecki>both ways :)
22:45:17  * blackorzar__joined
22:45:32  * blackorzarquit (Ping timeout: 246 seconds)
22:45:34  * dapquit (Quit: Leaving.)
22:48:21  * blackorzar_quit (Ping timeout: 264 seconds)
23:09:02  <bnoordhuis>so it has come to this
23:09:07  <bnoordhuis>the ruskies are taking over
23:09:34  <mmalecki>you could've expected this to happen, Ben
23:09:49  <bnoordhuis>we've been to soft on 'em
23:10:57  <mmalecki>bnoordhuis: so, I'm going to grab my gun, eh?
23:11:27  <bnoordhuis>we shall fight them on the beaches!
23:12:05  <mmalecki>that's what allies said too
23:17:14  <bnoordhuis>net: fix net.Server.listen({fd:x}) error reporting (Ben Noordhuis) <- that hasn't actually landed yet :/
23:17:52  <bnoordhuis>oh well, we'll just sneak it into the next release
23:17:52  <indutny>hahaha
23:18:13  <indutny>nice policy
23:18:30  <indutny>lets invent features
23:18:34  <indutny>add them to changelog
23:18:42  <indutny>and implement them if anyone asks
23:18:47  <indutny>bnoordhuis: ^
23:18:51  <bnoordhuis>then implement at our leisure or until the bug reports start coming in
23:18:53  * avalanche123joined
23:18:58  <bnoordhuis>demand driven development
23:19:07  <indutny>yes
23:19:19  <indutny>I think it'll work
23:19:29  <indutny>noone ever tests most of the stuff we're doing
23:19:33  <indutny>until it crashes in production
23:19:40  <indutny>:)
23:19:43  <indutny>ok
23:19:48  <indutny>3am is a good time to start sleeping
23:19:49  <indutny>ttyl
23:19:51  <indutny>;)
23:20:13  <bnoordhuis>sleep tight, fedor
23:25:40  * dapjoined
23:30:51  <TooTallNate>isaacs: are there plans for linux binaries ever?
23:30:59  <TooTallNate>or binaries for platforms other than mac and windows?
23:31:05  <TooTallNate>solaris, etc.
23:31:25  <bnoordhuis>oh, i lied - that listen() errno change did land
23:32:21  <bnoordhuis>TooTallNate: i guess it might work now that we have a mostly static build
23:32:38  <bnoordhuis>i.e. with libc being the only dependency
23:32:49  <TooTallNate>bnoordhuis: i would be +1 :)
23:33:08  <TooTallNate>makes a lot of sense considering it's what most people will be deploying to
23:33:23  * chobi_e_changed nick to chobi_e
23:34:26  <bnoordhuis>it'd be nice when someone reports a compiler bug to tell him "here, either download this binary or fix it yourself"
23:35:13  <TooTallNate>bnoordhuis: ^ you'd like that, wouldn't you :p
23:35:34  <bnoordhuis>i would love that :)
23:35:41  <TooTallNate>lol
23:35:50  <TooTallNate>i'll create a ticket
23:35:55  <TooTallNate>get some async feedback
23:36:23  * c4milo_quit (Remote host closed the connection)
23:36:59  <bnoordhuis>good idea
23:38:28  * chobi_echanged nick to chobi_e_
23:40:46  <bnoordhuis>https://github.com/bnoordhuis/node/commit/23cf4b2 <- one big per-tick HandleScope for performance reasons. opinions?
23:41:40  * AvianFlujoined
23:42:05  <bnoordhuis>a nice side effect is that it makes it easy to check how many handles are created on each tick
23:42:34  <mmalecki>bnoordhuis: what about memory usage cost?
23:43:03  <bnoordhuis>mmalecki: have you read the commit log? :)
23:43:04  <mmalecki>also, does it affect gc pauses?
23:43:10  <mmalecki>oh lol
23:43:30  <mmalecki>I tend to proceeed to code immediately :)
23:43:43  <mmalecki>ok, that sgtm
23:43:47  <bnoordhuis>re gc pauses, not that i've seen
23:44:00  <bnoordhuis>the gc in v0.8 is a lot better though
23:44:10  <mmalecki>oh, yeah, it is
23:44:26  <mmalecki>measured that with dtrace on production load balancers
23:45:23  <TooTallNate>wow, concomitant
23:45:25  <TooTallNate>good word
23:45:46  * loladiroquit (Quit: loladiro)
23:47:16  * paddybyersquit (Quit: paddybyers)
23:56:23  * joeandaverdejoined