00:02:49  * bnoordhuisquit (Ping timeout: 252 seconds)
00:02:58  * EhevuTov_quit (Ping timeout: 272 seconds)
00:05:01  * piscisaureus_quit (Ping timeout: 240 seconds)
00:06:10  * hzjoined
00:09:53  * hzquit (Client Quit)
00:19:06  <tjfontaine>because I'm a bofh I banned github's irc hook on oftc, spammy service.
00:34:03  * joshthecoderquit (Quit: Leaving...)
00:38:06  * Benviejoined
00:38:32  * piscisaureus_joined
00:49:27  * ericktjoined
00:54:09  * mikealquit (Quit: Leaving.)
00:55:19  * mikealjoined
00:57:19  * dapquit (Quit: Leaving.)
00:57:40  * mikealquit (Client Quit)
00:57:54  * dshaw_quit (Quit: Leaving.)
01:07:58  * EhevuTov__quit (Quit: This computer has gone to sleep)
01:17:10  * c4miloquit (Remote host closed the connection)
01:19:14  * piscisaureus_quit (Ping timeout: 268 seconds)
01:22:36  * igorzi_joined
01:26:03  * igorziquit (Ping timeout: 252 seconds)
01:29:18  * igorzi_quit (Ping timeout: 245 seconds)
01:34:34  * joshthecoderjoined
01:59:09  <isaacs>ircretary: tell piscisaureus_ i have questions for him about `npm flatten`
01:59:09  <ircretary>isaacs: I'll be sure to tell piscisaureus_
01:59:13  <isaacs>ircretary: thanks
01:59:13  <ircretary>isaacs: You're welcome :)
01:59:19  * mmaleckijoined
02:01:05  * mscdexjoined
02:01:41  <mscdex>does libuv provide a way to get notified when a socket (fd) disconnects?
02:05:04  * brsonquit (Ping timeout: 240 seconds)
02:08:18  <mscdex>like maybe checking for an event mask or something (e.g. UV_CLOSING) in the poll callback?
02:14:40  <joshthecoder>mscdex: think your read cb will get invoked w/ an error (UV_EOF)
02:14:56  <joshthecoder>https://github.com/joyent/libuv/blob/master/include/uv.h#L276
02:14:57  <mscdex>read cb?
02:15:16  <mscdex>i'm using uv_poll :-)
02:15:24  <mscdex>sorry
02:17:07  <joshthecoder>mscdex: polling a socket fd?
02:17:14  <mscdex>yep
02:17:56  * mmaleckiquit (Ping timeout: 244 seconds)
02:21:46  * lohkeypart
02:23:11  * mikealjoined
02:30:01  <joshthecoder>mscdex: any reason you use poll here?
02:30:27  <mscdex>yep, writing a binding to a third party async lib :-)
02:30:29  <joshthecoder>my guess you need to do a read() and check if it returns zero.
02:30:46  <mscdex>there is no read
02:31:03  <mscdex>i'm not using uv_stream
02:31:32  <joshthecoder>mscdex: I mean http://linux.die.net/man/2/read
02:32:10  <mscdex>that's not portable
02:33:18  <mscdex>i see libuv uses some UV_CLOSING, etc when working with a handle's flags, but i don't see those flags on my handle, it's constant inside my poll callback
02:33:39  <mscdex>hrmm
02:38:41  * mikealquit (Quit: Leaving.)
02:40:02  <joshthecoder>mscdex: that's only set when you close from what I see.
02:40:11  <joshthecoder>not if the other side hangs up
02:41:04  <mscdex>yeah
02:41:16  <joshthecoder>poll really only gives you read/write events. It has no concept of "connection". It just deals with any FD.
02:42:05  * Benviequit
02:49:48  * brsonjoined
02:55:01  * brsonquit (Ping timeout: 246 seconds)
02:56:56  * brsonjoined
03:04:20  * mikealjoined
03:12:11  * blackorzarjoined
03:29:26  * mscdexpart ("Leaving")
03:31:23  * bradleymeckquit (Quit: bradleymeck)
03:46:07  * mikealquit (Quit: Leaving.)
03:55:36  * dshaw_joined
04:33:53  * mikealjoined
04:39:46  * TooTallNatequit (Quit: ["Textual IRC Client: www.textualapp.com"])
04:42:50  * tomshredsjoined
05:03:18  * ericktquit (Quit: erickt)
05:16:13  * ArmyOfBrucequit (Excess Flood)
05:17:07  * ArmyOfBrucejoined
05:18:28  * ArmyOfBrucequit (Excess Flood)
05:19:22  * ArmyOfBrucejoined
05:29:37  * indexzerojoined
05:43:28  * Benviejoined
05:44:03  * ArmyOfBrucequit (Excess Flood)
05:44:30  * ArmyOfBrucejoined
05:54:35  * mikealquit (Quit: Leaving.)
06:00:37  * mikealjoined
06:05:44  <indutny>hey people
06:15:14  * mikealquit (Quit: Leaving.)
06:33:47  * blackorzarquit (Remote host closed the connection)
06:35:40  * brsonquit (Ping timeout: 240 seconds)
06:37:03  * indexzeroquit (Quit: indexzero)
06:52:55  * rendarjoined
07:09:00  * indutnyquit (Ping timeout: 240 seconds)
07:09:18  * indutnyjoined
07:17:27  * loladirojoined
07:18:17  * loladiroquit (Read error: Connection reset by peer)
07:18:23  * loladiro_joined
07:25:11  * AvianFluquit (Ping timeout: 246 seconds)
07:25:37  * AvianFlujoined
07:41:16  * joshthecoderquit (Quit: Linkinus - http://linkinus.com)
07:45:02  * `3rdEdenjoined
08:03:42  * indexzerojoined
08:04:48  * indexzeroquit (Client Quit)
08:14:24  * loladiro_quit (Ping timeout: 240 seconds)
08:43:30  * hzjoined
09:12:30  * loladirojoined
09:21:31  * `3rdEdenquit (Quit: Leaving...)
09:24:43  * `3rdEdenjoined
09:40:09  * mikealjoined
09:51:02  * mmaleckijoined
10:00:35  * loladiroquit (Quit: loladiro)
10:31:44  * bnoordhuisjoined
10:32:29  * mmalecki_joined
10:32:41  * mmaleckiquit (Ping timeout: 260 seconds)
10:33:02  * mmalecki_changed nick to mmalecki
10:34:44  * dshaw_quit (Quit: Leaving.)
10:47:42  * AvianFluquit (Quit: AvianFlu)
10:52:12  <CIA-131>node: Ben Noordhuis master * r2cdf427 / src/signal_wrap.cc : signal_wrap: fix unused variable compiler warning - http://git.io/shNTCg
11:19:33  * abraxasquit (Remote host closed the connection)
11:22:57  * loladirojoined
11:25:05  <rendar>wait, signal_wrap.cc ? .cc == C++ extension? isn't libuv written in C? :)
11:25:34  <deoxxa>that's node
11:25:40  <deoxxa>(that commit is node)
11:25:59  <rendar>i see
11:28:27  * piscisaureus_joined
11:29:27  * piscisaureus_quit (Read error: Connection reset by peer)
11:36:13  * piscisaureus_joined
11:36:43  <piscisaureus_>ircretary: tell isaacs I am happy to answer questions about npm flatten
11:36:43  <ircretary>piscisaureus_: I'll be sure to tell isaacs
11:37:16  * mmaleckichanged nick to mmalecki[away]
11:46:04  <CIA-131>node: Ben Noordhuis v0.8 * r7c75ca7 / doc/api/cluster.markdown : doc: cluster: setupMaster() can be called only once - http://git.io/B18U9A
12:01:05  <bnoordhuis>someone broke the internet...
12:06:52  * bnoordhuisquit (Read error: Operation timed out)
12:09:26  <piscisaureus_>yea
12:09:26  <piscisaureus_>my internet also breaks all teh time
12:09:26  <piscisaureus_>I should apply for a job at UPC
12:09:49  <`3rdEden>My UPC internet only breaks around midnight
12:12:04  <piscisaureus_>is that because your SO takes the power cable out? :-p
12:12:27  <`3rdEden>lol nope
12:14:09  * bnoordhuisjoined
12:14:16  <indutny>bnoordhuis: hey man
12:14:20  <indutny>piscisaureus_: good morning
12:14:31  <indutny>bnoordhuis: mind porting my select patch to 0.8 branch?
12:14:34  <piscisaureus_>good afternoon indutny
12:14:40  <indutny>I'll upgrade libuv in node afterwards
12:14:48  <indutny>piscisaureus_: how are you?
12:14:56  <piscisaureus_>indutny: is that possible without breaking the ABI?
12:15:04  <indutny>em...
12:15:06  <piscisaureus_>indutny: I'm good, ou?
12:15:06  <indutny>dunno
12:15:10  <indutny>piscisaureus_: fine
12:18:11  <piscisaureus_>hmm, benchmarking isaac's static file server:
12:18:12  <piscisaureus_>http://screencast.com/t/y4TwGK406FQ
12:18:17  <indutny>piscisaureus_: well, it adds void* pointer to a uv_stream_t
12:18:23  * bradleymeckjoined
12:18:25  <piscisaureus_>ai
12:18:45  <indutny>piscisaureus_: probably not good for ABI compatibility
12:18:47  <indutny>:)
12:18:53  <piscisaureus_>I don't think so :-)
12:18:55  <indutny>but how does it breaks it?
12:18:58  <piscisaureus_>maybe you can find a way around that
12:19:01  <indutny>aaah
12:19:07  * TheJHquit (Read error: Operation timed out)
12:19:09  <indutny>people may allocate uv_stream_t
12:19:11  <indutny>hm...
12:19:14  <indutny>yeah, let me think about it
12:20:57  <indutny>piscisaureus_: what do you think about adding `void* reserved `to every structure in next uv version? :D
12:23:20  <piscisaureus_>crap I profiled the debug build :-(
12:23:22  * bnoordhuisquit (Ping timeout: 252 seconds)
12:23:29  <piscisaureus_>indutny: not really that enthousiastic atm ...
12:23:50  <piscisaureus_>indutny: what are the fields present in an uv_stream_t ?
12:23:54  <piscisaureus_>hmm
12:24:01  <piscisaureus_>I can look that up myself :-)
12:25:36  <piscisaureus_>indutny: can you not overlap hte read_watcher field?
12:26:30  <piscisaureus_>indutny: connection_cb also seems like a good candidate to me
12:26:38  <piscisaureus_>indutny: since that is only used for servers
12:29:24  <indutny>hm...
12:29:34  <indutny>lets see
12:29:46  <indutny>actually, I feel very depressed
12:29:50  <piscisaureus_>indutny: why?
12:29:52  <indutny>Omsk makes me sad
12:30:06  <piscisaureus_>ah you're back in omsk
12:30:06  <indutny>not really an inspiring environment there
12:30:15  <indutny>well, I'm going back to Moscow at 24th
12:30:25  <piscisaureus_>I can imagine - but you are born there right?
12:30:46  <indutny>yeah
12:31:01  <piscisaureus_>indutny: typically people don't really hate the environment where they grew up - so how come you do?
12:31:46  * `3rdEdenchanged nick to `3E|BRB
12:32:11  <piscisaureus_>indutny: I mean, by all means, if you hate the place, then leave.
12:32:45  <piscisaureus_>but you shouldn't think that going somewhere is a solution to all your problems :-)
12:33:06  <indutny>haha
12:33:13  <indutny>I only hate Omsk actually
12:33:29  <indutny>Moscow and every other place is good
12:33:50  <indutny>it became awful place to live
12:33:54  <indutny>in past 3-4 years
12:34:08  <piscisaureus_>so what happened then?
12:34:30  <indutny>it's hard to describe, if you haven't been there :)
12:34:55  <indutny>all talented and smart people left this place
12:35:01  <indutny>almost
12:35:26  <piscisaureus_>right, ok
12:35:47  <indutny>and people do not like cleverness there
12:36:01  * TheJHjoined
12:36:01  <piscisaureus_>so they like frown upon you?
12:36:19  * loladiroquit (Quit: loladiro)
12:36:38  <indutny>yes
12:36:40  <indutny>everyone :D
12:36:50  <indutny>also I'm quite hairy in comparison to all people around :)
12:37:00  <piscisaureus_>hahaha
12:37:01  <indutny>and there're a lot of people who have been in prisons
12:37:32  * bradleymeckquit (Quit: bradleymeck)
12:37:58  * bnoordhuisjoined
12:41:38  <piscisaureus_>indutny: but I suppose that didnt really change in the past couple of years
12:41:54  <indutny>that always been like that
12:42:04  <indutny>but it was for minority of citizens
12:45:04  <indutny>time to hack :)
12:46:17  <bnoordhuis>anyone on os x that wants to test https://github.com/bnoordhuis/libuv/commit/3f085ac ?
12:46:32  <indutny>how to test?
12:46:43  <indutny>make -j4 test?
12:47:31  <bnoordhuis>indutny: yes. apply and run the tests
12:48:00  * TheJHquit (Ping timeout: 240 seconds)
13:07:12  <indutny>bnoordhuis: lgtm
13:07:19  <indutny>[% 100|+ 156|- 1]: Done.
13:07:23  <indutny>only process title failed
13:07:39  <bnoordhuis>cool, i'll land it later today
13:15:01  <indutny>piscisaureus_: hijacking those doesn't seem to be a really good idea for me
13:15:07  <indutny>I mean connection_cb
13:15:17  <indutny>it's used for tcp and pipe
13:15:36  <piscisaureus_>indutny: I know
13:15:43  <piscisaureus_>indutny: but only if it's used as a server
13:16:01  * mmalecki[away]changed nick to mmalecki
13:16:13  <indutny>anyway, you may try starting server using existing fd
13:16:20  <indutny>which may not work with kqueue
13:18:04  * hz_joined
13:19:29  <indutny>piscisaureus_: can I use ->data from read_watcher->io_watcher ?
13:19:45  <piscisaureus_>indutny: I think so but to be sure you'd have to ask bnoordhuis
13:19:55  <indutny>bnoordhuis: I need to talk to you!
13:19:57  <indutny>:)
13:20:00  <bnoordhuis>indutny: no
13:20:03  <indutny>bnoordhuis: why?
13:20:06  <bnoordhuis>it's going away
13:20:18  <indutny>well, I need it to backport patch to v0.8
13:20:19  <bnoordhuis>besides, it's currently used to stash the cb in
13:20:20  <indutny>without breaking ABI
13:20:30  <indutny>bnoordhuis: where?
13:20:44  <bnoordhuis>look at uv__io_set_cb in core.c
13:21:22  <indutny>ohhhh
13:21:39  <indutny>right
13:23:47  <indutny>bnoordhuis: do you see any field that I can hijack
13:24:03  <indutny>I suppose we might need to have one select() thread on os
13:24:04  <bnoordhuis>indutny: in uv_stream_t?
13:24:06  <indutny>s/os/osx
13:24:08  <indutny>bnoordhuis: yes
13:24:13  <bnoordhuis>no
13:24:27  <bnoordhuis>btw, i don't really care about back-porting that patch
13:24:37  <bnoordhuis>people that want to `node < /dev/tty` can upgrade to master
13:24:55  <indutny>hm...
13:25:02  * hz_quit (Quit: Yaaic - Yet another Android IRC client - http://www.yaaic.org)
13:25:11  * piscisaureus_quit (Read error: Connection reset by peer)
13:25:59  * piscisaureus_joined
13:26:40  <mmalecki>what'd be the purpose of doing `node < /dev/tty` anyway?
13:26:51  <mmalecki>I'm not too good with ttys, tbh
13:27:22  <bnoordhuis>what are the best spam blacklists right now? still spamhaus?
13:27:37  * piscisaureus_quit (Read error: Connection reset by peer)
13:27:50  <indutny>bnoordhuis: I think so
13:27:51  <mmalecki>bnoordhuis: https://github.com/joyent/node/blob/master/AUTHORS
13:28:16  <indutny>mmalecki: suppose you're doing following thing curl http:/... | sh
13:28:33  <bnoordhuis>mmalecki: was that an attempt at a joke?
13:28:35  <indutny>mmalecki: and want later to do a prompt inside this process
13:28:54  <indutny>bnoordhuis: he's just asking for a ban
13:29:03  <mmalecki>bnoordhuis: yeah. looks like I'm too hangover to do jokes tho
13:29:15  <bnoordhuis>mmalecki: leave it to the professionals :)
13:29:30  <mmalecki>bnoordhuis: hangovers or jokes?
13:29:37  <bnoordhuis>maybe both
13:30:03  <mmalecki>oh well, that won't really work, I'm afraid
13:31:04  * piscisaureus_joined
13:33:17  * `3E|BRBquit (Quit: Leaving...)
13:48:35  * c4milojoined
13:48:42  * c4miloquit (Remote host closed the connection)
13:50:22  * c4milojoined
13:50:59  * piscisaureus__joined
13:51:25  * piscisaureus_quit (Read error: No route to host)
13:51:54  * piscisaureus___joined
13:52:06  * piscisaureus___changed nick to piscisaureus_
13:55:41  * piscisaureus__quit (Ping timeout: 260 seconds)
14:01:11  * loladirojoined
14:05:30  <tjfontaine>bnoordhuis: spamhaus, psbl, and spameatingmonkey are generally the ones I rely on
14:08:14  * mmaleckiquit (Ping timeout: 248 seconds)
14:11:45  * piscisaureus_quit (Read error: Connection reset by peer)
14:11:57  * piscisaureus_joined
14:20:46  * piscisaureus_quit (Ping timeout: 260 seconds)
14:25:19  * TheJHjoined
14:30:02  * `3rdEdenjoined
14:46:32  * piscisaureus_joined
14:51:39  * tomshredsquit (Quit: Linkinus - http://linkinus.com)
15:03:07  * ericktjoined
15:04:12  <indutny>tjfontaine: are you running your own mail server?
15:04:40  * ericktquit (Client Quit)
15:07:35  <tjfontaine>indutny: I maintain a few mail stacks yes
15:10:00  <indutny>why?
15:10:39  <tjfontaine>indutny: you mean in comparison to using a service like google?
15:10:45  <indutny>yes
15:11:17  <tjfontaine>indutny: some companies just prefer the control and cost benefits
15:11:52  <indutny>over stability
15:12:06  <indutny>:)
15:12:06  <tjfontaine>you overestimate googles stability
15:15:44  * AvianFlujoined
15:17:30  * ericktjoined
15:21:56  * stagasquit (Ping timeout: 272 seconds)
15:26:33  * ericktquit (Quit: erickt)
15:29:10  * loladiroquit (Read error: Connection reset by peer)
15:29:16  * loladiro_joined
15:29:40  * dylangjoined
15:38:02  <indutny>well, they have a lot of failover servers
15:38:20  <indutny>while for maintaining personal email server I can't afford more than 1 machine
15:39:21  <tjfontaine>in personal world you rarely need more than 1 MTA, 1 MDA, and 1 backup MX
15:47:11  * dapjoined
15:50:23  <piscisaureus_>since people seem to be stubborn and use path.join all the time (<-- yes, you isaacs)
15:50:33  <piscisaureus_>I will now proceed and make it work for windows
15:54:35  <indutny>everyone use path.join
15:56:24  * piscisaureus_quit (Read error: Connection reset by peer)
15:57:31  * piscisaureus_joined
15:58:51  * dylangquit (Quit: dylang)
16:05:15  * ericktjoined
16:16:51  <isaacs>piscisaureus_: what's this?
16:17:18  <piscisaureus_>isaacs: what do you mean?
16:17:25  <piscisaureus_>isaacs: oh - the path thing
16:17:39  <isaacs>piscisaureus_: i only use path.join by mistake, or for security of path.join('/', userinput)
16:17:47  * mmaleckijoined
16:17:50  <piscisaureus_>isaacs: path.join has an "oddity" in windows and it breaks st
16:17:56  <isaacs>piscisaureus_: ah
16:18:01  <isaacs>piscisaureus_: what's the oddity?
16:18:06  <isaacs>er, what's the breakage in windows?
16:18:17  <isaacs>we should have a path.mount() function
16:18:27  <piscisaureus_>isaacs: well, path.join('/', '/bla') -> '//bla'
16:18:55  <isaacs>like, path.mount = function (root, p) { return path.resolve(root, '.' + path.join('/', p)) }
16:19:27  <piscisaureus_>isaacs: you mean, never go above "root" ?
16:19:31  <isaacs>piscisaureus_: yeah
16:19:35  <piscisaureus_>ah
16:19:41  <piscisaureus_>that function doesn't do it :-)
16:19:43  <piscisaureus_>but I get the point
16:19:45  <isaacs>so you can't do HTTP GET ../../../../etc/passwd
16:19:52  <piscisaureus_>yeah I agree that would be nice
16:20:18  <isaacs>what about path.resolve(path.join(root, path.join('/', p)))?
16:21:23  * joshthecoderjoined
16:22:07  <piscisaureus_>nope
16:22:19  <isaacs>so, how would you do that in windows?
16:22:25  <piscisaureus_>ehm
16:22:29  <piscisaureus_>I don't we have a function for that
16:22:30  <isaacs>just path.resolve() and check the result?
16:22:38  <piscisaureus_>well, there's ways of course
16:22:58  * bradleymeckjoined
16:22:59  <isaacs>short term, what's the bugfix for st?
16:23:18  <piscisaureus_>isaacs: I am fixing it it in node
16:23:22  <piscisaureus_>isaacs: even more short term
16:23:32  * piscisaureus_looks it up
16:23:34  <isaacs>for 0.8? or master?
16:23:44  <piscisaureus_>- u = path.join('/', url.parse(u).pathname).replace(/\\/g, '/')
16:23:45  <piscisaureus_>+ u = ('/' + url.parse(u).pathname).replace(/[\/\\]+/g, '/')
16:24:14  <piscisaureus_>isaacs: I think we can fix it in 0.8
16:24:42  <piscisaureus_>isaacs: Let's decide after I finish the tests
16:24:46  <isaacs>kk
16:25:24  <piscisaureus_>isaacs: st is nice btw. It makes very few syscalls
16:25:33  <isaacs>thanks
16:25:37  <piscisaureus_>isaacs: so that means that JS is the bottleneck here :-)
16:25:43  <isaacs>ahh
16:25:45  <isaacs>haha
16:25:48  <isaacs>yeah, that's the goal :)
16:25:59  <piscisaureus_>isaacs: and I also really like it's interface
16:26:04  <piscisaureus_>really easy to set up and to integrate
16:26:12  <piscisaureus_>isaacs: there was one thing I missed
16:26:32  <piscisaureus_>isaacs: it should be possible to redirect people from '/dir' to '/dir/' etc
16:26:33  <isaacs>that's what you get when you don't use a framework: everything has to be versatile and easy to use, because there's no cookie-cutter interfae.
16:26:41  <isaacs>hmm... yeah
16:26:46  <isaacs>i think there is something like that, isn't there?
16:26:50  <isaacs>the autoindex does that?
16:27:03  <piscisaureus_>isaacs: doesn't for me - maybe it's just a bug
16:27:16  <piscisaureus_>isaacs: but /static gives me an error and /static/ gives me the list
16:27:47  <isaacs>hm... weird
16:27:49  <isaacs>works for me
16:27:57  <isaacs>https://npmjs.org/doc redirects to https://npmjs.org/doc/
16:28:21  <piscisaureus_>hmm
16:28:28  <piscisaureus_>maybe another windows issue
16:28:31  <piscisaureus_>will investigate
16:28:33  <isaacs>k
16:28:34  <piscisaureus_>but... first things first
16:29:13  <isaacs>piscisaureus_: dunno if you saw, there's a first pass at `npm dedupe` in teh latest npm release
16:38:07  * bradleymeck_joined
16:40:46  * bradleymeckquit (Ping timeout: 260 seconds)
16:40:47  * bradleymeck_changed nick to bradleymeck
16:41:26  <piscisaureus_>isaacs: cool
16:41:34  <piscisaureus_>isaacs: I will try it out with our repos
16:42:46  <mmalecki>isaacs: hey, what do you think about doing git-style command invocation?
16:43:10  <mmalecki>like, `npm <command-which-doesnt-exist>` would call `npm-command-which-doesnt-exist`
16:44:01  <piscisaureus_>isaacs: so how do I use the thing
16:44:15  <piscisaureus_>C:\test>npm dedupe
16:44:16  <piscisaureus_>npm ERR! Error: ENOENT, readdir 'C:\Program Files\nodejs\node_modules\npm\man\ma
16:44:16  <piscisaureus_>n1'
16:44:35  <piscisaureus_>oh nvm - it's the dedupe branch
16:45:20  * `3rdEdenquit (Quit: Leaving...)
16:47:05  <piscisaureus_>isaacs: it works :-)
16:47:34  <piscisaureus_>isaacs: the only thing is - npm ls shows a lot of "UNMET DEPENDENCY" now
16:47:44  <isaacs>hmm.... well, that shouldn't happen :)
16:47:54  <piscisaureus_>isaacs: but it seems to work and apparently it's smart about versions as well
16:47:55  <piscisaureus_>isaacs++
16:48:22  <isaacs>piscisaureus_: if you get caught in umnets, it's not working
16:51:15  * dylangjoined
16:51:19  <piscisaureus_>isaacs: https://gist.github.com/3417223
16:51:46  <piscisaureus_>isaacs: sometimes it didn't move packages far enough up
16:52:10  <piscisaureus_>it may be my setup
16:53:07  <piscisaureus_>... trying again on a fresh box
16:53:25  <isaacs>ircretary: hi
16:53:26  <ircretary>isaacs: Hello :)
16:54:21  * zippjoined
16:54:27  <piscisaureus_>ircretary: insult CAPSLOCKBOT
16:54:27  <ircretary>piscisaureus_: I'm not sure what to do with that command. Ask for help in PM.
16:55:13  <isaacs>ircretary: bomb CAPSLOCKBOT
16:55:13  <ircretary>isaacs: Arming the charges...
16:57:53  <isaacs>piscisaureus_: that's a gnarly dep graph
17:00:03  <isaacs>piscisaureus_: how do i get that locally?
17:00:12  <piscisaureus_>isaacs: I'm sorry, you cant..
17:00:21  <isaacs>ok.
17:00:30  <isaacs>can you give me a list of your top-level deps, only?
17:01:53  <isaacs>piscisaureus_: all i really care about is the package.json anyway :)
17:02:10  <piscisaureus_>isaacs: ok... a sec
17:02:12  <isaacs>omg. that's so evil...
17:02:25  <isaacs>`npm install c9` locally installs it globally
17:02:45  <isaacs>fyi: you're going to have to move your install.js to gyp eventually, or be broken
17:03:38  * TooTallNatejoined
17:04:23  <isaacs>piscisaureus_: ok... so... i did `npm install c9`, and it did some script and now there's some global stuff.
17:04:24  <creationix>wait, npm is supposed to share common deep deps between deps?
17:04:29  <isaacs>piscisaureus_: how do i remove it?
17:04:33  <creationix>I thought that feature was removed
17:04:42  <isaacs>creationix: it's never been removed
17:04:49  <piscisaureus_>isaacs: ehm, I don't know
17:04:57  <piscisaureus_>isaacs: the c9 you just got isn't the same thing
17:05:06  <creationix>what is the c9 in npm?
17:05:17  <piscisaureus_>creationix: cloud9local "new"
17:05:20  <piscisaureus_>afaik
17:05:31  <isaacs>creationix: https://npmjs.org/package/c9
17:05:47  <creationix>ahh, yes c9local
17:06:14  <creationix>oh good, hoarders depends on it
17:06:21  <creationix>`npm install hoarders` is the new forkbomb
17:08:54  <piscisaureus_>isaacs: what "npm dedupe" did to me: https://gist.github.com/4d65d8f284ab62cefe0b#file_diff.txt
17:09:33  <creationix>isaacs, so when will npm test be integrated into npm? I noticed you made it required in the `npm init` script a while back?
17:10:12  <creationix>I've been using the same travis.yml for all my projects https://github.com/c9/vfs-socket/blob/master/.travis.yml
17:10:37  * arlolrajoined
17:15:18  * hzquit (Ping timeout: 272 seconds)
17:18:59  * ericktquit (Quit: erickt)
17:22:23  * ericktjoined
17:25:59  * stagasjoined
17:30:08  * brsonjoined
17:37:09  * mordy__part
17:43:46  * tomshredsjoined
17:45:00  * isaacstopic: Liberia. Uzbekistan. Velociraptors. http://piscisaureus.no.de/libuv/latest
17:46:34  * tomshredsquit (Quit: Linkinus - http://linkinus.com)
17:49:15  * `3rdEdenjoined
17:53:44  * lohkeyjoined
18:00:14  * mikealquit (Quit: Leaving.)
18:00:34  * mikealjoined
18:01:25  * mikealquit (Client Quit)
18:03:13  * mikealjoined
18:05:48  * mikealquit (Client Quit)
18:08:39  * tomshredsjoined
18:12:06  * EhevuTovjoined
18:21:38  * mikealjoined
18:33:36  * EhevuTov_joined
18:36:38  * EhevuTovquit (Ping timeout: 252 seconds)
18:37:51  * EhevuTov_quit (Client Quit)
18:40:29  * EhevuTovjoined
18:40:52  * EhevuTovquit (Remote host closed the connection)
18:41:47  * AvianFluquit (Quit: AvianFlu)
18:42:16  * isaacstopic: v0.8.8 release tomorrow
18:57:00  <bnoordhuis>tjfontaine: thanks
18:57:42  <bnoordhuis>moderate smtpd_recipient_restrictions in postfix + spamhaus seems to catch 99.999% of spam
19:00:30  <tjfontaine>bnoordhuis: depending on the amount of flow, I usually prefer to use the dnsbls at the SA level, so you can also train bayes
19:02:14  * bradleymeck_joined
19:02:27  <bnoordhuis>tjfontaine: i don't have spamassasin running, this mx only forwards mail to another host
19:02:32  <tjfontaine>nod
19:02:56  <bnoordhuis>and the spam volume is moderately low so far, 170 spam messages in ~ 5 hours
19:03:10  <tjfontaine>ya, no need to get carried away then
19:04:52  * bradleymeckquit (Ping timeout: 272 seconds)
19:04:52  * bradleymeck_changed nick to bradleymeck
19:08:41  * mikealquit (Quit: Leaving.)
19:20:20  * `3rdEdenquit (Quit: Linkinus - http://linkinus.com)
19:21:10  * AvianFlujoined
19:26:06  <bnoordhuis>piscisaureus_: any further remarks on that child_process.exec / execFile PR?
19:26:12  <piscisaureus_>bnoordhuis: nope
19:26:27  <bnoordhuis>cool, then i'll land it
19:28:15  <CIA-131>node: Ben Noordhuis master * r63d13e8 / (4 files in 3 dirs): child_process: emit error on exec failure - http://git.io/TxjJpA
19:40:42  * mikealjoined
19:48:26  * TooTallNatequit (Quit: Computer has gone to sleep.)
19:48:51  * arlolraquit (Read error: Connection reset by peer)
19:49:07  * arlolrajoined
19:49:45  * stagasquit (Quit: ChatZilla 0.9.88-rdmsoft [XULRunner 1.9.0.17/2009122204])
19:51:41  * dshaw_joined
19:52:34  * brsonquit (Ping timeout: 240 seconds)
19:53:09  * mikealquit (Quit: Leaving.)
19:54:25  <CIA-131>node: Bert Belder wip * r8568d7b / (lib/path.js test/simple/test-path.js): path.join stuff - http://git.io/tBv98Q
19:54:32  <piscisaureus_>^-- isaacs: take a look at the test
19:54:38  * brsonjoined
19:55:37  <piscisaureus_>isaacs: failing with 0.8.7: https://gist.github.com/3418840
20:13:52  * dshaw_quit (Quit: Leaving.)
20:14:48  * TooTallNatejoined
20:23:34  * bradleymeckquit (Quit: bradleymeck)
20:25:13  * joshthecoderquit (Read error: Connection reset by peer)
20:30:08  * loladiro_quit (Read error: Connection reset by peer)
20:30:14  * loladirojoined
20:31:11  <isaacs>piscisaureus_: it does scare me a little to change path.join at all on a stable branch
20:31:59  <isaacs>piscisaureus_: also, i'd almost prefer to just say path.join isn't windows safe, and make it the same for both windows and unix.
20:32:21  <isaacs>but that's probably not so great.
20:33:58  <bnoordhuis>https://github.com/bnoordhuis/node/commit/badbd1a <- v0.8, #3900. update tls cipher list. opinions?
20:36:03  * joshthecoderjoined
20:36:13  <piscisaureus_>isaacs: I am also not sure about changing it on a stable branch
20:36:23  <piscisaureus_>isaacs: but this was to fix `st` and https://github.com/joyent/node/issues/3610
20:37:31  <piscisaureus_>isaacs: but I prefer making it doing something sensible vs doing something random :-)
20:39:56  * TheJHquit (Ping timeout: 260 seconds)
20:40:26  <piscisaureus_>isaacs: it's the same for https://github.com/joyent/node/pull/3462 -
20:40:26  <piscisaureus_>I'd rather have it either *never* work or *always* work
20:40:30  * tomshredsquit (Ping timeout: 244 seconds)
20:40:49  <piscisaureus_>isaacs: "it works only when you have installed git or msys" is not very acceptable to me
20:48:06  * joshthecoderquit (Ping timeout: 272 seconds)
20:48:15  <CIA-131>node: Bert Belder wip * r6e71b65 / (lib/path.js test/simple/test-path.js): path.join stuff - http://git.io/SkklvQ
20:51:18  * joshthecoderjoined
20:56:15  <bnoordhuis>piscisaureus_: bertje, on a mac atm?
20:57:12  <piscisaureus_>bnoordhuis: the hardware only
20:57:26  <bnoordhuis>piscisaureus_: ah okay, nvm then
21:02:32  <bnoordhuis>hah, bloody smartos - ioctl(FIOCLEX) doesn't raise an error but it doesn't actually work...
21:03:19  <piscisaureus_>hmm
21:03:32  <piscisaureus_>maybe bryan cantrill can fix that :-)
21:07:10  <bnoordhuis>that doesn't help me
21:13:57  <bnoordhuis>piscisaureus_: can you try https://gist.github.com/3419436 when you're on OS X again?
21:14:07  <bnoordhuis>no rush, i'm just curious
21:14:34  <mmalecki>piscisaureus_: I'm on OS X now, want a hand?
21:14:35  <bnoordhuis>fun fact, the fionbio.c test with USE_IOCTL=1 consistently kills my shell on smartos
21:14:50  <bnoordhuis>maybe i'm breaking some rlimit
21:14:51  <piscisaureus_>mmalecki: I think you meant bnoordhuis ?
21:14:52  <bnoordhuis>mmalecki: yes please
21:15:00  * dshaw_joined
21:15:15  <mmalecki>er, yeah, I did mean bnoordhuis
21:15:22  <mmalecki>okay, running
21:16:43  <mmalecki>bnoordhuis:
21:16:43  <mmalecki>[[email protected] gist-3419436 (master)]$ USE_IOCTL=1 time ./a.out 2.95 real 1.37 user 1.56 sys
21:16:46  <mmalecki>[[email protected] gist-3419436 (master)]$ USE_IOCTL=0 time ./a.out 3.16 real 1.43 user 1.72 sys
21:17:03  <mmalecki>hm, for some reason those ended up on one line
21:17:05  <mmalecki>oh well
21:17:06  <bnoordhuis>mmalecki: is that fioclex.c or fionbio.c?
21:17:21  <mmalecki>bnoordhuis: that was fioclex
21:17:32  <bnoordhuis>okay, good
21:17:45  <bnoordhuis>there's a 50/50 chance that fionbio is slower
21:17:53  <mmalecki>indeed it is
21:17:59  * piscisaureus_quit (Read error: Connection reset by peer)
21:17:59  <mmalecki>[[email protected] gist-3419436 (master)]$ USE_IOCTL=0 time ./a.out 9.92 real 1.52 user 8.37 sys
21:18:02  <mmalecki>[[email protected] gist-3419436 (master)]$ USE_IOCTL=1 time ./a.out 7.65 real 1.45 user 6.19 sys
21:18:23  <bnoordhuis>ah, but it's still faster when USE_IOCTL=1
21:18:24  <bnoordhuis>good
21:18:28  <bnoordhuis>thanks maciej
21:18:31  <mmalecki>np
21:19:36  * piscisaureus_joined
21:22:52  <isaacs>piscisaureus_: hey
21:22:53  <CIA-131>libuv: Ben Noordhuis master * rb12b649 / src/unix/core.c : darwin, freebsd: use ioctl(FIOCLEX) and ioctl(FIONBIO) - http://git.io/oEmH8w
21:23:05  <isaacs>hey, btw, cia is back, w00t!
21:23:06  <piscisaureus_>hey isaacs
21:23:38  <isaacs>piscisaureus_: so, i think path.join on windows should be unix path.join(..).replace(/\//g, '\\')
21:24:09  <isaacs>piscisaureus_: that would make it consistent.
21:24:24  <piscisaureus_>isaacs: yeah, but that would make it actively destroy unc paths
21:24:26  <isaacs>piscisaureus_: as for the %Home% env thing, fuck that noise. ~/.node_modules was a mistake from day 1
21:24:32  * travis-cijoined
21:24:32  <travis-ci>[travis-ci] joyent/libuv#603 (master - b12b649 : Ben Noordhuis): The build passed.
21:24:32  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/37173f851524...b12b6498d612
21:24:32  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/2193903
21:24:32  * travis-cipart
21:24:36  <isaacs>piscisaureus_: so don't use it for unc paths.
21:24:41  <isaacs>piscisaureus_: it also destroys urls.
21:24:46  <isaacs>that's why we don't use it for url paths.
21:24:51  <isaacs>it should be much stupider.
21:24:55  <isaacs>path.resolve is the smart one
21:24:57  <piscisaureus_>isaacs: well .. hmmpf
21:25:04  <bnoordhuis>btw, any opinions on that tls cipher list thing? the 30 second merge window starts now
21:25:06  <piscisaureus_>isaacs: that was my opinion too
21:25:11  <isaacs>heh
21:25:28  <mjr_>I think path.join should just fix / vs \ depending on unix vs. win.
21:25:29  <isaacs>bnoordhuis: tls cipher thing?
21:25:33  <piscisaureus_>isaacs: but then it turns out that people are very stubborn and keep using path.join
21:25:39  <bnoordhuis>isaacs: https://github.com/bnoordhuis/node/commit/badbd1a
21:25:59  <isaacs>bnoordhuis: that commit hash makes me think it's going to be a bad day
21:26:09  <bnoordhuis>i thought it was awesome :)
21:26:13  <piscisaureus_>isaacs: so now what will happen if you *do* use unc paths it will blow up in terrible ways... So I made it smart enough
21:26:31  <piscisaureus_>to handle the only realistic case where it matters
21:26:35  <isaacs>bnoordhuis: yeah, beastly
21:26:41  <isaacs>bnoordhuis: +1 from me
21:26:44  <piscisaureus_>namely, "\\server" as the first component
21:26:50  <bnoordhuis>cool, i'll merge it
21:27:08  <isaacs>piscisaureus_: it'll also blow up http://foo.com/boo
21:27:13  <CIA-131>node: Ben Noordhuis v0.8 * rbadbd1a / (doc/api/tls.markdown lib/tls.js): tls: update default cipher list - http://git.io/qhJX2w
21:27:23  <piscisaureus_>isaacs: yeah but process.cwd() will never return that :-)
21:27:27  <isaacs>> path.join("http://foo.com/bar")
21:27:28  <isaacs>'http:/foo.com/bar'
21:27:40  <isaacs>piscisaureus_: path.join should not be used on process.cwd() ;)
21:28:05  <isaacs>but we could make it smart enough to not club \\ at the start of the path
21:28:15  <piscisaureus_>isaacs: that's basically what I did
21:28:28  <isaacs>if it does that, it should do it on unix as well.
21:28:35  <piscisaureus_>isaacs: there's a lot of "questionable" cases in the tests - which are just to catch changes when they happen
21:28:42  <piscisaureus_>isaacs: well, UNC paths don't work on unix right?
21:28:43  <isaacs>path.join shouldn't be platform aware except in the specific slash type it returns
21:28:58  <isaacs>piscisaureus_: no, but //etc/passwd === /etc/passwd on every unix i'm familiar with
21:29:07  <piscisaureus_>so, why bother?
21:29:29  <isaacs>mostly, path.join is useful for url paths, when you know that you're only dealing wiht the path portion
21:29:38  <isaacs>or known-relative paths.
21:29:42  <isaacs>that's why st uses it
21:30:02  <piscisaureus_>that's what I just made "just work"
21:30:06  <isaacs>because //foo//bar is *not* the same as /foo/bar in a url
21:30:08  <piscisaureus_>without even breaking unc paths
21:30:42  <piscisaureus_>isaacs: you could argue that it is still a little too clever
21:30:46  <isaacs>piscisaureus_: ok, it sounds like we mostly agree, so i'll stop debating you and just review the patch on your wip ;)
21:30:50  <piscisaureus_>isaacs: but making it dumber is more work :-)
21:32:26  <isaacs>timezones, character encodings, windows paths...
21:32:32  * isaacsmaking a list of things to set on fire
21:32:32  <piscisaureus_>ya
21:32:47  <piscisaureus_>my personal preference: signals and symlinks :-)
21:32:59  <piscisaureus_>version differences
21:33:04  * loladiroquit (Ping timeout: 272 seconds)
21:33:04  <piscisaureus_>^-- KILL IT!
21:33:20  <mmalecki>bnoordhuis: I think you had a tale to tell about signals
21:33:35  <bnoordhuis>mmalecki: yeah... don't get me started, i'll cry
21:33:38  <isaacs>seriously, though, why didn't microsoft convert to unix paths like 15 years ago when it was clear that their thing was idiotic?
21:33:44  <mmalecki>bnoordhuis: pleeeeeeaaaase!
21:33:52  <mmalecki>bnoordhuis: it's perfect time for scary stories
21:34:23  <bnoordhuis>it's not a really good story though, it's mostly that it seemed like a good idea at the time
21:34:36  <bnoordhuis>well, actually... not everyone agreed, even back then
21:34:42  <mmalecki>so why did it seem like a good idea? and why is it bac now?
21:34:58  <bnoordhuis>oh, i'm talking about unix signals in general
21:35:12  <bnoordhuis>not how node deals with them
21:35:14  <mmalecki>yeah, this is what I mean
21:35:36  <mmalecki>node's way of dealing with them is pretty pleasant
21:36:23  <bnoordhuis>yeah, it hides all the nastiness and do { r = syscall(); } while (r == -1 && errno == EINTR) stuff
21:37:35  <mmalecki>and why are signals wrong, exactly? except APIs. why were they a good idea back then?
21:38:15  <bnoordhuis>mmalecki: the idea was that you had to have some way to asynchronously notify programs
21:38:31  <bnoordhuis>e.g. a program that's blocked in select()
21:38:59  <bnoordhuis>back then, threads were what you got when you wore out your socks
21:39:07  <bnoordhuis>so they invented signals
21:39:12  <isaacs>piscisaureus_: you know what was awesome? when path.join() was arguments.join('/') and path.split() was p.split('/')
21:39:19  <isaacs>piscisaureus_: + dots
21:39:41  <isaacs>if you wanted a slash at the start, you put a '' at the start of the args
21:39:44  <isaacs>LIKE A GENTLEMAN.
21:40:01  * isaacsmost of the way through the tests... looks pretty sane so far..
21:40:06  <bnoordhuis>isaacs: re rpm and debs, do you want to do that for v0.8 or master?
21:40:29  <isaacs>bnoordhuis: v0.8
21:40:38  <mmalecki>bnoordhuis: I see. the concept appears to be quite appealing. what's wrong with it?
21:40:38  <bnoordhuis>okay
21:40:53  <isaacs>bnoordhuis: you shouldn't need to change any api, i would think. just some build makefile junk, right?
21:41:05  <bnoordhuis>mmalecki: well, for one, almost everyone gets it wrong
21:41:31  <bnoordhuis>mmalecki: second, now you need to guard every blocking syscall with 'if (errno == EINTR)'
21:41:36  <isaacs>piscisaureus_: so, the bit that i'm kind of iffy on starts with: [['c:'], 'c:.'],
21:41:45  <bnoordhuis>mmalecki: thirdly, they're really just a hack
21:41:59  <bnoordhuis>isaacs: yes
21:44:21  <mmalecki>bnoordhuis: I see. EINTR is what happens when a call is interrupted by a signal, right?
21:44:27  <piscisaureus_>isaacs: ya. I didn't change that
21:44:39  <piscisaureus_>isaacs: that is the status quo. It should be fixed
21:44:49  <isaacs>piscisaureus_: hm.
21:45:25  <isaacs>yeah, i'd think that ['c:', ''] or ['c:'] would be just 'c:'
21:45:52  <isaacs>[['c:', 'file'], 'c:/file'] is particularly weird
21:46:23  <isaacs>i mean, i'ts a join, so it should join with /, but c:file is a valid thing
21:46:31  <piscisaureus_>some more things on my hitlist:
21:46:32  <piscisaureus_>tabs, pointer alignment, line endings, apple keyboards, current working directories
21:46:57  * isaacshas a smaller hitlist than piscisaureus_
21:47:00  <piscisaureus_>isaacs: so you want join('c:', 'file') -> 'c:file'
21:47:00  <isaacs>piscisaureus_: you need to focus
21:47:07  <isaacs>my next two were citizenship and marriage.
21:47:26  <isaacs>the list is really, "dated wrong concepts that everyone sane should now admit were just big mistakes"
21:47:33  <piscisaureus_>isaacs: kids
21:47:36  <mmalecki>isaacs: immigration policies and visas
21:47:49  <piscisaureus_>isaacs: I think 'c:file' would be a huge surprise for many people
21:47:51  <isaacs>mmalecki: if there's no citizenship, then "immigration" doesn't mean anything
21:47:59  <isaacs>piscisaureus_: yeah, i guess it should be c:\file
21:48:07  <isaacs>piscisaureus_: but then why even have c:.?
21:48:16  <piscisaureus_>isaacs: i'd rather join('c:') -> 'c:/' if we have to change this
21:48:27  <isaacs>hm. yeah, that's kind of what you expect, probably
21:48:40  <isaacs>also, can we make path.join just *always* return / separators?
21:48:44  <isaacs>they work on all windows, right?
21:49:02  <isaacs>it's really annoying that using path.join on url paths has to have that stupid extra step in st to switch from \ to /
21:49:21  <piscisaureus_>isaacs: heh - that's what the dutch people would call a "question of conscience"
21:49:33  <piscisaureus_>isaacs: I'd like to pretend they always work :-)
21:50:16  <piscisaureus_>isaacs: I could fill an hourlong talk about windows path handling and everyone would fall asleep (including myself)
21:50:18  <isaacs>really, if you care about it referring to a real thing on the filesystem, and being correct for your platform, then you want path.resolve()
21:50:26  <piscisaureus_>ya
21:50:41  <piscisaureus_>let's path.join('c:') -> 'c:/'
21:50:47  <isaacs>ugh. ok.
21:50:49  <isaacs>yes.
21:51:13  <isaacs>if you want to update the tests (and, for bonus points, the impl) to that, then lgtm for master.
21:51:13  <piscisaureus_>isaacs: although I don't think 'c:\\' would be a big surprise and it looks more familiar
21:51:33  <isaacs>piscisaureus_: it's more that path.join(url.parse(req.url).pathname) should not give me \ chars on windows.
21:51:39  <isaacs>since i'm not even talking about the fs
21:51:54  <isaacs>i just don't want you going above the root i'm giving you access to
21:52:03  <isaacs>and path.join('/', input) is a nice way to do that
21:52:17  <piscisaureus_>hmm
21:52:18  <isaacs>path.mount(root, req.url) would be even nicer.
21:52:25  <piscisaureus_>let me sleep on it
21:52:32  <isaacs>and we could probably just do that, and use that instead.
21:52:43  <piscisaureus_>I would be okay with path.mount
21:52:45  <isaacs>and then path.join can use whatever /\ chars it wants, i don't care.
21:53:22  <isaacs>path.mount(root, request) is nicer than path.join(root, path.join('/', request)) anyway
21:53:28  <isaacs>the intent is clearer
21:53:40  <piscisaureus_>I agree
21:53:53  <piscisaureus_>isaacs: ok, I have to sit on this for a while (and I am about to go home, anyway)
21:53:59  <isaacs>i can also alwasy do x = path.resolve(root, request); and then test that x.indexOf(root) === 0
21:54:12  <piscisaureus_>isaacs: but yeah - it would be good to do this in core because it is so common and so easy to get wrong
21:54:22  <isaacs>and dangerous to get wrong.
21:54:31  <isaacs>kewl, thanks, man
21:54:39  <piscisaureus_>np
21:54:42  <isaacs>piscisaureus_: also, dedupe just fixed a problem for me
21:54:43  <piscisaureus_>I get paid for it
21:54:46  <isaacs>so i'm glad that you made me build that :)
21:54:49  <piscisaureus_>haha
21:54:53  <isaacs>very gratifying
21:54:54  <piscisaureus_>I am sure we will be too :-)
21:55:04  <piscisaureus_>good work isaacs
21:55:09  <isaacs>np
22:03:23  * dylang_joined
22:07:07  * c4miloquit (Remote host closed the connection)
22:07:51  * dylang_quit (Ping timeout: 244 seconds)
22:07:54  * dylangquit (Ping timeout: 272 seconds)
22:08:36  * brsonquit (Ping timeout: 268 seconds)
22:09:34  * brsonjoined
22:11:48  <piscisaureus_>isaacs: btw - yesterday I upgraded v8 so the object comparison bug is now fixed
22:11:58  <isaacs>piscisaureus_: i saw that, thanks
22:12:14  <isaacs>0.8.8 tomorrow, it'll make some confused people less confused :)
22:13:39  * rendarquit
22:17:14  * bradleymeckjoined
22:25:24  * tomshredsjoined
22:29:55  * tomshredsquit (Client Quit)
22:30:17  * tomshredsjoined
22:30:49  <CIA-131>node: isaacs v0.8 * r2bcb9ab / (170 files in 28 dirs): npm: Upgrade to 1.1.55 - http://git.io/N0xsIg
22:31:39  * AvianFluquit (Quit: AvianFlu)
22:38:12  * bradleymeckquit (Quit: bradleymeck)
22:47:22  <CIA-131>libuv: Bert Belder master * r564e7c7 / (4 files in 3 dirs): windows: emit SIGWINCH when the console size changes - http://git.io/0uKkHQ
22:49:06  * travis-cijoined
22:49:06  <travis-ci>[travis-ci] joyent/libuv#604 (master - 564e7c7 : Bert Belder): The build passed.
22:49:06  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/b12b6498d612...564e7c765cf7
22:49:06  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/2194777
22:49:06  * travis-cipart
22:54:47  * bradleymeckjoined
22:55:24  <CIA-131>node: Bert Belder master * red093f1 / (13 files in 5 dirs): uv: upgrade to 564e7c7 - http://git.io/f_E1dw
23:00:23  <piscisaureus_>hmm
23:00:29  <piscisaureus_>shit, forgot one thing
23:02:01  * brsonquit (Read error: Operation timed out)
23:02:19  * brsonjoined
23:03:35  <TooTallNate>piscisaureus_: bnoordhuis: indutny's apple select() thing made it into libuv already?
23:03:44  <piscisaureus_>TooTallNate: master only
23:03:47  * dylangjoined
23:03:59  <TooTallNate>piscisaureus_: how come?
23:04:05  <TooTallNate>not v0.8 I mean
23:04:31  <piscisaureus_>TooTallNate: needs work to fix it in a non-ABI-breaking way
23:04:52  <TooTallNate>ahhh, makes sense
23:05:03  <CIA-131>libuv: Bert Belder master * rf03f14c / src/win/tty.c : windows: lock tty output before updating the virtual window size - http://git.io/dHtDoQ
23:06:45  * travis-cijoined
23:06:45  <travis-ci>[travis-ci] joyent/libuv#605 (master - f03f14c : Bert Belder): The build passed.
23:06:45  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/564e7c765cf7...f03f14cf3df2
23:06:45  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/2194903
23:06:45  * travis-cipart
23:07:17  <piscisaureus_>within the 2 minute window... ?
23:07:17  <CIA-131>libuv: Bert Belder master * r03bdfd0 / src/win/tty.c : windows: lock tty output before updating the virtual window size - http://git.io/EslMSA
23:07:22  <piscisaureus_>yay
23:08:57  * travis-cijoined
23:08:57  <travis-ci>[travis-ci] joyent/libuv#606 (master - 03bdfd0 : Bert Belder): The build passed.
23:08:57  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/f03f14cf3df2...03bdfd08ccc7
23:08:57  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/2194943
23:08:57  * travis-cipart
23:10:33  <TooTallNate>wow
23:10:40  <TooTallNate>./node < /dev/tty works on master
23:10:43  <TooTallNate>this is a glorious moment
23:10:46  <TooTallNate>indutny++
23:10:49  <TooTallNate>indutny++
23:10:49  <TooTallNate>indutny++
23:11:09  <bnoordhuis>now explain why you care so much about that
23:11:26  <piscisaureus_>ya
23:11:28  <TooTallNate>bnoordhuis: piping to "cdir"
23:11:35  <piscisaureus_>why does that even matter ? :-/
23:11:48  <TooTallNate>piscisaureus_: do you look at json all day?
23:12:09  <piscisaureus_>TooTallNate: no, XML. Why?
23:13:19  <bnoordhuis>i'm partial to ber encoded asn.1 myself
23:14:04  <TooTallNate>"curl http://registry.npmjs.org/-/all?scripts=install,preinstall,postinstall | cdir"
23:14:15  <TooTallNate>it saves a step
23:14:28  <TooTallNate>"curl http://registry.npmjs.org/-/all?scripts=install,preinstall,postinstall > out.json; cdir out.json"
23:14:41  <TooTallNate>and doesn't create a file in my cwd :)
23:14:56  <TooTallNate>i know the other cdir'ers will be happy as well
23:15:06  * arlolraquit (Quit: Linkinus - http://linkinus.com)
23:15:36  <bnoordhuis>TooTallNate: i think i've met them both
23:15:47  <TooTallNate>lulz :p
23:16:05  <piscisaureus_>http://support.microsoft.com/kb/299349 <-- practical
23:16:15  <piscisaureus_>Q: how to remove duplicate emails
23:16:24  <piscisaureus_>A: select all duplicate emails and press delete
23:16:27  <piscisaureus_>Thanks, microsoft
23:20:06  * blackorzarjoined
23:21:22  * brsonquit (Ping timeout: 272 seconds)
23:27:46  * c4milojoined
23:28:38  <CIA-131>node: isaacs v0.8 * r024f4b1 / (104 files in 8 dirs): npm: Upgrade to 1.1.56 - http://git.io/T1P2Jw
23:30:42  * toothrotquit (Ping timeout: 246 seconds)
23:32:13  * brsonjoined
23:34:21  * dylangquit (Quit: dylang)
23:34:38  <isaacs>TooTallNate, indutny: nice
23:34:58  * bradleymeckquit (Quit: bradleymeck)
23:39:14  * mmaleckiquit (Ping timeout: 265 seconds)
23:41:56  * tomshredsquit (Read error: Connection reset by peer)
23:41:57  * tommybergeronjoined
23:42:21  * tommybergeronquit (Client Quit)
23:46:30  * bnoordhuisquit (Ping timeout: 244 seconds)
23:55:49  * piscisaureus_quit (Read error: Connection reset by peer)
23:57:43  * piscisaureus_joined