00:00:01  * c4miloquit (Ping timeout: 248 seconds)
00:00:01  * ircretaryquit (Remote host closed the connection)
00:00:08  * ircretaryjoined
00:00:56  * bradleymeckjoined
00:03:02  <isaacs>hm. no hope for a clean revert, and manually reverting doesn't fix it.
00:03:12  <isaacs>boy, when bisect works, it really works, and when it doesn't, it really doesn't.
00:04:44  * jmar777quit (Remote host closed the connection)
00:05:20  * jmar777joined
00:09:25  <isaacs>wait, it looks like thatmight not be the culprit...
00:09:42  * jmar777quit (Ping timeout: 264 seconds)
00:10:09  * paddybyersquit (Ping timeout: 248 seconds)
00:10:14  <isaacs>def broke between 0.9.0 and 0.9.1
00:11:48  * bradleymeckquit (Quit: bradleymeck)
00:12:12  * bnoordhuisjoined
00:17:51  * lohkeypart
00:23:27  <loladiro>Has anybody ever integrated libuv with Tcl?
00:24:18  <CoverSlide>that requires someone to actually like tcl
00:26:10  <isaacs>aha
00:26:11  <isaacs>6bec5440ebced1bc2b91c1c9f3fbc7a361995331 is the first bad commit
00:26:17  <isaacs>bnoordhuis: ping
00:26:48  <isaacs>bnoordhuis: https://gist.github.com/4506976 <-- this test fails ever since 6bec5440ebced1bc2b91c1c9f3fbc7a361995331 uv: upgrade to 8073a26
00:27:36  <isaacs>ie, the child ends immediately.
00:28:39  * karupaneruraquit (Excess Flood)
00:30:28  * karupanerurajoined
00:35:31  <bnoordhuis>isaacs: is it supposed to print ok=true ("SIGKILL") ?
00:35:36  <isaacs>bnoordhuis: yes.
00:35:45  <isaacs>https://gist.github.com/4507016
00:35:53  <isaacs>that's as far as i can get bisecting libuv
00:36:12  <isaacs>between the git-bisect-bad commit, and the bisect-good commit, the test crashes with an assertion failure.
00:37:05  <isaacs>bnoordhuis: do you see that test work on master?
00:37:09  <bnoordhuis>no
00:37:09  <isaacs>bnoordhuis: if so, what os?
00:37:12  <isaacs>oh, ok :)
00:37:28  <isaacs>it looks like things work properly when it's not a TTY
00:38:12  <bnoordhuis>or stdio:'inherit' is not specified
00:40:25  <isaacs>sure
00:40:56  <isaacs>looks like sharing tty fd's with the child proc broke
00:42:56  <bnoordhuis>well
00:43:01  <bnoordhuis>that's always been kind of broken
00:43:14  <bnoordhuis>node leaves the fds in non-blocking mode
00:43:22  <bnoordhuis>not all programs handle that well
00:43:44  <isaacs>i take that back, it fails when the fds are not tty's
00:43:56  <isaacs>bnoordhuis: no, that's been broken since 6bec5440ebced1bc2b91c1c9f3fbc7a361995331
00:44:03  <isaacs>bnoordhuis: `npm explore` depends on that
00:44:26  <bnoordhuis>that was just a general observation
00:44:30  <isaacs>bnoordhuis: almost anyone using {stdio:'inherit'} is going to expect that the child process doesn't immediately terminate.
00:44:33  <isaacs>sure.
00:44:37  <isaacs>it's been *a bit weird* forever.
00:44:57  <isaacs>but bash, vim, many other programs, work fine, as long as the stdio fd's are tty
00:45:02  <isaacs>nonblocking or not
00:45:25  <isaacs>(actually, running node as a child proc in vim using ! is not so great)
00:45:31  <isaacs>but running vim as an inherited child is fine
00:49:26  <bnoordhuis>something is installing a ton more signal handlers
00:49:59  <bnoordhuis>or causing bash to install a ton more
00:50:42  <bnoordhuis>and that test with v0.8 leaves the console in a half-broken state :/
00:50:52  <isaacs>really?
00:50:56  <isaacs>it works perfectly for me on 0.8
00:51:17  <isaacs>the child process starts up, and if i don't kill it, i can do my bash stuff, and the script gets a 'exit' event when i ^D
00:51:53  <isaacs>but `cat` should be just as good a test
00:52:48  <isaacs>is there something like `cat` on windows that'll not do much, butnot exit right away?
00:53:31  <isaacs>ah, `type`
00:54:33  <isaacs>ah, fails when it doesn't get an argument, though
00:54:36  * bradleymeckjoined
00:59:59  * mikealquit (Quit: Leaving.)
01:01:06  <MI6>joyent/node: Nathan Rajlich v0.8 * 6a9ca14 : make binary: allow custom config flags For a use-case, see: https://gist - http://git.io/MirMOw
01:01:58  <bnoordhuis>+5700 open("/dev/null", O_RDONLY) = 12
01:02:01  <bnoordhuis>+5700 dup2(12, 0) = 0
01:02:04  <bnoordhuis>+5700 close(12) = 0
01:02:04  <bnoordhuis>well, that kind of explains it
01:02:53  <isaacs>hm. that's odd.
01:03:13  <isaacs>bnoordhuis: so we're closing the child's stdin?
01:03:35  <bnoordhuis>yes
01:03:54  <bnoordhuis>let's see when and why that changed
01:04:23  <isaacs>bnoordhuis: i'll bet you a beer it was in one of these-> https://gist.github.com/4507016
01:04:53  <bnoordhuis>47f496a seems like a likely culprit
01:05:45  <isaacs>ah, yes, it does
01:06:57  <MI6>joyent/node: Nathan Rajlich v0.8 * 4fa3fd1 : make binary: allow custom config flags For example, to cross-compile fro - http://git.io/Al6dfA
01:11:42  * piscisaureus_joined
01:14:17  * bradleymeckquit (Quit: bradleymeck)
01:15:11  <MI6>joyent/node: Nathan Rajlich master * b916774 : configure: define "arm_fpu" and "arm_neon" for ARM v8's common.gypi file (+1 more commits) - http://git.io/uaM6UA
01:15:32  <piscisaureus_>yay
01:16:44  <MI6>joyent/node: isaacs created branch stdio-inherit-broken - http://git.io/nUOx-g
01:16:58  <isaacs>bnoordhuis: failing test is on that branch^
01:17:05  <isaacs>bnoordhuis: sounds like you've got an idea of what's up, though
01:17:21  <isaacs>piscisaureus_: good evening
01:17:32  <piscisaureus_>hello isaacs
01:21:28  <bnoordhuis>isaacs: https://gist.github.com/a314a7541601bdc47976 <- can you confirm that that patch fixes it for you?
01:22:26  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
01:22:28  <isaacs>bnoordhuis: nope.
01:22:35  <isaacs>$ ./node ../wstester/bash.js
01:22:35  <isaacs>ok=false (null)
01:22:36  <isaacs>fail 1 null
01:22:43  <isaacs>$ ./node test/simple/test-child-process-stdio-inherit.js
01:22:43  <isaacs>events.js:69
01:22:43  <isaacs> throw arguments[1]; // Unhandled 'error' event
01:22:43  <isaacs> ^
01:22:43  <isaacs>Error: write EPIPE
01:22:46  <isaacs> at errnoException (net.js:847:11)
01:22:46  * piscisaureus_joined
01:22:48  <isaacs> at Object.afterWrite (net.js:666:19)
01:22:51  * indexzerojoined
01:22:51  <isaacs> at process._makeCallback (node.js:299:20)
01:23:13  <bnoordhuis>isaacs: you recompiled, didn't you?
01:23:20  <isaacs>yeah
01:23:34  <bnoordhuis>i figured you did but i ask just in case :)
01:23:39  <bnoordhuis>can you try something?
01:23:49  * indexzeroquit (Client Quit)
01:24:06  <isaacs>bnoordhuis: it's a *different* error :)
01:24:09  * indexzerojoined
01:24:24  <isaacs>sure
01:24:40  <bnoordhuis>isaacs: https://gist.github.com/80b41ddc81345378e118 <- what does that print?
01:25:23  <isaacs>https://gist.github.com/4507219
01:25:32  <bnoordhuis>hmm, it fixes that one test case but it kills many test/simple ones
01:26:00  <bnoordhuis>isaacs: no, i mean when you run that test case you linked me to
01:26:04  <bnoordhuis>the one that spawns bash
01:26:45  <isaacs>oh, ok, one sec
01:28:00  <isaacs>$ ./node ../wstester/bash.js
01:28:01  <isaacs>fd=0 close_fd=-1 use_fd=0
01:28:01  <isaacs>fd=1 close_fd=-1 use_fd=1
01:28:01  <isaacs>fd=2 close_fd=-1 use_fd=2
01:28:01  <isaacs>ok=true ("SIGKILL")
01:28:03  <isaacs>yes, that works
01:28:17  <bnoordhuis>okay, cool
01:29:32  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
01:30:11  * mikealjoined
01:30:29  <bnoordhuis>isaacs: i'm off to bed. i'll look at it some more tomorrow
01:30:39  <isaacs>kewl
01:30:42  <isaacs>thanks
01:35:00  * bnoordhuisquit (Ping timeout: 252 seconds)
01:37:37  * abraxas_joined
01:46:38  * stagas_joined
01:48:48  * stagasquit (Ping timeout: 264 seconds)
01:51:30  * stagas_quit (Ping timeout: 252 seconds)
01:52:30  * dapquit (Quit: Leaving.)
01:58:55  <MI6>joyent/node: Andreas Madsen master * bb1c039 : path: fix bugs related to paths with trailing slashes - http://git.io/AwvLFw
02:12:50  * philips_quit (Excess Flood)
02:14:03  * bradleymeckjoined
02:17:10  <MI6>joyent/node: isaacs master * dc0c524 : http: Set _dumped=false initially The better to keep the IncomingMessage - http://git.io/qxHRNA
02:17:54  * pooyajoined
02:21:25  * philips_joined
02:25:12  * piscisaureus_joined
02:25:18  <piscisaureus_>:-(
02:36:47  * indexzeroquit (Quit: indexzero)
02:38:24  * indexzerojoined
02:39:11  * piscisaureus_quit (Ping timeout: 255 seconds)
02:48:05  * brsonquit (Quit: leaving)
02:48:35  * brsonjoined
03:50:39  * brsonquit (Ping timeout: 276 seconds)
03:51:29  * kazuponjoined
03:53:50  * kazuponquit (Remote host closed the connection)
03:54:07  * kazuponjoined
03:54:56  * kazuponquit (Read error: Connection reset by peer)
03:55:12  * kazuponjoined
03:59:56  * bradleymeckquit (Quit: bradleymeck)
04:00:29  * kazuponquit (Read error: Connection reset by peer)
04:01:41  * kazuponjoined
04:07:16  * Benviejoined
04:10:15  * kazuponquit (Ping timeout: 265 seconds)
04:20:32  * pooyaquit (Quit: pooya)
04:20:57  * kazuponjoined
04:34:12  * abraxas_quit (Remote host closed the connection)
04:39:02  * brsonjoined
05:01:11  * mikealquit (Quit: Leaving.)
05:17:02  * c4milojoined
05:22:39  * c4miloquit (Ping timeout: 276 seconds)
05:31:27  * mikealjoined
05:39:21  * perezdjoined
05:40:53  * perezdquit (Client Quit)
05:41:02  * perezdjoined
05:48:58  * perezdquit (Quit: perezd)
05:51:53  * abraxas_joined
05:52:06  * joshthecoderquit (Quit: Linkinus - http://linkinus.com)
05:57:43  * jroessleinjoined
05:58:12  * jroessleinquit (Remote host closed the connection)
06:00:25  * mikealquit (Quit: Leaving.)
06:12:16  * wolfeidauquit (Remote host closed the connection)
06:17:44  * loladiroquit (Quit: loladiro)
06:21:19  * mikealjoined
06:23:31  * loladirojoined
06:25:43  * loladiroquit (Client Quit)
06:34:53  * paddybyersjoined
06:40:56  * bradleymeckjoined
06:52:36  * wolfeidaujoined
07:01:59  * loladirojoined
07:05:56  * mikealquit (Quit: Leaving.)
07:12:35  * mikealjoined
07:20:40  * loladiroquit (Quit: loladiro)
07:25:50  * loladirojoined
07:26:56  * loladiroquit (Client Quit)
07:30:35  * brsonquit (Read error: Connection reset by peer)
07:35:02  * `3rdEdenjoined
07:37:25  * rendarjoined
07:45:48  * EhevuTovjoined
08:38:44  * AvianFluquit (Remote host closed the connection)
08:56:24  * bradleymeckquit (Quit: bradleymeck)
08:59:12  * bradleymeckjoined
09:08:41  * EhevuTovquit (Quit: This computer has gone to sleep)
09:33:26  * kazupon_joined
09:36:11  * kazuponquit (Ping timeout: 252 seconds)
10:14:05  * kazupon_quit (Remote host closed the connection)
10:14:08  * indexzeroquit (Quit: indexzero)
10:17:40  * abraxas_quit (Remote host closed the connection)
10:18:13  * abraxas_joined
10:22:38  * abraxas_quit (Ping timeout: 252 seconds)
10:34:28  * bradleymeckquit (Quit: bradleymeck)
10:35:06  * stagasjoined
10:57:28  * bnoordhuisjoined
10:58:19  <indutny>bnoordhuis: hey man
10:58:27  <bnoordhuis>indutny: hey
10:58:35  <indutny>bnoordhuis: I'm having a hard time trying to understand how this thing is working https://github.com/joyent/node/commit/46a489be73fd87fa3e17dd929c53d9e90860f80a
10:58:40  <indutny>bnoordhuis: can you help me a little?
10:58:55  <bnoordhuis>indutny: what do you want to know?
10:59:06  <indutny>bnoordhuis: why does it need negative sized arrays
10:59:17  <indutny>in typedefs
10:59:20  <bnoordhuis>indutny: it doesn't, it's a kind of static assert
10:59:29  <indutny>ah
10:59:32  <indutny>oh
10:59:38  <indutny>interesting
10:59:58  <bnoordhuis>what frankensystem is that where float != 32 bits?
11:00:05  <indutny>erm
11:00:07  <indutny>probably not this line
11:00:10  <indutny>let me look at build
11:00:17  <indutny>line 150
11:00:26  <indutny>ah, this line
11:00:33  <indutny>that is smartos
11:01:00  <bnoordhuis>indutny: check what sizeof(uint32_t) and sizeof(float) is
11:03:25  <indutny>bnoordhuis: one sec
11:05:07  <indutny>4 4
11:05:09  <indutny>odd
11:05:21  <indutny>../src/v8_typed_array_bswap.h: In function 'T v8_typed_array::SwapBytes(T) [with T = signed char]':
11:05:21  <indutny>../src/v8_typed_array_bswap.h:150:23: instantiated from 'T v8_typed_array::LoadAndSwapBytes(void*) [with T = signed char]'
11:05:21  <indutny>../src/v8_typed_array.cc:694:7: instantiated from 'static v8::Handle<v8::Value> {anonymous}::DataView::getGeneric(const v8::Arguments&) [with T = signed char]'
11:05:21  <indutny>../src/v8_typed_array.cc:738:40: instantiated from here
11:05:21  <indutny>../src/v8_typed_array_bswap.h:125:16: error: size of array is negative
11:05:42  <indutny>oh wait
11:05:46  <indutny>wrong line
11:06:42  <indutny>bnoordhuis: ^
11:09:47  <bnoordhuis>indutny: ah
11:09:56  <indutny>yeah, so matching has failed
11:10:13  <bnoordhuis>what gcc version is that?
11:10:20  <indutny>4.6.2
11:11:08  <bnoordhuis>indutny: what happens when you add a template with T==signed char?
11:11:23  <indutny>sec
11:12:17  <indutny>building
11:13:53  <indutny>template <>
11:13:53  <indutny>inline unsigned char SwapBytes(unsigned char x) { return x; }
11:13:55  <indutny>this ^ ?
11:14:08  <indutny>oh
11:14:10  <indutny>signed char
11:14:24  <indutny>yes, it works now
11:19:52  <bnoordhuis>okay
11:20:17  <bnoordhuis>odd that it doesn't pick up the int8_t template
11:20:32  <bnoordhuis>what is int8_t typedef'd as?
11:25:07  <indutny>too many questions :)
11:30:53  * hzjoined
11:49:21  * qmx|awaychanged nick to qmx
12:04:42  * paddybyersquit (Ping timeout: 240 seconds)
12:10:34  * stagas_joined
12:12:12  * stagasquit (Ping timeout: 240 seconds)
12:12:14  * stagas_changed nick to stagas
12:12:17  * sgallaghjoined
12:37:28  * paddybyersjoined
12:38:10  * stagas_joined
12:40:12  * stagasquit (Ping timeout: 244 seconds)
12:40:39  * c4milojoined
12:41:15  * stagasjoined
12:43:03  * stagas__joined
12:43:27  * stagas_quit (Ping timeout: 255 seconds)
12:45:41  * stagasquit (Ping timeout: 240 seconds)
12:45:43  * stagas__changed nick to stagas
13:00:31  <bnoordhuis>why is `git push` so slow today?
13:01:14  <MI6>joyent/node: Ben Noordhuis master * 31fc52a : deps: upgrade libuv to 9aab5d4 - http://git.io/g_hoGA
13:02:24  <MI6>joyent/libuv: Ben Noordhuis master * 9aab5d4 : Revert "unix: improve uv_guess_handle() implementation" This reverts com (+3 more commits) - http://git.io/8xw0xg
13:02:55  <bnoordhuis>pff, github is broken-ish again
13:04:04  <bnoordhuis>All systems operational <- lies1
13:04:51  * travis-cijoined
13:04:51  <travis-ci>[travis-ci] joyent/libuv#1017 (master - 9aab5d4 : Ben Noordhuis): The build is still failing.
13:04:51  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/2cecd0d3d1b2...9aab5d483794
13:04:51  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/4089278
13:04:51  * travis-cipart
13:27:41  * toothrjoined
13:32:44  * toothrchanged nick to toothrot
13:42:54  * AvianFlujoined
13:46:02  * jmar777joined
14:03:02  * felixgejoined
14:03:03  * felixgequit (Changing host)
14:03:03  * felixgejoined
14:08:30  * stagasquit (Ping timeout: 252 seconds)
14:17:18  * bradleymeckjoined
14:17:27  * bradleymeckquit (Client Quit)
14:49:08  * bentkusjoined
14:49:12  <bentkus>HALLO
14:56:35  <txdv>bnoordhuis: what do i have to do
15:03:25  <txdv>WHY IS NOBODY ANSWERING
15:03:27  <txdv>WHERE IS LOUDBOT
15:20:29  * piscisaureus_joined
15:20:52  <MI6>joyent/node: Bert Belder master * 7be2724 : Revert "path: fix bugs related to paths with trailing slashes" It only d - http://git.io/o7ADVg
15:26:50  <bnoordhuis>txdv: what do you have to do for what?
15:27:16  <txdv>To be ready for the future!
15:27:46  <txdv>I just want an answer
15:27:59  <txdv>like "well ipv6 is not here yet, we might do it in the future"
15:28:05  <txdv>but the implementation now is currently broken
15:28:46  <txdv>as i stated if you echo a 0 into /proc/sys/net/ipv6/ipv6_bind (or something similar) the behaviour of uv_udp_bind will differ
15:28:46  <bnoordhuis>do you mean ipv6 in general or in libuv?
15:28:56  <bnoordhuis>i would concur with the first
15:29:03  <txdv>first what
15:29:09  <bnoordhuis>statement
15:29:47  <txdv>first solution?
15:30:32  * piscisaureus_quit (Read error: Connection reset by peer)
15:30:33  * dapjoined
15:30:53  * piscisaureus_joined
15:30:54  * dapquit (Client Quit)
15:31:03  <bnoordhuis>txdv: ipv6 can no doubt be improved (in general in libuv) but it's not a top priority right now
15:31:18  * dapjoined
15:31:19  <txdv>IT DOESN'T HAVE TO BE
15:31:21  <txdv>just communicate to me
15:31:28  <txdv>i can do all the patches
15:31:38  <txdv>SAY: good boy or bad boy
15:33:44  <bnoordhuis>txdv: that means having to think about it
15:33:54  <bnoordhuis>i have only limited memory bandwidth
15:34:16  <bnoordhuis>not to mention a short attention sp- hey look, a butterfly!
15:35:32  <txdv>I would understand that female secondary genitalia would distract you, but butterflies
15:35:35  <txdv>come on
15:35:54  <bnoordhuis>interesting creatures, butterflies. interesting lifecycles
15:36:05  <tjfontaine>not to mention their effect
15:36:22  <bnoordhuis>then again, so does the malaria mosquito
15:37:16  <bnoordhuis>have an interesting lifecycle, i mean (it does. look it up on wikipedia)
15:37:27  * qmxchanged nick to qmx|lunch
15:37:47  <piscisaureus_>hello
15:37:53  <bnoordhuis>txdv: what's your proposal re: udp + bind?
15:38:09  <bnoordhuis>make ipv6-only the default?
15:38:19  <txdv>YES
15:39:27  <txdv>You know that ipv6-only is not 0 by default on all unices
15:39:40  <txdv>like on macosx it is 1
15:39:59  <txdv>and if you echo 1 to /proc/sys/net/.../ipv6only, it will be 1 on linux too
15:40:07  <txdv>and some distribution use the default value of 1
15:40:26  <bnoordhuis>okay, make it so
15:44:00  <bnoordhuis>$ out/Release/node -e 'for (var i = 0, junk = [123.456]; i < 26; ++i) junk = junk.concat(junk)'
15:44:03  <bnoordhuis>FATAL ERROR: CALL_AND_RETRY_0 Allocation failed - process out of memory
15:44:08  <bnoordhuis>^ that used to work before we upgraded v8
15:48:32  <indutny>hallo
15:48:44  <indutny>bnoordhuis: have you patched that byteswapping thing?
15:48:54  <bnoordhuis>indutny: no. have you?
15:48:59  <indutny>nope, but I'll do
15:49:46  <txdv>IMO bind/bind4/bind6 is the best solution
15:50:20  * dapquit (Quit: Leaving.)
15:50:22  <txdv>but I don't care, however the api will look like, I will model it in C# after the bind/bind4/bind6 solution
15:50:42  * dapjoined
15:51:40  <bnoordhuis>txdv: i don't want to add new api functions for this
15:52:21  <txdv>change existing api = add new api
15:52:34  <txdv>solution 1 adds a flag to the function
15:53:16  <MI6>joyent/node: Fedor Indutny master * 35a137c : typed arrays: fix build on sunos - http://git.io/NkIK3g
15:53:23  <bnoordhuis>hmm, i managed to make git abort()
15:54:13  <txdv>bnoordhuis: in this very particular case, I just want the feature, i don't care how the api will look like
15:54:38  <txdv>BUt you guys are totally silent about what to do
15:54:49  <txdv>even though I gathered all the information in that pr/issue
15:55:24  <bnoordhuis>txdv: i would move to make ipv6-only the default and unchangeable
15:55:40  <bnoordhuis>on freebsd, for example, mixed mode doesn't work all that great
15:56:05  <txdv>UNCHANGEABLE
15:56:07  <bnoordhuis>piscisaureus_: ^ agree/disagree?
15:56:15  <indutny>bnoordhuis: nice
15:56:20  <indutny>bnoordhuis: do you have core dump?
15:56:21  <bnoordhuis>that's something for 0.12 though
15:56:30  <bnoordhuis>indutny: no. it didn't repeat
15:56:40  <txdv>you mean the ipv6 change is only for 0.12?
15:56:43  <indutny>I'm pretty sure it's because of networking
15:56:49  <indutny>applications are always borked because of it
15:56:53  <piscisaureus_>bnoordhuis: agree with what?
15:57:06  <bnoordhuis>piscisaureus_: make ipv6-only the default and unchangeable
15:57:07  <piscisaureus_>RROR: CALL_AND_RETRY_0 <-- disagree
15:57:18  <piscisaureus_>bnoordhuis: ah, fine
15:57:23  <bnoordhuis>txdv: ^
15:57:25  <piscisaureus_>bnoordhuis: unchangeable seems fine for now
15:57:51  <txdv>NOOOOO
15:57:59  <txdv>you want to destroy the awesome feature of dualstacking
15:58:00  <piscisaureus_>bnoordhuis: you know that I want to revise the bind apis
15:58:13  <piscisaureus_>bnoordhuis: we can add a flag for that (both tcp and udp)
15:58:50  <bnoordhuis>txdv: you brought it up. now decisions are gonna be made
15:59:12  <txdv>You take away dualstacking, I will take away your candy
15:59:21  <bnoordhuis>that's okay, i don't like candy
15:59:29  <txdv>But your children do
15:59:40  <bnoordhuis>they never get any. we're all calvinists here
15:59:42  <txdv>And children without candy ... they are monsters
16:00:11  <bnoordhuis>piscisaureus_: still planning on coming to gouda today?
16:00:19  <piscisaureus_>bnoordhuis: no
16:00:26  <bnoordhuis>i figured as much :)
16:00:29  <txdv>what a cliche
16:00:35  <txdv>dutch people eating gouda
16:00:45  <txdv>why are you doing this
16:00:56  <txdv>You know in germany most of the people hate dutch people
16:01:16  <txdv>there are 2 things german people can't stand, racism and dutch people
16:01:37  <bnoordhuis>txdv: that's okay. the feeling is mutual
16:01:52  <bnoordhuis>actually, i believe most germans think dutch is cute
16:01:56  <bnoordhuis>cute!
16:02:04  <txdv>dutch is cute?
16:02:06  <txdv>like the language?
16:02:08  * Guest___joined
16:02:09  <txdv>or the country?
16:02:17  * sblomjoined
16:02:21  <bnoordhuis>the language. the country is mostly just flat
16:02:28  <Guest___>Pff we really need triplestack
16:02:28  <indutny>harhar
16:02:30  <piscisaureus_>Guest___: piss off, what are you doing here?
16:02:43  <txdv>bnoordhuis: the german language already sounds disgusting
16:02:49  <Guest___>@piscisaureus_ implement dualstack already lamre
16:02:49  <piscisaureus_>IP6_TRIPLESTACK
16:02:55  <txdv>and dutch makes it just worse
16:03:00  <txdv>piscisaureus_: STOP MAKING FUN OF ME
16:04:13  <txdv>piscisaureus_: I think we are save if we only support dualstack
16:04:29  <piscisaureus_>txdv: so you want to enable by default or disable by default?
16:04:40  <txdv>disable by default
16:04:43  <txdv>enable on request
16:04:43  <piscisaureus_>good
16:06:19  <txdv>No, it's excellent
16:06:31  <bnoordhuis>txdv: don't forget to test freebsd
16:06:46  <txdv>don't pull in any of my pr yet
16:06:51  <txdv>you wouldn't do it anyway
16:07:13  <txdv>bnoordhuis: well, I will install it
16:07:15  <indutny>txdv: ++ for german language
16:07:16  <kohai>txdv has 0 beer
16:07:17  * mikealquit (Quit: Leaving.)
16:07:26  <txdv>but only if you test it on macosx
16:07:59  * `3rdEdenquit (Remote host closed the connection)
16:08:28  * bradleymeckjoined
16:09:19  <txdv>bnoordhuis: i mean that the dutch language is worse than german
16:09:48  <piscisaureus_>txdv: kanker
16:10:32  <bnoordhuis>txdv: that was meant for indutny, i take it?
16:10:46  <indutny>no
16:10:47  <piscisaureus_>muhaha
16:10:48  <indutny>I disagree
16:10:55  <indutny>dutch is much better than german
16:11:00  <indutny>it's cute
16:11:05  <piscisaureus_>scheisse
16:11:13  <indutny>I know this word
16:11:13  <txdv>SCHEISS HOLLAENDISCH
16:11:21  <indutny>and the second one too
16:11:28  <indutny>you're disgusting people
16:11:37  * indutnychanged nick to gollum
16:11:48  <gollum>my precious
16:11:51  <txdv>I'm not disgusting
16:11:58  <txdv>come on, I am Lithuanian
16:11:59  <gollum>txdv: you're
16:12:34  <txdv>am I lithuanian or am I disgusting?
16:12:38  <txdv>Or is there no difference?
16:13:11  <gollum>Where would you be without me, gollum, gollum? I saved us! It was me! We survived because of me!
16:13:31  * gollumchanged nick to indutny
16:13:33  <indutny>just kidding
16:13:41  <indutny>bnoordhuis: you should try taking charges for messaging you http://t.co/yi1c3hAy
16:13:56  <indutny>bnoordhuis: it might end up beeing more profitable than software development
16:13:56  * sblomquit (Quit: leaving)
16:14:36  <bnoordhuis>indutny: does i need a facebook account for that? if yes, the deal is off
16:14:44  <indutny>idk
16:14:53  <indutny>never heard of paid irc messaging
16:16:27  <txdv>Brilliant idea
16:16:47  <txdv>Charge 100$ people who don't have money but want to get in contact with people who have millions of dollars
16:16:59  <indutny>txdv: inflation
16:17:07  <indutny>he's afraid of i
16:17:07  <txdv>My interpretation of this, zuckerberg supports slavery
16:17:09  <indutny>s/i/it
16:17:26  <txdv>Is zuckerberg a jew?
16:18:14  <txdv>Why do most of the american jews have german surnames
16:19:58  <indutny>because they've left germany in WW2
16:21:00  * mikealjoined
16:22:14  <txdv>even though they left germany and the german culture, they still use german expressions in the american english language
16:25:02  * qmx|lunchchanged nick to qmx
16:25:03  <txdv>bad topic
16:25:06  <txdv>lets talk about libuv
16:25:18  <txdv>bnoordhuis: piscisaureus_ what is your decision on dualstack?
16:26:45  <piscisaureus_>txdv: my decision is: I don't care so much. I am fine with making it configurable for both tcp and udp one day.
16:26:55  <piscisaureus_>txdv: but I will not land a buttload of new API to support it
16:30:10  * mmaleckichanged nick to mmalecki[out]
16:37:48  <isaacs>piscisaureus_: what's wrong with the trailing slash path.basename/dirname/etc change?
16:38:10  <piscisaureus_>isaacs: doesn't work on windows. the implementation is also too complicated
16:38:14  <piscisaureus_>isaacs: I'm making a better patch now
16:38:50  <isaacs>piscisaureus_: ok, great.
16:39:08  <isaacs>piscisaureus_: mostly, i like the behavior expressed in the test cases, and that the test cases pass with the patc.
16:39:34  <piscisaureus_>isaacs: yes, but I like consistency as well
16:39:39  <isaacs>agreed
16:39:54  <piscisaureus_>isaacs: if basename('foo/') === 'foo' then basename('foo\\') == 'foo' should be true on windows
16:40:18  <piscisaureus_>isaacs: there are indeed some edge cases that aren't covered yet (e.g. on unix basename('/') == '/')
16:40:20  <piscisaureus_>but that's for later
16:40:46  <piscisaureus_>sorry, I mean, the unix cli basename too
16:40:49  <isaacs>piscisaureus_: basename('/') SHOULD == '/'
16:40:51  <piscisaureus_>so $(basename /)
16:41:04  <piscisaureus_>isaacs: yes, but it's not the case now
16:41:18  <isaacs>indeed.
16:41:28  <piscisaureus_>isaacs: so THAT is for another patch
16:41:34  <isaacs>yep
16:42:12  <piscisaureus_>where I also demand consistency - so in that case basename('c:\\') == 'c:\\'
16:43:43  <isaacs>true
16:44:01  <isaacs>piscisaureus_: i support your goals in this matter
16:44:52  <indutny>isaacs: express works fine with node 0.9.6-pre
16:45:00  <indutny>isaacs: that was socket.io, and they've already figured out that problem
16:45:05  <indutny>isaacs: so, npm update has fixed it
16:45:28  <indutny>bnoordhuis: stdio thing is fixed, right?
16:45:37  <indutny>I mean you've fixed
16:45:53  <bnoordhuis>indutny: i probably traded one bug for another but yes
16:46:07  <indutny>bnoordhuis: wanna tell me more?
16:46:13  <indutny>probably I can look into it on weekend
16:46:46  <bnoordhuis>indutny: i was exhibiting some dry wit. or trying to. it's fixed
16:47:05  <bnoordhuis>but process.c kind of frightens me
16:47:16  <bnoordhuis>it's more complex than i would like for it to be
16:47:54  <indutny>yeah
16:47:56  <indutny>I agree
16:48:01  <indutny>especially stdio thing
16:48:07  <indutny>duping and all this coping with fds
16:49:23  <txdv>piscisaureus_: so you are okay with solution 3?
16:55:34  <isaacs>indutny: 35a137c breaks the build on os x
16:55:58  <isaacs>indutny: https://gist.github.com/4512254
16:56:10  <isaacs>bnoordhuis: doesn't fully fix the problem, still.
16:56:12  <indutny>shit come real
16:56:18  <indutny>let me think about it
16:57:02  <txdv>aren't you developing on a macosx machine indutny?
16:57:08  <indutny>yep, I am
16:57:11  <bnoordhuis>indutny: look up what int8_t is typedef'd as
16:57:12  <indutny>but I tested this patch on smartos server
16:57:18  <bnoordhuis>the compiler ought to match that to signed char
16:57:19  <indutny>bnoordhuis: doing it now
16:57:22  <isaacs>indutny: https://gist.github.com/4512266
16:57:39  <indutny>isaacs: and?
16:58:19  <isaacs>oh, sorry, that last one was for bnoordhuis ^_^
16:58:23  <bnoordhuis>isaacs: and?
16:58:28  <indutny>bnoordhuis you'll be surprised
16:58:38  <indutny>#if defined(_CHAR_IS_SIGNED)
16:58:38  <indutny>typedef char int8_t;
16:58:38  <indutny>#else
16:58:38  <indutny>#if defined(__STDC__)
16:58:38  <indutny>typedef signed char int8_t;
16:58:38  <indutny>#endif
16:59:00  <bnoordhuis>indutny: ah. i believe we compile with -fsigned-char though
16:59:11  <isaacs>bnoordhuis: that test still fails on master, passes prior to the libuv update i told you about last night
16:59:12  <indutny>can we just define __STDC__ in that header file?
16:59:20  <indutny>bnoordhuis: I mean typedarray_byteswap.h
16:59:37  <isaacs>oh, no, wait, nvm
16:59:41  <isaacs>bnoordhuis: early, sorry. test passes now
16:59:48  <bnoordhuis>isaacs: okay, good :)
16:59:52  <isaacs>bnoordhuis: the build had failed, so i still had yesterday's bin. duh
17:00:01  * isaacsis dangerously undercaffienated
17:00:17  <bnoordhuis>indutny: okay, we don't compile with -fsigned-char. forget i said that
17:00:23  <indutny>aha
17:00:29  <indutny>let me try patching node.gyp
17:00:33  <bnoordhuis>well
17:00:36  <isaacs>bnoordhuis: so.. that test. lgty? i'd like to land it so that we have some test of stdio:'inherit', because right now, it's never exercised except in user programs.
17:00:40  <bnoordhuis>it _is_ compiling right?
17:00:43  <indutny>meanwhile, lets revert it
17:00:48  <indutny>bnoordhuis: right now/
17:00:50  <indutny>?
17:00:52  <bnoordhuis>yes
17:00:56  <indutny>on smartos - yes
17:00:58  <indutny>on osx - no
17:01:04  <bnoordhuis>isaacs: one sec, let me review it
17:01:39  <MI6>joyent/node: Fedor Indutny master * d93346d : Revert "typed arrays: fix build on sunos" This reverts commit 35a137cf8d - http://git.io/lLn68A
17:02:05  <bnoordhuis>isaacs: yes, lgtm
17:03:21  <indutny>bnoordhuis: rebuilding with patched common.gypi
17:05:03  * bradleymeckquit (Quit: bradleymeck)
17:05:59  * mikealquit (Quit: Leaving.)
17:12:35  * Guest___quit (Quit: ["Textual IRC Client: www.textualapp.com"])
17:14:24  <MI6>joyent/node: isaacs master * 3383d77 : test: child process { stdio:'inherit' } - http://git.io/lUtOEQ
17:15:57  * mikealjoined
17:15:57  <indutny>bnoordhuis: hm..
17:16:05  <indutny>bnoordhuis: -fsigned-char doesn't seem to help
17:16:15  <indutny>g++ '-DNODE_WANT_INTERNALS=1' '-DARCH="ia32"' '-DNODE_TAG=""' '-DHAVE_OPENSSL=1' '-DHAVE_DTRACE=1' '-D__POSIX__' '-DPLATFORM="sunos"' '-D_LARGEFILE_SOURCE' '-D_FILE_OFFSET_BITS=64' '-D_POSIX_C_SOURCE=200112' -I../src -I../tools/msvs/genfiles -I../deps/uv/src/ares -I/root/node/out/Release/obj/gen -I../deps/openssl/openssl/include -I../deps/v8/include
17:16:15  <indutny>-I../deps/zlib -I../deps/http_parser -I../deps/cares/include -I../deps/uv/include -Wall -Wextra -Wno-unused-parameter -m32 -pthreads -O2 -fno-strict-aliasing -ffunction-sections -fdata-sections -fno-tree-vrp -fno-omit-frame-pointer -fno-rtti -fno-exceptions -fsigned-char -MMD -MF
17:16:15  <indutny>/root/node/out/Release/.deps//root/node/out/Release/obj.target/node/src/node_crypto.o.d.raw -c -o /root/node/out/Release/obj.target/node/src/node_crypto.o ../src/node_crypto.cc
17:16:15  <indutny>In file included from ../src/v8_typed_array.cc:26:0:
17:16:15  <indutny>../src/v8_typed_array_bswap.h: In function 'T v8_typed_array::SwapBytes(T) [with T = signed char]':
17:16:15  <indutny>../src/v8_typed_array_bswap.h:150:23: instantiated from 'T v8_typed_array::LoadAndSwapBytes(void*) [with T = signed char]'
17:16:15  <indutny>../src/v8_typed_array.cc:694:7: instantiated from 'static v8::Handle<v8::Value> {anonymous}::DataView::getGeneric(const v8::Arguments&) [with T = signed char]'
17:16:16  <kohai>g has 7 beers
17:16:16  <indutny>../src/v8_typed_array.cc:738:40: instantiated from here
17:16:16  <indutny>../src/v8_typed_array_bswap.h:125:16: error: size of array is negative
17:16:40  <isaacs>bnoordhuis: fixes `npm explore` also. good stuff.
17:18:35  <isaacs>piscisaureus_: are you working on the path changes at the moment? should i wait on 0.9.6 for you to finish?
17:18:43  <piscisaureus_>isaacs: yes, I am almost done
17:18:49  <indutny>isaacs: you should wait for this signed char thing
17:19:14  <isaacs>great
17:19:54  <bnoordhuis>i guess i'll log in on umcats...
17:20:09  <isaacs>bnoordhuis: the other drone is bigger
17:20:20  <bnoordhuis>yeah, but i always forget what its name/address is :/
17:20:25  <isaacs>bnoordhuis: 165.225.128.184
17:20:42  <isaacs>bnoordhuis: just give it a name in your .ssh/config
17:20:53  <bnoordhuis>doing that as we speak :)
17:21:00  <isaacs>that remidns me, though, i have access to nodejs.org's dns now
17:21:04  * bentkusquit (Quit: Lost terminal)
17:21:22  <indutny>finally
17:21:26  <bnoordhuis>isaacs: you finally wrestled it from ryan's clammy hands?
17:21:27  <indutny>bnoordhuis: umcats
17:21:35  <indutny>bnoordhuis: is it a smartos server?
17:21:46  <isaacs>indutny: yeah
17:21:49  <bnoordhuis>indutny: yes. a puny one though
17:21:49  <indutny>ok
17:22:02  <isaacs>we'll probably delete it.
17:22:13  <isaacs>i don't think we really use umcats for anything. it has a fun name, though
17:27:31  <isaacs>bnoordhuis: that TypedArray issue: it'd affect stuff like Uint16Array as well, right?
17:27:41  <bnoordhuis>isaacs: yes
17:28:09  <isaacs>k.
17:28:39  <indutny>bnoordhuis: it seems that I need -funsigned-char
17:28:51  <bnoordhuis>indutny: hm
17:29:07  <indutny>because if #if defined(_CHAR_IS_SIGNED)
17:29:10  * pooyajoined
17:29:14  <indutny>int8_t = char
17:29:16  <indutny>and if its' not
17:29:20  <indutny>int8_t = signed char
17:29:26  <indutny>shit come real
17:29:28  <bnoordhuis>still, there's templates for both
17:29:45  <indutny>well
17:30:50  <bnoordhuis>indutny: https://gist.github.com/d45fe163cee35cad79e0 <- that fixes it
17:31:03  <indutny>oh, interesting
17:31:09  <indutny>that lgtm
17:31:21  <indutny>nevertheless issue is pretty odd
17:35:16  <MI6>joyent/node: Ben Noordhuis master * b509ae6 : typed arrays: fix sunos signed/unsigned char issue The int8_t and uint8_ - http://git.io/HGj2MA
17:35:42  <indutny>good
17:40:41  <bnoordhuis>>>> libc.syscall(1, 1, "hello\n", 6)
17:40:41  <bnoordhuis>hello
17:40:41  <bnoordhuis>6
17:40:47  <indutny>python?
17:40:47  <bnoordhuis>^ i want that in node
17:40:49  <bnoordhuis>yes
17:40:52  <indutny>why?
17:41:03  <bnoordhuis>good ffi? what's not to like?
17:41:09  <indutny>bad ffi
17:41:21  <indutny>many instructions for one real instruction
17:41:37  <indutny>better call method which has a lot of instructions
17:41:48  <indutny>minimum waste of time for preparing stack to enter C++
17:42:23  <bnoordhuis>ah, actually we could shave off some cycles if we didn't go through v8's js/c++ thunks
17:42:43  <bnoordhuis>though i doubt it's critical
17:45:27  <MI6>joyent/node: piscisaureus created branch issue4538 - http://git.io/Tkva2w
17:45:30  <piscisaureus_>isaacs / bnoordhuis: review? ^
17:46:18  <bnoordhuis>ah, 40 char regexes
17:46:26  <piscisaureus_>they got shorter :-)
17:46:34  <bnoordhuis>it might be calculating fib(40) for all we know
17:48:03  <bnoordhuis>piscisaureus_: lgtm i think. i assume tests pass
17:48:09  <piscisaureus_>they do
17:48:25  <piscisaureus_>bnoordhuis: I want you to review the regexes
17:49:27  <bnoordhuis>piscisaureus_: i like that they're still regular
17:49:45  <bnoordhuis>none of that newfangled backtracking
17:50:25  <MI6>joyent/node: Bert Belder master * 6a91eab : path: make basename and extname ignore trailing slashes Fixes #4536 - http://git.io/zThWmw
17:50:59  <piscisaureus_>bnoordhuis: I think there is backtracing involved because the end groups are sticky and the early groups are nongreedy
17:52:02  <bnoordhuis>piscisaureus_: then go fix that
17:52:20  <piscisaureus_>bnoordhuis: I refuse
17:53:19  * stagasjoined
18:01:14  <isaacs>alright, once DNS updates, we've got build/test machines at smartos.nodejs.org ubuntu.nodejs.org centos.nodejs.org
18:02:28  * lohkeyjoined
18:02:51  <isaacs>piscisaureus_: any reason why you wrapped the trailing slashes in a non-capturing group? ie, why (?:[\\\/]*) instead of just [\\\/]*
18:05:41  <piscisaureus_>isaacs: well because I don't need it
18:06:18  * jmar777quit (Read error: Connection reset by peer)
18:06:45  <piscisaureus_>isaacs: but no specific reason
18:06:45  * jmar777joined
18:08:50  <isaacs>it's not relevant enough a change to make now, i guess.
18:09:01  <isaacs>just a review-nit, but i was not in time :)
18:09:29  <isaacs>doesn't affect correctness or speed, just the number of characters in the line
18:09:41  <isaacs>ok! so we're good for 0.9.6, then?>
18:10:13  <tjfontaine>we should fix setImmediate :)
18:11:11  * qmxchanged nick to qmx|brb
18:13:21  * piscisaureus_quit (Ping timeout: 248 seconds)
18:13:22  * sblomjoined
18:15:56  <isaacs>tjfontaine: yes, but not in this release
18:16:04  <isaacs>bnoordhuis: the change to src/v8_typed_array.cc was done in both master and v0.8, yes?
18:16:14  <isaacs>i'm merging v0.8 into master, and seeing a change there.
18:16:21  <isaacs>bnoordhuis: should src/v8_typed_array.cc just use the master version?
18:17:24  <isaacs>oh, it looks like the patch wasn't landed in master at all..
18:18:43  <isaacs>that answers that
18:19:09  <MI6>joyent/node: isaacs master * 97db62b : Merge remote-tracking branch 'ry/v0.8' into master Conflicts: AUTHORS (+13 more commits) - http://git.io/nhVo1g
18:21:40  <isaacs>lol, chopped off the top line of the changelog in that merge. whoops.
18:25:56  * loladirojoined
18:26:26  * qmx|brbchanged nick to qmx
18:30:08  * bradleymeckjoined
18:32:05  <sblom>When does the 0.10 train leave the station? Very simple improvement to Window perf counter support that I'd love to have in 0.10: https://github.com/joyent/node/pull/4560
18:33:13  <isaacs>sblom: it's more like "the train will leave when it's full, but hurry up anyway" kind of thing
18:33:20  <isaacs>sblom: rather than strict time-based release.
18:33:24  <sblom>isaacs: fair enough
18:33:51  <sblom>I'll ask piscisaureus_ for a train ticket. :)
18:34:25  <rendar>what about the shutdown(socket) function? it seems its called when a tcp connection is closed in libuv
18:34:38  <isaacs>sblom: from a prediction pov, though, it seems likely we'll be releasing 0.10 by the end of the month.
18:34:49  <isaacs>sblom: it's getting very close to done
18:35:10  <isaacs>bnoordhuis: any strictly libuv-related changes you think ought to be in the changelog?
18:35:45  * paddybyersquit (Ping timeout: 248 seconds)
18:35:48  <isaacs>bnoordhuis: there's a bunch, most have your name on them: https://gist.github.com/4512898
18:36:08  <isaacs>but it seems like mostly lowlevel cleanup type stuff. the stdio:inherit bugfix is already mentioned
18:37:57  <isaacs>review, plz: https://gist.github.com/4512916
18:38:40  * mralephjoined
18:49:41  <MI6>joyent/node: isaacs created branch v0.9.6-release - http://git.io/nr27wA
18:50:45  <bradleymeck>isaacs: how long before 0.10.
18:51:35  <CoverSlide>but only 52 more weeks
18:56:41  * mmalecki[out]changed nick to mmalecki
18:58:10  * felixge_joined
18:58:10  * felixge_quit (Changing host)
18:58:10  * felixge_joined
18:59:16  * felixgequit (Read error: Connection reset by peer)
18:59:16  * felixge_changed nick to felixge
18:59:37  <isaacs>bradleymeck: a few more week
18:59:38  <isaacs>*s
18:59:48  <bradleymeck>thanks
19:00:16  <mmalecki>bradleymeck: we got most of shit working anyway
19:00:34  <bradleymeck>mmalecki: yea but im trying to iron out execSync in my spare cycles
19:00:51  <mmalecki>oh... good luck, stay sane dawg
19:01:52  <isaacs>bradleymeck: i do not have much hope of that being done for 0.10
19:01:58  <isaacs>bradleymeck: but 0.12 will not be long behind it
19:02:09  <bradleymeck>isaacs: sure as long as its not 6months
19:02:10  <isaacs>so don't be discouraged
19:02:18  <isaacs>bradleymeck: THAT DEPENDS ON YOU MY FRIEND!
19:02:23  <isaacs>bradleymeck: and also, on us
19:02:24  <bradleymeck>lol
19:02:34  <isaacs>actually, mostly on us
19:02:45  <isaacs>but whether or not execSync is in it, you can probably have something to do with that
19:05:28  * `3rdEdenjoined
19:05:43  <mmalecki>isaacs: way to motivate ;)
19:05:54  <isaacs>mmalecki: thanks :)
19:05:58  <mmalecki>isaacs: did we ever decide what to do with `close` event and child processes?
19:06:20  <mmalecki>ya know, that thing when you broke all the API and stuff :D
19:06:27  <isaacs>mmalecki: yeah. you have until the child's 'exit' or 'error' event to read() from it, or else it gets dumped.
19:06:47  * loladiroquit (Quit: loladiro)
19:06:48  <isaacs>mmalecki: a better approach is mentioned in the issue, but requires libuv changing
19:07:07  <mmalecki>isaacs: did you ever get to implementing it?
19:07:12  <isaacs>yeah
19:07:16  <isaacs>the bugs are fixed now.
19:07:18  <mmalecki>OH WORD.
19:07:23  <isaacs>there's just this weird edge case if you don't read() until after exit.
19:07:37  <isaacs>because we have to dump it in order to get to the close event
19:10:41  <indutny>isaacs: so it's like debugger tests failing all over again
19:10:53  <indutny>I should try to fix it
19:10:59  <isaacs>yeah
19:11:00  <indutny>but I'm so lame
19:11:02  <isaacs>i'ts not urgent
19:11:09  <indutny>ok, what's urgent?
19:11:13  <isaacs>:)
19:11:16  <isaacs>getting 0.9.6 out
19:11:21  <isaacs>get them fixed before 0.9.7 :)
19:11:24  <indutny>ok
19:11:31  <indutny>so what can I do to get 0.9.6 out?
19:11:35  <isaacs>not much
19:11:37  <isaacs>it's pretty much done.
19:11:49  <indutny>oh no
19:11:50  <isaacs>all the other tests are passing, and debugger tests pass if run directly, just not when the test runner is running them
19:11:53  <isaacs>they're really flaky
19:12:04  <indutny>my life was vain
19:12:07  <isaacs>hahah
19:12:09  <isaacs>not at all!
19:12:11  * paddybyersjoined
19:12:16  <isaacs>indutny: you showed up just in time :)
19:12:23  <indutny>for what?
19:15:46  <indutny>isaacs: ^^
19:15:51  * brsonjoined
19:16:05  <indutny>oh, nothing
19:17:18  <isaacs>just waiting on the os x binary tarballs
19:18:01  <indutny>ah
19:18:01  <indutny>ok
19:18:18  * indutnylistens Dream Theater albums
19:18:29  <indutny>it's like getting home after long travel
19:18:44  <indutny>I always return to listening them after several years
19:19:10  <CoverSlide>Pull Me Underr!!!
19:20:06  <indutny>yeah
19:20:21  <indutny>Actually, my favorite album is Awake
19:20:38  <indutny>there're so many fresh ideas there
19:20:45  <CoverSlide>that's the only Dream Theater song I know of
19:21:16  <indutny>ah
19:21:17  <indutny>ok, then
19:21:39  <indutny>CoverSlide: http://www.youtube.com/watch?v=GTduOgLle-8
19:22:19  * lohkeyquit (Quit: lohkey)
19:22:31  * TheJHjoined
19:25:13  <MI6>joyent/node: isaacs created tag v0.9.6 - http://git.io/MURpRg
19:26:13  <MI6>joyent/node: isaacs master * 9a4a39f : Now working on 0.9.7 (+2 more commits) - http://git.io/w1saRw
19:26:26  <MI6>joyent/node: isaacs v0.8 * 4c9de9d : blog: release v0.9.6 - http://git.io/C0YeqA
19:29:36  <isaacs>ok, 0.9.6 is out http://blog.nodejs.org/2013/01/11/node-v0-9-6-unstable/
19:29:59  <isaacs>thanks for jumping on bugs at the last minute, everybody
19:30:15  <isaacs>indutny: would you like to update V8 in master?
19:30:50  <isaacs>there's a 3.15.11.7
19:31:16  <isaacs>(immediately after a release is the best time to do that, i've found)
19:33:03  <mmalecki>indutny: any chance we could have dtrace helper backported to v0.8?
19:33:05  <mmalecki>cc isaacs
19:37:26  * loladirojoined
19:38:16  * trevnorrisjoined
19:38:34  <isaacs>indutny: < trevnorris> isaacs: indutny added new class methods to include/v8.h that allow for the Isolate to be passed. But didn't include it for Number::New().
19:38:44  <isaacs>(moved here from #Node.js)
19:39:12  <trevnorris>indutny: just implemented it and it's saving me about 40% efficiency for some new buffer read methods I'm working on.
19:45:53  * isaacs&
19:47:52  * mjr_joined
19:48:03  * `3rdEdenquit (Remote host closed the connection)
19:49:36  <mjr_>color readline repl with tab completion over a TCP socket you guys.
19:49:49  <mjr_>Better than Cats. I'm going to see it again and again.
20:02:12  <indutny>isaacs: huh?
20:02:25  <indutny>isaacs: I can't remember adding new classes
20:02:35  <indutny>mmalecki: dunno, it's possible
20:02:42  <tjfontaine>not adding new classes, but passing the node_isolate so v8 doesn't look it up
20:02:43  <indutny>isaacs: I'll update v8
20:02:50  <tjfontaine>I think is what he means
20:02:52  <indutny>tjfontaine: huh, I did it?
20:03:14  <tjfontaine>I think ben did it in conjunction with your v8 upgrade
20:03:24  <indutny>ah, ok
20:03:29  <indutny>where is isaacs anyway
20:03:30  <trevnorris>indutny: yeah. sorry. just the members where you can pass the isolate.
20:04:17  <trevnorris>indutny: the latest changes you made in 7b4d95a. freakin awesome btw.
20:04:36  <indutny>aaaah
20:04:47  <indutny>ooook
20:05:07  <trevnorris>indutny: I made pretty much the same changes for Number and it improved my method time by 40%
20:05:19  <indutny>well, I didn't do that
20:05:22  <indutny>v8 team did it :)
20:05:26  <trevnorris>heh
20:05:44  <indutny>but yeah
20:05:48  <indutny>passing isolate is a big win
20:06:30  <trevnorris>yeah. especially on small functions like the buffer read/write methods.
20:13:17  * exit2joined
20:13:26  * exit2part
20:17:35  * TheJHquit (Ping timeout: 260 seconds)
20:23:48  <indutny>isaacs: updated v8
20:23:56  <MI6>joyent/node: Fedor Indutny master * d22bd9e : deps: update v8 to 3.15.11.7 - http://git.io/Y289Ig
20:27:40  <trevnorris>indutny: can you revert the cflags changes made in common.gypi? those instructions make it run much slower.
20:34:12  <indutny>aaah
20:34:13  <indutny>shit
20:34:15  <indutny>it came again
20:34:18  <indutny>thanks
20:35:30  <MI6>joyent/node: Ben Noordhuis master * 8018113 : v8: remove optimization switches Remove compiler switches from $(TOPLEVE - http://git.io/q1xWkw
20:35:32  * TheJHjoined
20:35:56  <bradleymeck>yes/no on https://github.com/joyent/node/issues/4570 , can patch myself and add test
20:37:51  <trevnorris>awesome. me and my benchmarks thank you.
20:39:17  * indexzerojoined
20:52:43  * EhevuTovjoined
20:54:45  * piscisaureus_joined
20:57:53  * sgallaghquit (Remote host closed the connection)
21:04:47  * brsonquit (Ping timeout: 252 seconds)
21:06:38  * c4miloquit (Remote host closed the connection)
21:07:05  * c4milojoined
21:08:47  * c4miloquit (Read error: Connection reset by peer)
21:08:52  * c4milo_joined
21:10:00  * piscisaureus_quit (Read error: Connection reset by peer)
21:10:43  * piscisaureus_joined
21:20:19  <indutny>isaacs: so where're we now with streams2 and performance?
21:21:28  <CoverSlide>does node have its own benchmark suite?
21:21:55  <trevnorris>CoverSlide: there are some in benchmark/
21:22:26  <CoverSlide>oh awesome
21:22:27  <indutny>if you want a module - then answer is no
21:22:32  <trevnorris>i'm working on making them extensive for Buffers now.
21:22:35  <indutny>it's just some scripts
21:22:41  <CoverSlide>i should start looking at the source
21:23:47  <CoverSlide>oh it even has an r script? nice
21:25:05  <MI6>joyent/node: Scott Blomquist master * 8e311d2 : windows: add support for '_Total' perf counters - http://git.io/wHK2-A
21:25:34  <piscisaureus_>sblom: ^-- your train ticket
21:26:15  <CoverSlide>he's got a ticket to ride!!
21:27:09  * bradleymeckquit (Quit: bradleymeck)
21:31:51  <trevnorris>just submitted a ticket to implement Isolate on Number::New(). hopefully they consider it.
21:36:50  * EhevuTovquit (Quit: This computer has gone to sleep)
21:38:36  * EhevuTovjoined
21:46:00  * jmar777quit (Remote host closed the connection)
21:46:34  * jmar777joined
21:50:36  <sblom>piscisaureus_: thanks.
21:50:44  * jmar777quit (Ping timeout: 248 seconds)
21:51:37  * rendarquit
21:54:01  * brsonjoined
21:55:59  * AndreasMadsenjoined
22:09:46  * perezdjoined
22:11:09  * jmar777joined
22:18:30  * pieternjoined
22:20:29  * AndreasMadsenquit (Remote host closed the connection)
22:26:23  * EhevuTovquit (Quit: This computer has gone to sleep)
22:29:32  * EhevuTovjoined
22:32:53  * felixgequit (Quit: felixge)
22:39:14  * pieternquit (Remote host closed the connection)
22:39:33  * pieternjoined
22:40:53  * felixgejoined
22:40:53  * felixgequit (Changing host)
22:40:53  * felixgejoined
22:56:54  * perezdquit (Quit: perezd)
22:59:13  * felixgequit (Quit: felixge)
23:05:37  * lohkeyjoined
23:06:45  * jmar777quit (Remote host closed the connection)
23:14:33  <sblom>Every time I build, I end up with deps/openssl/openssl.{props|targets|xml} in my working copy. Anyone have any reason not to .gitignore them?
23:15:55  <tjfontaine>hm are those used to generate the .def to re-export symbols?
23:16:00  * indexzeroquit (Quit: indexzero)
23:16:04  * paddybyersquit (Ping timeout: 246 seconds)
23:17:57  <isaacs>piscisaureus_: it *seems* like ws works fine on master.
23:18:06  <isaacs>piscisaureus_: the examples all work fine
23:18:08  <isaacs>einaros: ^
23:20:08  * indexzerojoined
23:23:36  * TheJHquit (Ping timeout: 276 seconds)
23:28:15  <piscisaureus_>isaacs: well, that's good news.
23:28:27  <piscisaureus_>tjfontaine: no. we're not generating .def files atm
23:28:41  <piscisaureus_>sblom: I never seen that but these files are VS files so you can add them to .gitignore
23:28:44  <tjfontaine>piscisaureus_: k
23:28:51  <piscisaureus_>isaacs: it seems alright to give sblom push access for this kind of stuff
23:33:06  <isaacs>piscisaureus_: right. on it
23:33:24  <isaacs>thanks for reminding me, that was on my todo list
23:34:08  <isaacs>sblom: So, you've got push access
23:34:53  <isaacs>sblom: Probably goes without saying, but just so it's said: All code should be reviewed by someone who is not the author. Don't push any patches that are in an area you don't grok. All submitters should have signed the CLA (I'll send you the google doc)
23:35:25  <isaacs>sblom: i'll send an email out to the mailing list soon. welcome to the team for real, it's good to have you :)
23:36:31  <isaacs>and no force-pushes to master! (unless it's really a good reason, and you have to feel bad if you do it)
23:36:47  <isaacs>basically, nothing changes, but if piscisaureus_ says "LGTM" you can just push it yourself now :)
23:37:21  <tjfontaine>force pushes for everyone!
23:38:11  <isaacs>force-pushing to feature branches is fine and good. encouraged, even.
23:39:51  <isaacs>piscisaureus_: https://github.com/joyent/node/wiki/Project-Organization says you work for c9 still
23:41:57  <piscisaureus_>isaacs: ah, good that you notice. I will update it once our website goes public (in a week or so)
23:41:58  <piscisaureus_>or less
23:42:03  * c4milo_quit (Remote host closed the connection)
23:42:41  <isaacs>kewl
23:43:29  * c4milojoined
23:46:26  * paddybyersjoined
23:48:17  * c4miloquit (Ping timeout: 248 seconds)
23:57:26  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
23:58:59  <isaacs>horns sounded :)