00:00:10  <piscisaureus>so windows thinks there are 2 references to the pipe, one from regedit and one from notepad
00:00:11  <DrPizza>so the last processto start
00:00:13  <DrPizza>has all the handles
00:00:18  <piscisaureus>yes
00:00:25  <DrPizza>(its own and all the previous process's)
00:00:27  <DrPizza>sowhen it exits
00:00:29  <DrPizza>it closes the whole lot
00:00:30  <piscisaureus>yes
00:00:36  <DrPizza>releasing all the waits
00:00:37  <DrPizza>hmmm
00:00:47  <DrPizza>this is awkward
00:00:53  <piscisaureus>so we need to use SetHandleInformation to make the handles non-inheritable again after spawning
00:00:55  <piscisaureus>not so much
00:01:06  <piscisaureus>we're opening everying non-inheritable by default
00:01:21  <piscisaureus>we're just setting the stdio stuff to inheritable just before CreateProcess
00:01:31  <piscisaureus>but this is never reverted after CreateProcess
00:01:33  <piscisaureus>easy to fix
00:01:38  <DrPizza>oh right
00:01:43  <DrPizza>and since the handles ar ekept open until the process exits
00:01:45  <DrPizza>yes I see
00:02:08  <DrPizza>but
00:02:29  <DrPizza>does it not wait for them to be closed prior to calling CloseHandle(process)?
00:02:37  <DrPizza>oh no
00:02:41  <DrPizza>it just loops through and closes them
00:08:57  <bnoordhuis>piscisaureus: https://github.com/bnoordhuis/node/commit/474de0d <- does that look like an acceptable solution to #1657?
00:09:44  <DrPizza>ugh errno
00:10:11  <piscisaureus>bnoordhuis: why is that hack needed?
00:10:35  <bnoordhuis>piscisaureus: you mean the FSError function?
00:10:42  <piscisaureus>yes
00:10:52  <bnoordhuis>because i didn't want to touch ErrnoException
00:11:21  <piscisaureus>what doesn't ErrnoException do that what we need anyway?
00:11:39  <piscisaureus>iow wat mist er in ErrnoExcepion, waardoor je dit nodig hebt?
00:12:00  <bnoordhuis>it's not aware of libuv
00:12:04  <DrPizza>oh dear bnoordhuis you havebroken piscisaureus
00:12:14  <DrPizza>he is now talking weirdness
00:12:34  <piscisaureus>bnoordhuis: I don't like it very much
00:12:43  <bnoordhuis>suggestions?
00:12:44  <piscisaureus>but if it's temporary I guess it's okay
00:13:00  <bnoordhuis>yeah, it'll go once we drop legacy
00:13:06  <piscisaureus>okay
00:13:11  <piscisaureus>+1 then
00:13:22  <bnoordhuis>maybe errnoexception should be killed off completely
00:13:28  <bnoordhuis>but that's something for another day
00:13:45  <DrPizza>let's just delete legacynow
00:15:39  <CIA-52>node: Ben Noordhuis master * r554dc63 / (12 files in 6 dirs): uv: upgrade to 58ef43e - http://git.io/0SqIlQ
00:15:40  <CIA-52>node: Ben Noordhuis master * r2d9ab49 / (3 files in 2 dirs):
00:15:40  <CIA-52>node: file: fix errno propagation in sync fs calls
00:15:40  <CIA-52>node: Include libuv fs error code in the JS exception.
00:15:40  <CIA-52>node: Fixes #1657. - http://git.io/z-tqMw
00:16:17  <bnoordhuis>not having legacy around would make things easier
00:16:33  <bnoordhuis>but hey, nobody ever promised us a rose garden
00:17:37  <dmkbot>joyent/node: Kos: Misleading console output for arrays with undefined - https://github.com/joyent/node/issues/1651
00:27:50  <piscisaureus>Kos: Misleading console output for arrays with undefined - <-- this is also a drama
00:30:07  <DrPizza>bnoordhuis: hey, we might as well do somethingexciting while ryah is away!
00:30:21  <bnoordhuis>like fixing bugs?
00:30:30  <bnoordhuis>[12:02|% 100|+ 527|- 27]: Done <- `make test-all` on linux :(
00:52:53  <CIA-52>libuv: Bert Belder childstdio * r3edb89a / (include/uv-private/uv-win.h src/win/process.c): win: correctly manage child process stdio - http://git.io/Z49ufg
00:52:58  <piscisaureus>DrPizza: ^-- that fixes it. can you look at it?
00:57:14  <CIA-52>libuv: Bert Belder childstdio * r2de1ac1 / (include/uv-private/uv-win.h src/win/process.c): win: correctly manage child process stdio handles - http://git.io/u-fZYw
00:57:39  <DrPizza>piscisaureus: looking
00:58:11  <piscisaureus>DrPizza: I also made it so that if you don't specify a stdin/stdout pipe, it inherits the parent's handle
00:58:17  <DrPizza>hmm
00:58:20  <piscisaureus>(unless obviously the parent has no handle)
00:59:32  <DrPizza>so
00:59:42  <DrPizza>we makethe process object only contain the parent-side handles
00:59:47  <DrPizza>the ones it needs to do I/O on
01:00:02  <piscisaureus>DrPizza: no it only contains child-side handles
01:00:06  <DrPizza>oh
01:00:08  <DrPizza>right
01:00:11  <DrPizza>because the parent side get passed in
01:00:16  <piscisaureus>yes
01:00:33  <piscisaureus>DrPizza: the child handles are not stored most of the time
01:01:08  <piscisaureus>DrPizza: only in the edge case that we need to start a thread and print the error to the child's stderr, that array is actually used
01:01:16  <DrPizza>ah
01:01:37  <piscisaureus>DrPizza: I have been wanting to fix up this for a long time. There were many loose ends
01:06:02  <piscisaureus>Chorale: Peter, der nicht denkt zurück
01:06:28  <DrPizza>wut
01:07:00  <DrPizza>piscisaureus: i think it looks right
01:07:38  <CIA-52>libuv: Ben Noordhuis master * r79d9f81 / (src/unix/fs.c test/test-fs.c test/test-list.h): unix: implement uv_fs_futime, add tests for uv_fs_utime and uv_fs_futime - http://git.io/chyQGg
01:07:39  <DrPizza>although
01:07:40  <DrPizza>hmm
01:07:45  <DrPizza>do we not have a race now
01:08:10  <piscisaureus>Nein
01:08:10  <piscisaureus>Aber, Dein Will gescheh, Herr Gott, zugleich
01:08:24  <DrPizza>piscisaureus: if spawning fails, the handles will stay open until the threadpool task gets performed, won't they?
01:08:37  <piscisaureus>DrPizza: hmm, yes good point
01:08:45  <DrPizza>which means that if another task spawns quickly
01:08:53  <DrPizza>it'll inherit the handles
01:08:56  <DrPizza>I guess it won't be such an issue
01:08:59  <DrPizza>sinc ehtey'll close "soon"
01:09:04  <piscisaureus>yeah
01:09:08  <DrPizza>because the threadpool will close its side
01:09:08  <piscisaureus>but good point
01:09:12  <DrPizza>and it doesn't atter about the process's own
01:09:18  <DrPizza>but it's still getting foreign handles that it doesn't want
01:09:22  <piscisaureus>Yes
01:09:52  <piscisaureus>We could use SetHandleInformation to prevent it from being duped to another child
01:09:54  <piscisaureus>yes
01:09:59  <piscisaureus>only marginally more ugly
01:10:01  <DrPizza>yeah
01:10:12  <piscisaureus>good catch!
01:15:20  <CIA-52>libuv: Bert Belder childstdio * r83311f9 / (include/uv-private/uv-win.h src/win/process.c): win: correctly manage child process stdio handles - http://git.io/VbuVxg
01:15:26  <piscisaureus>DrPizza: okay?
01:16:20  <DrPizza>yes, that looks fair
01:17:15  <piscisaureus>nice
01:17:51  <DrPizza>I take it it fixes the bug :)
01:18:05  <piscisaureus>DrPizza: yes, I wrote that in node first
01:18:18  <piscisaureus>then exported with git format-patch --relative=deps/uv
01:18:18  <DrPizza>ok cool
01:18:24  <DrPizza>oh that is handy
01:18:30  <DrPizza>I will have to ask you to remind me of that in th efuture
01:18:39  <piscisaureus>sure
01:18:54  <CIA-52>libuv: Bert Belder master * rde8ae8d / (include/uv-private/uv-win.h src/win/process.c): win: correctly manage child process stdio handles - http://git.io/5QA1lQ
01:19:28  <piscisaureus>ghe
01:19:34  <piscisaureus>bug cought
01:19:38  <piscisaureus>caught
01:19:44  <piscisaureus>bnoordhuis: I'm gonna force push
01:20:10  <bnoordhuis>piscisaureus: what?
01:20:17  <bnoordhuis>oh, libuv
01:20:29  <CIA-52>libuv: Bert Belder master * r3d538af / (include/uv-private/uv-win.h src/win/process.c): win: correctly manage child process stdio handles - http://git.io/1d0AQQ
01:20:54  <piscisaureus>- SetHandleInformation(child_stdio[1], HANDLE_FLAG_INHERIT, 0);
01:20:54  <piscisaureus>+ SetHandleInformation(child_stdio[i], HANDLE_FLAG_INHERIT, 0);
01:20:59  <piscisaureus>that --^
01:21:27  <DrPizza>doh
01:22:17  <piscisaureus>I like this patch
01:22:26  <piscisaureus>I knew this could be an issue for a long time
01:22:34  <piscisaureus>it feels like paying back technical debt
01:24:33  <piscisaureus>bnoordhuis:
01:24:34  <piscisaureus>test\test-fs.c(1227): error C2039: 'statbuf' : is not a member of 'uv_fs_s' [D:
01:24:34  <piscisaureus>\nodejs\libuv\run-tests.vcxproj]
01:24:34  <piscisaureus> D:\nodejs\libuv\include\uv.h(887) : see declaration of 'uv_fs_s'
01:24:34  <piscisaureus>:-( :-( :-(
01:24:50  <bnoordhuis>ow, sorry
01:25:45  <bnoordhuis>piscisaureus: how do the uv_fs_*stat functions work on windows?
01:26:19  <piscisaureus>bnoordhuis they use req->stat :-)
01:26:26  <piscisaureus>but the struct type is different
01:26:33  <piscisaureus>_stati64 to be exact
01:26:43  <piscisaureus>I think we should have uv_stat_t
01:26:49  <bnoordhuis>right, i see the #defines at the top of test-fs.c
01:26:58  <bnoordhuis>yes, this is confusing and ugly
01:27:26  <bnoordhuis>then again, copying the stat struct on every syscall... hmm
01:27:31  <bnoordhuis>i'll fix the test anyway
01:27:37  <dmkbot>joyent/node: geki007: windows: spawn process: close order shouldn't important - https://github.com/joyent/node/issues/1438
01:27:39  <dmkbot>joyent/node: Bonuspunkt: cant enter @{[]}~ at REPL on windows - https://github.com/joyent/node/issues/1359
01:32:50  <bnoordhuis>st_atime Time of last access of file. Valid on NTFS but not on FAT formatted disk drives.
01:32:54  <bnoordhuis>windows--
01:33:45  <piscisaureus>bnoordhuis: we will have our own implementation of that I think
01:33:56  <piscisaureus>but fat doesn't store it
01:34:14  <piscisaureus>I'm pretty sure it is not valid for FAT on linux either
01:34:34  <bnoordhuis>piscisaureus: it's not that surprising indeed because it simply isn't there
01:34:42  <bnoordhuis>still sucks though
01:40:13  <piscisaureus>bnoordhuis: what should dirname('/foo/bar/') return?
01:40:28  <piscisaureus>. /foo/bar or /foo ?
01:40:39  <bnoordhuis>piscisaureus: '/foo/bar'
01:41:07  <bnoordhuis>hmm, `dirname /foo/bar/` prints '/foo'
01:42:03  <piscisaureus>on the command line?
01:42:16  <piscisaureus>in that case the current implementation is correct :-)
01:42:51  * felixgequit (Quit: felixge)
01:43:03  <piscisaureus>assert.equal(path.dirname('/a/b/'), '/a');
01:43:10  <piscisaureus>yes, it's defined in the tests :-)
01:47:49  <bnoordhuis>piscisaureus: https://github.com/bnoordhuis/libuv/commit/18f8cc3 <- does that work for you?
01:48:54  <bnoordhuis>and let's fix uv_stat_t tomorrow
01:49:26  <piscisaureus>bnoordhuis: compiles, fs_file_noent, fs_utime, fs_futime fail
01:49:57  <bnoordhuis>do you know why?
01:50:27  <bnoordhuis>fs_file_noent fails because libuv on windows doesn't set UV_ENOENT
01:50:45  <piscisaureus>I think it doesn't propagate crt errors at all atm
01:51:09  <bnoordhuis>what's the resolution of st_atime and st_mtime btw? seconds, milliseconds?
01:51:39  <piscisaureus>bnoordhuis: on windows you mean
01:51:42  <piscisaureus>or in general
01:51:45  <piscisaureus>it is time_t
01:51:49  <bnoordhuis>piscisaureus: on windows
01:51:55  <bnoordhuis>oh, so it's seconds?
01:52:06  <bnoordhuis>tests should pass then
01:52:56  <piscisaureus>bnoordhuis:
01:52:56  <piscisaureus>`fs_utime` failed: exit code 3
01:52:56  <piscisaureus>Output from process `fs_utime`:
01:52:56  <piscisaureus>Assertion failed in test\test-fs.c on line 422: s->st_atime == atime
01:52:56  <piscisaureus>=============================================================
01:52:57  <piscisaureus>[% 0|+ 61|- 2]: fs_futime
01:52:57  <piscisaureus>`fs_futime` failed: exit code 3
01:52:58  <piscisaureus>Output from process `fs_futime`:
01:52:58  <piscisaureus>Assertion failed in test\test-fs.c on line 1231: req.result != -1
01:52:59  <piscisaureus>=============================================================
01:55:53  <bnoordhuis>req.result != -1 means the file to futime couldn't be opened
01:56:47  <piscisaureus>hmm
01:56:54  <piscisaureus>. // Regex to split a filename into [*, dir, basename, ext]
01:56:54  <piscisaureus>. // windows version
01:56:54  <piscisaureus>. var splitPathRe = /^(([a-zA-Z]:|[\\\/]{2}[^\\\/]+[\\\/][^\\\/]+)?.+(?:[\\\/](?!$)|:)|[\\\/])?((?:.+?)?(\.[^.]*)?)$/;
01:57:11  <piscisaureus>^-- please review :-) hehehehe
01:59:38  <bnoordhuis>haha, DrPizza!
01:59:44  <DrPizza>?
02:00:47  <bnoordhuis>DrPizza: two lines up :)
02:00:59  <DrPizza>yeah...
02:01:01  <DrPizza>not reviewing that :p
02:01:18  <piscisaureus>It's annoying
02:01:28  <piscisaureus>I don't want to specialize basename, extname etc
02:02:15  * piscisaureuspart
02:02:19  <CIA-52>libuv: Ben Noordhuis master * re8ab5cb / test/test-fs.c : test: fix utime and futime tests on windows - http://git.io/ATms-g
02:02:19  * piscisaureusjoined
02:08:08  <bnoordhuis>https://www.destroyallsoftware.com/blog/2011/one-base-class-to-rule-them-all <- pretty hilarious!
02:13:24  <bnoordhuis>[12:15|% 100|+ 528|- 26]: Done <- `make test-all` on linux again
02:41:59  <piscisaureus>bnoordhuis: still here?
02:42:22  <bnoordhuis>piscisaureus: yes
02:42:57  <CIA-52>node: Bert Belder pathfix * r372c208 / (lib/path.js test/simple/test-path.js): pathfix - http://git.io/U7tIYA
02:43:09  <piscisaureus>bnoordhuis: can you check for me if that (^^) stil passes test/simple/test-path.js on linux?
02:45:53  <bnoordhuis>piscisaureus: it does
02:45:57  <piscisaureus>:-)
02:45:59  <piscisaureus>phew
02:46:07  * piscisaureuswipes sweat off his forehead
02:47:23  <bnoordhuis>okay, off to bed
02:47:25  <bnoordhuis>sleep tight, guys
02:47:42  <piscisaureus>yes, me too
02:47:51  <piscisaureus>same to you bnoordhuis
02:47:55  <CIA-52>node: Bert Belder master * r8153a21 / (lib/path.js test/simple/test-path.js):
02:47:55  <CIA-52>node: Improve path parsing on windows
02:47:55  <CIA-52>node: Closes #650 - http://git.io/lrR9-A
02:51:53  * bnoordhuisquit (Ping timeout: 258 seconds)
02:57:37  <dmkbot>joyent/node: ry: DNS stuff needs _WIN32_WINNT to be 0x0501 but libeio\xthread.h defines it to be 0x0400. - https://github.com/joyent/node/issues/1236
03:12:37  <dmkbot>joyent/node: ry: DNS stuff needs _WIN32_WINNT to be 0x0501 but libeio\xthread.h defines it to be 0x0400. - https://github.com/joyent/node/issues/1236
03:14:29  <piscisaureus>DrPizza: can you point to a commit where you fixed #1236?
03:14:35  <DrPizza>hmmm
03:14:45  <DrPizza>'twas ages ago
03:16:07  <piscisaureus>DrPizza: otherwise, should I just close it then?
03:16:14  <DrPizza>well
03:16:38  <piscisaureus>maybe you can look if the offending define has been removed? (I am not sure where it is/was myself)
03:17:18  <DrPizza>well
03:17:21  <DrPizza>it was in xthread.h
03:17:25  <DrPizza>which windows buildsdon't even use now
03:17:33  <DrPizza>since xthread.h is an eio thing
03:18:17  <DrPizza>https://github.com/joyent/libuv/blob/master/src/unix/eio/xthread.h
03:18:18  <piscisaureus>" once we go through libuv for async calls we can remove the define in node.h"
03:18:19  <DrPizza>it's not there
03:18:21  <piscisaureus>said ryan
03:18:27  <DrPizza>node.h huh
03:18:31  <piscisaureus>well, ok then :-)
03:18:37  <piscisaureus>should I close it
03:18:37  <DrPizza>let's hav ea look there
03:19:25  <DrPizza>hmm
03:19:32  <DrPizza>the node one is still there
03:19:33  <DrPizza>what the hell
03:19:38  <DrPizza>I'm sure I switched to using sdkddkver.h
03:20:34  <DrPizza>piscisaureus: can you search the entire node project for an include of sdkddkver.h ?
03:20:45  <piscisaureus>sure
03:22:49  <piscisaureus>Binary file node.sdf matches
03:22:52  <piscisaureus>nothing else
03:23:25  <piscisaureus>and a lot of object files but I filtered them out
03:23:27  <DrPizza>huh
03:23:28  <DrPizza>ok
03:23:29  <DrPizza>that's odd
03:23:38  <DrPizza>I'm going to have to look into this
03:23:45  <DrPizza>it's possible the change got reverted for some reason
03:24:27  <piscisaureus>no
03:24:32  <piscisaureus>definitely not there
03:24:49  <DrPizza>hrm
03:32:37  <dmkbot>joyent/node: aseemk: path.split() needed, as counterpart to path.join() - https://github.com/joyent/node/issues/1224
03:36:20  <DrPizza>piscisaureus: anyway the bug can be closed
03:36:37  <piscisaureus>ok
03:42:37  <dmkbot>joyent/node: Skomski: Added new win32 platform function: getNetworkInterfaces() - https://github.com/joyent/node/issues/1652
06:00:14  * mralephjoined
06:10:18  * dmkbotquit (Remote host closed the connection)
06:11:24  * dmkbotjoined
07:53:57  * mralephquit (Quit: Leaving.)
08:01:22  <dmkbot>joyent/node: herby: runInThisContext does not run in context of the caller - https://github.com/joyent/node/issues/898
08:16:22  <dmkbot>joyent/node: Kos: Misleading console output for arrays with undefined - https://github.com/joyent/node/issues/1651
08:23:48  * felixgejoined
08:24:02  * felixgequit (Client Quit)
08:36:22  <dmkbot>joyent/node: geki007: windows: spawn process: close order shouldn't important - https://github.com/joyent/node/issues/1438
08:46:22  <dmkbot>joyent/node: Kos: Misleading console output for arrays with undefined - https://github.com/joyent/node/issues/1651
11:01:22  <dmkbot>joyent/node: PS86K: Problem with connection 3g - https://github.com/joyent/node/issues/1660
11:31:22  <dmkbot>joyent/node: fprijate: Wrong fs.stat size on win32 - https://github.com/joyent/node/issues/1647
12:43:27  * bnoordhuisjoined
13:11:22  <dmkbot>joyent/node: chowey: net_uv.js errno not defined - https://github.com/joyent/node/issues/1661
13:11:32  <bnoordhuis>piscisaureus: https://github.com/Skomski/node/commit/238a86a1219f94e04f0d8150b0e0e5ee14d89980 *sigh*
13:48:02  <CIA-52>libuv: Ben Noordhuis master * rcc91989 / src/unix/core.c :
13:48:02  <CIA-52>libuv: unix: fix buffer overrun in uv__strlcpy()
13:48:02  <CIA-52>libuv: Reported by Thomas Shinnick. - http://git.io/kGVOOA
13:58:05  * dmkbotquit (Ping timeout: 260 seconds)
14:38:54  <CIA-52>node: Ben Noordhuis master * r94b0481 / (6 files in 4 dirs): uv: upgrade to cc91989 - http://git.io/Qg4DRw
14:38:54  <CIA-52>node: Ben Noordhuis master * r56efe9c / src/node_file.cc :
14:38:55  <CIA-52>node: file: translate libuv error codes to errno names
14:38:55  <CIA-52>node: Fixes incorrect error names (ex. ENOEXEC instead of EBADF,
14:38:55  <CIA-52>node: EMLINK instead of ENOENT).
14:38:55  <CIA-52>node: Fixes #1656. - http://git.io/91KL0g
15:05:14  <CIA-52>node: Maciej Małecki master * rd055294 / lib/url.js : url: add plus sign to protocol pattern - http://git.io/fgm_IQ
15:05:15  <CIA-52>node: Maciej Małecki master * rfb93ab4 / test/simple/test-url.js : test: test for correct parsing of URLs with plus character in protocol - http://git.io/_NHUow
15:11:13  * dmkbotjoined
15:28:02  <bnoordhuis>https://github.com/tshinnic/node-1/commit/c425bc8a467332bd7ea8619bf9258df184da0475 <- longest commit log evah!
15:36:10  <dmkbot>joyent/node: Kos: Misleading console output for arrays with undefined - https://github.com/joyent/node/issues/1651
15:46:10  <dmkbot>joyent/node: jetlan: node v0.5.5 + mingw 'PROCESS_MEMORY_COUNTERS' was not declared in this scope - https://github.com/joyent/node/issues/1662
15:46:12  <dmkbot>joyent/node: jetlan: node v0.5.5 + mingw undefined reference to `[email protected]' - https://github.com/joyent/node/issues/1663
15:46:48  * dmkbotquit (Remote host closed the connection)
15:47:18  * dmkbotjoined
15:57:16  <dmkbot>joyent/node: fprijate: Wrong fs.stat size on win32 - https://github.com/joyent/node/issues/1647
15:57:17  <dmkbot>joyent/node: koichik: Buffer.write() should always set Buffer._charsWritten - https://github.com/joyent/node/issues/1633
16:17:16  <dmkbot>joyent/node: koichik: Buffer.write() should always set Buffer._charsWritten - https://github.com/joyent/node/issues/1633
16:21:30  <CIA-52>node: Thomas Shinnick master * r4cf0ce5 / (9 files): (log message trimmed)
16:21:31  <CIA-52>node: docs: typos and minor edits in several modules
16:21:31  <CIA-52>node: Mostly quite minor edits. Those possibly of more interest are:
16:21:31  <CIA-52>node: emitter.setMaxListeners(n)
16:21:31  <CIA-52>node: That the limit is per event name for an emitter.
16:21:31  <CIA-52>node: fs.readlink()
16:21:31  <CIA-52>node: Not a path, but rather the symbolic link's string value, which
16:27:16  <dmkbot>joyent/node: jordansissel: Add '.' '+' and '-' in url protocol pattern - https://github.com/joyent/node/issues/1603
17:02:16  <dmkbot>joyent/node: jordansissel: Add '.' '+' and '-' in url protocol pattern - https://github.com/joyent/node/issues/1603
17:17:16  <dmkbot>joyent/node: jordansissel: Add '.' '+' and '-' in url protocol pattern - https://github.com/joyent/node/issues/1603
17:17:54  * graydonjoined
17:32:16  <dmkbot>joyent/node: jordansissel: Add '.' '+' and '-' in url protocol pattern - https://github.com/joyent/node/issues/1603
17:33:25  * brsonjoined
17:49:12  <bnoordhuis>no scrum call today right?
17:49:42  <bnoordhuis>i'll take that as a yes
17:53:24  <piscisaureus>bnoordhuis: there's never a scum on tuesday
17:53:29  <piscisaureus>it's msft call
17:53:59  <piscisaureus>bnoordhuis: but that got moved to thursday
17:57:16  <dmkbot>joyent/node: kuebk: Unexpected behavior of console.log under certain condition - https://github.com/joyent/node/issues/1634
18:15:49  * brsonquit (Read error: Connection reset by peer)
18:16:29  * brsonjoined
18:57:16  <dmkbot>joyent/node: Kos: Misleading console output for arrays with undefined - https://github.com/joyent/node/issues/1651
19:07:16  <dmkbot>joyent/node: Kos: Misleading console output for arrays with undefined - https://github.com/joyent/node/issues/1651
19:23:44  * mralephjoined
19:32:16  <dmkbot>joyent/node: Skomski: New win32 platform function: GetCPUInfo - https://github.com/joyent/node/issues/1664
19:43:08  * mralephquit (Read error: Connection reset by peer)
19:43:08  * mraleph1joined
19:49:32  <piscisaureus>bnoordhuis: http://webwereld.nl/nieuws/107838/linux-kernel-verhuist-tijdelijk-naar-github.html
20:06:29  <piscisaureus>mraleph1: so, did you chat with your colleagues?
20:06:50  <mraleph1>piscisaureus: oh yeah
20:07:07  <mraleph1>so caching is fragile
20:07:12  <mraleph1>better force flattening
20:07:22  <mraleph1>instead of trying and ignoring failure
20:07:47  <piscisaureus>mraleph1: that obviously works
20:08:17  <piscisaureus>mraleph1: but in many cases this UTF8Write call is really a one-off
20:08:24  <mraleph1>maybe you should fle an issue
20:08:43  <mraleph1>otherwise all this might be lost in discussion ;-)
20:08:59  <piscisaureus>mraleph1: actually we now set the HINT_MANY_WRITES_EXPECTED hint but I doubt that is the right thing to do
20:09:28  <piscisaureus>yes
20:16:23  <bnoordhuis>piscisaureus: what are you working on tonight?
20:17:01  <piscisaureus>nothing
20:18:20  <bnoordhuis>so it's a night like any other?
20:18:27  <piscisaureus>yes
20:18:31  <bnoordhuis>good
20:18:52  <bnoordhuis>thinking of clocking out early today myself
20:19:49  <piscisaureus>mraleph: http://code.google.com/p/v8/issues/detail?id=1665
20:20:06  <piscisaureus>I will too
20:20:12  <mraleph1>good
20:20:38  <piscisaureus>mraleph1: the problem with issues is that they're a sinkhole for problems that never get fixed
20:20:48  <piscisaureus>mraleph1: (this is true for node too, btw)
20:21:48  <bnoordhuis>:)
20:22:41  <mraleph1>well… no comments ;-)
20:23:09  <piscisaureus>mraleph1: :-S
20:24:00  <piscisaureus>mraleph1: I always consider "no comments" as a hint that you're actually working on it
20:24:19  <piscisaureus>mraleph1: but should I now take the hint that you're closing the issue tracker?
20:24:26  <mraleph1>heh then I should find another catch phrase :-)
20:24:44  <mraleph1>nah I just don't share your view on issue tracker
20:24:55  <mraleph1>issue tracker is the pile of tasks
20:24:57  <mraleph1>kinda
20:25:13  <piscisaureus>mraleph1: in a way, yes
20:25:22  <piscisaureus>mraleph1: this is a problem for us too
20:25:29  <mraleph1>and those tasks get done ;-)
20:25:34  <mraleph1>sooner or later
20:25:41  <piscisaureus>I contents that
20:25:46  <piscisaureus>some of them get done
20:25:54  <piscisaureus>but I mean, look: https://github.com/joyent/node/issues?sort=created&direction=desc&state=open
20:26:11  <piscisaureus>there are serious and important issues there
20:26:30  <piscisaureus>but there are also many somewhat-valid problems that nobody knows how to deal with
20:26:33  <mraleph1>you need a vacuum cleaning there
20:26:42  <piscisaureus>they just blur our view
20:27:00  <bnoordhuis>augean stables
20:27:08  <bnoordhuis>it's coming in faster than we can clean it out :/
20:27:38  <piscisaureus>not really or?
20:27:49  <piscisaureus>bnoordhuis: we were at 500 open 2 months ago right?
20:28:00  <bnoordhuis>were we?
20:28:05  <bnoordhuis>over 400 anyway
20:28:05  <piscisaureus>I'm not sure
20:28:09  <piscisaureus>hmm maybe not
20:28:14  <bnoordhuis>okay, so maybe it's not that bad
20:28:32  <piscisaureus>bnoordhuis: we should solve issues in fifo order
20:28:35  <piscisaureus>I think that would help
20:29:01  <bnoordhuis>yes
20:29:16  <bnoordhuis>but not now :)
20:31:33  <DrPizza>bnoordhuis: oh good
20:31:40  <DrPizza>bnoordhuis: 1638 was the stupidest bug report ever
20:32:02  <bnoordhuis>oh, i don't know about that
20:32:10  <bnoordhuis>have you seen the one about printing arrays?
20:32:15  <DrPizza>lol
20:32:16  <dmkbot>joyent/node: miksago: TODO: Use getopt_long() in ParseArgs - https://github.com/joyent/node/issues/344
20:32:20  <DrPizza>that one is less stupid imo
20:32:49  <bnoordhuis>maybe not
20:33:03  <DrPizza>well, isn't the array length still 3?
20:33:04  <bnoordhuis>but it's attracting too much comments for such a stupid simple thing
20:33:11  <DrPizza>if so, it should still print 3 things.
20:33:23  <bnoordhuis>yes, it's not an entirely invalid issue
20:33:26  <DrPizza>although "don't use delete on arrays" seems like a good workaround in the meantime.
20:33:36  <bnoordhuis>but it quickly turned into a bikeshed
20:34:02  <piscisaureus>https://github.com/joyent/node/issues/596 <-- close this?
20:35:36  <bnoordhuis>see, that's one of the though ones
20:35:48  <piscisaureus>why?
20:35:49  <bnoordhuis>the issue itself is valid, i just don't think we'll ever fix it
20:36:04  <piscisaureus>bnoordhuis: but _ is used only in the repl right?
20:36:07  <bnoordhuis>yes
20:36:20  <piscisaureus>so, wtf
20:36:34  <piscisaureus>this guy is coding in the repl?
20:36:42  <bnoordhuis>you mean you don't? :)
20:36:52  <bnoordhuis>lots of people try out stuff in the repl, yours truly included
20:36:55  <piscisaureus>well, I don't load underscore.js in the repl
20:37:16  <dmkbot>joyent/node: miksago: TODO: Use getopt_long() in ParseArgs - https://github.com/joyent/node/issues/344
20:37:26  <bnoordhuis>piscisaureus: okay, close it
20:39:47  <piscisaureus>bnoordhuis: "We use `_` because we need to use *something*, and it has to be a valid javascript name. There are not so many good options. $ would be another possibility, but that would clash with jquery. So we can't fix this; if you want to use underscore from the REPL, assign it to another variable"
20:39:49  <piscisaureus>right?
20:40:02  <bnoordhuis>yes
20:48:19  * ryahjoined
21:02:16  <dmkbot>joyent/node: ArtS: Weird 'require' behaviour - https://github.com/joyent/node/issues/410
21:12:39  <CIA-52>node: Ben Noordhuis master * rb66d225 / README.md :
21:12:39  <CIA-52>node: docs: document that the build system depends on python >= 2.5.2
21:12:39  <CIA-52>node: Fixes #354. - http://git.io/auvTAQ
21:13:08  * ryahquit (Quit: Lost terminal)
21:17:16  <dmkbot>joyent/node: ArtS: Weird 'require' behaviour - https://github.com/joyent/node/issues/410
21:18:32  <piscisaureus>bnoordhuis: http://pastebin.com/u/ComodoHacker
21:18:44  <piscisaureus>looks genuine
21:19:06  <piscisaureus>this guy signed windows calc with a diginotar cert, i verified that
21:19:21  <bnoordhuis>yeah, read about that
21:20:42  <bnoordhuis>seems he's slightly peeved about srebrenica
21:25:04  <piscisaureus>bnoordhuis: maybe there were more hackers?
21:25:39  <piscisaureus>bnoordhuis: I mean, FoxIT said there were many malwares found on Diginotar's servers
21:25:40  <piscisaureus>ghg
21:26:05  <bnoordhuis>maybe, who can say?
21:27:00  <bnoordhuis>at any rate, it's a good thing that diginotar got crushed
21:27:26  <bnoordhuis>i just hope its board members get prosecuted from criminal liability
21:34:15  * dmkbotquit (Remote host closed the connection)
21:34:33  * dmkbotjoined
21:44:27  * mraleph1quit (Quit: Leaving.)
22:09:31  <dmkbot>joyent/node: ry: setTimeout(fn) does not work as expected - https://github.com/joyent/node/issues/593
22:24:31  <dmkbot>joyent/node: Skomski: New win32 platform function: GetCPUInfo - https://github.com/joyent/node/issues/1664
22:34:31  <dmkbot>joyent/node: andrewschaaf: Very incorrect column numbers in exception stack text - https://github.com/joyent/node/issues/527
23:11:39  <piscisaureus>bnoordhuis: cpuinfo lgtm behalve de perror calls dan
23:13:50  <CIA-52>node: Ryan Dahl zlib * r20e6bb6 / (43 files in 3 dirs): import zlib - http://git.io/x61sqg
23:14:01  * ryahjoined
23:14:07  * ryahquit (Client Quit)
23:14:17  * ryahjoined
23:15:29  * bnoordhuisquit (Ping timeout: 258 seconds)
23:39:31  <dmkbot>joyent/node: scunningham: Support ssl session get/set in tls. Useful for client side session resume. - https://github.com/joyent/node/issues/1606
23:39:56  <DrPizza>ryah: oh gosh
23:40:04  <DrPizza>ryah: I was going to give you a zlib branch myself
23:49:31  <dmkbot>joyent/node: georgesnelling: util.assert long message strings are truncated on mac osx - https://github.com/joyent/node/issues/1590