00:00:00  * ircretaryquit (Remote host closed the connection)
00:00:08  * ircretaryjoined
00:04:00  * wolfeidauquit (Remote host closed the connection)
00:05:28  * wolfeidaujoined
00:12:43  * brsonquit (Quit: leaving)
00:22:07  * seldoquit
00:26:56  * wolfeidauquit (Remote host closed the connection)
00:34:10  * a_le_quit (Remote host closed the connection)
00:35:26  * wolfeidaujoined
00:39:30  * mkrufkyjoined
00:42:37  <mkrufky>indutny: not sure if you recall... i asked a question a few weeks back asking how i could schedule a job to run but to be able to queue it to a specific thread rather than the uv threadpool ... you suggested that i create my own thread and to use a uv_async_t to notify main thread upon completion .....
00:42:50  <indutny>yeah
00:42:55  <indutny>mkrufky: I do remember
00:42:56  <indutny>what's up?
00:42:56  <mkrufky>anyway, I got it all done that weekend and it's working nicely. i just wanted to thank you
00:43:00  <indutny>oh
00:43:01  <indutny>great!
00:43:02  <indutny>congrats
00:43:03  <mkrufky>:-)
00:43:21  <mkrufky>and i figured, if you were here and active, id mention the shortcomings that i am still having
00:43:24  <mkrufky>:-)
00:43:37  <mkrufky>for my purposes, this solution *is* good enough
00:43:49  <mkrufky>but... in the grand scheme of things, something more robust would be cool
00:43:54  <mkrufky>for instance...
00:44:38  * a_lejoined
00:44:48  <mkrufky>when using I schedule a job using uv_queue_work , the delay for it to actually run and call back the main thread is usually very short
00:45:22  <mkrufky>but when i have to use my own thread, and using uv_asynv_t to call back the main thread, depending on what else is happening on the system, the uv_async_t style callback can be delayed by a long time
00:45:42  <mkrufky>again, for my purposes, i have it working well enough. .. but for the sake of scratching this itch ............
00:46:51  <indutny>mkrufky: hm...
00:46:52  <indutny>this is very odd
00:47:02  <indutny>mkrufky: uv_queue_work does work the same way internally
00:47:06  <mkrufky>you mentioned that some other open issues may help the future of this situation, but that maybe i should open a LEP for a new feature
00:47:42  <srl295>jgi: ping
00:47:46  <mkrufky>hmm ... i thought, that because uv_async_t is thread safe, there are only fixed times in which the main thread is able to be woken from an uv_async_t
00:48:21  <mkrufky>but in the case of uv_queue_work, the callback is expected and explicitly checked for ....
00:48:35  <mkrufky>ill admit that i havent delved deep enough in those internals to actually understand the mechanics
00:49:08  <mkrufky>the specific case where i see this happening is as follows:
00:49:25  <mkrufky>an embedded device running nodejs outputting to the console connected via serial cable
00:49:57  * a_lequit (Remote host closed the connection)
00:50:06  <mkrufky>jobs scheduled using uv_queue_work tend to execute 0 - 3 ms and call back the main thread also 0-3 ms later, worst case scenario
00:50:33  * a_lejoined
00:50:47  <mkrufky>meanwhile ..... jobs in a separate thread calling back the main thread using a uv_async_t .... well, initial jobs call back main thread quickly, but the longer the process runs the longer the callback delay becomes
00:50:59  <mkrufky>ive seen it take 45 seconds to wake the main thread
00:51:18  <mkrufky>meanwhile, other uv_queue_work style jobs running at the same time still exec and callback in 0-3 ms
00:51:49  <mkrufky>now, this is an exaggerated situation, obviously nodejs needed to be detached from the console. i did that and performance problem disappeared entirely
00:52:39  <mkrufky>... but its interesting to see how the uv_async_t style callbacks get delayed when the system is busied up and blocking on the console
00:53:49  <mkrufky>oh, and again, this is the version of libuv that is shipped with nodejs 0.10.26 ... i believe that is 0.10
00:58:48  <jgi>srl295: pong
00:59:14  <srl295>jgi: congrats on the .15 - but i see some extra files there
00:59:22  <srl295>jgi: and, you pinged
01:00:39  <jgi>srl295: what extra files? iculslocs.* and icupkg.&?
01:00:45  <jgi>srl295: sorry, icupkg.*
01:00:46  <srl295>jgi: right
01:01:17  * crispyzealjoined
01:01:27  <jgi>srl295: right, sorry for that
01:01:34  <jgi>srl295: and thanks for letting me know
01:01:36  <srl295>jgi: np
01:01:57  <jgi>srl295: I’m trying to remember why I pinged you :)
01:02:27  <srl295>jgi: gen* and icu* sholdn't be there., not sure if something different needs to happen in the source base
01:03:12  <jgi>srl295: regarding extra files, I’m going to investigate why they ended up in the dist directory
01:03:25  <jgi>srl295: re node summit, that’s great, I don’t know yet if I’ll be there
01:03:44  <crispyzeal>hij1nx, surprised to see you working with C++
01:03:45  <jgi>srl295: but I would be very happy to meet you in person!
01:04:49  <srl295>jgi: fwiw!
01:22:20  * chris_99quit (Remote host closed the connection)
01:33:49  * toastynerdjoined
01:37:11  * inolenquit (Read error: Connection reset by peer)
01:37:38  * inolenjoined
01:38:41  * wolfeidauquit (Ping timeout: 246 seconds)
01:50:01  * wolfeidaujoined
01:54:12  * dap_quit (Quit: Leaving.)
01:54:29  * a_le_joined
01:55:08  * a_lequit (Read error: Connection reset by peer)
01:56:20  * brsonjoined
02:01:27  * brsonquit (Quit: leaving)
02:17:38  * thlorenzquit (Remote host closed the connection)
02:25:26  <crispyzeal>is the documentation for libuv in uv.h gone now? i remember it being more thorough
02:34:51  * iarnaquit (Remote host closed the connection)
02:45:14  * AvianFluquit (Ping timeout: 245 seconds)
02:47:44  * iarnajoined
03:14:00  * iarnaquit (Remote host closed the connection)
03:16:42  * inolenquit (Remote host closed the connection)
03:17:27  * inolenjoined
03:18:24  * thlorenzjoined
03:22:53  * thlorenzquit (Ping timeout: 245 seconds)
03:27:45  * crispyzealquit (Quit: bye)
03:29:32  * qard_quit (Quit: leaving)
03:29:36  * rmgquit (Remote host closed the connection)
03:30:47  * ferossquit (Read error: Connection reset by peer)
03:31:35  * wwicks_quit (Ping timeout: 272 seconds)
03:31:52  * Tux64quit (Read error: Connection reset by peer)
03:35:54  * groundwaterquit (Ping timeout: 244 seconds)
03:42:23  * wolfeidauquit (Ping timeout: 240 seconds)
03:44:53  * ferossjoined
03:47:14  * groundwaterjoined
03:51:04  * groundwaterquit (Write error: Connection reset by peer)
03:52:46  * octetcloudquit (Ping timeout: 255 seconds)
03:53:17  * Left_Turnquit (Remote host closed the connection)
03:55:01  * ferossquit (Ping timeout: 272 seconds)
03:58:23  * abraxas_quit (Remote host closed the connection)
03:59:00  * abraxas_joined
03:59:33  * wolfeidaujoined
03:59:41  * dsantiag_joined
04:00:52  * dsantiagoquit (Ping timeout: 255 seconds)
04:03:22  * abraxas_quit (Ping timeout: 240 seconds)
04:03:34  * mkrufkyquit (Ping timeout: 255 seconds)
04:03:45  * Fishrock123quit (Quit: Leaving...)
04:08:31  * groundwaterjoined
04:09:04  * wwicks_joined
04:14:54  * ferossjoined
04:19:14  * thlorenzjoined
04:21:57  * wolfeidauquit (Remote host closed the connection)
04:24:30  * thlorenzquit (Ping timeout: 264 seconds)
04:37:30  * kevinswiberquit (Read error: Connection reset by peer)
04:48:43  * Tux64joined
05:04:53  * brsonjoined
05:07:01  * brsonquit (Client Quit)
05:10:17  * abraxas_joined
05:52:16  * kevinswiberjoined
06:05:04  * a_le_quit (Remote host closed the connection)
06:05:42  * a_lejoined
06:06:05  * thlorenzjoined
06:08:34  * thlorenzquit (Remote host closed the connection)
06:14:30  * iarnajoined
06:15:03  * seishunjoined
06:19:33  * iarnaquit (Ping timeout: 276 seconds)
06:32:22  * iarnajoined
06:34:57  * jgiquit (Quit: jgi)
06:37:03  * Tux64quit (Quit: Connection closed for inactivity)
06:48:47  * crispyzealjoined
06:52:33  * kevinswiberquit (Remote host closed the connection)
06:57:03  * toastynerdquit (Ping timeout: 264 seconds)
07:04:17  <crispyzeal>I'm getting an EXC_BAD_ACCESS in threadpool.c:91@v1.x on OS X 10.10.1 using XCode 6.1.1, does anyone know if that could be caused by my code? it's the call to w->work() after QUEUE_DATA. maybe there's some way to disable worker threads?
07:05:04  <crispyzeal>there's almost no backtrace information, since it's started on a new thread in uv__thread_start
07:05:40  <crispyzeal>should I try on master?
07:06:48  * iarnaquit (Remote host closed the connection)
07:09:23  * thlorenzjoined
07:14:08  * thlorenzquit (Ping timeout: 245 seconds)
07:20:43  * jgijoined
07:24:40  <crispyzeal>nvm, I guess I have to allocate requests on the heap? the examples show it on the stack, so I assumed that's how it was done
07:26:21  * jgiquit (Quit: jgi)
07:28:06  * jreyno40part
07:29:11  <crispyzeal>maybe i should just do a big documentation pull request, lots of things need to be updated in there
07:31:29  * seishunquit (Ping timeout: 246 seconds)
08:07:12  * iarnajoined
08:10:11  * thlorenzjoined
08:15:07  * thlorenzquit (Ping timeout: 255 seconds)
08:15:38  <saghul>crispyzeal: yes, you need to allocate the request on the heap, part of it is used in another thread, to run the work
08:15:52  <saghul>crispyzeal: a documentation PR would be awesome
08:18:31  <crispyzeal>saghul, yeah, it really needs some work... also considering hosting some windows installers for binary libuv builds, if that doesn't already exist
08:18:59  <saghul>it doesn't currently exist
08:19:28  <saghul>I was contacted by MS on the matter, can you please send an email to the mailing list?
08:24:04  <crispyzeal>okay
08:24:39  <crispyzeal>once I get a chance, I'll do that
08:31:11  * SergeiRNDjoined
08:32:23  * iarnaquit (Ping timeout: 246 seconds)
08:35:36  <saghul>crispyzeal: thanks!
09:08:31  * rendarjoined
09:27:26  * SplinterOfChaosquit (Ping timeout: 244 seconds)
09:43:55  * bajtosjoined
09:59:20  * thlorenzjoined
10:03:22  * thlorenzquit (Ping timeout: 240 seconds)
10:06:36  * chris_99joined
10:17:29  * iarnajoined
10:21:54  * iarnaquit (Ping timeout: 245 seconds)
10:35:47  * abraxas_quit (Remote host closed the connection)
10:36:29  * abraxas_joined
10:38:18  * crispyzealquit (Quit: bye)
10:40:53  * Left_Turnjoined
10:41:30  * abraxas_quit (Ping timeout: 276 seconds)
10:45:26  <brucem>The comments in uv.h about loop_new and loop_delete being deprecated .... they refer to a 0.12, but libuv jumped from 0.10 to 1.0, 1.1 and now 1.2 ...
10:45:47  <brucem>(also, there's a deprecated attribute that would've been useful for them)
10:54:56  * SergeiRNDquit (Quit: Leaving.)
10:58:01  * glycoljoined
11:15:17  * bajtosquit (Quit: bajtos)
11:15:27  * glycolpart
11:19:09  * SergeiRNDjoined
11:21:07  * glycoljoined
11:28:12  * Tux64joined
11:44:34  <saghul>brucem: they are deprecated in v1.x, but we won't remove them until we bump the major number
11:44:43  <saghul>sorry for the confusion!
11:48:10  * thlorenzjoined
11:53:17  * thlorenzquit (Ping timeout: 256 seconds)
11:58:45  * glycolpart
12:00:24  * tarrudajoined
12:17:18  * tarrudaquit (Ping timeout: 264 seconds)
12:20:33  <hij1nx>saghul: do you have any experience with precompiled headers? specifically `-include-pch`?
12:20:48  <hij1nx>(clang or gcc)
12:24:56  * abraxas_joined
12:25:17  <hij1nx>oh, maybe i want a precompiled preamble :-/
12:30:06  * abraxas_quit (Ping timeout: 265 seconds)
12:38:56  <brucem>saghul: The comment should get updated!
12:39:10  <brucem>(by someone who knows what it should say, which isn't me)
12:39:43  * mkrufkyjoined
12:40:23  <txdv>how do I get the file name which changed in uv_fs_poll?
12:40:24  <brucem>saghul: but I started updating the Dylan bindings (which were never completed) ... and things seem to be going smoothly now.
12:40:34  <txdv>I dont seem to see anything related to it in the struct that is returned
12:40:43  <brucem>txdv: that stack trace from yesterday was atrocious. Someone should be ashamed.
12:40:53  * mkrufkypart
12:41:27  <txdv>since mono is opensource i have to be ashamed
12:43:30  <txdv>but whats the use of uv_fs_poll if it doesn't return a path to what changed
12:43:31  <txdv>:/
13:06:27  <saghul>txdv: it just monitors the given path, not anything below it
13:06:43  <saghul>so the path that changed is uv_fs_poll_get...
13:06:51  * [spoiler]joined
13:06:54  <saghul>hij1nx: nope, sorry :-(
13:11:08  * tarrudajoined
13:11:30  <saghul>brucem: where was that comment again?
13:11:33  <hij1nx>saghul: yeah, seems like not a lot of people do, thx anyway :)
13:12:21  <txdv>saghul: I have speficially monitor a file?
13:12:34  <txdv>i just put ./ and it still registers file changes within ./
13:12:38  <txdv>but i cant know what file changed
13:14:48  <brucem>saghul: in uv.h
13:25:14  <saghul>txdv: it monitors just the given path, I guess somthing changes in the directory if you add a file to it
13:25:25  <saghul>txdv: all it does is stat() on a timer
13:26:47  <saghul>brucem: for 0.10, I assume. The migration path is documented here: http://docs.libuv.org/en/latest/migration_010_100.html#loop-initialization-and-closing
13:27:28  <saghul>many coments are not right in uv.h and they are all gone in v1.x, I'd rather not go back there...
13:36:52  <brucem>saghul: https://github.com/libuv/libuv/blob/v1.x/include/uv.h#L251
13:37:02  * thlorenzjoined
13:39:33  <saghul>brucem: oh, I see. I thought those were gone. Will fix.
13:41:32  * thlorenzquit (Ping timeout: 245 seconds)
13:54:45  * iarnajoined
13:58:56  * iarnaquit (Ping timeout: 246 seconds)
14:13:46  * abraxas_joined
14:14:14  * lance|afkchanged nick to lanceball
14:17:24  <txdv>but i cant know what file changed
14:17:26  <txdv>o srry
14:19:06  * abraxas_quit (Ping timeout: 264 seconds)
14:21:01  <saghul>probably atime or mtime changed in the directory
14:24:12  <txdv>O god, it is so hard to access the fields within that uv_fs_t
14:24:20  <txdv>how do i get the offset in ffi?
14:33:50  * thlorenzjoined
14:34:22  <brucem>saghul: Got a idle callback happening again. :)
14:43:59  * SergeiRNDquit (Quit: Leaving.)
14:45:35  * Fishrock123joined
14:46:39  * kevinswiberjoined
14:54:42  * SplinterOfChaosjoined
14:56:04  * iarnajoined
14:58:14  <txdv>I'm too stupid to conver timespec_t to a datetime
15:00:23  * iarnaquit (Ping timeout: 252 seconds)
15:11:28  * thlorenzquit (Remote host closed the connection)
15:31:44  <saghul>brucem: \o/
15:33:12  * dshaw_joined
15:48:33  * piscisaureusjoined
15:51:10  * a_lequit (Ping timeout: 265 seconds)
15:53:34  * dshaw_quit (Quit: Leaving.)
15:54:40  * dshaw_joined
15:55:54  * rmgjoined
15:56:27  * a_lejoined
16:03:58  * toastynerdjoined
16:11:58  * thlorenzjoined
16:12:19  * dshaw_quit (Quit: Leaving.)
16:14:31  * seishunjoined
16:17:21  * thlorenzquit (Ping timeout: 256 seconds)
16:27:50  * kevinswiberquit (Remote host closed the connection)
16:28:18  * kevinswiberjoined
16:31:38  * AvianFlujoined
16:32:52  * kevinswiberquit (Ping timeout: 240 seconds)
16:37:27  * tarrudaquit (Read error: Connection reset by peer)
16:39:15  <txdv>I guess I shouldn't close the default loop
16:44:39  * bajtosjoined
16:46:09  * AlexisMochajoined
16:47:28  <txdv>saghul: do I have to close all handles in a loop to delete it?
16:56:19  * glycoljoined
17:02:12  * octetcloudjoined
17:02:13  <glycol>how are rename events fired when watching a directory?
17:08:08  * iarnajoined
17:16:21  * kevinswiberjoined
17:17:05  <srl295>tjfontaine: trevnorris: 0.11.15's default Intl works great, thanks! finally tried it
17:17:33  <tjfontaine>glad -- now we just have ot make everything else work ;)
17:19:43  * dap_joined
17:22:25  * AlexisMochaquit (Read error: Connection reset by peer)
17:25:01  * Ralithquit (Ping timeout: 255 seconds)
17:34:15  <saghul>txdv: yes
17:34:31  <saghul>txdv: you can use uv_walk to help you on that
17:34:54  * brsonjoined
17:35:32  <saghul>glycol: check the events in the callback http://docs.libuv.org/en/latest/fs_event.html#c.uv_fs_event_cb
17:35:38  <saghul>you want UV_RENAME
17:40:51  * AvianFluquit (Ping timeout: 244 seconds)
17:44:53  * thlorenzjoined
17:45:36  * thlorenzquit (Remote host closed the connection)
17:48:47  * jgijoined
17:49:12  * DarkUraniumjoined
17:49:13  <DarkUranium>hey guys
17:49:30  <srl295>jgi: good mornign, was just saying that Intl worked for me in 0.11.15 ( as small-icu)
17:49:31  <DarkUranium>I've a question; does the library work w/ MinGW in Windows?
17:50:03  <jgi>srl295: that’s good news, thank you for sharing that
17:50:14  <jgi>srl295: and good morning to you too :)
17:50:58  <jgi>srl295: and of course congratulations and thank you for making ICU work!
17:51:14  * abraxas_joined
17:51:19  <srl295>jgi: welcome!
17:51:26  <saghul>DarkUranium: yes, it does
17:51:44  <creationix>saghul: any gotchas? I seem to have misconfigured dukluv
17:51:54  <creationix>(re mingw)
17:52:04  <saghul>not that I'm aware of
17:52:10  <saghul>though I recommend using MinGW-w64
17:52:14  <srl295>jgi: btw apropos of packaging stuff.. I'd like to see ICU use CLDR's JSON format so people can leverage things like this http://cldr.unicode.org/development/development-process/design-proposals/json-packaging
17:52:16  <creationix>ok, I must have done something dumb then
17:52:43  <saghul>creationix: the mingw you download from mingw.org has very outdated headers
17:53:06  <saghul>MSYS2 is pretty cool: http://msys2.github.io/
17:53:49  <DarkUranium>saghul, thanks
17:54:03  <jgi>srl295: that looks very interesting
17:54:27  <saghul>DarkUranium: sure!
17:54:30  * Fishrock123quit (Remote host closed the connection)
17:56:09  <jgi>srl295: what’s the current status of that? Draft spec as the title says? Do you know when implementation work would start?
17:56:13  * abraxas_quit (Ping timeout: 245 seconds)
17:56:28  <srl295>jgi: how about "just discussed in committee 30 minutes ago"?
17:56:39  <jgi>srl295: :)
17:57:16  <srl295>jgi: but also for example https://github.com/rxaviers/cldr-data-npm - the JSON data has been around for a while
17:57:23  <DarkUranium>saghul, I was asking because of another project, the author didn't know if it was a bug in libuv, or his bindings (it wouldn't compile in MinGW in Windoze)
17:58:09  <jgi>trevnorris, tjfontaine, chrisdickinson, cjihrig: Do you have some time to review https://github.com/joyent/node/pull/9063?
17:59:20  <tjfontaine>sure -- which parts in partciular, I don't presume to be able to vett openssl core fixes ;)
17:59:32  <cjihrig>jgi: all the changes are to openssl except the one test. i can rubber stamp it for you if you'd like
17:59:36  <saghul>DarkUranium: if it doesn't compile please do report it
17:59:48  <tjfontaine>god whitespace diffs
18:00:33  <tjfontaine>oh look, new define OPENSSL_NO_SSL3_METHOD
18:00:42  <DarkUranium>saghul, I will if it turns out to be a libuv, and not a bindings, problem
18:00:43  <cjihrig>tjfontaine: you can throw ?w=0 on the end of the PR url, but it kinda messes with commenting last time i checked
18:00:52  * Ralithjoined
18:01:10  <tjfontaine>cjihrig: I did -- it's just bothersome
18:01:26  <tjfontaine>I have to becareful of my bloodpressure
18:01:36  <tjfontaine>getting angry at github will make me lose my day
18:02:12  * glycolpart
18:02:30  <saghul>DarkUranium: great!
18:02:56  <jgi>tjfontaine, cjihrig: the upgrade to OpenSSL 1.0.1k is cherry-picked from io.js and has been done by indutny
18:03:30  <cjihrig>jgi: LGTM
18:03:31  <tjfontaine>jgi: which commit yours or the other that removes the .s
18:03:45  <jgi>tjfontaine, cjihrig: so it’s more about reviewing the way all the commits are split and the upgrade from openssl 1.0.1k to 1.0.1l
18:04:37  <tjfontaine>jgi: only thing it seems like mips.s was removed
18:04:52  <tjfontaine>and deps/openssl/openssl/crypto/bn/asm/x86_64-win32-masm.asm
18:06:19  * jasnelljoined
18:06:32  <srl295>jgi: jasnell: hi
18:06:41  <jasnell>hey there
18:06:44  <jgi>jasnell: hi!
18:07:07  <tjfontaine>hey!
18:09:36  * seldojoined
18:12:10  <jgi>cjihrig: thanks for the review!
18:13:56  <tjfontaine>there's a script to generate the asm stuff, and it would be nice to make sure there are no new changes
18:13:59  * Fishrock123joined
18:14:01  <jgi>cjihrig, tjfontaine: I’ll look into why mips.s was removed, the x86_64-win32-masm.asm has been removed because of https://github.com/iojs/io.js/pull/289#issuecomment-69511165, if i understand correctly
18:14:39  <tjfontaine>ok
18:21:16  * creationixpart
18:21:56  <tjfontaine>jgi: presuming you were able to type 'make' in the 'asm' directory I'm groovy with it
18:22:43  <jgi>tjfontaine: alright, will do and let you know
18:22:50  <jgi>tjfontaine: thank you for the review!
18:23:13  <tjfontaine>my pleasure
18:24:44  * creationixjoined
18:26:15  <jasnell>tjfontaine: just fyi... I'm currently down at the peets coffee in #2 emb.center. I'll be heading up to the office shortly
18:26:31  <jasnell>should be in about a half hour
18:26:58  <tjfontaine>jasnell: ok, we're doing an all hands at the moment, should be done around 11 as well
18:28:20  * AvianFlujoined
18:36:31  * a_lequit (Remote host closed the connection)
18:45:32  * thlorenzjoined
18:50:05  * jgiquit (Quit: jgi)
18:50:47  * jgijoined
18:51:01  * kevinswiberquit (Remote host closed the connection)
18:54:32  <jasnell>tjfontaine: finishing a build now really quick (hopefully quick) then will head up
18:54:35  * kevinswiberjoined
18:55:59  * jasnellquit (Remote host closed the connection)
19:01:58  * kevinswiberquit (Remote host closed the connection)
19:02:35  * kevinswiberjoined
19:06:27  * rkowalski___quit (Quit: Page closed)
19:06:47  * kevinswiberquit (Ping timeout: 252 seconds)
19:10:10  <jgi>tjfontaine: cd deps/openssl/asm && make && git status -s succeeds and doesn’t show any new file
19:10:22  * seldoquit (Remote host closed the connection)
19:18:54  * bajtosquit (Quit: bajtos)
19:22:27  * reqsharkjoined
19:23:40  * seldojoined
19:40:25  * abraxas_joined
19:40:55  <MI6>joyent/node: Fedor Indutny v0.10 * 9e387fb : test: fix crypto-stream after openssl update (+3 more commits) - http://git.io/iC4n9g
19:43:19  * stagasjoined
19:44:34  * wolfeidaujoined
19:45:22  * abraxas_quit (Ping timeout: 244 seconds)
19:54:52  * wolfeidauquit (Remote host closed the connection)
19:55:41  * brsonquit (Quit: Lost terminal)
19:55:49  * brsonjoined
19:55:53  * a_lejoined
19:57:17  * a_lequit (Read error: Connection reset by peer)
19:57:24  * a_le_joined
20:00:05  * brsonquit (Client Quit)
20:00:22  * brsonjoined
20:09:18  * a_le_quit (Remote host closed the connection)
20:09:52  * a_lejoined
20:10:06  * jgiquit (Quit: jgi)
20:12:17  * jgijoined
20:12:40  * wolfeidaujoined
20:16:40  * wolfeidauquit (Remote host closed the connection)
20:31:50  * kevinswiberjoined
20:41:25  * SergeiRNDjoined
20:48:47  <jgi>chrisdickinson: ping
20:50:15  * wolfeidaujoined
20:50:28  * SergeiRNDquit (Quit: Leaving.)
20:52:26  * wolfeidauquit (Remote host closed the connection)
20:54:44  * wolfeidaujoined
20:55:45  <chrisdickinson>jgi: pong -- 9070 is "working as expected"
20:56:34  <DarkUranium>hey, saghul
20:56:37  * wolfeidauquit (Remote host closed the connection)
21:01:35  * jgiquit (Quit: jgi)
21:14:09  * [spoiler]quit (Quit: I'm running to save my life!)
21:16:02  * thlorenzquit
21:17:42  <piscisaureus>jgi: tjfontaine: yes make-ing the asm turned out to make no difference, the output didn't change
21:22:36  * thlorenzjoined
21:25:30  * brsonquit (Quit: leaving)
21:26:48  * kevinswiberquit (Read error: Connection reset by peer)
21:29:18  * abraxas_joined
21:33:59  * abraxas_quit (Ping timeout: 245 seconds)
21:46:02  * jgijoined
21:46:13  * seldoquit
21:46:57  * toastynerdquit (Ping timeout: 245 seconds)
21:47:13  <jgi>chrisdickinson: thanks and sorry for not being connected, I went out for lunch
21:53:15  <tjfontaine>we'll just have to get you a shell somewhere ;)
21:53:24  <tjfontaine>screen+irssi is the only way to irc ;)
21:54:59  <chrisdickinson>i'm on a textual + znc setup
21:55:17  * kevinswiberjoined
21:55:18  <chrisdickinson>and i heartily halfway recommend it, sort of, kind of
21:58:24  <DarkUranium>saghul, found one problem w/ MinGW compiles in libuv: https://github.com/libuv/libuv/issues/153
22:03:14  <trevnorris>afternoon
22:03:21  <DarkUranium>hi
22:03:23  <trevnorris>jgi: get reviewed?
22:04:17  <jgi>trevnorris: yep, already done, thanks!
22:04:28  <jgi>trevnorris: and hi!
22:04:53  <trevnorris>hey :)
22:05:22  <trevnorris>srl295: great. thank you for all the hard work getting that working :)
22:05:34  <srl295>trevnorris: welcome
22:06:13  * dshaw_joined
22:06:39  <trevnorris>jgi: what's the v0.11.15 regression?
22:06:49  <tjfontaine>we broke the npms
22:07:26  <jgi>trevnorris: basically, dependencies using git+ssh seem to be broken, I need to find some time to create an issue with a good description for that
22:08:50  <trevnorris>ah. that sucks.
22:08:56  <trevnorris>tjfontaine: how's it been? :)
22:09:15  <tjfontaine>busy as per usual
22:09:19  <tjfontaine>how are things for you?
22:11:21  * toastynerdjoined
22:11:58  * kevinswiberquit (Remote host closed the connection)
22:12:41  * kevinswiberjoined
22:12:44  <piscisaureus>tjfontaine: jgi: tip: check out https://github.com/iojs/io.js/commit/913addbff5481567262c387cef9594f809e4ef83
22:12:44  <piscisaureus>and
22:12:44  <piscisaureus>https://github.com/iojs/io.js/commit/a1e54d6fb72ef0e0d3aab9ed2bf88d60a4a419d9
22:12:46  * Fishrock123quit (Remote host closed the connection)
22:13:06  * seishunquit (Ping timeout: 276 seconds)
22:13:26  <trevnorris>well. went to google today and spoke w/ the v8 team lead.
22:16:57  <jgi>piscisaureus: thanks!
22:17:36  * wolfeidaujoined
22:17:44  <tjfontaine>did the path path then reland after the second?
22:24:29  * brsonjoined
22:26:14  <piscisaureus>A fixed version of it
22:26:15  <piscisaureus>https://github.com/iojs/io.js/pull/299
22:27:05  <tjfontaine>honestly the whole state of the api is kinda bunk -- the intent is that format and parse round trip -- but quietly doing the wrong thing will always lead to these problems
22:27:17  <tjfontaine>I can't change it such that if you supply both path and pathname to throw
22:27:31  <tjfontaine>but we can at least generate the combinations and say -- these two thigns don't match -- someone fucked up
22:27:36  <piscisaureus>Too many cooks spoil the broth as they say
22:27:38  <tjfontaine>then we can actually give useful information to the programmer
22:32:39  <piscisaureus>tjfontaine: it's mostly slow and devastating feature creep
22:32:53  <piscisaureus>tjfontaine: I remember the first "overlapping" field was added for browser compat
22:33:04  <piscisaureus>what's up with all the other parts now (query and path) I don't know
22:36:09  * rendarquit (Quit: Leaving)
22:41:14  <jgi>tjfontaine: I’m not sure I understand when you say “we can at least generate the combinations”
22:45:36  * toastynerdquit (Ping timeout: 276 seconds)
22:46:09  * kevinswi_joined
22:46:15  * kevinswiberquit (Read error: Connection reset by peer)
22:54:04  * reqshark_joined
22:57:11  * lanceballchanged nick to lance|afk
22:57:59  * reqsharkquit (Ping timeout: 256 seconds)
22:59:15  * reqshark_quit (Ping timeout: 276 seconds)
22:59:29  * toastynerdjoined
23:00:47  * kevinswi_quit (Remote host closed the connection)
23:01:20  <tjfontaine>jgi: we can generate the result as if .path had not be set, and we can generate it as if it had, if they do not match -- throw
23:01:23  * kevinswiberjoined
23:02:03  <tjfontaine>jgi: even with having preferences for the members -- it makes it really hard to know what the programmer meant to do
23:05:17  * kevinswiberquit (Read error: Connection reset by peer)
23:05:27  * kevinswiberjoined
23:13:09  <jgi>tjfontaine: according to this diagram: https://github.com/joyent/node/issues/9070#issuecomment-70920180, and if I understand correctly what you’re saying (which I’m afraid I’m not), that would mean that it would throw every time “search” or “query” is set right?
23:13:48  <tjfontaine>well I think query is search without the ? maybe
23:13:53  * Fishrock123joined
23:13:53  <tjfontaine>anyway it's not if they're both set
23:14:06  <tjfontaine>but if they are set, and if we format both cases it would result in different output
23:14:31  <tjfontaine>consider { host: "foobar.com:8080", hostname: "foobar", port: 8080 }
23:14:34  <tjfontaine>no throw
23:14:42  <tjfontaine>but if port is 8081 -- throw
23:15:12  <jgi>yep
23:15:38  <tjfontaine>because we actually don't know what to do with that input (aside from having have variable precedence -- which is frankly too magical imesho)
23:16:05  <jgi>tjfontaine: yes, I see what you mean now
23:16:09  * dshaw_quit (Quit: Leaving.)
23:16:44  <jgi>tjfontaine: thanks for taking the time to explain it :)
23:16:53  <tjfontaine>my pleasure -- sorry for being cryptic :D
23:17:03  <tjfontaine>I'm mostly just trolled by the state
23:17:14  <tjfontaine>and that I didn't catch it earlier
23:17:18  <tjfontaine>I need to be better.
23:18:04  * abraxas_joined
23:18:27  * Fishrock123quit (Ping timeout: 272 seconds)
23:20:39  * chris_99quit (Quit: Ex-Chat)
23:21:03  <jgi>tjfontaine: so just to confirm that I understood properly what you meant, you’re saying that “url.format(parsed, true)” here: https://github.com/joyent/node/issues/9070#issuecomment-70918084 should throw right?
23:22:53  * abraxas_quit (Ping timeout: 245 seconds)
23:23:20  <tjfontaine>jgi: yes *I* think so
23:24:00  * dshaw_joined
23:26:46  <jgi>tjfontaine: ok, I still don’t know how I feel about that
23:26:59  <tjfontaine>it's controversial I know
23:27:30  <jgi>tjfontaine: I also still don’t know how I feel about the current state :)
23:27:37  <tjfontaine>:)
23:34:31  * a_lequit (Remote host closed the connection)
23:39:52  * bradleymeckjoined
23:44:30  * a_lejoined
23:45:01  * a_lequit (Read error: Connection reset by peer)
23:45:55  * a_lejoined
23:48:34  * wolfeidauquit (Read error: Connection reset by peer)
23:48:35  * wolfeidau_joined
23:50:07  * toastynerdquit (Ping timeout: 272 seconds)
23:53:50  * kevinswiberquit (Remote host closed the connection)
23:55:48  * kevinswiberjoined
23:57:06  * a_lequit (Remote host closed the connection)
23:58:11  * toastynerdjoined
23:58:45  * kevinswiberquit (Remote host closed the connection)
23:59:18  * kevinswiberjoined