00:00:58  <piscisaureus_>bnoordhuis: the uv_dlerror patch looks fine to me.
00:01:01  <mjr_>Oh yeah, here's an article about it: http://www.amigareport.com/ar137/p1-12.html
00:01:10  <piscisaureus_>bnoordhuis: that is, almost, I added a comment
00:01:25  <bnoordhuis>mjr_: you are now my personal hero
00:01:59  <mjr_>When I was 12, there was no internet.
00:02:24  <bnoordhuis>yeah, same for me - only at uni
00:02:35  <mmalecki>I got internet when I was 6!
00:03:08  <mmalecki>started coding when I was 9 or so, first HTML page. I done some silly ShowMessage in Delphi before, but that doesn't really count
00:04:08  <mmalecki>also, I'm going to sleep. night everyone.
00:04:14  <bnoordhuis>sleep tight, maciej
00:04:29  <mmalecki>you too, Ben!
00:04:36  * mmaleckichanged nick to mmalecki[zzz]
00:10:05  <isaacs>bnoordhuis, piscisaureus_, igorzi: anything else outstanding for v0.6.12?
00:10:15  <isaacs>i'm going to do the release tomorrow morning, now that the npm stuff is finished.
00:10:23  <isaacs>and pull in that ustack thing dap did.
00:10:35  <bnoordhuis>isaacs: -fno-strict-aliasing
00:12:00  <isaacs>bnoordhuis: i remember that, but got a link?
00:12:04  <isaacs>bnoordhuis: can't recall the specifics
00:12:15  <bnoordhuis>wait, let me push it
00:12:39  <bnoordhuis>isaacs: https://github.com/bnoordhuis/node/compare/strict-aliasing
00:12:48  <bnoordhuis>that fixes the v8 hangup on smartos
00:13:17  <isaacs>right
00:13:29  <isaacs>wait, that's right, piscisaureus_ sent me something about that
00:13:46  <isaacs>but diesn't 'cflags': [ '-fomit-frame-pointer' ] do something awful?
00:13:50  <isaacs>dap: ^
00:14:02  <isaacs>like, disable all the jsstack stuff?
00:14:10  <piscisaureus_>bnoordhuis: I think isaacs was working on -fno-strict-aliasing.
00:14:29  <piscisaureus_>bnoordhuis: it has to be used on all gcc 4.5 except gcc-llvm
00:14:32  <bnoordhuis>isaacs: -fomit-frame-pointer is still disabled on sunos
00:14:39  <isaacs>yeah, -fno-strict-aliasing fixes it as well.
00:14:52  <isaacs>ohhh, i get it
00:14:54  <isaacs>sorry, missed a !=
00:14:58  <bnoordhuis>heh
00:15:04  <isaacs>so if it's not SunOS you get -fno-strict-aliasing
00:15:10  <isaacs>er, -fomit-frame-pointer
00:15:16  <bnoordhuis>yes
00:15:20  <isaacs>but if it is SunOS you get '-O3', '-fdata-sections', '-ffunction-sections', '-fno-strict-aliasing'
00:15:24  <bnoordhuis>and -fno-strict-aliasing on all platforms
00:15:33  <bnoordhuis>which probably sucks from a performance perspective
00:15:39  <isaacs>sure, whatever.
00:15:59  <bnoordhuis>but... there's a lot of aliasing going on in node and libuv
00:16:03  <piscisaureus_>bnoordhuis: http://codereview.chromium.org/6539006 <-- this has to be ported to gyp. Isaacs had a patch partially done.
00:16:06  <bnoordhuis>so it has the potential to break at any time
00:16:15  <isaacs>ok
00:16:57  <bnoordhuis>piscisaureus_: is it a bug in v8 or gcc?
00:17:05  <piscisaureus_>bnoordhuis: GCC bug
00:17:12  <isaacs>i've gotta run for a few hours. i just made a note of this, i'll check it out and read the logs when i get back.
00:17:14  <bnoordhuis>do you have a link to the bug report for me?
00:18:01  <bnoordhuis>ah, found it
00:18:06  <bnoordhuis>http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45883
00:18:13  <bnoordhuis>okay, so it's a compiler bug
00:18:38  <bnoordhuis>so -fstrict-aliasing -fno-builtin-memcpy should work...
00:18:40  * bnoordhuistests
00:19:59  <piscisaureus_>bnoordhuis: http://code.google.com/p/v8/issues/detail?id=884
00:20:20  <piscisaureus_>bnoordhuis: no-builtin-memcpy makes it terribly slow and is not needed
00:21:14  <bnoordhuis>oh, i'm just curious if -fno-builtin-memcpy fixes it
00:21:23  <piscisaureus_>bnoordhuis: it does
00:21:32  <piscisaureus_>bnoordhuis: at least, it did on cygwin :-)
00:21:37  <piscisaureus_>a year ago
00:21:45  <bnoordhuis>right, now let's test sunos
00:21:57  <bnoordhuis>and lo and behold, it does
00:23:01  <bnoordhuis>fixed in 4.6.0, hmm? okay, i'll add (steal!) a gcc version check
00:24:23  <piscisaureus_>bnoordhuis: ok. Be careful to not detect gcc-llvm-4.2 as affected
00:24:37  <piscisaureus_>bnoordhuis: that's what you have on your mac and it is not affected
00:24:56  <bnoordhuis>"on your mac"... you offend me, sir!
00:26:10  <AvianFlu>bnoordhuis, ++
00:26:10  <kohai>bnoordhuis has 10 beers
00:26:19  <AvianFlu>damn, no kohai >.<
00:27:02  <piscisaureus_>*(&bnoordhuis)--
00:29:35  <piscisaureus_>mraleph: what is the best moment to bug Erik?
00:30:02  <mraleph>piscisaureus_: well somewhere during the day probably
00:30:14  <mraleph>piscisaureus_: just drop him a mail
00:31:22  <piscisaureus_>mraleph: right. It's unfortunate he lives in europe
00:31:49  <mraleph>said guy from Netherlands to guy who lives in Denmark
00:32:27  <mraleph>u should be sleeping
00:32:45  <piscisaureus_>It would be best if all morning people lived in the US and night people in europe
00:32:53  <piscisaureus_>mraleph: you should too.
00:33:34  <mraleph>I am. This is all my dream
00:36:00  <piscisaureus_>right. Yes I am going to sleep too.
00:36:08  <piscisaureus_>bnoordhuis: what time are you in a'dam tomorrow?
00:36:58  <bnoordhuis>piscisaureus_: depends on the guy with the shoveling machine
00:37:25  <bnoordhuis>there's this guy with a very loud shoveling machine that's been waking me up at 07:30 AM the last two days
00:37:37  <piscisaureus_>bnoordhuis: so you will be here at 8.30 most likely?
00:37:58  <bnoordhuis>maybe a little later
00:38:08  <bnoordhuis>i'll try to be there around noon, give or take an hour
00:39:12  <piscisaureus_>right. night all.
00:39:16  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
00:44:19  * Ariajoined
00:48:19  * mralephquit (Quit: Leaving.)
01:06:39  * sh1mmerquit (Quit: sh1mmer)
01:26:41  * dapquit (Quit: Leaving.)
01:30:05  * dapjoined
01:44:42  <bnoordhuis>isaacs: https://github.com/bnoordhuis/node/compare/strict-aliasing <- you may want to double-check it
01:47:53  <pfox___>so what are you supposed to do, in the body of a uv_connect_cb, that is used in a call to uv_tcp_connect() ?
01:48:15  <pfox___>what operation can you do with the now initialized handle... ?
01:51:23  <TooTallNate>bnoordhuis: thank you for remembering the default value <3
01:51:39  <TooTallNate>was about to send a patch for gcc_optimize_level
01:51:46  <TooTallNate>but now it won't be needed :)
01:53:17  <bnoordhuis>pfox___: start reading and writing
01:53:41  <bnoordhuis>TooTallNate: that patch should make joyent happy
01:54:08  <bnoordhuis>which is good because i still haven't returned my laptop
01:54:21  <TooTallNate>haha
01:56:04  * abraxasjoined
01:57:35  <bnoordhuis>fancy that, php 5.4 adds traits
01:58:04  <bnoordhuis>what's also nice is that the built-in http server is powered by our very own http-parser
01:58:19  <abraxas>nice, but no cigar :)
02:01:29  <abraxas>i've used php for many many years, but all the OOP features feel like a big pile of magic on top of a big pile of don't-do-OOP
02:02:21  <abraxas>gotta love prototypal inheritence
02:05:21  <bnoordhuis>oh, oop
02:05:34  <bnoordhuis>i mostly write c and i've never missed oop
02:06:01  <bnoordhuis>in the end it's all functions anyway
02:06:03  <abraxas>:) i guess you don't mind passing pointers around and around and around?
02:06:40  <bnoordhuis>not at all, i like pointers
02:11:14  <bnoordhuis>pfox___: so what are you working on?
02:25:18  * pfox___quit (Ping timeout: 248 seconds)
02:30:03  * mikealquit (Quit: Leaving.)
02:34:19  * dapquit (Quit: Leaving.)
02:51:06  * perezdjoined
02:51:58  * bnoordhuisquit (Ping timeout: 248 seconds)
02:53:38  * brsonquit (Quit: leaving)
03:01:29  * philipsquit (Excess Flood)
03:02:38  * philipsjoined
03:04:09  * txdv_quit (Quit: No Ping reply in 180 seconds.)
03:04:24  * txdvjoined
03:06:40  * dshaw_quit (Read error: Operation timed out)
03:17:03  * TooTallNatequit (Quit: Linkinus - http://linkinus.com)
03:22:36  * mikealjoined
04:05:50  * pfox___joined
04:35:11  * Ariaquit (Remote host closed the connection)
04:53:10  * sh1mmerjoined
05:20:36  * elijah-mbpjoined
05:55:38  * felixgejoined
05:55:38  * felixgequit (Changing host)
05:55:38  * felixgejoined
06:01:48  * felixgequit (Quit: felixge)
06:22:12  * perezdquit (Quit: perezd)
06:25:43  * felixgejoined
06:25:43  * felixgequit (Changing host)
06:25:43  * felixgejoined
06:34:20  * sh1mmerquit (Quit: sh1mmer)
06:44:33  * pfox___quit (Ping timeout: 255 seconds)
07:19:43  * mikealquit (Quit: Leaving.)
07:32:07  * mikealjoined
07:39:31  * isaacsquit (Remote host closed the connection)
07:48:56  * rendarjoined
07:51:28  * stephankquit (Quit: *Poof!*)
08:13:11  * mralephjoined
08:21:03  * mmalecki[zzz]changed nick to mmalecki
08:23:02  * abraxasquit
08:42:13  * mralephquit (Quit: Leaving.)
08:44:31  * txdv_joined
08:48:26  * txdvquit (Ping timeout: 252 seconds)
08:48:44  * mikealquit (Ping timeout: 260 seconds)
08:49:45  * xaqjoined
08:50:34  * mmaleckichanged nick to mmalecki[brb]
09:09:32  * mikealjoined
09:25:42  * xaqquit (Remote host closed the connection)
09:25:51  * mikealquit (Quit: Leaving.)
09:29:36  * mikealjoined
10:12:28  * mmalecki[brb]changed nick to mmalecki
10:51:22  * mmaleckichanged nick to mmalecki[away]
11:29:55  * piscisaureus_joined
12:57:06  * paddybyersjoined
14:42:37  * bnoordhuisjoined
14:46:17  * mmalecki[away]changed nick to mmalecki
14:51:44  <bnoordhuis>piscisaureus_: https://github.com/shigeki/libuv/commit/e082861e03d0190755906f530c71556332b09817#commitcomment-1035705
14:56:28  * pfox___joined
15:03:46  <piscisaureus_>ik ruik brokjes
15:06:58  <bnoordhuis>je moet je vaker wassen, bertje
15:07:23  <piscisaureus_>ik zal jou eens de oren wassen bennetje
15:07:46  <piscisaureus_>misschien eerst de oren waxen, man wat een oerwoud komt daar uit
15:08:07  <bnoordhuis>zo'n klein ventje, zo'n grote mond
15:09:00  <piscisaureus_>bnoordhuis: nou ja in fysiek opzicht ben je niet echt klein hoor. Maar als je vooral doelde op je persoonlijkheid, ja idd
15:27:33  <CIA-99>node: Ben Noordhuis master * r0613af0 / configure :
15:27:33  <CIA-99>node: build: remove --shared-cares configure switch
15:27:33  <CIA-99>node: c-ares is a) bundled with libuv now, and b) contains out-of-tree patches. It no
15:27:33  <CIA-99>node: longer makes sense to link against a shared library.
15:27:33  <CIA-99>node: Fixes #2117. - http://git.io/nAxEgA
15:32:40  <piscisaureus_>bnoordhuis: floating patches on cares?
15:42:11  <CIA-99>libuv: Shigeki Ohtsu master * rb55801f / (8 files in 6 dirs): win, unix: add uv_dlerror() and uv_dlerror_free() - http://git.io/Zl2vdw
15:42:44  * travis-cijoined
15:42:44  <travis-ci>[travis-ci] joyent/node#533 (master - 0613af0 : Ben Noordhuis): The build is still failing.
15:42:44  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/5e80321...0613af0
15:42:44  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/780212
15:42:44  * travis-cipart
15:43:48  <bnoordhuis>piscisaureus_: windows rogue dns server
15:44:10  * travis-cijoined
15:44:10  <travis-ci>[travis-ci] joyent/libuv#114 (master - b55801f : Shigeki Ohtsu): The build is still failing.
15:44:10  <travis-ci>[travis-ci] Change view : https://github.com/joyent/libuv/compare/702f905...b55801f
15:44:10  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/libuv/builds/780307
15:44:10  * travis-cipart
15:44:18  <tjfontaine>"rogue"?
15:44:46  <piscisaureus_>ruig
15:45:31  <bnoordhuis>tjfontaine: https://github.com/joyent/libuv/commit/cb474b2
15:45:33  <pfox___>bnoordhuis: hey, about the tcp stuff
15:45:50  <pfox___>im starting to work on mapping the uv_tcp bidness, starting w/ requests
15:45:52  <tjfontaine>bnoordhuis: oh evil
15:46:37  <pfox___>so, continuing my newb q: once you're in the uv_connect_cb .. you can assume that the uv_connect_t's handle field is live and can be written to?
15:46:49  <bnoordhuis>pfox___: yes
15:46:56  <bnoordhuis>provided status==0 of course
15:47:00  <pfox___>granted
15:47:03  <pfox___>also:
15:47:09  <pfox___>libertarian ultraviolence
15:47:22  <bnoordhuis>i like it
15:47:42  * bnoordhuistopic: libertarian ultraviolence
15:49:03  <piscisaureus_>lib ultimat vodka
15:49:42  <pfox___>lithe uniformed vixen
15:54:18  <creationix>bnoordhuis, I fear what the title will become if I give you ops in #luvmonkey ;)
16:34:37  <piscisaureus_>bnoordhuis: http://msdn.microsoft.com/en-us/library/by56e477%28v=vs.80%29.aspx
16:37:42  * isaacsjoined
16:52:27  * philipsquit (Excess Flood)
16:52:42  * philips_joined
16:52:59  * philips_changed nick to philips
16:53:06  * philipsquit (Changing host)
16:53:06  * philipsjoined
16:55:00  * mralephjoined
17:02:32  * xaqjoined
17:03:32  * xaqquit (Read error: Connection reset by peer)
17:03:49  * xaqjoined
17:06:30  * bnoordhuisquit (Remote host closed the connection)
17:07:19  * xaqquit (Remote host closed the connection)
17:08:32  * piscisaureus_quit (Ping timeout: 240 seconds)
17:14:14  * indutny_7_dayschanged nick to indutny
17:14:18  <indutny>yo
17:16:59  <isaacs>ello :)
17:17:21  <indutny>isaacs: how're ya?
17:17:34  <isaacs>good, good
17:17:38  <isaacs>just cutting 0.6.12 now
17:17:43  <indutny>cool!
17:17:44  <isaacs>er, merging in the last few things
17:17:53  <isaacs>how's things in omsk?
17:18:15  * dapjoined
17:18:26  <indutny>well, all is fine, thanks
17:27:08  * stephankjoined
17:31:17  * paddybyersquit (Quit: paddybyers)
17:33:12  <CIA-99>node: isaacs v0.6 * r0541271 / (220 files in 12 dirs): (log message trimmed)
17:33:13  <CIA-99>node: Upgrade npm to 1.1.3
17:33:13  <CIA-99>node: * Update request to support HTTPS-over-HTTP proxy tunneling
17:33:13  <CIA-99>node: * Throw on undefined envs in config settings
17:33:13  <CIA-99>node: * Update which to 1.0.5
17:33:13  <CIA-99>node: * Fix windows UNC busyloop in findPrefix
17:33:13  <CIA-99>node: * Bundle nested bundleDependencies properly
17:33:14  <CIA-99>node: Dave Pacheco v0.6 * redea94c / (src/v8constants.h src/v8ustack.d):
17:33:14  <CIA-99>node: dtrace ustack helper improvements
17:33:15  <CIA-99>node: Fixes #2852 - http://git.io/FVUXpg
17:33:16  <CIA-99>node: isaacs v0.6 * rd8c5ba2 / (test/fixtures/a1.js test/fixtures/a1.js):
17:33:16  <CIA-99>node: Change test fixture from symlink to regular file
17:33:16  <CIA-99>node: The only test using this is test/simple/test-fs-chmod.js, and it was
17:33:17  <CIA-99>node: treating a.js and a1.js as two separate files, resulting in a race
17:33:32  <isaacs>dap: thanks! ^
17:33:45  <indutny>yeah
17:39:21  <dap>np. thanks for landing it.
17:41:51  * travis-cijoined
17:41:51  <travis-ci>[travis-ci] joyent/node#534 (v0.6 - 0541271 : isaacs): The build passed.
17:41:51  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/3aa2fd3...0541271
17:41:51  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/781356
17:41:51  * travis-cipart
17:49:06  <CIA-99>node: isaacs v0.6 * r1bf3a14 / (100 files in 7 dirs): Upgrade npm to 1.1.4 (minor bugfix) - http://git.io/O4ONUg
17:52:36  <isaacs>so, anything else for 0.6.12?
17:53:15  <isaacs>ircretary: tell bnoordhuis That -fno-strict-aliasing patch is for master. did something similar need to go in v0.6?
17:53:15  <ircretary>isaacs: I'll be sure to tell bnoordhuis
17:58:29  * travis-cijoined
17:58:29  <travis-ci>[travis-ci] joyent/node#535 (v0.6 - 1bf3a14 : isaacs): The build was broken.
17:58:29  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/0541271...1bf3a14
17:58:29  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/781453
17:58:29  * travis-cipart
18:01:30  * mralephquit (Quit: Leaving.)
18:04:52  * `3rdEdenjoined
18:08:28  * mralephjoined
18:08:48  * xaqjoined
18:24:24  * TooTallNatejoined
18:27:39  * xaqquit (Remote host closed the connection)
18:29:16  * dshaw_joined
18:31:23  * xaqjoined
18:31:24  * xaqquit (Remote host closed the connection)
18:31:35  <isaacs>review, please: https://github.com/isaacs/node/commit/35b36394f16c4e5eb08fa4fb21d797ac11439dc9
18:31:58  <isaacs>test, please: http://nodejs.org/dist/v0.6.12/node-v0.6.12-RC0.tar.gz
18:36:47  <isaacs>ack, tabs in the changelog. fixed: https://github.com/isaacs/node/commit/v0.6.12-release
18:37:29  <mmalecki>.12 is giving me bad dreams
18:38:03  <mmalecki>isaacs: lgtm tho
18:38:18  <isaacs>bad dreams?
18:38:20  <TooTallNate>the fat binary is win
18:38:51  <tjfontaine>TooTallNate: is your gyp thing going to do similar? :)
18:38:57  <mmalecki>isaacs: 0.4.12 was a nightmare ;)
18:39:22  <TooTallNate>tjfontaine: as soon as gyp supports it
18:39:53  <isaacs>oh, heh
18:39:53  <TooTallNate>tjfontaine: but in the master branch we have a similar double compilation + lipo setup
18:40:04  <TooTallNate>what was wrong with 0.4.12?
18:40:09  <tjfontaine>TooTallNate: ok
18:40:32  <mmalecki>TooTallNate: porting all the stuff from 0.4.12 to 0.6 :)
18:41:08  <isaacs>mmalecki: would you be willing/able test that the code in the tarball works on travis?
18:42:18  <isaacs>TooTallNate: yeah, i'm building the fatso now
18:42:56  <mmalecki>isaacs: well, technically, yes, but not sure if that makes any sense - not many projects would run on RC version and we can't really rebuild all of them. I can run tests locally tho
18:43:36  <isaacs>mmalecki: sure, i mean, just make sure that it doesn't blow up horribly in a similarly travis-ish setup
18:43:46  <mmalecki>isaacs: sure, I can do that :)
18:44:01  <isaacs>the if there's no problems anywhere, i just strip the RC from the filename
18:46:33  <benvie>it is a candidate for release after all!
18:46:49  <benvie>it just needs to win the election
18:49:20  <isaacs>benvie: exactly!
18:51:05  <isaacs>TooTallNate: baller: https://gist.github.com/1960344
18:51:16  * mikealquit (Quit: Leaving.)
18:51:31  <isaacs>fat magic: cafebabe :)
18:51:40  <TooTallNate>hahaha
18:51:46  <TooTallNate>that's classic
18:51:49  <TooTallNate>did you put that?
18:51:55  <benvie>rofl haha
18:52:08  <benvie>that is awesome
18:52:30  <benvie>lipo
18:53:07  <benvie>that whole thing is offensive to fatchicks
18:53:42  <tjfontaine>actually it's also used to remove unwanted archs, so it's like "lipo suction"
18:53:45  * `3rdEdenquit (Quit: rebootin)
18:54:17  <benvie>wow
18:54:44  <isaacs>TooTallNate: no, it generates it
18:55:12  <isaacs>TooTallNate: it's just a random string of hex to separate the different bins it contains
18:55:26  <benvie>well that indicates that lipo is the thing that injects fat? cause that's not accurate
18:55:32  <TooTallNate>isaacs: so that's just a coincidence?
18:55:42  <isaacs>woohoo!! [01:14|% 100|+ 337|- 1]: Done
18:55:52  <isaacs>TooTallNate: nono, i'm sure it's got a selection of cutesy things to have there.
18:56:01  <tjfontaine>benvie: the fact that lipo works in reverse is just coincidence, generally you use the compiler to achieve multiarch
18:56:06  <isaacs>TooTallNate: since, if "0xCAFEBABE" was in either bin, it'd fail
18:56:21  <benvie>haha
18:56:41  <TooTallNate>oh i see, haha
18:57:29  <benvie>now I wonder what other words it has made out of a-e
18:57:37  <benvie>on the topic of fat stuff
18:57:44  <benvie>a-f
18:58:03  * brsonjoined
18:58:25  <benvie>feedboob
18:59:23  * `3rdEdenjoined
19:01:06  * mikealjoined
19:14:00  <isaacs>igorzi: hey
19:14:30  <isaacs>release going frighteningly smoothly. wanna check windows? i've got 11 failures, but down from 15 in 0.6.11, it looks like.
19:18:38  * xaqjoined
19:22:31  * xaqquit (Remote host closed the connection)
19:38:35  <isaacs>test, please: test, please: http://nodejs.org/dist/v0.6.12/node-v0.6.12-RC1.tar.gz
19:47:56  <benvie>I'm off to the see the wizard, the wonderful wizard of node-0.6.12-releasecandidate1
19:49:09  <AvianFlu>benvie, ++
19:49:09  <kohai>benvie has 1 beer
19:49:24  <mmalecki>node hyphen oh dot six dot twelve hyphen release candidate one
19:49:34  <benvie>exactly how you say it
19:50:12  <mmalecki>it should be written like that!
19:50:25  <benvie> I'm off to the see the wizard, the wonderful wizard of node hyphen oh dot six dot twelve hyphen release candidate one
19:50:29  <AvianFlu>don't forget the "point javascript"
19:50:53  <benvie> I'm off to the see the wizard, the wonderful wizard of node hyphen oh dot six dot twelve hyphen release candidate one of house javascript
19:58:37  * paddybyersjoined
19:58:57  <creationix>wait, you say "hyphen", but skip the "v"?
19:59:12  <mmalecki>OH DAMN
19:59:38  <creationix>"node jay ess version oh six twelve release candidate one" is how I would say it
20:00:48  <creationix>but probably skip the "jay ess"
20:01:11  <mmalecki>s/jay ess/javascript/ and it's fine
20:01:36  <isaacs>mmalecki: s/dot/point/
20:01:56  <mmalecki>isaacs: ++
20:01:56  <kohai>isaacs has 8 beers
20:01:56  <creationix>right, I never bought into the "point javascript" thing
20:02:08  <isaacs>creationix: i don't htink you're really supposed to
20:02:10  <isaacs>it's kind of a joke
20:02:22  <benvie>the only use for it is that node is a dictionary word
20:02:31  <benvie>so searching for it helps with .js
20:02:39  <creationix>better to just leave it off entirely and then look at people like they are idiots if they don't know what just "node" means :P
20:03:04  <creationix>hence why I google search before naming projects usually
20:03:08  <benvie>searching google for node only works because it personalizes my results
20:03:10  <creationix>luvit and luvmonkey were pretty unique
20:03:16  <benvie>yeah hah
20:03:23  <benvie>agree on that tactic
20:03:43  <creationix>though bnoordius still wants to s/luvmonkey/monkeyluv/
20:04:04  <benvie>you run into postfacto pornos when you stray too far
20:04:22  <benvie>maybe nerd pornos by authors obsessed with lua?
20:04:52  <isaacs>creationix: why the "monkey" bit?
20:04:57  <creationix>luvmonkey isn't lua
20:04:59  <creationix>it's spidermonkey
20:05:03  <isaacs>ohh, ok
20:05:03  <benvie>oh sweet
20:05:06  <benvie>now I'm interested
20:05:13  <creationix>it's early alpha
20:05:18  <isaacs>what's it do?
20:05:19  <creationix>but the initial ram overhead is half of v8
20:05:30  <benvie>is it restart from scratch or are you using spidermonkey's work?
20:05:31  <creationix>I need to implement more to have an idea of request per second
20:05:41  <benvie>er spidernode
20:05:44  <creationix>spidernode is using v8 APis
20:05:48  <creationix>I'm using vanilla sm APis
20:05:52  <creationix>so fresh project
20:05:56  <benvie>yeah I thought they went down the wrong road with that
20:06:04  <creationix>afaik, the spidernode project is getting revived, but slowly
20:06:13  <benvie>yeah I saw there was actually a commit or two to it
20:06:24  <igorzi>isaacs: hey.. yep i'll try windows
20:06:25  <benvie>but they went down the wrong road
20:06:45  <creationix>I didn't want to wait, so I started my own
20:06:50  <creationix>benvie, https://github.com/creationix/luvmonkey
20:06:55  <benvie>it's way way easier to port node to spidermonkey than to port spidermonkey to v8
20:07:09  <creationix>benvie, well, to be fair, the sm API is pretty old and crufty, they want a new API anyway
20:07:10  <benvie>that seems obvious
20:07:25  <creationix>spidernode is an excuse to implement v8's api in sm
20:07:28  <benvie>yeah it is but even so
20:07:34  <benvie>oh ha that does make sense
20:07:44  <creationix>I just don't want a C++ api
20:07:45  <benvie>from that angle, but not from the node angle
20:08:02  <creationix>right, I'm taking the easier path and just binding libuv to vanilla sm
20:08:15  <isaacs>igorzi: i'm building RC1 now
20:08:21  <benvie>which is awesome, and I bet there's a whole lot of people associated with rust that are ready to jump on board that
20:09:01  <creationix>dunno, there has been very little interest so far
20:09:13  <creationix>though I haven't pushed it yet either
20:09:26  <creationix>I'm waiting till I have something substantial to benchmark against v8 node
20:09:27  <isaacs>it'd be nice to port v8 to a clean C api
20:09:34  <creationix>indeed
20:09:42  <benvie>I suppose in one way they are already kind of like
20:09:53  <benvie>busy with something that's basically that
20:09:56  <creationix>isaacs, luvit is such a joy to write, the lua C api is squeaky clean
20:09:58  <benvie>rust is its own language
20:10:01  <isaacs>but like, that'd be basically a rewrite, to do it properly.
20:10:27  <creationix>isaacs, have you ever looked at spidermonkey's C api?
20:10:34  <benvie>haha
20:10:52  <creationix>it's reminds me of the 90's and when js was new
20:11:02  <creationix>so much cruft from all the years of evolution
20:11:23  <benvie>I will say, I wish that ffi was a core offering in v8 as it is luajit and spidermonkey
20:11:48  <creationix>sm has ffi?
20:11:59  <benvie>it's optional but yeah it's a core component that can be compiled in
20:12:06  <benvie>and often is
20:12:30  <benvie>ctypes
20:12:40  <benvie>or jsctypes I guess
20:12:52  <benvie>but it's libffi underneath
20:13:23  <benvie>nothing like the awesomeness of the lua ffi
20:13:28  <benvie>but it gets you on the bus
20:13:37  <benvie>riding to c town
20:13:38  <isaacs>creationix: i've hacked on sm a bit, yeah
20:13:50  <creationix>isaacs, cool
20:13:57  <isaacs>creationix: but not for a very long time
20:14:04  <creationix>I hear jsc has a nice API
20:14:15  <creationix>I'll try that next once luvmonkey is bootstrapped and has a community
20:14:17  <isaacs>creationix: who did you hear that from?
20:14:25  <isaacs>creationix: did you help them get psychiatric help?
20:14:28  <benvie>they've had the luxury of rewriting their shit multiple times without much notice so it wouldn't ne surprising
20:14:32  <creationix>qt folks, so it may be biased
20:14:56  <creationix>who else uses jsc
20:15:06  <benvie>they just launched their new thing
20:15:18  <creationix>though, I am impressed with the recent speed boost jsc got with the new jit/optimization layer
20:17:53  <benvie>marketing technobabble around js engines is pretty ridiculous sometimes
20:19:58  <isaacs>creationix: at kno, we used jsc in our app framework
20:20:05  <isaacs>creationix: the v8 api is much much nicer
20:20:13  <isaacs>(/opinion)
20:20:15  <creationix>v8 is clean, it's just C++
20:20:20  <isaacs>yeah
20:20:31  <creationix>personally C++ just makes me sad
20:20:34  <creationix>not sure why
20:20:35  <isaacs>it's a good example of how you can create really good software, even in an objectively bad language.
20:21:01  <creationix>after I'm done implementing node on 3 new engines, I'll propose a sane C js api
20:21:11  <benvie>**cue linus torvalds** http://harmful.cat-v.org/software/c++/linus
20:22:44  * paddybyersquit (Quit: paddybyers)
20:22:54  <isaacs>ugh. documentation css bug.
20:22:55  * isaacsannoyed
20:22:59  <isaacs>rc2 comin right up!
20:23:04  <creationix>wohoo
20:23:17  <creationix>benvie, I've always liked that response
20:23:22  <benvie>the bin works great though...ha
20:23:30  <isaacs>ah
20:23:40  <mmalecki>yeah, this response rocks
20:23:42  <benvie>yeah I do too because it doesn't just cover the technical merits, but the emotional ones. "I want to piss you off to keep you away"
20:23:48  <benvie>I always like things that touch on both
20:25:41  <creationix>I do think it would be cool to have a common C api between v8, jsc, and sm and just use that for node
20:25:41  <creationix>probably never going to happen though
20:26:29  <benvie>no way won't happen, that goes down a similar road as the js bytecode one
20:26:49  <benvie>well maybe
20:27:04  <benvie>I guess it's not the same road though it's similar
20:27:35  <benvie>it would have to be some based pared down one that each would have custom extensions to
20:27:46  * AndreasMadsenjoined
20:28:40  <benvie>or just limited to basically js itself as accessed through c
20:30:33  * felixgequit (Quit: felixge)
20:30:41  <benvie>I have ill-omens on the es-discuss unicode discussion that isaacs inadvertently kickstarted
20:31:31  <benvie>black ravens beat the air, StringIterators pace the windows
20:31:36  <isaacs>haha
20:32:31  <benvie>java's ghost glares at us from the mantle, long dead yet still breathing...glaring
20:34:46  <creationix>right, I'm not proposing anything till I've first successfully re-implemented node in atleast 3 new vms
20:35:06  <benvie>yeah at least then you'll have some unique authority to speak on the matter
20:35:14  <creationix>hmm, what would ruby be like
20:35:21  <creationix>anything single threaded will work
20:35:31  <creationix>just need to prune all the io libraries
20:35:44  * `3rdEdenquit (Quit: Leaving...)
20:35:49  <benvie>if there's at least one good thing about academics, it's that they generally (generally...) listen to demonstratable work on the topic
20:37:07  <benvie>as you said spidermonkey's api is ready for a redo from the ground up already at least
20:37:35  <benvie>maybe starting at one implementation of said proposed api would be an undeniable carrot
20:42:24  <igorzi>isaacs: [10:35|% 100|+ 354|- 13]: Done <-- win2k8
20:42:40  * felixgejoined
20:42:40  * felixgequit (Changing host)
20:42:40  * felixgejoined
20:46:31  <isaacs>igorzi: does this meet with your approval?
20:46:42  <isaacs>igorzi: i'm building RC2 now, but it's just for a CSS bug.
20:48:10  <igorzi>isaacs: i don't see any regressions, so i think we're good
20:48:13  <isaacs>kewl
20:48:20  <isaacs>once it finishes building the msi etc i'll upload
20:48:24  <isaacs>and merge in
20:48:36  <isaacs>igorzi: you were testing rc0 or rc1?
20:48:51  <igorzi>isaacs: rc1
20:49:00  <isaacs>awesome
20:49:17  <isaacs>i bumped v8 between 0 and 1, so it's good to know that didn't break anything
20:50:23  * AndreasMadsenquit (Remote host closed the connection)
20:56:47  * felixgequit (Quit: http://www.debuggable.com/)
21:13:56  <CIA-99>node: isaacs v0.6 * r2f256af / src/node_version.h : Now working on v0.6.13 (+6 more commits...) - http://git.io/o_hZjA
21:23:02  * travis-cijoined
21:23:02  <travis-ci>[travis-ci] joyent/node#536 (v0.6 - 2f256af : isaacs): The build was fixed.
21:23:02  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/1bf3a14...2f256af
21:23:02  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/782853
21:23:02  * travis-cipart
21:25:06  <isaacs>0.6.12 is out.
21:25:57  <isaacs>retweet every tweet! https://twitter.com/nodejs/status/175692884320849920
21:41:00  <benvie>js with static typing as fast as c++?
21:41:20  <benvie>I guess that's what dart is
21:57:59  * bnoordhuisjoined
22:01:40  <isaacs>bnoordhuis: hullo :)
22:01:57  <bnoordhuis>isaacs: hey isaac
22:02:21  <isaacs>dunno if ircretary told you, but that compiler thing looks like it's just for master, right?
22:02:24  <isaacs>wasn't required for 0.6.12?
22:02:40  <bnoordhuis>well, i don't know yet - does v0.6 have the same problem?
22:02:45  <bnoordhuis>if so, i'll backport it
22:02:49  <isaacs>no, it's apparently working by accident
22:03:12  <isaacs>don't sweat it
22:03:14  <isaacs>:)
22:03:15  <bnoordhuis>that's good, i suppose
22:03:22  <isaacs>also, i released 0.6.12 already
22:03:29  <bnoordhuis>ah okay
22:03:46  <isaacs>it's not worth having a 0.6.13 to fix a bug we only have in theory :)
22:03:54  <bnoordhuis>agreed :)
22:03:58  <tjfontaine>nah, :P
22:04:05  <bnoordhuis>and if the bug does pop up, backporting it is pretty trivial
22:05:11  <bnoordhuis>isaacs: have you spoken with pquerna about openssl performance recently?
22:05:15  <bnoordhuis>after the mailing post i mean
22:05:25  <isaacs>yeah, a little bit
22:05:43  <isaacs>he seemed to think it was simply a matter of apache2.2 doing its rsa calculations in a multicore manner
22:05:52  <bnoordhuis>possibly
22:06:02  <bnoordhuis>i did a benchmark this week with apache in single process mode
22:06:07  <bnoordhuis>(httpd -X)
22:06:12  * rendarquit
22:06:20  <bnoordhuis>and it's still ~30% faster...
22:06:22  <isaacs>hm
22:06:39  <bnoordhuis>but it's linked against a newer version of openssl, that could make a difference
22:06:46  <isaacs>oh, also, he suggested that we probably ought to update our openssl version that we bundle, since it's 1.0 now, and we're on like 0.9.2 or something
22:06:52  <isaacs>right ^
22:08:00  <bnoordhuis>0.9.8.r i think but yes, upgrading is a good idea :)
22:08:22  <bnoordhuis>i'll check next week if chromium is shipping a newer version
22:08:58  * toothrquit (Read error: Connection reset by peer)
22:09:09  * toothrjoined
22:39:45  <bnoordhuis>isaacs: btw, do you want me to land that strict aliasing patch?
22:40:15  <isaacs>bnoordhuis: i thought i already did..
22:40:36  <CIA-99>node: Ben Noordhuis master * r30b29d8 / (common.gypi configure):
22:40:36  <CIA-99>node: build: disable -fstrict-aliasing if gcc < 4.6.0
22:40:36  <CIA-99>node: A compiler bug in older versions of gcc makes it do unsafe optimizations at -O1
22:40:36  <CIA-99>node: and higher. This manifested itself with (at least) gcc 4.5.2 on SmartOS because
22:40:36  <CIA-99>node: it made V8 hang in a busy loop.
22:40:36  <CIA-99>node: See http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45883 - http://git.io/lneWew
22:40:49  <isaacs>BAM
22:40:53  <bnoordhuis>very good :)
22:41:03  <isaacs>i'd pulled it, but didn't push yet :)
22:44:20  * wankdankerquit (Remote host closed the connection)
22:46:21  <TooTallNate>^ which means I have to do "installVersion" 2 for node-gyp
22:46:54  <TooTallNate>:) we'll see how this goes
22:49:13  <bnoordhuis>the nodecamp.eu 2011 footage is really embarrassing to look at...
22:50:22  <bnoordhuis>god, i'm glad i'm only visible for one or two seconds
22:50:32  <TooTallNate>where's the footage :p
22:50:47  <bnoordhuis>TooTallNate: http://www.youtube.com/watch?v=oo4VySUF-g4
22:50:55  <bnoordhuis>there's probably more :/
22:51:19  <TooTallNate>i'll try to find spot you :p
22:51:27  <TooTallNate>though i don't know what i'm looking for :D
22:51:33  <TooTallNate>since you refuse to put up a gravatar :D
22:51:41  <bnoordhuis>yeah
22:52:03  <bnoordhuis>it's at moments like these that i know why i never uploaded one
22:52:09  <TooTallNate>hahahaha
22:52:15  <bnoordhuis>you can see me in the back when bert's being interviewed
22:52:15  * TooTallNatestalker
22:54:01  <TooTallNate>i remember watching that video months ago, haha
22:54:58  * travis-cijoined
22:54:58  <travis-ci>[travis-ci] joyent/node#537 (master - 30b29d8 : Ben Noordhuis): The build is still failing.
22:54:58  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/0613af0...30b29d8
22:54:58  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/783618
22:54:58  * travis-cipart
22:56:24  <TooTallNate>nice, node-gyp v0.2.1 has the solaris fix :)
22:56:56  <TooTallNate>'twas broken because the gcc_optimization_level variable had no default value :O
22:57:32  <TooTallNate>so that's why I was happy that ben remembered the default for the strict_aliasing
22:57:55  * bnoordhuisblushes
22:58:11  <TooTallNate>though i suppose technically i should do a similar gcc version check in node-gyp...
22:58:12  <TooTallNate>damn
22:58:32  <TooTallNate>the whole ./configure script is stuff i wish gyp took care of for us
22:58:39  <bnoordhuis>yeah...
22:58:53  <bnoordhuis>gyp decidedly takes the 'build system only'-approach
23:05:20  * pfox___quit (Remote host closed the connection)
23:29:17  <CIA-99>node: Jimb Esser master * r78db187 / (lib/tls.js test/simple/test-https-socket-options.js): (log message trimmed)
23:29:17  <CIA-99>node: tls: proxy set(Timeout|NoDelay|KeepAlive) methods
23:29:17  <CIA-99>node: - fix crash calling ClientRequest::setKeepAlive if the underlying request is
23:29:17  <CIA-99>node: HTTPS.
23:29:17  <CIA-99>node: - fix discarding of callback parameter when calling ClientRequest::setTimeout on
23:29:18  <CIA-99>node: HTTPS requests.
23:29:19  <CIA-99>node: - fix discarding of noDelay parameter when calling ClientRequest::setNoDelay on
23:32:13  <igorzi>bnoordhuis: hey yt?
23:33:45  * dshaw_quit (Ping timeout: 260 seconds)
23:36:57  * mmaleckichanged nick to mmalecki[zzz]
23:41:46  <tjfontaine>what is more "node" .createServer that accepts parameters to deliniate tcp/udp, or .createTCPServer .createUDPServer
23:44:11  * travis-cijoined
23:44:11  <travis-ci>[travis-ci] joyent/node#538 (master - 78db187 : Jimb Esser): The build is still failing.
23:44:11  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/30b29d8...78db187
23:44:11  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/783837
23:44:11  * travis-cipart
23:48:26  <TooTallNate>tjfontaine: assuming you're talking about a service that uses those as a transport
23:48:31  <TooTallNate>i'd say .createServer
23:48:39  <TooTallNate>since the underlying transport is an implementation detail
23:49:03  <TooTallNate>that also allows you to pick a sensible default
23:49:11  <TooTallNate>that your users dont have to think about
23:49:45  <tjfontaine>ya, that's what other out-of-band conversation came to as well
23:50:55  * paddybyersjoined
23:59:14  * dapquit (Quit: Leaving.)