00:03:20  * piscisaureus_changed nick to piscisaureus
00:13:55  * mmaleckichanged nick to mmalecki[zzz]
00:24:56  <piscisaureus>Right. Now that my beloved slurp is back, it's time to go home :-)
00:25:26  <piscisaureus>bnoordhuis: when are you coming to the office this week?
00:25:36  <bnoordhuis>piscisaureus: friday
00:25:43  <piscisaureus>bnoordhuis: a ha
00:25:52  <bnoordhuis>miss me?
00:26:35  <piscisaureus>bnoordhuis: Yeah, I like how you always warm up up the toilet seat for me.
00:26:54  <piscisaureus>anyway, later guys
00:27:24  <bnoordhuis>sleep tight, bertje
00:28:12  * piscisaureusquit (Quit: ~ Trillian Astra - www.trillian.im ~)
00:28:17  <tjfontaine>such an interesting work dynamic you have
00:29:24  <benvie>making ffi bindings to libclang is hard
00:30:22  <tjfontaine>benvie: why? btw they fixed the previous bug with the missing symbol, so it should work now on winders
00:31:08  <benvie>just because things don't don't work and don't give any feedback as to whether what you're doing wrong
00:31:19  <benvie>I've managed to work through a lot of it
00:32:31  <benvie>and yeah I had some fun with your bindings, I iterated I made a thing and iterated on it for a coupoe weeks and ended up with nearly functional 30,000 lines bindings to windows.h
00:32:56  <tjfontaine>heh
00:33:14  <benvie>https://gist.github.com/1973743
00:33:25  <benvie>it's entertaining
00:34:04  <tjfontaine>ah yes, I meant to do something with enums
00:34:40  <tjfontaine>this is pretty hilarious though
00:34:45  <benvie>it really is
00:35:00  <benvie>it takes like 7 seconds to make too
00:35:29  <tjfontaine>ya clang's pretty awesome so we all get to reap from it
00:35:38  <benvie>I wanted to take switch over to using ffi bindings
00:35:53  <benvie>and the raw generated bindings are great
00:36:03  <CIA-99>node: Shigeki Ohtsu master * r59c3923 / (3 files in 3 dirs): process: show detailed error message in process.dlopen() - http://git.io/fGCoOA
00:36:05  <benvie>but using them just isn't easy
00:36:26  <benvie>even little things like passing in the arguments, had to patch in array support for ffi
00:36:42  <benvie>I've ultimately reimplemented most of ffi itself along the way
00:36:49  <benvie>the js part
00:36:51  <tjfontaine>heh
00:37:36  <benvie>but I'm almost there. And then it will be able to code itself and I can retired
00:37:41  <benvie>the code will become sentient
00:38:50  <tjfontaine>that's the idea anyway :P
00:50:46  * travis-cijoined
00:50:46  <travis-ci>[travis-ci] joyent/node#555 (master - 59c3923 : Shigeki Ohtsu): The build is still failing.
00:50:46  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/39ce942...59c3923
00:50:46  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/802822
00:50:46  * travis-cipart
00:52:20  <CIA-99>node: Igor Zinkovsky master * r9ea5a4c / test/simple/test-process-argv-0.js : fix simple/test-process-argv-0 on windows - http://git.io/9a_fmw
00:54:02  * piscisaureus_joined
01:07:38  * travis-cijoined
01:07:38  <travis-ci>[travis-ci] joyent/node#556 (master - 9ea5a4c : Igor Zinkovsky): The build is still failing.
01:07:38  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/59c3923...9ea5a4c
01:07:38  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/802872
01:07:38  * travis-cipart
01:09:49  * mikealjoined
01:21:25  * pfox___quit (Ping timeout: 265 seconds)
01:29:57  * txdvjoined
01:35:01  * bnoordhuisquit (Ping timeout: 252 seconds)
01:39:53  * sj26quit (Excess Flood)
01:43:26  * sj26joined
01:43:29  * pfox___joined
02:06:27  * mikealquit (Quit: Leaving.)
02:08:39  * TooTallNatejoined
02:10:41  * isaacsquit (Remote host closed the connection)
02:36:06  * Ariaquit (Ping timeout: 276 seconds)
02:39:21  * dshaw_quit (Ping timeout: 276 seconds)
02:42:45  * abraxasjoined
03:10:33  * brsonquit (Quit: leaving)
03:15:02  * perezdquit (Quit: perezd)
04:14:30  * txdvquit (Ping timeout: 272 seconds)
04:15:24  * txdvjoined
04:17:36  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
04:29:02  * TooTallNatequit (Quit: Linkinus - http://linkinus.com)
04:36:34  * perezdjoined
04:49:58  * perezdquit (Quit: perezd)
05:30:29  * pfox___quit (Ping timeout: 245 seconds)
05:44:14  * pfox___joined
06:23:42  * pfox___quit (Ping timeout: 272 seconds)
06:26:39  * paddybyersjoined
07:45:21  * mjr_quit (Quit: mjr_)
07:54:20  * rendarjoined
08:11:06  * paddybyersquit (Quit: paddybyers)
08:13:02  * mjr_joined
08:14:05  * mikealjoined
08:17:41  * paddybyersjoined
09:11:22  * mmalecki[zzz]quit (Quit: Reconnecting)
09:11:40  * mmalecki[zzz]joined
09:11:50  * mmalecki[zzz]changed nick to mmalecki
09:11:56  * mmaleckiquit (Client Quit)
09:12:13  * mmaleckijoined
09:12:34  * mjr_quit (Quit: mjr_)
09:14:13  * mmaleckichanged nick to away
09:14:20  * awaychanged nick to mmalecki[away]
09:45:13  * mikealquit (Quit: Leaving.)
09:46:37  * mikealjoined
09:50:04  * mikealquit (Client Quit)
09:50:25  * mikealjoined
09:54:51  * mikealquit (Client Quit)
09:56:01  * mikealjoined
09:56:49  * mikealquit (Client Quit)
11:48:54  * abraxasquit (Remote host closed the connection)
11:49:28  * abraxasjoined
11:52:55  * bnoordhuisjoined
11:54:05  * abraxasquit (Ping timeout: 244 seconds)
12:38:27  <CIA-99>node: Rlidwka master * r443071d / lib/readline.js : readline: add multiline support - http://git.io/w1I3Jw
12:38:28  <CIA-99>node: Dmitry Nizovtsev master * r1e9bcf2 / (8 files in 3 dirs):
12:38:29  <CIA-99>node: net, http, https: add localAddress option
12:38:29  <CIA-99>node: Binds to a local address before making the outgoing connection. - http://git.io/KlYMAQ
12:42:33  * elijah-mbpquit (Ping timeout: 276 seconds)
12:45:17  * elijah-mbpjoined
12:53:17  * travis-cijoined
12:53:18  <travis-ci>[travis-ci] joyent/node#557 (master - 443071d : Rlidwka): The build is still failing.
12:53:18  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/9ea5a4c...443071d
12:53:18  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/805744
12:53:18  * travis-cipart
12:54:25  * AndreasMadsenjoined
12:57:37  <AndreasMadsen>ircretary: ask isaacs if he would review that cluster patch #2740
12:57:37  <ircretary>AndreasMadsen: I'm not sure what to do with that command. Ask for help in PM.
12:58:30  <AndreasMadsen>ircretary: tell isaacs would you care to review that cluster patch #2740
12:58:31  <ircretary>AndreasMadsen: I'll be sure to tell isaacs
13:18:36  * indutny_sleepingchanged nick to indutny
14:17:15  * piscisaureus_joined
14:45:19  * rendarquit
14:47:37  * rendarjoined
15:31:38  * hij1nxjoined
15:53:12  * piscisaureus__joined
15:53:24  * piscisaureus_quit (Read error: Connection reset by peer)
15:59:12  * piscisaureus__quit (Ping timeout: 248 seconds)
16:04:21  * isaacsjoined
16:16:36  * mmalecki[away]changed nick to mmalecki
16:41:30  * hij1nxquit (Quit: hij1nx)
16:47:40  * stephankjoined
16:52:52  * piscisaureus_joined
17:00:35  <bnoordhuis>piscisaureus_: ping
17:03:04  <bnoordhuis>piscisaureus_: http_simple with bytes/1024 seems to spend over 25% of its time in String::WriteToFlat()
17:03:08  <bnoordhuis>that doesn't look kosher
17:03:52  <bnoordhuis>is your patch addressing that issue? if so, can i test it?
17:14:51  <CIA-99>node: Ben Noordhuis master * r544e5ee / benchmark/http_simple_auto.js :
17:14:51  <CIA-99>node: bench: add http_simple_auto benchmark
17:14:51  <CIA-99>node: Starts a server and benchmarks it with ab. - http://git.io/DH6K3A
17:17:55  <bnoordhuis>isaacs: msft call tonight?
17:19:02  * piscisaureus_quit (Quit: ~ Trillian Astra - www.trillian.im ~)
17:20:25  * piscisaureus_joined
17:25:39  <isaacs>bnoordhuis: yes, please
17:25:52  <bnoordhuis>isaacs: okay, i'll try to make it :)
17:26:02  <isaacs>thanks. i miss your lovely baritone.
17:26:18  <bnoordhuis>haha
17:28:22  * indutnychanged nick to indutny_31456
17:28:28  * indutny_31456changed nick to indutny_sleeping
17:30:16  <piscisaureus_>scary
18:00:19  * travis-cijoined
18:00:19  <travis-ci>[travis-ci] joyent/node#558 (master - 544e5ee : Ben Noordhuis): The build is still failing.
18:00:19  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/443071d...544e5ee
18:00:19  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/807493
18:00:19  * travis-cipart
18:06:29  * mjr_joined
18:08:34  * mikealjoined
18:10:27  * TooTallNatejoined
18:11:05  * `3rdEdenjoined
18:25:14  * `3rdEdenquit (Quit: Leaving...)
18:27:14  * dshaw_joined
18:33:35  * AvianFluquit (Quit: Leaving)
18:39:58  * brsonjoined
18:42:55  * `3rdEdenjoined
18:59:54  * `3rdEdenquit (Quit: Leaving...)
19:00:39  <bnoordhuis>call time?
19:02:22  <isaacs>yessir
19:05:14  * pieternjoined
19:33:09  <TooTallNate>bnoordhuis: should be ready --> https://github.com/joyent/node/pull/2871
19:39:12  <bnoordhuis>piscisaureus_: did you read what i said about String::WriteToFlat() earlier today?
19:40:37  <mmalecki>actually, TooTallNate's pull request means that we can totally get rid of any compiled modules from forever
19:44:45  <CIA-99>node: Nathan Rajlich master * r07c886f / (3 files in 3 dirs):
19:44:45  <CIA-99>node: process: add `process.hrtime()`
19:44:45  <CIA-99>node: This commit adds a high-resolution timer function. - http://git.io/6LKq6g
19:46:26  * Skomskijoined
19:47:17  <CIA-99>node: Ming Liu master * rb602390 / doc/api/repl.markdown : docs: correct repl eval callback signature - http://git.io/ZdlV2w
19:48:48  * igorziquit (Ping timeout: 245 seconds)
19:49:27  <TooTallNate>mmalecki: nice, what are you guys using the high res timer for?
19:49:55  <mmalecki>TooTallNate: creating sockets
19:50:22  <mmalecki>(to avoid race conditions when 2 forever workers are being spawned at the same time, I suppose)
19:56:27  <piscisaureus_>bnoordhuis: I just looked it up
19:57:56  <bnoordhuis>piscisaureus_: i tried your patch btw, but it didn't make a difference on this particular benchmark
19:59:12  <piscisaureus_>bnoordhuis: sure - because the buffer constructor doesn't use it (yet)
19:59:48  <piscisaureus_>bnoordhuis: although I find it funny because v8 is supposed to buffer the string with FLAG_MANY_WRITES_EXPECTED so it should just flatten it once.
20:00:02  * travis-cijoined
20:00:02  <travis-ci>[travis-ci] joyent/node#559 (master - 07c886f : Nathan Rajlich): The build is still failing.
20:00:02  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/544e5ee...07c886f
20:00:02  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/808236
20:00:02  * travis-cipart
20:00:08  <piscisaureus_>bnoordhuis: (which is completely "unreal" btw)
20:00:49  <bnoordhuis>piscisaureus_: https://gist.github.com/47db6e5656dc4f07ccee <- that's what it looks like in callgrind
20:01:07  <bnoordhuis>so how do i make the buffer constructor use your patch?
20:01:20  <bnoordhuis>also, we never landed that string concatenation patch, did we?
20:01:55  <piscisaureus_>bnoordhuis: I don't think my patch will improve it
20:02:05  <piscisaureus_>bnoordhuis: but I will try to use it
20:02:20  * travis-cijoined
20:02:20  <travis-ci>[travis-ci] joyent/node#560 (master - b602390 : Ming Liu): The build is still failing.
20:02:20  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/07c886f...b602390
20:02:20  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/808256
20:02:20  * travis-cipart
20:10:42  <piscisaureus_>bnoordhuis: what are we using Buffer._charsWritten for?
20:11:28  * benviequit (Read error: No route to host)
20:11:45  * benviejoined
20:15:51  * AvianFlujoined
20:17:25  * mikealquit (Quit: Leaving.)
20:20:55  <piscisaureus_>bnoordhuis: what was that ab command that you used again?
20:24:37  <bnoordhuis>piscisaureus_: ab -c 100 -t 30 http://127.0.0.1:8000/bytes/1024
20:24:58  <bnoordhuis>or you can use `node benchmark/http_simple_auto.js -c 100 -t 30 bytes/1024`
20:25:53  <piscisaureus_>bnoordhuis: I am trying with my patch but it does not seem to make any difference
20:26:47  <bnoordhuis>alas, no
20:27:30  <piscisaureus_>bnoordhuis: it now spends 25% of the time in node::string_to_utf8<2>
20:28:23  <bnoordhuis>piscisaureus_: i didn't see that function in the call graph when i tested your patch from last friday...
20:28:31  <piscisaureus_>bnoordhuis: try my patch again
20:29:42  <piscisaureus_>bnoordhuis: btw, I didn't land the gcc fixes we did friday
20:29:47  <piscisaureus_>let me land those too
20:30:23  <bnoordhuis>and fix up the merge conflicts while you're at it :)
20:30:29  * benviequit (Ping timeout: 245 seconds)
20:30:58  * benviejoined
20:31:09  <piscisaureus_>0
20:32:08  <piscisaureus_>bnoordhuis: k, updated my branch
20:33:57  <bnoordhuis>recompiling...
20:40:00  <bnoordhuis>piscisaureus_: https://gist.github.com/3b6d363f58ba03820e2b
20:40:20  <bnoordhuis>at least it seems to be executing more instructions in the same time frame...
20:41:06  * elijah-mbpquit (Remote host closed the connection)
20:47:58  <bnoordhuis>piscisaureus_: actually... your patch seems to make it slower
20:48:44  * mikealjoined
20:53:52  <piscisaureus_>bnoordhuis: ill get back to you in a couple of minutes
20:55:21  <bnoordhuis>piscisaureus_: sure
20:55:39  <bnoordhuis>some stats, benchmarked with -c 100 -n 3000: https://gist.github.com/b6103fc9cf61590c2961
21:05:03  <tjfontaine>bnoordhuis: did this guy ended up satisfying the contributor stuff? https://github.com/joyent/node/pull/2869
21:07:36  <bnoordhuis>tjfontaine: yes
21:08:10  <bnoordhuis>i assume the implicit question was 'can you merge it?'?
21:08:32  <tjfontaine>perhaps :)
21:09:37  <piscisaureus_>bnoordhuis: about my patch making it slower - how big is the difference?
21:10:19  <bnoordhuis>piscisaureus_: see the gist
21:10:51  <bnoordhuis>1,008,503,931 vs 870,971,631 instructions
21:11:26  <piscisaureus_>bnoordhuis: that makes sense because v8 flattens the string,
21:11:50  <piscisaureus_>bnoordhuis: it has no link with reality because you would not often write exactly the same string over and over again
21:15:21  * isaacsquit (Remote host closed the connection)
21:16:20  <piscisaureus_>bnoordhuis: try with this patch applied to http_simple.js
21:16:20  <piscisaureus_>https://gist.github.com/1989015
21:21:16  <CIA-99>node: ssuda master * rb72d43c / (lib/dgram.js test/simple/test-dgram-bind.js): dgram: fix when we call .close() inside 'listening' - http://git.io/hz2rrA
21:21:52  * txdvquit (Ping timeout: 248 seconds)
21:22:15  <tjfontaine>bnoordhuis: thanks
21:23:50  <bnoordhuis>piscisaureus_: i get what you're getting at if you get what i mean
21:24:34  <bnoordhuis>if i make http_simple send a 1024 chars long string literal, String::WriteToFlat() doesn't show up in the call graph
21:26:22  <piscisaureus_>bnoordhuis: yes, of course. Neither will you see v8::String::ReadMemory::next
21:27:29  <bnoordhuis>so it's really a poor benchmark...
21:29:24  <piscisaureus_>bnoordhuis: http_simple with bytes/xxx is poor
21:29:55  <piscisaureus_>bnoordhuis: I would say generating the string every time is much more realistic.
21:31:24  * isaacsjoined
21:36:22  * dshaw_1joined
21:36:27  * travis-cijoined
21:36:27  <travis-ci>[travis-ci] joyent/node#561 (master - b72d43c : ssuda): The build is still failing.
21:36:27  <travis-ci>[travis-ci] Change view : https://github.com/joyent/node/compare/b602390...b72d43c
21:36:27  <travis-ci>[travis-ci] Build details : http://travis-ci.org/joyent/node/builds/809182
21:36:27  * travis-cipart
21:38:26  * dshaw_quit (Ping timeout: 244 seconds)
21:53:40  * dshaw_joined
21:55:17  * Skomskiquit (Quit: Nettalk6 - www.ntalk.de)
21:55:29  * dshaw_1quit (Ping timeout: 244 seconds)
21:56:58  * igozijoined
22:10:34  * bnoordhuisquit (Ping timeout: 252 seconds)
22:15:34  * mjr_quit (Quit: mjr_)
22:21:26  * AvianFluquit (Quit: Leaving)
22:25:28  <igozi>isaacs: what was that module that you mentioned earlier today? (i think you said it was 'toss'?)
22:26:21  <isaacs>toss yes
22:26:26  <isaacs>it's one of SubStack's
22:26:35  <isaacs>https://github.com/substack/node-toss#readme
22:27:06  <igozi>isaacs: thx
22:38:00  <AndreasMadsen>when can this land -> https://github.com/joyent/node/pull/2743
22:41:14  <AndreasMadsen>piscisaureus_: ^
22:44:20  * mikealquit (Quit: Leaving.)
22:45:21  * rendarquit
22:54:04  * mikealjoined
22:59:21  * paddybyersquit (Quit: paddybyers)
23:15:35  * AndreasMadsenquit (Remote host closed the connection)
23:25:42  * hij1nxjoined
23:45:03  <piscisaureus_>yay, it works: https://github.com/piscisaureus/slurp-logs/commits/master
23:45:14  <piscisaureus_>hey igozi, did you have your name changed?
23:47:12  <igozi>oops
23:47:15  * igoziquit (Quit: Page closed)
23:47:40  * igorzijoined
23:48:25  <piscisaureus_>igorzi: you should get an actual irc client :-)
23:51:45  <igorzi>piscisaureus_: i can only get in on the web front end.. the irc port is blocked
23:52:13  * piscisaureus_rolls eyes
23:52:24  * piscisaureus_changed nick to piscisaureus
23:54:16  <piscisaureus>All freenode servers listen on ports 6665, 6666, 6667, 6697 (SSL only), 7000 (SSL only), 7070 (SSL only), 8000, 8001 and 8002.
23:54:27  <piscisaureus>I bet one of these will be open :-)
23:54:41  <igorzi>piscisaureus: i'll try :)
23:55:50  <igorzi>piscisaureus: is there a good irc client written in node?
23:56:05  <piscisaureus>igorzi: like, one that you would run in a browser?
23:56:27  <igorzi>piscisaureus: console
23:56:33  <piscisaureus>hehe
23:56:36  <piscisaureus>maybe
23:56:45  <piscisaureus>if you want browser, you could deploy subway
23:56:50  <piscisaureus>https://github.com/thedjpetersen/subway
23:57:05  <piscisaureus>console would be nice
23:57:36  <piscisaureus>nah, I don't see any console ones
23:58:28  <piscisaureus>people tend to use irssi if they like consoles
23:58:32  <piscisaureus>I use trillian