00:02:48  <pquerna>rphillips: https://gist.github.com/d5f76453282285362248
00:02:58  <pquerna>rphillips: so, i never get client or server data callbacks
00:03:01  <pquerna>ever
00:03:18  <pquerna>on('data') never fires for either
00:04:56  <pquerna>rphillips: for example test-tls-client-reject.lua does this too, it has an assert inside :on('data')... but it doesn't assert that its actually called
00:05:00  <pquerna>rphillips: and its not
00:06:43  <pquerna>https://gist.github.com/a7bd326a393f6cb63ff0
00:06:49  <pquerna>causes a test failure for example
01:01:29  * xmingquit (Read error: Connection reset by peer)
02:34:10  * tim_smartchanged nick to tim_smart|away
02:49:43  * philipsquit (Excess Flood)
02:52:48  * philipsjoined
03:00:54  * xmingjoined
03:25:09  * dvv-androidjoined
03:28:10  * pancakejoined
03:32:28  * pancakequit (Ping timeout: 255 seconds)
03:48:18  * dvv-androidquit (Ping timeout: 246 seconds)
04:15:24  * dvvquit (Ping timeout: 240 seconds)
04:17:49  * dvvjoined
04:23:01  <rphillips>pquerna: https://gist.github.com/b8f5375c5a2559635eef
04:23:12  <rphillips>couple tweaks
04:23:18  <rphillips>and I applied this patch to luvit: https://gist.github.com/bc3c536d72f147f41c27
04:23:26  <rphillips>the CryptoStream:write didn't support buffers
04:25:08  <rphillips>it still is failing (18138 ~= 17710)
04:27:47  * dvvquit (Ping timeout: 248 seconds)
04:34:53  <rphillips>in fact... I wonder if the regular socket:write supports buffers
04:41:27  * philipsquit (Excess Flood)
04:42:40  <rphillips>i'll look at it more in the morning... thx
04:44:49  * philipsjoined
05:20:13  * tim_smart|awaychanged nick to tim_smart
06:29:14  * tsingjoined
06:34:02  * tsing_joined
06:37:38  * tsingquit (Ping timeout: 244 seconds)
06:55:27  * tsing_quit (Quit: Linkinus - http://linkinus.com)
07:56:56  * xmingquit (Changing host)
07:56:56  * xmingjoined
09:56:10  * mmaleckijoined
09:59:09  * spionWjoined
10:00:12  * tim_smartchanged nick to tim_smart|away
11:49:54  * tokyodanjoined
11:56:45  * tokyodanquit (Quit: tokyodan)
12:33:08  * mmaleckiquit (Ping timeout: 272 seconds)
12:34:25  * TheJHjoined
13:04:01  * mmaleckijoined
13:37:46  * TheJHquit (Ping timeout: 260 seconds)
14:04:09  * dvv-androidjoined
14:10:36  * dvv-androidquit (Ping timeout: 246 seconds)
14:36:55  <pquerna>rphillips: morning
14:37:52  <rphillips>pquerna: hey
14:47:05  <rphillips>i'm thinking those last 513 bytes are missing because there is a lack of buffering in luvit
14:49:31  * mmaleckiquit (Ping timeout: 260 seconds)
15:07:58  * mmaleckijoined
15:23:50  <rphillips>pquerna: figured out part of it
15:24:01  <rphillips>:close() is the wrong function to call should be :done()
15:24:34  <rphillips>in keystone/init.lua:80
15:26:20  <rphillips>other issue is the http.request function returns the socket and not the http request object
15:26:34  <rphillips>http module needs major help
15:33:34  * spionWquit (Ping timeout: 246 seconds)
15:36:04  * mmaleckiquit (Ping timeout: 245 seconds)
15:48:48  <pquerna>rphillips: ahhh
15:51:08  * dvv-androidjoined
16:20:27  * dvv-androidquit (Ping timeout: 246 seconds)
16:24:18  <creationix>rphillips: http needs to be a standalone project
16:24:28  <creationix>it needs proper attention
16:26:53  <rphillips>working on it
16:34:16  * mmaleckijoined
16:36:10  <levi>I think I may have hit a quirk with http responses while trying to get dvv's sockjs server working, too. I haven't had time to pin it down, though.
16:46:07  <creationix>rphillips: so does this mean you guys are using http more now?
16:46:13  <creationix>I know before it was all json over tcp
16:47:21  <xming>is ltin for serializing?
16:47:34  <creationix>xming: serializing and config files
16:47:42  <creationix>I prefer msgpack for serializing
16:47:55  <creationix>but ltin is more lua specefic
16:48:03  <creationix>(tables as keys, for example)
16:48:24  <rphillips>creationix: yeah, we are using http for a few things now
16:48:31  <creationix>rphillips: fabulous!
16:48:42  <creationix>rphillips: sorry for the mess
16:48:50  <creationix>life has gotten in the way of my hobby
16:49:05  <creationix>hopefully soon I get to work on luvit some at my day job
16:49:07  <rphillips>np. it'll get fixed
16:49:36  <xming>while you guys are all here, does luvit use luacrypt for tls? So luvit's depends on luacrypt, and luacrypt on opeensl
17:18:29  <pquerna>it uses openssl directly for some features
17:18:37  <pquerna>and uses luacrypt for some features too
17:21:32  * dvv-androidjoined
17:41:11  * `3rdEdenjoined
17:45:23  <creationix>so if I understand right, there is no no such thing as arrays in lua
17:45:29  <creationix>just a standard library called "table"
17:45:51  <creationix>if I wanted to index my arrays starting at 0, I could just replace the table library right?
17:45:56  <creationix>or make a new array lib?
17:46:22  * TheJHjoined
17:46:58  <creationix>hmm, well I guess the {1, 2, 3} syntax hard-coded the starting-at-one semantic
17:47:03  <creationix>oh well
17:50:09  <levi>Metatable tweaks might be able to get you to 0-indexed arrays, but the benefit seems dubious to me.
17:52:32  <rphillips>creationix: just start at 1
17:52:44  <creationix>yeah, I'll stick to 1
17:52:54  <creationix>someone was asking me how hard it would be to change it to 0
17:53:14  <creationix>I'm often amazing how seriously small lua is when you remove the standard libraries
17:54:38  <creationix>OSX pro tip: put this in your .profile
17:54:39  <creationix>PROMPT_COMMAND='echo -ne "\033]0;${USER}@${HOSTNAME}: ${PWD}\007"'
18:03:38  * `3rdEdenquit (Quit: Leaving...)
18:06:24  * `3rdEdenjoined
18:06:31  * `3rdEdenquit (Client Quit)
18:07:29  <levi>So, it seems that it would be fairly straightforward to implement a trait-based object system and multi-dispatch generic functions.
18:08:16  <levi>And they can all co-exist in the same system, although interoperability might be interesting.
19:06:48  * tim_smart|awaychanged nick to tim_smart
19:07:38  * spionWjoined
19:17:23  * mmaleckiquit (Ping timeout: 248 seconds)
19:18:53  <creationix>yeah, lua OOP is whatever you want
19:19:01  <creationix>I implemented basic prototypal for luvit objects
19:19:05  * mmaleckijoined
19:39:29  * dvv-androidquit (Read error: Connection reset by peer)
20:01:07  <pquerna>rphillips: heya
20:02:07  <pquerna>rphillips: out of meetings and interviews
20:04:23  <rphillips>pquerna: awesome
20:04:44  <rphillips>saw Max in the lunchroom earlier
20:20:41  <pquerna>rphillips: so, anything i should do to help? you just hacking on the test cases?
20:21:03  <rphillips>i'm hacking on porting the http functionality from node
20:21:08  <pquerna>okay
20:21:16  <rphillips>we could divide on conqer
20:21:46  <pquerna>sure, if there is a clear line or module i should do
20:23:10  <pquerna>i might try using miniz too, see if it gets our code even smaller, but thats really a distraction
20:23:17  <pquerna>or should i attack a different todo?
20:23:28  <CIA-113>Ryan Phillips features/improve_http_module * r385f884 / lib/luvit/http.lua : http: imrpovements part 1 - http://git.io/Ck1WiQ
20:23:56  <rphillips>is there a better way? I was porting the http module almost verbatim
20:24:14  <pquerna>rphillips: i mean, its not a bad thing
20:24:21  <pquerna>its prolly all good
20:24:46  <pquerna>going to fix a bunch of issues we were just ignoring before
20:24:50  <pquerna>all the header stuff is good too
20:25:01  <rphillips>yeah. it'll fix a lot of things for everyone
20:25:35  <pquerna>seems best maybe if you just keep attacking it, i take something else form the gdoc?
20:25:42  <rphillips>ok.
20:25:45  <rphillips>np
20:25:59  <pquerna>mmm
20:26:06  <pquerna>i'll add test.skip i guess to start :x
20:26:10  <rphillips>hah
20:26:16  <pquerna>dont' see others that are good, thoughts?
20:26:57  <rphillips>looking
20:27:15  <rphillips>pquerna: sending the virgo version and bundle version would be helpful
20:28:11  <rphillips>starting on windows support might be good
20:28:16  <rphillips>wdyt?
20:28:51  <rphillips>custom plugins is something Kami_ may or may not have started yet
20:29:30  <pquerna>windows support in what regard?
20:29:47  <rphillips>service and installer would be a good start, IMO
20:29:54  <pquerna>ah
20:30:00  <pquerna>fml is my response
20:30:14  <pquerna>i guess i could do the RPM too
20:30:15  <rphillips>ROTFL
20:30:20  <rphillips>yah
20:30:24  <pquerna>would be good to get RHEL going
20:30:29  <pquerna>okay, doing test.skip
20:30:35  <pquerna>then will start on general packaging
20:30:40  <rphillips>ok
20:30:47  <rphillips>it's getting close
20:30:58  <rphillips>it's always the last 10%
20:31:02  <pquerna>always :)
20:32:33  <rphillips>pquerna: oh, rch had a good idea to put fb303 metrics into the agent endpoint to propogate those up to the dashboard
20:32:57  <rphillips>number of clients connected, rate, etc
20:34:15  <pquerna>ew, node.js
20:35:14  <rphillips>ew?
20:36:36  <rphillips>oh, for the dashboard
20:36:38  <rphillips>heh
20:47:07  * dvv-androidjoined
20:52:17  * mmaleckiquit (Ping timeout: 265 seconds)
20:57:26  * dvv-androidquit (Ping timeout: 276 seconds)
20:58:08  * mmaleckijoined
20:58:21  * leviquit (Ping timeout: 252 seconds)
21:01:53  <pquerna>rphillips: https://github.com/racker/luvit-bourbon/pull/6
21:02:21  * levijoined
21:04:45  <rphillips>pquerna: lgtm
21:05:05  <pquerna>rphillips: thanks
21:05:43  <rphillips>interesting. github had an ajaxy update of you closing the PR
21:05:59  <pquerna>neat
21:06:55  <rphillips>bb in about 30... going home to get a package
21:07:33  <CIA-113>Ryan Phillips features/improve_http_module * r1508455 / lib/luvit/http.lua : http: more functionality #2 - http://git.io/yPgccg
21:07:48  <pquerna>gl hf
21:14:15  * mmaleckiquit (Ping timeout: 245 seconds)
21:28:11  * spionWchanged nick to spionL
22:13:54  * TheJHquit (Ping timeout: 240 seconds)
22:36:43  * neomantraquit (Quit: Leaving.)
23:11:21  * spionLquit (Ping timeout: 246 seconds)
23:24:46  * aliem_joined