00:08:10  <rje>rphillips, they pack nw.exe in there
00:08:36  <rje>https://github.com/nwjs/nw.js/wiki/How-to-package-and-distribute-your-apps#windows-1
00:21:06  <rje>rphillips, got it
00:21:17  <rje>the dll, needs the failure handle too
00:21:33  <rje>creationix, ^^^^
00:27:30  <rphillips>interesting
00:28:08  <rje>https://gist.github.com/rjemanuele/e9de7b9274f900f7ee47
00:29:22  <rje>if that gets included in sigar.dll all is well
00:29:35  <rphillips>oh interesting
00:29:55  <rphillips>so the plugins could include a snippet to put this code in there
00:30:24  <rje>i'd suggest providing a .o file with the luvi.lib
00:31:42  <rje>OR the plugins could have an exported call to set this up
00:31:53  <rphillips>good idea
00:32:04  <rje>we should talk about this tomorrow
00:32:38  <rphillips>+1
00:32:39  <rje>but right now i have a rackspace-monitoring-agent.exe that can load sigar.dll without luvi.exe
00:35:55  * UniOnquit (Remote host closed the connection)
00:43:56  <rphillips>https://github.com/luvit/luvit/issues/267#issuecomment-78399186
00:44:01  <rphillips>interesting issue
00:44:12  <rphillips>uv.walk(p) doesn't show an active tcp handle
01:06:32  * joconnorquit (Ping timeout: 244 seconds)
01:07:55  <rphillips>it's the end event that blocks the process exit
01:29:41  <rphillips>https://github.com/luvit/luvit/issues/665
01:30:02  <rphillips>woops: https://github.com/luvit/luvit/pull/666
01:30:08  * travis-cijoined
01:30:09  <travis-ci>luvit/luvit#1835 (fixes/flush_http - 9ff4bf0 : Ryan Phillips): The build has errored.
01:30:09  <travis-ci>Change view : https://github.com/luvit/luvit/commit/9ff4bf05eb63
01:30:09  <travis-ci>Build details : http://travis-ci.org/luvit/luvit/builds/54038002
01:30:09  * travis-cipart
01:30:45  * travis-cijoined
01:30:46  <travis-ci>luvit/luvit#1836 (fixes/flush_http - e3e535f : Ryan Phillips): The build passed.
01:30:46  <travis-ci>Change view : https://github.com/luvit/luvit/compare/9ff4bf05eb63...e3e535fe7c42
01:30:46  <travis-ci>Build details : http://travis-ci.org/luvit/luvit/builds/54038060
01:30:46  * travis-cipart
01:30:48  <rphillips>failed because of my force push
01:30:52  <rphillips>there it goes
01:37:55  * nateratorquit (Quit: naterator)
01:38:22  * nateratorjoined
01:38:40  * nateratorquit (Client Quit)
03:08:38  <rphillips>hmm. tis strange... getting a certificate error from our auth endpoints with luvit
03:08:50  <rphillips>google.com passes
03:18:38  * nateratorjoined
03:29:41  * nateratorquit (Quit: naterator)
03:52:28  <rphillips>wow that is a weird one
03:52:50  <rphillips>the thawte root server ca was not in the ca store
03:52:54  <rphillips>https://github.com/luvit/luvit/pull/667
03:53:17  * travis-cijoined
03:53:18  <travis-ci>luvit/luvit#1838 (fixes/add_thawte_premium_server_ca - 8499709 : Ryan Phillips): The build passed.
03:53:18  <travis-ci>Change view : https://github.com/luvit/luvit/commit/84997092d01b
03:53:18  <travis-ci>Build details : http://travis-ci.org/luvit/luvit/builds/54047315
03:53:18  * travis-cipart
03:53:21  <rphillips>thawte premium server ca*
04:07:14  * travis-cijoined
04:07:15  <travis-ci>luvit/luvi#440 (bump/openssl_1.0.2 - 79e206a : Ryan Phillips): The build has errored.
04:07:15  <travis-ci>Change view : https://github.com/luvit/luvi/commit/79e206ac1412
04:07:15  <travis-ci>Build details : http://travis-ci.org/luvit/luvi/builds/54047974
04:07:15  * travis-cipart
04:09:40  * travis-cijoined
04:09:41  <travis-ci>luvit/luvi#440 (bump/openssl_1.0.2 - 79e206a : Ryan Phillips): The build passed.
04:09:41  <travis-ci>Change view : https://github.com/luvit/luvi/commit/79e206ac1412
04:09:41  <travis-ci>Build details : http://travis-ci.org/luvit/luvi/builds/54047974
04:09:41  * travis-cipart
05:41:39  <rphillips>rje: https://github.com/luvit/luvit/issues/668
05:42:14  <rphillips>i'm updating the keystone (identity client) and the response never emits the data response
06:49:52  * Akagi201joined
07:46:29  * hdmsquit (Quit: hdms)
09:20:58  * Akagi201_joined
09:23:59  * Akagi201quit (Ping timeout: 245 seconds)
10:03:36  * Akagi201_quit (Remote host closed the connection)
10:09:12  * SouLjoined
11:41:43  <rphillips>good morning
12:32:08  * SouLquit (Ping timeout: 272 seconds)
13:09:08  * UniOnjoined
13:09:09  * UniOnquit (Remote host closed the connection)
13:09:33  * UniOnjoined
13:12:31  <rphillips>creationix: https://github.com/luvit/luvit/blob/master/deps/http.lua#L166-L177
13:12:45  <rphillips>is there something that needs to be done for chunked responses?
13:33:21  * hdmsjoined
13:40:13  * hdmsquit (Remote host closed the connection)
13:41:01  * hdmsjoined
14:49:34  * SouLjoined
15:06:50  <creationix>rphillips: http-codec handles chunked encoding for you on both decoding and encoding
15:07:01  <creationix>(on encoding make sure to set the proper header manually
15:23:31  <creationix>if you set “Transfer-Encoding: chunked” http-codec will encode with chunked encoding.
15:39:38  * SouLquit (Ping timeout: 246 seconds)
15:45:52  <rphillips>hmm. i'm passing that header down
16:20:18  * joconnorjoined
16:23:38  <rphillips>creationix: does the incomingmessage object need the 'extra' data pushed into it?
16:23:55  <rphillips>the extra data is the start of the payload
16:37:45  <rje>.
16:37:54  <rphillips>hey rje
16:38:00  <rje>hey
16:38:04  <rphillips>could you help me with this http post bug?
16:38:36  * SouLjoined
16:39:15  <rje>rphillips, sure
16:39:53  <rphillips>make sure to get master luvit
16:40:30  <rje>do you have a test in luvit?
16:40:52  <rphillips>gist https://gist.github.com/rphillips/384641f12b8580ff643f
16:41:18  <rphillips>it's one of luvit's tests we haven't ported yet
16:49:39  <rphillips>rje: what is happening is the chunked response is lost
16:50:06  <rje>rphillips: i'm porting that test now
16:53:25  * nateratorjoined
16:53:32  * travis-cijoined
16:53:33  <travis-ci>luvit/luvit#1841 (master - 2141e69 : Ryan Phillips): The build passed.
16:53:33  <travis-ci>Change view : https://github.com/luvit/luvit/compare/d27771fc095b...2141e69c49a0
16:53:33  <travis-ci>Build details : http://travis-ci.org/luvit/luvit/builds/54121987
16:53:33  * travis-cipart
16:57:20  <rphillips>https://tessel.io/blog/112888410737/moving-faster-with-io-js
17:10:34  * nateratorquit (Quit: naterator)
17:26:19  <creationix>I’ve been watching colony (js in lua for tessel) for a while, they are constantly hitting edge cases
17:26:24  <creationix>I wonder how real v8 compares
17:33:09  <rphillips>yeah, never been a fan of coffeescript either
17:47:41  <rje>rphillips: chunked data gets to the server just fine. i don't think the chunks are terminated right with a 0 len chunk
17:48:25  <rje>the done method will need to do that
17:58:17  <creationix>write an empty string to the encoder
17:58:19  <rphillips>rje: the response comes back
17:58:34  <rphillips>i see the payload from the identity server, just not in the 'data' event
18:00:43  <rje>creationix: that's what i did
18:11:58  * SouL_joined
18:13:17  * SouLquit (Read error: Connection reset by peer)
18:21:12  <rphillips>rje: https://github.com/virgo-agent-toolkit/luvit-rackspace-monitoring-client/pull/9
18:21:26  <rphillips>here is a branch that also tests the http functionality
18:22:12  <rje>k, thanks
18:29:49  <rje>rphillips: https://github.com/luvit/luvit/commit/0c84c7e3211476bb2b45fafaf5d6891d50f38bba
18:30:20  <rje>rphillips: are you saying you won't recieve data here? https://github.com/luvit/luvit/commit/0c84c7e3211476bb2b45fafaf5d6891d50f38bba#diff-635b10e498fcb8ff519110018109f0dbR65
18:30:38  <rphillips>yes, presently... let me try this patch
18:38:01  * SouLjoined
18:38:25  * travis-cijoined
18:38:26  <travis-ci>luvit/luvit#1842 (http_post_error - 0c84c7e : Rob Emanuele): The build passed.
18:38:26  <travis-ci>Change view : https://github.com/luvit/luvit/commit/0c84c7e32114
18:38:26  <travis-ci>Build details : http://travis-ci.org/luvit/luvit/builds/54137832
18:38:26  * travis-cipart
18:41:17  * SouL_quit (Ping timeout: 246 seconds)
18:49:12  * SouL_joined
18:53:08  * SouLquit (Ping timeout: 265 seconds)
19:17:19  <rphillips>strange... still seeing the issue
19:55:44  * nateratorjoined
20:00:07  * DarkGodquit (Remote host closed the connection)
20:03:18  * DarkGodjoined
20:12:09  * nateratorquit (Quit: naterator)
20:14:49  * nateratorjoined
20:17:04  <creationix>Instructions for a typical luvit/lit based app https://github.com/luvit/blog.luvit.io#local-development
20:19:53  * travis-cijoined
20:19:54  <travis-ci>luvit/luvit#1844 (master - 8e5c6df : Tim Caswell): The build passed.
20:19:54  <travis-ci>Change view : https://github.com/luvit/luvit/compare/2141e69c49a0...8e5c6dfa069a
20:19:54  <travis-ci>Build details : http://travis-ci.org/luvit/luvit/builds/54150760
20:19:54  * travis-cipart
20:30:43  * SouLjoined
20:31:21  * SouL_quit (Read error: Connection reset by peer)
20:31:45  * nateratorquit (Quit: naterator)
20:50:08  * SouL_joined
20:53:37  <rphillips>wow. that is really slick tim :)
20:53:44  * SouLquit (Ping timeout: 272 seconds)
20:56:56  <rphillips>rje: [email protected]:/racker/luvit-keystone-client.git
20:57:02  <rphillips>luvi-up branch
20:57:14  <rphillips>can you help me look at this and to see why it isn't behaving the same as the test case
20:58:01  <rphillips>test file I'm using https://gist.github.com/rphillips/5914501bc5cd86e712af
20:58:59  <rphillips>i see the payload in the http.lua decoder() return
20:59:34  <rphillips>creationix: what does 'private' mean in the package.lua
20:59:35  <rphillips>?
20:59:43  <creationix>it prevents lit from publishing it
20:59:49  <creationix>same as npm I think
21:00:41  <rphillips>gotcha
21:08:02  * nateratorjoined
21:18:32  * nateratorquit (Quit: naterator)
21:19:27  <rphillips>creationix: tracked it down to this check https://github.com/luvit/luvit/blob/master/deps/http-codec.lua#L245
21:20:21  <rphillips>i have a chunk that is a partial read
21:20:31  <creationix>hmm, I wonder if it should be +2 instead of +4
21:20:31  <rphillips>that is 3717 bytes
21:21:01  <rphillips>length is 7793
21:21:07  <creationix>so encoding is hex + ‘\r\n’ + data + ‘\r\n’ ?
21:21:31  <rphillips>correct... but this chunk does not have the complete '\r\n' at the end of it
21:21:49  <creationix>ahh, so it’s invalid
21:21:58  <rphillips>it's a partial chunk
21:22:13  <creationix>well if it’s partial, then that condition should trigger
21:22:17  <creationix>that means to wait for more data
21:23:46  <rphillips>1e71\r\n{"... data ... ] packet ends in the middle of the json
21:24:26  <rphillips>it appears to be checking to see if chunk is the entire chunked payload
21:26:20  <rphillips>maybe a decodeChunkContinue
21:27:27  <creationix>so what is the problem? Do you never get more chunks after that?
21:27:40  <creationix>it’s supposed to return nothing when there is more data to be read before parsing
21:28:23  <creationix>I tested the endpoint with curl and it doesn’t complain, so I think the server is behaving properly
21:28:32  <rphillips>so the issue is the chunk that I have, is the beginning of the chunked payload
21:28:43  <rphillips>the next read will bring in the next chunk of data
21:29:00  <rphillips>the function is assuming the entire chunk of data is one read
21:29:02  <creationix>where is the code using http-codec?
21:29:20  <creationix>I’ll bet we’re not concatenating the leftover data when new data comes in
21:29:32  <rphillips>https://github.com/luvit/luvit/blob/master/deps/http.lua#L168
21:30:35  <creationix>hmm, that looks right https://github.com/luvit/luvit/blob/master/deps/http.lua#L166
21:31:50  <creationix>I still don’t understand the problem. Does it not finally emit the chunk when the rest of it is read and concatenated?
21:32:54  <creationix>I suppose the decoder could emit what it has already and remember how much it’s already emitted
21:33:10  <creationix>currently you won’t get anything till the entire chunk is read
21:33:58  <rphillips>hmm
21:34:32  <rphillips>the 'extra' data is the chunked payload... it goes around the loop and back into decode and the state is decodeChunk
21:34:55  <rphillips>basically, the chunkedDecode isn't streaming
21:35:15  <rphillips>decodeChunked*
21:36:04  <creationix>https://github.com/luvit/luvit/blob/master/tests/test-http-decoder.lua#L167-L189
21:36:19  <creationix>it emits the chunks one at a time, but it doesn’t emit partial chunks, it will buffer till a chunk is done
21:37:26  <rphillips>hmm. checking
21:37:59  <creationix>this behavior can be changed if we want to stream even smaller chunks as they come in
21:38:14  <creationix>I don’t see any danger in non buffering till we get full chunks
21:39:27  <rphillips>hmm. ok
21:40:18  <creationix>so is it working as designed or is not working like I described? Also would you like the behavior changed to buffer less?
21:43:18  <rphillips>so, I don't see the extra 'data' event to read the next block of data
21:43:39  <creationix>that’s probably your root problem
21:43:57  <creationix>if all your data was emitting, you wouldn’t notice the decoder was buffering to whole chunks
21:50:13  <rje>rphillips: heading back from lunch, had a meeting with andrew at lunch
21:51:00  <rphillips>np
22:01:21  <rphillips>So that is the problem
22:01:36  <rphillips>The stream is getting closed
22:03:45  * SouL_quit (Remote host closed the connection)
22:45:41  <rje>rphillips, so was that sorted or is there still an issue?
22:58:04  * bjornquit (Remote host closed the connection)
23:01:05  <creationix>dinner time. I talked with ken and got lots of feedback for the new luvit site, still working on it.
23:22:17  <rphillips>rje: still working it
23:23:31  <rje>rphillips, is the data there and its not getting emitted by http.lua?
23:28:47  <rphillips>the _done() call destroys the socket
23:29:25  <rphillips>https://github.com/luvit/luvit/blob/master/deps/http.lua#L329
23:39:48  <rje>rphillips, should it not?
23:41:49  <rphillips>so _done() fires the write of the headers and payload, but doesn't wait for the response
23:41:59  <rphillips>currently
23:42:06  <rje>right
23:42:09  <rphillips>https://github.com/joyent/node/blob/master/lib/_http_outgoing.js#L504
23:42:43  <rphillips>it should probably just write the data at that point... but the changes I made aren't working
23:45:22  <rje>rphillips, can we work up a unit test that spaces out the io?
23:53:17  <rphillips>yeah, let me see
23:57:24  <rphillips>rje: https://github.com/luvit/luvit/blob/http_post_error/tests/test-http-post-1mb.lua#L44
23:57:30  <rphillips>try changing this line to write data