ruote tmp/log_2013-03-17.html

2013-03-17 08:21:53 utc phaeron jmettraux: https://github.com/MeeGoIntegration/ruote/commit/554fda0f9c60b0be13962c1ed650a5810a57a224
2013-03-17 08:22:06 utc phaeron jmettraux: github fails to create pull request
2013-03-17 08:24:54 utc jmettraux phaeron: hello
2013-03-17 08:26:29 utc jmettraux well, I'll do it manually
2013-03-17 08:28:30 utc jmettraux wait, I realize now that those D shouldn't be there
2013-03-17 08:28:43 utc jmettraux argh
2013-03-17 08:28:46 utc phaeron jmettraux: it goes into redirect loop (github)
2013-03-17 08:29:02 utc jmettraux yeah, sorry, I won't be able to debug github
2013-03-17 08:29:27 utc phaeron jmettraux: :) wasn't asking to
2013-03-17 08:29:39 utc phaeron I gave up on it after couple of tries
2013-03-17 08:33:00 utc jmettraux I removed this special logging: https://github.com/jmettraux/ruote/commit/f42f1181f074d5634ae59e5dbf104f4a27728923
2013-03-17 08:33:06 utc jmettraux sorry for the nuisance
2013-03-17 08:33:21 utc jmettraux no need to divert to /dev/null
2013-03-17 08:33:55 utc jmettraux I misunderstood you, I thought you wanted to use it out of the blue, I didn't notice it went to master
2013-03-17 08:34:01 utc jmettraux sorry again
2013-03-17 08:34:29 utc phaeron jmettraux: oh , that's even better , I was thinking the logger will have to do a lot of work and then it goes to /dev/null
2013-03-17 08:34:39 utc phaeron thanks :)
2013-03-17 08:35:31 utc jmettraux you'r welcome
2013-03-17 08:35:34 utc jmettraux you're
2013-03-17 08:35:53 utc phaeron jmettraux: much better , but that doesn't remove the logger itself (which still writes to cwd)
2013-03-17 08:36:04 utc phaeron it might still benefit from setting logfile name
2013-03-17 08:36:28 utc jmettraux it's really about debugging
2013-03-17 08:36:35 utc jmettraux mostly only for me
2013-03-17 08:37:41 utc phaeron ok , no problems :)
2013-03-17 08:40:08 utc phaeron second thing is trying to make receiver.flunk take a trace to set , without changing the function signature
2013-03-17 08:42:58 utc jmettraux well,
2013-03-17 08:44:46 utc jmettraux simply pass an exception
2013-03-17 08:46:10 utc phaeron https://github.com/jmettraux/ruote/blob/master/lib/ruote/receiver/base.rb#L86
2013-03-17 08:46:33 utc jmettraux pass an exception
2013-03-17 08:46:55 utc phaeron aha
2013-03-17 08:47:00 utc phaeron sorry misunderstood
2013-03-17 08:47:04 utc phaeron will try that
2013-03-17 08:47:20 utc jmettraux exceptions have a #set_backtrace method
2013-03-17 08:48:01 utc jmettraux https://gist.github.com/anonymous/5180689
2013-03-17 08:49:44 utc phaeron yes that's what is used in ruote's flunk
2013-03-17 08:50:03 utc phaeron but ruote-amqp's flunk sends the class name
2013-03-17 08:50:09 utc phaeron I'll do the change there
2013-03-17 08:52:09 utc jmettraux https://github.com/kennethkalmer/ruote-amqp/blob/master/lib/ruote/amqp/receiver.rb#L152 says it sends the class and the message
2013-03-17 08:52:35 utc phaeron yeah I want the remote backtrace as well
2013-03-17 08:52:43 utc jmettraux ah
2013-03-17 08:53:32 utc jmettraux let's modify ruote-amqp and ruote so that err['trace'] is accepted
2013-03-17 08:53:57 utc jmettraux funny how people focus on the technical details before explaining the use case
2013-03-17 08:54:22 utc phaeron jmettraux: sorry :(
2013-03-17 08:54:30 utc jmettraux no worries
2013-03-17 08:54:49 utc jmettraux guilty as well
2013-03-17 08:54:56 utc phaeron I have a patch , will push it in a moment
2013-03-17 09:06:19 utc phaeron https://github.com/MeeGoIntegration/ruote-amqp/commit/5c911035c511a9e7a580ff231ba67a56cdd263f8
2013-03-17 09:06:19 utc jmettraux phaeron: ok, patch to ruote or ruote-amqp?
2013-03-17 09:06:26 utc phaeron ruote-amqp
2013-03-17 09:06:26 utc jmettraux ruote-amqp
2013-03-17 09:07:24 utc jmettraux sorry, I can't accept it
2013-03-17 09:07:33 utc jmettraux I actually also have a patch
2013-03-17 09:07:43 utc jmettraux this needs to be solved at the ruote level
2013-03-17 09:08:02 utc phaeron jmettraux: that's ok , as long as we solve it somewhere
2013-03-17 09:08:12 utc phaeron jmettraux: what did I do wrong ? :)
2013-03-17 09:08:32 utc jmettraux it's not wrong
2013-03-17 09:09:39 utc jmettraux the idea is good, I think it's better at the ruote level, not limited to ruote-amqp
2013-03-17 09:10:10 utc phaeron cool , no problems :)
2013-03-17 09:10:21 utc jmettraux running the test suite before pushing
2013-03-17 09:10:38 utc jmettraux it'll probably still need some ruote-amqp adjustment
2013-03-17 09:10:49 utc jmettraux hence my proposition of using err['trace']
2013-03-17 09:11:50 utc phaeron ( I had done it that way earlier , but you asked to send an exception instance :) )
2013-03-17 09:12:10 utc jmettraux I didn't know the whole story ;-)
2013-03-17 09:13:23 utc jmettraux please try to follow the target project's convention when submitting, I don't do object.method( x ), but object.method(x)
2013-03-17 09:13:24 utc jmettraux ;-)
2013-03-17 09:13:36 utc jmettraux also patch should come with tests/specs
2013-03-17 09:13:40 utc jmettraux patches
2013-03-17 09:14:32 utc jmettraux https://github.com/jmettraux/ruote/commit/b7eeff0d789bfe878b26e3139804744fcade864c
2013-03-17 09:15:03 utc jmettraux now need to adapt ruote-amqp, but dinner here
2013-03-17 09:15:35 utc phaeron looking
2013-03-17 09:18:13 utc phaeron yeah. I still like sending exception instance imho. more clean
2013-03-17 09:19:25 utc phaeron bon_appetit , breakfast here
2013-03-17 22:23:03 utc phaeron jmettraux: thanks for merging. can I get a pointer on how to run the specs ? :)
2013-03-17 22:23:14 utc phaeron jmettraux: I am still new to ruby ways
2013-03-17 22:23:18 utc jmettraux bundle exec rspec
2013-03-17 22:23:26 utc jmettraux hello, good evening
2013-03-17 22:26:09 utc phaeron jmettraux: was out for a long walk. just back :)
2013-03-17 22:26:28 utc jmettraux how many hours of sunlight these days?
2013-03-17 22:26:56 utc phaeron sunset is an almost natural 7pm :)
2013-03-17 22:29:51 utc jmettraux how is the new venture coming along?
2013-03-17 22:30:47 utc phaeron it's quite nice. Hopefully it works this time :)
2013-03-17 22:36:10 utc phaeron installing a fresh bundle to see about the spec
2013-03-17 22:36:54 utc jmettraux adapted the pull request and added a spec for it: https://github.com/kennethkalmer/ruote-amqp/commits/master
2013-03-17 22:36:58 utc jmettraux many thanks!
2013-03-17 22:40:00 utc phaeron I should learn this spec stuff , sounds important in ruby :)
2013-03-17 22:40:59 utc jmettraux tests in shiny clothes
2013-03-17 22:42:24 utc phaeron jmettraux: you don't like h['trace'] ? :(
2013-03-17 22:42:42 utc jmettraux what's the use case for it?
2013-03-17 22:43:06 utc jmettraux the err as Hash should be sufficient
2013-03-17 22:43:23 utc jmettraux a trace without at least an error message is, well, a trace
2013-03-17 22:43:25 utc phaeron same as err = h.delete('error')
2013-03-17 22:43:39 utc jmettraux look at the whole routine
2013-03-17 22:43:53 utc phaeron ack , I misread sorry
2013-03-17 22:44:21 utc jmettraux I understand, you want to pass "error" and "trace" together, but for now I accept a "error" Hash
2013-03-17 22:44:37 utc jmettraux do you guys really want to pass "error" and "trace" together at the top level?
2013-03-17 22:44:57 utc phaeron jmettraux: it was being abused as such for some time
2013-03-17 22:45:16 utc jmettraux ok, I did not know
2013-03-17 22:45:19 utc phaeron hash is cleaner of course, I will adapt the participant side
2013-03-17 22:45:29 utc phaeron no worries :)
2013-03-17 23:06:02 utc jmettraux if you need help to get into the test/spec first groove, I could take the time to prepare your next pull request with you
2013-03-17 23:11:14 utc phaeron jmettraux: that would be cool , once I find a new feature :)