| 2010-08-20 07:33:32 utc | tosch_le | hello folks! |
| 2010-08-20 07:33:54 utc | jmettraux | tosch_le: hello ! |
| 2010-08-20 07:34:30 utc | tosch_le | jmettraux: do we have a release day today or is there still anything to do?# |
| 2010-08-20 07:34:59 utc | tosch_le | (i'll be on holidays the next 10 days) |
| 2010-08-20 07:35:07 utc | jmettraux | good for you |
| 2010-08-20 07:35:13 utc | tosch_le | :-) |
| 2010-08-20 07:35:36 utc | jmettraux | I still need to finish adapt ruote-beanstalk |
| 2010-08-20 07:35:57 utc | jmettraux | then run the tests (extensive) for ruote-dm, ruote-redis, ruote-beanstalk |
| 2010-08-20 07:36:16 utc | jmettraux | ruote-couch looks green, except for a few timing issues |
| 2010-08-20 07:36:20 utc | jmettraux | 2 |
| 2010-08-20 07:37:42 utc | tosch_le | ah, you're working hard, i see. i'll try to update ruote-on-rails today and will sent some love to rk-client. |
| 2010-08-20 07:37:51 utc | tosch_le | s/sent/send/ |
| 2010-08-20 07:37:52 utc | jmettraux | oh cool |
| 2010-08-20 07:38:21 utc | tosch_le | and honestly: i'm no bot ;-) |
| 2010-08-20 07:38:26 utc | jmettraux | :-) |
| 2010-08-20 07:38:41 utc | jmettraux | http://ruote.rubyforge.org/participants.html I documented engine.register { catchall } and co |
| 2010-08-20 07:41:02 utc | tosch_le | great, thanks |
| 2010-08-20 07:47:43 utc | jmettraux | it's great if you update ruote-on-rails and rk-client |
| 2010-08-20 07:57:46 utc | tosch_le | hmm, http://github.com/tosch/ruote-on-rails/blob/master/lib/tasks/ruote_kit.rake won't work anymore |
| 2010-08-20 07:58:49 utc | jmettraux | symptom ? |
| 2010-08-20 07:58:54 utc | jmettraux | aaah |
| 2010-08-20 07:59:27 utc | tosch_le | run_worker! isn't anymore… |
| 2010-08-20 07:59:44 utc | tosch_le | run_worker is there, but it doesn't join the engine's thread |
| 2010-08-20 08:00:13 utc | jmettraux | RuoteKit.run_worker(storage) |
| 2010-08-20 08:00:14 utc | tosch_le | so that the engine is shutdown just after is has been instanciated – when the rake task is done |
| 2010-08-20 08:00:33 utc | jmettraux | do you pass the storage ? |
| 2010-08-20 08:01:41 utc | tosch_le | i'll do that, though i don't know yet how to solve that in a convenient way. i'd like to have all the config options in one place (http://github.com/tosch/ruote-on-rails/blob/master/config/initializers/ruote_kit.rb) |
| 2010-08-20 08:02:51 utc | tosch_le | the initializer is run in the rake task, too. so RuoteKit.engine will be already set, but without the worker. |
| 2010-08-20 08:02:55 utc | jmettraux | nice challenge |
| 2010-08-20 08:03:42 utc | jmettraux | in my config/initializers/ruote.rb |
| 2010-08-20 08:03:52 utc | jmettraux | I have something like $storage = xyz |
| 2010-08-20 08:03:59 utc | tosch_le | thinking about either a RUOTE_STORAGE constant or doing something like RuoteKit.engine = Ruote::Engine.new(Ruote::Worker.new(RuoteKit.engine.context.storage))) |
| 2010-08-20 08:04:34 utc | jmettraux | your second variant will not work |
| 2010-08-20 08:04:40 utc | tosch_le | i supposed that |
| 2010-08-20 08:04:58 utc | tosch_le | but that would be a nice one :-) |
| 2010-08-20 08:05:25 utc | jmettraux | e = E.new(W.new(e.context.storage)) |
| 2010-08-20 08:05:50 utc | jmettraux | initially, e is nil |
| 2010-08-20 08:06:03 utc | jmettraux | e.context will break |
| 2010-08-20 08:06:06 utc | tosch_le | no, it's set in the initializer |
| 2010-08-20 08:06:22 utc | jmettraux | aaaaaaah |
| 2010-08-20 08:06:28 utc | tosch_le | http://github.com/tosch/ruote-on-rails/blob/master/config/initializers/ruote_kit.rb will be run before the rake task is run |
| 2010-08-20 08:06:55 utc | tosch_le | (and in the initializer, there will be a call RuoteKit.engine = foo) |
| 2010-08-20 08:07:05 utc | jmettraux | RuoteKit.configure is dead |
| 2010-08-20 08:07:28 utc | tosch_le | yeah, the initializer is to be reworked. that's what i'm doing right now |
| 2010-08-20 08:08:18 utc | tosch_le | it will look more like that: http://gist.github.com/539864 |
| 2010-08-20 08:08:35 utc | jmettraux | I have something like this : http://gist.github.com/539865 |
| 2010-08-20 08:08:40 utc | jmettraux | nice |
| 2010-08-20 08:08:47 utc | jmettraux | n + 1 for me |
| 2010-08-20 08:09:48 utc | tosch_le | i like your version more |
| 2010-08-20 08:10:19 utc | tosch_le | thanks for sharing that |
| 2010-08-20 08:11:04 utc | jmettraux | thanks for ruote-on-rails ! |
| 2010-08-20 08:11:40 utc | tosch_le | you're running the worker along with the rails app? do you use passenger? |
| 2010-08-20 08:12:03 utc | jmettraux | thin |
| 2010-08-20 08:12:10 utc | jmettraux | with nginx in front |
| 2010-08-20 08:13:14 utc | tosch_le | ok, that way you have no worries the app will be stopped after some time. |
| 2010-08-20 08:13:54 utc | tosch_le | i'd like to have the worker in a separate instance to be passenger-friendly |
| 2010-08-20 08:14:01 utc | jmettraux | :-) |
| 2010-08-20 08:14:20 utc | jmettraux | makes 100% sense |
| 2010-08-20 08:15:50 utc | tosch_le | after a quick look into ruote's api, i'll use RuoteKit.engine.join within the rake task and everything will be fine :-) |
| 2010-08-20 08:18:04 utc | jmettraux | should we integrate that in RuoteKit.run_worker(storage) ? |
| 2010-08-20 08:18:23 utc | tosch_le | unsure. |
| 2010-08-20 08:18:32 utc | tosch_le | would make sense. |
| 2010-08-20 08:19:13 utc | tosch_le | i won't need run_worker in another way and i don't see other needs for that method anyway |
| 2010-08-20 08:20:52 utc | jmettraux | yeah, it's a bit artificial anyway |
| 2010-08-20 08:22:48 utc | jmettraux | RuoteKit.run_worker(STORAGE, :join => true) |
| 2010-08-20 08:23:15 utc | tosch_le | or drop run_worker entirely? |
| 2010-08-20 08:23:36 utc | tosch_le | would be consequent |
| 2010-08-20 08:24:07 utc | tosch_le | s/consequent/consistent |
| 2010-08-20 08:24:41 utc | jmettraux | in fact, the better way is |
| 2010-08-20 08:25:58 utc | jmettraux | RuoteKit.engine = Ruote::Engine.new(Ruote::Worker.new(STORAGE), false); RuoteKit.engine.worker.run |
| 2010-08-20 08:26:06 utc | jmettraux | which runs the worker in the current thread |
| 2010-08-20 08:26:31 utc | jmettraux | RuoteKit.run_worker could wrap that logic |
| 2010-08-20 08:27:12 utc | tosch_le | fine, let's go that way. |
| 2010-08-20 08:27:32 utc | jmettraux | or I could modify the engine initializer so that we can |
| 2010-08-20 08:27:41 utc | jmettraux | RuoteKit.engine = Ruote::Engine.new(Ruote::Worker.new(STORAGE), :join => true)) |
| 2010-08-20 08:28:08 utc | tosch_le | i like run_worker more. it's just more … explicit |
| 2010-08-20 08:28:15 utc | jmettraux | OK |
| 2010-08-20 08:28:20 utc | jmettraux | let me change that |
| 2010-08-20 08:42:02 utc | jmettraux | ok, first step http://github.com/jmettraux/ruote/commit/d0af2e0ab7498a4463c80f190b02246f3170c456 |
| 2010-08-20 08:44:21 utc | jmettraux | second step http://github.com/jmettraux/ruote-kit/commit/12652d3e6475404f6b31f54040e15e7b490c2238 |
| 2010-08-20 08:44:33 utc | jmettraux | OK if run_worker 'blocks' ? |
| 2010-08-20 08:45:35 utc | tosch_le | it's fine for me |
| 2010-08-20 08:46:50 utc | jmettraux | please tell me how those two commits work for you |
| 2010-08-20 09:30:16 utc | tosch_le | master branch of ruote-on-rails (rails2) is updated |
| 2010-08-20 09:30:31 utc | tosch_le | run_worker is working fine :-) |
| 2010-08-20 09:32:19 utc | jmettraux | ah great ! Many thanks ! |
| 2010-08-20 09:57:24 utc | tosch_le | out for lunch now, back for rails3 branch later :-) |
| 2010-08-20 10:01:27 utc | jmettraux | Guten Appetit ! |
| 2010-08-20 10:51:44 utc | tosch_le | back from lunch, rails3 branch of ruote-on-rails is ready for v2.1.11 :-) |
| 2010-08-20 11:49:41 utc | jmettraux | that means I have to hurry |
| 2010-08-20 12:13:04 utc | jmettraux | very nice, well done |
| 2010-08-20 14:46:47 utc | tosch_le | sorry, ruote-kit-client has to wait till after the holidays |
| 2010-08-20 14:47:07 utc | jmettraux | no worries, thanks anyway ! |
| 2010-08-20 14:47:28 utc | tosch_le | leaving now, have a good time and see you in september! |
| 2010-08-20 14:48:04 utc | jmettraux | ciao ! Enjoy the summer ! |