| *** etoews has joined #openstack-sdks | 00:13 | |
| *** etoews has quit IRC | 00:18 | |
| *** bknudson has joined #openstack-sdks | 00:21 | |
| *** etoews has joined #openstack-sdks | 00:32 | |
| *** etoews has quit IRC | 00:37 | |
| *** HenryG has joined #openstack-sdks | 01:21 | |
| *** samchoi has quit IRC | 01:38 | |
| *** etoews has joined #openstack-sdks | 02:01 | |
| *** etoews has quit IRC | 02:05 | |
| *** etoews has joined #openstack-sdks | 02:18 | |
| *** etoews has quit IRC | 02:23 | |
| *** bknudson has quit IRC | 02:24 | |
| *** wchrisj has quit IRC | 04:27 | |
| *** wchrisj has joined #openstack-sdks | 05:11 | |
| *** wchrisj has quit IRC | 05:28 | |
| *** sharwell__ has quit IRC | 05:29 | |
| *** sharwell__ has joined #openstack-sdks | 05:30 | |
| *** jamie_h_ has joined #openstack-sdks | 09:47 | |
| *** mfer has joined #openstack-sdks | 10:24 | |
| *** jamielennox is now known as jamielennox|away | 11:15 | |
| *** krames has quit IRC | 12:44 | |
| *** krames has joined #openstack-sdks | 13:16 | |
| *** bknudson has joined #openstack-sdks | 13:45 | |
| *** etoews has joined #openstack-sdks | 13:48 | |
| *** wchrisj has joined #openstack-sdks | 14:06 | |
| *** openstack has joined #openstack-sdks | 14:21 | |
| wchrisj | elight: I hit a snag last night that you may be able to assist with... | 14:30 |
|---|---|---|
| *** krames has quit IRC | 14:48 | |
| *** etoews has quit IRC | 14:49 | |
| *** krames has joined #openstack-sdks | 14:51 | |
| *** rgbkrk has joined #openstack-sdks | 15:19 | |
| *** rgbkrk_ has joined #openstack-sdks | 15:22 | |
| *** rgbkrk has quit IRC | 15:23 | |
| *** etoews has joined #openstack-sdks | 15:23 | |
| *** rgbkrk_ has quit IRC | 15:24 | |
| *** rgbkrk has joined #openstack-sdks | 15:25 | |
| *** etoews has quit IRC | 15:28 | |
| *** etoews has joined #openstack-sdks | 15:43 | |
| *** etoews has quit IRC | 15:47 | |
| *** krames has quit IRC | 16:04 | |
| *** rgbkrk has quit IRC | 16:20 | |
| wchrisj | elight: krames: anyone got a few minutes to look at something with me? | 16:22 |
| *** HenryG has quit IRC | 16:44 | |
| *** HenryG has joined #openstack-sdks | 16:45 | |
| *** bknudson has left #openstack-sdks | 17:34 | |
| *** samchoi has joined #openstack-sdks | 17:35 | |
| *** krames has joined #openstack-sdks | 17:36 | |
| *** bknudson has joined #openstack-sdks | 17:46 | |
| *** krames has quit IRC | 17:51 | |
| *** krames has joined #openstack-sdks | 17:57 | |
| *** jamie_h_ has quit IRC | 18:35 | |
| *** jamie_h_ has joined #openstack-sdks | 19:06 | |
| edleafe | OK, I know it's Friday afternoon for most people, but I want to put this out there as a starting point for some more discussion about the Python SDK. | 19:21 |
| edleafe | I have a Blueprint registered at https://blueprints.launchpad.net/unifiedsdk/+spec/overall-class-design, which links to the wiki page: https://wiki.openstack.org/wiki/PythonOpenStackSDK/ClassDesignDiscussion#Flexibility. | 19:21 |
| edleafe | I'd like to have everyone review this, and possibly comment/update before next week's meeting. | 19:21 |
| *** jamie_h_ has quit IRC | 20:09 | |
| *** rgbkrk has joined #openstack-sdks | 20:18 | |
| bknudson | edleafe: one issue we had with the 'managers' classes in current clients -- | 20:24 |
| bknudson | they have a reference back to the client | 20:24 |
| bknudson | so now the client isn't collected easily | 20:25 |
| bknudson | and connections stay open | 20:25 |
| bknudson | maybe jamielennox|away could explain it better. | 20:25 |
| edleafe | bknudson: my current client does that, too, because the HTTP stuff was in the client | 20:25 |
| edleafe | once I separated it out, it really wasn't necessary | 20:25 |
| edleafe | the client initialized the manager with the info it needed | 20:26 |
| bknudson | ok. | 20:26 |
| bknudson | one of the proposed solutions was weakrefs from manager to client | 20:26 |
| edleafe | why would that be necessary? | 20:26 |
| bknudson | then when the client was released it could be cleaned up | 20:27 |
| edleafe | most of the use cases I've seen have been long-lived: clients and managers created early in the app's life, and persist while the app does | 20:28 |
| edleafe | I haven't seen a case where they are created/destroyed repeatedly | 20:28 |
| bknudson | this was affecting Horizon. | 20:29 |
| bknudson | I should be able to find the bug... | 20:29 |
| bknudson | https://bugs.launchpad.net/python-keystoneclient/+bug/1282089 | 20:30 |
| edleafe | Ah, well, identity is really a separate case, especially for Horizon | 20:32 |
| edleafe | but managers really shouldn't need a reference to the client if they are created properly | 20:33 |
| bknudson | edleafe: is resource going to need a reference to the client, to do resource.some_method() ? | 20:33 |
| edleafe | They will need to have the reference to the manager | 20:34 |
| edleafe | IOW, client and resource call the same method of manager | 20:35 |
| edleafe | resource simply passes itself as the entity to act on | 20:35 |
| elight | wchrisj: Sorry. Been doing outreach all day today | 20:44 |
| wchrisj | elight: np | 20:44 |
| wchrisj | elight: Do you have a few minutes before the end of the day? I need to overcome an issue in order to get velocity back... | 20:45 |
| wchrisj | elight: or are you outta here? | 20:45 |
| wchrisj | ;-) | 20:45 |
| wchrisj | elight: understandable if so... | 20:46 |
| elight | Give me a few minutes... | 20:46 |
| wchrisj | elight: yeah, I'm in a mtg ATM too | 20:47 |
| wchrisj | elight: ping me when you are free | 20:47 |
| elight | wchrisj: Ugh. When is your meeting over? | 20:47 |
| wchrisj | elight: I can bail anytime | 20:47 |
| wchrisj | ;-) | 20:47 |
| mfer | samchoi ycombinator glenc jamie_h note, the first PHP meeting will be April 9th | 20:53 |
| mfer | I updated the wiki | 20:53 |
| mfer | ycombinator thanks for pulling all those things together | 20:53 |
| samchoi | Thanks for the heads up. It makes it difficult for me to attend meetings if the invite is sent out very late since I usually begin work after our meeting time... | 20:54 |
| ycombinator | thanks mfer | 20:55 |
| ycombinator | samchoi: you should've received a recurring google invite from me | 20:55 |
| samchoi | yep, got the most recent one, thanks ycombinator | 20:55 |
| ycombinator | mfer: I'll be on vacation that week but I'll try to hide from my wife and get on IRC :) | 20:56 |
| mfer | ycombinator ha. really, don't do that | 20:56 |
| samchoi | mfer: that sounds like a Rajeev thing to do | 20:56 |
| mfer | ycombinator is there a meeting calendar for all openstack meetings somewhere? | 20:56 |
| ycombinator | mfer: yes, one secon | 20:57 |
| ycombinator | second* | 20:57 |
| ycombinator | mfer: https://www.google.com/calendar/ical/bj05mroquq28jhud58esggqmh4@group.calendar.google.com/public/basic.ics | 20:57 |
| ycombinator | not sure if thierry has had a chance to update it with ours yet | 20:57 |
| mfer | ycombinator who manages that one? | 20:59 |
| ycombinator | mfer: Thierry Carrez | 20:59 |
| mfer | thanks | 20:59 |
| ycombinator | looks like he did our meeting to it | 20:59 |
| ycombinator | add our* | 20:59 |
| mfer | ok, i need to run to dinner. have a good night | 20:59 |
| mfer | ycombinator i see that. thanks | 20:59 |
| *** mfer has quit IRC | 21:00 | |
| *** mfer has joined #openstack-sdks | 21:00 | |
| *** mfer has quit IRC | 21:01 | |
| *** krames has quit IRC | 21:13 | |
| *** etoews has joined #openstack-sdks | 21:21 | |
| *** etoews has quit IRC | 21:25 | |
| *** etoews has joined #openstack-sdks | 21:39 | |
| *** etoews has quit IRC | 21:44 | |
| *** wchrisj has quit IRC | 22:06 | |
| *** bknudson has quit IRC | 22:06 | |
| *** rgbkrk has quit IRC | 22:45 | |
| *** dhellmann is now known as dhellmann_ | 23:13 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!