| *** jamielennox|away is now known as jamielennox | 00:51 | |
| *** jamie_h has joined #openstack-sdks | 07:24 | |
| *** jamie_h has quit IRC | 07:25 | |
| *** openstackgerrit has quit IRC | 07:53 | |
| *** jamielennox is now known as jamielennox|away | 08:28 | |
| *** jamielennox|away is now known as jamielennox | 10:34 | |
| *** openstackgerrit has joined #openstack-sdks | 13:37 | |
| *** openstackgerrit has quit IRC | 13:51 | |
| *** terrylhowe has joined #openstack-sdks | 14:20 | |
| *** TravT has joined #openstack-sdks | 14:40 | |
| *** wchrisj has joined #openstack-sdks | 14:51 | |
| *** TravT has quit IRC | 15:05 | |
| *** TravT has joined #openstack-sdks | 15:05 | |
| *** TravT has quit IRC | 15:05 | |
| *** TravT has joined #openstack-sdks | 15:06 | |
| wchrisj | elight Good morning! | 15:08 |
|---|---|---|
| elight | wchrisj: hey, man | 15:09 |
| wchrisj | elight Was getting specs in shape around compute, and thinking about the change to RequestCommon we discussed last week. | 15:10 |
| elight | wchrisj: yeah? | 15:10 |
| wchrisj | elight Were you thinking of a standard ruby class instance, or a Singleton approach? | 15:10 |
| elight | wchrisj: Module with module methods is my gut feeling because it's stateless. | 15:12 |
| wchrisj | elight: is that going to impact the way that requests are handled? those are mixd into the service class... | 15:13 |
| elight | wchrisj: it would in that the services would need to send a message to the module explicitly rather than implicitly. | 15:13 |
| elight | wchrisj: perhaps I'm being too much of a purist in that regard. I'm growing shy of mixins. But I think it's more to do with how they're abused. | 15:14 |
| elight | wchrisj: the Core Connection change should help here. We may be able to finally kill request_params. | 15:14 |
| wchrisj | elight: understood. just want to make sure we arent overomplicating things | 15:14 |
| elight | wchrisj: agreed | 15:15 |
| wchrisj | elight: at what level would we send that message? in the request? in the service class? | 15:15 |
| elight | wchrisj: I meant extracting the mixin into a service so that calls to "request" are clearly to a different abstraction. The services shouldn't know how to send HTTP requests but instead ask another object to do it for them. | 15:18 |
| elight | wchrisj: sorry. Overloaded word: service. | 15:18 |
| elight | The Fog Service classes are almost God objects already. | 15:19 |
| wchrisj | elight: agreed | 15:19 |
| wchrisj | elight OK | 15:20 |
| elight | wchrisj: it's probably easier if I write a gist, for clarity, and link it on the issue. | 15:20 |
| wchrisj | That's fine - I think I understand better now what you were thinking in terms of a ServiceRequest (my name for it) | 15:21 |
| wchrisj | elight ^^ | 15:21 |
| wchrisj | elight: A gist would be good, actually | 15:27 |
| *** openstackgerrit has joined #openstack-sdks | 15:43 | |
| elight | wchrisj: Odd. I had thought that I opened an issue about RequestCommon... | 16:26 |
| *** mhagedorn has joined #openstack-sdks | 16:37 | |
| *** openstackgerrit has quit IRC | 17:20 | |
| *** openstackgerrit has joined #openstack-sdks | 17:25 | |
| briancurtin | terrylhowe: looking at the "Authentication from keystoneclient" change. save for the pep8 fail, is this ready to be reviewed? The first thing i noticed in openstack/auth/access.py is a fixme for a deprecated arg | 18:27 |
| terrylhowe | briancurtin, I have fixed the pep8 errors, but I was trying to get some better tests before uploading a patch | 18:27 |
| terrylhowe | I attempted not to change anything but I what I had to change from keystoneclient, so there are some fixme and things like that in there | 18:28 |
| *** TravT has quit IRC | 18:51 | |
| *** TravT has joined #openstack-sdks | 18:59 | |
| *** stevemar has joined #openstack-sdks | 19:09 | |
| stevemar | terrylhowe ping | 19:10 |
| terrylhowe | what's up stevemar ? | 19:12 |
| stevemar | had a question about https://review.openstack.org/#/c/89609/1/openstackclient/volume/v1/type.py | 19:12 |
| stevemar | terrylhowe, taking a closer look at cinder client, i guess set_keys doesn't return the full volume_type object? | 19:13 |
| stevemar | just the keys? | 19:13 |
| terrylhowe | yeh, that is the way I remember it. | 19:14 |
| terrylhowe | the create doesn't have properties set because you can't set properties there | 19:14 |
| terrylhowe | https://github.com/openstack/python-cinderclient/blob/master/cinderclient/v2/volume_types.py#L44 | 19:16 |
| stevemar | terrylhowe, i get that, i'm just wondering why set_keys only returns the properties | 19:18 |
| *** etoews has joined #openstack-sdks | 19:30 | |
| terrylhowe | logically, it makes sense to me, what does not make sense to me is why they implemented it that way | 19:31 |
| *** mhagedorn has quit IRC | 19:36 | |
| *** wchrisj has quit IRC | 19:47 | |
| *** wchrisj has joined #openstack-sdks | 20:04 | |
| *** mhagedorn has joined #openstack-sdks | 20:11 | |
| *** wchrisj has quit IRC | 21:15 | |
| *** wchrisj has joined #openstack-sdks | 21:17 | |
| *** jamielennox is now known as jamielennox|away | 21:23 | |
| *** wchrisj has quit IRC | 21:41 | |
| *** wchrisj has joined #openstack-sdks | 21:51 | |
| *** wchrisj has quit IRC | 21:52 | |
| *** wchrisj has joined #openstack-sdks | 21:53 | |
| *** wchrisj has quit IRC | 22:01 | |
| *** wchrisj has joined #openstack-sdks | 22:04 | |
| *** wchrisj has quit IRC | 22:06 | |
| *** wchrisj has joined #openstack-sdks | 22:27 | |
| wchrisj | hey elight - did you have a chance to do that gist for me today? | 23:00 |
| elight | wchrisj: afraid not. first thing tomorrow morning. | 23:02 |
| wchrisj | elight - thanks! | 23:02 |
| wchrisj | elight - realized when I started down the path that the existing RequestCommon is already a module with module methods; just being mixed in. | 23:04 |
| *** etoews has quit IRC | 23:10 | |
| *** samchoi has joined #openstack-sdks | 23:13 | |
| *** jamielennox|away is now known as jamielennox | 23:16 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!