18:59:56 #startmeeting python-openstacksdk 18:59:57 Meeting started Tue Nov 24 18:59:56 2015 UTC and is due to finish in 60 minutes. The chair is briancurtin. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:59:58 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 19:00:01 The meeting name has been set to 'python_openstacksdk' 19:00:22 o/ 19:00:41 o/ 19:00:49 o/ 19:01:44 i don’t have any specific agenda here, so it’s fairly open. 0.7.1 just went out and i’m upgrading the rackspace extension to work with KSA for the auth plugin, and putting out a release of that with two service extensions, which should gain some users (though mostly internal people) 19:03:06 terrylhowe: anything you want to cover here? i’ve seen some chatter about OSC using SDK’s network stuff 19:03:06 I’m working on the integration with osc, seems to be working 19:03:37 I’ll probably remove the ‘do not merge’ soon 19:05:42 i've been working on getting the docs published to developer.openstack.org https://review.openstack.org/#/c/246591/ 19:06:08 one thing that came up talking to annegentle is how do we handle doc versions? 19:06:34 which begs the question, what are our requirements? 19:06:49 i don’t understand either of those questions 19:07:11 so we agreed to publish to developer.openstack.org/sdks/python/openstacksdk 19:07:16 er, maybe on the first one, build the docs by tag? 19:07:25 right 19:07:45 is that something infra can do? 19:07:58 do we want to have a docs published for every version we release? 19:08:21 i think infras capabilities might be limited here but i haven't really dug into it yet. 19:08:31 I would think latest would be best 19:08:47 atm, only a couple releases per cycle 19:08:49 i would hope we can get docs for probably every X.Y, and that is always the latest in the X.Y series 19:09:55 briancurtin: i'm not 100% sure what you mean by "that is always the latest in the X.Y series" 19:09:58 example? 19:10:37 right now we have 0.7.0 and 0.7.1 versions out there. docs.blah/sdk/python/0.7 would be the 0.7.1 docs. when 0.7.2 comes out, it would be those 19:10:54 yep. that sgtm. 19:11:19 if that is doable, sounds better than just latest 19:11:29 so we would always have /0.6 /0.7 /1.0 /1.1 etc. 19:11:39 and maybe a /latest ??? 19:11:49 in an idea world, yes to all of that 19:11:57 that's what i would like too. no idea if it's possible. 19:12:12 I’m guessing, based on what i remember, that we only get latest right now 19:12:53 or that you can build based on *branch* not tag (because some reason) 19:13:03 that's what i haven't dug into yet. 19:13:52 but if we can only get latest for now, do we consider that good enough until more publishing options come along? 19:15:00 yeah, as i think right now latest is all anyone probably wants. once we start moving on down the road, stability and pinning will matter more. someone’s going to want to hang onto 1.3 19:15:43 we probably also need to have stable branches, or something like that 19:17:16 okay. i'll see what's possible. actually getting things out there will likely have to wait until annegentle gets back next week because of the theme change https://review.openstack.org/#/c/247059 19:18:19 i think we can just make that requirement change and push it through 19:19:04 if it's urgent, sure. 19:21:58 etoews: i just pushed https://review.openstack.org/#/c/247059/ which i think completes it 19:22:29 k 19:25:32 anything else? 19:25:56 i don’t really have anything. plugging away at this extension thing for right now now that the release is out 19:26:36 I’m going to guess we’ll hit a slowdown the rest of the week, at least thurs/fri 19:27:29 sounds good 19:27:56 anything to review in https://launchpad.net/python-openstacksdk/+milestone/1.0 19:30:06 there’s https://bugs.launchpad.net/python-openstacksdk/+bug/1467732 which has https://review.openstack.org/#/c/194767/ (pretty complete) and https://review.openstack.org/#/c/242694/ (a potentially different direction, or rather extension of that one that will allow it to be used for both types of metadata) 19:30:06 Launchpad bug 1467732 in OpenStack SDK "Improve server metadata usability" [Medium,In progress] - Assigned to Brian Curtin (brian.curtin) 19:30:38 should probably move the wishlist stuff out of M1 (except for the occ one). 19:31:08 ah, yeah. i had a separate query i’ve been looking at that excludes it 19:31:33 (it gets rid of the released/committed stuff so it’s more of a todo list) 19:32:18 ya. launchpad is difficult by default. 19:32:34 ok, those are all out of m1 19:33:49 so if it's down to confirmed and in progress we're in reasonably good shape. 19:34:11 still trying to get 1.0 out before the end of the year? 19:34:45 i hope so 19:35:28 ++ 19:36:31 does anyone know Richard Theis? 19:36:45 sounds good, I think just play with it a bit more and make sure things are working. Qiming was having some issues with selin (sp?) client today, but I think it had dependencies on the old transport 19:36:59 the remaining Confirmed bugs seem relatively easy to get through. of course things are going to pop up as we complete the rest of this stuff, but assuming relatively the same amount of activity from everyone, i think end of year is probably doable 19:38:04 ah. i see he's in #openstack-sdks. 19:38:58 he is interested in contributing to osc for network stuff and he has been contributing around 19:39:14 that's awesome. 19:39:34 terrylhowe: yeah, I’m hoping some of the work i’m doing for the rackspace extension should put some good mileage on it. i had a guy in our UK office enable support for our Cloud Backup and Cloud Databases stuff, and he’s working on a few other things we offer. his team wants to start using SDK for some tools 19:43:54 This osc stuff should exercise at least the network stuff 19:44:23 and that’s probably the biggest set of APIs we have, so that’s pretty good 19:47:33 #endmeeting