13:02:15 <dtroyer> #startmeeting openstackclient
13:02:16 <openstack> Meeting started Thu Jun 16 13:02:15 2016 UTC and is due to finish in 60 minutes.  The chair is dtroyer. Information about MeetBot at http://wiki.debian.org/MeetBot.
13:02:17 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
13:02:19 <openstack> The meeting name has been set to 'openstackclient'
13:02:25 <dtroyer> Who is here for OSC today?
13:02:53 <dhellmann> o/
13:03:40 <rtheis> o/
13:04:28 <dtroyer> Well, 33% increase over last week :)
13:04:35 <dtroyer> out of context that sounds great!
13:05:19 <dtroyer> #topic 3.0 reviews
13:05:41 <dtroyer> The initial KSA review and almost all of the osc-lib reviews have merged
13:06:08 <dtroyer> tangchen's IP commands are getting close
13:07:45 <dtroyer> https://review.openstack.org/#/c/325593/ is a re-work of our release notes to use a major-version organization rather than release cycle.
13:08:13 <dtroyer> dhellmann is looking at an enhancement to reno that will make this a bit easier, but what we have looks like it will work until then
13:08:25 <rtheis> lgtm
13:09:00 <dtroyer> specifically specifying a range of versions to include by adding a latest-version directive
13:09:04 <dhellmann> approved
13:09:53 <dhellmann> yes, that's on our to-do list, though it doesn't have a high priority right now so it may be a little while before we get to it
13:10:00 <dhellmann> our == the release team
13:10:01 <dtroyer> anything else specific to the 3.x changes?  I know there are a couple more ksa-related things, after which I will osc-lib-ify that too
13:10:22 <dtroyer> thanks dhellmann, no hurry here
13:10:33 <rtheis> nothing else
13:12:17 <dtroyer> I should note, also, in the release note change series I added some directives to simplify bug and doc pointers (back in https://review.openstack.org/#/c/325571).
13:13:09 <dtroyer> I don't recall if I've mentioned it in a meeting yet, this gets us out of needing full URLs to point back into our docs, and simplifies bug pointers
13:13:33 <dtroyer> #topic reviews
13:13:39 <dhellmann> those custom roles don't translate well to the text version of the release notes used for email announcements, though :-/
13:13:42 <dtroyer> Any other reviews we should talk about?
13:14:19 <rtheis> dtroyer: since they don't translate well should we still use them in the release notes?
13:14:29 <dtroyer> ah, I didn't know that… should it be different?  I didn't want hard-coded URLs in the doc to be able to link back to our dev docs
13:14:42 <dhellmann> it's a common enough thing to want to do, so it might be something we want to support in reno
13:15:06 <dhellmann> but yeah, I remember stevemar pointing out issues in the last release announcement email
13:15:38 <dhellmann> http://lists.openstack.org/pipermail/openstack-announce/2016-June/001183.html
13:15:41 <dtroyer> I'll go back and look at it, I didn't read the notes in the email closely
13:15:57 <rtheis> I had one review: https://review.openstack.org/#/c/330223/
13:16:06 <rtheis> Sounds like some good stuff here
13:16:24 <rtheis> Do we play to leverage that in OSC to address the output formatting issues
13:16:28 <rtheis> *plan
13:16:53 <dhellmann> rtheis : that's the idea, this was a POC as much as anything
13:17:07 <rtheis> ok
13:17:47 <dtroyer> the driver was being able to format human-readable and machine-readable separately, specifcally for the formatters that do not have intelligent dump methods (shell, value, etc)
13:18:12 <dtroyer> but it would also encapsulate some of the other common things we do
13:18:34 * dhellmann notes that cliff 2.1.0 is building
13:18:43 <dtroyer> dhellmann: would you see some of the generic FormattableColumn subclasses going into cliff too?
13:18:54 <dhellmann> dtroyer : sure
13:19:21 <dtroyer> ok, I'll plan for that as we develop them
13:19:23 <dhellmann> I'm certain we could come up with some generic things like ObjectWithNameAndID or whatever
13:19:38 <dhellmann> yeah, start with them in OSC and then move them to cliff as we figure out what's generic
13:20:15 <dtroyer> the ones that are osc-specific will go to osc-lib, but we'll skip that step for generic ones then
13:21:16 <dhellmann> well, wherever it makes sense to develop them out of cliff is fine, I just meant you'd iterate faster without needing releases of libs
13:21:49 <dhellmann> once we figure out what the base class API needs to look like, we'll need to write docs in cliff too
13:22:55 <dtroyer> other reviews?
13:23:08 <rtheis> nothing else from me
13:23:18 <dtroyer> #topic bugs
13:23:30 <dtroyer> I don't have any specific bugs to discuss...
13:24:09 <rtheis> me either
13:24:34 <dtroyer> #topic open discussion
13:24:39 <dtroyer> What else is going on?
13:25:49 <rtheis> Slow progress on neutronclient supporting OSC plugin from me, but amotoki was planning to pick it back up
13:26:41 <dtroyer> rtheis: I saw something from briancurtin about more refactoring in the SDK, do you know how much he las left to do?
13:27:02 <rtheis> He has compute out for review.  I am hoping to look at that this week
13:27:28 <rtheis> Once compute is worked out then hopefully things can proceed quickly with the others
13:27:32 <briancurtin> dtroyer: unfortunately a lot, but it'll accelerate as I do more. working on image now
13:27:56 <dtroyer> ok, thanks briancurtin.  image is the one I am anxious to try out ;)
13:28:14 <dtroyer> getting rid of glanceclient is high on my list due to its specific depenencies
13:30:06 <dtroyer> We have had a couple of reviews for stable branches that are not critical or security things… I think I need to write down our release policy more clearly somewhere…
13:30:46 <rtheis> sounds good
13:31:01 <dtroyer> ok, if nothing else, it's time for breakfast…
13:31:33 <dtroyer> Thanks guys
13:31:34 <rtheis> nothing else
13:31:35 <rtheis> thanks
13:31:38 <dtroyer> #endmeeting