19:00:22 <briancurtin> #startmeeting python-openstacksdk
19:00:22 <openstack> Meeting started Tue Aug 11 19:00:22 2015 UTC and is due to finish in 60 minutes.  The chair is briancurtin. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:00:23 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
19:00:26 <openstack> The meeting name has been set to 'python_openstacksdk'
19:00:35 <briancurtin> if you're here for the SDK meeting, say hi
19:00:44 <etoews> o/
19:02:38 <briancurtin> terrylhowe: you around?
19:03:14 <terrylhowe> o/
19:03:50 <briancurtin> i put together a small agenda but it's really just figuring out what to work on for releases - https://wiki.openstack.org/wiki/Meetings/PythonOpenStackSDK
19:03:55 <briancurtin> #topic Plans for next release, likely 0.6.2
19:04:18 <briancurtin> there's two things i want to try and get done and out the door:
19:04:26 <briancurtin> * Create should update all attributes returned in response (https://bugs.launchpad.net/python-openstacksdk/+bug/1474478)
19:04:26 <openstack> Launchpad bug 1474478 in OpenStack SDK "Why doesn't create update all attributes" [Critical,In progress] - Assigned to Brian Curtin (brian.curtin)
19:04:31 <briancurtin> * object_store.create_object not working with container instances (not converting instance to ID)
19:04:46 <briancurtin> i mentioned the latter earlier this morning and started working on it
19:05:07 <briancurtin> is there anything else that immediately stands out that we should try and put into a release as soon as its ready?
19:05:20 <terrylhowe> not that I can think of
19:05:42 <etoews> i wouldn't mind getting https://review.openstack.org/#/c/196221/ in a release
19:06:18 <briancurtin> i'll take a quick look at that right now
19:06:41 <etoews> thx
19:07:39 <briancurtin> etoews: yeah i remember this one, looks good. approved
19:08:51 <briancurtin> ok so if that's in, then the two things above, that's probably good to just get those things in and then release again. after that...is there anything big that's outstanding feature-wise? a lot of the bug tracker is doc and functional testing, which is certainly important, but slightly less prioritized than actual usable features
19:09:50 <briancurtin> now that plugins are done and working, it feels like we're really mostly there except for rounding out docs, tests, and a bunch of bugs
19:10:14 <terrylhowe> the only thing I’ve been working on is ksa and that won’t be ready for a long time
19:10:58 <briancurtin> i will have to look back into it, but i was running into some issues with multi-version services and getting the right one to work. identity was the tricky one
19:11:36 <terrylhowe> there may still be issues with picking versions vs version path.
19:12:15 <briancurtin> https://bugs.launchpad.net/python-openstacksdk/+bug/1466279 was one that i had entered
19:12:15 <openstack> Launchpad bug 1466279 in OpenStack SDK "ValidVersions.path not respected" [Critical,Confirmed] - Assigned to Brian Curtin (brian.curtin)
19:12:27 <terrylhowe> yeh
19:12:36 <etoews> i've got to rebase https://review.openstack.org/#/c/199318/ but then i can dig into bug tracker et al
19:13:53 <briancurtin> etoews: yeah that'll be a good one
19:16:13 <briancurtin> so beyond these few things, it's probably mostly on to the tracker to start tackling more of the functional tests...and then fixing whatever bugs we find out of that because there surely will be some
19:16:29 <briancurtin> oh, and i guess i need to get on to user-provided metadata for object_store. i have most of that done locally]
19:18:23 <terrylhowe> The foreign key thing, is that totally awful to you guys https://review.openstack.org/#/c/207972/
19:18:37 <terrylhowe> I kind of liked the mapping from resource to path
19:18:39 <briancurtin> terrylhowe: i dont understand why it's needed
19:19:02 <briancurtin> especially since %-style string interpolation was already handling a dictionary
19:19:39 <terrylhowe> It allows us to have a property that is a resource and use it in the path
19:19:41 <briancurtin> it seems like something we should be checking before we do the interpolation - if we have something that is a resource, send it through, otherwise create {"id": the_value} and send that through
19:20:52 <briancurtin> it seems pretty heavy handed for something that can just be an if-statement before doing url = base_path % blah
19:22:02 <etoews> i confess i found it pretty confusing (as in, i didn't totally get why it was needed but i thought it was just me)
19:23:43 <terrylhowe> well, it addresses exactly that bug you found
19:23:45 <briancurtin> side note: another "feature" release we should tackle soon is converting as many things over to accepting resources instead of forcing IDs, since there's still a lot left
19:24:19 <etoews> terrylhowe: which bug was that?
19:24:31 <terrylhowe> object create
19:24:56 <terrylhowe> not sure brian actually openned a bug
19:25:15 <briancurtin> terrylhowe: it probably would address that, but i think the fix for what i encountered would actually come before it hits any of that code. we should do something in the area you're working in, but it's probably a trivial function that just looks what's being sent into interpolation
19:25:20 <terrylhowe> I saw the same thing when I was messing with the foreign key thing, obj.py was broken
19:26:19 <terrylhowe> well there are several pieces to the foreign key thing that may be worth saving and they will probably make more sense one at a time
19:26:44 <briancurtin> i guess i'd rather split it up because the idea of a foreign key here doesn't even make sense to me
19:26:59 <briancurtin> it's just sending one thing or another through string interpolation
19:29:30 <terrylhowe> well, foreign key is a bit weird since I picked swift, but it makes more sense for other services
19:31:18 <briancurtin> i guess i'd have to see how it works in practice then, but we can also carry this out with %-style formatting anyway
19:32:06 <terrylhowe> yeh, that is creating some vitriol I see
19:32:31 <briancurtin> consistency matters a lot, and i say this as someone who prefers str.format(...)
19:33:17 <briancurtin> whatever way we pick on that we just have to stay with, and i think it makes sense to stick with %-style as it already works with logging and i think we found another thing a  few weeks ago it worked with automatically
19:34:17 <terrylhowe> it doesn’t have to use format it just looks cooler that way
19:39:48 <etoews> i'd prefer consistent over cool
19:40:45 <briancurtin> i guess we can figure it out when we determine how and where it's needed. i think i can solve the object_store.create_object thing without it, but we'll see. still toying with it
19:43:10 <briancurtin> is there anything else near-term to focus on outside of testing stuff?
19:46:18 <etoews> docs, bugs, and testing. that's it.
19:49:35 <etoews> should we take a minute to cull the review queue?
19:49:53 <briancurtin> i was just going to end the meeting and hten do that
19:50:20 <etoews> sgtm
19:50:52 <briancurtin> #endmeeting