| thomasb06 | gtema: Hello. The file openstack/proxy.py is ready. To push it I need to `checkout master`, `pull origin master`, and then copy my file and commit it? | 08:45 |
|---|---|---|
| gtema | yes, should be | 08:46 |
| thomasb06 | let's go | 08:46 |
| opendevreview | Thomas Bucaioni proposed openstack/openstacksdk master: Indentation of the docstrings https://review.opendev.org/c/openstack/openstacksdk/+/829862 | 09:14 |
| thomasb06 | ah, it worked... | 09:15 |
| thomasb06 | now I check in the UI, sometimes I miss some details | 09:15 |
| thomasb06 | gtema: the other file is ressources? | 09:18 |
| thomasb06 | more precisely, openstack/ressource.py | 09:19 |
| gtema | there should be more. check proactively others | 09:22 |
| thomasb06 | it's not the root only then? Understood | 09:30 |
| opendevreview | Artem Goncharov proposed openstack/openstacksdk master: Merge 'origin/feature/r1' into master https://review.opendev.org/c/openstack/openstacksdk/+/829871 | 09:53 |
| frickler | I haven't seen this in gerrit before, nice summary of what is happening https://review.opendev.org/c/openstack/openstacksdk/+/829871/-1..1//MERGE_LIST | 10:48 |
| opendevreview | Mamatisa Nurmatov proposed openstack/python-openstackclient master: Refactor network fakes to sdk properties PART 1 https://review.opendev.org/c/openstack/python-openstackclient/+/829881 | 12:04 |
| opendevreview | Mamatisa Nurmatov proposed openstack/python-openstackclient master: Refactor network fakes to sdk properties PART 1 https://review.opendev.org/c/openstack/python-openstackclient/+/829881 | 13:22 |
| thomasb06 | gtema: Zuul returns an error, but I guess it's not me: https://review.opendev.org/c/openstack/openstacksdk/+/829862 | 13:33 |
| thomasb06 | After a couple of days, I recheck? | 13:34 |
| isabek_ | thomasb06: Hi! I think you can recheck it now, there some timeout | 13:37 |
| thomasb06 | isabek_: great, thank you very much | 13:38 |
| thomasb06 | since it's the main branch, I'm a bit cautious this time | 13:39 |
| isabek_ | thomasb06: Welcome! Now job is success : https://zuul.openstack.org/status/change/829862,1 | 13:47 |
| gtema | thomasb06 - all fine. This was a "normal" sporadic failure | 13:48 |
| thomasb06 | isabek_: gtema: alright then, nothing is broken | 13:49 |
| isabek_ | gtema: Hi! Can we remove tenant_id from network resources? example: https://github.com/openstack/openstacksdk/blob/master/openstack/network/v2/address_scope.py#L43 | 14:49 |
| gtema | isabek_ for now we can not drop it completely | 14:50 |
| gtema | we have old clients potentially still trying to pass it | 14:50 |
| gtema | sdk is used by osc/ansible which depending on their version may also result in very old things still being passed | 14:51 |
| isabek_ | At this moment I'm working on network_fakes to use sdk properties. https://review.opendev.org/c/openstack/python-openstackclient/+/829881 | 14:51 |
| isabek_ | Because of tenant_id property is failing https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_72d/829881/2/check/osc-tox-py36-tips/72dc317/testr_results.html | 14:52 |
| isabek_ | gtema: What do you recommend for it? Just fix unit tests? | 14:53 |
| gtema | yes, tests need to be fixed. I actually always wanted that SDK also maintains "fakes" that can be then consumed by osc | 14:56 |
| gtema | I really dislike current approach, especially that we will move more and more OSC stuff to sdk | 14:57 |
| isabek_ | gtema: Can I simply hide with hidden_columns? | 15:01 |
| gtema | honestly I feel like I was already fixing this | 15:02 |
| gtema | https://review.opendev.org/c/openstack/python-openstackclient/+/824746 | 15:02 |
| gtema | I am surprised you have it now | 15:02 |
| isabek_ | Because now I'm using sdk properties where is tenant_id | 15:03 |
| isabek_ | such as https://github.com/openstack/openstacksdk/blob/master/openstack/network/v2/address_scope.py#L43 | 15:04 |
| gtema | and what was before? | 15:07 |
| gtema | ah, before it was simply FakeResource | 15:08 |
| isabek_ | yes | 15:10 |
| gtema | also this function approach "create_availability_zone" is not how I think it should be in the long run | 15:10 |
| gtema | I image | 15:10 |
| gtema | from openstack.fakes.network.v2 import availability_zone | 15:10 |
| gtema | but before those appear in sdk we can prepare things in osc. Afterwards we simply relocate them to sdk | 15:11 |
| gtema | I would even start implementing it in SDK, but I have no feeling we can release that in next days/weeks. Rather few months still to finalise r1 changes | 15:12 |
| opendevreview | James Kirsch proposed openstack/openstacksdk master: Identity: Add support for system role assignment https://review.opendev.org/c/openstack/openstacksdk/+/826193 | 19:01 |
| opendevreview | James Kirsch proposed openstack/openstacksdk master: Identity: Add support for system role assignment https://review.opendev.org/c/openstack/openstacksdk/+/826193 | 22:23 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!