| opendevreview | Merged opendev/git-review master: Fix use of removed --preserve-merges option https://review.opendev.org/c/opendev/git-review/+/818219 | 03:33 |
|---|---|---|
| opendevreview | Jeremy Stanley proposed opendev/bindep master: DNM: Test bindep with PYTHONWARNINGS=error https://review.opendev.org/c/opendev/bindep/+/818672 | 14:20 |
| opendevreview | Jeremy Stanley proposed opendev/bindep master: DNM: Test bindep with PYTHONWARNINGS=error https://review.opendev.org/c/opendev/bindep/+/818672 | 14:56 |
| opendevreview | Merged openstack/project-config master: Retire training-labs: remove project infra https://review.opendev.org/c/openstack/project-config/+/817507 | 17:16 |
| fungi | clarkb: not urgent, but i replicated the `setup.py install` deprecation warning with the bindep change above, parented to your pep-517 config. looks like pbr.util.setup_cfg_to_setup_kwargs() is calling into setuptools in such a way that it ends up traversing setupt | 17:25 |
| fungi | er, traversing setuptools.command.install | 17:26 |
| fungi | also winds up hitting easy_install from the same point | 17:28 |
| opendevreview | Merged openstack/project-config master: Fix ACLs for deliverables under release management https://review.opendev.org/c/openstack/project-config/+/818356 | 17:31 |
| fungi | pretty sure it has to do with how the setuptools command classes are being iterated through. we might need to weed out some of what's in data_files there to avoid calling deprecated classes? | 17:32 |
| Clark[m] | From your traceback it was through the cmdclass argument which I can't see being set anywhere though | 17:40 |
| Clark[m] | And it makes sense that cmdclass would trigger the deprecation warning because cmdclass extends setup.py commands | 17:41 |
| Clark[m] | Something in pbr might be setting that implicitly for some pbr specific extension? Maybe for testing commands? | 17:41 |
| Clark[m] | If that is the case we can probably avoid that if invoked via pep517 | 17:42 |
| Clark[m] | fungi: https://opendev.org/openstack/pbr/src/branch/master/pbr/hooks/commands.py is doing it I think | 17:45 |
| Clark[m] | We should be able to set a flag in the pep517 build entry point and not load commands when we do that | 17:46 |
| Clark[m] | That might make some things change behavior like entry point writing though | 17:48 |
| Clark[m] | fungi: I think we may need to move LocalSDist into the pep517 hook somehow so that it doesn't rely on commands to override | 17:55 |
| fungi | oh, that might make sense | 18:05 |
| opendevreview | Jeremy Stanley proposed opendev/bindep master: Update formatting of author_email in setup.cfg https://review.opendev.org/c/opendev/bindep/+/818675 | 19:01 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!