Hi Xqt. Sorry to reopen but changing my user-config.py in the way mentioned above surfaced a new test failure and I think this one is legitimate...

Just use Site('wikipedia:test') for test.wikipedia.org. The old test_family just duplicated the same site. (Ok the test family was prior)

test.wikipedia.org is part of wikipedia_family.py. Therefore the test family was dropped. See T228375

Xqt requested that I move information about https://gerrit.wikimedia.org/r/c/pywikibot/core/+/707869 onto this ticket. The 'drev' value with the family mentioned above is...

They've all been merged or dropped. Resolving.

Gotcha. I think we've concluded this ticket so closing. Feel free to reopen if you'd like to further discuss anything.

But I am also fine with your proposal and it might be better to have one big change with 7.0 than douzands every month. I explicitly aggree with your points above.

Xqt merged our FutureWarning patch but hasn't said if he approves of the above (ie. to begin enforcing a symantic versioning policy). Since he merges most patches this will be a lost cause without his agreement so leaving this open to get his thoughts.

I amended our deprecation policy in the following ways...

Hi Xqt. Sorry to reopen but on reflection I'd like to propose a different policy.

Hi Xqt. It's a bit unclear to me what the task is...

Hi Xqt, I added a deprecation policy to our guidelines. When I get external link permission I'll link the tools functions to their API docs.

Don't see any advantage.

If we don't have existing epytext lint checks then why is reST linting a requirement to proceed with your change? Or did we discard the change for reasons aside from darglint?

rstcheck validates the reSt syntax. It does not validate the content.

Currently the only check is made by sphinx but there is no code validation.

Hi Xqt. I really like your CR 673836 and would love to move this along. Unfortunately your CI link no longer works - what kind of errors did you experience and can we reproduce them locally?

Hi Xqt. For what it's worth I have the ui_tests inconsistently fail locally (it seems to happen about 70% of the time).

Thanks Xqt! Yesterday I finished evaluating all our deprecations prior to the end of 2019, removing FutureWarnings and changing DeprecationWarnings to FutureWarnings. In retrospect I should've discussed it here before digging in. Sorry about that.

Thanks for the detailed explanation Xqt! I don't have sufficient familiarity with our frameworks usage to determine the best class hierarchy to provide so I'm going to focus elsewhere for now. I might revisit this in the future. In the meantime I'll simply drop the deprecated bit of this module.

That wiki fixed their certificate. Closing.

Hi Xqt. I'm going to reopen this because T106121 did not answer this. This ticket is about documenting when deprecated code can be removed.

There seems to be a Python 3.5 issue regarding ModuleDeprecationWrapper. Would you like for me to install Python 3.5 to troubleshoot or should we continue with this?

That's... really odd. I have Python 3.8 and unfortunately installing a deprecated 3.5 interpreter would be a pita. Do you have any idea what's going on or should I try to get one working?

That's odd. I do get it...

Thanks @JJMC89! Great point. Code review updated with a better approach which fixes that.