change python requirements to 3.10 - 3.12

alan-turing-institute / DTBase

A starting point from which digital twins can be developed.

MIT License

11 stars 4 forks source link

change python requirements to 3.10 - 3.12 #134

Closed EdwinB12 closed 10 months ago

EdwinB12 commented 10 months ago

Checklist:

[x] Put on minimum versions of python dependancies : resolve #119
[x] Remove unneccessary python installs in dockerfiles
[x] Change python versions in pyproject.toml and readme to document supported python versions : resolve #124
[x] Change github actions to run tests on all python versions
[ ] Remove easy warnings : resolve #125
[ ] Consider building backend and frontend apps with all supported python versions

EdwinB12 commented 10 months ago

@mhauru @GiorgioCerro - Could you look at the warnings spat out from the tests and see if there are any super easy ones you can address?

You may need to flag the warnings as errors to see the full warning: pytest -W error tests

I'm going to merge this branch with test-actions branch to run the docker containers and check whether those actions pass okay.

EdwinB12 commented 10 months ago

All actions have passed (including docker deployments on test-actions branch).

I'm essentially happy for this to be merged after review as lots of the warnings have now been supressed. However, please feel free to have a go at some of the remaining warnings.

Last thing @mhauru - We discussed whether we want to run a docker deployment for each python version. I think we leave this for now. I'm quite happy that the tests are being run for all supported python versions but let me know what you think.

mhauru commented 10 months ago

Last thing @mhauru - We discussed whether we want to run a docker deployment for each python version. I think we leave this for now. I'm quite happy that the tests are being run for all supported python versions but let me know what you think.

Sounds good.

The SQLAlchemy warning came from a legitimate bug (nice warning SQLAlchemy!), I fixed it.

mhauru commented 10 months ago

I also silenced a warning coming from pandas that we were losing nanosecond precision when converting to python datetimes. With that we are down to 4 warnings, which all seem to come from statsmodels, and would probably require a deeper dive.

codecov-commenter commented 10 months ago

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Comparison is base (f432422) 86.29% compared to head (ed8c1bb) 86.29%.

:exclamation: Current head ed8c1bb differs from pull request most recent head 82d390d. Consider uploading reports for the commit 82d390d to get more accurate results

Additional details and impacted files

```diff @@ Coverage Diff @@ ## main #134 +/- ## ========================================== - Coverage 86.29% 86.29% -0.01% ========================================== Files 54 54 Lines 4101 4100 -1 ========================================== - Hits 3539 3538 -1 Misses 562 562 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.