Closed eddiebergman closed 9 months ago
Some open TODOs for this refactor (I will edit this comment while working on this):
Attention: 140 lines
in your changes are missing coverage. Please review.
Comparison is base (
56895c2
) 84.92% compared to head (52c4d1f
) 71.14%.:exclamation: Current head 52c4d1f differs from pull request most recent head 42b34e7. Consider uploading reports for the commit 42b34e7 to get more accurate results
Files | Patch % | Lines |
---|---|---|
openml/runs/functions.py | 56.86% | 22 Missing :warning: |
openml/extensions/sklearn/extension.py | 81.69% | 13 Missing :warning: |
openml/testing.py | 72.91% | 13 Missing :warning: |
openml/datasets/dataset.py | 70.58% | 10 Missing :warning: |
openml/runs/run.py | 60.00% | 10 Missing :warning: |
openml/utils.py | 84.48% | 9 Missing :warning: |
openml/config.py | 82.60% | 8 Missing :warning: |
openml/tasks/split.py | 71.42% | 8 Missing :warning: |
openml/tasks/functions.py | 86.27% | 7 Missing :warning: |
openml/datasets/functions.py | 83.78% | 6 Missing :warning: |
... and 14 more |
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
I reset this branch to an earlier commit as I messed up something fixing lint errors. Perhaps it's better to do that in another PR and have this as a clean state where the tests pass (minus the 3.6 version)
One failing test, seems unerlated but not sure
FAILED tests/test_datasets/test_dataset_functions.py::TestOpenMLDataset::test_data_status - openml.exceptions.OpenMLServerException: https://test.openml.org/api/v1/xml/data/status/update returned code 694: Illegal status transition - None
Tested the failing test, it was just flaky and due to the test server
WIP
What does this PR implement/fix? Explain your changes.
This PR aims to update tooling to ruff. This PR stemmed from cloning the repo and immediately having black format to
88
in my editor due to no file specification.In general, there's also no import sorting which is probably not problematic for an established repo but good to have.
There should be no breaking changes other than rule
FBT
which discourages positional boolean arguments. It might be worth disabling this in public user facing functions if it's arleady done to prevent any backwards breaking changes.Please let me know if I should justify ruff more, there's a lot of good things and very few negatives I could really think of.
How should this PR be tested?
make check
which runspre-commit run --all-files
Any other comments?