Remove pytest-lazy-fixture due to incompatibility with pytest v8.0.0
#369
Conversation
|
Looks like tests on Ubuntu for Python 3.10 reproducibly encounter a segmentation fault 🤔 would you be able to investigate further, @willGraham01 , please? |
Will do this afternoon / Monday after DocSkills teardown admin 👍 |
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
|
Tests seem to be passing now, but holding off merging this PR due to the presence of #372 and an investigation can take place. But if people need CI to pass, consider merging or rebasing onto this branch for the time being. |
* improve detection debug logs * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * remove too wordy debug logs --------- Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
|
Investigations over on #376 seem to imply that the garbage collection fix was not necessary (all tests seem to be passing without the need for the garbage collection cleanup statement). I suspect there might be a faulty cache, so have deleted the cache for this branch and requested a re-run of the original CI that threw the seg-fault error: https://github.com/brainglobe/cellfinder/actions/runs/7743237973/job/21314877990 |
Reverting apparently un-neccessary changes due to conflation with a bad cache fetch. * Revert "try attempting garbage collector to run at end of test" This reverts commit 2753abb. * Revert "Actually raise the exception for not enough CPUs" This reverts commit 1545b1b. * Revert "[pre-commit.ci] auto fixes from pre-commit.com hooks" This reverts commit bdd0d91. * Revert "Remove fixtures and get pytest to check for cpu availability at runtime" This reverts commit 125fa9d. * Revert "[pre-commit.ci] auto fixes from pre-commit.com hooks" This reverts commit dd0e519. * Revert "Clarify problem fixture docstrings" This reverts commit 67caf82. * Add a timeout to CI for runs
for more information, see https://pre-commit.ci
|
Is the 20 min timout limit still sensible here, @willGraham01 ? |
Good catch, will bump this to 1 hour (should be more than enough time for the Windows tests to complete as they're the slowest). |
Resolves #368 |
Turns out we were only using
pytest-lazy-fixturefor one test anyway.Workaround is to pass the fixture name to
pytest.paramas a string, and then use therequestsfixture (built in topytest) to fetch the value of that fixture.As such,
pytest-lazy-fixtureis no longer a dependency.Also fixes:
scikit-imageeven though cellfinder depends on it?