Closed cottsay closed 1 year ago
Base: 81.75% // Head: 82.01% // Increases project coverage by +0.25%
:tada:
Coverage data is based on head (
970a90d
) compared to base (a6c04e0
). Patch coverage: 100.00% of modified lines in pull request are covered.
:umbrella: View full report at Codecov.
:loudspeaker: Do you have feedback about the report comment? Let us know in this issue.
is there any way we can add a test to catch this situation?
We can't test for it here because colcon-ros
is where then shenanigans are happening. Even then, we don't have a regular infrastructure CI job where the changes would have been tested together.
It turns out that this mechanism is used to communicate the Python-type dependencies from the setuptools options when processing ament_python packages, which otherwise have ROS-type dependencies. So while we're seeing
python3-pytest
in the test dependency list, the pure python code here looks for the pure python dependencypytest
.Since this doesn't actually save us much time and is clearly still needed, I'll take the same approach in the Python project code as was taken in ament_python so that we can all use the same Python testing task.
This reverts #536.