Closed mcara closed 3 months ago
Attention: Patch coverage is 30.43478%
with 16 lines
in your changes are missing coverage. Please review.
Project coverage is 80.33%. Comparing base (
5de7fc7
) to head (dd32324
). Report is 4 commits behind head on master.
Files | Patch % | Lines |
---|---|---|
spherical_geometry/great_circle_arc.py | 28.57% | 15 Missing :warning: |
spherical_geometry/polygon.py | 50.00% | 1 Missing :warning: |
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
This does fix the devdeps job, so if you don't want to add a test, feel free to merge. Thanks!
I think this should fix failing tests and the Python versions of length
and inner1d
. I disabled one test: I don't think it is critical and I will think about it in a week.
Unit tests for would be a plus but no time now.
Tests with numpy>2.0.0
are passing after https://github.com/spacetelescope/spherical_geometry/pull/270 and https://github.com/spacetelescope/spherical_geometry/blob/7028f15eff18caee6c56219ff774c3c534b6ea67/pyproject.toml#L57 uncommented.
After commit c99e74e4440f4d40cae9a409dc1ff8f3ea179453 most tests (except for one that we skip) with numpy
implementation of quad-accuracy functions from math_util
pass.
I will now remove tricks used for testing purpose and squash most commits.
This PR should fix (for now some) errors in unit tests when C
math_util
is not available.length()
needs more investigation (but the biggest difference in hownan
are treated in comparisons: C code considers twonan
equal) and I will do that in about 1 week from now.Fix #261
CC: @jhunkeler @pllim