Tech Debt
Several tests that used to pass (at least for some matrix cases) now always get skipped. The tests are
src/nitypes/_arguments.py::nitypes._arguments.arg_to_float SKIPPED (...) [ 0%]
src/nitypes/_arguments.py::nitypes._arguments.is_dtype SKIPPED (coul...) [ 0%]
src/nitypes/_arguments.py::nitypes._arguments.validate_dtype SKIPPED [ 0%]
src/nitypes/complex/init.py::nitypes.complex SKIPPED (could not ...) [ 0%]
The root cause are changes in pytest-doctestplus behavior introduced in 1.6.0 and not fully rectified in 1.7.0.
I will separately propose a workaround, but that workaround will be much more verbose and contain code duplication, making it worse than the original. The workaround will give us two choices:
- Put the workaround in place, and keep this technical debt to put the code back the way it was after the root issue gets fixed.
- Live with limited coverage until the root cause gets fixed.
AB#3703865
Tech Debt
Several tests that used to pass (at least for some matrix cases) now always get skipped. The tests are
src/nitypes/_arguments.py::nitypes._arguments.arg_to_float SKIPPED (...) [ 0%]
src/nitypes/_arguments.py::nitypes._arguments.is_dtype SKIPPED (coul...) [ 0%]
src/nitypes/_arguments.py::nitypes._arguments.validate_dtype SKIPPED [ 0%]
src/nitypes/complex/init.py::nitypes.complex SKIPPED (could not ...) [ 0%]
The root cause are changes in pytest-doctestplus behavior introduced in 1.6.0 and not fully rectified in 1.7.0.
I will separately propose a workaround, but that workaround will be much more verbose and contain code duplication, making it worse than the original. The workaround will give us two choices:
AB#3703865