At time of writing, testing coverage according to Codecov is 64%. Additionally, some of the tests we're currently running use "magic number" correct answers. While this is fine to some extent, it would also be nice to have some easily hand-verifiable answers for:
easier testing over a variety of edge cases
testing on multiple, small, minimal grids for speed
an easier time debugging
easier development process where intermediate results are more easily verifiable
documentation and examples in the future
lower barrier to entry for contributors in the future
Some specific areas of proposed improvement:
creation of "test data" that can be imported with uxarray.test_data(...) or similar
Also mentioned by @anissa111 in https://github.com/UXARRAY/uxarray/discussions/88#discussioncomment-3335717
At time of writing, testing coverage according to Codecov is 64%. Additionally, some of the tests we're currently running use "magic number" correct answers. While this is fine to some extent, it would also be nice to have some easily hand-verifiable answers for:
Some specific areas of proposed improvement:
uxarray.test_data(...)
or similarhelpers.py
, which currently has testing coverage of 27%__from_verts__
to include entire, proper, grid initializationverify
function