Closed 1uc closed 2 months ago
If it helps reviewing, we could split out the commits that move code, e.g. compare_arrays
into a separate PR. Then this one will be removing code and reimplementing the odd test; but no refactoring.
Attention: Patch coverage is 84.81013%
with 12 lines
in your changes are missing coverage. Please review.
Project coverage is 86.40%. Comparing base (
5f3ded6
) to head (2e4a41a
).
Files | Patch % | Lines |
---|---|---|
tests/unit/compary_arrays.hpp | 52.00% | 12 Missing :warning: |
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
Nice to split test_boost
and test_stl
as test_opencv
and test_xtensor
already are.
For the next time, maybe a test_eigen
?
Interesting, I thought I'd split them out already, but yes, they can get split away too.
There's (at least) three attempts at testing read/write cycles for multi-dimensional arrays. The first is whatever is in
tests_*multi_dims.cpp
. Since it's partial, a second attempt was made: the first half oftest_all_types.cpp
. Since this was still partial and not quite DRY enough, a third attempt was made.This PR removes the duplicated tests from the first and second attempt, one-by-one. Each commit removes one test. Or moves/reimplements a test if it's not yet covered by the third attempt.
There's a fourth attempt in
test_high_five_base.cpp
; but that's left for another PR.