Closed fitoprincipe closed 1 month ago
test failure is weird since the failing test is not using the modified method at all 🤔
The test failure is related to GEE slightly changing it's rounding method all the time, you must nuke the nox folder from your side and start the complete test suit again to regenerate the file. I would like to have a better way of testing this with pytest-gee but never found the time to work on it yet.
ee.Projection
not only stores the CRS but also the nominal scale, and although the CRS is usually constant across all bands, is not the same for the nominal scale (example here). So, one of the modification of this PR is to copy each band's projection to the image generated byee.Image.fullLike
.In the previous version of
fullLike
the footprint of the original images was passed to the resulting image. Now the user can choose whether to copy it or not, giving more flexibility to the user.