Romain Hugonnet

Sure! Don't forget to mention that we will likely take glacier-specific tools out of xDEM in the long-term :wink:

@adehecq @atedstone @erikmannerfelt This PR is also ready for your first review! It is not finalized, but at a good stage to hear your feedback, questions, recommendations, and then move...

Thanks @guillaumeeb, some very good points! I'll account for the other, more specific comments while I dive back into the PR. For DataArray vs Dataset: That's an important point that...

@adehecq @erikmannerfelt @adebardo @belletva This PR is ready for your review, see description above for details.

> Do you plan to add the "edge_method" problem in a new issue? You're right, opened #747!

Thanks @marinebcht! For MyPy: We can leave "# type: ignore" at the end of every decorator I think, it's not an issue. For functions to decorate: Yes, also `interp_points()` would...

@marinebcht OK thanks, I hadn't realized. If I understood properly: I don't think it's a good idea to override the name of the class with the module name. Why not...

@marinebcht Amazing description and summary, I grasped everything immediately. :wink: And great idea for the extra consistency tests! (crop/uncropped and pointcloud/raster) Here are possible explanations: - **For uncropped/crop:** In the...

Quick answers: - **Inlier mask out of bounds:** I think our only option is to raise an error: if the user is passing an `inlier_mask` that does not intersect the...

@marinebcht Yes indeed, the behaviour of 1/ is not correct, it would have to raise an error... I thought an error/warning was raised by `reproject()` when Rasters did not intersect,...