Benoit Bovy

Technically it *could* be possible to implement an Xarray index that keeps track of the original CRSs and that encapsulates custom CoordinateTransform objects, such that from the user point of...

Hi @shoyer -- Yes! I started looking again at this earlier today actually, I opened #10000 as a side PR. Apart from unit tests I don't think there is more...

@dcherian Yes let's do this (unless we want to avoid breaking changes in the next related PRs... although I'm already pretty happy with the API here). I'll add unit tests...

Hmm maybe we should fix CI before merging it, though? I can have a look tomorrow.

I added the tests here (mostly complete for the features added). This should be ready for review (or for merging :-)).

The issue above is actually duplicate of #1647, but I wonder if there could be a more generic fix for dtypes other than variable length strings?

Briefly chiming in here... > The "native" dataset concept is increasingly outdated This may be true for Discrete Global Grid Systems (DGGS) as well. > The reprocessing and refactoring of...

Would an overall structure like below (example for sentinel-2) reconcile the different points of view? ``` /measurements/ ├── r10m/ │ ├── band1 │ ├── band2 │ └── spatial_ref └── overviews/...

> It's an open question whether various combinations of these should be compacted into one extension type with metadata Yes I think it would be more future-proof to treat DGGS...

What does mean "here is data encoded as GeoArrow" exactly? If this means "here are simple feature geometries encoded in Apache Arrow", then I agree with @martinfleis too. But if...