issue_comments: 1042909091

This data as json

html_url issue_url id node_id user created_at updated_at author_association body reactions performed_via_github_app issue
https://github.com/pydata/xarray/pull/5692#issuecomment-1042909091 https://api.github.com/repos/pydata/xarray/issues/5692 1042909091 IC_kwDOAMm_X84-KYej 4160723 2022-02-17T12:40:31Z 2022-02-17T12:41:06Z MEMBER

Ok, this is ready for review!

I’m not sure what’s happening with the failing benchmark test unstacking.UnstackingSparse.time_unstack_to_sparse_2d, it seems flaky (or I missed something when trying to profile this benchmark + setup).

I haven’t updated the release notes yet as I’m not sure how best to proceed. I guess separate items should be added for each substantial change, but maybe not for all the fixes?

There’s still a number of Dataset methods (and DataArray counterparts) that may not be (fully) compatible with custom indexes (especially multi-coordinate / multi-dimensional indexes). After a quick check:

drop_sel, drop_isel, drop_dims, transpose, interpolate_na, ffill, bfill, reduce, map, apply, quantile, rank, integrate, cumulative_integrate, filter_by_attrs, idxmin, idxmax, argmin, argmax...

I haven’t looked more into those methods as this PR is already way too big and the current tests are passing. This is something that we could do progressively I think. Some of those methods may be pretty straightforward to refactor (e.g., just pass through custom indexes instead of re-computing default ones, when this is possible). It may be trickier for other methods, though.

 
{
    "total_count": 2,
    "+1": 0,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 2,
    "eyes": 0
}
   966983801