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-1069677202,https://api.github.com/repos/pydata/xarray/issues/5692,1069677202,IC_kwDOAMm_X84_wfqS,14808389,2022-03-16T22:00:53Z,2022-03-16T22:00:53Z,MEMBER,sounds good to me!,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,966983801
https://github.com/pydata/xarray/pull/5692#issuecomment-1065066894,https://api.github.com/repos/pydata/xarray/issues/5692,1065066894,IC_kwDOAMm_X84_e6GO,14808389,2022-03-11T12:24:32Z,2022-03-11T12:24:32Z,MEMBER,"the doc build actually fails because `nbconvert` relied on the implicit dependency of `nbformat` on `ipython_genutils`, but that got removed in the most recent release of `nbformat`. This is fixed in #6350.","{""total_count"": 1, ""+1"": 1, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,966983801
https://github.com/pydata/xarray/pull/5692#issuecomment-1041843391,https://api.github.com/repos/pydata/xarray/issues/5692,1041843391,IC_kwDOAMm_X84-GUS_,14808389,2022-02-16T16:23:56Z,2022-02-16T16:23:56Z,MEMBER,"> For `arr.roll`, the index is dropped if `roll_coords=True` and `CustomIndex.roll()` is not implemented.
right, I can confirm that, too","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,966983801
https://github.com/pydata/xarray/pull/5692#issuecomment-1041781329,https://api.github.com/repos/pydata/xarray/issues/5692,1041781329,IC_kwDOAMm_X84-GFJR,14808389,2022-02-16T15:46:18Z,2022-02-16T15:56:59Z,MEMBER,"> I wonder how it goes when applying other operations for which the custom index has no implementation.
let's try it:
- `arr.roll(x=5)`: index is not dropped, the result for `sel(indexers)` is different from before the `roll` (I guess that makes sense? `pandas` indexes behave the same way...)
- `copy`: raises `NotImplementedError`
- `arr.stack(z=(""x"", ""y""))`: index is dropped
- `arr.isel(x=slice(5, 10), y=slice(10, 20))`: index is dropped
I don't have ideas on how to test the other operations right now.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,966983801
https://github.com/pydata/xarray/pull/5692#issuecomment-1039079452,https://api.github.com/repos/pydata/xarray/issues/5692,1039079452,IC_kwDOAMm_X8497xgc,14808389,2022-02-14T13:20:18Z,2022-02-14T13:20:18Z,MEMBER,"thanks, @benbovy, those changes look good to me for now. After this is merged I'll have to update the tests anyways.
In the meantime, if you find any tests where the behavior changed so much that it is tricky to fix them I'd be fine with xfailing / skipping them for now.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,966983801
https://github.com/pydata/xarray/pull/5692#issuecomment-925966420,https://api.github.com/repos/pydata/xarray/issues/5692,925966420,IC_kwDOAMm_X843MSBU,14808389,2021-09-23T16:23:05Z,2021-09-23T16:24:58Z,MEMBER,"> FYI, this was forbidden in #3645
The discussion for that is in #3438. In short, in addition to renaming, `rename*` did replace other variables, which I think we should only allow when explicitly requested (e.g. by requiring `replace=True`), regardless of what it does with indexes. It also contained a bug related to the indexes, which I think @dcherian was trying to fix in #4108.
In any case, it would really be good to come up with a consistent API in #4825, and I agree that the index refactor seems like a good time for that.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,966983801