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/issues/5562#issuecomment-873548277,https://api.github.com/repos/pydata/xarray/issues/5562,873548277,MDEyOklzc3VlQ29tbWVudDg3MzU0ODI3Nw==,14808389,2021-07-04T08:46:34Z,2021-07-04T08:46:34Z,MEMBER,"that makes sense, but I think long term it might be better to find a way to delegate this to duck array extension libraries (it really should call `.pint.dequantify()`).","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,935531700
https://github.com/pydata/xarray/issues/5562#issuecomment-873490045,https://api.github.com/repos/pydata/xarray/issues/5562,873490045,MDEyOklzc3VlQ29tbWVudDg3MzQ5MDA0NQ==,2448579,2021-07-04T00:23:46Z,2021-07-04T00:23:46Z,MEMBER,"> I think we should special-case pint and add the units attribute on the DataArray like .pint.dequantify

We would also want this to write pint-backed xarray datasets to netCDF.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,935531700
https://github.com/pydata/xarray/issues/5562#issuecomment-873486576,https://api.github.com/repos/pydata/xarray/issues/5562,873486576,MDEyOklzc3VlQ29tbWVudDg3MzQ4NjU3Ng==,2448579,2021-07-03T23:56:54Z,2021-07-03T23:56:54Z,MEMBER,Oh I see. `as_numpy` is using `data.magnitude` and dropping the units. I think we should special-case pint and add the `units` attribute on the DataArray like `.pint.dequantify`,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,935531700
https://github.com/pydata/xarray/issues/5562#issuecomment-873482727,https://api.github.com/repos/pydata/xarray/issues/5562,873482727,MDEyOklzc3VlQ29tbWVudDg3MzQ4MjcyNw==,2448579,2021-07-03T23:12:31Z,2021-07-03T23:12:31Z,MEMBER,wouldn't `as_numpy` on a pint-backed DataArray return a DataArray with numpy data and units in the attribute?,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,935531700
https://github.com/pydata/xarray/issues/5562#issuecomment-873478528,https://api.github.com/repos/pydata/xarray/issues/5562,873478528,MDEyOklzc3VlQ29tbWVudDg3MzQ3ODUyOA==,14808389,2021-07-03T22:23:50Z,2021-07-03T22:23:50Z,MEMBER,"that will work if we want to remove the units, but `.pint.dequantify()` also adds the units as string-valued attributes. In other words, I'd like the new hook to work on `xarray` objects instead of the wrapped duck array.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,935531700
https://github.com/pydata/xarray/issues/5562#issuecomment-872966327,https://api.github.com/repos/pydata/xarray/issues/5562,872966327,MDEyOklzc3VlQ29tbWVudDg3Mjk2NjMyNw==,2448579,2021-07-02T12:38:09Z,2021-07-02T12:38:09Z,MEMBER,"> For example for pint the hook would return the result of .pint.dequantify() but it could also be used to explicitly call .get on cupy arrays or .todense on sparse arrays.

We discussed `as_numpy` or `to_numpy_data` here: https://github.com/pydata/xarray/issues/3245","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,935531700