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/4014#issuecomment-621102297,https://api.github.com/repos/pydata/xarray/issues/4014,621102297,MDEyOklzc3VlQ29tbWVudDYyMTEwMjI5Nw==,1700203,2020-04-29T09:56:11Z,2020-04-29T09:56:11Z,CONTRIBUTOR,"I'd be happy to open a PR, if such a behaviour is desired.

It seems like extending the [`_nc3_dtype_coercions` dict](https://github.com/pydata/xarray/blob/33a66d6380c26a59923922ee11e8ffcf0b4f379f/xarray/backends/netcdf3.py#L31) (and extending tests) is all that needs to be done here?

Suggestion: Add all [`uint` types supported by numpy](https://numpy.org/devdocs/user/basics.types.html):
```python
_nc3_dtype_coercions = {
    ""int64"": ""int32"",
    ""bool"": ""int8"",
    ""uint64"": ""int32"",
    ""uint32"": ""int32"",
    ""uint16"": ""int16"",
    ""uint8"": ""int8"",
}
```","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,608923222