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/1576#issuecomment-330277364,https://api.github.com/repos/pydata/xarray/issues/1576,330277364,MDEyOklzc3VlQ29tbWVudDMzMDI3NzM2NA==,2443309,2017-09-18T16:26:36Z,2017-09-18T16:26:36Z,MEMBER,"> Why can't xarray used masked arrays, that would retain the original dtype?

We have an open issue for this topic (#1194). A lot of it comes down to performance, dask is part of that but the other issue is that masked arrays in numpy are quite slow.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,258500654
https://github.com/pydata/xarray/issues/1576#issuecomment-330271058,https://api.github.com/repos/pydata/xarray/issues/1576,330271058,MDEyOklzc3VlQ29tbWVudDMzMDI3MTA1OA==,2443309,2017-09-18T16:03:49Z,2017-09-18T16:03:49Z,MEMBER,"Right, since xarray uses `np.nan` as its fill value, any array with a `_FillValue` will be promoted to a float dtype.

Out of curiosity, what is the meaning `_NoFill = ""true""`?","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,258500654