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/1163#issuecomment-268943572,https://api.github.com/repos/pydata/xarray/issues/1163,268943572,MDEyOklzc3VlQ29tbWVudDI2ODk0MzU3Mg==,1217238,2016-12-23T05:44:00Z,2016-12-23T05:44:00Z,MEMBER,Fixed by #1165,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,195576963
https://github.com/pydata/xarray/issues/1163#issuecomment-267086215,https://api.github.com/repos/pydata/xarray/issues/1163,267086215,MDEyOklzc3VlQ29tbWVudDI2NzA4NjIxNQ==,1217238,2016-12-14T16:46:05Z,2016-12-14T16:46:05Z,MEMBER,"I'm a little surprised that CDO cannot handle a NaN `_FillValue` -- are you sure this is actually the case? If so, it might be worth raising the issue with the CDO developers.

But actually, xarray currently does not even set `_FillValue` by default, though you can do this yourself on a per-variable basis [by setting `encoding`](http://xarray.pydata.org/en/stable/io.html#scaling-and-type-conversions). I think it would be reasonable to default to a `_FillValue` of NaN for floating point dtypes if that would help third party tools, given that that is how xarray's data model works (any other default choice would run the risk of over-writing user data).","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,195576963