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/849#issuecomment-220884679,https://api.github.com/repos/pydata/xarray/issues/849,220884679,MDEyOklzc3VlQ29tbWVudDIyMDg4NDY3OQ==,1217238,2016-05-23T04:21:48Z,2016-05-23T04:21:48Z,MEMBER,"I think this is good to merge. Appveyor is definitely doing something strange here with the Python 3 build.
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,155181260
https://github.com/pydata/xarray/pull/849#issuecomment-220884502,https://api.github.com/repos/pydata/xarray/issues/849,220884502,MDEyOklzc3VlQ29tbWVudDIyMDg4NDUwMg==,1217238,2016-05-23T04:19:31Z,2016-05-23T04:19:31Z,MEMBER,"I think it's fine to write all booleans data using this encoding. It's pretty obvious how it works, and it's compatible with all versions of netCDf.

Unfortunately, netCDF-C libraries can't read variables with the enum dtype created by h5py for boolean data (https://github.com/Unidata/netcdf-c/issues/267), so I don't think we should encourage that option.
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,155181260
https://github.com/pydata/xarray/pull/849#issuecomment-220652754,https://api.github.com/repos/pydata/xarray/issues/849,220652754,MDEyOklzc3VlQ29tbWVudDIyMDY1Mjc1NA==,1217238,2016-05-20T16:24:48Z,2016-05-20T16:24:48Z,MEMBER,"One annoying potential inconsistency here is that it _is_ possible to write boolean dtypes using h5py (which uses HDF5 enums), which is different from how we do things here.
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,155181260