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/678#issuecomment-356081997,https://api.github.com/repos/pydata/xarray/issues/678,356081997,MDEyOklzc3VlQ29tbWVudDM1NjA4MTk5Nw==,2443309,2018-01-08T20:11:27Z,2018-01-08T20:11:27Z,MEMBER,"Yes, thanks @gerritholl.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,121740837
https://github.com/pydata/xarray/issues/678#issuecomment-355914496,https://api.github.com/repos/pydata/xarray/issues/678,355914496,MDEyOklzc3VlQ29tbWVudDM1NTkxNDQ5Ng==,500246,2018-01-08T09:13:59Z,2018-01-08T09:13:59Z,CONTRIBUTOR,Is this fixed by #1170?,"{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,121740837
https://github.com/pydata/xarray/issues/678#issuecomment-164083490,https://api.github.com/repos/pydata/xarray/issues/678,164083490,MDEyOklzc3VlQ29tbWVudDE2NDA4MzQ5MA==,1217238,2015-12-11T23:53:19Z,2015-12-11T23:53:19Z,MEMBER,"`encoding` only exists on `Variable` objects, with the user facing API exposed via `DataArray`. We don't even have it on `Dataset` objects, though I suppose that could be a logical addition.

Currently dimensions are just strings. I would be reluctant to use some sort of more complex object instead.
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,121740837
https://github.com/pydata/xarray/issues/678#issuecomment-164082151,https://api.github.com/repos/pydata/xarray/issues/678,164082151,MDEyOklzc3VlQ29tbWVudDE2NDA4MjE1MQ==,2443309,2015-12-11T23:45:06Z,2015-12-11T23:45:06Z,MEMBER,"I'd like to keep the solution out of `to_netcdf` if possible.  To me, not maintaining the kind of dimension means that we are not ""round tripping"" netCDF4 files faithfully. Thinking about the `encoding` option, IIRC, `encoding` just exists on `DataArrays` and `DataSets`, it would be nice to have something on the dimension itself. Would it may fit better in the `Variable.dim` space?
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,121740837
https://github.com/pydata/xarray/issues/678#issuecomment-164010099,https://api.github.com/repos/pydata/xarray/issues/678,164010099,MDEyOklzc3VlQ29tbWVudDE2NDAxMDA5OQ==,1217238,2015-12-11T18:27:34Z,2015-12-11T18:27:34Z,MEMBER,"We could certainly add an option to `to_netcdf` to explicitly indicate record dimensions. In principle, we could probably also save this information via some flag in the `encoding` dict associated with the coordinate variables.
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,121740837
https://github.com/pydata/xarray/issues/678#issuecomment-164002110,https://api.github.com/repos/pydata/xarray/issues/678,164002110,MDEyOklzc3VlQ29tbWVudDE2NDAwMjExMA==,2443309,2015-12-11T17:49:39Z,2015-12-11T17:49:39Z,MEMBER,"We've noticed this as well and it is something we find out how to do.

@shoyer - Any thoughts on how to make this happen?  In `netCDF4`, this is done by setting the size of a dimension to `None`.

cc @lizaclark
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,121740837