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/1046#issuecomment-253929482,https://api.github.com/repos/pydata/xarray/issues/1046,253929482,MDEyOklzc3VlQ29tbWVudDI1MzkyOTQ4Mg==,1217238,2016-10-14T21:56:42Z,2016-10-14T21:56:51Z,MEMBER,"@chunweiyuan I agree, this seems worth doing, and I think you have a pretty sensible approach here. For large arrays (especially with ndim > 1), this should add only minimal performance overhead. If you can fit this into the [existing framework](https://github.com/pydata/xarray/blob/master/xarray/core/rolling.py) for `rolling` that would be awesome!
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,182667672
https://github.com/pydata/xarray/issues/1046#issuecomment-253405068,https://api.github.com/repos/pydata/xarray/issues/1046,253405068,MDEyOklzc3VlQ29tbWVudDI1MzQwNTA2OA==,1217238,2016-10-13T03:37:55Z,2016-10-13T03:37:55Z,MEMBER,"I think we mostly tried to make this consistent with pandas. To be honest I don't entirely understand the logic myself.

Cc @jhamman 
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,182667672