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/1041#issuecomment-343530337,https://api.github.com/repos/pydata/xarray/issues/1041,343530337,MDEyOklzc3VlQ29tbWVudDM0MzUzMDMzNw==,1217238,2017-11-10T17:07:17Z,2017-11-10T17:07:17Z,MEMBER,"@chrwerner Yes, exactly.","{""total_count"": 1, ""+1"": 1, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,181534708
https://github.com/pydata/xarray/issues/1041#issuecomment-343525358,https://api.github.com/repos/pydata/xarray/issues/1041,343525358,MDEyOklzc3VlQ29tbWVudDM0MzUyNTM1OA==,1217238,2017-11-10T16:48:10Z,2017-11-10T16:48:10Z,MEMBER,"@chrwerner From an implementation perspective, resample works virtually identically to passing in an explicit array of values to `groupby()`. If you can make your arrays of ""10 year interval"" labels, you should be able to make this work.","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,181534708
https://github.com/pydata/xarray/issues/1041#issuecomment-252307570,https://api.github.com/repos/pydata/xarray/issues/1041,252307570,MDEyOklzc3VlQ29tbWVudDI1MjMwNzU3MA==,1217238,2016-10-07T17:07:40Z,2016-10-07T17:07:40Z,MEMBER,"It seems like we should just make `resample()` work for non-dimension coordinates. I don't think there's any reason in principle why that shouldn't work.

To make `sel()` efficient, we need to compute a hash table. So I think it's somewhat reasonable to require setting levels you want to select on to be part of a (multi)index.
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,181534708
https://github.com/pydata/xarray/issues/1041#issuecomment-252294581,https://api.github.com/repos/pydata/xarray/issues/1041,252294581,MDEyOklzc3VlQ29tbWVudDI1MjI5NDU4MQ==,10050469,2016-10-07T16:13:00Z,2016-10-07T16:13:00Z,MEMBER,"Just an idea: what about adding a keyword to `resample()` (and maybe also to `sel()`)? Something like:

``` python
ds.resample('D', by_coord='time')
```
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,181534708
https://github.com/pydata/xarray/issues/1041#issuecomment-252293208,https://api.github.com/repos/pydata/xarray/issues/1041,252293208,MDEyOklzc3VlQ29tbWVudDI1MjI5MzIwOA==,10050469,2016-10-07T16:07:00Z,2016-10-07T16:07:41Z,MEMBER,"> but I feel like it searches for the coordinate with the same name as the supplied dimension. Probably searching for a coordinate containing only the dimension specified in resample could do the trick.

Note that this is not limited to `.resample()` but also to `.sel()`. Indexing over a non-dimension coordinate is something I also needed to do (https://groups.google.com/forum/#!searchin/xarray/fabien%7Csort:relevance/xarray/KTlG2snZabg/5V1vs3ODBAAJ). But after looking into the implementation, I understand that it's quite complicated for xarray to keep track of all possible coordinates linked to one specific dimension. And what if there are more than one coordinate?
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,181534708
https://github.com/pydata/xarray/issues/1041#issuecomment-252150731,https://api.github.com/repos/pydata/xarray/issues/1041,252150731,MDEyOklzc3VlQ29tbWVudDI1MjE1MDczMQ==,1217238,2016-10-07T04:32:25Z,2016-10-07T04:32:25Z,MEMBER,"Does `a2.resample(""D"", ""time"")` work in your second example? I agree that the error message could be more helpful here, but I think this is working as intended -- if you only supply the name of a dimension with integer labels, xarray has no way to know how what times to use for resampling. 
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,181534708