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/837#issuecomment-228419690,https://api.github.com/repos/pydata/xarray/issues/837,228419690,MDEyOklzc3VlQ29tbWVudDIyODQxOTY5MA==,1217238,2016-06-24T18:13:36Z,2016-06-24T18:13:36Z,MEMBER,"Merged via https://github.com/pydata/xarray/commit/0d0ae9d3e766c3af9dd98383ab3b33dfea9494dc
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,151483622
https://github.com/pydata/xarray/pull/837#issuecomment-222873697,https://api.github.com/repos/pydata/xarray/issues/837,222873697,MDEyOklzc3VlQ29tbWVudDIyMjg3MzY5Nw==,1217238,2016-06-01T02:07:42Z,2016-06-01T02:07:42Z,MEMBER,"This looks good to me. The test could probably still be made more succinct but this covers the essentials. I like that it provides an integration test for using rec arrays.
@jhamman any more comments before I merge?
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,151483622
https://github.com/pydata/xarray/pull/837#issuecomment-222017597,https://api.github.com/repos/pydata/xarray/issues/837,222017597,MDEyOklzc3VlQ29tbWVudDIyMjAxNzU5Nw==,5384661,2016-05-26T22:53:00Z,2016-05-26T22:53:00Z,CONTRIBUTOR,"Added bugfix entry in what's new
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,151483622
https://github.com/pydata/xarray/pull/837#issuecomment-218359954,https://api.github.com/repos/pydata/xarray/issues/837,218359954,MDEyOklzc3VlQ29tbWVudDIxODM1OTk1NA==,2443309,2016-05-11T04:38:51Z,2016-05-11T04:38:51Z,MEMBER,"@maciekswat - this is almost there. @shoyer and I each have comments remaining that should be addressed and then we need a bug fix entry in ""what's new"". Thanks!
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,151483622
https://github.com/pydata/xarray/pull/837#issuecomment-216556267,https://api.github.com/repos/pydata/xarray/issues/837,216556267,MDEyOklzc3VlQ29tbWVudDIxNjU1NjI2Nw==,5384661,2016-05-03T15:00:17Z,2016-05-03T15:00:17Z,CONTRIBUTOR,"Added a comment explaining why we need to catch typeError in array_equiv
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,151483622
https://github.com/pydata/xarray/pull/837#issuecomment-216338679,https://api.github.com/repos/pydata/xarray/issues/837,216338679,MDEyOklzc3VlQ29tbWVudDIxNjMzODY3OQ==,5384661,2016-05-02T19:35:04Z,2016-05-02T19:35:04Z,CONTRIBUTOR,"I simplified the test - please feel free to do further edits, also added doc string explaining what the test checks for - essentially it checks if DataArray subtraction behaves correctly when dimension is a structured numpy array.
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,151483622
https://github.com/pydata/xarray/pull/837#issuecomment-216313208,https://api.github.com/repos/pydata/xarray/issues/837,216313208,MDEyOklzc3VlQ29tbWVudDIxNjMxMzIwOA==,5384661,2016-05-02T18:06:38Z,2016-05-02T18:06:38Z,CONTRIBUTOR,"just fixed the formatting according to PEP8 guidelines
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,151483622
https://github.com/pydata/xarray/pull/837#issuecomment-216310274,https://api.github.com/repos/pydata/xarray/issues/837,216310274,MDEyOklzc3VlQ29tbWVudDIxNjMxMDI3NA==,1217238,2016-05-02T17:56:03Z,2016-05-02T17:56:03Z,MEMBER,"@maciekswat awesome -- thanks!
can you please run your code through a PEP8 checker like pep8 or pyflakes? some of your spacing is off.
also, please add a note to ""What's new"" giving yourself credit for the bug fix :).
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,151483622
https://github.com/pydata/xarray/pull/837#issuecomment-216307904,https://api.github.com/repos/pydata/xarray/issues/837,216307904,MDEyOklzc3VlQ29tbWVudDIxNjMwNzkwNA==,5384661,2016-05-02T17:48:02Z,2016-05-02T17:48:02Z,CONTRIBUTOR,"I have rewrote array_equiv fcn from ops.py to use single comparison (the one which tests from null is inside try/except clause). Also added test case to test_DataArray.py
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,151483622
https://github.com/pydata/xarray/pull/837#issuecomment-215291343,https://api.github.com/repos/pydata/xarray/issues/837,215291343,MDEyOklzc3VlQ29tbWVudDIxNTI5MTM0Mw==,1217238,2016-04-28T02:36:01Z,2016-04-28T02:36:01Z,MEMBER,"Thanks for looking into this!
Just to summarize, the issue is that `isnull` doesn't work on structured dtypes:
```
>>> pd.isnull(p_data)
TypeError: Not implemented for this type
```
The problem with your current patch is that it will actually compare all array elements twice. Because checks for equality are done quite often in xarray, this will have prohibitive performance cost.
So, we'll either need to check for structured dtypes, or catch `TypeError` exceptions and try doing the comparison without checking for nulls.
Also, we need tests, to verify that the bug is fixed.
","{""total_count"": 0, ""+1"": 0, ""-1"": 0, ""laugh"": 0, ""hooray"": 0, ""confused"": 0, ""heart"": 0, ""rocket"": 0, ""eyes"": 0}",,151483622