id,node_id,number,state,locked,title,user,body,created_at,updated_at,closed_at,merged_at,merge_commit_sha,assignee,milestone,draft,head,base,author_association,auto_merge,repo,url,merged_by
307673781,MDExOlB1bGxSZXF1ZXN0MzA3NjczNzgx,3221,closed,0,Allow invalid_netcdf=True in to_netcdf(),13190237,"Hi all,

I prepared a little PR which could close #2243 and would allow for a IMO clean way of writing data with complex dtypes (and others).

What do you think? 

TODOs:
----------
 - [X] Closes #2243
 - [X] Tests added
 - [x] Passes `black . && mypy . && flake8`
 - [x] Fully documented, including `whats-new.rst` for all changes and `api.rst` for new API
",2019-08-15T11:32:56Z,2019-08-22T20:20:42Z,2019-08-22T20:12:11Z,2019-08-22T20:12:11Z,76d4a6710a1e22c6af112d0c3f5a81a7b435448d,,,0,50b52a07bdaa82e6dae8324b6eefc2d134612468,2fdcbb6feccb01a97b8b51160b81e238486fff00,CONTRIBUTOR,,13221727,https://github.com/pydata/xarray/pull/3221,
310140126,MDExOlB1bGxSZXF1ZXN0MzEwMTQwMTI2,3244,closed,0,Make argmin/max work lazy with dask,13190237,"As @shoyer pointed out in https://github.com/pydata/xarray/issues/3237, nanargmax/min from numpy or dask should be used when not working on object arrays. Also, nanargmin/max were added to the nputils module so they should be using bottleneck if available.


 - [x] Closes #3237
 - [x] Tests added
 - [x] Passes `black . && mypy . && flake8`
 - [x] Fully documented, including `whats-new.rst` for all changes and `api.rst` for new API
",2019-08-22T20:55:49Z,2019-10-25T11:53:42Z,2019-09-06T23:15:19Z,2019-09-06T23:15:19Z,0a046dbdeff409728b4e9bf55fba9d2aae9acd07,,,0,7fd62b827f1ea504618efde42d75c502e5e6f73c,3f4b0250e386f08233f4e11dc1cd4b12cfa953b0,CONTRIBUTOR,,13221727,https://github.com/pydata/xarray/pull/3244,