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
841520137,PR_kwDOAMm_X84yKJQJ,6249,closed,0,Fix pickling issue,10580038,"<!-- Feel free to remove check-list items aren't relevant to your change -->

- [x] Closes #6226
- [x] Added tests for opening netcdf files with cftime axis in parallel

This PR tries to address some updates in the pandas library on how datetime objects are pickled. Since I was told that the datatype of the CFtimeIndex object will always be `object` I just instructed the `__new__` method to receive any additional arguments but not to evaluate them. Which feels a little awkward but does the job. If there are better ways to deal with this please let me know.  @spencerkclark @aidanheerdegen @mathause ",2022-02-07T09:45:46Z,2022-02-10T22:37:54Z,2022-02-10T22:37:38Z,2022-02-10T22:37:37Z,d994273b5d61bbdc2fde5050af5922deb66db274,,,0,6efd3bdf89da98aa429b742f2ec55023276e00b9,44a3e3bb76fc1266cbaef6e6fa84fa0c146af846,CONTRIBUTOR,,13221727,https://github.com/pydata/xarray/pull/6249,