xr.Dataset.drop

[markusritschel]

Currently, drop throws an error if one of the labels doesn't exist. It would be nice to have a parameter in the drop method for optionally ignoring errors like in the pandas.DataFrame.
From the pandas documentation:

errors : {‘ignore’, ‘raise’}, default ‘raise’
If ‘ignore’, suppress error and only existing labels are dropped.

[shoyer]

This sounds fine to me. Any interest in putting together a pull request? See http://xarray.pydata.org/en/stable/contributing.html for guidance.

(It might also make sense to put this in xarray's drop_dims as well, for consistency)