Expecting column values to be of type array or list raises ValueError

[sam-hoffman]

Mentioned in Slack

Adding an expectation that a column in Pandas be of type list or array raises an error:

import great_expectations as ge
import pandas as pd


df = pd.DataFrame(
    {'a':[1,2,3,4,5], 
     'b':[[1,2], [3,4], [5], [6,7,8], [9, 10, 11]]}
)
geset = ge.dataset.PandasDataset(df)

#these two work
geset.expect_column_values_to_be_of_type('a', 'int')
geset.expect_column_values_to_be_of_type('a', 'int64')


#these don't work :(
geset.expect_column_values_to_be_of_type('b', 'array')
geset.expect_column_values_to_be_of_type('b', 'list')

Traceback here:

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-11-17211834476d> in <module>()
     11 
     12 #these don't work :(
---> 13 geset.expect_column_values_to_be_of_type('b', 'array')
     14 geset.expect_column_values_to_be_of_type('b', 'list')

/Users/Shared/anaconda3/lib/python3.6/site-packages/great_expectations/data_asset/util.py in f(*args, **kwargs)
     83         @wraps(self.mthd, assigned=('__name__', '__module__'))
     84         def f(*args, **kwargs):
---> 85             return self.mthd(obj, *args, **kwargs)
     86 
     87         f.__doc__ = doc

/Users/Shared/anaconda3/lib/python3.6/site-packages/great_expectations/dataset/pandas_dataset.py in expect_column_values_to_be_of_type(self, column, type_, **kwargs)
    507         else:
    508             res = self._expect_column_values_to_be_of_type__map(
--> 509                 column, type_, **kwargs
    510             )
    511             # Note: this logic is similar to the logic in _append_expectation for deciding when to overwrite an

/Users/Shared/anaconda3/lib/python3.6/site-packages/great_expectations/data_asset/data_asset.py in wrapper(self, *args, **kwargs)
    234 
    235                         else:
--> 236                             raise err
    237 
    238                 else:

/Users/Shared/anaconda3/lib/python3.6/site-packages/great_expectations/data_asset/data_asset.py in wrapper(self, *args, **kwargs)
    221                 if self._config.get("interactive_evaluation", True) or self._active_validation:
    222                     try:
--> 223                         return_obj = func(self, **evaluation_args)
    224 
    225                     except Exception as err:

/Users/Shared/anaconda3/lib/python3.6/site-packages/great_expectations/dataset/pandas_dataset.py in inner_wrapper(self, column, mostly, result_format, *args, **kwargs)
     82             else:
     83                 boolean_mapped_null_values = np.array([True if (value in ignore_values) or (pd.isnull(value)) else False
---> 84                                                        for value in series])
     85 
     86             element_count = int(len(series))

/Users/Shared/anaconda3/lib/python3.6/site-packages/great_expectations/dataset/pandas_dataset.py in <listcomp>(.0)
     82             else:
     83                 boolean_mapped_null_values = np.array([True if (value in ignore_values) or (pd.isnull(value)) else False
---> 84                                                        for value in series])
     85 
     86             element_count = int(len(series))

ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

[eugmandel]

Thank you for the fix, @sam-hoffman ! It is released in 0.9.2.