comparison dataframes in assertions.py

[t3rn0]

DISCLAIMER
I came here without deep knowledge of data-detective. Following question is purely based on the contents of the assertions.py file
(https://github.com/Tinkoff/data-detective/blob/master/data-detective-airflow/data_detective_airflow/test_utilities/assertions.py)

Function assert_frame_equal compares two dataframes. How? By computing two scalar values with to_bytes and comparing them. This method has some limitations:

a = pd.DataFrame([1, 1, 2, 4])  
b = pd.DataFrame([3, 2, 4, 3])  
c = pd.DataFrame([2, 4])  
d = pd.DataFrame([1, [1, 2], 4])  
e = pd.DataFrame([{"a": 2}, {"a": 4}])  
     
assert to_bytes(a) == to_bytes(b) == to_bytes(c) == to_bytes(d) == to_bytes(e)  # True

Probably you compare dataframes of the same size or the same type. So there's no chance you'll have dataframes C, D, E in test simultaneously. But having dataframes A and B equal makes one wondering.
Have you considered such cases?

[mebelousov]

You're right that the function compares dataframes with identical columns.
I'm not sure it's a good idea to add this check to assert_frame_equal. It seems that other checks may be required, and the function will become too complicated.