Error while writing unit tests for nested function in python

Question

I have a function which reads data from db, modifies it and writes it back to another db. I am trying to write unit tests for this

import file1
import file2
def update_func(query):
 max_date = file1.get_cutover_date()
 data_dataframe = file2.get_dataframe_from_db(query)
 transformed_dataframe = modify_df(data_dataframe)
 df1 = transformed_dataframe.filter(transformed_dataframe['date']>max_date)
 write_to_db(df1)

I tried to patch the function calls.

@patch('file2.get_dataframe_from_db')
@patch('file1.get_cutover_date')
test_update_func(patched_get_cutover, patched_get_df):
 patched_get_cutover.return_value = '2015-01-01 00:00:00'
 response = [Row(id=1,date='2016-10-12 00:00:00')]
 response_df = spark.createDataFrame(response)
 patched_get_df.return_value = response_df
 update_func('test_query')
 patched_get_cutover.assert_called
 patched_get_df.assert_called

But I am getting TypeError: '>' not supported between instances of MagicMock and MagicMock

• User051209: May be you are patching functions in file1 and file2 while update_func() is in an other file. Try to change @patch('file2.get_dataframe_from_db') to @patch('<file-of-update_func>.get_dataframe_from_db'). But I'm not sure.
• RedFox: All my functions are in different files. Updated my question to include imports
• User051209: Change the import to: from file1 import get_cutover_date and from file2 import get_dataframe_from_db. After that change the patches to: @patch('<file-of-update_func>.get_dataframe_from_db') and @patch('<file-of-update_func>.get_cutover_date'). I think that with these changes you can solve your problem.
• Kashyap: 1. Post full stack trace, is the exception thrown from transformed_dataframe['date']>max_date or elsewhere? 2. If you're using the values returned by a patched function, then you need to ensure the value it returns is "usable". E.g. in this case you're using the output of file2.get_dataframe_from_db. So patch it to return a proper dataframe if you really want to use it.