test_import_cleaning.py

### Testing of import_cleaning function
import unittest
from import_cleaning import *
import pickle

data = pd.read_pickle('dataframe.pkl')
columns =data.columns

# unittest test case
class StringConverting(unittest.TestCase):

    def test_convert_column_types(self):
        # Creating a sample DataFrame for testing
        sample_data = pd.DataFrame({
            'sample_col1': ['NaN', 'string1', 'string2'],
            'sample_col2': ['string3', 'NaN', 'string4']
        })
        
        # Call the function to be tested
        result = convert_column_types(sample_data, ['sample_col1', 'sample_col2'])
        
        # Check if 'Not Applicable' is replaced with pd.NA and column type is 'string'
        #if we remove a letter in the expected output eg type "sting1", we get an error
        self.assertEqual(result['sample_col1'].tolist(), [pd.NA, 'string1', 'string2']) 
        self.assertEqual(result['sample_col2'].tolist(), ['string3', pd.NA, 'string4'])
        self.assertEqual(result['sample_col1'].dtype, 'string')
        self.assertEqual(result['sample_col2'].dtype, 'string')
        
        # Check if the number of columns in our actual dataframe matches our desired output
        assert (len(columns) == 16)

        # Check that no more 'NaN' nor 'Not Applicable' are in the actual dataframe    
        for col in data:
            self.assertFalse(((data == 'Not Applicable').any().any()) 
                             &(data == 'NaN').any().any())

if __name__ =='__main__':
    unittest.main()