part_4_assert_our_data.py

df_ge = ge.dataset.PandasDataset(df) #Load pd dataframe into great_expectations (GE)

#Define a rule to define the order and fields that should exist
df_ge.expect_table_columns_to_match_ordered_list(['flight_number',
 'date',
 'time_utc',
 'booster_version',
 'launch_site',
 'payload',
 'customer',
 'outcome'])
#flight_number should ALWAYS be unique. We don't want duplicate data in our warehouse
df_ge.expect_column_values_to_be_unique('flight_number')

#flight_number should never be null
df_ge.expect_column_values_to_not_be_null('flight_number')

#flight_number should be an integer
df_ge.expect_column_values_to_be_of_type('flight_number', 'int64')

#Validate our data
validation_results = df_ge.validate()

#If our data is valid validation_results["success"] should return True
if(validation_results["success"]):
    print ('All assertions have passed! :)')
else:
    print ('Some assertions have failed! :(')