- orbis_match_api_call.py: script to run that executes query againt the Orbis Match API
- src/utils.py: file containing "OrbisMatchAPIQueryClient" class with methods used in "orbis_match_api_call.py" script
- conf/config.py: file containing Orbis API credentials (expecting name "orbis_api_key")
- In the "orbis_match_api_call.py", make sure the variables set within the 'if name == "main"' block are set correctly (e.g., "path_to_data", "company_name_col", etc.).
- Make sure you are using the appropriate read and write methods for the environment you are using. If the data is parquet format, use the "load_data_spark_parquet" and "write_data_spark_parquet" methods. If csv, use the "load_data_spark_csv" and "write_data_pandas_csv".
- Execute the "orbis_match_api_call.py" script.