How do you keep only one record of duplicated records?

My pipeline is a simple pipeline that execute an Oracle query then parse the data to a CSV file. The result of the query gives some of the duplicate records (Same ID), that because the database has ...