sql: insert on duplicate key update (update only specific columns if a unique key already exists; insert new rows if unique key does not exist)

gistfile1.sql

INSERT INTO _customers (`User ID`, Username, `Billing: Full Name` )
SELECT `User ID`, Username, `Billing: Full Name`
FROM _customers_import
ON DUPLICATE KEY UPDATE Username=values(Username), `Billing: Full Name`=values(`Billing: Full Name`)