* moved unique value naming to datasource (credit: @bergquist) * merge rows based on same column-values and empty values * expanded tests
