e.g. Selecting distinct date from date column using Magic ETL?
I believe you can achieve this via either the edit data -> remove duplicates or edit columns -> group by transformations. Your preference might be driven by what you want to do with the data that gets thrown out. Selecting just the date column and removing duplicates should give you a set of distinct dates in the original DataSet. Using group by lets you attach aggregates based on the other columns in addition to getting the distinct set.
@magicdust, did cr1ckt's reply help you out?