Originally posted by Marilyn de Queiroz:
I think that if I did
select (distinct *)
I would get every row, right?
Correct.
Do you think that "group by" is a better way than "distinct" in general because I can't do a "distinct" on just one field?
In general, it is preferable to use a standard SQL construct rather than trying to create your own because that is what database vendors optimize for. Don't worry about the extra sort. Sorting 1 row (in each group) takes an almost trivial amount of time.