Yes there is! A table should have a primary key, which ensures no two rows in a table are exactly equal. So the simple solution with ‘a except b’ should work.

Yes there is! A table should have a primary key, which ensures no two rows in a table are exactly equal. So the simple solution with ‘a except b’ should work. Lots of otherwise simple things get complicated with duplicates, e.g. try to change or delete one row out of a set of duplicates. A table with duplicates is not even a relation, since relations are defined as sets. SQL is in a weird place because it technically allows duplicates, but many operations are not able to distinguish between duplicates.