If, for a given table, the value of a column c is determined by the values of columns (a, b), is there a way to declare this fact to postgres??
I'm asking because the examples given for group by mention that, if you group by the primary key, you can use any other column of the table in the select-list, because it will be functionally dependent on the primary key. However, I have not been able to find anything in the documentation that extends this to other columns or column combinations.