"Colin 't Hart" <colin@sharpheart.org> writes:
> On 11 November 2013 14:34, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> No, and it probably won't ever be, since the semantics aren't the same.
>> EXCEPT/INTERSECT imply duplicate elimination.
> Can't we just use DISTINCT for that?
If you have to do a DISTINCT it's not clear to me that you're going to get
much win.
(The bigger picture here is that pretty much zero optimization effort has
been spent on EXCEPT/INTERSECT, because they're just not used that much
compared to other places where we could put that effort.)
regards, tom lane