"David G. Johnston" <david.g.johnston@gmail.com> writes:
> On Thu, Dec 5, 2019 at 3:18 PM John Lumby <johnlumby@hotmail.com> wrote:
>> whereas
>> select count(DISTINCT ( parent_id , name) ) from mytable
>> is accepted.
> Correct, converting the two individual columns into a "tuple" allows the
> default tuple distinct-making infrastructure to be used to execute the
> query.
Yeah. This might be more intelligible if it were written
select count(DISTINCT ROW(parent_id, name) ) from mytable
However, the SQL committee in their finite wisdom have decreed that
the ROW keyword is optional. (As long as there's more than one
column expression; the need for that special case is another reason
why omitting ROW isn't really a nice thing to do.)
regards, tom lane