Re: "SELECT ... FROM DUAL" is not quite as silly as it appears

Ashutosh Bapat <ashutosh(dot)bapat(at)enterprisedb(dot)com> writes:
> On Thu, Mar 15, 2018 at 8:57 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> Suppose that, either in the rewriter or early in the planner, we were
>> to replace such cases with nonempty FromExprs, by adding a dummy RTE
>> representing a table with no columns and one row. This would in turn
>> give rise to an ordinary Path that converts to a Result plan, so that
>> the case is handled without any special contortions later.

> Since table in the dummy FROM clause returns one row without any
> column, I guess, there will be at least one row in the output. I am
> curious how would we handle cases which do not return any row
> like

> create function set_ret_func() returns setof record as $$select * from
> pg_class where oid = 0;$$ language sql;
> select set_ret_func();
> set_ret_func
> --------------
> (0 rows)

It'd be the same as now, so far as the executor is concerned:

regression=# explain select set_ret_func();
QUERY PLAN
--------------------------------------------------
ProjectSet (cost=0.00..5.27 rows=1000 width=32)
-> Result (cost=0.00..0.01 rows=1 width=0)
(2 rows)

The difference is that, within the planner, the ResultPath would be
associated with a "real" base relation instead of needing its very
own code path in query_planner().

regards, tom lane