Testing for

Question

Hi,
Is there a more efficient way of writing this SQL, I know I can do it with 1 exist using an aggregate function such as count(1), just wonder if there's any other way;
select 1
from tableA a
where :param1 = 'true'
and exists (select 1
from tableB b
where b.colx = a.colx
or :param1 = 'false'
and not exists (select 1
from tableB b
where b.colx = a.colx
cheers

Answer 1

Hmmm... Then it ought to be possible to push the CASE into the where clause.
select distinct a.*
from tableA a, tableB b
where a.colx = b.colx(+)
AND 1 = (case
         when :param1 = 'true' and b.colx IS NOT NULL THEN 1
         when :param1 = 'false' and b.colx IS NULL    THEN 1
                                                      ELSE 0
         end)The DISTINCT should deal with those cases where b has multiple rows.