blob: 99fc8541449a0235f62456d9574a1a28d34a9dbf [file] [log] [blame]
-- left outer join
-- group by on null producing side
-- having clause with the case statement
-- aggregation on top of subquery
-- having clause in the outer query
select x3
from
(
select t2.a2,
t2.b2,
count(*)
from
t1 left outer join t2 on t1.a1 = t2.a2
group by
t2.a2,
t2.b2
having
case when t2.a2 is null then 100 else t2.a2 end > 5
) as dt(x1, x2, x3)
group by
x3
having x3 > 1
order by
x3;