When a SubQuery is used in a FROM clause instead of a Table name, Oracle executes
the subquery and then uses the resulting rows as a view in the FROM
clause. It's a very useful method for performing reports with various types of counts.
Consider the following classic query, which counts the number of awards
accepted or rejected by employees. It does an outer join to take care of
employees who might have only accepted or only rejected awards. This will
give one row per employee in an emp_status table:
select a.id, nvl(accepted, 0) TotAccepted,
nvl(rejected, 0) TotRejected,
fname
from emp_status a, (select count(*) accepted, id,
code , status
from emp_status
where code = 'AWARD'
and status = 'A'
group by id, code, status ) b ,
(select count(*) rejected, id,
code , status
from emp_status
where code = 'AWARD'
and status = 'R'
group by id, code, status) c ,
(select id, fname
from employee) d
where a. code = 'AWARD'
and a.status in ('R', 'A')
and a.id = b.id(+)
and a.id = c.id(+)
and a.id = d.id
group by a.id, accepted, rejected, fname