Hi!

SELECT
    e.emp_no
FROM employee e
    JOIN employee_project p ON p.emp_no = e.emp_no
GROUP BY e.emp_no
HAVING MAX((SELECT
                SUM(e2.salary)
            FROM employee e2
                JOIN employee_project p2 ON p2.emp_no = e2.emp_no
            WHERE e2.emp_no = e.emp_no AND
                  p2.proj_id IS NOT DISTINCT FROM p.proj_id)) <> 0

This is accepted by Firebird (also by Oracle).

In the subselect p.proj_id should not be accepted - group does not contains it, and has no aggregate on it. What value the engine choose when it runs, and why?

Is this equivalent with it?

SELECT
    e.emp_no
FROM employee e
    JOIN employee_project p ON p.emp_no = e.emp_no
GROUP BY e.emp_no
HAVING (SELECT
                SUM(e2.salary)
            FROM employee e2
                JOIN employee_project p2 ON p2.emp_no = e2.emp_no
            WHERE e2.emp_no = e.emp_no AND
                  p2.proj_id IS NOT DISTINCT FROM MAX(p.proj_id)) <> 0