在 SQL 中使用 HAVING 子句

Question

I am trying to get the employees whose income is greater than 80% of the total average. The problem is that the HAVING clause doesn't works as it throws the following error.

This is the error in line of the HAVING (column 38):
ORA-00935: group function is nested too deeply

How can I fix the query to use HAVING correctly?

SELECT 
  e.FIRST_NAME || ' ' || e.LAST_NAME as name,
  sum(o.order_total) as income
FROM  EMPLOYEES e, ORDERS o
WHERE e.employee_id = o.sales_rep_id 
GROUP BY e.FIRST_NAME, e.LAST_NAME
HAVING sum(o.order_total)*0.8 > avg(sum(o.order_total))
ORDER BY sum(o.order_total) DESC;

Answer 1

Starting from your existing query, you could use a window function to compute that grand average, and use it for filtering in the having clause:

select 
    e.first_name || ' ' || e.last_name name,
    sum(o.order_total) income
from employees e
inner join orders o on e.employee_id = o.sales_rep_id 
group by e.employee_id, e.first_name, e.last_name
having sum(o.order_total) > 0.8 * avg(sum(o.order_total)) over()
order by sum(o.order_total) desc;

Notes:

• always use explicit joins (with the on keyword) rather than old-school, implicit joins
• I added the id of the employee to the group by clause to prevent wrongly grouping possible homonyms

---

Alternatively, you can use a subquery:

select name, income
from (
    select 
        e.first_name || ' ' || e.last_name name,
        sum(o.order_total) income,
        avg(sum(o.order_total)) over() avg_income
    from employees e
    inner join orders o on e.employee_id = o.sales_rep_id 
    group by e.employee_id, e.first_name, e.last_name
) t
where income > 0.8 * avg_income
order by income desc;