创建子查询的麻烦

Question

我在使用子查询创建查询时遇到了一些问题。我必须使用子查询，因为这是学校的作业。

我有两张桌子，一张是员工桌子，一张是部门桌子。从这些表中，我必须返回收入超过部门平均工资的员工。

这就是我目前所拥有的：

SELECT 
    ename,
    salary,
    (SELECT 
            department.depname
        FROM
            department
        WHERE
            department.depno = employee.depno) AS depname,
    (SELECT 
            AVG(salary)
        FROM
            employee
        WHERE
            employee.depno IN (SELECT 
                    department.depno
                FROM
                    department)) avgSalary
FROM
    employee
WHERE
    salary > (SELECT 
            AVG(salary)
        FROM
            employee
        WHERE
            employee.depno IN (SELECT 
                    department.depno
                FROM
                    department))

返回：

'JONES', '2975.00', 'RESEARCH', '2073.214286'
'BLAKE', '2850.00', 'SALES', '2073.214286'
'CLARK', '2450.00', 'ACCOUNTING', '2073.214286'
'SCOTT', '3000.00', 'RESEARCH', '2073.214286'
'KING', '5000.00', 'ACCOUNTING', '2073.214286'
'FORD', '3000.00', 'RESEARCH', '2073.214286'

但是平均工资返回所有工资的平均值。但我需要每个部门都这样做。

我知道如何通过执行以下查询来做到这一点：

SELECT 
    department.depname,
    (SELECT 
            AVG(salary)
        FROM
            employee
        WHERE
            department.depno = employee.depno) avgSalary
FROM
    department

返回：

ACCOUNTING  2916.666667
RESEARCH    2175.000000
SALES       1566.666667
OPERATIONS  

但我不知道如何组合它们，或者不使用连接是否可以。

任何帮助将不胜感激

编辑：

Table: employee
Columns:
empno decimal(4,0) PK 
ename varchar(10) 
efunction varchar(10) 
boss decimal(4,0) 
employed date 
salary decimal(7,2) 
commission decimal(7,2) 
depno decimal(2,0)

Table: department
Columns:
depno decimal(2,0) PK 
depname varchar(14) 
location varchar(13)

Answer 1

有点像这样...？

mysql> select   *
    -> from     employee e
    -> where    e.employee_salary > (
    ->             select   avg(employee_salary) avg
    ->             from     employee ee
    ->             -- this is how you join them without a join...
    ->             where    ee.department_id=e.department_id
    ->          );
+-------------+---------------+---------------+-----------------+
| employee_id | department_id | employee_name | employee_salary |
+-------------+---------------+---------------+-----------------+
|           2 |             1 | employee 2    |              40 |
|           6 |             1 | employee 6    |              50 |
+-------------+---------------+---------------+-----------------+
2 rows in set (0.04 sec)

在哪里……

mysql> select * from employee;
+-------------+---------------+---------------+-----------------+
| employee_id | department_id | employee_name | employee_salary |
+-------------+---------------+---------------+-----------------+
|           1 |             1 | employee 1    |              20 |
|           2 |             1 | employee 2    |              40 |
|           3 |             1 | employee 3    |              30 |
|           4 |             1 | employee 4    |              30 |
|           5 |             1 | employee 5    |              30 |
|           6 |             1 | employee 6    |              50 |
+-------------+---------------+---------------+-----------------+
6 rows in set (0.00 sec)

没有连接...

mysql> select   e.*,
    ->
    ->          (select  department_name
    ->           from    department d
    ->           where   d.department_id=e.department_id) as department_name,
    ->
    ->          (select   avg(employee_salary) avg
    ->             from     employee ee
    ->             where    ee.department_id=e.department_id) as department_average
    ->
    -> from     employee e
    ->
    -> where    e.employee_salary > (
    ->             select   avg(employee_salary) avg
    ->             from     employee ee
    ->             where    ee.department_id=e.department_id
    ->          );
+-------------+---------------+---------------+-----------------+-----------------+--------------------+
| employee_id | department_id | employee_name | employee_salary | department_name | department_average |
+-------------+---------------+---------------+-----------------+-----------------+--------------------+
|           2 |             1 | employee 2    |              40 | department 1    | 33.333333333333336 |
|           6 |             1 | employee 6    |              50 | department 1    | 33.333333333333336 |
+-------------+---------------+---------------+-----------------+-----------------+--------------------+
2 rows in set (0.00 sec)

除非真的有必要，否则您真的应该避免在选择中进行这样的连接，因为在现实生活中最好只使用连接（对于更大数据集的效率/性能）。不过，where 中的子查询是完全合适的。