数据库_06_分组查询

#进阶5：分组查询

语法：
select 分组函数，列（要求出现在group by的后面）
from 表
【where 筛选条件】
group by 分组的列表
【order by 子句】

注：
查询列表必须特殊，要求是分组函数和group by后出现的字段

特点：
1.分组查询筛选分为两类
分组前筛选原始表 group by子句的前面 where
分组后筛选分组后的结果集 group by子句的后面 having
2.分组函数做条件肯定放在having子句中
3.能用分组前筛选的，优先考虑
4.支持单个字段分组，多个字段分组，表达式或函数
5.也可以添加排序（放在最后）

#案例1：查询每个工种的最高工资

SELECT MAX(salary) 最高工资,job_id 工种 
FROM employees 
GROUP BY job_id 
ORDER BY 最高工资;

#案例2：查询每个位置上的部门个数

SELECT COUNT(*),location_id 
FROM departments 
GROUP BY location_id;

#添加分组前筛选条件
#案例1：查询邮箱中包含a字符的，每个部门的平均工资

SELECT AVG(salary),department_id
FROM employees 
WHERE email LIKE '%a%'
GROUP BY department_id;

#案例2：查询有奖金的每个领导手下员工的最高工资

SELECT MAX(salary),manager_id
FROM employees
WHERE commission_pct IS NOT NULL
GROUP BY manager_id;

#添加分组后的筛选条件
#案例1：查询哪个部门的员工个数>2

SELECT COUNT(*),department_id
FROM employees
GROUP BY department_id
HAVING COUNT(*)>2;

#案例2：查询每个工种有奖金的员工的最高工资>12000的工种编号和工资
#1.查询每个工种有奖金的最高工资

SELECT MAX(salary),job_id
FROM employees
WHERE commission_pct IS NOT NULL
GROUP BY job_id;

#2.由1继续筛选，最高工资>12000

扫描二维码关注公众号，回复： 12323278 查看本文章

SELECT MAX(salary),job_id
FROM employees
WHERE commission_pct IS NOT NULL
GROUP BY job_id
HAVING MAX(salary)>12000;

#案例3：查询领导编号>102的每个领导手下的最低工资>5000的领导编号是哪个，以及其最低工资

SELECT MIN(salary),manager_id
FROM employees
WHERE manager_id>102
GROUP BY manager_id
HAVING MIN(salary)>5000;

#按表达式或函数分组
#案例：按员工姓名的长度分组，查询每一组的员工个数，筛选员工个数>5的有哪些

SELECT COUNT(*),LENGTH(last_name)
FROM employees
GROUP BY LENGTH(last_name)
HAVING COUNT(*)>5;

#按多个字段分组
#案例：查询每个部门每个工种的员工的平均工资

SELECT AVG(salary),department_id,job_id
FROM employees
GROUP BY department_id,job_id;/*两者一样分为一组*/

#添加排序
#案例：查询每个部门每个工种的员工的平均工资，并按工资高低显示

SELECT AVG(salary),department_id,job_id
FROM employees
GROUP BY department_id,job_id
ORDER BY AVG(salary) DESC;

#测验

SELECT MAX(salary),MIN(salary),AVG(salary),SUM(salary),job_id
FROM employees
GROUP BY job_id
ORDER BY job_id;

SELECT MIN(salary),manager_id
FROM employees
WHERE manager_id IS NOT NULL
GROUP BY manager_id
HAVING MIN(salary)>=6000;

SELECT department_id,AVG(salary),COUNT(*)
FROM employees
GROUP BY department_id
ORDER BY AVG(salary);

SELECT COUNT(*),job_id
FROM employees
GROUP BY job_id;

数据库_06_分组查询

猜你喜欢