«

mysql group by having 实例代码

时间:2024-3-2 00:22     作者:韩俊     分类: Mysql


mysql group by having 实例

注意:使用group by的时候,SELECT子句中的列名必须为分组列。 如下实例必须包括name列名,因为name是作为group by分组的条件。

实例:

我的数据库中有一张员工工作记录表,表中的数据库如下:

mysql> SELECT * FROM employee_tbl;
+------+------+------------+--------------------+
| id  | name | work_date | daily_typing_pages |
+------+------+------------+--------------------+
|  1 | John | 2007-01-24 |        250 |
|  2 | Ram | 2007-05-27 |        220 |
|  3 | Jack | 2007-05-06 |        170 |
|  3 | Jack | 2007-04-06 |        100 |
|  4 | Jill | 2007-04-06 |        220 |
|  5 | Zara | 2007-06-06 |        300 |
|  5 | Zara | 2007-02-06 |        350 |
+------+------+------------+--------------------+
7 rows in set (0.00 sec)

需求一:现在我需要查找出每个员工工作了多少天,这时就需要用到group by语句:

mysql> SELECT name, COUNT(*)
  -> FROM  employee_tbl 
  -> GROUP BY name;
+------+----------+
| name | COUNT(*) |
+------+----------+
| Jack |    2 |
| Jill |    1 |
| John |    1 |
| Ram |    1 |
| Zara |    2 |
+------+----------+
5 rows in set (0.04 sec)

需求二:我现在要查找工作天数大于一天的所有员工,这个时候就需要用到group by having语句

mysql> SELECT name, COUNT(*)
  -> FROM  employee_tbl 
  -> GROUP BY name having COUNT(*)>1 ;
+------+----------+
| name | COUNT(*) |
+------+----------+
| Jack |    2 |
| Zara |    2 |
+------+----------+
2 rows in set (0.04 sec)

感谢阅读,希望能帮助到大家,谢谢大家对本站的支持!

标签: mysql

热门推荐