mysql删除重复记录，保存Id最小的一条

Posted 2020-08-07 风与叶子

方法1：
1、创建一个临时表，选取需要的数据。
2、清空原表。
3、临时表数据导入到原表。
4、删除临时表。
mysql> select * from student;
+----+------+
| ID | NAME |
+----+------+
| 11 | aa |
| 12 | aa |
| 13 | bb |
| 14 | bb |
| 15 | bb |
| 16 | cc |
+----+------+
6 rows in set

mysql> create temporary table temp as select min(id),name from student group by name;
Query OK, 3 rows affected
Records: 3 Duplicates: 0 Warnings: 0

mysql> truncate table student;
Query OK, 0 rows affected

mysql> insert into student select * from temp;
Query OK, 3 rows affected
Records: 3 Duplicates: 0 Warnings: 0

mysql> select * from student;
+----+------+
| ID | NAME |
+----+------+
| 11 | aa |
| 13 | bb |
| 16 | cc |
+----+------+
3 rows in set

mysql> drop temporary table temp;
Query OK, 0 rows affected
这个方法，显然存在效率问题。

---

方法2：按name分组，把最小的id保存到临时表，删除id不在最小id集合的记录，如下：
mysql> create temporary table temp as select min(id) as MINID from student group by name;
Query OK, 3 rows affected
Records: 3 Duplicates: 0 Warnings: 0

mysql> delete from student where id not in (select minid from temp);
Query OK, 3 rows affected

mysql> select * from student;
+----+------+
| ID | NAME |
+----+------+
| 11 | aa |
| 13 | bb |
| 16 | cc |
+----+------+
3 rows in set 

---

方法3：直接在原表上操作，容易想到的sql语句如下：

mysql> delete from student where id not in (select min(id) from student group by name);
执行报错：1093 - You can\'t specify target table \'student\' for update in FROM clause
原因是：更新数据时使用了查询，而查询的数据又做了更新的条件，mysql不支持这种方式。
怎么规避这个问题？
再加一层封装，如下：
mysql> delete from student where id not in (select minid from (select min(id) as minid from student group by name) b);
Query OK, 3 rows affected

mysql> select * from student;
+----+------+
| ID | NAME |
+----+------+
| 11 | aa |
| 13 | bb |
| 16 | cc |
+----+------+
3 rows in set