在生产环境中，我们有的列是不允许出现重复值的，亦或是某两列不允许同时重复，但由于前端未做限制，或者没限制住，出现了单列重复值，或者两列本应组成唯一组合却也出现重复，这两种情况都是不允许的。现在由于前端应用限制不住，要做删除操作后，添加唯一索引，从数据库层面进行限制，以下是处理过程：
mysql> select * from aixuan1;
+----+------+-------+
| id | text | text1 |
+----+------+-------+
| 1 | aa | 11 |
| 2 | bb | 22 |
| 3 | cc | 33 |
| 4 | cc | 44 |
| 5 | bb | 22 |
| 6 | aa | 11 |
| 7 | dd | 55 |
+----+------+-------+
7 rows in set (0.00 sec)
text字段全部重复的有：
mysql> select * from aixuan1 where text in (select text from aixuan1 GROUP BY text having count(*) > 1);
+----+------+-------+
| id | text | text1 |
+----+------+-------+
| 1 | aa | 11 |
| 2 | bb | 22 |
| 3 | cc | 33 |
| 4 | cc | 44 |
| 5 | bb | 22 |
| 6 | aa | 11 |
+----+------+-------+
6 rows in set (0.00 sec)
筛选出text单列重复值
select * from aixuan1 where `text` in (select `text` from aixuan1 GROUP BY `text` having count(*) > 1) and id not in (select min(id) from aixuan1 group by text having count(*)>1)
+----+------+-------+
| id | text | text1 |
+----+------+-------+
| 4 | cc | 44 |
| 5 | bb | 22 |
| 6 | aa | 11 |
+----+------+-------+
3 rows in set (0.00 sec)
还可以这么查
mysql> select * FROM aixuan1 WHERE id NOT IN ( SELECT temp.mid FROM ( SELECT min(id) as mid FROM aixuan1 em GROUP BY em.text) AS temp);
+----+------+-------+
| id | text | text1 |
+----+------+-------+
| 4 | cc | 44 |
| 5 | bb | 22 |
| 6 | aa | 11 |
+----+------+-------+
3 rows in set (0.00 sec)
筛选出text和text1同时重复的字段：
mysql> select * FROM aixuan1 WHERE id NOT IN ( SELECT temp.mid FROM ( SELECT min(id) as mid FROM aixuan1 em GROUP BY em.text,em.text1) AS temp);
+----+------+-------+
| id | text | text1 |
+----+------+-------+
| 5 | bb | 22 |
| 6 | aa | 11 |
+----+------+-------+
2 rows in set (0.00 sec)
查出来了，删就好办了，把select换成delete就Ok了，具体说保留大的id还是保留小的id那条，只要子查询的id函数用min(id)或者max(id)即可
本文出自 “岁伏” 博客，请务必保留此出处http://suifu.blog.51cto.com/9167728/1796055
