删除表中的重复行，首先重复出现的行必须满足的条件是与其他行的值相同，另外还需要判断空行，再进一步进行处理 

表结构

SQL>desc test1

Name                 Null?     Type

---------------------------------------------------

ID                                    NUMBER

DEPARTMENT                    VARCHAR2(33)

如下共有11行数据 第3，5行和第7,9行数据相同 现在删除重复的行

SQL> select * from test1;

 ID DEPARTMENT
---------- ---------------------------------
  5 afa啊打发add阿发afda
  6 啊发发afda阿佛奥
  1 111
  3 售后
  1 111
  7 ddddd
  9
  8 小的
  9
  4 afa阿凡达
 10 afafa

11 rows selected.

SQL>

如下只删除了id=1 没有删除id=9的重复行

SQL> delete test1 a 
  2  where a.rowid >
  3  (select min(rowid) from test1 b
  4  where b.id=a.id and b.department=a.department);

1 row deleted.

SQL> select * from test1;

 ID DEPARTMENT
---------- ---------------------------------
  5 afa啊打发add阿发afda
  6 啊发发afda阿佛奥
  1 111
  3 售后
  7 ddddd
  9
  8 小的
  9
  4 afa阿凡达
 10 afafa

10 rows selected.

SQL>

rollback回滚

SQL> rollback;

Rollback complete.

SQL> select * from test1;

 ID DEPARTMENT
---------- ---------------------------------
  5 afa啊打发add阿发afda
  6 啊发发afda阿佛奥
  1 111
  3 售后
  1 111
  7 ddddd
  9
  8 小的
  9
  4 afa阿凡达
 10 afafa

11 rows selected.

SQL>

当值为null 上面的删除用=无效 需要另作判断

SQL> delete test1 a where a.rowid
  2  > (select min(rowid)
  3  from test1 b
  4  where ((b.id=a.id)or(b.id is null and a.id is null))
  5  and ((b.department=a.department)or(b.department is null and a.department is null)));

2 rows deleted.

SQL> select * from test1;

 ID DEPARTMENT
---------- ---------------------------------
  5 afa啊打发add阿发afda
  6 啊发发afda阿佛奥
  1 111
  3 售后
  7 ddddd
  9
  8 小的
  4 afa阿凡达
 10 afafa

9 rows selected.

SQL>