由于一次操作失误，给表中插入了多条重复的数据，所以就需要删除重复的数据只保留一条，一时想不到好方法，各种查资料，终于找到了，特意写到这里，方便以后自己用~
查询：
select A.n_PatentID,A.s_OtherFileName,A.s_FileName FROM TB_OtherFiles A, (SELECT n_PatentID,s_OtherFileName,s_FileName FROM TB_OtherFiles GROUP BY n_PatentID,s_OtherFileName,s_FileName HAVING COUNT(1)>=2 ) AS B WHERE A.s_OtherFileName=B.s_OtherFileName AND A.n_PatentID=B.n_PatentID AND A.s_FileName=B.s_FileName AND A.n_PatentID >0 order by n_PatentID
删除：
DELETE T FROM (SELECT n_PatentID,s_OtherFileName,s_FileName,s_recycle, ROW_NUMBER()OVER(PARTITION BY n_PatentID,s_OtherFileName,s_FileName ORDER BY RAND()) AS RN FROM TB_OtherFiles where n_PatentID>0) AS t WHERE RN>1
原文：http://www.cnblogs.com/yixuanhan/p/6677530.html