急急急 ,如何删除重复数据
有ceshi表 :


需要把表中TITLE,BOOK_SEARC这两个字段重复的记录删除,只保留一条数据 .

写了这个SQL语句 :
delete ceshi where (TITLE,BOOK_SEARC) in (select TITLE,BOOK_SEARC from ceshi group by TITLE,BOOK_SEARC having count(*)>1)
and BOOK_BARCO not in(select min(BOOK_BARCO) from ceshi group by TITLE,BOOK_SEARC having count(*)>1)

会提示语法错误,请各位大侠指点一下.该如何实现这个功能?还有就是我表中的数据是百万条的记录,如何删除才能提高执行效率,缩短时间.谢谢各位了!
[解决办法]

delete ceshi 
where BOOK_BARCO in 
(select BOOK_BARCO 
from (select *,rn=row_number() over(partition by TITLE,BOOK_SEARC order by BOOK_BARCO) from tb)t
where rn>1
)

[解决办法]

 DELETE ceshi
 WHERE  EXISTS ( SELECT 1
                 FROM   ( SELECT    TITLE ,
                                    BOOK_SEARC
                          FROM      ceshi
                          GROUP BY  TITLE ,
                                    BOOK_SEARC
                          HAVING    COUNT(*) > 1
                        ) b
                 WHERE  ceshi.title = b.title
                        AND ceshi.book_searc = b.book_searc )

[解决办法]
再改一下,是大于1的，都删除掉：

with t
as
(
select *,
       row_number() over(paritition by TITLE,BOOK_SEARC 
                         order by TITLE) as rown
from ceshi
)


delete from t
where rown > 1

[解决办法]

select *
into temp_ceshi
from
(
select *,ROW_NUMBER() over(partition by TITLE,BOOK_SEARC ,order by getdate()) rn
from ceshi
) t
where rn=1

用上面这个语句把结果存放到 temp_ceshi表。这个表里就是你需要的结果了。然后把表名改成ceshi。重复数据就去掉了