平常工作,尤其是面试中经常遇到这样一个问题,查询表A中31到40条的记录,ID可能是不连续的。

    1. 如果ID连续

      select * from A where ID between 31 and 40

       

    2. 如果ID不连续,提供三种写法

       

      --两次对表A查询效率较低
      select top 10 * from A where ID not in (select top 30 ID from A)

       

      --外层查询没有对表A查询,效率大有提高
      select top 10 * from (select top 40 * from A order by ID) as t order by t.ID desc

       

      --ROW_NUMBER()函数效率更高,sqlserver2005以及以上版本中才可以使用
      select * from (select ROW_NUMBER() over(order by ID) as 'sequence',A.*  from A ) as t where t.sequence between 31 and 40

        


 

 

@金色海洋(jyk)阳光男孩

--如果是海量数据做查询的话,这个是更高效的,这个不错
select
top 10 * from A where id in (select top 10 id from (select top 40 id from A order by ID desc) as t order by t.ID ) order by A.ID desc

 

       @害怕飞的鸟

--这哥们给出了sql2012的新写法,我机器上没装这么前卫的工具,在我老大的机器上测试可行,性能效率暂不明确
SELECT * FROM A Order by ID OFFSET 30 ROWS FETCH NEXT 10 ROWS ONLY

 from:http://www.cnblogs.com/wenyang-rio/archive/2012/10/25/2738499.html

 posted on 2013-01-26 14:17  纳米程序员  阅读(257)  评论(0编辑  收藏  举报