"SQL" uses row_number () to implement group queries

650) this.width=650; "src=" Http://s3.51cto.com/wyfs02/M01/5A/24/wKiom1T3v8zzQQ6rAACos1rMbc4635.jpg "title=" QQ picture 20150305103036.png "alt=" wkiom1t3v8zzqq6raacos1rmbc4635.jpg "/> There is a table as left, and now you need to implement the first two records that query each group (GROUPID). How is it implemented?

The first method is not explained with UNION all. However, the row_number () function can be implemented in a higher version of SQL Server with the following statement:

SELECT * FROM (select Row_number () Up (PARTITION by GroupID ORDER by id) as rowid,* from dbo.test) x WHERE ro Wid<3

Principle: The contents of the X table is based on the results of the GroupID grouping, each ROWID index is incremented in each group, such as Group 1 ROWID 1 2 3, Group 2 ROWID not 4 5 6 also starting from 1 2 3. So, the query results are as follows:

650) this.width=650; "src=" Http://s3.51cto.com/wyfs02/M02/5A/24/wKiom1T3wMnh89cWAAC-p8RFbVY360.jpg "title=" QQ picture 20150305103445.png "alt=" Wkiom1t3wmnh89cwaac-p8rfbvy360.jpg "/>

This article is from the "PZ Technical training base" blog, please be sure to keep this source http://panzi.blog.51cto.com/4738203/1617511

"SQL" uses row_number () to implement group queries