Sql分页

1.

SELECT TOP 页大小 * FROM table WHERE 主键 NOT IN
(
    SELECT TOP (页码-1)*页大小 主键 FROM table WHERE 查询条件 ORDER BY 排序条件
)
ORDER BY 排序条件

 SELECT  TOP 10* FROM HumanResources.Employee
 WHERE BusinessEntityID NOT IN
 (SELECT top 10 BusinessEntityID FROM HumanResources.Employee)

 按着上面的模板写个示例,

发现，子查询中会返回10行数据，而主查询会返回所以得数据。俩个数据再相互比较，返回符合条件的数据。

在大数据量的时候，计算量会很大。

2.

--顺序写法：
 SELECT TOP 页大小 *
 FROM table1
 WHERE id >=
 (
 SELECT ISNULL(MAX(id),0) 
 FROM 
 (
 SELECT TOP 页大小*(页数-1)+1 id FROM table1 ORDER BY id
 ) A
 )
 ORDER BY id
 
 --降序写法：
 SELECT TOP 页大小 *
 FROM table1
 WHERE id <=
 (
 SELECT ISNULL(MIN(id),0) 
 FROM 
 (
 SELECT TOP 页大小*(页数-1)+1 id FROM table1 ORDER BY id Desc
 ) A
 )
 ORDER BY id Desc

 SELECT TOP 10 * FROM HumanResources.Employee
 WHERE BusinessEntityID>=(SELECT isnull(MAX(BusinessEntityID),0) from
  (SELECT top 11 BusinessEntityID from HumanResources.Employee 
  ORDER BY BusinessEntityID)
  A)
  ORDER BY BusinessEntityID

子查询返回本页的第一个Id.

主查询直接根据条件返回记录。操作的数据减少了很多。即使是大数据量也不存在效率问题。

但是如何主键不能按照某种顺序进行排序，这个方法就不怎么好用了。

3.但是可以使用row_number函数为不能排序的列建立一个行号。

 SELECT TOP 页大小 * 
 FROM (SELECT ROW_NUMBER() OVER (ORDER BY id) AS RowNumber,* FROM table1) A
 WHERE RowNumber > 页大小*(页数-1)

 SELECT TOP 10 * from
  (SELECT ROW_NUMBER()OVER (ORDER BY BusinessEntityID) as RowNumber,* from HumanResources.Employee)A
  where RowNumber> 10

 根据第一中方法写的存储过程，测试一下是否好用：

CREATE proc GetNextPage
@rows int,
@pageIndex int
as
begin
select TOP (@rows) * from HumanResources.Employee
where BusinessEntityID
not in
(
SELECT TOP (@rows*@pageIndex) BusinessEntityID from HumanResources.Employee
)
end

exec getnextPage 10, 1