今天被机试了

今天被机试了，题目是分页存储过程，记录一下：

创建数据库：

USE [master]
GO

/****** Object:  Database [UserDB]    Script Date: 04/25/2012 15:22:56 ******/
CREATE DATABASE [UserDB] ON  PRIMARY 
( NAME = N'UserDB', FILENAME = N'C:\Program Files\Microsoft SQL Server\MSSQL10_50.MSSQLSERVER\MSSQL\DATA\UserDB.mdf' , SIZE = 4352KB , MAXSIZE = UNLIMITED, FILEGROWTH = 1024KB )
 LOG ON 
( NAME = N'UserDB_log', FILENAME = N'C:\Program Files\Microsoft SQL Server\MSSQL10_50.MSSQLSERVER\MSSQL\DATA\UserDB_log.LDF' , SIZE = 832KB , MAXSIZE = 2048GB , FILEGROWTH = 10%)
GO

创建数据表的语句：

USE [UserDB]
GO

/****** Object:  Table [dbo].[UserBaseinfo]    Script Date: 04/25/2012 15:21:02 ******/
SET ANSI_NULLS ON
GO

SET QUOTED_IDENTIFIER ON
GO

SET ANSI_PADDING ON
GO

CREATE TABLE [dbo].[UserBaseinfo](
    [UserId] [int] IDENTITY(1,1) NOT NULL,
    [Name] [varchar](128) NULL,
    [Description] [varchar](1024) NULL,
    [Title] [varchar](64) NULL,
    [Address] [varchar](128) NULL
) ON [PRIMARY]

GO

SET ANSI_PADDING OFF
GO

以下有几个存储过程：

select * from UserDB.dbo.UserBaseinfo


alter procedure proc_paged_with_notin  ----利用select top and select not in 
( 
    @pageIndex int,  --页索引 
    @pageSize int    --每页记录数 
) 
as 
begin 
    set nocount on; 
    declare @timediff datetime --耗时 
    declare @sql nvarchar(500) 
    select @timediff=Getdate() 
    set

 @sql='select top '+str(@pageSize)+' * from 
    UserDB.dbo.UserBaseinfo 
    where(UserId not in(select top '+str(@pageSize*@pageIndex)+' UserId
     from UserDB.dbo.UserBaseinfo order by UserId ASC)) order by UserId' 
----上面这个@sql的意思是:假定pagesize=10,pageindex=2,那么先找前面20条数据，然后再提取不在这20条里面的数据，也就是从21条开始找10条取出来，pageindex记住是pageindex-1的，从0数起，其他的都比较容易理解

    execute(@sql)  --因select top后不支技直接接参数，所以写成了字符串@sql 
    select datediff(ms,@timediff,GetDate()) as 耗时 
    set nocount off; 
end

耗时 :0-80ms

--利用select top and select max(列) 
alter procedure proc_paged_with_selectMax  
( 
    @pageIndex int,  --页索引 
    @pageSize int    --页记录数 
) 
as 
begin 
set nocount on; 
    declare @timediff datetime 
    declare @sql nvarchar(500) 
    select @timediff=Getdate() 
    set @sql='select top '+str(@pageSize)+' * From UserDB.dbo.UserBaseinfo
     where
     (  UserId>(select max(UserId) 
     From (select top '+str(@pageSize*@pageIndex)+' UserId 
     From UserDB.dbo.UserBaseinfo order by UserId) 
     as TempTable
           )
     ) order by UserId' 


    execute(@sql) 
    select datediff(ms,@timediff,GetDate()) as 耗时 
set nocount off; 
end

A=select top 20 UserId 
     From UserDB.dbo.UserBaseinfo order by UserId 意思是先找到前面20条

UserId>(select max(UserId) 
     From A 就是说在这20条里面打一个最大号的ID，也就是20
然后再去找比20大的，找10条

耗时：0-40ms