SQL Server 排序函数 ROW_NUMBER和RANK 用法总结
下面的例子和SQL语句均在SQL Server 2008环境下运行通过，使用SQL Server自带的AdventureWorks数据库。
转载请注明此文原创自
CSDN TJVictor的专栏：http://blog.csdn.net/tjvictor/archive/2009/07/08/4331039.aspx
1.ROW_NUMBER()基本用法：
SELECT
SalesOrderID,
CustomerID,
ROW_NUMBER() OVER (ORDER BY SalesOrderID) AS RowNumber
FROM Sales.SalesOrderHeader
结果集：
SalesOrderID CustomerID RowNumber
--------------- ------------- ---------------
43659 676 1
43660 117 2
43661 442 3
43662 227 4
43663 510 5
43664 397 6
43665 146 7
43666 511 8
43667 646 9
:
2.RANK()基本用法:
SELECT
SalesOrderID,
CustomerID,
RANK() OVER (ORDER BY CustomerID) AS Rank
FROM Sales.SalesOrderHeader
结果集：
SalesOrderID CustomerID Rank
--------------- ------------- ----------------
43860 1 1
44501 1 1
45283 1 1
46042 1 1
46976 2 5
47997 2 5
49054 2 5
50216 2 5
51728 2 5
57044 2 5
63198 2 5
69488 2 5
44124 3 13
:
3.利用CTE来过滤ROW_NUMBER()的用法：
WITH NumberedRows AS
(
SELECT
SalesOrderID,
CustomerID,
ROW_NUMBER() OVER (ORDER BY SalesOrderID) AS RowNumber
FROM Sales.SalesOrderHeader
)
SELECT * FROM NumberedRows
WHERE RowNumber BETWEEN 100 AND 200
结果集：
SalesOrderID CustomerID RowNumber
--------------- ------------- --------------
43759 13257 100
43760 16352 101
43761 16493 102
:
43857 533 199
43858 36 200
4.带Group by的ROW_NUMBER()用法：
WITH CustomerSum
AS
(
SELECT CustomerID, SUM(TotalDue) AS TotalAmt
FROM Sales.SalesOrderHeader
GROUP BY CustomerID
)
SELECT
*,
ROW_NUMBER() OVER (ORDER BY TotalAmt DESC) AS RowNumber
FROM CustomerSum
结果集：
CustomerID TotalAmt RowNumber
------------- --------------- ---------------
678 1179857.4657 1
697 1179475.8399 2
170 1134747.4413 3
328 1084439.0265 4
514 1074154.3035 5
155 1045197.0498 6
72 1005539.7181 7
:
5.ROW_NUMBER()或是RANK()聚合用法：
WITH CustomerSum AS
(
SELECT CustomerID, SUM(TotalDue) AS TotalAmt
FROM Sales.SalesOrderHeader
GROUP BY CustomerID
)
SELECT *,
RANK() OVER (ORDER BY TotalAmt DESC) AS Rank
--或者是ROW_NUMBER() OVER (ORDER BY TotalAmt DESC) AS Row_Number
FROM CustomerSum
RANK()的结果集：
CustomerID TotalAmt Rank
----------- --------------------- --------------------
678 1179857.4657 1
697 1179475.8399 2
170 1134747.4413 3
328 1084439.0265 4
514 1074154.3035 5
:
6.DENSE_RANK()基本用法：
SELECT
SalesOrderID,
CustomerID,
DENSE_RANK() OVER (ORDER BY CustomerID) AS DenseRank
FROM Sales.SalesOrderHeader
WHERE CustomerID > 100
结果集：
SalesOrderID CustomerID DenseRank
------------ ----------- --------------------
46950 101 1
47979 101 1
49048 101 1
50200 101 1
51700 101 1
57022 101 1
63138 101 1
69400 101 1
43855 102 2
44498 102 2
45280 102 2
46038 102 2
46951 102 2
47978 102 2
49103 102 2
50199 102 2
51733 103 3
57058 103 3
:
7.RANK()与DENSE_RANK()的比较：
WITH CustomerSum AS
(
SELECT
CustomerID,
ROUND(CONVERT(int, SUM(TotalDue)) / 100, 8) * 100 AS TotalAmt
FROM Sales.SalesOrderHeader
GROUP BY CustomerID
)
SELECT *,
RANK() OVER (ORDER BY TotalAmt DESC) AS Rank,
DENSE_RANK() OVER (ORDER BY TotalAmt DESC) AS DenseRank
FROM CustomerSum
结果集：
CustomerID TotalAmt Rank DenseRank
----------- ----------- ------- --------------------
697 1272500 1 1
678 1179800 2 2
170 1134700 3 3
328 1084400 4 4
:
87 213300 170 170
667 210600 171 171
196 207700 172 172
451 206100 173 173
672 206100 173 173
27 205200 175 174
687 205200 175 174
163 204000 177 175
102 203900 178 176
:
8.NTILE()基本用法：
SELECT
SalesOrderID,
CustomerID,
NTILE(10000) OVER (ORDER BY CustomerID) AS NTile
FROM Sales.SalesOrderHeader
结果集：
SalesOrderID CustomerID NTile
--------------- ------------- ---------------
43860 1 1
44501 1 1
45283 1 1
46042 1 1
46976 2 2
47997 2 2
49054 2 2
50216 2 2
51728 2 3
57044 2 3
63198 2 3
69488 2 3
44124 3 4
:
45024 29475 9998
45199 29476 9998
60449 29477 9998
60955 29478 9999
49617 29479 9999
62341 29480 9999
45427 29481 10000
49746 29482 10000
49665 29483 10000
9.所有排序方法对比：
SELECT
SalesOrderID AS OrderID,
CustomerID,
ROW_NUMBER() OVER (ORDER BY CustomerID) AS RowNumber,
RANK() OVER (ORDER BY CustomerID) AS Rank,
DENSE_RANK() OVER (ORDER BY CustomerID) AS DenseRank,
NTILE(10000) OVER (ORDER BY CustomerID) AS NTile
FROM Sales.SalesOrderHeader
结果集：
OrderID CustomerID RowNumber Rank DenseRank NTile
-------- ------------- --------- ------- --------- --------
43860 1 1 1 1 1
44501 1 2 1 1 1
45283 1 3 1 1 1
46042 1 4 1 1 1
46976 2 5 5 2 2
47997 2 6 5 2 2
49054 2 7 5 2 2
50216 2 8 5 2 2
51728 2 9 5 2 3
57044 2 10 5 2 3
63198 2 11 5 2 3
69488 2 12 5 2 3
44124 3 13 13 3 4
44791 3 14 13 3 4
:
10.PARTITION BY基本使用方法：
SELECT
SalesOrderID,
SalesPersonID,
OrderDate,
ROW_NUMBER() OVER (PARTITION BY SalesPersonID ORDER BY OrderDate) AS OrderRank
FROM Sales.SalesOrderHeader
WHERE SalesPersonID IS NOT NULL
结果集：
SalesOrderID SalesPersonID OrderDate OrderRank
--------------- ---------------- ------------ --------------
:
43659 279 2001-07-01 00:00:00.000 1
43660 279 2001-07-01 00:00:00.000 2
43681 279 2001-07-01 00:00:00.000 3
43684 279 2001-07-01 00:00:00.000 4
43685 279 2001-07-01 00:00:00.000 5
43694 279 2001-07-01 00:00:00.000 6
43695 279 2001-07-01 00:00:00.000 7
43696 279 2001-07-01 00:00:00.000 8
43845 279 2001-08-01 00:00:00.000 9
43861 279 2001-08-01 00:00:00.000 10
:
48079 287 2002-11-01 00:00:00.000 1
48064 287 2002-11-01 00:00:00.000 2
48057 287 2002-11-01 00:00:00.000 3
47998 287 2002-11-01 00:00:00.000 4
48001 287 2002-11-01 00:00:00.000 5
48014 287 2002-11-01 00:00:00.000 6
47982 287 2002-11-01 00:00:00.000 7
47992 287 2002-11-01 00:00:00.000 8
48390 287 2002-12-01 00:00:00.000 9
48308 287 2002-12-01 00:00:00.000 10
:
11.PARTITION BY聚合使用方法：
WITH CTETerritory AS
(
SELECT
cr.Name AS CountryName,
CustomerID,
SUM(TotalDue) AS TotalAmt
FROM
Sales.SalesOrderHeader AS soh
INNER JOIN Sales.SalesTerritory AS ter ON soh.TerritoryID = ter.TerritoryID
INNER JOIN Person.CountryRegion AS cr ON cr.CountryRegionCode = ter.
CountryRegionCode
GROUP BY
cr.Name, CustomerID
)
SELECT
*,
RANK() OVER(PARTITION BY CountryName ORDER BY TotalAmt, CustomerID DESC) AS Rank
FROM CTETerritory
结果集：
CountryName CustomerID TotalAmt Rank
-------------- ------------- ----------- --------------
Australia 29083 4.409 1
Australia 29061 4.409 2
Australia 29290 5.514 3
Australia 29287 5.514 4
Australia 28924 5.514 5
:
Canada 29267 5.514 1
Canada 29230 5.514 2
Canada 28248 5.514 3
Canada 27628 5.514 4
Canada 27414 5.514 5
:
France 24538 4.409 1
France 24535 4.409 2
France 23623 4.409 3
France 23611 4.409 4
France 20961 4.409 5
:
12.PARTITION BY求平均数使用方法：
WITH CTETerritory AS
(
SELECT
cr.Name AS CountryName,
CustomerID,
SUM(TotalDue) AS TotalAmt
FROM
Sales.SalesOrderHeader AS soh
INNER JOIN Sales.SalesTerritory AS ter ON soh.TerritoryID = ter.TerritoryID
INNER JOIN Person.CountryRegion AS cr ON cr.CountryRegionCode = ter.
CountryRegionCode
GROUP BY
cr.Name, CustomerID
)
SELECT
*,
RANK() OVER (PARTITION BY CountryName ORDER BY TotalAmt, CustomerID DESC) AS Rank,
AVG(TotalAmt) OVER(PARTITION BY CountryName) AS Average
FROM CTETerritory
结果集：
CountryName CustomerID TotalAmt Rank Average
-------------- ------------- ----------- ------- ------------------
Australia 29083 4.409 1 3364.8318
Australia 29061 4.409 2 3364.8318
Australia 29290 5.514 3 3364.8318
:
Canada 29267 5.514 1 12824.756
Canada 29230 5.514 2 12824.756
Canada 28248 5.514 3 12824.756
:
转载请注明此文原创自CSDN TJVictor的专栏：http://blog.csdn.net/tjvictor/archive/2009/07/08/4331039.aspx
SQL Server 排序函数 ROW_NUMBER和RANK 用法总结的更多相关文章
- SQL Server排序函数row_number和rank的区别
SQL Server排序函数row_number和rank的区别 直接看测试结果 declare @table table(name varchar(100),amount int, memo var ...
- SQL Server 排名函数( ROW_NUMBER、RANK、DENSE_RANK、NTILE )
排名函数是Sql Server2005新增的功能,下面简单介绍一下他们各自的用法和区别.我们新建一张Order表并添加一些初始数据方便我们查看效果. CREATE TABLE [dbo].[Order ...
- sql server 2000 对应 sql server 2005的row_number()、rank()、DENSE_RANK( )、ntile( )等用法
转自CSDN:http://blog.csdn.net/htl258/article/details/4006717 SQL server 2005新增的几个函数,分别是row_number( ).r ...
- mssql sqlserver 分组排序函数row_number、rank、dense_rank用法简介及说明
在实际的项目开发中,我们经常使用分组函数,对组内数据进行群组后,然后进行组内排序:如:1:取出一个客户一段时间内,最大订单数的行记录2: 取出一个客户一段时间内,最后一次销售记录的行记录——————— ...
- SQL SERVER排序函数
排名函数是SQL Server2005新加的功能.在SQL Server2005中有如下四个排名函数: 1.row_number 2.rank 3.dense_rank 4.ntile 下面分别介绍一 ...
- MySQL实现SQL Server排名函数
最近在MySQL中遇到分组排序查询时,突然发现MySQL中没有row_number() over(partition by colname)这样的分组排序.并且由于MySQL中没有类似于SQL Ser ...
- Sql Server REPLACE函数的使用;SQL中 patindex函数的用法
Sql Server REPLACE函数的使用 REPLACE用第三个表达式替换第一个字符串表达式中出现的所有第二个给定字符串表达式. 语法REPLACE ( ''string_replace1'' ...
- 数据库开发基础-SQl Server 聚合函数、数学函数、字符串函数、时间日期函数
SQL 拥有很多可用于计数和计算的内建函数. 函数的语法 内建 SQL 函数的语法是: SELECT function(列) FROM 表 函数的类型 在 SQL 中,基本的函数类型和种类有若干种.函 ...
- SQL SERVER 2008- 字符串函数
/* 1,ASCII返回字符表达式中最左侧字符的ASCII代码值 仅返回首字母的ASCII码值 parameter char或varchar returns integer */ SELECT ASC ...
随机推荐
- Python PIL库之Image注解(API)
http://blog.csdn.net/xiezhiyong3621/article/details/8499543 class Image Methods defined here: __geta ...
- 中文输入法在vs2010中失效解决方案
这样你就可以用切换输入法的方式,输入中文咯. 后来用了2次发现还是有问题,后来我就直接把输入法的切换改成ctrl+1,后来使用就一直没有问题.总之,解决方案视具体情况解决.
- UVALive 5903 Piece it together
一开始用的STL一直超时不能过,后来发现AC的代码基本都用的普通邻接表,然后改了一下13s,T=T,效率太低了.然后把某大神,详情戳链接http://acm.hust.edu.cn/vjudge/pr ...
- 【Quick 3.3】资源脚本加密及热更新(三)热更新模块
[Quick 3.3]资源脚本加密及热更新(三)热更新模块 注:本文基于Quick-cocos2dx-3.3版本编写 一.介绍 lua相对于c++开发的优点之一是代码可以在运行的时候才加载,基于此我们 ...
- C#中的OLEDB连接2
在通过ADO对Excel对象进行连接时(此时Excel则认为是一个数据源),需要配置对Excel数据源对应的连接串,这个连接串中包括了Provider信息(其实类似对数据库进行连接操作时,都需要指定连 ...
- PHP libxml RSHUTDOWN安全限制绕过漏洞(CVE-2012-1171)
漏洞版本: PHP PHP 5.5.x 漏洞描述: BUGTRAQ ID: 65673 CVE(CAN) ID: CVE-2012-1171 PHP是一种HTML内嵌式的语言. PHP 5.x版本内的 ...
- [HUST 1017] Exact cover
Exact cover Time Limit: 15s Memory Limit: 128MB Special Judge Submissions: 6012 Solved: 3185 DESCRIP ...
- 【转】Android版本和API Level对应关系
原文网址:http://blog.csdn.net/huiguixian/article/details/39928027 从Android developer copy过来,留作笔记的. Platf ...
- 我的第一个项目：用kinect录视频库
kinect深度视频去噪 kinectmod32.dll http://pan.baidu.com/s/1DsGqX 下载后改名kinect.dll 替换掉Redist\OpenNI2\Drivers ...
- 基于React服务器端渲染的博客系统
系统目录及源码由此进入 目录 1. 开发前准备 1.1 技术选型1.2 整体设计1.3 构建开发 2. 技术点 2.1 react2.2 redux, react-router2.3 server-r ...