SQL Server 2008中的Pivot和UnPivot

SQL Server 2008中SQL应用系列--目录索引

今天给新成员讲解PIVOT 和 UNPIVOT示例，顺便整理了一下其用法。这是自SQL Server 2005起提供的新功能。

官方示例：http://msdn.microsoft.com/zh-cn/library/ms177410%28v=sql.105%29.aspx

首先看PIVOT示例：

基本表数据：

View Code

IF NOT OBJECT_ID('tb_Income') IS NULL
    DROP TABLE [tb_Income]

/****** Object:  Table [dbo].[tb_Income]   Script Date: 2012/4/5 8:19:21 ******/


CREATE TABLE [dbo].[tb_Income](
----[PKID] int primary key identity(101,1),
 [PName] [Nvarchar](20) NOT NULL,
    [CYear] Smallint NOT NULL,
    [CMonth] TinyInt NOT NULL,
    [CMoney] Decimal (10,2) Not Null

    )

GO
INSERT [dbo].[tb_Income] 
SELECT '胡一刀',2011,2,5600
union ALL SELECT '胡一刀',2011,1,5678
union ALL SELECT '胡一刀',2011,3,6798
union ALL SELECT '胡一刀',2011,4,7800
union ALL SELECT '胡一刀',2011,5,8899
union ALL SELECT '胡一刀',2011,8,8877
union ALL SELECT '胡一刀',2011,6,7788
union ALL SELECT '胡一刀',2011,7,6798
union ALL SELECT '胡一刀',2011,10,10000
union ALL SELECT '胡一刀',2011,9,12021
union ALL SELECT '胡一刀',2011,11,8799
union ALL SELECT '胡一刀',2011,12,10002

union ALL SELECT '苗人凤',2011,1,3455
union ALL SELECT '苗人凤',2011,2,4567
union ALL SELECT '苗人凤',2011,3,5676
union ALL SELECT '苗人凤',2011,4,5600
union ALL SELECT '苗人凤',2011,5,6788
union ALL SELECT '苗人凤',2011,6,5679
union ALL SELECT '苗人凤',2011,7,6785
union ALL SELECT '苗人凤',2011,8,7896
union ALL SELECT '苗人凤',2011,9,7890
union ALL SELECT '苗人凤',2011,10,7799
union ALL SELECT '苗人凤',2011,11,9988

union ALL SELECT '郑希来',2011,2,5600
union ALL SELECT '郑希来',2011,3,2345
union ALL SELECT '郑希来',2011,5,12000
union ALL SELECT '郑希来',2011,4,23456
union ALL SELECT '郑希来',2011,6,4567
union ALL SELECT '郑希来',2011,7,6789
union ALL SELECT '郑希来',2011,8,9998
union ALL SELECT '郑希来',2011,9,34567
union ALL SELECT '郑希来',2011,12,5609

GO

测试结果如下：

SELECT * FROM tb_Income
GO

/*

PName    CYear    CMonth    CMoney
胡一刀    2011    2    5600.00
胡一刀    2011    1    5678.00
胡一刀    2011    3    6798.00
胡一刀    2011    4    7800.00
胡一刀    2011    5    8899.00
胡一刀    2011    8    8877.00
胡一刀    2011    6    7788.00
胡一刀    2011    7    6798.00
胡一刀    2011    10    10000.00
胡一刀    2011    9    12021.00
胡一刀    2011    11    8799.00
胡一刀    2011    12    10002.00
苗人凤    2011    1    3455.00
苗人凤    2011    2    4567.00
苗人凤    2011    3    5676.00
苗人凤    2011    4    5600.00
苗人凤    2011    5    6788.00
苗人凤    2011    6    5679.00
苗人凤    2011    7    6785.00
苗人凤    2011    8    7896.00
苗人凤    2011    9    7890.00
苗人凤    2011    10    7799.00
苗人凤    2011    11    9988.00
郑希来    2011    2    5600.00
郑希来    2011    3    2345.00
郑希来    2011    5    12000.00
郑希来    2011    4    23456.00
郑希来    2011    6    4567.00
郑希来    2011    7    6789.00
郑希来    2011    8    9998.00
郑希来    2011    9    34567.00
郑希来    2011    12    5609.00
*/

现在需要统计2011年的个人总工资，使用Group by 即可：

SELECT PName,sum(Cmoney) as YearMoney from tb_Income
GROUP BY PName
ORDER BY sum(Cmoney) desc

/*
PName    YearMoney
郑希来    104931.00
胡一刀    99060.00
苗人凤    72123.00
*/

现在我们来进行行列转换：

SELECT CYear,胡一刀,苗人凤,郑希来 FROM tb_Income 
PIVOT(sum(CMoney) 
FOR PName IN
 (胡一刀,苗人凤,郑希来)) t

 /*
 CYear    胡一刀    苗人凤    郑希来
2011    5678.00    3455.00    NULL
2011    5600.00    4567.00    5600.00
2011    6798.00    5676.00    2345.00
2011    7800.00    5600.00    23456.00
2011    8899.00    6788.00    12000.00
2011    7788.00    5679.00    4567.00
2011    6798.00    6785.00    6789.00
2011    8877.00    7896.00    9998.00
2011    12021.00    7890.00    34567.00
2011    10000.00    7799.00    NULL
2011    8799.00    9988.00    NULL
2011    10002.00    NULL    5609.00
*/

注意行列已经转换，再汇总,关键是去除干扰列，重新构建新数据集X：

SELECT 胡一刀,苗人凤,郑希来 FROM 
(SELECT PName,CMoney FROM tb_Income) X 
PIVOT(sum(CMoney) 
FOR PName IN
 (胡一刀,苗人凤,郑希来)) t

 /*
    胡一刀    苗人凤    郑希来
    99060.00    72123.00    104931.00
*/

UNPIVOT的示例更简单一些：

生成基本数据：

View Code

IF NOT OBJECT_ID('tb_Tel') IS NULL
    DROP TABLE [tb_Tel]

CREATE TABLE [dbo].[tb_Tel](
----[PKID] int primary key identity(101,1),
 [PName] [Nvarchar](20) NOT NULL,
    [Mobile1] [Nvarchar](20) NOT NULL,
    [Mobile2] [Nvarchar](20) NOT NULL,
    [Mobile3] [Nvarchar](20) Not Null
    )
GO
INSERT [dbo].[tb_Tel] 
SELECT '胡一刀','13067894562','13567889667','16767894562'
union ALL SELECT '苗人凤','1507894562','15267889667','15367894562'
union ALL SELECT '郑希来','18067894562','18567889667','18767894562'
GO

结果：

SELECT * FROM tb_Tel

/*
PName    Mobile1    Mobile2    Mobile3
胡一刀    13067894562    13567889667    16767894562
苗人凤    1507894562    15267889667    15367894562
郑希来    18067894562    18567889667    18767894562
*/

行列转换：

SELECT PName,电话类型,电话号码
FROM tb_Tel
UNPIVOT(电话类型 FOR 电话号码 IN (Mobile1,Mobile2,Mobile3) ) p

/*
PName    电话类型    电话号码
胡一刀    13067894562    Mobile1
胡一刀    13567889667    Mobile2
胡一刀    16767894562    Mobile3
苗人凤    1507894562    Mobile1
苗人凤    15267889667    Mobile2
苗人凤    15367894562    Mobile3
郑希来    18067894562    Mobile1
郑希来    18567889667    Mobile2
郑希来    18767894562    Mobile3
*/