行转列,列转行是我们在开发过程中经常碰到的问题。行转列一般通过CASE WHEN 语句来实现,也可以通过 SQL SERVER 2005 新增的运算符PIVOT来实现。用传统的方法,比较好理解。层次清晰,而且比较习惯。 但是PIVOT 、UNPIVOT提供的语法比一系列复杂的SELECT...CASE 语句中所指定的语法更简单、更具可读性。下面我们通过几个简单的例子来介绍一下列转行、行转列问题。
我们首先先通过一个老生常谈的例子,学生成绩表(下面简化了些)来形象了解下行转列
CREATE TABLE [StudentScores]
(
[UserName] NVARCHAR(20), --学生姓名
[Subject] NVARCHAR(30), --科目
[Score] FLOAT, --成绩
)
INSERT INTO [StudentScores] SELECT 'Nick', '语文', 80
INSERT INTO [StudentScores] SELECT 'Nick', '数学', 90
INSERT INTO [StudentScores] SELECT 'Nick', '英语', 70
INSERT INTO [StudentScores] SELECT 'Nick', '生物', 85
INSERT INTO [StudentScores] SELECT 'Kent', '语文', 80
INSERT INTO [StudentScores] SELECT 'Kent', '数学', 90
INSERT INTO [StudentScores] SELECT 'Kent', '英语', 70
INSERT INTO [StudentScores] SELECT 'Kent', '生物', 85
如果我想知道每位学生的每科成绩,而且每个学生的全部成绩排成一行,这样方便我查看、统计,导出数据
SELECT
UserName,
MAX(CASE Subject WHEN '语文' THEN Score ELSE 0 END) AS '语文',
MAX(CASE Subject WHEN '数学' THEN Score ELSE 0 END) AS '数学',
MAX(CASE Subject WHEN '英语' THEN Score ELSE 0 END) AS '英语',
MAX(CASE Subject WHEN '生物' THEN Score ELSE 0 END) AS '生物'
FROM dbo.[StudentScores]
GROUP BY UserName
查询结果如图所示,这样我们就能很清楚的了解每位学生所有的成绩了
接下来我们来看看第二个小列子。有一个游戏玩家充值表(仅仅为了说明,举的一个小例子),
CREATE TABLE [Inpours]
(
[ID] INT IDENTITY(1,1),
[UserName] NVARCHAR(20), --游戏玩家
[CreateTime] DATETIME, --充值时间
[PayType] NVARCHAR(20), --充值类型
[Money] DECIMAL, --充值金额
[IsSuccess] BIT, --是否成功 1表示成功, 0表示失败
CONSTRAINT [PK_Inpours_ID] PRIMARY KEY(ID)
)
INSERT INTO Inpours SELECT '张三', '2010-05-01', '支付宝', 50, 1
INSERT INTO Inpours SELECT '张三', '2010-06-14', '支付宝', 50, 1
INSERT INTO Inpours SELECT '张三', '2010-06-14', '手机短信', 100, 1
INSERT INTO Inpours SELECT '李四', '2010-06-14', '手机短信', 100, 1
INSERT INTO Inpours SELECT '李四', '2010-07-14', '支付宝', 100, 1
INSERT INTO Inpours SELECT '王五', '2010-07-14', '工商银行卡', 100, 1
INSERT INTO Inpours SELECT '赵六', '2010-07-14', '建设银行卡', 100, 1
下面来了一个统计数据的需求,要求按日期、支付方式来统计充值金额信息。这也是一个典型的行转列的例子。我们可以通过下面的脚本来达到目的
代码
SELECT CONVERT(VARCHAR(10), CreateTime, 120) AS CreateTime,
CASE PayType WHEN '支付宝' THEN SUM(Money) ELSE 0 END AS '支付宝',
CASE PayType WHEN '手机短信' THEN SUM(Money) ELSE 0 END AS '手机短信',
CASE PayType WHEN '工商银行卡' THEN SUM(Money) ELSE 0 END AS '工商银行卡',
CASE PayType WHEN '建设银行卡' THEN SUM(Money) ELSE 0 END AS '建设银行卡'
FROM Inpours
GROUP BY CreateTime, PayType
CASE PayType WHEN '支付宝' THEN SUM(Money) ELSE 0 END AS '支付宝',
CASE PayType WHEN '手机短信' THEN SUM(Money) ELSE 0 END AS '手机短信',
CASE PayType WHEN '工商银行卡' THEN SUM(Money) ELSE 0 END AS '工商银行卡',
CASE PayType WHEN '建设银行卡' THEN SUM(Money) ELSE 0 END AS '建设银行卡'
FROM Inpours
GROUP BY CreateTime, PayType
如图所示,我们这样只是得到了这样的输出结果,还需进一步处理,才能得到想要的结果
在做报表的时候经常会遇到统计 1.2.3.4月的从一月到当前月的总金额等这类的需求,常规的做法就是
select sum(t1.当月金额) as 到目前为止总金额,date as 日期 from TableName t1 inner join TableName as t2 on t1.id=t2.id and t1.time-'12 Month'>=t2.time
这种写法 性能会比较差,借助分组来做统计 性能上就会提升很多倍
=============================================================================================================
下面我们来看看列转行,主要是通过UNION ALL ,MAX来实现。假如有下面这么一个表
代码
CREATE TABLE ProgrectDetail
(
ProgrectName NVARCHAR(20), --工程名称
OverseaSupply INT, --海外供应商供给数量
NativeSupply INT, --国内供应商供给数量
SouthSupply INT, --南方供应商供给数量
NorthSupply INT --北方供应商供给数量
)
INSERT INTO ProgrectDetail
SELECT 'A', 100, 200, 50, 50
UNION ALL
SELECT 'B', 200, 300, 150, 150
UNION ALL
SELECT 'C', 159, 400, 20, 320
UNION ALL
SELECT 'D', 250, 30, 15, 15
(
ProgrectName NVARCHAR(20), --工程名称
OverseaSupply INT, --海外供应商供给数量
NativeSupply INT, --国内供应商供给数量
SouthSupply INT, --南方供应商供给数量
NorthSupply INT --北方供应商供给数量
)
INSERT INTO ProgrectDetail
SELECT 'A', 100, 200, 50, 50
UNION ALL
SELECT 'B', 200, 300, 150, 150
UNION ALL
SELECT 'C', 159, 400, 20, 320
UNION ALL
SELECT 'D', 250, 30, 15, 15
我们可以通过下面的脚本来实现,查询结果如下图所示
代码
SELECT ProgrectName, 'OverseaSupply' AS Supplier,
MAX(OverseaSupply) AS 'SupplyNum'
FROM ProgrectDetail
GROUP BY ProgrectName
UNION ALL
SELECT ProgrectName, 'NativeSupply' AS Supplier,
MAX(NativeSupply) AS 'SupplyNum'
FROM ProgrectDetail
GROUP BY ProgrectName
UNION ALL
SELECT ProgrectName, 'SouthSupply' AS Supplier,
MAX(SouthSupply) AS 'SupplyNum'
FROM ProgrectDetail
GROUP BY ProgrectName
UNION ALL
SELECT ProgrectName, 'NorthSupply' AS Supplier,
MAX(NorthSupply) AS 'SupplyNum'
FROM ProgrectDetail
GROUP BY ProgrectName
MAX(OverseaSupply) AS 'SupplyNum'
FROM ProgrectDetail
GROUP BY ProgrectName
UNION ALL
SELECT ProgrectName, 'NativeSupply' AS Supplier,
MAX(NativeSupply) AS 'SupplyNum'
FROM ProgrectDetail
GROUP BY ProgrectName
UNION ALL
SELECT ProgrectName, 'SouthSupply' AS Supplier,
MAX(SouthSupply) AS 'SupplyNum'
FROM ProgrectDetail
GROUP BY ProgrectName
UNION ALL
SELECT ProgrectName, 'NorthSupply' AS Supplier,
MAX(NorthSupply) AS 'SupplyNum'
FROM ProgrectDetail
GROUP BY ProgrectName
用UNPIVOT 实现如下:
代码
SELECT ProgrectName,Supplier,SupplyNum
FROM
(
SELECT ProgrectName, OverseaSupply, NativeSupply,
SouthSupply, NorthSupply
FROM ProgrectDetail
)T
UNPIVOT
(
SupplyNum FOR Supplier IN
(OverseaSupply, NativeSupply, SouthSupply, NorthSupply )
) P
FROM
(
SELECT ProgrectName, OverseaSupply, NativeSupply,
SouthSupply, NorthSupply
FROM ProgrectDetail
)T
UNPIVOT
(
SupplyNum FOR Supplier IN
(OverseaSupply, NativeSupply, SouthSupply, NorthSupply )
) P