--行转列问题
--建立測試環境
Create Table TEST
(DATES Varchar(6),
EMPNO Varchar(5),
STYPE Varchar(1),
AMOUNT Int)
--插入數據
Insert TEST Select '200605', '02436', 'A', 5
Union All Select '200605', '02436', 'B', 3
Union All Select '200605', '02436', 'C', 3
Union All Select '200605', '02436', 'D', 2
Union All Select '200605', '02436', 'E', 9
Union All Select '200605', '02436', 'F', 7
Union All Select '200605', '02436', 'G', 6
Union All Select '200605', '02438', 'A', 7
Union All Select '200605', '02438', 'B', 8
Union All Select '200605', '02438', 'C', 0
Union All Select '200605', '02438', 'D', 3
Union All Select '200605', '02438', 'E', 4
Union All Select '200605', '02438', 'F', 5
Union All Select '200605', '02438', 'G', 1
GO
--測試
--如果STYPE固定，可以這麼寫
Select
DATES,
EMPNO,
SUM(Case STYPE When 'A' Then AMOUNT Else 0 End) As A,
SUM(Case STYPE When 'B' Then AMOUNT Else 0 End) As B,
SUM(Case STYPE When 'C' Then AMOUNT Else 0 End) As C,
SUM(Case STYPE When 'D' Then AMOUNT Else 0 End) As D,
SUM(Case STYPE When 'E' Then AMOUNT Else 0 End) As E,
SUM(Case STYPE When 'F' Then AMOUNT Else 0 End) As F,
SUM(Case STYPE When 'G' Then AMOUNT Else 0 End) As G
From TEST
Group By DATES,EMPNO
Order By DATES,EMPNO

--如果STYPE不固定，用動態語句
Declare @S Varchar(1000)
Set @S=''
Select @S=@S+',SUM(Case STYPE When '''+STYPE+''' Then AMOUNT Else 0 End) As '+STYPE From (Select Distinct STYPE From TEST) A Order By STYPE
Set @S='Select DATES,EMPNO'+@S+' From TEST Group By DATES,EMPNO Order By DATES,EMPNO'
EXEC(@S)
GO
--如果被转置的是数字类型的话，应用下列语句
DECLARE @S VARCHAR(1000)
SET @S='SELECT DATES,EMPNO '
SELECT @S=@S+',['+STYPE+']=SUM(CASE WHEN STYPE='''+STYPE+''' THEN AMOUNT ELSE 0 END)'
FROM (Select Distinct STYPE From TEST) A Order By STYPE
SET @S=@S+' FROM TEST GROUP BY DATES,EMPNO'
EXEC(@S)