Question

如何使用以下数据在Dynamic SQL中创建简单的数据透视表：

 ID Month   AssignmentMonth Designation
 1  5            May         Joe Blow
 2  5            May         Available
 3  5            May         Available
 4  5            May         Available
 5  6            June        Carry over
 6  6            June        Christopher Freeberg
 7  6            June        Ringo Starr
 8  6            June        Robert L Testcustomer
 9  7            July        Carry over
10  7            July        Carry over
11  7            July        Carry over
12  7            July        Carry over
13  8            August      Available
14  8            August      Carry over
15  8            August      Carry over
16  8            August      Carry over

所以当它完成时它看起来像下面的例子：

May               June                 July                 August

Joe Blow         Carry Over            Carry over          Available
Available        Christopher Freeburg  Carry over          Carry over
Available        Ringo Starr           Carry over          Carry over
Available        Robert L Testcustomer Carry over          Carry over

这是我的尝试，但它确实产生了正确的结果。它只返回一行。

   August        July       June                    May
   Carry over   Carry over  Robert L Testcustomer   Joe Blow

    DECLARE @cols AS NVARCHAR(MAX),
    @query  AS NVARCHAR(MAX)

    select @cols = STUFF((SELECT distinct ',' +   QUOTENAME(AssignmentMonth) 
                from @CardAssigns
        FOR XML PATH(''), TYPE
        ).value('.', 'NVARCHAR(MAX)') 
    ,1,1,'') 
SET @query = 'WITH PivotData AS
(
SELECT   Designation, AssignmentMonth      
FROM    #Temp1
)
SELECT  '+ @cols +'
FROM PivotData
PIVOT
(
MAX(Designation)
FOR AssignmentMonth
IN ( '+ @cols +' )
) as PIVOTResult'

EXECUTE(@query)

Answer 1

您的查询存在的问题是，透视查询没有要分组的列。 pivot表运算符会自动GROUP BY来自锚点查询的部分MAX(Designation) FOR AssignmentMonth IN ...中未列出的列，因此在您的查询中，您只有两列来自锚定查询Designation和AssignmentMonth，因此您的查询没有任何额外的列可以分组，以便它为您提供只有一行的MAX(Designation)。但是，您可以通过再添加一列来进行解决方法，因为您无法按ID或Month进行分组，您可以添加行号列，它将会给你想要的结果，如下：

SET @query = 'WITH PivotData AS
(
  SELECT   
    Designation, 
    AssignmentMonth, 
    ROW_NUMBER() OVER(PARTITION BY   AssignmentMonth ORDER BY ID) AS RN
  FROM   CardAssigns
)
SELECT  '+ @cols +'
FROM PivotData
PIVOT
(
  MAX(Designation)
  FOR AssignmentMonth
  IN ( '+ @cols +' )
) as PIVOTResult';

EXECUTE(@query);

SQL Fiddle Demo

这会给你：

|     August |       July |                  June |       May |
|------------|------------|-----------------------|-----------|
|  Available | Carry over |            Carry over |  Joe Blow |
| Carry over | Carry over |  Christopher Freeberg | Available |
| Carry over | Carry over |           Ringo Starr | Available |
| Carry over | Carry over | Robert L Testcustomer | Available |

Answer 2

下面的示例使用动态查询和ROW_NUMBER来排列您所描述的值。

CREATE TABLE #Data (
    ID INT,
    Month INT,
     AssignmentMonth VARCHAR(50),
     Designation VARCHAR(50)

)
INSERT #Data VALUES
(1,5,'May','Joe Blow'),
(2,5,'May','Available'),
(3,5,'May','Available'),
(4,5,'May','Available'),
(5,6,'June','Carry over'),
(6,6,'June','Christopher Freeberg'),
(7,6,'June','Ringo Starr'),
(8,6,'June','Robert L Testcustomer'),
(9,7,'July','Carry over'),
(10,7,'July','Carry over'),
(11,7,'July','Carry over'),
(12,7,'July','Carry over'),
(13,8,'August','Available'),
(14,8,'August','Carry over'),
(15,8,'August','Carry over'),
(16,8,'August','Carry over')

DECLARE @ColumnList VARCHAR(MAX) =
    SUBSTRING((
        SELECT
            ',[' + AssignmentMonth + ']' AS [text()]
        FROM #Data
        GROUP BY AssignmentMonth
        ORDER BY MIN(ID)
        FOR XML PATH('')
    ), 2, 4000)
DECLARE @ColumnAggs VARCHAR(MAX) =
    SUBSTRING((
        SELECT
            ',MAX([' + AssignmentMonth + ']) AS [' + AssignmentMonth + ']' AS [text()]
        FROM #Data
        GROUP BY AssignmentMonth
        ORDER BY MIN(ID)
        FOR XML PATH('')
    ), 2, 4000)

DECLARE @Sql VARCHAR(MAX) =
    'SELECT ' + @ColumnAggs + ' FROM ('
    + 'SELECT ROW_NUMBER() OVER (PARTITION BY Month ORDER BY ID) AS RowId, ' + @ColumnList
    + 'FROM #Data D PIVOT (MAX(Designation) FOR AssignmentMonth IN (' + @ColumnList + ')) P'
    + ') T GROUP BY RowId'
    PRINT @Sql
EXEC (@Sql)

输出

May                            June                           July                           August
------------------------------ ------------------------------ ------------------------------ ------------------------------
Joe Blow                       Carry over                     Carry over                     Available
Available                      Christopher Freeberg           Carry over                     Carry over
Available                      Ringo Starr                    Carry over                     Carry over
Available                      Robert L Testcustomer          Carry over                     Carry over

简单的数据透视表与月和文本数据动态SQL

2 个答案: