Question

我有带有动态TabHeader的表，例如：

RecordNum | TabHeader | tabvalue
1              A         data1
1              B         data2
1              C         data3
2              A         Data4
2              B         data5
2              c         data6
3              A         Data7

我想创建要实现的查询表：

   recordNum |   A      |  B    |  C  |
      1        data1     data2   data3
      2        data4     data5   data6
      3        data7

我尝试使用数据透视，但没有运气，拜托任何人都可以帮助我查询：

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

select @cols = STUFF((SELECT distinct ',' + QUOTENAME(tablecoloumnmap)

                    from temptes
            FOR XML PATH(''), TYPE
            ).value('.', 'NVARCHAR(MAX)') 
        ,1,1,'')

set @query = 'SELECT recordnum ' + @cols + ' from 
             (
                select *
                from temptes
            ) x
            pivot 
            (
                max(tablevalue)
                for tablecoloumnmap in (' + @cols + ')
            ) p '

execute(@query)

Answer 1

您可以改为使用条件聚合：

select RecordNum, 
       max(case when tabheader = 'A' then tabvalue end) as A,
       max(case when tabheader = 'B' then tabvalue end) as B,
       max(case when tabheader = 'C' then tabvalue end) as C
from table t
group by RecordNum;

编辑：：这是Pivot的动态版本：

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

SELECT @cols = STUFF((SELECT DISTINCT ',' + QUOTENAME(TabHeader)
                      FROM temptes
                      FOR XML PATH(''), TYPE
                     ).value('.', 'NVARCHAR(MAX)') , 1, 1, ''
                    )


SET @query = N'select *
               from temptes x pivot 
               ( max(tabvalue)
                 for TabHeader in (' + @cols + ')
               ) p '

EXECUTE (@query)

Answer 2

您好，对您现有的sql脚本进行了一些更正，它将提供您期望的确切结果

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

select @cols = STUFF((SELECT distinct ',' + QUOTENAME(TabHeader)

                    from temptes
            FOR XML PATH(''), TYPE
            ).value('.', 'NVARCHAR(MAX)') 
        ,1,1,'')

set @query = 'SELECT recordnum ,' + @cols + ' from 
             (
                select *
                from temptes
            ) x
            pivot 
            (
                max(tabvalue)
                for TabHeader in (' + @cols + ')
            ) p '
PRINT @query
execute(@query)

结果

Recordnum    A        B       C
----------------------------------
1           data1   data2   data3
2           Data4   data5   data6
3           Data7   NULL    NULL

在SQL Server中使用数据透视创建矩阵表

2 个答案: