Question

我有以下数据：

---------------------------------------------------
oGroup     oDate        Value1      Value2
---------------------------------------------------
A          2014-01-01   10          100
A          2014-01-02   12          108
B          2014-01-01   20          120
B          2014-01-02   22          140
C          2014-01-01   5           160
C          2014-01-02   9           175

我在pivot中有以下脚本：

select 
   *
from 
   (
    select 
       Group
       , oDate
       , Value1
       , Value2
    from #ret
   ) src
   pivot
   (
      sum(value1)
      for group in ([X1], [X2], [X3])
   ) piv
   order by oDate;

然后结果应该是这样的：

------------------------------------------------
oDate         X1     X2     X3
------------------------------------------------
2014-01-01    10     20     5
2014-01-02    12     22     9

我的问题是如何在X1，X2，X3之后将value2添加到另外3列。比方说，Y1 Y2和Y3。

所以结果应该是：

------------------------------------------------------
oDate         X1     X2     X3     Y1     Y2     Y3
------------------------------------------------------
2014-01-01    10     20     5      100    120    160  
2014-01-02    12     22     9      108    140    175
------------------------------------------------------

我怎样才能做到这一点？谢谢。

Answer 1

使用PIVOT：

DECLARE @t TABLE
    (
      oGroup CHAR(1) ,
      oDate DATE ,
      Value1 INT ,
      Value2 INT
    )

INSERT  INTO @t
VALUES  ( 'A', '20140101', 10, 100 ),
        ( 'A', '20140102', 12, 108 ),
        ( 'B', '20140101', 20, 120 ),
        ( 'B', '20140102', 22, 140 ),
        ( 'C', '20140101', 5, 160 ),
        ( 'C', '20140102', 9, 175 );
WITH    cte1
          AS ( SELECT   oGroup ,
                        oDate ,
                        Value1
               FROM     @t
             ),
        cte2
          AS ( SELECT   oGroup ,
                        oDate ,
                        Value2
               FROM     @t
             )
    SELECT  a1.* ,
            a2.[A] ,
            a2.[B] ,
            a2.[C]
    FROM    ( SELECT    *
              FROM      cte1 PIVOT( SUM(Value1) FOR oGroup IN ( [A], [B],
                                                          [C] ) ) AS p
            ) a1
            JOIN ( SELECT   *
                   FROM     cte2 PIVOT( SUM(Value2) FOR oGroup IN ( [A],
                                                          [B], [C] ) ) AS p
                 ) a2 ON a1.oDate = a2.oDate

输出：

oDate       A   B   C   A   B   C
2014-01-01  10  20  5   100 120 160
2014-01-02  12  22  9   108 140 175

Answer 2

select 
   *
into #tb1
from 
   (
    select 
       Group
       , oDate
       , Value1
    from #ret
   ) src
   pivot
   (
      sum(value1)
      for group in ([X1], [X2], [X3])
   ) piv
   order by oDate;

select 
   *
into #tb2
from 
   (
    select 
       Group
       , oDate
       , Value2
    from #ret
   ) src
   pivot
   (
      sum(value2)
      for group in ([Y1], [Y2], [Y3])
   ) piv
   order by oDate;



select * from #tb1 inner join #tb2 on #tb1.oDate = #tb2.oDate

使用两个FOR SQL旋转两列

2 个答案: