Question

我有一张表格，上面有客户编号，订单编号和等级字段。每个客户编号可以包含具有不同等级值的多个订单。实施例

cust#  order#  rank

1       12      1
1       13      3
1       14      2
2       15      2
2       16      1
3       17      3
3       18      4
3       19      1
3       20      2

我正在使用此表来填充另一个看起来像这样的表。

cust   order1 order2 order3 order4
1       12      14     13
2       16      15
3       19      20     17    18

那么，如何选择填充order2, order3, order4字段的第二个，第三个最低排名？

编辑：我不想做1,2,3等等，因为有时可能没有等级2所以order2字段将为空，但将填充order3字段。我想以最低排名来做。 order1中的第一个最低排名，order2中的第二个最低排名等。

Answer 1

SELECT R1.cust#, R1.order#, R2.order#, r3.order#, r4.order#
FROM MyTable AS R1
LEFT JOIN MyTable AS R2 ON R1.cust# = R2.cust# AND R2.rank = 2
LEFT JOIN MyTable AS R3 ON R1.cust# = R3.cust# AND R3.rank = 3
LEFT JOIN MyTable AS R4 ON R1.cust# = R4.cust# AND R4.rank = 4
WHERE R1.rank = 1

Answer 2

也许这会有所帮助：

SELECT cust#,
       MAX(CASE WHEN rank = 1 THEN order# END) AS order1,
       MAX(CASE WHEN rank = 2 THEN order# END) AS order2,
       MAX(CASE WHEN rank = 3 THEN order# END) AS order3,
       MAX(CASE WHEN rank = 4 THEN order# END) AS order4     
  FROM(SELECT cust#,
              order#,
              ROW_NUMBER() OVER (PARTITION BY cust# ORDER BY order#) rank
         FROM your_table
      )
 GROUP
    by cust#

Answer 3

使用ROW_NUMBER()和PIVOT的组合，如下所示：

DECLARE @tmp TABLE (Customer INT, OrderNumber INT, Ranking INT)
INSERT INTO @tmp (Customer, OrderNumber, Ranking)
SELECT 1,       12,      1 UNION
SELECT 1,       13,      3 UNION
--SELECT 1,       14,      2 UNION
SELECT 2,       15,      2 UNION
SELECT 2,       16,      1 UNION
SELECT 3,       17,      3 UNION
SELECT 3,       18,      4 UNION
SELECT 3,       19,      1 UNION
SELECT 3,      20,      2


SELECT 
    Customer, 
    MAX(Order1) AS Order1, 
    MAX(Order2) AS Order2, 
    MAX(Order3) AS Order3, 
    MAX(Order4) AS Order4
FROM 
  (
    SELECT 
        *, 
        'Order' + CAST(ROW_NUMBER() OVER (PARTITION BY Customer ORDER BY Ranking) AS VARCHAR(4)) AS rn 
    FROM @tmp 
  ) d
PIVOT 
  (
    MAX(OrderNumber) FOR rn IN ([Order1], [Order2], [Order3], [Order4])
  ) p
GROUP BY Customer

在SQL Server中选择具有第二最低等级的行

3 个答案: