Question

我想使用Rank()查询每个客户的最高折扣级别。但是有些数据丢失了，我不知道为什么...... （我使用了数据库 Northwind ）

这是我试过的脚本..

WITH CTE AS 
(SELECT ProductID, CustomerID, Discount,    
       RANK() over (partition by CustomerID order by Discount desc) as rank
FROM Orders
    JOIN [Order Details] 
        ON Orders.OrderID= [Order Details].OrderID
)

SELECT Distinct ProductID, CustomerID, Discount   

FROM CTE
    WHERE 
            Discount>0.1 and
            ProductID between 60 and 65 and
            rank=1
    order by CustomerID

在此过程之前，我事先尝试过这个脚本

    select ProductID, customerID, Discount
from Orders
    join [Order Details]
        on orders.OrderID = [Order details].OrderID
where 
    Discount>0.1 and
    ProductID between 60 and 65
 order by ProductID, CustomerID
go

我手动编辑以获得每个客户的最高折扣级别。我应该有大约32行（这取决于我是否删除重复的数据）但是当我运行第一个脚本时，我只得到15行.. 我怎么解决这个问题？任何帮助将不胜感激。

Answer 1

您在应用过滤器之前计算RANK()，这意味着过滤器可以删除某些Rank() = 1行。尝试将过滤器移到CTE。

;WITH cte
     AS (SELECT productid,
                customerid,
                discount,
                Rank()
                  OVER (
                    PARTITION BY customerid
                    ORDER BY discount DESC) AS [rank]
         FROM   orders
                JOIN [order details]
                  ON orders.orderid = [order details].orderid
         WHERE  discount > 0.1
                AND productid BETWEEN 60 AND 65)
SELECT DISTINCT productid,
                customerid,
                discount
FROM   cte
WHERE  [rank] = 1
ORDER  BY customerid

我想使用Northwind查询每个客户的最高折扣级别

1 个答案: