Question

可能最好我用一个例子来解释我所追求的内容：）

说我有下表

SELECT *
INTO #OrderRanking
FROM (
    SELECT 'Jim', 1000
    UNION ALL
    SELECT 'Jim', 500
    UNION ALL
    SELECT 'Jim', 500
    UNION ALL
    SELECT 'Jim', 3000
    UNION ALL
    SELECT 'Bob', 1000
    UNION ALL
    SELECT 'Bob', 2000
    UNION ALL
    SELECT 'Bob', 500
    UNION ALL
    SELECT 'Bob', 500
    UNION ALL
    SELECT 'Fred', 500
) AS t(Customer, OrderTotal)

我想要的是SELECT声明中的以下结果

Customer, OrderTotal, CustomerId
'Jim', 1000, 1
'Jim', 500, 1
'Jim', 500, 1
'Jim', 3000, 1
'Bob', 1000, 2
'Bob', 2000, 2
'Bob', 500, 2
'Bob', 500, 2
'Fred', 500, 3

我的实际查询有点复杂，其中源列超过2。

Answer 1

我是个傻瓜，

SELECT *, DENSE_RANK() OVER(ORDER BY Customer DESC)
FROM #OrderRanking

我认为这应涵盖所有情况！

Answer 2

select customer, ordertotal, dense_rank() over(order by total desc, customer asc)
from (select *, sum(ordertotal) over(partition by customer) total from  #orderranking) t

SQL根据其他列生成顺序标识

2 个答案: