Question

我有两张桌子：

customers表，列customer_id，created_at
orders表，包含order_id列，customer_id，paid_at，amount

我必须编写一个SQL查询，根据他们注册的年份（他们的队列年份）分解客户，并确定多年来每个队列的年度总收入（例如，2011年的队列总收入为第1年$ x，第2年$ y等。）

select c.customer_id, c.created_at, SUM(o.amount) as Tot_amt
from customers c inner join orders o on c.customer_id = o.customer_id
group by c.created_at, Tot_amt;

Answer 1

GROUP BY年（日期栏）

select   c.customer_id, YEAR(c.created_at), SUM(o.amount) as Tot_amt
from     customers c inner join orders o on c.customer_id = o.customer_id
group by c.customer_id, YEAR(c.created_at)

Answer 2

由于它是客户的分手，customer_id在最终输出中毫无意义，只有队列年代才会出现问题。

select YEAR(c.created_at), SUM(o.amount) as Tot_amt
from customers c inner join orders o using(customer_id)
group by YEAR(c.created_at);

Answer 3

支持你的陈述

2011年队列的第1年总收入为$ x，第2年为$ y

您的订单表中应该有orderDate

select 
  YEAR(c.created_at) as cohortYear, 
(YEAR(orderDate)-YEAR(c.created_at)+1) as YearNum,
SUM(o.amount) OVER( PARTITION by YEAR(c.created_at),(YEAR(orderDate)-YEAR(c.created_at)) ORDER BY YEAR(c.created_at),(YEAR(orderDate)-YEAR(c.created_at))) as Tot_amt
from customers c inner join orders o on c.customer_id = o.customer_id

如果您需要使用分组

的解决方案

select 
  YEAR(c.created_at) as cohortYear, 
(YEAR(orderDate)-YEAR(c.created_at)+1) as YearNum,
SUM(o.amount) as Tot_amt
from customers c inner join orders o on c.customer_id = o.customer_id
group by YEAR(c.created_at),(YEAR(orderDate)-YEAR(c.created_at)+1)
order by YEAR(c.created_at),(YEAR(orderDate)-YEAR(c.created_at))

需要按开始年份和之后每年分组的事实行的总和

3 个答案: