Question

我有三张桌子：

**Bookie**:

------------

Id : int (Primary Key)

Bookie : varchar

**Transactions**:

-------------------

Id : int (Primary Key)

Date : date

Bookie : int (Foreign Key)

Amount : decimal

Wagers:

Id : int (Primary Key)

Stake : decimal

Bookie : int (Foreign Key)

Wagers表不是完整的，但对于sql这些列就足够了。

到目前为止，我尝试过这个sql代码：

SELECT Bookie.Bookie, SUM(Amount), Wagers.Stake FROM Transactions
JOIN Bookie on Transactions.Bookie = Bookie.Id
JOIN Wagers on Bookie.Id = Wagers.Bookie
GROUP BY Bookie.Bookie, Wagers.Stake

但我的输出看起来像这样：

Bookie | Amount | Stake
William Hill | 600.00 | 14.70
Bet at Home | 500.00 | 30.00
Bet at Home | 500.00 | 45.00
Bet at Home | 500.00 | 50.00
Bet365 | 10200.00 | 100.00

它应该是这样的：

Bookie | Amount | Stake
William Hill | 600.00 | 14.70
Bet at Home | 500.00 | 125.00
Bet365 | 10200.00 | 100.00

有谁知道如何解决这个问题？

由于

Answer 1

您需要汇总Stake：

SELECT Bookie.Bookie, SUM(Amount) AS Amount, SUM(Wagers.Stake) AS Stake
FROM Transactions
JOIN Bookie 
  ON Transactions.Bookie = Bookie.Id
JOIN Wagers 
  ON Bookie.Id = Wagers.Bookie
GROUP BY Bookie.Bookie;

修改

SELECT Bookie, MAX(Amount) AS Amount, SUM(Stake) AS Stake FROM ( SELECT Bookie.Bookie, SUM(Amount) AS Amount, Wagers.Stake FROM Transactions JOIN Bookie on Transactions.Bookie = Bookie.Id JOIN Wagers on Bookie.Id = Wagers.Bookie GROUP BY Bookie.Bookie, Wagers.Stake ) AS sub GROUP BY Bookie;

Answer 2

使用来自两个不同表的聚合，最好预先聚合数据，然后才加入：

select 
  b.bookie, 
  t.sum_amount, 
  w.sum_stake
from bookie b 
join
(
  select bookie, sum(amount) as sum_amount
  from transactions
  group by bookie
) t on t.bookie = b.id
join       
(
  select bookie, sum(stake) as sum_stake
  from wagers 
  group by bookie
) w on w.bookie = b.id;

由于您只对每个表中的一个聚合感兴趣，您甚至可以在select子句中使用相关子查询。易于编写和易于阅读： - ）

select 
  b.bookie, 
  (select sum(t.amount) from transactions t where t.bookie = b.id),
  (select sum(w.stake) from wagers w where w.bookie = b.id)
from bookie b

SQL如何正确分组行

2 个答案: