Sql 为什么父子表上的这种连接会导致错误的汇总？

我在写这个查询来汇总一些表字段的值时感到困惑。结果摘要是错误的。我简化了设计以便于理解。 表和字段如下所示：

Table [City] => {ID, CityName}
Table [A]    => {ID, CityID, Date, Value}
Table [B_1]  => {ID, AID, Value}
Table [B_2]  => {ID, AID, Value}

如您所见，[city]的表是[A]的父表，[A]是[B_1]和[B_2]的父表

以下是查询：

SELECT     City.CityName, A.Date, SUM(A.Value), SUM(B_1.Value), SUM(B_2.Value)
FROM         A INNER JOIN
                      B_1 ON A.ID = B_1.AID INNER JOIN
                      B_2 ON A.ID = B_2.AID INNER JOIN
                      City ON A.CityID = City.ID
GROUP BY City.CityName, A.Date

假设我们有这些值：

Data for Table [City] =>
ID  CityName
--  --------
1   LA
2   NYC

Data for Table [A] =>
ID  CityID  Date  Value 
--  ------  ----  -----
1   1       2014  8
2   1       2014  5 
3   2       2014  3

Data for Table [B_1] =>
ID  AID  Value 
--  ---  -----
1   1    3
2   1    6 
3   2    7
4   2    2

Data for Table [B_2] =>
ID  AID  Value 
--  ---  -----
1   1    8
2   1    5 
3   2    3
4   2    1

我正在努力实现这一结果：

CityName  Date  AValue  B1Value  B2Value
--------  ----  ------  -------  -------
LA        2014  13      18       17

但结果是：

CityName  Date  AValue  B1Value  B2Value
--------  ----  ------  -------  -------
LA        2014  52      36       34

---

有什么问题吗？

您正在沿不同维度进行聚合，并得到每个“a”的叉积。您可以通过预聚合结果来解决此问题：

SELECT City.CityName, A.Date, SUM(A.Value), SUM(BVALUE1), SUM(BVALUE2)
FROM  A INNER JOIN
      (SELECT AID, SUM(VALUE) as BVALUE1
       FROM B_1
       GROUP BY AID
      ) B_1
      ON A.ID = B_1.AID INNER JOIN
      (SELECT AID, SUM(VALUE) as BVALUE2
       FROM B_2
       GROUP BY AID
      ) B_2
      ON A.ID = B_2.AID INNER JOIN
      City
      ON A.CityID = City.ID
GROUP BY City.CityName, A.Date;