如何从SQL中的2个表计算？

我有以下2个表格：

A (ID, No1, No2,D) B (ID, No3, No4,D) 

表数据如下：

表格数据

 ID = 1, No1 = 23, No2 = 45, D = 1 ID = 2, No1 = 12, No2 = 5, D = 1 ID = 3, No1 = 14, No2 = 4, D = 1 ID = 4, No1 = 41, No2 = 12, D = 1 ID = 5, No1 = 2, No2 = 25, D = 1 ID = 6, No1 = 3, No2 = 96, D = 1 

B表数据没有任何数据

如果我通过以下SQL查询

 SELECT ISNULL(SUM(dbo.A.No1), 0) AS No1Sum, ISNULL(SUM(dbo.B.No3), 0) AS No3Sum, ISNULL(SUM(dbo.A.No1), 0) - ISNULL(SUM(dbo.B.No3), 0) AS Diffrence FROM dbo.A LEFT OUTER JOIN dbo.B ON dbo.A.id = dbo.B.id GROUP BY dbo.B.No3 

随后数据即将到来

 No1Sum = 95, No3Sum = 0, Diffrence = 95 

但是当我添加WHERE条件并写入查询时，以下查询中没有数据

 SELECT ISNULL(SUM(dbo.A.No1), 0) AS No1Sum, ISNULL(SUM(dbo.B.No3), 0) AS No3Sum, ISNULL(SUM(dbo.A.No1), 0) - ISNULL(SUM(dbo.B.No3), 0) AS Diffrence FROM dbo.A LEFT OUTER JOIN dbo.B ON dbo.A.id = dbo.B.id WHERE (dbo.AD = 1) AND (dbo.BD = 1) GROUP BY dbo.B.No3 

如果数据将在两个表中都显示，那么结果将显示为例如

 No1Sum = 95, No3Sum = 10, Diffrence = 85, 

如果我给出以下查询

 SELECT ISNULL(SUM(dbo.A.No1), 0) AS No1Sum, ISNULL(SUM(dbo.B.No3), 0) AS No3Sum, ISNULL(SUM(dbo.A.No1), 0) - ISNULL(SUM(dbo.B.No3), 0) AS Diffrence FROM dbo.A LEFT OUTER JOIN dbo.B ON dbo.A.id = dbo.B.id AND dbo.AD = 1 AND dbo.BD = 1 GROUP BY dbo.B.No3 

然后关注结果显示

 No1Sum = 570, No3Sum = 0, Diffrence = 570 

（由于A表有6行，它总计6次）

那么如何从两个表中计算出两个表中的数据结果是否准确，但是当A表有数据且B没有数据时，结果就不会到来。

提前致谢

Answer 1

此查询不起作用的原因是因为它只返回dbo.bd为1的记录。但在查询中，它有时可以为null。

 SELECT ISNULL(SUM(dbo.A.No1), 0) AS No1Sum, ISNULL(SUM(dbo.B.No3), 0) AS No3Sum, ISNULL(SUM(dbo.A.No1), 0) - ISNULL(SUM(dbo.B.No3), 0) AS Diffrence FROM dbo.A LEFT OUTER JOIN dbo.B ON dbo.A.id = dbo.B.id WHERE (dbo.AD = 1) AND (dbo.BD = 1) GROUP BY dbo.B.No3 

所以改变这一行

 (dbo.AD = 1) AND (dbo.BD = 1) GROUP BY dbo.B.No3 

对此

 (dbo.AD = 1) AND (isnull(dbo.BD, 1) = 1) GROUP BY dbo.B.No3 

如果arg不为null，则将isnull（arg，default）设置为arg，否则将其设置为default。

Answer 2

我认为这样做，即使Second表有数据。

 SELECT ISNULL(SUM(A.No1), 0) AS No1Sum , ISNULL(SUM(B.No3), 0) AS No3Sum , ISNULL(SUM(A.No1), 0) - ISNULL(SUM(B.No3), 0) AS Diffrence FROM A LEFT OUTER JOIN B ON A.id = B.id 

---

 SELECT ISNULL(SUM(dbo.A.No1), 0) AS No1Sum, ISNULL(SUM(dbo.B.No3), 0) AS No3Sum, ISNULL(SUM(dbo.A.No1), 0) - ISNULL(SUM(dbo.B.No3), 0) AS Diffrence FROM dbo.A LEFT OUTER JOIN dbo.B ON dbo.A.id = dbo.B.id WHERE (dbo.AD = 1) AND (dbo.BD = 1) GROUP BY dbo.B.No3 

如果表B没有任何数据，则此查询将不会获取任何记录，因为这种情况。

  WHERE (dbo.AD = 1) AND (dbo.BD = 1) 

我认为在连接上应用多个条件是开销。

 LEFT OUTER JOIN dbo.B ON dbo.A.id = dbo.B.id AND dbo.AD = 1 --Overhead AND dbo.BD = 1 --Overhead