执行递归SQL查询（SQL Server 2005中）

Question

我有这个表：

ID_Articulo ID_Componente Tipo Cantidad 
1   5    2  1.5 
5   3    1  3 
1   6    1  6 
2   3    1  3.5 
6   8    2  4.2 
8   9    1  2 

我需要找到Cantidad字段的总和对于一个给定ID_Articulo和TIPO = 1。例如，对于ID_Articulo = 1应该是1.5 * 3 + 6 + 6 * 4.2 * 2 = 60.9。

每个ID_Articulo的级别（深度）数量是可变的。

这可以用SQL查询吗？我的数据库是SQL Server 2005中

aditional的信息

场ID_Articulo和ID_Compenente都与同一个表Articulos。这就是为什么数据是递归的。因此，对于给定的例子，我有：

ReC#1 is not Tipo=1, but relates ID 1 with ID 5 
ReC#2 relates ID 5 with ID 3 and Tipo=1 So, I have 1.5 (of ReC#1) * 3 
ReC#3 relates ID 1 with ID 6 and Tipo=1, so i have 6 
ReC#4 is from another ID 
ReC#5 relates ID 6 with ID 8 but Tipo!=1 so i don't sum 
ReC#6 relates ID 8 and ID 9 and Tipo=1, so I have 6 * 4.2 * 2 

最终解决

这是最后的代码（有些字段名称不同）：

;WITH CTE AS (
SELECT b.ID_Articulo, ID_Componente, Tipo, Cantidad, 
     CAST(Cantidad AS DECIMAL(6,2)) AS partialSum 
FROM Arbol b inner join articulos a on a.ID_Articulo=b.ID_Componente 
WHERE b.ID_Articulo = 2716 

UNION ALL 

SELECT t.ID_Articulo, t.ID_Componente, t.Tipo, t.Cantidad, 
     CAST(c.partialSum * t.Cantidad AS DECIMAL(6,2)) AS partialSum 
FROM (SELECT b.ID_Articulo, ID_Componente, A.Tipo, Cantidad FROM Arbol b inner join articulos a on a.ID_Articulo=b.ID_Componente inner join Articulos a2 on a2.ID_Articulo=b.ID_Articulo where a2.Tipo<>'I') as t 
INNER JOIN CTE AS c ON c.ID_Componente = t.ID_Articulo 
) 
SELECT SUM(partialSum) 
FROM CTE 
WHERE Tipo = 'I' 

感谢@ giorgos，betsos为最终解决方案

Answer 1

您可以使用以下递归CTE来获得预期结果：

;WITH CTE AS (
    SELECT ID_Articulo, ID_Componente, Tipo, Cantidad, 
      CAST(Cantidad AS DECIMAL(6,1)) AS partialSum 
    FROM mytable 
    WHERE ID_Articulo = 1 

    UNION ALL 

    SELECT t.ID_Articulo, t.ID_Componente, t.Tipo, t.Cantidad, 
      CAST(c.partialSum * t.Cantidad AS DECIMAL(6,1)) AS partialSum 
    FROM mytable AS t 
    INNER JOIN CTE AS c ON t.ID_Articulo = c.ID_Componente 
) 
SELECT SUM(partialSum) 
FROM CTE 
WHERE Tipo = 1 

什么上述递归呢，是返回从ID_Articulo = 1发出所有分支，与Cantidad值相乘的累积产物沿着：

ID_Articulo ID_Componente Tipo Cantidad partialSum 
------------------------------------------------------------------ 
1    5    2  1.5   1.5 
1    6    1  6.0   6.0 
6    8    2  4.2   25.2 
8    9    1  2.0   50.4 
5    3    1  3.0   4.5 

结果是使用SUM上partialSum计算只有对于那些行Tipo = 1。

Demo here