Question

我有一个SQL表，如下所示：

Id  Quantity
1   24.857    
2   24.84
3   24.881
4   24.823

我需要选择数量的累积总和，直到达到指定值 - 比方说50。所以使用上面的表格，我需要查询返回前3行。（前两行的数量总和不会达到50，所以我也需要第三行。）

现在我有下面的代码，它虽然接近但不太正确。此查询的结果仅返回表的前两行，这是不够的。

DECLARE @point INT = 50;

WITH x(Id, Quantity, RowNum, TotalQuantity) AS
(
  SELECT
    id, quantity, ROW_NUMBER() OVER (ORDER BY id),
    SUM(quantity) OVER (ORDER BY id RANGE UNBOUNDED PRECEDING)
  FROM inventory
)

SELECT x.Id, x.RowNum, x.Quantity, x.TotalQuantity
FROM x
WHERE x.TotalQuantity <= @point
ORDER BY x.Id

此查询会产生以下结果：

Id  RowNum  Quantity   TotalQuantity
1   1       24.857     24.857
2   2       24.84      49.697

我觉得好像我在这里需要一个WHILE循环，我试图无济于事。

非常感谢任何帮助，谢谢。

Answer 1

当总数量为＆gt; = @point时，您可以再使用一个cte来获取第一行。然后加入现有的cte。

WITH x(Id, Quantity, RowNum, TotalQuantity) AS
(
  SELECT
    id, quantity, ROW_NUMBER() OVER (ORDER BY id),
    SUM(quantity) OVER (ORDER BY id RANGE UNBOUNDED PRECEDING)
  FROM inventory
)
, y as (select min(rownum) as minrownum from x where totalquantity >= @point)
SELECT x.Id, x.RowNum, x.Quantity, x.TotalQuantity
FROM x
JOIN y on x.rownum <= y.minrownum
ORDER BY x.Id

Answer 2

您可以使用LAG查看前一行。

WITH x(Id, Quantity, RowNum, TotalQuantity, TotalQuantityPrior) AS
(
  SELECT
    id, quantity, ROW_NUMBER() OVER (ORDER BY id),
    SUM(quantity) OVER (ORDER BY id RANGE UNBOUNDED PRECEDING) AS TotalQuantity,
    LAG(TotalQuantity,1,0) OVER (ORDER BY id) AS TotalQuantityPrior
  FROM inventory
)

SELECT x.Id, x.RowNum, x.Quantity, x.TotalQuantity
FROM x
WHERE x.TotalQuantityPrior <= @point
ORDER BY x.Id

我默认使用0 - 如果您可以将0作为@point，或者根据您的业务规则可能有其他内容，您可能需要-1。

https://msdn.microsoft.com/en-us/library/hh231256.aspx

如何在SQL中查询累积总和达到特定值

2 个答案: