Question

我有一个包含5列的表格：

PK_Column, FK_MeterID, Date, Index, Value

我有这个查询工作正常：

SELECT FK_MeterID,
       sum(Value) as 'Total',
       @p_Month,
       @p_Year
FROM ValuesHistory
WHERE month(Date) = @p_Month
      and year(Date)  = @p_Year
GROUP BY  FK_MeterID

然而，我想做

，而不是将值超过一个月

月份的最后一个索引与

的第一个索引之间的索引差异

月。（它应该返回相同的结果）

如何使用差异进行此查询？

由于

Answer 1

听起来好像在寻找aggregate functions。

WITH CTE AS
(
    SELECT RN = ROW_NUMBER() OVER (
                  PARTITION BY FK_MeterID
                  ORDER BY Date, PK_Column),
           TOTAL  = SUM(Value) OVER (PARTITION BY FK_MeterID),
           MINVAL = MIN(Value)  OVER (PARTITION BY FK_MeterID),
           MAXVAL = MAX(Value)  OVER (PARTITION BY FK_MeterID),
           PK_Column, FK_MeterID, [Date], [Index], Value
    FROM ValuesHistory
    WHERE month(Date) = @p_Month
      AND year(Date)  = @p_Year
)
SELECT DIFFERENCE = MAXVAL - MINVAL, CTE.*
FROM CTE

Sql server差异而不是sum

1 个答案: