我有一个包含nazwa,wartosc,czas的过程数据的表。表数据如下所示
现在我写了查询:
;WITH [cteRows] AS
(
SELECT [nazwa],
[wartosc],
[czas],
ROW_NUMBER() OVER (ORDER BY [czas]) AS [RowNumber]
FROM [test].[dbo].[coldbox7]
WHERE [nazwa] = 'coldbox7/WymianaDanychDB/Hydraulic'
AND [czas] >= '2016-07-14 22:00:00.000'
AND [czas] <= '2016-07-14 22:30:00.000'
)
SELECT mc.[RowNumber],
mc.[czas],
mc.[nazwa],
mc.[wartosc],
DATEDIFF(SECOND, mc.[czas], ISNULL(mp.[czas], mc.[czas])) AS [Wsekundach]
FROM [cteRows] mc
LEFT JOIN [cteRows] mp
ON mc.[RowNumber] = mp.[RowNumber] + 1
WHERE mc.[wartosc] = 0
输出如下所示:
我有问题:如何使 [Wsekundach] 的一行总和而不是当前输出?
尝试以下操作:
;WITH [cteRows] AS
(
SELECT [nazwa],
[wartosc],
[czas],
ROW_NUMBER() OVER (ORDER BY [czas]) AS [RowNumber]
FROM [test].[dbo].[coldbox7]
WHERE [nazwa] = 'coldbox7/WymianaDanychDB/Hydraulic'
AND [czas] >= '2016-07-14 22:00:00.000'
AND [czas] <= '2016-07-14 22:30:00.000'
)
, cteNewRows AS
(
SELECT mc.[RowNumber],
mc.[czas],
mc.[nazwa],
mc.[wartosc],
DATEDIFF(SECOND, mc.[czas], ISNULL(mp.[czas], mc.[czas])) AS [Wsekundach]
FROM [cteRows] mc
LEFT JOIN [cteRows] mp ON mc.[RowNumber] = mp.[RowNumber] + 1
WHERE mc.[wartosc] = 0
)
SELECT SUM(Wsekundach) AS SumOfWsekundach
FROM cteNewRows
如果你想要一行是差异,我认为你可以使用一个更简单的查询:
SELECT DATEDIFF(SECOND, MIN(c.[czas]), MAX(c.[czas])
FROM [test].[dbo].[coldbox7] c
WHERE [nazwa] = 'coldbox7/WymianaDanychDB/Hydraulic' AND
[czas] >= '2016-07-14 22:00:00.000' AND
[czas] <= '2016-07-14 22:30:00.000' AND
c.[wartosc] = 0;
仅根据最大值和最小值进行求和比对中间总和求和更有效。