SQL語句來計算同一表的兩行之間的DateTime差異

Question

我在Table1中有2列： Time_Stamp和RunTimeMinute 。 如何從與RunTimeMinute = 1相對應的Time_Stamp值中減去其中RunTimeMinute = 0的Time_Stamp值（這將使我花費時間來使機器運行）？

Time_Stamp                      RunTimeMinute
    2016-03-01 04:32:10.0000000 1
    2016-03-01 04:33:11.0000000 2
    2016-03-01 04:34:13.0000000 3
    2016-03-01 04:35:15.0000000 4
    2016-03-01 04:36:16.0000000 5
    2016-03-01 04:37:18.0000000 6
    2016-03-01 04:38:20.0000000 7
    2016-03-01 04:39:22.0000000 8
    2016-03-01 04:40:23.0000000 9
    2016-03-01 04:41:16.0000000 0
    2016-03-01 04:45:36.0000000 10

Answer 1

SELECT DATEDIFF（mi，（以（cast_Time_STAMP）as Datetime），運行時間）

Answer 2

要完成您描述的任務（假設Time_Stamp字段是DateTime類型，而RunTimeMinute是一個整數），則可以使用SQL SELECT查詢/子查詢技術和DATEDIFF()函數，如以下示例所示：

SELECT [YourTable].Time_Stamp AS t1,
(SELECT YourTable.Time_Stamp FROM YourTable WHERE YourTable.RunTimeMinute=1) AS t0, 
DateDiff("n",[t0],[t1]) AS ElapsedTimeMin
FROM YourTable
WHERE ([YourTable].RunTimeMinute)=0;

ElapsedTimeMin將以分鍾為單位顯示結果。 您可以將DATEDIFF()函數的“ datepart”指定為“ s”，以秒為單位獲取結果。

希望這會有所幫助。

Answer 3

例如，我們有一個這樣的示例：

CREATE TABLE Table1 (
  Time_Stamp datetime,
  RunTimeMinute int
)
INSERT INTO Table1 VALUES                                         
('2016-03-01 04:32:10.000', 1),
('2016-03-01 04:33:11.000', 2),
('2016-03-01 04:34:13.000', 3),
('2016-03-01 04:35:15.000', 4),
('2016-03-01 04:36:16.000', 5),
('2016-03-01 04:37:18.000', 6),
('2016-03-01 04:38:20.000', 7),
('2016-03-01 04:39:22.000', 8),
('2016-03-01 04:40:23.000', 9),
('2016-03-01 04:41:16.000', 0),
('2016-03-01 04:45:36.000', 10),
('2016-03-01 05:31:10.000', 1),
('2016-03-01 05:35:11.000', 2),
('2016-03-01 05:37:13.000', 3),
('2016-03-01 05:39:15.000', 4),
('2016-03-01 05:41:16.000', 5),
('2016-03-01 05:46:18.000', 6),
('2016-03-01 05:48:20.000', 7),
('2016-03-01 05:51:22.000', 8),
('2016-03-01 05:53:23.000', 9),
('2016-03-01 05:55:16.000', 0),
('2016-03-01 05:57:36.000', 10),
('2016-03-02 05:34:09.000', 1),
('2016-03-02 05:35:14.000', 2),
('2016-03-02 05:36:11.000', 3),
('2016-03-02 05:37:18.000', 4),
('2016-03-02 05:38:20.000', 5),
('2016-03-02 05:39:38.000', 6),
('2016-03-02 05:40:40.000', 7),
('2016-03-02 05:41:12.000', 8),
('2016-03-02 05:42:32.000', 9),
('2016-03-02 05:44:11.000', 0),
('2016-03-02 05:47:38.000', 10)

然后我們這樣做：

;WITH cte AS (
    SELECT  Time_Stamp,
            RunTimeMinute,
            ROW_NUMBER() OVER (PARTITION BY RunTimeMinute ORDER BY Time_Stamp) AS rnum
    FROM Table1
    WHERE RunTimeMinute IN (0,1)
  )

SELECT  MIN(Time_Stamp) as StartTime,
        DATEDIFF(minute, MIN(Time_Stamp), MAX(Time_Stamp)) AS ElapsedTimeMin
FROM cte
GROUP BY rnum

並得到這個：

|               StartTime | ElapsedTimeMin |
|-------------------------|----------------|
| March, 01 2016 04:32:10 |              9 |
| March, 01 2016 05:31:10 |             24 |
| March, 02 2016 05:34:09 |             10 |