Question

我的源表中有这种类型的数据。

site              date_time        data
1016771     2016-06-04 00:00:00    LBV
1016771     2016-06-04 00:00:00     SOB
1016771     2016-06-04 01:29:00     LBV
1016771     2016-06-04 01:29:00     SOB
1016771     2016-06-04 05:51:00     SOB
1016771     2016-06-04 05:51:00     LBV
1016771     2016-06-04 08:35:00     SOB
1016771     2016-06-04 08:35:00     LBV
1016771     2016-06-04 11:18:00     LBV
1016771     2016-06-04 11:18:00     SOB
1016771     2016-06-04 12:20:00     LBV
1016771     2016-06-04 12:20:00     SOB
1016771     2016-06-04 13:24:00     LBV
1016771     2016-06-04 13:24:00     SOB
1016771     2016-06-04 14:04:00     LBV
1016771     2016-06-04 14:04:00     SOB
1016771     2016-06-04 15:27:00     LBV
1016773     2016-06-04 15:27:00     SOB
1016773     2016-06-04 16:55:00     LBV
1016773     2016-06-04 16:55:00     SOB
1016773     2016-06-04 18:15:00     LBV
1016773     2016-06-04 18:15:00     SOB

现在我想对这个数据模型执行一些操作。我想要在同一行中的下一行数据。输出是这样的：

site            SOB_date_time             LBV_date_time 
1016771     2016-06-04 00:00:00         2016-06-04 01:29:00    
1016771     2016-06-04 05:51:00         2016-06-04 05:51:00 
1016771     2016-06-04 08:35:00         2016-06-04 08:35:00     
1016771     2016-06-04 11:18:00         2016-06-04 12:20:00    
1016771     2016-06-04 12:20:00         2016-06-04 13:24:00     
1016771     2016-06-04 13:24:00         2016-06-04 14:04:00    
1016771     2016-06-04 14:04:00         2016-06-04 15:27:00 
1016773     2016-06-04 15:27:00         2016-06-04 16:55:00 
1016773     2016-06-04 16:55:00         2016-06-04 18:15:00

请你帮我解决这个问题。

Answer 1

如果您只有2个不同的data，则可以使用OUTER APPLY：

SELECT  y.[site],   
        y.date_time as SOB_date_time,
        t.date_time as LBV_date_time
FROM YourTable y
OUTER APPLY (
    SELECT TOP 1 *
    FROM YourTable
    WHERE y.[site] = [site]
        AND y.date_time < date_time
        AND [data] = 'LBV'
    ORDER BY date_time
) as t
WHERE y.[data] = 'SOB'

输出：

site    SOB_date_time       LBV_date_time
1016771 2016-06-04 00:00:00 2016-06-04 01:29:00
1016771 2016-06-04 01:29:00 2016-06-04 05:51:00
1016771 2016-06-04 05:51:00 2016-06-04 08:35:00
1016771 2016-06-04 08:35:00 2016-06-04 11:18:00
1016771 2016-06-04 11:18:00 2016-06-04 12:20:00
1016771 2016-06-04 12:20:00 2016-06-04 13:24:00
1016771 2016-06-04 13:24:00 2016-06-04 14:04:00
1016771 2016-06-04 14:04:00 2016-06-04 15:27:00
1016773 2016-06-04 15:27:00 2016-06-04 16:55:00
1016773 2016-06-04 16:55:00 2016-06-04 18:15:00
1016773 2016-06-04 18:15:00 NULL

TOP 1 WITH TIES和ROW_NUMBER（）的另一种方式：

SELECT  TOP 1 WITH TIES 
        y.[site],   
        y.date_time as SOB_date_time,
        y1.date_time as LBV_date_time
FROM YourTable y
LEFT JOIN YourTable y1
    ON y.[site] = y1.[site] and y1.[data] = 'LBV' AND y.date_time < y1.date_time
WHERE y.[data] = 'SOB'
ORDER BY ROW_NUMBER() OVER (PARTITION BY y.[site],y.date_time ORDER BY   y1.date_time)

需要在一行中显示具有不同条件的下一行数据

1 个答案: