Row_Number() 在分区上 - 更简单的方法 [英] Row_Number() Over Partition - Easier Way

问题描述

我有以下查询:

INSERT into @TEST (col1,staffnumber,rn)
SELECT starttime as col1,staffnumber,ROW_NUMBER() OVER (PARTITION BY StaffNumber ORDER BY STARTTIME DESC) AS rn
  FROM @TempTable2 t2
UNION 
SELECT eventdate as col1, staffnumber,ROW_NUMBER() OVER (PARTITION BY StaffNumber ORDER BY EVENTDATE DESC) AS rn
  FROM @TempTable1 t1

但是，显然，这给了我以下值:

But, obviously, this gives me the following values:

COL1                | staffnumber    | rn
10/02/2014 10:29:24         1          1
10/02/2014 09:08:19         1          3
10/02/2014 10:00:23         1          2
10/02/2014 10:00:29         1          1

然而，我需要它显示:

COL1                | staffnumber    | rn
10/02/2014 10:29:24         1          1
10/02/2014 09:08:19         1          4
10/02/2014 10:00:23         1          3
10/02/2014 10:00:29         1          2

我可以通过以下方式做到这一点:

I can do this by doing the following:

INSERT into @TEST1 (col1,staffnumber,rn)
SELECT convert(time,col1),staffnumber,ROW_NUMBER() OVER (PARTITION BY StaffNumber ORDER BY convert(time,COL1) DESC) AS rn
FROM @TEST

但是必须有一种更简单、更有效的方法来做到这一点.有什么想法吗?

But there must be an easier more efficient way of doing it. Any ideas?

推荐答案

计算UNION之后的ROW_NUMBER.

SELECT *, ROW_NUMBER() OVER (PARTITION BY StaffNumber ORDER BY STARTTIME DESC) AS rn
FROM (
 SELECT * FROM T1
 UNION ALL
 SELECT * FROM T2
) x

我怀疑你需要UNION ALL.

这篇关于Row_Number() 在分区上 - 更简单的方法的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！