SQL Server-汇总数据

Question

以下是我的数据。如何根据ID和会议室汇总数据，但如果会议室重复，则不应汇总到首次出现？

应该像下面的结果集一样进行汇总。

数据集：

ID      Room    Length  AD
-----------------------------------
1001    MM        2     2018-06-26
1001    MM        3     2018-06-26
1001    MM        0.5   2018-06-26
1001    MM        3     2018-06-28
1001    A5        3.5   2018-06-29
1001    MM        4.4   2018-06-29
1001    MM        3.2   2018-06-30
1001    A5        2.1   2018-07-02
1001    A5        1.7   2018-07-03

预期结果：

ID     Room   Sum_Length
-----------------------
1001    MM      8.5
1001    A5      3.5
1001    MM      7.6
1001    A5      3.8

Answer 1

似乎您需要row_number()的区别：

select id, room, sum(Length)
from (select t.*, row_number() over (partition by id order by ad) seq1,
                  row_number() over (partition by id, room order by ad) seq2
      from table t
     ) t
group by id, room, (seq1 - seq2);