Question

这是我的表

Id         |   ReceiverId  |   SenderId   | Text
-----------+---------------+--------------+-----------
1          |   5           |   1          | text
2          |   5           |   1          | text
3          |   1           |   5          | text
4          |   2           |   5          | text
5          |   2           |   5          | text
6          |   5           |   3          | text
7          |   5           |   4          | text
9          |   5           |   6          | text
10         |   5           |   4          | text
11         |   10          |   5          | text
12         |   5           |   10         | text
13         |   10          |   5          | text
14         |   5           |   10         | text

如何根据[ReceiverId，SenderId]对和按ID排序 降序顺序选择没有重复的行。那就是：[5,1] = [1,5]是重复的。 OR [5,1] = [5,1]也是副本。

所以最终的结果应该是：

Id         |   ReceiverId  |   SenderId   | Text
-----------+---------------+--------------+-----------
14         |   10          |   5          | text
9          |   5           |   6          | text
7          |   5           |   4          | text
6          |   5           |   3          | text
5          |   2           |   5          | text
3          |   1           |   5          | text

Answer 1

假设在记录中，您只需检查SenderId和ReceiverId（顺序无关紧要）就认为是相同的，那么您需要具有最大Id的记录（这可能是最新的）。然后，此查询将为您提供结果：

select Id, ReceiverId, SenderId, [Text]
from MyTable t
where t.Id in (
    select top 1 tt.Id
    from MyTable tt
    where (tt.SenderId = t.SenderId and tt.ReceiverId = t.ReceiverId) or
         (tt.SenderId = t.ReceiverId and tt.ReceiverId = t.SenderId)
    order by tt.Id desc
)
order by t.Id desc

将MyTable替换为您的表名。

Answer 2

one of the displayed URL's a substring "test"

使用sql从表中选择unqiue值

2 个答案: