MYSQL多组by和max

这是一张名为posts_vots的表格

id|discussion_id|post_id|user_id|vote_sign|
__________________________________________
1 |      1      |   1   |   1   |     1   |
2 |      1      |   1   |   2   |    -1   |
3 |      1      |   2   |   3   |     1   |
4 |      1      |   2   |   4   |     1   |
5 |      2      |   3   |   1   |    -1   |
6 |      2      |   4   |   2   |     1   |

我想用这些结果创建一个视图：

discussion_id|post_id|score
      1      |   2   | 2
      2      |   4   | 1

与：

• post_id是得分最高的帖子
• 分数是总和（投票号）

我和group by和max一起折磨我的心，但我找不到办法=(

如果有人有主意

---

谢谢；）

首先使用子查询计算分数并为每个讨论id选择最大分数。然后

加入结果集以获得每个讨论id的最大分数的帖子

select t1.*
from (select discussion_id,post_id,sum(vote_sign) as score
      from posts_votes
      group by discussion_id,post_id) t1
join (select discussion_id,max(score) as maxscore 
      from (select discussion_id,post_id,sum(vote_sign) as score
            from posts_votes
            group by discussion_id,post_id) t
      group by discussion_id) t2 
on t1.discussion_id = t2.discussion_id and t1.score = t2.maxscore

这里是与您的问题相对应的查询

select
  discussion_id, 
  max(post_id) as max_post_id,
  sum(vote_sign) as score
from
  posts_votes 
group by
  discussion_id;

这恰好为示例数据提供了正确的解决方案，但是max（post_id）并不总是返回正确的post。是的。这是必需的@草莓..如果使用其他支持窗口功能的dbms，它会更容易。耶！非常感谢。这很好，只有当两个项目的分数相同时才有缺点，但我会用代码来处理它！
select
  discussion_id, 
  max(post_id) as max_post_id,
  sum(vote_sign) as score
from
  posts_votes 
group by
  discussion_id;

select SUBSTRING_INDEX(GROUP_CONCAT(post_id ORDER BY sm DESC), ',', 1)  AS top_post, discussion_id, max(score) as score
from (
    select discussion_id, post_id, sum(vote_sign) as score
    from posts_votes
    group by post_id, discussion_id
) c
group by discussion_id