通过另一列的MAX值进行分组

Question

表FieldStudies是：

ID          Name
---|-----------------------|
1  | Industrial Engineering| 
2  | Civil Engineering     |
3  | Architecture          |
4  | Chemistry             |

和表Eductionals是：

ID  UserID  Degree  FieldStudy_ID
---|------|--------|------------|
1  | 100  | 3      | 4          | 
2  | 101  | 2      | 2          |
3  | 101  | 3      | 2          |
4  | 101  | 4      | 3          |
5  | 103  | 3      | 4          |
6  | 103  | 4      | 2          |

我想找到每个FieldStudies的学生FieldStudies ，但要考虑到最高的Degree 。

所需的输出：

ID          Name              Count
---|-----------------------|--------|
1  | Industrial Engineering|   0    |
2  | Civil Engineering     |   0    |
3  | Architecture          |   1    |
4  | Chemistry             |   2    |

我努力了：

select Temptable2.* , count(*) As CountField from
    (select fs.*
    from FieldStudies fs
    left outer join
        (select e.UserID , Max(e.Degree) As ID_Degree , e.FieldStudy_ID
        from Eductionals e
        group by e.UserID) Temptable
    ON fs.ID = Temptable.FieldStudy_ID) Temptable2
group by Temptable2.ID

但是我收到以下错误：

在选择列表中，列“ Eductionals.FieldStudy_ID”无效，因为该列既未包含在聚合函数中，也未包含在GROUP BY子句中。

Answer 1

如果我理解正确，那么每个人只需要最高的学位。 如果是这样，您可以使用row_number()缩减给定人员的多行，其余的是聚合和join ：

select fs.id, fs.Name, count(e.id)
from fieldstudies fs left join
     (select e.*,
             row_number() over (partition by userid order by degree desc) as seqnum
      from educationals e
     ) e
     on e.FieldStudy_ID = fs.id and seqnum = 1
group by fs.id, fs.Name
order by fs.id;