對兩個相關表中的列求和並按另一列分組（postgresql）

Question

給定兩個表：

• 教師：ID，school_id
• 學校：ID、student_population、地區。

...什么查詢可以按地區匯總師生人數？

樣本數據：

Teachers
ID, school_id
1, 1
2, 1
3, 2
4, 3

Schools
ID, student_population, district
1, 45, Falaba
2, 23, Falaba
3, 87, Bonth
4, 10, Moyamba

預期結果：

teachers, students, district
3, 68, Falaba
1, 87, Bonth
0, 10, Moyamba

我的第一次嘗試是：

select count(*) as teachers, sum(student_population), district from teachers join schools on teachers.school_id = schools.id group by district.

...然而，這個查詢導致每個學校的 student_population 被分解為屬於它的每個老師。 因此，法拉巴的學生總數為 113 (45+45+23) 而不是 68。

Answer 1

假設表teachers有（至少）包含老師的id的第二列，你可以這樣做：

select
    count(distinct t.teacher_id) teachers
    sum(s.student_population) students,
    s.district
from schools s
inner join teachers t on t.school_id = s.id
group by s.district

更好的選擇是預先聚合teachers表：

select
    sum(t.teachers) teachers,
    sum(s.student_population) students,
    s.district
from schools s
inner join (select school_id, count(*) teachers from teachers group by school_id) t 
    on t.school_id = s.id
group by s.district

DB Fiddle 上的演示：

teachers | students | district
-------: | -------: | :-------
       1 |       87 | Bonth   
       3 |       68 | Falaba