Question

我不知道标题是否有意义，但情况如下：见下表2

人员（顺便说一下，如果有用的话，这个表永远不会有> 1000行）

+----+---------+-------------------+---------+
| id |  name   |      address      | city_id |
+----+---------+-------------------+---------+
|  1 | person1 | some address      |     123 |
|  2 | person2 | another address   |     542 |
|  3 | person3 | different address |     623 |
+----+---------+-------------------+---------+

城市（这个城市可能包含全球所有州（以及国家/地区的addl。列））

+-----+-------+--------+
| id  | city  | state  |
+-----+-------+--------+
| 123 | city1 | state1 |
| 542 | city2 | state1 |
| 623 | city3 | state2 |
+-----+-------+--------+

首先，我只知道people.id。使用此功能，我需要找到属于同一state的所有人（不同city ）。例如，如果我有people.id=1，我需要让所有来自州的人员（person.id = 1）属于：

输出：

+----+---------+-----------------+---------+
| id |  name   |     address     | city_id |
+----+---------+-----------------+---------+
|  1 | person1 | some address    |     123 |     /*Both the people are from state1*/
|  2 | person2 | another address |     542 |
+----+---------+-----------------+---------+

我能够在两个查询中实现这一点：存储

输出的变量$state

SELECT c.state from people p INNER JOIN cities c ON p.city_id=c.id where p.id=<my input>;

然后是另一个查询

SELECT a.* FROM {人{1}} {城市{1}}

使用单个JOIN是否有更有效的方法来实现这一目标？我尝试将两个查询与JOIN（在子查询中）中的JOIN组合到a INNER JOIN，这在某种程度上感觉不对。

P.S：我不是在寻找关于规范化或其他模式更改的建议。所有这些都已经考虑到另一个开发分支以供以后升级。

Answer 1

您可以尝试以下查询 -

SELECT p2.*
FROM people p 
JOIN cities c ON p.city_id=c.id 
JOIN cities c2 ON c.state=c2.state  
JOIN people p2 ON p2.city_id=c2.id 
WHERE p.id=<my input>;

注意：对于人员表中的性能ID和city_id，以及城市中的id和状态表应该被编入索引。

另外，对于更多优化，您应该使用state_id而不是state来进行连接，为此您必须在表中创建state_id字段。

Answer 2

你可以试试这个。

select * from people
where city_id in(
    select city from cities c
    inner join(
        select c.state from people p
        left join cities c on c.city = p.city_id
        where p.id = '1'
    ) s on s.state = c.state
)

Answer 3

SELECT * 
FROM people p1 
WHERE p1.city_id in (
    SELECT c1.id 
    FROM cities c1 
    WHERE c1.state IN (
        SELECT c2.state 
        FROM people p2,
             cities c2 
        WHERE c2.id = p2.city_id 
          AND p2.id = 1
        )
    );

而不是查询中的p2.id = 1，请从人员表中提供您要获取数据的人员ID。

Answer 4

Select a.* from 
people as a
inner join (
select cityid from people where personid = <my input>
) as b on b.cityid = a.cityid

- 或 -

WITH CTE as
(
    select cityid from people where personid = <my input>
)

select a.* from people as a
inner join CTE as b on b.cityid = a.cityid;

SQL从2个表

4 个答案: