数据库是关于公交车的,所以有三个表:
停止
stop_id | stop_name
--------------------
1 | station_1
2 | station_2
3 | station_3
ROUTE
route_id | route_num
--------------------
1 | route_1
2 | route_2
3 | route_3
ROUTE_STOP
stop_id | route_id
------------------
1 | 1
2 | 1
1 | 2
3 | 2
1 | 3
2 | 3
3 | 3
所以第一条路线有1号和2号站点,第2条路线有1号和3号站点,而第3条路线有站点。
尝试通过station_1和station_3获取route_num
:
SELECT distinct(r.route_num) from STOP s
JOIN ROUTE_STOP rs
ON s.stop_id = rs.stop_id
JOIN ROUTE_STOP r_s
ON rs.stop_id = r_s.stop_id
JOIN ROUTE r
ON rs.route_id = r.route_id
WHERE s.stop_name='station_1' OR s.stop_name='station_3'
AND rs.stop_id <> r_s.stop_id
,结果应该是route_2和route_3,但它不起作用。如果站之间没有路线,则应该没有结果。
如果两个电台之间没有路由,如何获得经过2个电台的route_num
?
答案 0 :(得分:0)
您的查询应该是
SELECT r.route_num from route r
JOIN ROUTE_STOP rs
ON r.route_id = rs.route_id
JOIN STOP s ON rs.stop_id = s.stop_id
WHERE s.stop_name in ('station_1','station_3');
答案 1 :(得分:0)
执行此操作的一种方法是加入表并使用in
子句与group by
和having count()
来查找匹配的组,或者按route_num分组并使用用于过滤具有所需停靠点的组的子句:
-- query 1
SELECT r.route_num
FROM ROUTE r
JOIN ROUTE_STOP rs ON r.route_id = rs.route_id
JOIN STOP s ON rs.stop_id = s.stop_id
WHERE s.stop_name IN ('station_1','station_3')
GROUP BY r.route_num
HAVING COUNT(DISTINCT s.stop_id) = 2;
-- query 2
SELECT r.route_num
FROM ROUTE r
JOIN ROUTE_STOP rs ON r.route_id = rs.route_id
JOIN STOP s ON rs.stop_id = s.stop_id
GROUP BY r.route_num
HAVING (SUM(CASE WHEN s.stop_name = 'station_1' THEN 1 ELSE 0 END) = 1)
AND (SUM(CASE WHEN s.stop_name = 'station_3' THEN 1 ELSE 0 END) = 1);
因为MySQL将布尔表达式计算为1或0,所以可以将上一个查询中的having
子句减少为:
HAVING (SUM(s.stop_name = 'station_1') = 1)
AND (SUM(s.stop_name = 'station_3') = 1);
两个查询都假设在路线中可能会多次停止(如果路线在同一站点开始和结束)。