我有以下表格: MOVIES
+-------------+-------------+------+-----+---------+----------------+
| Field | Type | Null | Key | Default | Extra |
+-------------+-------------+------+-----+---------+----------------+
| id | int(11) | NO | PRI | NULL | auto_increment |
| user_id | int(11) | YES | MUL | NULL | |
| title | varchar(40) | NO | | NULL | |
| description | text | NO | | NULL | |
| created_at | datetime | YES | | NULL | |
| updated_at | datetime | YES | | NULL | |
+-------------+-------------+------+-----+---------+----------------+
MOVIE_VOTES
+------------+------------+------+-----+---------+----------------+
| Field | Type | Null | Key | Default | Extra |
+------------+------------+------+-----+---------+----------------+
| id | int(11) | NO | PRI | NULL | auto_increment |
| user_id | int(11) | YES | MUL | NULL | |
| movie_id | int(11) | YES | MUL | NULL | |
| like | tinyint(1) | YES | | NULL | |
| created_at | datetime | YES | | NULL | |
| updated_at | datetime | YES | | NULL | |
+------------+------------+------+-----+---------+----------------+
我想查询按照他们喜欢的数量排序的所有电影。我有这个问题:
SELECT *, COUNT(*) AS count_likes
FROM movie_votes
RIGHT JOIN movies
ON movies.id=movie_votes.movie_id
WHERE movie_votes.like = 1
GROUP BY movie_id
ORDER BY count_likes;
这将按照喜欢的数量排序电影但不是尚未投票(喜欢)的电影。出于某种原因,右连接用作简单连接。 我搜索了一些相关的答案,但找不到任何有用的东西。
答案 0 :(得分:1)
在执行外部连接时应指定列名,以便它只计算非空值
SELECT movies.id,
movies.user_id,
movies.title,
movies.description,
movies.created_at,
movies.updated_at,
COUNT(movie_votes.movie_id) AS count_likes
FROM movies
LEFT JOIN movie_votes
ON movies.id = movie_votes.movie_id
AND movie_votes.like = 1
GROUP BY movies.id,
movies.user_id,
movies.title,
movies.description,
movies.created_at,
movies.updated_at
ORDER BY count_likes
条件movie_votes.like = 1
应放在ON
子句中,以便在将其加入表movie_votes
之前先过滤表movies
中的行。