使用MySQL，我可以执行以下操作：
SELECT hobbies FROM peoples_hobbies WHERE person_id = 5;
我的输出：
shopping
fishing
coding
但我只想要1行1列：
预期输出：
shopping, fishing, coding
原因是我从多个表中选择了多个值，在所有的连接之后，我得到了比我想要的多得多的行。
我在MySQL Doc上查找了一个函数，它看起来不像CONCAT或CONCAT_WS函数接受结果集。
这里有人知道怎么做吗？
当前回答
您可以使用GROUP_CONCAT:
SELECT person_id,
GROUP_CONCAT(hobbies SEPARATOR ', ')
FROM peoples_hobbies
GROUP BY person_id;
正如Ludwig在评论中所说，您可以添加DISTINCT运算符以避免重复：
SELECT person_id,
GROUP_CONCAT(DISTINCT hobbies SEPARATOR ', ')
FROM peoples_hobbies
GROUP BY person_id;
正如Jan在他们的评论中所说，您也可以在使用ORDER BY将值内爆之前对其进行排序：
SELECT person_id,
GROUP_CONCAT(hobbies ORDER BY hobbies ASC SEPARATOR ', ')
FROM peoples_hobbies
GROUP BY person_id;
正如Dag在评论中所说，结果有1024字节的限制。要解决此问题，请在查询之前运行此查询：
SET group_concat_max_len = 2048;
当然，您可以根据需要更改2048。要计算和分配值：
SET group_concat_max_len = CAST(
(SELECT SUM(LENGTH(hobbies)) + COUNT(*) * LENGTH(', ')
FROM peoples_hobbies
GROUP BY person_id) AS UNSIGNED);
其他回答
本例中另一个有趣的例子-
下面是people_hobbies示例表的结构-
DESCRIBE people_hobbies;
+---------+--------------+------+-----+---------+----------------+
| Field | Type | Null | Key | Default | Extra |
+---------+--------------+------+-----+---------+----------------+
| id | int unsigned | NO | PRI | NULL | auto_increment |
| ppl_id | int unsigned | YES | MUL | NULL | |
| name | varchar(200) | YES | | NULL | |
| hby_id | int unsigned | YES | MUL | NULL | |
| hobbies | varchar(50) | YES | | NULL | |
+---------+--------------+------+-----+---------+----------------+
该表的填充方式如下-
SELECT * FROM people_hobbies;
+----+--------+-----------------+--------+-----------+
| id | ppl_id | name | hby_id | hobbies |
+----+--------+-----------------+--------+-----------+
| 1 | 1 | Shriya Jain | 1 | reading |
| 2 | 4 | Shirley Setia | 4 | coding |
| 3 | 2 | Varsha Tripathi | 7 | gardening |
| 4 | 3 | Diya Ghosh | 2 | fishing |
| 5 | 4 | Shirley Setia | 3 | gaming |
| 6 | 1 | Shriya Jain | 6 | cycling |
| 7 | 2 | Varsha Tripathi | 1 | reading |
| 8 | 3 | Diya Ghosh | 5 | shopping |
| 9 | 3 | Diya Ghosh | 4 | coding |
| 10 | 4 | Shirley Setia | 1 | reading |
| 11 | 1 | Shriya Jain | 4 | coding |
| 12 | 1 | Shriya Jain | 3 | gaming |
| 13 | 4 | Shirley Setia | 2 | fishing |
| 14 | 4 | Shirley Setia | 7 | gardening |
| 15 | 2 | Varsha Tripathi | 3 | gaming |
| 16 | 2 | Varsha Tripathi | 2 | fishing |
| 17 | 1 | Shriya Jain | 5 | shopping |
| 18 | 1 | Shriya Jain | 7 | gardening |
| 19 | 3 | Diya Ghosh | 1 | reading |
| 20 | 4 | Shirley Setia | 5 | shopping |
+----+--------+-----------------+--------+-----------+
现在，生成了一个表hobby_list，其中包含所有人的列表和每个人的爱好列表，每个爱好在一行中-
CREATE TABLE hobby_list AS
-> SELECT ppl_id, name,
-> GROUP_CONCAT(hobbies ORDER BY hby_id SEPARATOR "\n")
-> AS hobbies
-> FROM people_hobbies
-> GROUP BY ppl_id
-> ORDER BY ppl_id;
SELECT * FROM hobby_list;
您可以使用GROUP_CONCAT:
SELECT person_id,
GROUP_CONCAT(hobbies SEPARATOR ', ')
FROM peoples_hobbies
GROUP BY person_id;
正如Ludwig在评论中所说，您可以添加DISTINCT运算符以避免重复：
SELECT person_id,
GROUP_CONCAT(DISTINCT hobbies SEPARATOR ', ')
FROM peoples_hobbies
GROUP BY person_id;
正如Jan在他们的评论中所说，您也可以在使用ORDER BY将值内爆之前对其进行排序：
SELECT person_id,
GROUP_CONCAT(hobbies ORDER BY hobbies ASC SEPARATOR ', ')
FROM peoples_hobbies
GROUP BY person_id;
正如Dag在评论中所说，结果有1024字节的限制。要解决此问题，请在查询之前运行此查询：
SET group_concat_max_len = 2048;
当然，您可以根据需要更改2048。要计算和分配值：
SET group_concat_max_len = CAST(
(SELECT SUM(LENGTH(hobbies)) + COUNT(*) * LENGTH(', ')
FROM peoples_hobbies
GROUP BY person_id) AS UNSIGNED);
现在已经很晚了，但对于那些正在搜索“使用透视表将多个MySQL行连接到一个字段中”的人来说会很有帮助：）
查询：
SELECT pm.id, pm.name, GROUP_CONCAT(c.name) as channel_names
FROM payment_methods pm
LEFT JOIN payment_methods_channels_pivot pmcp ON pmcp.payment_method_id = pm.id
LEFT JOIN channels c ON c.id = pmcp.channel_id
GROUP BY pm.id
桌子
payment_methods
id | name
1 | PayPal
channels
id | name
1 | Google
2 | Faceook
payment_methods_channels_pivot
payment_method_id | channel_id
1 | 1
1 | 2
输出：
有一个GROUP聚合函数GROUP_CONCAT。
如果您的MySQL版本（4.1）支持GROUP_CONCAT，请查看GROUP_CONNAT。有关详细信息，请参阅文档。
它看起来像：
SELECT GROUP_CONCAT(hobbies SEPARATOR ', ')
FROM peoples_hobbies
WHERE person_id = 5
GROUP BY 'all';
