问题背景

项目需求用到了一个字段是一个集合类型，但表关联是多对多的形式，所以在join之后出现多条数据
注意事项：

sql执行优先级：where > group by > having > order by

例如：

需要变成

建表语句：


SET NAMES utf8mb4;
SET FOREIGN_KEY_CHECKS = 0;

-- ----------------------------
-- Table structure for Student
-- ----------------------------
DROP TABLE IF EXISTS `Student`;
CREATE TABLE `Student`  (
  `Name` varchar(255) CHARACTER SET utf8mb4 COLLATE utf8mb4_bin NULL DEFAULT NULL,
  `Course` varchar(255) CHARACTER SET utf8mb4 COLLATE utf8mb4_bin NULL DEFAULT NULL
) ENGINE = InnoDB CHARACTER SET = utf8mb4 COLLATE = utf8mb4_bin ROW_FORMAT = Dynamic;

-- ----------------------------
-- Records of Student
-- ----------------------------
INSERT INTO `Student` VALUES ('张三', '大学语文');
INSERT INTO `Student` VALUES ('李四', '大学语文');
INSERT INTO `Student` VALUES ('张三', '钢琴');
INSERT INTO `Student` VALUES ('张三', '吉他');
INSERT INTO `Student` VALUES ('李四', '架子鼓');

SET FOREIGN_KEY_CHECKS = 1;

组装sql

1 方案一：https://www.cnblogs.com/feiyuhuo/p/10748112.html
这个方案尝试过，但是会报错，也许你们会成功

You have an error in your SQL syntax;
check the manual that corresponds to your MySQL server version 
for the right syntax to use near 'XML PATH('')), 1, 1, '') AS Course FROM student AS T' at line 1

2 方案二，使用高级函数GROUP_CONCA和分组group by，可以进行合并

SELECT Name, GROUP_CONCAT(Course SEPARATOR ',') as Course from Student GROUP BY `Name`;

应用到项目

1 项目中多表关联非常复杂，需要排序，合并，分组等，这里提一下思路
原本表与表关联之后，可以看到id有两条，这两条只有majorName字段不同，我需要做的就是把这两条合并成一条，majorName合并在一起，逗号分隔

这里需求是我需要把所有的查询出来之后，还要进行排序，还可以进行模糊查询，LEFT JOIN就是左边的表所有值都需要包含

SELECT
	* 
FROM
	(
	SELECT
		m.id AS id,
		m.`name` AS NAME,
		m.`code` AS CODE,
		m.data_type AS dataType,
		m.note AS note,
		m.type AS type,
		s.`name` AS sourceName,
		i.`name` AS interfaceName,
		m.`interface_id` AS interfaceId,
		GROUP_CONCAT( d.`code` SEPARATOR ',' ) AS majorName 
	FROM
		datax_metric m
		LEFT JOIN source s ON m.source_id = s.id
		LEFT JOIN source_interface i ON m.interface_id = i.id
		LEFT JOIN metric_field f ON m.interface_id = f.interface_id
		LEFT JOIN field d ON f.master_field = d.id 
	WHERE
		m.STATUS = 1 
	GROUP BY
		m.id 
	ORDER BY
		m.create_time DESC,
		m.id DESC 
	) AS a 
WHERE
	majorName LIKE '%nowtime%' 
	OR majorName LIKE '%nub%'