MySQL行列转化

初始化表结构：

CREATE TABLE `student_scores` (
  `student_id` int NOT NULL,
  `student_name` varchar(50) DEFAULT NULL,
  `math_score` int DEFAULT NULL,
  `english_score` int DEFAULT NULL,
  `science_score` int DEFAULT NULL,
  PRIMARY KEY (`student_id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_0900_ai_ci;

INSERT INTO student_scores (student_id, student_name, math_score, english_score, science_score) VALUES
(1, 'Alice', 85, 90, 78),
(2, 'Bob', 76, 88, 92),
(3, 'Charlie', 90, 85, 80);

查询表数据：
 

行转化为列： 

由于不是我们想要的格式，我们将其转化为列式结构：

-- 行数转化为列
SELECT student_id, student_name, 'Math' AS subject, math_score AS score FROM student_scores
UNION ALL
SELECT student_id, student_name, 'English' AS subject, english_score AS score FROM student_scores
UNION ALL
SELECT student_id, student_name, 'Science' AS subject, science_score AS score FROM student_scores;

执行结果： 

列转化为行： 

 将其作为一张临时表，对其进行行列转化：

select student_id,student_name,
MIN(Case when subject = 'Math' then score end ) as math_score,
MIN(case when subject = 'English' then score end )as english_score,
MIN(case when subject = 'Science' then score end )as science_score 
from  (
    SELECT student_id, student_name, 'Math' AS subject, math_score AS score FROM student_scores
    UNION ALL
    SELECT student_id, student_name, 'English' AS subject, english_score AS score FROM student_scores
    UNION ALL
    SELECT student_id, student_name, 'Science' AS subject, science_score AS score FROM student_scores
) AS unpivoted
GROUP BY unpivoted.student_id,unpivoted.student_name

执行结果：

扩展：

union 与 union all区别

1. UNION：会自动去除合并结果集中的重复记录，只返回唯一的记录。

2. UNION ALL：会返回所有记录，包括重复的记录。