您的位置:首页技术文章

MySQL中列转行和行转列总结解决思路

【字号: 日期:2023-02-05 11:01:23浏览:2作者:猪猪
目录
  • 引言
  • 列转行
  • 行转列
  • 总结

引言

在学习sql中遇到了列转行和行转列的题目,这里总结一下如何在对应的情景下解决不同的题目;

列转行

创建一个表stu_score_01:

SET NAMES utf8mb4;
SET FOREIGN_KEY_CHECKS = 0;
-- ----------------------------
-- Table structure for stu_score_01
-- ----------------------------
DROP TABLE IF EXISTS `stu_score_01`;
CREATE TABLE `stu_score_01` (
  `id` varchar(255) NOT NULL,
  `name` varchar(255) NOT NULL,
  `chinese` varchar(255) DEFAULT NULL,
  `math` varchar(255) DEFAULT NULL,
  `english` varchar(255) DEFAULT NULL,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
-- ----------------------------
-- Records of stu_score_01
-- ----------------------------
BEGIN;
INSERT INTO `stu_score_01` VALUES ("1", "张三", "111", "109", "98");
INSERT INTO `stu_score_01` VALUES ("2", "李四", "89", "119", "109");
INSERT INTO `stu_score_01` VALUES ("3", "王五", "96", "102", "107");
INSERT INTO `stu_score_01` VALUES ("4", "小六", "56", "78", "88");
COMMIT;
SET FOREIGN_KEY_CHECKS = 1;

如果想要把这个表转为下面的形式:

+--------+---------+-------+
| name | project | score |
+--------+---------+-------+
| 张三 | chinese | 111 |
| 李四 | chinese | 89 |
| 王五 | chinese | 96 |
| 小六 | chinese | 56 |
| 张三 | math | 109 |
| 李四 | math | 119 |
| 王五 | math | 102 |
| 小六 | math | 78 |
| 张三 | english | 98 |
| 李四 | english | 109 |
| 王五 | english | 107 |
| 小六 | english | 88 |
+--------+---------+-------+

那么就可以使用union或者union all来实现列转行操作:

select name, "chinese" as project, chinese as score from stu_score_01
union all
select name, "math" as project, math as score from stu_score_01
union all
select name, "english" as project, english as score from stu_score_01;

简单解释一下:分别查询每一个科目的所有情况,求并集即可;比如单独执行一下sql:

select name, "chinese" as project, chinese as score from stu_score_01;
#结果
+--------+---------+-------+
| name   | project | score |
+--------+---------+-------+
| 张三 	| chinese | 111   |
| 李四 	| chinese | 89    |
| 王五 	| chinese | 96    |
| 小六 	| chinese | 56    |
+--------+---------+-------+

接下来只需要一次类推求出所有情况集合求并即可;

union和union all都是求的表的并集,但是union会有去重和排序操作,效率低于union all,这里不需要去重,所以使用union all保证效率;

行转列

创建一个表stu_score_03:

SET NAMES utf8mb4;
SET FOREIGN_KEY_CHECKS = 0;
-- ----------------------------
-- Table structure for stu_score_03
-- ----------------------------
DROP TABLE IF EXISTS `stu_score_03`;
CREATE TABLE `stu_score_03` (
  `id` varchar(255) NOT NULL,
  `name` varchar(255) NOT NULL,
  `project` varchar(255) DEFAULT NULL,
  `score` varchar(255) DEFAULT NULL,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
-- ----------------------------
-- Records of stu_score_03
-- ----------------------------
BEGIN;
INSERT INTO `stu_score_03` VALUES ("1", "张三", "chinese", "111");
INSERT INTO `stu_score_03` VALUES ("10", "李四", "english", "109");
INSERT INTO `stu_score_03` VALUES ("11", "王五", "english", "107");
INSERT INTO `stu_score_03` VALUES ("12", "小六", "english", "88");
INSERT INTO `stu_score_03` VALUES ("2", "李四", "chinese", "89");
INSERT INTO `stu_score_03` VALUES ("3", "王五", "chinese", "96");
INSERT INTO `stu_score_03` VALUES ("4", "小六", "chinese", "56");
INSERT INTO `stu_score_03` VALUES ("5", "张三", "math", "109");
INSERT INTO `stu_score_03` VALUES ("6", "李四", "math", "119");
INSERT INTO `stu_score_03` VALUES ("7", "王五", "math", "102");
INSERT INTO `stu_score_03` VALUES ("8", "小六", "math", "78");
INSERT INTO `stu_score_03` VALUES ("9", "张三", "english", "98");
COMMIT;
SET FOREIGN_KEY_CHECKS = 1;

如果想要单独把每一行科目分别转化为不同的列,如:

+--------+---------+------+---------+
| name   | chinese | math | english |
+--------+---------+------+---------+
| 小六 	| 56      | 78   | 88      |
| 张三 	| 111     | 109  | 98      |
| 李四 	| 89      | 119  | 109     |
| 王五 	| 96      | 102  | 107     |
+--------+---------+------+---------+

可以使用case…when和max/sum和group by来实现:

select name,
max(case when project = "chinese" then score else 0 end) as "chinese",
max(case when project = "math" then score else 0 end) as "math",
max(case when project = "english" then score else 0 end) as "english"
from stu_score_03
group by name;
# 或者使用sum
select name,
sum(case when project = "chinese" then score else 0 end) as "chinese",
sum(case when project = "math" then score else 0 end) as "math",
sum(case when project = "english" then score else 0 end) as "english"
from stu_score_03
group by name;

简单解释一下:

因为要查询每个人的不同科目成绩,所以需要对不同的人进行分组,所以需要使用group by,不然查出来的成绩谁都不知道知道是谁的;

对于每一个case when,比如:case when project = 'chinese' then score else 0 end

意思就是当project为chinese时获取score,否则就取0;其他也是一样的意思

还有为什么需要加上max或者sum,先想象一下如果不加上max或者sum会有什么样的效果:

因为先判断的是chinese科目,如果张三首先出现的科目是math,那么他先走chinese科目的判断,因为math不等于chinese,

所以给chinese科目赋值为0;

所以会看到如下效果:

select name,
case when project = "chinese" then score else 0 end as "chinese",
case when project = "math" then score else 0 end as "math",
case when project = "english" then score else 0 end as "english"
from stu_score_03
group by name;
#执行结果
+--------+---------+------+---------+
| name   | chinese | math | english |
+--------+---------+------+---------+
| 小六 	| 0       | 0    | 88      |
| 张三 	| 111     | 0    | 0       |
| 李四 	| 0       | 0    | 109     |
| 王五 	| 0       | 0    | 107     |
+--------+---------+------+---------+

因为小六最先出现的是english成绩,所以他的chinese和math成绩都被赋予值为0,

而张三最先出现的是chinese成绩,所以他的math和english成绩也被赋予值为0;

如果使用max或者sum,那么max会在出现的所有值的情况下(包括0)去最大的值,其实就是实际的分数,只是把0的情况去除了;

而sum是加上了所有值,因为除了实际分数外其他都是0,所以可以直接相加;

总结

说了这么多,其实可以总结两句话:

列转行,查询需要的每列数据使用union或者union all求并集

行转列,使用case…when分情况查询数据,group by和sum/max进行筛选

到此这篇关于MySQL中列转行和行转列总结解决思路的文章就介绍到这了,更多相关MySQL列转行和行转列内容请搜索以前的文章或继续浏览下面的相关文章希望大家以后多多支持!

标签: MySQL