MySQL系列 | MySQL5.7特性：JSON数据类型 [转载]

概述

MySQL5.7的发行声明中，官方称之为里程碑式的版本，除了运行速度大幅度提升之外，还添加了之前版本没有的功能，如本文所述的原生JSON数据类型功能。
在此版本之前，MySQL所有的JSON数据类型，全部是使用text等文本类型来实现的，数据的处理只能在应用代码级来实现，十分不方便。

什么是JSON类型

作为DBA，可能会对这个概念稍微有点陌生，但是对于开发者来说，这是一个十分熟悉的事物。

JSON(JavaScript Object Notation, JS 对象简谱) 是一种轻量级的数据交换格式。它基于 ECMAScript (欧洲计算机协会制定的js规范)的一个子集，采用完全独立于编程语言的文本格式来存储和表示数据。简洁和清晰的层次结构使得 JSON 成为理想的数据交换语言。 易于人阅读和编写，同时也易于机器解析和生成，并有效地提升网络传输效率。

MySQL原始JSON类型的优势在哪？

原生的JSON优势如下：

  1. 存储上类似text,可以存非常大的数据。
  2. 存储在JSON列中的JSON文档的自动验证 。无效的文档会产生错误。
  3. 优化的存储格式。存储在JSON列中的JSON文档将 转换为内部格式，以允许对文档元素进行快速读取访问。
  4. 相比于传统形式，不需要遍历所有字符串才能找到数据。
  5. 支持索引：通过虚拟列的功能可以对JSON中部分的数据进行索引。

MySQL的JSON类型

创建JSON类型表

创建一个基础的员工表，除了工号字段外，还有一个个人基础信息字段和一个个人能力信息字段

?
1
2
3
4
5
6
7
8
9
10
11
12
MySQL [test]> CREATE TABLE  employee (
    ->
    ->     `empno` int(10) unsigned NOT NULL AUTO_INCREMENT,
    ->
    ->     `basic_info` JSON NOT NULL,
    ->
    ->     `skill_info` JSON NOT NULL,
    ->
    ->     PRIMARY KEY (`empno`)
    ->
    -> );
Query OK, 0 rows affected (0.02 sec)

表的基础信息，其中JSON类型的字段，是不可以有默认值的，这点需要注意

?
1
2
3
4
5
6
7
8
9
MySQL [test]> desc employee;
+------------+------------------+------+-----+---------+----------------+
| Field      | Type             | Null | Key | Default | Extra          |
+------------+------------------+------+-----+---------+----------------+
| empno      | int(10) unsigned | NO   | PRI | NULL    | auto_increment |
| basic_info | json             | NO   |     | NULL    |                |
| skill_info | json             | NO   |     | NULL    |                |
+------------+------------------+------+-----+---------+----------------+
3 rows in set (0.00 sec)

试着插入几条数据

我们手动插入几条数据进这张表中，在前两条数据中，在个人能力信息上使用的是数组的方式，，后面两条则是使用对象的形式。在MySQL5.7.8版本后的JSON类型中，这两种都是可以的

?
1
2
3
4
5
6
7
INSERT INTO `employee`  VALUES (1,'{"name": "wangyiyi", "age": "23" ,"from": "hangzhou"}', '["java", "go", "python"]');
 
INSERT INTO `employee`  VALUES (2,'{"name": "linxue", "age": 24 ,"from": "shanghai"}', '["mysql", "oracle", "python"]');
 
INSERT INTO `employee`  VALUES (3,'{"name": "zhaoqing", "age": 24 ,"from": "shanghai"}', '{"system": "linux","database": "mysql", "language": "python"}');
 
INSERT INTO `employee`  VALUES (4,'{"name": "zhouxixi", "age": 30 ,"from": "nanjing"}', '{"system": ["linux","windows"],"database": ["mysql","oracle","postgresql"], "language": ["python","java","go"]}');

插入多个数据后,表中内容为如下

?
1
2
3
4
5
6
7
8
9
10
MySQL [test]> select * from employee;
+-------+-------------------------------------------------------+-----------------------------------------------------------------------------------------------------------------------+
| empno | basic_info                                            | skill_info                                                                                                            |
+-------+-------------------------------------------------------+-----------------------------------------------------------------------------------------------------------------------+
|     1 | {"age": "23", "from": "hangzhou", "name": "wangyiyi"} | ["java", "go", "python"]                                                                                              |
|     2 | {"age": 24, "from": "shanghai", "name": "linxue"}     | ["mysql", "oracle", "python"]                                                                                         |
|     3 | {"age": 24, "from": "shanghai", "name": "zhaoqing"}   | {"system": "linux", "database": "mysql", "language": "python"}                                                        |
|     4 | {"age": 30, "from": "nanjing", "name": "zhouxixi"}    | {"system": ["linux", "windows"], "database": ["mysql", "oracle", "postgresql"], "language": ["python", "java", "go"]} |
+-------+-------------------------------------------------------+-----------------------------------------------------------------------------------------------------------------------+
4 rows in set (0.00 sec)

更新数据

 

UPDATE resty_user SET basic_info='{"age": "23","moblie": "13669361999", "email": "tinywan@aliyun.com", "id_card": "620000988221230198",  "from": "甘肃兰州", "name": "王玉玺"}',
skill_info='{"system": ["linux", "windows"],"database": ["mysql", "oracle", "postgresql"], "language": ["python", "java", "go"]}' where id = 120

 

json数据查询方式

在插入了json类型的数据之后，可以针对JSON类型做一些特定的查询，如查询年龄大于20的记录
在SQL的语句中使用 字段->.键名 就可以查询出所对应的键值

?
1
2
3
4
5
6
7
8
9
10
11
12
13
MySQL [test]>  select * from employee  WHERE basic_info->'$.age'> 20;
+-------+-------------------------------------------------------+-----------------------------------------------------------------------------------------------------------------------+
| empno | basic_info                                            | skill_info                                                                                                            |
+-------+-------------------------------------------------------+-----------------------------------------------------------------------------------------------------------------------+
|     1 | {"age": "23", "from": "hangzhou", "name": "wangyiyi"} | ["java", "go", "python"]                                                                                              |
|     2 | {"age": 24, "from": "shanghai", "name": "linxue"}     | ["mysql", "oracle", "python"]                                                                                         |
|     3 | {"age": 28, "from": "shanghai", "name": "zhaoqing"}   | {"system": "linux", "database": "mysql", "language": "go"}                                                            |
|     4 | {"age": 30, "from": "nanjing", "name": "zhouxixi"}    | {"system": ["linux", "windows"], "database": ["mysql", "oracle", "postgresql"], "language": ["python", "java", "go"]} |
+-------+-------------------------------------------------------+-----------------------------------------------------------------------------------------------------------------------+
4 rows in set (0.00 sec)
 
MySQL [test]> select * from employee  WHERE basic_info->'$.age'< 20;
Empty set (0.00 sec)

除了使用上述方式外，也可使用 提取json值的 函数 json_extract （使用函数的方式）

?
1
2
3
4
5
6
7
8
MySQL [test]> select * from employee where json_extract(basic_info,'$.age') = 24;
+-------+-----------------------------------------------------+----------------------------------------------------------------+
| empno | basic_info                                          | skill_info                                                     |
+-------+-----------------------------------------------------+----------------------------------------------------------------+
|     2 | {"age": 24, "from": "shanghai", "name": "linxue"}   | ["mysql", "oracle", "python"]                                  |
|     3 | {"age": 24, "from": "shanghai", "name": "zhaoqing"} | {"system": "linux", "database": "mysql", "language": "python"} |
+-------+-----------------------------------------------------+----------------------------------------------------------------+
2 rows in set (0.00 sec)

对于数值查询也可做一个范围内查询，如下：

复制代码
MySQL [test]> select * from employee  WHERE basic_info->'$.age' in (23, 24);
+-------+-------------------------------------------------------+----------------------------------------------------------------+
| empno | basic_info                                            | skill_info                                                     |
+-------+-------------------------------------------------------+----------------------------------------------------------------+
|     1 | {"age": "23", "from": "hangzhou", "name": "wangyiyi"} | ["java", "go", "python"]                                       |
|     2 | {"age": 24, "from": "shanghai", "name": "linxue"}     | ["mysql", "oracle", "python"]                                  |
|     3 | {"age": 24, "from": "shanghai", "name": "zhaoqing"}   | {"system": "linux", "database": "mysql", "language": "python"} |
+-------+-------------------------------------------------------+----------------------------------------------------------------+
3 rows in set, 1 warning (0.00 sec)
复制代码

因为 JSON 不同于字符串，所以如果用字符串和 JSON 字段比较，是不会相等的：

如下，直接使用字符串查询，查询不出来内容

?
1
2
MySQL [test]> select * from employee where basic_info = '{"age": 24, "from": "shanghai", "name": "linxue"}';
Empty set (0.00 sec)

可以通过 CAST 将字符串转成 JSON 的形式，如下：

复制代码
MySQL [test]> select * from employee where basic_info = CAST('{"age": 24, "from": "shanghai", "name": "linxue"}' AS JSON);
+-------+---------------------------------------------------+-------------------------------+
| empno | basic_info                                        | skill_info                    |
+-------+---------------------------------------------------+-------------------------------+
|     2 | {"age": 24, "from": "shanghai", "name": "linxue"} | ["mysql", "oracle", "python"] |
+-------+---------------------------------------------------+-------------------------------+
1 row in set (0.00 sec)
复制代码

查看单纯数组类型的函数JSON_CONTAINS

复制代码
MySQL [test]> select * from employee where JSON_CONTAINS (skill_info,'"mysql"');
+-------+---------------------------------------------------+-------------------------------+
| empno | basic_info                                        | skill_info                    |
+-------+---------------------------------------------------+-------------------------------+
|     2 | {"age": 24, "from": "shanghai", "name": "linxue"} | ["mysql", "oracle", "python"] |
+-------+---------------------------------------------------+-------------------------------+
1 row in set (0.00 sec)
复制代码

JSON_PRETTY函数： 以易于阅读的格式打印出JSON值
便于在一些外部应用引用数据时，更方便的使用它

复制代码
MySQL [test]> select JSON_PRETTY(basic_info) from employee;
+---------------------------------------------------------------+
| JSON_PRETTY(basic_info)                                       |
+---------------------------------------------------------------+
| {
  "age": "23",
  "from": "hangzhou",
  "name": "wangyiyi"
} |
| {
  "age": 24,
  "from": "shanghai",
  "name": "linxue"
}     |
| {
  "age": 28,
  "from": "shanghai",
  "name": "zhaoqing"
}   |
| {
  "age": 30,
  "from": "nanjing",
  "name": "zhouxixi"
}    |
+---------------------------------------------------------------+
4 rows in set (0.00 sec)
复制代码

MySQL 5.7.22中添加了此功能，此函数返回用于存储JSON文档的二进制表示的字节数，用于查看当前JSON字段的存储大小

复制代码
MySQL [test]> select skill_info,JSON_STORAGE_SIZE(skill_info) AS Size  from employee;
+-----------------------------------------------------------------------------------------------------------------------+------+
| skill_info                                                                                                            | Size |
+-----------------------------------------------------------------------------------------------------------------------+------+
| ["java", "go", "python"]                                                                                              |   29 |
| ["mysql", "oracle", "python"]                                                                                         |   34 |
| {"system": "linux", "database": "mysql", "language": "go"}                                                            |   63 |
| {"system": ["linux", "windows"], "database": ["mysql", "oracle", "postgresql"], "language": ["python", "java", "go"]} |  137 |
+-----------------------------------------------------------------------------------------------------------------------+------+
4 rows in set (0.00 sec)
复制代码

查询JSON字段的长度

复制代码
MySQL [test]> select JSON_LENGTH(basic_info) from employee;
+-------------------------+
| JSON_LENGTH(basic_info) |
+-------------------------+
|                       3 |
|                       3 |
|                       3 |
|                       3 |
+-------------------------+
4 rows in set (0.00 sec)
复制代码

查看数据的类型：可以是对象，数组或标量类型

复制代码
MySQL [test]> select JSON_TYPE(skill_info) from employee;
+-----------------------+
| JSON_TYPE(skill_info) |
+-----------------------+
| ARRAY                 |
| ARRAY                 |
| OBJECT                |
| OBJECT                |
+-----------------------+
4 rows in set (0.00 sec)
复制代码

json数据修改方式

如果是整个 json 更新的话，和一般类型插入是一样的

json_array_insert是在指定下标插入，这是插入一般数组类型时的操作

复制代码
MySQL [test]> SELECT json_array_insert(skill_info, '$[1]', 'php') from employee;
+-----------------------------------------------------------------------------------------------------------------------+
| json_array_insert(skill_info, '$[1]', 'php')                                                                          |
+-----------------------------------------------------------------------------------------------------------------------+
| ["java", "php", "go", "python"]                                                                                       |
| ["mysql", "php", "oracle", "python"]                                                                                  |
| {"system": "linux", "database": "mysql", "language": "python"}                                                        |
| {"system": ["linux", "windows"], "database": ["mysql", "oracle", "postgresql"], "language": ["python", "java", "go"]} |
+-----------------------------------------------------------------------------------------------------------------------+
4 rows in set (0.00 sec)
复制代码

替换操作，也就是修改update操作，使用的是 json_replace 函数
json_replace：只替换已经存在的旧值，不存在则忽略；

复制代码
MySQL [test]> update employee set  skill_info = json_replace(skill_info, "$.language", "go") where empno = 3;
Query OK, 1 row affected (0.00 sec)
Rows matched: 1  Changed: 1  Warnings: 0

MySQL [test]> select * from employee where empno = 3;
+-------+-----------------------------------------------------+------------------------------------------------------------+
| empno | basic_info                                          | skill_info                                                 |
+-------+-----------------------------------------------------+------------------------------------------------------------+
|     3 | {"age": 24, "from": "shanghai", "name": "zhaoqing"} | {"system": "linux", "database": "mysql", "language": "go"} |
+-------+-----------------------------------------------------+------------------------------------------------------------+
1 row in set (0.00 sec)
复制代码

json_set：替换旧值，并插入不存在的新值；

复制代码
MySQL [test]> update employee set  basic_info = json_set(basic_info, "$.age", 28,"$.sex" ,"man") where empno = 3;
Query OK, 1 row affected (0.01 sec)
Rows matched: 1  Changed: 1  Warnings: 0

MySQL [test]> 
MySQL [test]> select * from employee where empno = 3;
+-------+-------------------------------------------------------------------+------------------------------------------------------------+
| empno | basic_info                                                        | skill_info                                                 |
+-------+-------------------------------------------------------------------+------------------------------------------------------------+
|     3 | {"age": 28, "sex": "man", "from": "shanghai", "name": "zhaoqing"} | {"system": "linux", "database": "mysql", "language": "go"} |
+-------+-------------------------------------------------------------------+------------------------------------------------------------+
1 row in set (0.00 sec)
复制代码

json_insert：插入新值，但不替换已经存在的旧值；

复制代码
MySQL [test]> update  employee  set  basic_info = json_insert (basic_info, "$.age", 30, "$.phone" ,"123456789") where empno = 3;
Query OK, 1 row affected (0.01 sec)
Rows matched: 1  Changed: 1  Warnings: 0

MySQL [test]> select * from employee where empno = 3;
+-------+-----------------------------------------------------------------------------------------+------------------------------------------------------------+
| empno | basic_info                                                                              | skill_info                                                 |
+-------+-----------------------------------------------------------------------------------------+------------------------------------------------------------+
|     3 | {"age": 28, "sex": "man", "from": "shanghai", "name": "zhaoqing", "phone": "123456789"} | {"system": "linux", "database": "mysql", "language": "go"} |
+-------+-----------------------------------------------------------------------------------------+------------------------------------------------------------+
1 row in set (0.00 sec)
复制代码

json_remove() 删除元素函数。

?
1
2
3
4
5
6
7
8
9
10
11
12
MySQL [test]> update  employee  set  basic_info = json_remove (basic_info, "$.sex", "$.phone") where empno = 3;
Query OK, 1 row affected (0.01 sec)
Rows matched: 1  Changed: 1  Warnings: 0
 
MySQL [test]>
MySQL [test]> select * from employee where empno = 3;
+-------+-----------------------------------------------------+------------------------------------------------------------+
| empno | basic_info                                          | skill_info                                                 |
+-------+-----------------------------------------------------+------------------------------------------------------------+
|     3 | {"age": 28, "from": "shanghai", "name": "zhaoqing"} | {"system": "linux", "database": "mysql", "language": "go"} |
+-------+-----------------------------------------------------+------------------------------------------------------------+
1 row in set (0.00 sec)

结语

JSON数据类型是一个对开发十分友好的功能，有了它，MySQL的功能才更趋于完善。

posted @   Tinywan  阅读(2020)  评论(0)  编辑  收藏  举报
编辑推荐：
· 10年+ .NET Coder 心语，封装的思维：从隐藏、稳定开始理解其本质意义
· .NET Core 中如何实现缓存的预热？
· 从 HTTP 原因短语缺失研究 HTTP/2 和 HTTP/3 的设计差异
· AI与.NET技术实操系列：向量存储与相似性搜索在 .NET 中的实现
· 基于Microsoft.Extensions.AI核心库实现RAG应用
阅读排行：
· TypeScript + Deepseek 打造卜卦网站：技术与玄学的结合
· 阿里巴巴 QwQ-32B真的超越了 DeepSeek R-1吗？
· 【译】Visual Studio 中新的强大生产力特性
· 10年+ .NET Coder 心语 ── 封装的思维：从隐藏、稳定开始理解其本质意义
· 【设计模式】告别冗长if-else语句：使用策略模式优化代码结构
历史上的今天：
2017-03-04 设计模式之————依赖注入(Dependency Injection)与控制反转(Inversion of Controller)
点击右上角即可分享
微信分享提示
AI IDE Trae