您的位置: 首页  >  IT文章  >  MySQL修改有存量数据的数据库字符集实战

MySQL修改有存量数据的数据库字符集实战

分类: IT文章 • 2024-03-30 18:03:05

原文链接：https://www.modb.pro/db/22722?cyn

我们已经知道数据库或者表，可以在创建后再修改字符集。但是修改字符集不会将已有的数据按新字符集重新进行处理。那么已有存量数据的数据库改如何调整字符集呢？可以使用命令进行转换，也可以像早期先将数据导出，调整字符集，再将数据导入。

我们假设有个业务，在英语国家诞生，当时其研发团队创建一个latin1字符集的数据库进行支撑。

root@database-one 13:25:  [(none)]> create database DiscountStore default charset latin1;
Query OK, 1 row affected (0.01 sec)

root@database-one 13:33:  [(none)]> use discountstore;
Database changed
root@database-one 13:33:  [discountstore]> create table orders (no int,Buyer varchar(30),Seller varchar(30),InstallationDate datetime) engine=innodb charset latin1;
Query OK, 0 rows affected (0.02 sec)

root@database-one 13:41:  [discountstore]> insert into orders values(666,'Steve','Tom',now()+2);
Query OK, 1 row affected (0.00 sec)

root@database-one 13:41:  [discountstore]> insert into orders values(777,'Jeff','Bill',now()+3);
Query OK, 1 row affected (0.00 sec)

root@database-one 13:42:  [discountstore]> select * from orders;
+------+-------+--------+---------------------+
| no   | Buyer | Seller | InstallationDate    |
+------+-------+--------+---------------------+
|  666 | Steve | Tom    | 2020-03-18 13:41:26 |
|  777 | Jeff  | Bill   | 2020-03-18 13:42:28 |
+------+-------+--------+---------------------+
2 rows in set (0.00 sec)

业务开展的很好，现在要拓展到中国，系统需要能够支持中文，同时还得考虑继续扩展到其它国家的可能，所以研发团队选utf8做为数据库新的字符集，因为utf8兼容latin1，所以只需修改数据库、表的默认字符集，转换列的字符集即可，不需要其它额外处理。

root@database-one 14:15:  [discountstore]> alter database discountstore character set utf8;
Query OK, 1 row affected (0.00 sec)

root@database-one 14:16:  [discountstore]> show create database discountstore G
*************************** 1. row ***************************
       Database: discountstore
Create Database: CREATE DATABASE `discountstore` /*!40100 DEFAULT CHARACTER SET utf8 */
1 row in set (0.00 sec)

root@database-one 14:16:  [discountstore]> alter table orders character set utf8;
Query OK, 0 rows affected (0.00 sec)
Records: 0  Duplicates: 0  Warnings: 0

root@database-one 14:20:  [discountstore]> show create table orders G
*************************** 1. row ***************************
       Table: orders
Create Table: CREATE TABLE `orders` (
  `no` int(11) DEFAULT NULL,
  `Buyer` varchar(30) CHARACTER SET latin1 DEFAULT NULL,
  `Seller` varchar(30) CHARACTER SET latin1 DEFAULT NULL,
  `InstallationDate` datetime DEFAULT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8
1 row in set (0.00 sec)

可以看到，库和表的字符集都变过来了，但是列的字符集还是latin1，对其也进行转换。

root@database-one 14:25:  [discountstore]> alter table orders convert to character set utf8;
Query OK, 2 rows affected (0.04 sec)
Records: 2  Duplicates: 0  Warnings: 0

root@database-one 14:28:  [discountstore]> show create table orders G
*************************** 1. row ***************************
       Table: orders
Create Table: CREATE TABLE `orders` (
  `no` int(11) DEFAULT NULL,
  `Buyer` varchar(30) DEFAULT NULL,
  `Seller` varchar(30) DEFAULT NULL,
  `InstallationDate` datetime DEFAULT NULL
) ENGINE=InnoDB DEFAULT CHARSET=utf8
1 row in set (0.00 sec)

root@database-one 14:32:  [discountstore]> select * from orders;
+------+-------+--------+---------------------+
| no   | Buyer | Seller | InstallationDate    |
+------+-------+--------+---------------------+
|  666 | Steve | Tom    | 2020-03-18 13:41:26 |
|  777 | Jeff  | Bill   | 2020-03-18 13:42:28 |
+------+-------+--------+---------------------+
2 rows in set (0.00 sec)

root@database-one 14:32:  [discountstore]> insert into orders values(888,'肖杰','郭伟',now()+4);
Query OK, 1 row affected (0.00 sec)

root@database-one 14:32:  [discountstore]> select * from orders;
+------+--------+--------+---------------------+
| no   | Buyer  | Seller | InstallationDate    |
+------+--------+--------+---------------------+
|  666 | Steve  | Tom    | 2020-03-18 13:41:26 |
|  777 | Jeff   | Bill   | 2020-03-18 13:42:28 |
|  888 | 肖杰   | 郭伟   | 2020-03-18 14:32:40 |
+------+--------+--------+---------------------+
3 rows in set (0.00 sec)

可以看到，转换完毕后，原数据正常，新的中文数据也可以存入了。

在上面的例子中，字符集从latin1转换为utf8，目标字符集是原字符集的超集，完全兼容，所以才能用命令转换。如果反过来，业务收缩，要从utf8将回latin1，就要先进行数据清洗，然后再修改和转换字符集，否则会因存量数据导致转换失败。

root@database-one 14:43:  [discountstore]> alter database discountstore character set latin1;
Query OK, 1 row affected (0.00 sec)

root@database-one 14:43:  [discountstore]> show create database discountstore G
*************************** 1. row ***************************
       Database: discountstore
Create Database: CREATE DATABASE `discountstore` /*!40100 DEFAULT CHARACTER SET latin1 */
1 row in set (0.00 sec)

root@database-one 14:44:  [discountstore]> alter table orders character set latin1;
Query OK, 0 rows affected (0.01 sec)
Records: 0  Duplicates: 0  Warnings: 0

root@database-one 14:44:  [discountstore]> show create table orders G
*************************** 1. row ***************************
       Table: orders
Create Table: CREATE TABLE `orders` (
  `no` int(11) DEFAULT NULL,
  `Buyer` varchar(30) CHARACTER SET utf8 DEFAULT NULL,
  `Seller` varchar(30) CHARACTER SET utf8 DEFAULT NULL,
  `InstallationDate` datetime DEFAULT NULL
) ENGINE=InnoDB DEFAULT CHARSET=latin1
1 row in set (0.00 sec)

root@database-one 14:44:  [discountstore]> select * from orders;
+------+--------+--------+---------------------+
| no   | Buyer  | Seller | InstallationDate    |
+------+--------+--------+---------------------+
|  666 | Steve  | Tom    | 2020-03-18 13:41:26 |
|  777 | Jeff   | Bill   | 2020-03-18 13:42:28 |
|  888 | 肖杰   | 郭伟   | 2020-03-18 14:32:40 |
+------+--------+--------+---------------------+
3 rows in set (0.00 sec)

root@database-one 14:46:  [discountstore]> alter table orders convert to character set latin1;
ERROR 1366 (HY000): Incorrect string value: 'xE8x82x96xE6x9DxB0' for column 'Buyer' at row 3
root@database-one 14:46:  [discountstore]> delete from orders where no=888;
Query OK, 1 row affected (0.00 sec)

root@database-one 14:47:  [discountstore]> alter table orders convert to character set latin1;
Query OK, 2 rows affected (0.24 sec)
Records: 2  Duplicates: 0  Warnings: 0

root@database-one 14:47:  [discountstore]> select * from orders;
+------+-------+--------+---------------------+
| no   | Buyer | Seller | InstallationDate    |
+------+-------+--------+---------------------+
|  666 | Steve | Tom    | 2020-03-18 13:41:26 |
|  777 | Jeff  | Bill   | 2020-03-18 13:42:28 |
+------+-------+--------+---------------------+
2 rows in set (0.00 sec)

root@database-one 14:47:  [discountstore]> show create table orders G
*************************** 1. row ***************************
       Table: orders
Create Table: CREATE TABLE `orders` (
  `no` int(11) DEFAULT NULL,
  `Buyer` varchar(30) DEFAULT NULL,
  `Seller` varchar(30) DEFAULT NULL,
  `InstallationDate` datetime DEFAULT NULL
) ENGINE=InnoDB DEFAULT CHARSET=latin1
1 row in set (0.00 sec)

先将数据导出，调整字符集，再将数据导入的方法我们也测试一下。

（可跳转至https://www.modb.pro/db/22722?cyn查看）

特别注意：
选择目标字符集的时候，最好是源字符集的超集，否则目标字符集中不支持的字符会变成乱码。

相关推荐