MySql字符串编码问题

mysql编码设置
查看系统的字符集和排序方式的设定可以通过下面的两条命令或mysql> status 。 


mysql> SHOW VARIABLES LIKE 'character_set_%'; 
+--------------------------+----------------------------+ 
| Variable_name | Value | 
+--------------------------+----------------------------+ 
| character_set_client | latin1 | 
| character_set_connection | latin1 | 
| character_set_database | latin1 | 
| character_set_results | latin1 | 
| character_set_server | latin1 | 
| character_set_system | utf8 | 
| character_sets_dir | /usr/share/mysql/charsets/ | 
+--------------------------+----------------------------+ 
7 rows in set (0.00 sec) 


mysql> SHOW VARIABLES LIKE 'collation_%'; 
+----------------------+-------------------+ 
| Variable_name | Value | 
+----------------------+-------------------+ 
| collation_connection | latin1_swedish_ci | 
| collation_database | latin1_swedish_ci | 
| collation_server | latin1_swedish_ci | 
+----------------------+-------------------+ 
3 rows in set (0.00 sec) 


上面列出的值就是系统的默认值。(很奇怪系统怎么默认是latin1的瑞典语排序方式)... 


当我们按照原来的方式通过PHP存取MySQL数据库时，就算设置了表的默认字符集为utf8并且通过UTF-8编码发送查询，你会发现存入数据库的仍然是乱码。问题就出在这个connection连接层上。解决方法是在发送查询前执行一下下面这句： 


1.  SET NAMES 'utf8'; 


它相当于下面的三句指令： 
SET character_set_client = utf8; 
SET character_set_results = utf8; 
SET character_set_connection = utf8; 


2. 创建数据库 
mysql> create database name character set utf8; 


3. 创建表 
CREATE TABLE `type` ( 
`id` int(10) unsigned NOT NULL auto_increment, 
`flag_deleted` enum('Y','N') character set utf8 NOT NULL default 'N', 
`flag_type` int(5) NOT NULL default '0', 
`type_name` varchar(50) character set utf8 NOT NULL default '', 
PRIMARY KEY (`id`) 
)  DEFAULT CHARSET=utf8; 


4. 修改数据库成utf8的. 
mysql> alter database name character set utf8; 


5. 修改表默认用utf8. 
mysql> alter table type character set utf8; 


6. 修改字段用utf8 
mysql> alter table type modify type_name varchar(50) CHARACTER SET utf8;