Why is MySQL treating é the same as e? - mysql

Question

I'm storing Unicode strings in a MySQL database with a Django web application. I can store Unicode data fine, but when querying, I find that é and e are treated as if they were the same character:
In [1]: User.objects.filter(last_name='Jildén')
Out[1]: [<User: Anders Jildén>]
In [2]: User.objects.filter(last_name='Jilden')
Out[2]: [<User: Anders Jildén>]
This is also the case when using the MySQL shell directly:
mysql> select last_name from auth_user where last_name = 'Jildén';
+-----------+
| last_name |
+-----------+
| Jildén |
+-----------+
1 row in set (0.00 sec)
mysql> select last_name from auth_user where last_name = 'Jilden';
+-----------+
| last_name |
+-----------+
| Jildén |
+-----------+
1 row in set (0.01 sec)
Here are the database charset settings:
mysql> SHOW variables LIKE '%character_set%';
+--------------------------+------------------------------------------------------+
| Variable_name | Value |
+--------------------------+------------------------------------------------------+
| character_set_client | latin1 |
| character_set_connection | latin1 |
| character_set_database | utf8 |
| character_set_filesystem | binary |
| character_set_results | latin1 |
| character_set_server | latin1 |
| character_set_system | utf8 |
| character_sets_dir | /usr/local/Cellar/mysql/5.1.54/share/mysql/charsets/ |
+--------------------------+------------------------------------------------------+
here's the table schema:
CREATE TABLE `auth_user` (
`id` int(11) NOT NULL AUTO_INCREMENT,
`username` varchar(30) CHARACTER SET utf8 NOT NULL,
`first_name` varchar(30) CHARACTER SET utf8 NOT NULL,
`last_name` varchar(30) CHARACTER SET utf8 NOT NULL,
`email` varchar(200) CHARACTER SET utf8 NOT NULL,
`password` varchar(128) CHARACTER SET utf8 NOT NULL,
`is_staff` tinyint(1) NOT NULL,
`is_active` tinyint(1) NOT NULL,
`is_superuser` tinyint(1) NOT NULL,
`last_login` datetime NOT NULL,
`date_joined` datetime NOT NULL,
PRIMARY KEY (`id`),
UNIQUE KEY `username` (`username`)
) ENGINE=InnoDB AUTO_INCREMENT=7952 DEFAULT CHARSET=utf8 COLLATE=utf8_bin
and here are the options I'm passing via Django's DATABASES setting:
DATABASES = {
'default': {
# ...
'OPTIONS': {
'charset': 'utf8',
'init_command': 'SET storage_engine=INNODB;',
},
},
}
Note that I have tried setting the table collation to utf8_bin, with no effect:
mysql> alter table auth_user collate utf8_bin;
mysql> select last_name from auth_user where last_name = 'Jilden';
+-----------+
| last_name |
+-----------+
| Jildén |
+-----------+
1 row in set (0.00 sec)
How can I get MySQL to treat these as different characters?

Answer 1

You were nearly there when you changed the table collation, but not quite. In MySQL, each column in a table has its own character set and collation. The table has its own character set and collation, but this does not override the column collations; it only determines what the collation will be for new columns that are added for which you don't specify the collation. So you haven't changed the collation of the column that you're interested in.
ALTER TABLE tablename MODIFY columnname
varchar(???) CHARACTER SET utf8 COLLATE utf8_bin NOT NULL

Answer 2

You need to set a collation that treats diacritics as significant. Try using utf8_bin

Answer 3

I think it would be important to know the charset of the table and the field you are querying.
The answer to your question could be found here
http://dev.mysql.com/doc/refman/5.0/en/charset-unicode-sets.html
Maybe the field you are querying has the utf8_general_ci charset.
To obtain what you want you should set the charset of that field as utf8_unicode_ci
Remember that, as the manual says, queries on utf8_unicode_ci charset fields are slower than queries on utf8_general_ci fields