REGEX in MySQL to find Japanese matches

I need to find all entries in a table that consist only out of certain Japanese UTF-8 characters.

For example, I want all fields that consist only out of 1 (一) and 2 (二).

I am using

SELECT combi_id, keb FROM combi WHERE keb REGEXP '[二一]+'

but it matches lots of other fields that contain different characters, what am I doing wrong?

This is the table:

CREATE TABLE IF NOT EXISTS `combi` (
      `combi_id` int(11) NOT NULL auto_increment,
      `ent_seq` int(11) NOT NULL,
      `reb` text NOT NULL,
      `keb` text NOT NULL,
      `ant` text NOT NULL,
      `ke_pri` text NOT NULL,
      `re_pri` text NOT NULL,
      `re_restr` text NOT NULL,
      `stagr` text NOT NULL,
      `s_inf` text NOT NULL,
      `lsource` text NOT NULL,
      `gloss` text NOT NULL,
      `xref` text NOT NULL,
      `stagk` text NOT NULL,
      PRIMARY KEY  (`combi_id`)
    ) ENGINE=MyISAM  DEFAULT CHARSET=utf8 AUTO_INCREMENT=146740 ;

and this is a sample line of data:

(22, 1000225, 'あからさま', '明白|偸閑|白地', '', '', '', '', '', '', '', 'plain|frank|candid|open|direct|straightforward|unabashed|blatant|flagrant', '开发者_如何学C', ''),

thanks a lot for the help!

If you want to match columns with only those characters, you should use

SELECT combi_id, keb FROM combi WHERE keb REGEXP '^[二一]+$'

Note the ^ in the beginning, and the $ in the end, meaning respectively "start of string" and "end of string". Without those, the regexp could match at any position.

Edit: testing it

mysql> select * from test;
+--------+
| f1     |
+--------+
| 二     |
| 東京   |
| 人     |
| 丸     |
+--------+
4 rows in set (0.00 sec)

mysql> select * from test where f1 regexp _utf8'[一二]';
+--------+
| f1     |
+--------+
| 二     |
| 東京   |
| 人     |
| 丸     |
+--------+
4 rows in set (0.00 sec)

Wow, indeed, character classes in mysql regexps sound seriously broken... This works, though:

mysql> select * from test where f1 regexp _utf8'(一|二)';
+------+
| f1   |
+------+
| 二   |
+------+
1 row in set (0.00 sec)