I have a table that stores strings in utf8, using the default collation utf8_genera_ci. During comparison using LIKE, MySQL 4.1.1 seems to be able to handle case insensitivity of latin characters correctly but not other characters such as Greek. For example, I have a varchar column (mycol) with data:
a (that is "a <not equal to sign> <greek small letter alpha>", 3 characters) if I do select * from mytable where mycol LIKE '%A%' (giving it latin capital A) will return that row but select * from mytable where mycol LIKE '%%' (giving it greek capital letter alpha) doesn't return that row and doing select * from mytable where my col LIKE '%%' (giving it greek small letter alpha) also returns that row. So it looks like case insensitivity only works with the latin part of the unicode set but doesn't work with others such as greek. Is that the expected behavior? If so, that is really undesirable, making it impossible to truly support multiple character sets at once. Thanks, Des Note: I did do SET CHARACTER SET utf8 for the client before sending strings over. --------------------------------- Do you Yahoo!? Yahoo! Hotjobs: Enter the "Signing Bonus" Sweepstakes
