Hi, I am using MySQL for a textual database in portuguese (brasilian).
When I use the FTS index, the search behavior is strange, and when I dump the index using the fts_dump, there are several words that repeat itself in the index. For my surprise, the performance of the FTS for boolean search was too low for the application that I am constructing, and I decided to implement a full FTS engine using MySQL MyIsam binary search capacibilities. When I start to construct the dictionary table, and looked inside, I had the same surprise, the behavior is still erratic. This behavior occoured in the 3.23 and 4.1 (both last versions tested - March 6th). Let me exemplify: There are a simple dictionary table, that have unique words associated with a numerical index, like: TABLE dictionary ( id int primary_key autoincrement not null, word char(50) ) First I inserted non equal words, like: SÃO, SAO and SAÕ, that returns discret ids, like: +--------+-------+ | id | word | +--------+-------+ | 76 | são | | 116223 | sao | | 222943 | saõ | +--------+-------+ But when I do a select for specifcly word like sao, I get all the three !!! Internally MySQL isn't distinguing the different words, assuming that à = A, and O = Õ. I read the manual, and I have a clue that this is a CHARACTER SET, COLLATE behavior, is this right ? I am using the default CHARACTER SET and COLLATION. Thanks Slepetys -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
