In general, a chinese keyword contain 2-7 chinese character, mostly 2-4. The auto-generated keyword is useless, because it can not find right separator of chinese words.
It's not the issure of version 3.0, i haved tested v1.4e, v2.0d, they have the same problem.
Bye the way, i am a programmer, i think to separate chinese text to keywords is not easy, because there is no simple rule to do this, it need a database include all chinese words. But the full text search is simple, please make sure the searched text and the phrase have the same encoding, they can be compared using 8-byte char.
Attached file contain one note 'note1' and one saved search, note1 include some chinese text and in it's Item Note there are some description, 'saved search' item should find something but in fact not(before search, please remove the Item Note of note1 or remove all auto-generated keywords).
Last edited by slangmgh; 03-29-2007 at 11:42 PM.
|