I have tried utilizing the full text searh feature and works as charm, except for one think which I would like to suggest to improve.
When a document is split into tokens the parser removes any whitespace or punctuation charactures not otherwise recognized. But some punctuation charactures such as a dot, slash and hyphen may be used to denote abbreviations and removing them from a token may result in changing the meaning of a word.
Here are some examples from the Russian language.
1. Parsing "р-н" returns two separate meaningless charactures but the combination denotes "район" (a district).
2. Parsing "с/с" returns two identical charactures which are later recognized as stop words but the combination donotes "сельсовет" (a subdistrict)
3. Parsing "o." returns a single characture which is later recognized as a stop word but the conbination denotes "остров" (an island).
It would be nice to have a possibility to assign a dictionary to the parser to recognize such cases.