This is some scala code to help deducing a language from text.
It is simple, it (with unicode standard) find the alphabet in which the text is write.
Then (if necessary, some language are written in a specific alphabet) it match a number of words to your database.
And find the most language name returned.
This could be irrelevant if you use a few number of words.

** setting up the database
see Mysql in DB.scala

** adding dictionary
see the function  DB.init
see the format of file in dict

** the number of words match
see language.maxWordsToTest
see language.textSizeLimit
see language.phraseSizeLimit