adaptions for german language and utf-8

I did some changes to adapt the script to utf-8 encoding and german language:

1. insert

mb_internal_encoding("UTF-8");

class SpellCorrector {


2. substitute

preg_match_all("/[a-zäöüß]+/",mb_strtolower($text),$matches);


3. substitute

$alphabet = 'abcdefghijklmnopqrstuvwxyzäöüß';


4.

change some string-functions to mb_ substitution:

strlen -> mb_strlen
strtolower -> mb_strtolower
substr -> mb_substr

that works well