The best way to match this would be a phonetic match using the soundex module. The soundex module determines the sound of a word, and then whether it is eqivalent to another word, in phonetic sound. So looking at viagra and comparing it to V��gr= a
It determines that the hyphenated i and hyphenated a sound similar to the i and a in viagra. and assigns no sound to the = and therefore determines the sound matches. The steps would be: determine equivalent sounds for letters. determine letters that have no sound such as . or = or similar. This is not currently implemented but would appear to be the best way to stop spam mispellings of spammy words. Just determine the phonetic sound of the word and then the phonetic sound of common mispellings of the word, and if they match, then you have a rule. Eduardo Bejar said: > Hi, > > Which rule works for catching this kind of words? > > I = just found something I thought you might be interested in. V�g�l= is > the > top�cal f�mal� equivalent to V��gr= a. It is the r�volut�onary new pr�d�ct > for w&oac= ute;m�n that has been f��tured in all�r�, = c�smo, gl�mour, > m�xim > and shape m�gaz�ne= .A= ll ord�rs back�d by our, > n� ri�k, mon�y ba�k G�ar�nt�e= ! > > Thanks, > > Edo > -- Luke Computer Science System Administrator Security Administrator,College of Engineering Montana State University-Bozeman,Montana
