This email sounds right. substCount variable always increases and never gets reset to zero and it seems that it should be reset before every substitution, so that its value reflects the number of characters substituted in each token.
I will commit the fix now. Gerhard, please correct me if I'm wrong. Thanks, Otis --- Bernhard Messer <[EMAIL PROTECTED]> wrote: > Hi all, > > I am using the lucene German Stemmer/Analyzer. There seems to be a > bug > within the GermanStemmer class. As far as i understand the algorithm > the > count variable "substCount" > should be set to 0 before processing the next token. > In the current implementation, the stemmed result will differ for the > same > terms after a while. > The easiest solution would be to reset that counter variable in the > method: > "private StringBuffer substitute( StringBuffer buffer )" . > > best regards > Bernhard > > > -- > To unsubscribe, e-mail: > <mailto:[EMAIL PROTECTED]> > For additional commands, e-mail: > <mailto:[EMAIL PROTECTED]> > __________________________________________________ Do You Yahoo!? Send FREE Valentine eCards with Yahoo! Greetings! http://greetings.yahoo.com -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
