Hello Lucene experts, as you might have seen in my previous postings, I am bound to use not more than Lucene 1.2 (due to hardware limitations I can only use Java 1.1 or 1.2).
I would like to do my own Similarity implementation which, I think, would allow me to insert other algorithms in Lucene which would enable me to compare different IR algoritms with Lucene as a framework. I got advice from here to use DefaultSimiliarity. However, this class was, accoding to the JavaDoc and the CHANGE text file, only introduced in Lucene 1.3. In Lucene 1.2 it seems to be all hard and solid coded without the intension to have flexiblility at this point. Now some questions: 1) Am I wrong here and does somebody know a strategy which could allow me to have an enhancement that could allow me to add my own IR similarity algoritm? 2) Does somebody know how to hack the Lucene code in order to make it happen? I presume some code changes could do that: Either changes in the Similarity class directly or a change of Similarity to non-final (plus setters to set new similarities like in Lucene 1.3). Any other suggestions? I generally like the small-ness of Lucene 1.2 but it is just that I would like to have a little bit more flexiblity at one point... Karl -- Lust, ein paar Euro nebenbei zu verdienen? Ohne Kosten, ohne Risiko! Satte Provisionen für GMX Partner: http://www.gmx.net/de/go/partner --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
