|| at Mar 27, 2007 at 7:46 am
27 mar 2007 kl. 09.33 skrev SK R:
I'm speaking about term positions.
In Lucene File Format, the .prx file contains the lists of
that each term occurs at within documents.
I asked what's the purpose of this .prx file?
It is generally used for phrase/span queries.
Also how phrasequery handled?
If the matching terms are within slop distance, the query is a match.
Or what do you ask here?
Slop and distance can affect the score. For exact meachanism see
PhraseScorerer.java, SloppyPhraseScorer.java or await a better answer
from someone else :)