Roy schrieb:
While studying the search code, I got a question: in BooleanScorer,
when collecting docids and scores from subscorers, the docids are
hashed through buckettable first. What's the purpose of doing the
hashing?
The purpose of buckettable is to gather all scores from the subscorers
for a certain interval of documents. Only after evaluating all
subscorers on the current interval of documents we have enough info
to decide whether a document is a hit or not.
With the current scorer API one could get rid of buckettable and
advance all subscores only by one document each time. I am not sure
whether the bucketable implementation is really much more efficient.
I only see the advantage of inlining some of the scorer.next and
score.score code.
Christoph
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
