Module org.apache.lucene.core

Package org.apache.lucene.search

Class BlockMaxConjunctionBulkScorer

java.lang.Object

org.apache.lucene.search.BulkScorer

org.apache.lucene.search.BlockMaxConjunctionBulkScorer

final class BlockMaxConjunctionBulkScorer
extends BulkScorer

BulkScorer implementation of BlockMaxConjunctionScorer that focuses on top-level conjunctions over clauses that do not have two-phase iterators. Use a Weight.DefaultBulkScorer around a BlockMaxConjunctionScorer if you need two-phase support. Another difference with BlockMaxConjunctionScorer is that this scorer computes scores on the fly in order to be able to skip evaluating more clauses if the total score would be under the minimum competitive score anyway. This generally works well because computing a score is cheaper than decoding a block of postings.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
private static class	BlockMaxConjunctionBulkScorer.DocAndScore

Field Summary

Fields

Modifier and Type	Field	Description
private final DocIdSetIterator[]	iterators
private final DocIdSetIterator	lead1
private final DocIdSetIterator	lead2
private final int	maxDoc
private final BlockMaxConjunctionBulkScorer.DocAndScore	scorable
private final Scorer	scorer1
private final Scorer	scorer2
private final Scorer[]	scorers
private final double[]	sumOfOtherClauses

Constructor Summary

Constructors

Constructor	Description
BlockMaxConjunctionBulkScorer(int maxDoc, List<Scorer> scorers)

Method Summary

Modifier and Type	Method	Description
private float	computeMaxScore(int windowMin, int windowMax)
long	cost()	Same as DocIdSetIterator.cost() for bulk scorers.
int	score(LeafCollector collector, Bits acceptDocs, int min, int max)	Collects matching documents in a range and return an estimation of the next matching document which is on or after max.
private void	scoreWindow(LeafCollector collector, Bits acceptDocs, int min, int max, float maxWindowScore)

Methods inherited from class org.apache.lucene.search.BulkScorer

score

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

scorers

private final Scorer[] scorers

iterators

private final DocIdSetIterator[] iterators

lead1

private final DocIdSetIterator lead1

lead2

private final DocIdSetIterator lead2

scorer1

private final Scorer scorer1

scorer2

private final Scorer scorer2

scorable

private final BlockMaxConjunctionBulkScorer.DocAndScore scorable

sumOfOtherClauses

private final double[] sumOfOtherClauses

maxDoc

private final int maxDoc

Constructor Details

BlockMaxConjunctionBulkScorer

BlockMaxConjunctionBulkScorer(int maxDoc,
 List<Scorer> scorers)
                       throws IOException

Throws:

IOException

Method Details

computeMaxScore

private float computeMaxScore(int windowMin,
 int windowMax)
                       throws IOException

Throws:

IOException

score

public int score(LeafCollector collector,
 Bits acceptDocs,
 int min,
 int max)
          throws IOException

Description copied from class: BulkScorer

Collects matching documents in a range and return an estimation of the next matching document which is on or after max.

The return value must be:

• >= max,
• DocIdSetIterator.NO_MORE_DOCS if there are no more matches,
• <= the first matching document that is >= max otherwise.

min is the minimum document to be considered for matching. All documents strictly before this value must be ignored.

Although max would be a legal return value for this method, higher values might help callers skip more efficiently over non-matching portions of the docID space.

For instance, a Scorer-based implementation could look like below:

 private final Scorer scorer; // set via constructor

 public int score(LeafCollector collector, Bits acceptDocs, int min, int max) throws IOException {
   collector.setScorer(scorer);
   int doc = scorer.docID();
   if (doc < min) {
     doc = scorer.advance(min);
   }
   while (doc < max) {
     if (acceptDocs == null || acceptDocs.get(doc)) {
       collector.collect(doc);
     }
     doc = scorer.nextDoc();
   }
   return doc;
 }
 

Specified by:

score in class BulkScorer

Parameters:

collector - The collector to which all matching documents are passed.

acceptDocs - Bits that represents the allowed documents to match, or null if they are all allowed to match.

min - Score starting at, including, this document

max - Score up to, but not including, this doc

Returns:

an under-estimation of the next matching doc after max

Throws:

IOException

scoreWindow

private void scoreWindow(LeafCollector collector,
 Bits acceptDocs,
 int min,
 int max,
 float maxWindowScore)
                  throws IOException

Throws:

IOException

cost

public long cost()

Description copied from class: BulkScorer

Same as DocIdSetIterator.cost() for bulk scorers.

Specified by:

cost in class BulkScorer