Apache
/
lucene
mirror of https://github.com/apache/lucene.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

reformat spacing 

git-svn-id: https://svn.apache.org/repos/asf/lucene/java/trunk@150016 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Erik Hatcher 2003-09-11 01:25:47 +00:00
parent b058f63c30
commit 798fc0f0ef
2 changed files with 111 additions and 110 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

140
src/java/org/apache/lucene/search/PhraseScorer.java
View File

@ -60,87 +60,89 @@ import org.apache.lucene.util.*;
import org.apache.lucene.index.*; import org.apache.lucene.index.*;
abstract class PhraseScorer extends Scorer { abstract class PhraseScorer extends Scorer {
private Weight weight; private Weight weight;
protected byte[] norms; protected byte[] norms;
protected float value; protected float value;
protected PhraseQueue pq; protected PhraseQueue pq;
protected PhrasePositions first, last; protected PhrasePositions first, last;
private float freq; private float freq;
PhraseScorer(Weight weight, TermPositions[] tps, Similarity similarity, PhraseScorer(Weight weight, TermPositions[] tps, Similarity similarity,
byte[] norms) throws IOException { byte[] norms) throws IOException {
super(similarity); super(similarity);
this.norms = norms; this.norms = norms;
this.weight = weight; this.weight = weight;
this.value = weight.getValue(); this.value = weight.getValue();
// use PQ to build a sorted list of PhrasePositions // use PQ to build a sorted list of PhrasePositions
pq = new PhraseQueue(tps.length); pq = new PhraseQueue(tps.length);
for (int i = 0; i < tps.length; i++) for (int i = 0; i < tps.length; i++) {
pq.put(new PhrasePositions(tps[i], i)); pq.put(new PhrasePositions(tps[i], i));
pqToList(); }
} pqToList();
public final void score(HitCollector results, int end) throws IOException {
Similarity similarity = getSimilarity();
while (last.doc < end) { // find doc w/ all the terms
while (first.doc < last.doc) { // scan forward in first
do {
first.next();
} while (first.doc < last.doc);
firstToLast();
if (last.doc >= end)
return;
}
// found doc with all terms
freq = phraseFreq(); // check for phrase
if (freq > 0.0) {
float score = similarity.tf(freq)*value; // compute score
score *= Similarity.decodeNorm(norms[first.doc]); // normalize
results.collect(first.doc, score); // add to results
}
last.next(); // resume scanning
} }
}
protected abstract float phraseFreq() throws IOException; public final void score(HitCollector results, int end) throws IOException {
Similarity similarity = getSimilarity();
while (last.doc < end) { // find doc w/ all the terms
while (first.doc < last.doc) { // scan forward in first
do {
first.next();
} while (first.doc < last.doc);
firstToLast();
if (last.doc >= end)
return;
}
protected final void pqToList() { // found doc with all terms
last = first = null; freq = phraseFreq(); // check for phrase
while (pq.top() != null) {
PhrasePositions pp = (PhrasePositions)pq.pop(); if (freq > 0.0) {
if (last != null) { // add next to end of list float score = similarity.tf(freq) * value; // compute score
last.next = pp; score *= Similarity.decodeNorm(norms[first.doc]); // normalize
} else results.collect(first.doc, score); // add to results
first = pp; }
last = pp; last.next(); // resume scanning
pp.next = null; }
} }
}
protected final void firstToLast() { protected abstract float phraseFreq() throws IOException;
last.next = first; // move first to end of list
last = first;
first = first.next;
last.next = null;
}
public Explanation explain(final int doc) throws IOException { protected final void pqToList() {
Explanation tfExplanation = new Explanation(); last = first = null;
while (pq.top() != null) {
PhrasePositions pp = (PhrasePositions) pq.pop();
if (last != null) { // add next to end of list
last.next = pp;
} else
first = pp;
last = pp;
pp.next = null;
}
}
score(new HitCollector() { protected final void firstToLast() {
public final void collect(int d, float score) {} last.next = first; // move first to end of list
}, doc+1); last = first;
first = first.next;
last.next = null;
}
float phraseFreq = (first.doc == doc) ? freq : 0.0f; public Explanation explain(final int doc) throws IOException {
tfExplanation.setValue(getSimilarity().tf(phraseFreq)); Explanation tfExplanation = new Explanation();
tfExplanation.setDescription("tf(phraseFreq=" + phraseFreq + ")");
return tfExplanation; score(new HitCollector() {
} public final void collect(int d, float score) {
}
}, doc + 1);
float phraseFreq = (first.doc == doc) ? freq : 0.0f;
tfExplanation.setValue(getSimilarity().tf(phraseFreq));
tfExplanation.setDescription("tf(phraseFreq=" + phraseFreq + ")");
return tfExplanation;
}
} }

81
src/java/org/apache/lucene/search/SloppyPhraseScorer.java
View File

@ -54,53 +54,52 @@ package org.apache.lucene.search;
* <http://www.apache.org/>. * <http://www.apache.org/>.
*/ */
import org.apache.lucene.index.TermPositions;
import java.io.IOException; import java.io.IOException;
import org.apache.lucene.util.*;
import org.apache.lucene.index.*;
final class SloppyPhraseScorer extends PhraseScorer { final class SloppyPhraseScorer extends PhraseScorer {
private int slop; private int slop;
SloppyPhraseScorer(Weight weight, TermPositions[] tps, Similarity similarity, SloppyPhraseScorer(Weight weight, TermPositions[] tps, Similarity similarity,
int slop, byte[] norms) throws IOException { int slop, byte[] norms) throws IOException {
super(weight, tps, similarity, norms); super(weight, tps, similarity, norms);
this.slop = slop; this.slop = slop;
}
protected final float phraseFreq() throws IOException {
pq.clear();
int end = 0;
for (PhrasePositions pp = first; pp != null; pp = pp.next) {
pp.firstPosition();
if (pp.position > end)
end = pp.position;
pq.put(pp); // build pq from list
} }
float freq = 0.0f; protected final float phraseFreq() throws IOException {
boolean done = false; pq.clear();
do { int end = 0;
PhrasePositions pp = (PhrasePositions)pq.pop(); for (PhrasePositions pp = first; pp != null; pp = pp.next) {
int start = pp.position; pp.firstPosition();
int next = ((PhrasePositions)pq.top()).position; if (pp.position > end)
for (int pos = start; pos <= next; pos = pp.position) { end = pp.position;
start = pos; // advance pp to min window pq.put(pp); // build pq from list
if (!pp.nextPosition()) { }
done = true; // ran out of a term -- done
break;
}
}
int matchLength = end - start; float freq = 0.0f;
if (matchLength <= slop) boolean done = false;
freq += getSimilarity().sloppyFreq(matchLength); // score match do {
PhrasePositions pp = (PhrasePositions) pq.pop();
int start = pp.position;
int next = ((PhrasePositions) pq.top()).position;
for (int pos = start; pos <= next; pos = pp.position) {
start = pos; // advance pp to min window
if (!pp.nextPosition()) {
done = true; // ran out of a term -- done
break;
}
}
if (pp.position > end) int matchLength = end - start;
end = pp.position; if (matchLength <= slop)
pq.put(pp); // restore pq freq += getSimilarity().sloppyFreq(matchLength); // score match
} while (!done);
if (pp.position > end)
return freq; end = pp.position;
} pq.put(pp); // restore pq
} while (!done);
return freq;
}
} }