=================================================================== --- D:/lucene-2.2.0/src/java/org/apache/lucene/search/payloads/BoostingTermQuery.java (revision 581681) +++ D:/lucene-2.2.0/src/java/org/apache/lucene/search/payloads/BoostingTermQuery.java (working copy) @@ -2,13 +2,11 @@ import org.apache.lucene.index.IndexReader; import org.apache.lucene.index.Term; +import org.apache.lucene.index.TermDocs; import org.apache.lucene.index.TermPositions; import org.apache.lucene.search.*; -import org.apache.lucene.search.spans.SpanScorer; -import org.apache.lucene.search.spans.SpanTermQuery; -import org.apache.lucene.search.spans.SpanWeight; -import org.apache.lucene.search.spans.TermSpans; + import java.io.IOException; /** @@ -28,7 +26,7 @@ */ /** - * The BoostingTermQuery is very similar to the {@link org.apache.lucene.search.spans.SpanTermQuery} except + * The BoostingTermQuery is very similar to the {@link org.apache.lucene.search.TermQuery} except * that it factors in the value of the payload located at each of the positions where the * {@link org.apache.lucene.index.Term} occurs. *

@@ -44,145 +42,136 @@ * * @see org.apache.lucene.search.Similarity#scorePayload(byte[], int, int) */ -public class BoostingTermQuery extends SpanTermQuery{ +public class BoostingTermQuery extends TermQuery{ + Term term; + Similarity similarity; + public BoostingTermQuery(Term term) { + super(term); + this.term = term; - public BoostingTermQuery(Term term) { - super(term); - } + } + protected Weight createWeight(Searcher searcher) throws IOException { + this.similarity = getSimilarity(searcher); + return new BoostingTermWeight(this, searcher); + } - protected Weight createWeight(Searcher searcher) throws IOException { - return new BoostingTermWeight(this, searcher); - } + protected class BoostingTermWeight extends TermWeight implements Weight { + public BoostingTermWeight(BoostingTermQuery query, Searcher searcher) throws IOException { + super(searcher); + } - protected class BoostingTermWeight extends SpanWeight implements Weight { + public Scorer scorer(IndexReader reader) throws IOException { + return new BoostingTermScorer(reader.termDocs(term), reader.termPositions(term), this, similarity, + reader.norms(term.field())); + } + class BoostingTermScorer extends TermScorer { + //TODO: is this the best way to allocate this? + byte[] payload = new byte[256]; + private TermPositions positions; + protected float payloadScore; + private int payloadsSeen; + boolean more; + protected boolean firstTime = true; + protected int doc; + protected int freq; + protected int position; + public BoostingTermScorer(TermDocs termDocs, TermPositions termPositions, Weight weight, + Similarity similarity, byte[] norms) throws IOException { + super(weight, termDocs, similarity, norms); + positions = termPositions; - public BoostingTermWeight(BoostingTermQuery query, Searcher searcher) throws IOException { - super(query, searcher); - } + } + /** + * Go to the next document + * + */ + public boolean next() throws IOException { + if (super.next()) { + return getPayloads(); + } else { + return false; + } + } + public boolean skipTo(int target) throws IOException { + if (super.skipTo(target)) { + return getPayloads(); + } else { + return false; + } + } + protected boolean getPayloads() throws IOException { + doc = super.doc(); + payloadScore = 0; + payloadsSeen = 0; + Similarity similarity1 = getSimilarity(); + // set term positions to current doc + positions.skipTo(doc); + freq = positions.freq(); + for (int i=0;i 0 ? (payloadScore / payloadsSeen) : 1); + } - //TODO: is this the best way to allocate this? - byte[] payload = new byte[256]; - private TermPositions positions; - protected float payloadScore; - private int payloadsSeen; + public void score(HitCollector hc) throws IOException { + while (next()) { + hc.collect(doc, score()); + } + } - public BoostingSpanScorer(TermSpans spans, Weight weight, - Similarity similarity, byte[] norms) throws IOException { - super(spans, weight, similarity, norms); - positions = spans.getPositions(); + public Explanation explain(final int doc) throws IOException { + Explanation result = new Explanation(); + Explanation nonPayloadExpl = super.explain(doc); + result.addDetail(nonPayloadExpl); + //QUESTION: Is there a wau to avoid this skipTo call? We need to know whether to load the payload or not - } - - /** - * Go to the next document - * - */ - /*public boolean next() throws IOException { - - boolean result = super.next(); - //set the payload. super.next() properly increments the term positions - if (result) { - //Load the payloads for all - processPayload(); - } - - return result; - } - - public boolean skipTo(int target) throws IOException { - boolean result = super.skipTo(target); - - if (result) { - processPayload(); - } - - return result; - }*/ - - protected boolean setFreqCurrentDoc() throws IOException { - if (!more) { - return false; - } - doc = spans.doc(); - freq = 0.0f; - payloadScore = 0; - payloadsSeen = 0; - Similarity similarity1 = getSimilarity(); - while (more && doc == spans.doc()) { - int matchLength = spans.end() - spans.start(); - - freq += similarity1.sloppyFreq(matchLength); - processPayload(similarity1); - - more = spans.next();//this moves positions to the next match in this document - } - return more || (freq != 0); - } - - - protected void processPayload(Similarity similarity) throws IOException { - if (positions.isPayloadAvailable()) { - payload = positions.getPayload(payload, 0); - payloadScore += similarity.scorePayload(payload, 0, positions.getPayloadLength()); - payloadsSeen++; - - } else { - //zero out the payload? - } - - } - - public float score() throws IOException { - - return super.score() * (payloadsSeen > 0 ? (payloadScore / payloadsSeen) : 1); - } - - - public Explanation explain(final int doc) throws IOException { - Explanation result = new Explanation(); - Explanation nonPayloadExpl = super.explain(doc); - result.addDetail(nonPayloadExpl); - //QUESTION: Is there a wau to avoid this skipTo call? We need to know whether to load the payload or not - - Explanation payloadBoost = new Explanation(); - result.addDetail(payloadBoost); -/* + Explanation payloadBoost = new Explanation(); + result.addDetail(payloadBoost); + /* if (skipTo(doc) == true) { processPayload(); } -*/ + */ - float avgPayloadScore = (payloadsSeen > 0 ? (payloadScore / payloadsSeen) : 1); - payloadBoost.setValue(avgPayloadScore); - //GSI: I suppose we could toString the payload, but I don't think that would be a good idea - payloadBoost.setDescription("scorePayload(...)"); - result.setValue(nonPayloadExpl.getValue() * avgPayloadScore); - result.setDescription("btq, product of:"); - return result; - } - } + float avgPayloadScore = (payloadsSeen > 0 ? (payloadScore / payloadsSeen) : 1); + payloadBoost.setValue(avgPayloadScore); + //GSI: I suppose we could toString the payload, but I don't think that would be a good idea + payloadBoost.setDescription("scorePayload(...)"); + result.setValue(nonPayloadExpl.getValue() * avgPayloadScore); + result.setDescription("btq, product of:"); + return result; + } + } - } + } - public boolean equals(Object o) { - if (!(o instanceof BoostingTermQuery)) - return false; - BoostingTermQuery other = (BoostingTermQuery) o; - return (this.getBoost() == other.getBoost()) - && this.term.equals(other.term); - } + public boolean equals(Object o) { + if (!(o instanceof BoostingTermQuery)) + return false; + BoostingTermQuery other = (BoostingTermQuery) o; + return (this.getBoost() == other.getBoost()) + && this.term.equals(other.term); + } }