Subclasses must apply their scoring formula in this class.
+ * @param stats the corpus level statistics. + * @param doc the document id. + * @param freq the term frequency. + * @param norms the field norms. + * @return the score. + */ + protected abstract float score( + EasyStats stats, int doc, float freq, byte[] norms); + + @Override + public ExactDocScorer exactDocScorer(Stats stats, String fieldName, + AtomicReaderContext context) throws IOException { + return new EasyExactDocScorer((EasyStats) stats, + context.reader.norms(fieldName)); + } + + @Override + public SloppyDocScorer sloppyDocScorer(Stats stats, String fieldName, + AtomicReaderContext context) throws IOException { + return new EasySloppyDocScorer((EasyStats) stats, + context.reader.norms(fieldName)); + } + + // --------------------------------- Classes --------------------------------- + + /** Calls the {@code score()} method. */ + // nocommit Javadoc + private class EasyExactDocScorer extends ExactDocScorer { + private final EasyStats stats; + private final byte[] norms; + + EasyExactDocScorer(EasyStats stats, byte norms[]) { + this.stats = stats; + this.norms = norms; + } + + @Override + public float score(int doc, int freq) { + return EasySimilarity.this.score(stats, doc, freq, norms); + } + } + + /** Calls the {@code score()} method. */ + // nocommit Javadoc + private class EasySloppyDocScorer extends SloppyDocScorer { + private final EasyStats stats; + private final byte[] norms; + + EasySloppyDocScorer(EasyStats stats, byte norms[]) { + this.stats = stats; + this.norms = norms; + } + + // todo: optimize + @Override + public float score(int doc, float freq) { + return EasySimilarity.this.score(stats, doc, freq, norms); + } + } +} diff --git lucene/src/test/org/apache/lucene/search/similarities/EasyStats.java lucene/src/test/org/apache/lucene/search/similarities/EasyStats.java new file mode 100644 index 0000000..f84fe95 --- /dev/null +++ lucene/src/test/org/apache/lucene/search/similarities/EasyStats.java @@ -0,0 +1,137 @@ +package org.apache.lucene.search.similarities; + +import org.apache.lucene.search.Similarity; + +/** + * Stores all statistics commonly used ranking methods. + * @lucene.experimental + */ +public class EasyStats extends Similarity.Stats { + /** The number of documents. */ + protected int numberOfDocuments; // TODO: to long? + /** The total number of tokens in the field. */ + protected long numberOfFieldTokens; + /** The average field length. */ + protected float avgFieldLength; + /** The document frequency. */ + protected int docFreq; + /** The total number of occurrences of this term across all documents. */ + // TODO: same field? + protected long totalTermFreq; + /** The total number of terms across all documents. */ + // TODO: same field? + protected long sumTotalTermFreq; + /** The number of unique terms. */ + // nocommit might be per-segment only + protected long uniqueTermCount; + + // -------------------------- Boost-related stuff -------------------------- + + /** Query's inner boost. */ + protected final float queryBoost; + /** Any outer query's boost. */ + protected float topLevelBoost; + + /** Constructor. Sets the query boost. */ + public EasyStats(float queryBoost) { + this.queryBoost = queryBoost; + } + + // ------------------------- Getter/setter methods ------------------------- + + /** Returns the number of documents. */ + public int getNumberOfDocuments() { + return numberOfDocuments; + } + + /** Sets the number of documents. */ + public void setNumberOfDocuments(int numberOfDocuments) { + this.numberOfDocuments = numberOfDocuments; + } + + /** Returns the total number of tokens in the field. */ + public long getNumberOfFieldTokens() { + return numberOfFieldTokens; + } + + /** Sets the total number of tokens in the field. */ + public void setNumberOfFieldTokens(long numberOfFieldTokens) { + this.numberOfFieldTokens = numberOfFieldTokens; + } + + /** Returns the average field length. */ + public float getAvgFieldLength() { + return avgFieldLength; + } + + /** Sets the average field length. */ + public void setAvgFieldLength(float avgFieldLength) { + this.avgFieldLength = avgFieldLength; + } + + /** Returns the document frequency. */ + public int getDocFreq() { + return docFreq; + } + + /** Sets the document frequency. */ + public void setDocFreq(int docFreq) { + this.docFreq = docFreq; + } + + /** Returns the total number of occurrences of this term across all documents. */ + public long getTotalTermFreq() { + return totalTermFreq; + } + + /** Sets the total number of occurrences of this term across all documents. */ + public void setTotalTermFreq(long totalTermFreq) { + this.totalTermFreq = totalTermFreq; + } + + /** Returns the total number of terms across all documents. */ + public long getSumTotalTermFreq() { + return sumTotalTermFreq; + } + + /** Sets the total number of terms across all documents. */ + public void setSumTotalTermFreq(long sumTotalTermFreq) { + this.sumTotalTermFreq = sumTotalTermFreq; + } + + /** Returns the number of unique terms. */ + public long getUniqueTermCount() { + return uniqueTermCount; + } + + /** Sets the number of unique terms. */ + public void setUniqueTermCount(long uniqueTermCount) { + this.uniqueTermCount = uniqueTermCount; + } + + // -------------------------- Boost-related stuff -------------------------- + + /** The square of the raw normalization value. + * @see #rawNormalizationValue() */ + @Override + public float getValueForNormalization() { + float rawValue = rawNormalizationValue(); + return rawValue * rawValue; + } + + /** Computes the raw normalization value. This basic implementation returns + * the query boost. Subclasses may override this method to include other + * factors (such as idf), or to save the value for inclusion in + * {@link #normalize(float, float)}, etc. + */ + protected float rawNormalizationValue() { + return queryBoost; + } + + /** No normalization is done. {@code topLevelBoost} is saved in the object, + * however. */ + @Override + public void normalize(float queryNorm, float topLevelBoost) { + this.topLevelBoost = topLevelBoost; + } +} diff --git my_stuff/Eclipse-Lucene-Codestyle.xml my_stuff/Eclipse-Lucene-Codestyle.xml new file mode 100644 index 0000000..07d4966 --- /dev/null +++ my_stuff/Eclipse-Lucene-Codestyle.xml @@ -0,0 +1,269 @@ + +