package org.apache.lucene.queryparser.xml.builders;
import java.io.IOException;
import java.util.HashSet;
import java.util.Set;
import org.apache.lucene.analysis.Analyzer;
import org.apache.lucene.analysis.TokenStream;
import org.apache.lucene.analysis.tokenattributes.CharTermAttribute;
import org.apache.lucene.queries.mlt.MoreLikeThisQuery;
import org.apache.lucene.queryparser.xml.QueryBuilder;
import org.apache.lucene.search.BoostQuery;
import org.apache.lucene.search.Query;
import org.apache.lucene.queryparser.xml.DOMUtils;
import org.apache.lucene.queryparser.xml.ParserException;
import org.w3c.dom.Element;
public class LikeThisQueryBuilder implements QueryBuilder {
private static final int DEFAULT_MAX_QUERY_TERMS = 20;
private static final int DEFAULT_MIN_TERM_FREQUENCY = 1;
private static final float DEFAULT_PERCENT_TERMS_TO_MATCH = 30;
private final Analyzer analyzer;
private final String defaultFieldNames[];
public LikeThisQueryBuilder(Analyzer analyzer, String[] defaultFieldNames) {
this.analyzer = analyzer;
this.defaultFieldNames = defaultFieldNames;
}
@Override
public Query getQuery(Element e) throws ParserException {
String fieldsList = e.getAttribute("fieldNames");
String fields[] = defaultFieldNames;
if ((fieldsList != null) && (fieldsList.trim().length() > 0)) {
fields = fieldsList.trim().split(",");
for (int i = 0; i < fields.length; i++) {
fields[i] = fields[i].trim();
}
}
String stopWords = e.getAttribute("stopWords");
Set<String> stopWordsSet = null;
if ((stopWords != null) && (fields != null)) {
stopWordsSet = new HashSet<>();
for (String field : fields) {
try (TokenStream ts = analyzer.tokenStream(field, stopWords)) {
CharTermAttribute termAtt = ts.addAttribute(CharTermAttribute.class);
ts.reset();
while (ts.incrementToken()) {
stopWordsSet.add(termAtt.toString());
}
ts.end();
} catch (IOException ioe) {
throw new ParserException("IoException parsing stop words list in "
+ getClass().getName() + ":" + ioe.getLocalizedMessage());
}
}
}
MoreLikeThisQuery mlt = new MoreLikeThisQuery(DOMUtils.getText(e), fields, analyzer, fields[0]);
mlt.setMaxQueryTerms(DOMUtils.getAttribute(e, "maxQueryTerms", DEFAULT_MAX_QUERY_TERMS));
mlt.setMinTermFrequency(DOMUtils.getAttribute(e, "minTermFrequency", DEFAULT_MIN_TERM_FREQUENCY));
mlt.setPercentTermsToMatch(DOMUtils.getAttribute(e, "percentTermsToMatch", DEFAULT_PERCENT_TERMS_TO_MATCH) / 100);
mlt.setStopWords(stopWordsSet);
int minDocFreq = DOMUtils.getAttribute(e, "minDocFreq", -1);
if (minDocFreq >= 0) {
mlt.setMinDocFreq(minDocFreq);
}
Query q = mlt;
float boost = DOMUtils.getAttribute(e, "boost", 1.0f);
if (boost != 1f) {
q = new BoostQuery(mlt, boost);
}
return q;
}
}