Examples with MoreLikeThisQuery org.apache.lucene.queries.mlt.MoreLikeThisQuery used on opensource projects

Example 1 with MoreLikeThisQuery

use of org.apache.lucene.queries.mlt.MoreLikeThisQuery in project lucene-solr by apache.

the class LikeThisQueryBuilder method getQuery.

/* (non-Javadoc)
    * @see org.apache.lucene.xmlparser.QueryObjectBuilder#process(org.w3c.dom.Element)
    */
@Override
public Query getQuery(Element e) throws ParserException {
    //a comma-delimited list of fields
    String fieldsList = e.getAttribute("fieldNames");
    String[] fields = defaultFieldNames;
    if ((fieldsList != null) && (fieldsList.trim().length() > 0)) {
        fields = fieldsList.trim().split(",");
        //trim the fieldnames
        for (int i = 0; i < fields.length; i++) {
            fields[i] = fields[i].trim();
        }
    }
    //Parse any "stopWords" attribute
    //TODO MoreLikeThis needs to ideally have per-field stopWords lists - until then
    //I use all analyzers/fields to generate multi-field compatible stop list
    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;
}