Lucene - Java Users - ShingleFilter

Need to set outputUnigrams = false with something like:

      StandardTokenizer source = new StandardTokenizer(Version.LUCENE_43, reader);
      TokenStream tokenStream = new StandardFilter(Version.LUCENE_43, source);
      tokenStream = new LowerCaseFilter(Version.LUCENE_43, tokenStream);

      TokenFilter sf = new ShingleFilter(tokenStream, 3,3);
      ((ShingleFilter)sf).setOutputUnigrams(false);

      sf = new StopFilter(Version.LUCENE_43,sf,StopAnalyzer.ENGLISH_STOP_WORDS_SET);
     
      return new Analyzer.TokenStreamComponents(source, sf);

Read full article from Lucene - Java Users - ShingleFilter