Java Code Examples for org.apache.lucene.search.MultiPhraseQuery#Builder

public void testQueryScorerMultiPhraseQueryHighlightingWithGap() throws Exception {
  MultiPhraseQuery.Builder mpqb = new MultiPhraseQuery.Builder();

  /*
   * The toString of MultiPhraseQuery doesn't work so well with these
   * out-of-order additions, but the Query itself seems to match accurately.
   */

  mpqb.add(new Term[] { new Term(FIELD_NAME, "wordz") }, 2);
  mpqb.add(new Term[] { new Term(FIELD_NAME, "wordx") }, 0);

  doSearching(mpqb.build());

  final int maxNumFragmentsRequired = 1;
  final int expectedHighlights = 2;

  assertExpectedHighlightCount(maxNumFragmentsRequired, expectedHighlights);
}
 

public void testMultiPhraseQueryParsing() throws Exception {
  TokenAndPos[] INCR_0_QUERY_TOKENS_AND = new TokenAndPos[]{
      new TokenAndPos("a", 0),
      new TokenAndPos("1", 0),
      new TokenAndPos("b", 1),
      new TokenAndPos("1", 1),
      new TokenAndPos("c", 2)
  };

  QueryParser qp = new QueryParser("field", new CannedAnalyzer(INCR_0_QUERY_TOKENS_AND));
  Query q = qp.parse("\"this text is acually ignored\"");
  assertTrue("wrong query type!", q instanceof MultiPhraseQuery);

  MultiPhraseQuery.Builder multiPhraseQueryBuilder = new MultiPhraseQuery.Builder();
  multiPhraseQueryBuilder.add(new Term[]{ new Term("field", "a"), new Term("field", "1") }, -1);
  multiPhraseQueryBuilder.add(new Term[]{ new Term("field", "b"), new Term("field", "1") }, 0);
  multiPhraseQueryBuilder.add(new Term[]{ new Term("field", "c") }, 1);

  assertEquals(multiPhraseQueryBuilder.build(), q);
}
 

public void testQueryScorerMultiPhraseQueryHighlighting() throws Exception {
  MultiPhraseQuery.Builder mpqb = new MultiPhraseQuery.Builder();

  mpqb.add(new Term[] { new Term(FIELD_NAME, "wordx"), new Term(FIELD_NAME, "wordb") });
  mpqb.add(new Term(FIELD_NAME, "wordy"));

  doSearching(mpqb.build());

  final int maxNumFragmentsRequired = 2;
  assertExpectedHighlightCount(maxNumFragmentsRequired, 6);
}
 

/** 
 * Creates complex phrase query from the cached tokenstream contents 
 */
protected Query analyzeMultiPhrase(String field, TokenStream stream, int slop) throws IOException {
  MultiPhraseQuery.Builder mpqb = newMultiPhraseQueryBuilder();
  mpqb.setSlop(slop);
  
  TermToBytesRefAttribute termAtt = stream.getAttribute(TermToBytesRefAttribute.class);

  PositionIncrementAttribute posIncrAtt = stream.getAttribute(PositionIncrementAttribute.class);
  int position = -1;  
  
  List<Term> multiTerms = new ArrayList<>();
  stream.reset();
  while (stream.incrementToken()) {
    int positionIncrement = posIncrAtt.getPositionIncrement();
    
    if (positionIncrement > 0 && multiTerms.size() > 0) {
      if (enablePositionIncrements) {
        mpqb.add(multiTerms.toArray(new Term[0]), position);
      } else {
        mpqb.add(multiTerms.toArray(new Term[0]));
      }
      multiTerms.clear();
    }
    position += positionIncrement;
    multiTerms.add(new Term(field, termAtt.getBytesRef()));
  }
  
  if (enablePositionIncrements) {
    mpqb.add(multiTerms.toArray(new Term[0]), position);
  } else {
    mpqb.add(multiTerms.toArray(new Term[0]));
  }
  return mpqb.build();
}
 

/** forms multiphrase query */
public void testSynonymsPhrase() throws Exception {
  MultiPhraseQuery.Builder expectedBuilder = new MultiPhraseQuery.Builder();
  expectedBuilder.add(new Term("field", "old"));
  expectedBuilder.add(new Term[] { new Term("field", "dogs"), new Term("field", "dog") });
  QueryBuilder builder = new QueryBuilder(new MockSynonymAnalyzer());
  assertEquals(expectedBuilder.build(), builder.createPhraseQuery("field", "old dogs"));
}
 

/** forms multiphrase query */
public void testCJKSynonymsPhrase() throws Exception {
  MultiPhraseQuery.Builder expectedBuilder = new MultiPhraseQuery.Builder();
  expectedBuilder.add(new Term("field", "中"));
  expectedBuilder.add(new Term[] { new Term("field", "国"), new Term("field", "國")});
  QueryBuilder builder = new QueryBuilder(new MockCJKSynonymAnalyzer());
  assertEquals(expectedBuilder.build(), builder.createPhraseQuery("field", "中国"));
  expectedBuilder.setSlop(3);
  assertEquals(expectedBuilder.build(), builder.createPhraseQuery("field", "中国", 3));
}
 

@Override
public Query multiPhraseQuery(String field, TokenStream stream, int slop, boolean enablePositionIncrements) throws IOException {

    MultiPhraseQuery.Builder mpqb = new MultiPhraseQuery.Builder();
    mpqb.setSlop(slop);

    TermToBytesRefAttribute termAtt = stream.getAttribute(TermToBytesRefAttribute.class);

    PositionIncrementAttribute posIncrAtt = stream.getAttribute(PositionIncrementAttribute.class);
    int position = -1;

    List<Term> multiTerms = new ArrayList<>();
    stream.reset();
    while (stream.incrementToken()) {
        int positionIncrement = posIncrAtt.getPositionIncrement();

        if (positionIncrement > 0 && multiTerms.size() > 0) {
            if (enablePositionIncrements) {
                mpqb.add(multiTerms.toArray(new Term[0]), position);
            } else {
                mpqb.add(multiTerms.toArray(new Term[0]));
            }
            multiTerms.clear();
        }
        position += positionIncrement;
        multiTerms.add(new Term(field, termAtt.getBytesRef()));
    }

    if (enablePositionIncrements) {
        mpqb.add(multiTerms.toArray(new Term[0]), position);
    } else {
        mpqb.add(multiTerms.toArray(new Term[0]));
    }
    return mpqb.build();
}
 

@Override
public Query rewrite(IndexReader reader) throws IOException {
    Query rewritten = super.rewrite(reader);
    if (rewritten != this) {
        return rewritten;
    }
    if (termArrays.isEmpty()) {
        return new MatchNoDocsQuery();
    }
    MultiPhraseQuery.Builder query = new MultiPhraseQuery.Builder();
    query.setSlop(slop);
    int sizeMinus1 = termArrays.size() - 1;
    for (int i = 0; i < sizeMinus1; i++) {
        query.add(termArrays.get(i), positions.get(i));
    }
    Term[] suffixTerms = termArrays.get(sizeMinus1);
    int position = positions.get(sizeMinus1);
    ObjectHashSet<Term> terms = new ObjectHashSet<>();
    for (Term term : suffixTerms) {
        getPrefixTerms(terms, term, reader);
        if (terms.size() > maxExpansions) {
            break;
        }
    }
    if (terms.isEmpty()) {
        if (sizeMinus1 == 0) {
            // no prefix and the phrase query is empty
            return Queries.newMatchNoDocsQuery("No terms supplied for " + MultiPhrasePrefixQuery.class.getName());
        }

        // if the terms does not exist we could return a MatchNoDocsQuery but this would break the unified highlighter
        // which rewrites query with an empty reader.
        return new BooleanQuery.Builder()
            .add(query.build(), BooleanClause.Occur.MUST)
            .add(Queries.newMatchNoDocsQuery("No terms supplied for " + MultiPhrasePrefixQuery.class.getName()),
                BooleanClause.Occur.MUST).build();
    }
    query.add(terms.toArray(Term.class), position);
    return query.build();
}
 

/**
 * Builds a new MultiPhraseQuery instance.
 * <p>
 * This is intended for subclasses that wish to customize the generated queries.
 * @return new MultiPhraseQuery instance
 */
protected MultiPhraseQuery.Builder newMultiPhraseQueryBuilder() {
  return new MultiPhraseQuery.Builder();
}