Java PhraseQuery.Builder方法代码示例

import org.apache.lucene.search.PhraseQuery; //导入方法依赖的package包/类
private Query applySlop(Query q, int slop) {
    if (q instanceof PhraseQuery) {
        PhraseQuery pq = (PhraseQuery) q;
        PhraseQuery.Builder builder = new PhraseQuery.Builder();
        builder.setSlop(slop);
        final Term[] terms = pq.getTerms();
        final int[] positions = pq.getPositions();
        for (int i = 0; i < terms.length; ++i) {
            builder.add(terms[i], positions[i]);
        }
        pq = builder.build();
        pq.setBoost(q.getBoost());
        return pq;
    } else if (q instanceof MultiPhraseQuery) {
        ((MultiPhraseQuery) q).setSlop(slop);
        return q;
    } else {
        return q;
    }
}
 

import org.apache.lucene.search.PhraseQuery; //导入方法依赖的package包/类
/**
 * Creates simple phrase query from the cached tokenstream contents
 */
protected Query analyzePhrase(String field, TokenStream stream, int slop) throws IOException {
	PhraseQuery.Builder builder = new PhraseQuery.Builder();
	builder.setSlop(slop);

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

	stream.reset();
	while (stream.incrementToken()) {
		if (enablePositionIncrements) {
			position += posIncrAtt.getPositionIncrement();
		} else {
			position += 1;
		}
		builder.add(new Term(field, termAtt.getBytesRef()), position);
	}

	return builder.build();
}
 

import org.apache.lucene.search.PhraseQuery; //导入方法依赖的package包/类
private Query applySlop(Query q, int slop) {
	if (q instanceof PhraseQuery) {
		PhraseQuery.Builder builder = new PhraseQuery.Builder();
		builder.setSlop(slop);
		PhraseQuery pq = (PhraseQuery) q;
		org.apache.lucene.index.Term[] terms = pq.getTerms();
		int[] positions = pq.getPositions();
		for (int i = 0; i < terms.length; ++i) {
			builder.add(terms[i], positions[i]);
		}
		q = builder.build();
	}
	else if (q instanceof MultiPhraseQuery) {
		MultiPhraseQuery mpq = (MultiPhraseQuery) q;

		if (slop != mpq.getSlop()) {
			q = new MultiPhraseQuery.Builder(mpq).setSlop(slop).build();
		}
	}
	return q;
}
 

import org.apache.lucene.search.PhraseQuery; //导入方法依赖的package包/类
/**
 * Rebuild a phrase query with a slop value
 */
private PhraseQuery addSlopToPhrase(PhraseQuery query, int slop) {
    PhraseQuery.Builder builder = new PhraseQuery.Builder();
    builder.setSlop(slop);
    final Term[] terms = query.getTerms();
    final int[] positions = query.getPositions();
    for (int i = 0; i < terms.length; ++i) {
        builder.add(terms[i], positions[i]);
    }

    return builder.build();
}
 

import org.apache.lucene.search.PhraseQuery; //导入方法依赖的package包/类
private void createChainProximityQueryClauses(BooleanQuery.Builder builder, String field, CachingTokenFilter stream, TermToBytesRefAttribute termAtt, OffsetAttribute offsetAtt) throws IOException {
    Term termArr[] = new Term[2];
    long offsetArr[] = new long[2];
    for(int i=0;i<2;i++) {
        termArr[i] = null;
        offsetArr[i] = 0;
    }

    while (stream.incrementToken()) {
        Term t = new Term(field, BytesRef.deepCopyOf(termAtt.getBytesRef()));
        if(termArr[0] == null) {
            termArr[0] = t;
            offsetArr[0] = offsetAtt.startOffset();
        } else if(termArr[1] == null) {
            termArr[1] = t;
            offsetArr[1] = offsetAtt.startOffset();
        } else {
            // shift
            termArr[0] = termArr[1];
            offsetArr[0] = offsetArr[1];
            // fill
            termArr[1] = t;
            offsetArr[1] = offsetAtt.startOffset();
        }
        
        if(termArr[0] != null && termArr[1] != null) {
            long offsetDiff = offsetArr[1] - offsetArr[0];
            if(offsetDiff > 0) {
                PhraseQuery.Builder pq = new PhraseQuery.Builder();

                pq.setSlop((int) (offsetDiff) + 1);
                pq.add(termArr[0]);
                pq.add(termArr[1]);

                builder.add(pq.build(), BooleanClause.Occur.SHOULD);
            }
        }
    }
}
 

import org.apache.lucene.search.PhraseQuery; //导入方法依赖的package包/类
@Override
public List<Song> search(IndexType indexType, String searchString, FieldName... fieldsToSearchIn) {
	try {
		Directory directory = getIndex(indexType);
		IndexReader indexReader = DirectoryReader.open(directory);
		IndexSearcher indexSearcher = new IndexSearcher(indexReader);
		
		BooleanQuery.Builder outerBooleanQueryBuilder = new BooleanQuery.Builder();
		for (FieldName field : fieldsToSearchIn) {
			PhraseQuery.Builder phraseQueryBuilder = new PhraseQuery.Builder();
			for (String searchTerm : searchString.toLowerCase().split(TERM_SPLIT_REGEX)) {
				phraseQueryBuilder.add(new Term(field.name(), searchTerm));
			}
			PhraseQuery phraseQuery = phraseQueryBuilder.build();
			outerBooleanQueryBuilder.add(phraseQuery, Occur.SHOULD);
		}
		BooleanQuery outerBooleanQuery = outerBooleanQueryBuilder.build();
		TopDocs hits = indexSearcher.search(outerBooleanQuery, Integer.MAX_VALUE);
		
		LOG.debug("{} hits for filter \"{}\"", hits.totalHits, outerBooleanQueryBuilder);
		
		List<Song> ret = new LinkedList<>();
		for (ScoreDoc scoreDocument : hits.scoreDocs) {
			Document document;
			document = indexSearcher.doc(scoreDocument.doc);
			String uuid = document.get(FieldName.UUID.name());
			ret.add(songByUuid.get(uuid));
		}
		return ret;
	} catch (Exception e) {
		LOG.warn("problem while searching", e);
		return new ArrayList<>(0);
	}
}
 

import org.apache.lucene.search.PhraseQuery; //导入方法依赖的package包/类
/**
 * Rebuild a phrase query with a slop value
 */
private PhraseQuery addSlopToPhrase(PhraseQuery query, int slop) {
	PhraseQuery.Builder builder = new PhraseQuery.Builder();
	builder.setSlop(slop);
	org.apache.lucene.index.Term[] terms = query.getTerms();
	int[] positions = query.getPositions();
	for (int i = 0; i < terms.length; ++i) {
		builder.add(terms[i], positions[i]);
	}

	return builder.build();
}
 

import org.apache.lucene.search.PhraseQuery; //导入方法依赖的package包/类
@Test
public void testMultiWordSynonyms() throws Exception {
	QueryParser dumb = new QueryParser("field", new Analyzer1());
	dumb.setSplitOnWhitespace(false);

	TermQuery guinea = new TermQuery(new Term("field", "guinea"));
	TermQuery pig = new TermQuery(new Term("field", "pig"));
	TermQuery cavy = new TermQuery(new Term("field", "cavy"));

	// A multi-word synonym source will form a graph query for synonyms that
	// formed the graph token stream
	BooleanQuery.Builder synonym = new BooleanQuery.Builder();
	synonym.add(guinea, BooleanClause.Occur.SHOULD);
	synonym.add(pig, BooleanClause.Occur.SHOULD);
	BooleanQuery guineaPig = synonym.build();

	// With the phrase operator, a multi-word synonym source will form a
	// graph query with inner phrase queries.
	PhraseQuery.Builder phraseSynonym = new PhraseQuery.Builder();
	phraseSynonym.add(new Term("field", "guinea"));
	phraseSynonym.add(new Term("field", "pig"));
	PhraseQuery guineaPigPhrase = phraseSynonym.build();

	// custom behavior, the synonyms are expanded, unless you use quote
	// operator
	QueryParser smart = new SmartQueryParser();
	smart.setSplitOnWhitespace(false);
	Assert.assertEquals(guineaPigPhrase, smart.parse("\"guinea pig\""));
}
 

import org.apache.lucene.search.PhraseQuery; //导入方法依赖的package包/类
@Test
public void testPhraseQueryPositionIncrements() throws Exception {
	CharacterRunAutomaton stopStopList = new CharacterRunAutomaton(new RegExp("[sS][tT][oO][pP]").toAutomaton());

	QueryParser qp = getParserConfig(new MockAnalyzer(random(), MockTokenizer.WHITESPACE, false, stopStopList));
	qp.setEnablePositionIncrements(true);

	PhraseQuery.Builder phraseQuery = new PhraseQuery.Builder();
	phraseQuery.add(new Term("field", "1"));
	phraseQuery.add(new Term("field", "2"), 2);
	Assert.assertEquals(phraseQuery.build(), getQuery("\"1 stop 2\"", qp));
}
 

import org.apache.lucene.search.PhraseQuery; //导入方法依赖的package包/类
public void testPhraseQueryPositionIncrements() throws Exception {
  CharacterRunAutomaton stopStopList =
      new CharacterRunAutomaton(new RegExp("[sS][tT][oO][pP]").toAutomaton());

  CommonQueryParserConfiguration qp = getParserConfig(new MockAnalyzer(random(), MockTokenizer.WHITESPACE, false, stopStopList));

  qp = getParserConfig(
      new MockAnalyzer(random(), MockTokenizer.WHITESPACE, false, stopStopList));
  qp.setEnablePositionIncrements(true);

  PhraseQuery.Builder phraseQuery = new PhraseQuery.Builder();
  phraseQuery.add(new Term("field", "1"));
  phraseQuery.add(new Term("field", "2"), 2);
  assertEquals(phraseQuery.build(), getQuery("\"1 stop 2\"",qp));
}
 

import org.apache.lucene.search.PhraseQuery; //导入方法依赖的package包/类
private void createPairedProximityQueryClauses(BooleanQuery.Builder builder, String field, CachingTokenFilter stream, TermToBytesRefAttribute termAtt, OffsetAttribute offsetAtt) throws IOException {
    Term termArr[] = new Term[2];
    long offsetArr[] = new long[2];
    for(int i=0;i<2;i++) {
        termArr[i] = null;
        offsetArr[i] = 0;
    }

    int count = 0;
    while (stream.incrementToken()) {
        Term t = new Term(field, BytesRef.deepCopyOf(termAtt.getBytesRef()));
        if(count % 2 == 0) {
            termArr[0] = t;
            offsetArr[0] = offsetAtt.startOffset();
        } else {
            termArr[1] = t;
            offsetArr[1] = offsetAtt.startOffset();

            long offsetDiff = offsetArr[1] - offsetArr[0];
            if(offsetDiff > 0) {
                PhraseQuery.Builder pq = new PhraseQuery.Builder();

                pq.setSlop((int) (offsetDiff) + 1);
                pq.add(termArr[0]);
                pq.add(termArr[1]);

                builder.add(pq.build(), BooleanClause.Occur.SHOULD);
            }

            termArr[0] = null;
            termArr[1] = null;
        }

        count++;
    }
    
    if(termArr[0] != null) {
        builder.add(new TermQuery(termArr[0]), BooleanClause.Occur.SHOULD);
        termArr[0] = null;
    }
}