Examples of org.apache.lucene.search.FuzzyQuery

org.apache.lucene.search.FuzzyQuery
Implements the fuzzy search query. The similarity measurement is based on the Levenshtein (edit distance) algorithm. Warning: this query is not very scalable with its default prefix length of 0 - in this case, *every* term will be enumerated and cause an edit score calculation.


    assertTrue(getQuery("term*", null) instanceof PrefixQuery);
    assertTrue(getQuery("term*^2", null) instanceof PrefixQuery);
    assertTrue(getQuery("term~", null) instanceof FuzzyQuery);
    assertTrue(getQuery("term~0.7", null) instanceof FuzzyQuery);
    FuzzyQuery fq = (FuzzyQuery) getQuery("term~0.7", null);
    assertEquals(1, fq.getMaxEdits());
    assertEquals(FuzzyQuery.defaultPrefixLength, fq.getPrefixLength());
    fq = (FuzzyQuery) getQuery("term~", null);
    assertEquals(2, fq.getMaxEdits());
    assertEquals(FuzzyQuery.defaultPrefixLength, fq.getPrefixLength());


    assertQueryNodeException("term~1.1"); // value > 1, throws exception


    assertTrue(getQuery("term*germ", null) instanceof WildcardQuery);

View Full Code Here

        break;
      case WILDCARD:
        query = new WildcardQuery( new Term( fieldName, term ) );
        break;
      case FUZZY:
        query = new FuzzyQuery(
            new Term( fieldName, term ),
            termContext.getThreshold(),
            termContext.getPrefixLength()
        );
        break;

View Full Code Here

        break;
      case WILDCARD:
        query = new WildcardQuery( new Term( fieldName, term ) );
        break;
      case FUZZY:
        query = new FuzzyQuery(
            new Term( fieldName, term ),
            termContext.getThreshold(),
            termContext.getPrefixLength()
        );
        break;

View Full Code Here

        break;
      case WILDCARD:
        query = new WildcardQuery( new Term( fieldName, term ) );
        break;
      case FUZZY:
        query = new FuzzyQuery(
            new Term( fieldName, term ),
            termContext.getThreshold(),
            termContext.getPrefixLength()
        );
        break;

View Full Code Here

      @Override
      protected Analyzer getIndexAnalyzer(String field) {
        return analyzer;
      }
    };
    Query query = new FuzzyQuery(new Term("body", "tets"), 1);
    TopDocs topDocs = searcher.search(query, null, 10, Sort.INDEXORDER);
    assertEquals(2, topDocs.totalHits);
    String snippets[] = highlighter.highlight("body", query, searcher, topDocs);
    assertEquals(2, snippets.length);
    assertEquals("This is a <b>test</b>.", snippets[0]);
    assertEquals("<b>Test</b> a one sentence document.", snippets[1]);
    
    // with prefix
    query = new FuzzyQuery(new Term("body", "tets"), 1, 2);
    topDocs = searcher.search(query, null, 10, Sort.INDEXORDER);
    assertEquals(2, topDocs.totalHits);
    snippets = highlighter.highlight("body", query, searcher, topDocs);
    assertEquals(2, snippets.length);
    assertEquals("This is a <b>test</b>.", snippets[0]);
    assertEquals("<b>Test</b> a one sentence document.", snippets[1]);
    
    // wrong field
    BooleanQuery bq = new BooleanQuery();
    bq.add(new MatchAllDocsQuery(), BooleanClause.Occur.SHOULD);
    bq.add(new FuzzyQuery(new Term("bogus", "tets"), 1), BooleanClause.Occur.SHOULD);
    topDocs = searcher.search(bq, null, 10, Sort.INDEXORDER);
    assertEquals(2, topDocs.totalHits);
    snippets = highlighter.highlight("body", bq, searcher, topDocs);
    assertEquals(2, snippets.length);
    assertEquals("This is a test.", snippets[0]);

View Full Code Here

            return pq.toString();
          }
        });
      }
    } else if (query instanceof FuzzyQuery) {
      final FuzzyQuery fq = (FuzzyQuery) query;
      if (fq.getField().equals(field)) {
        String utf16 = fq.getTerm().text();
        int termText[] = new int[utf16.codePointCount(0, utf16.length())];
        for (int cp, i = 0, j = 0; i < utf16.length(); i += Character.charCount(cp)) {
          termText[j++] = cp = utf16.codePointAt(i);
        }
        int termLength = termText.length;
        int prefixLength = Math.min(fq.getPrefixLength(), termLength);
        String suffix = UnicodeUtil.newString(termText, prefixLength, termText.length - prefixLength);
        LevenshteinAutomata builder = new LevenshteinAutomata(suffix, fq.getTranspositions());
        Automaton automaton = builder.toAutomaton(fq.getMaxEdits());
        if (prefixLength > 0) {
          Automaton prefix = BasicAutomata.makeString(UnicodeUtil.newString(termText, 0, prefixLength));
          automaton = BasicOperations.concatenate(prefix, automaton);
        }
        list.add(new CharacterRunAutomaton(automaton) {
          @Override
          public String toString() {
            return fq.toString();
          }
        });
      }
    } else if (query instanceof TermRangeQuery) {
      final TermRangeQuery tq = (TermRangeQuery) query;

View Full Code Here

  }


  /** test a fuzzy query */
  public void testFuzzy() throws Exception {
    Query regular = new TermQuery(new Term("field", "foobar"));
    Query expected = new FuzzyQuery(new Term("field", "foobar"), 2);


    assertEquals(expected, parse("foobar~2"));
    assertEquals(regular, parse("foobar~"));
    assertEquals(regular, parse("foobar~a"));
    assertEquals(regular, parse("foobar~1a"));


    BooleanQuery bool = new BooleanQuery();
    FuzzyQuery fuzzy = new FuzzyQuery(new Term("field", "foo"), LevenshteinAutomata.MAXIMUM_SUPPORTED_DISTANCE);
    bool.add(fuzzy, Occur.MUST);
    bool.add(new TermQuery(new Term("field", "bar")), Occur.MUST);


    assertEquals(bool, parse("foo~" + LevenshteinAutomata.MAXIMUM_SUPPORTED_DISTANCE + 1 + " bar"));
  }

View Full Code Here

    SpanFirstQuery sfq = new SpanFirstQuery(swq, 3);
    assertEquals(1, searcher.search(sfq, 10).totalHits);
  }
  
  public void testFuzzy() throws Exception {
    FuzzyQuery fq = new FuzzyQuery(new Term("field", "broan"));
    SpanQuery sfq = new SpanMultiTermQueryWrapper<FuzzyQuery>(fq);
    // will not match quick brown fox
    SpanPositionRangeQuery sprq = new SpanPositionRangeQuery(sfq, 3, 6);
    assertEquals(2, searcher.search(sprq, 10).totalHits);
  }

View Full Code Here

    assertEquals(2, searcher.search(sprq, 10).totalHits);
  }
  
  public void testFuzzy2() throws Exception {
    // maximum of 1 term expansion
    FuzzyQuery fq = new FuzzyQuery(new Term("field", "broan"), 1, 0, 1, false);
    SpanQuery sfq = new SpanMultiTermQueryWrapper<FuzzyQuery>(fq);
    // will only match jumps over lazy broun dog
    SpanPositionRangeQuery sprq = new SpanPositionRangeQuery(sfq, 0, 100);
    assertEquals(1, searcher.search(sprq, 10).totalHits);
  }

View Full Code Here

    SpanPositionRangeQuery sprq = new SpanPositionRangeQuery(sfq, 0, 100);
    assertEquals(1, searcher.search(sprq, 10).totalHits);
  }
  public void testNoSuchMultiTermsInNear() throws Exception {
    //test to make sure non existent multiterms aren't throwing null pointer exceptions  
    FuzzyQuery fuzzyNoSuch = new FuzzyQuery(new Term("field", "noSuch"), 1, 0, 1, false);
    SpanQuery spanNoSuch = new SpanMultiTermQueryWrapper<FuzzyQuery>(fuzzyNoSuch);
    SpanQuery term = new SpanTermQuery(new Term("field", "brown"));
    SpanQuery near = new SpanNearQuery(new SpanQuery[]{term, spanNoSuch}, 1, true);
    assertEquals(0, searcher.search(near, 10).totalHits);
    //flip order

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of org.apache.lucene.search.FuzzyQuery

com.browseengine.bobo.test.ProtoBufConvertTest

com.esri.gpt.catalog.lucene.QueryProvider

com.mysema.query.lucene.LuceneExpressions

com.stratio.cassandra.index.query.FuzzyCondition

it.eng.spagobi.commons.utilities.indexing.LuceneSearcher

it.unibz.instasearch.indexing.querying.QueryFuzzifier

no.priv.garshol.duke.databases.LuceneDatabase

org.apache.blur.lucene.serializer.FuzzyQueryWritable

org.apache.cocoon.bean.query.SimpleLuceneCriterionBean

org.apache.lucene.index.SingleTermsEnum

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.