Java Code Examples for org.apache.lucene.search.Query#visit()

public void testNonExistingField() throws Throwable {
  FuzzyLikeThisQuery flt = new FuzzyLikeThisQuery(10, analyzer);
  flt.addTerms("jonathin smoth", "name", 2, 1);
  flt.addTerms("jonathin smoth", "this field does not exist", 2, 1);
  // don't fail here just because the field doesn't exits
  Query q = flt.rewrite(searcher.getIndexReader());
  HashSet<Term> queryTerms = new HashSet<>();
  q.visit(QueryVisitor.termCollector(queryTerms));
  assertTrue("Should have variant jonathan", queryTerms.contains(new Term("name", "jonathan")));
  assertTrue("Should have variant smith", queryTerms.contains(new Term("name", "smith")));
  TopDocs topDocs = searcher.search(flt, 1);
  ScoreDoc[] sd = topDocs.scoreDocs;
  assertTrue("score docs must match 1 doc", (sd != null) && (sd.length > 0));
  Document doc = searcher.doc(sd[0].doc);
  assertEquals("Should match most similar when using 2 words", "2", doc.get("id"));
}
 

public void testClosestEditDistanceMatchComesFirst() throws Throwable {
  FuzzyLikeThisQuery flt = new FuzzyLikeThisQuery(10, analyzer);
  flt.addTerms("smith", "name", 2, 1);
  Query q = flt.rewrite(searcher.getIndexReader());
  HashSet<Term> queryTerms = new HashSet<>();
  q.visit(QueryVisitor.termCollector(queryTerms));
  assertTrue("Should have variant smythe", queryTerms.contains(new Term("name", "smythe")));
  assertTrue("Should have variant smith", queryTerms.contains(new Term("name", "smith")));
  assertTrue("Should have variant smyth", queryTerms.contains(new Term("name", "smyth")));
  TopDocs topDocs = searcher.search(flt, 1);
  ScoreDoc[] sd = topDocs.scoreDocs;
  assertTrue("score docs must match 1 doc", (sd != null) && (sd.length > 0));
  Document doc = searcher.doc(sd[0].doc);
  assertEquals("Should match most similar not most rare variant", "2", doc.get("id"));
}
 

public void testMultiWord() throws Throwable {
  FuzzyLikeThisQuery flt = new FuzzyLikeThisQuery(10, analyzer);
  flt.addTerms("jonathin smoth", "name", 2, 1);
  Query q = flt.rewrite(searcher.getIndexReader());
  HashSet<Term> queryTerms = new HashSet<>();
  q.visit(QueryVisitor.termCollector(queryTerms));
  assertTrue("Should have variant jonathan", queryTerms.contains(new Term("name", "jonathan")));
  assertTrue("Should have variant smith", queryTerms.contains(new Term("name", "smith")));
  TopDocs topDocs = searcher.search(flt, 1);
  ScoreDoc[] sd = topDocs.scoreDocs;
  assertTrue("score docs must match 1 doc", (sd != null) && (sd.length > 0));
  Document doc = searcher.doc(sd[0].doc);
  assertEquals("Should match most similar when using 2 words", "2", doc.get("id"));
}
 

public void testNoMatchFirstWordBug() throws Throwable {
  FuzzyLikeThisQuery flt = new FuzzyLikeThisQuery(10, analyzer);
  flt.addTerms("fernando smith", "name", 2, 1);
  Query q = flt.rewrite(searcher.getIndexReader());
  HashSet<Term> queryTerms = new HashSet<>();
  q.visit(QueryVisitor.termCollector(queryTerms));
  assertTrue("Should have variant smith", queryTerms.contains(new Term("name", "smith")));
  TopDocs topDocs = searcher.search(flt, 1);
  ScoreDoc[] sd = topDocs.scoreDocs;
  assertTrue("score docs must match 1 doc", (sd != null) && (sd.length > 0));
  Document doc = searcher.doc(sd[0].doc);
  assertEquals("Should match most similar when using 2 words", "2", doc.get("id"));
}
 

/**
 * Returns the size in bytes of a Query object. Unknown query types will be estimated
 * using {@link #shallowSizeOf(Object)}, or using the supplied <code>defSize</code> parameter
 * if its value is greater than 0.
 */
public static long sizeOf(Query q, long defSize) {
  if (q instanceof Accountable) {
    return ((Accountable)q).ramBytesUsed();
  } else {
    RamUsageQueryVisitor visitor = new RamUsageQueryVisitor(q, defSize);
    q.visit(visitor);
    return alignObjectSize(visitor.total);
  }
}
 

public void testRewrite2() throws Exception {
  SpanQuery q1 = new SpanTermQuery(new Term("last", "smith"));
  SpanQuery q2 = new SpanTermQuery(new Term("last", "jones"));
  SpanQuery q = new SpanNearQuery(new SpanQuery[]
    { q1, new FieldMaskingSpanQuery(q2, "last")}, 1, true );
  Query qr = searcher.rewrite(q);

  QueryUtils.checkEqual(q, qr);

  HashSet<Term> set = new HashSet<>();
  qr.visit(QueryVisitor.termCollector(set));
  assertEquals(2, set.size());
}
 

/**
 * Extracts MultiTermQueries that match the provided field predicate.
 * Returns equivalent automata that will match terms.
 */
static LabelledCharArrayMatcher[] extractAutomata(Query query, Predicate<String> fieldMatcher, boolean lookInSpan) {
  AutomataCollector collector = new AutomataCollector(lookInSpan, fieldMatcher);
  query.visit(collector);
  return collector.runAutomata.toArray(new LabelledCharArrayMatcher[0]);
}
 

/**
 * Extracts all terms texts of a given Query into an array of WeightedTerms
 *
 * @param query      Query to extract term texts from
 * @param prohibited <code>true</code> to extract "prohibited" terms, too
 * @param fieldName  The fieldName used to filter query terms
 * @return an array of the terms used in a query, plus their weights.
 */
public static WeightedTerm[] getTerms(Query query, boolean prohibited, String fieldName) {
  HashSet<WeightedTerm> terms = new HashSet<>();
  Predicate<String> fieldSelector = fieldName == null ? f -> true : fieldName::equals;
  query.visit(new BoostedTermExtractor(1, terms, prohibited, fieldSelector));
  return terms.toArray(new WeightedTerm[0]);
}
 

/**
 * Create a {@link QueryTree} from a passed in Query or Filter
 *
 * @param luceneQuery the query to analyze
 * @return a QueryTree describing the analyzed query
 */
QueryTree buildTree(Query luceneQuery, TermWeightor weightor) {
  QueryBuilder builder = new QueryBuilder();
  luceneQuery.visit(builder);
  return builder.apply(weightor);
}