Examples with SpanQuery org.apache.lucene.search.spans.SpanQuery used on opensource projects

Example 61 with SpanQuery

use of org.apache.lucene.search.spans.SpanQuery in project lucene-solr by apache.

the class TestPayloadSpans method testNestedSpans.

public void testNestedSpans() throws Exception {
    SpanTermQuery stq;
    Spans spans;
    IndexSearcher searcher = getSearcher();
    stq = new SpanTermQuery(new Term(PayloadHelper.FIELD, "mark"));
    spans = stq.createWeight(searcher, false, 1f).getSpans(searcher.getIndexReader().leaves().get(0), SpanWeight.Postings.PAYLOADS);
    assertNull(spans);
    SpanQuery[] clauses = new SpanQuery[3];
    clauses[0] = new SpanTermQuery(new Term(PayloadHelper.FIELD, "rr"));
    clauses[1] = new SpanTermQuery(new Term(PayloadHelper.FIELD, "yy"));
    clauses[2] = new SpanTermQuery(new Term(PayloadHelper.FIELD, "xx"));
    SpanNearQuery spanNearQuery = new SpanNearQuery(clauses, 12, false);
    spans = spanNearQuery.createWeight(searcher, false, 1f).getSpans(searcher.getIndexReader().leaves().get(0), SpanWeight.Postings.PAYLOADS);
    assertTrue("spans is null and it shouldn't be", spans != null);
    checkSpans(spans, 2, new int[] { 3, 3 });
    clauses[0] = new SpanTermQuery(new Term(PayloadHelper.FIELD, "xx"));
    clauses[1] = new SpanTermQuery(new Term(PayloadHelper.FIELD, "rr"));
    clauses[2] = new SpanTermQuery(new Term(PayloadHelper.FIELD, "yy"));
    spanNearQuery = new SpanNearQuery(clauses, 6, true);
    spans = spanNearQuery.createWeight(searcher, false, 1f).getSpans(searcher.getIndexReader().leaves().get(0), SpanWeight.Postings.PAYLOADS);
    assertTrue("spans is null and it shouldn't be", spans != null);
    checkSpans(spans, 1, new int[] { 3 });
    clauses = new SpanQuery[2];
    clauses[0] = new SpanTermQuery(new Term(PayloadHelper.FIELD, "xx"));
    clauses[1] = new SpanTermQuery(new Term(PayloadHelper.FIELD, "rr"));
    spanNearQuery = new SpanNearQuery(clauses, 6, true);
    // xx within 6 of rr
    SpanQuery[] clauses2 = new SpanQuery[2];
    clauses2[0] = new SpanTermQuery(new Term(PayloadHelper.FIELD, "yy"));
    clauses2[1] = spanNearQuery;
    SpanNearQuery nestedSpanNearQuery = new SpanNearQuery(clauses2, 6, false);
    // yy within 6 of xx within 6 of rr
    spans = nestedSpanNearQuery.createWeight(searcher, false, 1f).getSpans(searcher.getIndexReader().leaves().get(0), SpanWeight.Postings.PAYLOADS);
    assertTrue("spans is null and it shouldn't be", spans != null);
    checkSpans(spans, 2, new int[] { 3, 3 });
    closeIndexReader.close();
    directory.close();
}

Example 62 with SpanQuery

use of org.apache.lucene.search.spans.SpanQuery in project lucene-solr by apache.

the class TestPayloadSpans method testShrinkToAfterShortestMatch3.

public void testShrinkToAfterShortestMatch3() throws IOException {
    Directory directory = newDirectory();
    RandomIndexWriter writer = new RandomIndexWriter(random(), directory, newIndexWriterConfig(new TestPayloadAnalyzer()));
    Document doc = new Document();
    doc.add(new TextField("content", new StringReader("j k a l f k k p a t a k l k t a")));
    writer.addDocument(doc);
    IndexReader reader = writer.getReader();
    IndexSearcher is = newSearcher(getOnlyLeafReader(reader), false);
    writer.close();
    SpanTermQuery stq1 = new SpanTermQuery(new Term("content", "a"));
    SpanTermQuery stq2 = new SpanTermQuery(new Term("content", "k"));
    SpanQuery[] sqs = { stq1, stq2 };
    SpanNearQuery snq = new SpanNearQuery(sqs, 0, true);
    Spans spans = snq.createWeight(is, false, 1f).getSpans(is.getIndexReader().leaves().get(0), SpanWeight.Postings.PAYLOADS);
    TopDocs topDocs = is.search(snq, 1);
    Set<String> payloadSet = new HashSet<>();
    VerifyingCollector collector = new VerifyingCollector();
    for (int i = 0; i < topDocs.scoreDocs.length; i++) {
        while (spans.nextDoc() != Spans.NO_MORE_DOCS) {
            while (spans.nextStartPosition() != Spans.NO_MORE_POSITIONS) {
                collector.reset();
                spans.collect(collector);
                for (final BytesRef payload : collector.payloads) {
                    payloadSet.add(Term.toString(payload));
                }
            }
        }
    }
    assertEquals(2, payloadSet.size());
    if (VERBOSE) {
        for (final String payload : payloadSet) System.out.println("match:" + payload);
    }
    assertTrue(payloadSet.contains("a:Noise:10"));
    assertTrue(payloadSet.contains("k:Noise:11"));
    reader.close();
    directory.close();
}

Example 63 with SpanQuery

use of org.apache.lucene.search.spans.SpanQuery in project lucene-solr by apache.

the class TestPayloadSpans method testSpanNot.

public void testSpanNot() throws Exception {
    SpanQuery[] clauses = new SpanQuery[2];
    clauses[0] = new SpanTermQuery(new Term(PayloadHelper.FIELD, "one"));
    clauses[1] = new SpanTermQuery(new Term(PayloadHelper.FIELD, "three"));
    SpanQuery spq = new SpanNearQuery(clauses, 5, true);
    SpanNotQuery snq = new SpanNotQuery(spq, new SpanTermQuery(new Term(PayloadHelper.FIELD, "two")));
    Directory directory = newDirectory();
    RandomIndexWriter writer = new RandomIndexWriter(random(), directory, newIndexWriterConfig(new PayloadAnalyzer()).setSimilarity(similarity));
    Document doc = new Document();
    doc.add(newTextField(PayloadHelper.FIELD, "one two three one four three", Field.Store.YES));
    writer.addDocument(doc);
    IndexReader reader = getOnlyLeafReader(writer.getReader());
    writer.close();
    checkSpans(snq.createWeight(newSearcher(reader, false), false, 1f).getSpans(reader.leaves().get(0), SpanWeight.Postings.PAYLOADS), 1, new int[] { 2 });
    reader.close();
    directory.close();
}

Example 64 with SpanQuery

use of org.apache.lucene.search.spans.SpanQuery in project lucene-solr by apache.

the class TestPayloadSpans method testSpanFirst.

public void testSpanFirst() throws IOException {
    SpanQuery match;
    SpanFirstQuery sfq;
    match = new SpanTermQuery(new Term(PayloadHelper.FIELD, "one"));
    sfq = new SpanFirstQuery(match, 2);
    Spans spans = sfq.createWeight(searcher, false, 1f).getSpans(searcher.getIndexReader().leaves().get(0), SpanWeight.Postings.PAYLOADS);
    checkSpans(spans, 109, 1, 1, 1);
    //Test more complicated subclause
    SpanQuery[] clauses = new SpanQuery[2];
    clauses[0] = new SpanTermQuery(new Term(PayloadHelper.FIELD, "one"));
    clauses[1] = new SpanTermQuery(new Term(PayloadHelper.FIELD, "hundred"));
    match = new SpanNearQuery(clauses, 0, true);
    sfq = new SpanFirstQuery(match, 2);
    checkSpans(sfq.createWeight(searcher, false, 1f).getSpans(searcher.getIndexReader().leaves().get(0), SpanWeight.Postings.PAYLOADS), 100, 2, 1, 1);
    match = new SpanNearQuery(clauses, 0, false);
    sfq = new SpanFirstQuery(match, 2);
    checkSpans(sfq.createWeight(searcher, false, 1f).getSpans(searcher.getIndexReader().leaves().get(0), SpanWeight.Postings.PAYLOADS), 100, 2, 1, 1);
}

Example 65 with SpanQuery

use of org.apache.lucene.search.spans.SpanQuery in project lucene-solr by apache.

the class HighlighterPhraseTest method testSparseSpan.

public void testSparseSpan() throws IOException, InvalidTokenOffsetsException {
    final String TEXT = "the fox did not jump";
    final Directory directory = newDirectory();
    final IndexWriter indexWriter = new IndexWriter(directory, newIndexWriterConfig(new MockAnalyzer(random(), MockTokenizer.WHITESPACE, false)));
    try {
        final Document document = new Document();
        FieldType customType = new FieldType(TextField.TYPE_NOT_STORED);
        customType.setStoreTermVectorOffsets(true);
        customType.setStoreTermVectorPositions(true);
        customType.setStoreTermVectors(true);
        document.add(new Field(FIELD, new TokenStreamSparse(), customType));
        indexWriter.addDocument(document);
    } finally {
        indexWriter.close();
    }
    final IndexReader indexReader = DirectoryReader.open(directory);
    try {
        assertEquals(1, indexReader.numDocs());
        final IndexSearcher indexSearcher = newSearcher(indexReader);
        final Query phraseQuery = new SpanNearQuery(new SpanQuery[] { new SpanTermQuery(new Term(FIELD, "did")), new SpanTermQuery(new Term(FIELD, "jump")) }, 0, true);
        TopDocs hits = indexSearcher.search(phraseQuery, 1);
        assertEquals(0, hits.totalHits);
        final Highlighter highlighter = new Highlighter(new SimpleHTMLFormatter(), new SimpleHTMLEncoder(), new QueryScorer(phraseQuery));
        final TokenStream tokenStream = TokenSources.getTermVectorTokenStreamOrNull(FIELD, indexReader.getTermVectors(0), -1);
        assertEquals(highlighter.getBestFragment(new TokenStreamSparse(), TEXT), highlighter.getBestFragment(tokenStream, TEXT));
    } finally {
        indexReader.close();
        directory.close();
    }
}