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

Example 76 with SpanTermQuery

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

the class SynonymTokenizer method testPayloadQuery.

/** We can highlight based on payloads. It's supported both via term vectors and MemoryIndex since Lucene 5. */
public void testPayloadQuery() throws IOException, InvalidTokenOffsetsException {
    //"words" at positions 1 & 4
    final String text = "random words and words";
    //sets payload to "pos: X" (where X is position #)
    Analyzer analyzer = new MockPayloadAnalyzer();
    try (IndexWriter writer = new IndexWriter(dir, new IndexWriterConfig(analyzer))) {
        writer.deleteAll();
        Document doc = new Document();
        doc.add(new Field(FIELD_NAME, text, fieldType));
        writer.addDocument(doc);
        writer.commit();
    }
    try (IndexReader reader = DirectoryReader.open(dir)) {
        Query query = new SpanPayloadCheckQuery(new SpanTermQuery(new Term(FIELD_NAME, "words")), //just match the first "word" occurrence
        Collections.singletonList(new BytesRef("pos: 1")));
        IndexSearcher searcher = newSearcher(reader);
        QueryScorer scorer = new QueryScorer(query, searcher.getIndexReader(), FIELD_NAME);
        scorer.setUsePayloads(true);
        Highlighter h = new Highlighter(scorer);
        TopDocs hits = searcher.search(query, 10);
        assertEquals(1, hits.scoreDocs.length);
        TokenStream stream = TokenSources.getAnyTokenStream(searcher.getIndexReader(), 0, FIELD_NAME, analyzer);
        if (random().nextBoolean()) {
            //conceals detection of TokenStreamFromTermVector
            stream = new CachingTokenFilter(stream);
        }
        String result = h.getBestFragment(stream, text);
        //only highlight first "word"
        assertEquals("random <B>words</B> and words", result);
    }
}

Example 77 with SpanTermQuery

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

the class TestSolrCoreParser method checkApacheLuceneSolr.

private static void checkApacheLuceneSolr(Query query, String fieldName) {
    assertTrue(query instanceof SpanNearQuery);
    final SpanNearQuery snq = (SpanNearQuery) query;
    assertEquals(fieldName, snq.getField());
    assertEquals(42, snq.getSlop());
    assertFalse(snq.isInOrder());
    assertEquals(3, snq.getClauses().length);
    assertTrue(snq.getClauses()[0] instanceof SpanTermQuery);
    assertTrue(snq.getClauses()[1] instanceof SpanTermQuery);
    assertTrue(snq.getClauses()[2] instanceof SpanTermQuery);
}

Example 78 with SpanTermQuery

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

the class TestSolrCoreParser method checkChooseOneWordQuery.

private static void checkChooseOneWordQuery(boolean span, Query query, String fieldName, String... expectedTermTexts) {
    final Term term;
    if (span) {
        assertTrue(query instanceof SpanTermQuery);
        final SpanTermQuery stq = (SpanTermQuery) query;
        term = stq.getTerm();
    } else {
        assertTrue(query instanceof TermQuery);
        final TermQuery tq = (TermQuery) query;
        term = tq.getTerm();
    }
    final String text = term.text();
    boolean foundExpected = false;
    for (String expected : expectedTermTexts) {
        foundExpected |= expected.equals(text);
    }
    assertEquals(fieldName, term.field());
    assertTrue("expected term text (" + text + ") not found in (" + expectedTermTexts + ")", foundExpected);
}

Example 79 with SpanTermQuery

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

the class TestPositionIncrement method testPayloadsPos0.

public void testPayloadsPos0() throws Exception {
    Directory dir = newDirectory();
    RandomIndexWriter writer = new RandomIndexWriter(random(), dir, new MockPayloadAnalyzer());
    Document doc = new Document();
    doc.add(new TextField("content", new StringReader("a a b c d e a f g h i j a b k k")));
    writer.addDocument(doc);
    final IndexReader readerFromWriter = writer.getReader();
    LeafReader r = getOnlyLeafReader(readerFromWriter);
    PostingsEnum tp = r.postings(new Term("content", "a"), PostingsEnum.ALL);
    int count = 0;
    assertTrue(tp.nextDoc() != DocIdSetIterator.NO_MORE_DOCS);
    // "a" occurs 4 times
    assertEquals(4, tp.freq());
    assertEquals(0, tp.nextPosition());
    assertEquals(1, tp.nextPosition());
    assertEquals(3, tp.nextPosition());
    assertEquals(6, tp.nextPosition());
    // only one doc has "a"
    assertEquals(DocIdSetIterator.NO_MORE_DOCS, tp.nextDoc());
    IndexSearcher is = newSearcher(getOnlyLeafReader(readerFromWriter));
    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, 30, false);
    count = 0;
    boolean sawZero = false;
    if (VERBOSE) {
        System.out.println("\ngetPayloadSpans test");
    }
    PayloadSpanCollector collector = new PayloadSpanCollector();
    Spans pspans = snq.createWeight(is, false, 1f).getSpans(is.getIndexReader().leaves().get(0), SpanWeight.Postings.PAYLOADS);
    while (pspans.nextDoc() != Spans.NO_MORE_DOCS) {
        while (pspans.nextStartPosition() != Spans.NO_MORE_POSITIONS) {
            if (VERBOSE) {
                System.out.println("doc " + pspans.docID() + ": span " + pspans.startPosition() + " to " + pspans.endPosition());
            }
            collector.reset();
            pspans.collect(collector);
            sawZero |= pspans.startPosition() == 0;
            for (BytesRef payload : collector.payloads) {
                count++;
                if (VERBOSE) {
                    System.out.println("  payload: " + Term.toString(payload));
                }
            }
        }
    }
    assertTrue(sawZero);
    assertEquals(8, count);
    // System.out.println("\ngetSpans test");
    Spans spans = snq.createWeight(is, false, 1f).getSpans(is.getIndexReader().leaves().get(0), SpanWeight.Postings.POSITIONS);
    count = 0;
    sawZero = false;
    while (spans.nextDoc() != Spans.NO_MORE_DOCS) {
        while (spans.nextStartPosition() != Spans.NO_MORE_POSITIONS) {
            count++;
            sawZero |= spans.startPosition() == 0;
        // System.out.println(spans.doc() + " - " + spans.start() + " - " +
        // spans.end());
        }
    }
    assertEquals(4, count);
    assertTrue(sawZero);
    writer.close();
    is.getIndexReader().close();
    dir.close();
}

Example 80 with SpanTermQuery

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

the class TestPayloadExplanations method testSimpleTerm.

public void testSimpleTerm() throws Exception {
    SpanTermQuery q = new SpanTermQuery(new Term(FIELD, "w2"));
    testAllFunctions(q, new int[] { 0, 1, 2, 3 });
}