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

Example 11 with SpanTermQuery

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

the class SynonymTokenizer method testSpanHighlighting.

public void testSpanHighlighting() throws Exception {
    Query query1 = new SpanNearQuery(new SpanQuery[] { new SpanTermQuery(new Term(FIELD_NAME, "wordx")), new SpanTermQuery(new Term(FIELD_NAME, "wordy")) }, 1, false);
    Query query2 = new SpanNearQuery(new SpanQuery[] { new SpanTermQuery(new Term(FIELD_NAME, "wordy")), new SpanTermQuery(new Term(FIELD_NAME, "wordc")) }, 1, false);
    BooleanQuery.Builder bquery = new BooleanQuery.Builder();
    bquery.add(query1, Occur.SHOULD);
    bquery.add(query2, Occur.SHOULD);
    doSearching(bquery.build());
    TestHighlightRunner helper = new TestHighlightRunner() {

        @Override
        public void run() throws Exception {
            mode = QUERY;
            doStandardHighlights(analyzer, searcher, hits, query, HighlighterTest.this);
        }
    };
    helper.run();
    assertTrue("Failed to find correct number of highlights " + numHighlights + " found", numHighlights == 7);
}

Example 12 with SpanTermQuery

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

the class MissesTest method testSpanNearQuery.

public void testSpanNearQuery() throws IOException, InvalidTokenOffsetsException {
    try (Analyzer analyzer = new MockAnalyzer(random(), MockTokenizer.WHITESPACE, false)) {
        final Query query = new SpanNearQuery(new SpanQuery[] { new SpanTermQuery(new Term("test", "foo")), new SpanTermQuery(new Term("test", "bar")) }, 0, true);
        final Highlighter highlighter = new Highlighter(new SimpleHTMLFormatter(), new QueryScorer(query));
        assertEquals("this is a <B>foo</B> <B>bar</B> example", highlighter.getBestFragment(analyzer, "test", "this is a foo bar example"));
        assertNull(highlighter.getBestFragment(analyzer, "test", "this does not match"));
    }
}

Example 13 with SpanTermQuery

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

the class TokenSourcesTest method testOverlapWithOffset.

public void testOverlapWithOffset() throws IOException, InvalidTokenOffsetsException {
    final String TEXT = "the fox did not jump";
    final Directory directory = newDirectory();
    final IndexWriter indexWriter = new IndexWriter(directory, newIndexWriterConfig(null));
    try {
        final Document document = new Document();
        FieldType customType = new FieldType(TextField.TYPE_NOT_STORED);
        customType.setStoreTermVectors(true);
        // no positions!
        customType.setStoreTermVectorOffsets(true);
        document.add(new Field(FIELD, new OverlappingTokenStream(), customType));
        indexWriter.addDocument(document);
    } finally {
        indexWriter.close();
    }
    final IndexReader indexReader = DirectoryReader.open(directory);
    assertEquals(1, indexReader.numDocs());
    final IndexSearcher indexSearcher = newSearcher(indexReader);
    try {
        final DisjunctionMaxQuery query = new DisjunctionMaxQuery(Arrays.asList(new SpanTermQuery(new Term(FIELD, "{fox}")), new SpanTermQuery(new Term(FIELD, "fox"))), 1);
        // final Query phraseQuery = new SpanNearQuery(new SpanQuery[] {
        // new SpanTermQuery(new Term(FIELD, "{fox}")),
        // new SpanTermQuery(new Term(FIELD, "fox")) }, 0, true);
        TopDocs hits = indexSearcher.search(query, 1);
        assertEquals(1, hits.totalHits);
        final Highlighter highlighter = new Highlighter(new SimpleHTMLFormatter(), new SimpleHTMLEncoder(), new QueryScorer(query));
        final TokenStream tokenStream = TokenSources.getTermVectorTokenStreamOrNull(FIELD, indexReader.getTermVectors(0), -1);
        assertEquals("<B>the fox</B> did not jump", highlighter.getBestFragment(tokenStream, TEXT));
    } finally {
        indexReader.close();
        directory.close();
    }
}

Example 14 with SpanTermQuery

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

the class TokenSourcesTest method testOverlapWithOffsetExactPhrase.

public void testOverlapWithOffsetExactPhrase() throws IOException, InvalidTokenOffsetsException {
    final String TEXT = "the fox did not jump";
    final Directory directory = newDirectory();
    final IndexWriter indexWriter = new IndexWriter(directory, newIndexWriterConfig(null));
    try {
        final Document document = new Document();
        FieldType customType = new FieldType(TextField.TYPE_NOT_STORED);
        customType.setStoreTermVectors(true);
        // no positions!
        customType.setStoreTermVectorOffsets(true);
        document.add(new Field(FIELD, new OverlappingTokenStream(), customType));
        indexWriter.addDocument(document);
    } finally {
        indexWriter.close();
    }
    final IndexReader indexReader = DirectoryReader.open(directory);
    try {
        assertEquals(1, indexReader.numDocs());
        final IndexSearcher indexSearcher = newSearcher(indexReader);
        // final DisjunctionMaxQuery query = new DisjunctionMaxQuery(1);
        // query.add(new SpanTermQuery(new Term(FIELD, "{fox}")));
        // query.add(new SpanTermQuery(new Term(FIELD, "fox")));
        final Query phraseQuery = new SpanNearQuery(new SpanQuery[] { new SpanTermQuery(new Term(FIELD, "the")), new SpanTermQuery(new Term(FIELD, "fox")) }, 0, true);
        TopDocs hits = indexSearcher.search(phraseQuery, 1);
        assertEquals(1, 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("<B>the fox</B> did not jump", highlighter.getBestFragment(tokenStream, TEXT));
    } finally {
        indexReader.close();
        directory.close();
    }
}

Example 15 with SpanTermQuery

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

the class TokenSourcesTest method testOverlapWithPositionsAndOffset.

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