Examples with FieldType org.apache.lucene.document.FieldType used on opensource projects

Example 21 with FieldType

use of org.apache.lucene.document.FieldType in project lucene-solr by apache.

the class TokenSourcesTest method testMaxStartOffsetConsistency.

public void testMaxStartOffsetConsistency() throws IOException {
    FieldType tvFieldType = new FieldType(TextField.TYPE_NOT_STORED);
    tvFieldType.setStoreTermVectors(true);
    tvFieldType.setStoreTermVectorOffsets(true);
    tvFieldType.setStoreTermVectorPositions(true);
    Directory dir = newDirectory();
    MockAnalyzer analyzer = new MockAnalyzer(random());
    //we don't necessarily consume the whole stream because of limiting by startOffset
    analyzer.setEnableChecks(false);
    Document doc = new Document();
    final String TEXT = " f gg h";
    doc.add(new Field("fld_tv", analyzer.tokenStream("fooFld", TEXT), tvFieldType));
    doc.add(new TextField("fld_notv", analyzer.tokenStream("barFld", TEXT)));
    IndexReader reader;
    try (RandomIndexWriter writer = new RandomIndexWriter(random(), dir)) {
        writer.addDocument(doc);
        reader = writer.getReader();
    }
    try {
        Fields tvFields = reader.getTermVectors(0);
        for (int maxStartOffset = -1; maxStartOffset <= TEXT.length(); maxStartOffset++) {
            TokenStream tvStream = TokenSources.getTokenStream("fld_tv", tvFields, TEXT, analyzer, maxStartOffset);
            TokenStream anaStream = TokenSources.getTokenStream("fld_notv", tvFields, TEXT, analyzer, maxStartOffset);
            //assert have same tokens, none of which has a start offset > maxStartOffset
            final OffsetAttribute tvOffAtt = tvStream.addAttribute(OffsetAttribute.class);
            final OffsetAttribute anaOffAtt = anaStream.addAttribute(OffsetAttribute.class);
            tvStream.reset();
            anaStream.reset();
            while (tvStream.incrementToken()) {
                assertTrue(anaStream.incrementToken());
                assertEquals(tvOffAtt.startOffset(), anaOffAtt.startOffset());
                if (maxStartOffset >= 0)
                    assertTrue(tvOffAtt.startOffset() <= maxStartOffset);
            }
            assertTrue(anaStream.incrementToken() == false);
            tvStream.end();
            anaStream.end();
            tvStream.close();
            anaStream.close();
        }
    } finally {
        reader.close();
    }
    dir.close();
}

Example 22 with FieldType

use of org.apache.lucene.document.FieldType 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 23 with FieldType

use of org.apache.lucene.document.FieldType 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 24 with FieldType

use of org.apache.lucene.document.FieldType in project lucene-solr by apache.

the class TokenSourcesTest method testPayloads.

// LUCENE-5294
public void testPayloads() throws Exception {
    Directory dir = newDirectory();
    RandomIndexWriter writer = new RandomIndexWriter(random(), dir);
    FieldType myFieldType = new FieldType(TextField.TYPE_NOT_STORED);
    myFieldType.setStoreTermVectors(true);
    myFieldType.setStoreTermVectorOffsets(true);
    myFieldType.setStoreTermVectorPositions(true);
    myFieldType.setStoreTermVectorPayloads(true);
    curOffset = 0;
    Token[] tokens = new Token[] { getToken("foxes"), getToken("can"), getToken("jump"), getToken("high") };
    Document doc = new Document();
    doc.add(new Field("field", new CannedTokenStream(tokens), myFieldType));
    writer.addDocument(doc);
    IndexReader reader = writer.getReader();
    writer.close();
    assertEquals(1, reader.numDocs());
    TokenStream ts = TokenSources.getTermVectorTokenStreamOrNull("field", reader.getTermVectors(0), -1);
    CharTermAttribute termAtt = ts.getAttribute(CharTermAttribute.class);
    PositionIncrementAttribute posIncAtt = ts.getAttribute(PositionIncrementAttribute.class);
    OffsetAttribute offsetAtt = ts.getAttribute(OffsetAttribute.class);
    PayloadAttribute payloadAtt = ts.addAttribute(PayloadAttribute.class);
    ts.reset();
    for (Token token : tokens) {
        assertTrue(ts.incrementToken());
        assertEquals(token.toString(), termAtt.toString());
        assertEquals(token.getPositionIncrement(), posIncAtt.getPositionIncrement());
        assertEquals(token.getPayload(), payloadAtt.getPayload());
        assertEquals(token.startOffset(), offsetAtt.startOffset());
        assertEquals(token.endOffset(), offsetAtt.endOffset());
    }
    assertFalse(ts.incrementToken());
    reader.close();
    dir.close();
}

Example 25 with FieldType

use of org.apache.lucene.document.FieldType 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();
    }
}