Examples with TermVectorsReader org.apache.lucene.codecs.TermVectorsReader used on opensource projects

Example 6 with TermVectorsReader

use of org.apache.lucene.codecs.TermVectorsReader in project lucene-solr by apache.

the class BaseIndexFileFormatTestCase method testMultiClose.

/** Calls close multiple times on closeable codec apis */
public void testMultiClose() throws IOException {
    // first make a one doc index
    Directory oneDocIndex = applyCreatedVersionMajor(newDirectory());
    IndexWriter iw = new IndexWriter(oneDocIndex, new IndexWriterConfig(new MockAnalyzer(random())));
    Document oneDoc = new Document();
    FieldType customType = new FieldType(TextField.TYPE_STORED);
    customType.setStoreTermVectors(true);
    Field customField = new Field("field", "contents", customType);
    oneDoc.add(customField);
    oneDoc.add(new NumericDocValuesField("field", 5));
    iw.addDocument(oneDoc);
    LeafReader oneDocReader = getOnlyLeafReader(DirectoryReader.open(iw));
    iw.close();
    // now feed to codec apis manually
    // we use FSDir, things like ramdir are not guaranteed to cause fails if you write to them after close(), etc
    Directory dir = newFSDirectory(createTempDir("justSoYouGetSomeChannelErrors"));
    Codec codec = getCodec();
    SegmentInfo segmentInfo = new SegmentInfo(dir, Version.LATEST, Version.LATEST, "_0", 1, false, codec, Collections.emptyMap(), StringHelper.randomId(), new HashMap<>(), null);
    FieldInfo proto = oneDocReader.getFieldInfos().fieldInfo("field");
    FieldInfo field = new FieldInfo(proto.name, proto.number, proto.hasVectors(), proto.omitsNorms(), proto.hasPayloads(), proto.getIndexOptions(), proto.getDocValuesType(), proto.getDocValuesGen(), new HashMap<>(), proto.getPointDimensionCount(), proto.getPointNumBytes());
    FieldInfos fieldInfos = new FieldInfos(new FieldInfo[] { field });
    SegmentWriteState writeState = new SegmentWriteState(null, dir, segmentInfo, fieldInfos, null, new IOContext(new FlushInfo(1, 20)));
    SegmentReadState readState = new SegmentReadState(dir, segmentInfo, fieldInfos, IOContext.READ);
    // PostingsFormat
    try (FieldsConsumer consumer = codec.postingsFormat().fieldsConsumer(writeState)) {
        consumer.write(oneDocReader.fields());
        IOUtils.close(consumer);
        IOUtils.close(consumer);
    }
    try (FieldsProducer producer = codec.postingsFormat().fieldsProducer(readState)) {
        IOUtils.close(producer);
        IOUtils.close(producer);
    }
    // DocValuesFormat
    try (DocValuesConsumer consumer = codec.docValuesFormat().fieldsConsumer(writeState)) {
        consumer.addNumericField(field, new EmptyDocValuesProducer() {

            @Override
            public NumericDocValues getNumeric(FieldInfo field) {
                return new NumericDocValues() {

                    int docID = -1;

                    @Override
                    public int docID() {
                        return docID;
                    }

                    @Override
                    public int nextDoc() {
                        docID++;
                        if (docID == 1) {
                            docID = NO_MORE_DOCS;
                        }
                        return docID;
                    }

                    @Override
                    public int advance(int target) {
                        if (docID <= 0 && target == 0) {
                            docID = 0;
                        } else {
                            docID = NO_MORE_DOCS;
                        }
                        return docID;
                    }

                    @Override
                    public boolean advanceExact(int target) throws IOException {
                        docID = target;
                        return target == 0;
                    }

                    @Override
                    public long cost() {
                        return 1;
                    }

                    @Override
                    public long longValue() {
                        return 5;
                    }
                };
            }
        });
        IOUtils.close(consumer);
        IOUtils.close(consumer);
    }
    try (DocValuesProducer producer = codec.docValuesFormat().fieldsProducer(readState)) {
        IOUtils.close(producer);
        IOUtils.close(producer);
    }
    // NormsFormat
    try (NormsConsumer consumer = codec.normsFormat().normsConsumer(writeState)) {
        consumer.addNormsField(field, new NormsProducer() {

            @Override
            public NumericDocValues getNorms(FieldInfo field) {
                return new NumericDocValues() {

                    int docID = -1;

                    @Override
                    public int docID() {
                        return docID;
                    }

                    @Override
                    public int nextDoc() {
                        docID++;
                        if (docID == 1) {
                            docID = NO_MORE_DOCS;
                        }
                        return docID;
                    }

                    @Override
                    public int advance(int target) {
                        if (docID <= 0 && target == 0) {
                            docID = 0;
                        } else {
                            docID = NO_MORE_DOCS;
                        }
                        return docID;
                    }

                    @Override
                    public boolean advanceExact(int target) throws IOException {
                        docID = target;
                        return target == 0;
                    }

                    @Override
                    public long cost() {
                        return 1;
                    }

                    @Override
                    public long longValue() {
                        return 5;
                    }
                };
            }

            @Override
            public void checkIntegrity() {
            }

            @Override
            public void close() {
            }

            @Override
            public long ramBytesUsed() {
                return 0;
            }
        });
        IOUtils.close(consumer);
        IOUtils.close(consumer);
    }
    try (NormsProducer producer = codec.normsFormat().normsProducer(readState)) {
        IOUtils.close(producer);
        IOUtils.close(producer);
    }
    // TermVectorsFormat
    try (TermVectorsWriter consumer = codec.termVectorsFormat().vectorsWriter(dir, segmentInfo, writeState.context)) {
        consumer.startDocument(1);
        consumer.startField(field, 1, false, false, false);
        consumer.startTerm(new BytesRef("testing"), 2);
        consumer.finishTerm();
        consumer.finishField();
        consumer.finishDocument();
        consumer.finish(fieldInfos, 1);
        IOUtils.close(consumer);
        IOUtils.close(consumer);
    }
    try (TermVectorsReader producer = codec.termVectorsFormat().vectorsReader(dir, segmentInfo, fieldInfos, readState.context)) {
        IOUtils.close(producer);
        IOUtils.close(producer);
    }
    // StoredFieldsFormat
    try (StoredFieldsWriter consumer = codec.storedFieldsFormat().fieldsWriter(dir, segmentInfo, writeState.context)) {
        consumer.startDocument();
        consumer.writeField(field, customField);
        consumer.finishDocument();
        consumer.finish(fieldInfos, 1);
        IOUtils.close(consumer);
        IOUtils.close(consumer);
    }
    try (StoredFieldsReader producer = codec.storedFieldsFormat().fieldsReader(dir, segmentInfo, fieldInfos, readState.context)) {
        IOUtils.close(producer);
        IOUtils.close(producer);
    }
    IOUtils.close(oneDocReader, oneDocIndex, dir);
}

Example 7 with TermVectorsReader

use of org.apache.lucene.codecs.TermVectorsReader in project lucene-solr by apache.

the class TestTermVectorsReader method testDocsEnum.

public void testDocsEnum() throws IOException {
    TermVectorsReader reader = Codec.getDefault().termVectorsFormat().vectorsReader(dir, seg.info, fieldInfos, newIOContext(random()));
    for (int j = 0; j < 5; j++) {
        Terms vector = reader.get(j).terms(testFields[0]);
        assertNotNull(vector);
        assertEquals(testTerms.length, vector.size());
        TermsEnum termsEnum = vector.iterator();
        PostingsEnum postingsEnum = null;
        for (int i = 0; i < testTerms.length; i++) {
            final BytesRef text = termsEnum.next();
            assertNotNull(text);
            String term = text.utf8ToString();
            //System.out.println("Term: " + term);
            assertEquals(testTerms[i], term);
            postingsEnum = TestUtil.docs(random(), termsEnum, postingsEnum, PostingsEnum.NONE);
            assertNotNull(postingsEnum);
            int doc = postingsEnum.docID();
            assertEquals(-1, doc);
            assertTrue(postingsEnum.nextDoc() != DocIdSetIterator.NO_MORE_DOCS);
            assertEquals(DocIdSetIterator.NO_MORE_DOCS, postingsEnum.nextDoc());
        }
        assertNull(termsEnum.next());
    }
    reader.close();
}

Example 8 with TermVectorsReader

use of org.apache.lucene.codecs.TermVectorsReader in project lucene-solr by apache.

the class TestTermVectorsReader method testPositionReader.

public void testPositionReader() throws IOException {
    TermVectorsReader reader = Codec.getDefault().termVectorsFormat().vectorsReader(dir, seg.info, fieldInfos, newIOContext(random()));
    BytesRef[] terms;
    Terms vector = reader.get(0).terms(testFields[0]);
    assertNotNull(vector);
    assertEquals(testTerms.length, vector.size());
    TermsEnum termsEnum = vector.iterator();
    PostingsEnum dpEnum = null;
    for (int i = 0; i < testTerms.length; i++) {
        final BytesRef text = termsEnum.next();
        assertNotNull(text);
        String term = text.utf8ToString();
        //System.out.println("Term: " + term);
        assertEquals(testTerms[i], term);
        dpEnum = termsEnum.postings(dpEnum, PostingsEnum.ALL);
        assertNotNull(dpEnum);
        int doc = dpEnum.docID();
        assertEquals(-1, doc);
        assertTrue(dpEnum.nextDoc() != DocIdSetIterator.NO_MORE_DOCS);
        assertEquals(dpEnum.freq(), positions[i].length);
        for (int j = 0; j < positions[i].length; j++) {
            assertEquals(positions[i][j], dpEnum.nextPosition());
        }
        assertEquals(DocIdSetIterator.NO_MORE_DOCS, dpEnum.nextDoc());
        dpEnum = termsEnum.postings(dpEnum, PostingsEnum.ALL);
        doc = dpEnum.docID();
        assertEquals(-1, doc);
        assertTrue(dpEnum.nextDoc() != DocIdSetIterator.NO_MORE_DOCS);
        assertNotNull(dpEnum);
        assertEquals(dpEnum.freq(), positions[i].length);
        for (int j = 0; j < positions[i].length; j++) {
            assertEquals(positions[i][j], dpEnum.nextPosition());
            assertEquals(j * 10, dpEnum.startOffset());
            assertEquals(j * 10 + testTerms[i].length(), dpEnum.endOffset());
        }
        assertEquals(DocIdSetIterator.NO_MORE_DOCS, dpEnum.nextDoc());
    }
    //no pos, no offset
    Terms freqVector = reader.get(0).terms(testFields[1]);
    assertNotNull(freqVector);
    assertEquals(testTerms.length, freqVector.size());
    termsEnum = freqVector.iterator();
    assertNotNull(termsEnum);
    for (int i = 0; i < testTerms.length; i++) {
        final BytesRef text = termsEnum.next();
        assertNotNull(text);
        String term = text.utf8ToString();
        //System.out.println("Term: " + term);
        assertEquals(testTerms[i], term);
        assertNotNull(termsEnum.postings(null));
        assertNotNull(termsEnum.postings(null, PostingsEnum.ALL));
    }
    reader.close();
}

Example 9 with TermVectorsReader

use of org.apache.lucene.codecs.TermVectorsReader in project lucene-solr by apache.

the class CodecReader method getTermVectors.

@Override
public final Fields getTermVectors(int docID) throws IOException {
    TermVectorsReader termVectorsReader = getTermVectorsReader();
    if (termVectorsReader == null) {
        return null;
    }
    checkBounds(docID);
    return termVectorsReader.get(docID);
}