Examples with ColumnPageDecoder org.apache.carbondata.core.datastore.page.encoding.ColumnPageDecoder used on opensource projects

Example 11 with ColumnPageDecoder

use of org.apache.carbondata.core.datastore.page.encoding.ColumnPageDecoder in project carbondata by apache.

the class MeasureChunkReaderV3 method decodeMeasure.

/**
 * Decode measure column page with page header and raw data starting from offset
 */
protected ColumnPage decodeMeasure(DataChunk2 pageMetadata, ByteBuffer pageData, int offset, ColumnVectorInfo vectorInfo, BitSet nullBitSet, ReusableDataBuffer reusableDataBuffer) throws IOException {
    List<Encoding> encodings = pageMetadata.getEncoders();
    org.apache.carbondata.core.metadata.encoder.Encoding.validateEncodingTypes(encodings);
    List<ByteBuffer> encoderMetas = pageMetadata.getEncoder_meta();
    String compressorName = CarbonMetadataUtil.getCompressorNameFromChunkMeta(pageMetadata.getChunk_meta());
    ColumnPageDecoder codec = encodingFactory.createDecoder(encodings, encoderMetas, compressorName, vectorInfo != null);
    if (vectorInfo != null) {
        vectorInfo.vector.setCarbonDataFileWrittenVersion(vectorInfo.carbonDataFileWrittenVersion);
        codec.decodeAndFillVector(pageData.array(), offset, pageMetadata.data_page_length, vectorInfo, nullBitSet, false, pageMetadata.numberOfRowsInpage, reusableDataBuffer);
        return null;
    } else {
        return codec.decode(pageData.array(), offset, pageMetadata.data_page_length);
    }
}

Example 12 with ColumnPageDecoder

use of org.apache.carbondata.core.datastore.page.encoding.ColumnPageDecoder in project carbondata by apache.

the class TestPageLevelDictionary method testPageLevelDictionaryContainsOnlyUsedDictionaryValuesWhenMultiplePagesUseSameDictionary.

@Test
public void testPageLevelDictionaryContainsOnlyUsedDictionaryValuesWhenMultiplePagesUseSameDictionary() {
    LocalDictionaryGenerator generator = new ColumnLocalDictionaryGenerator(1000, 2);
    String columnName = "column1";
    PageLevelDictionary pageLevelDictionary1 = new PageLevelDictionary(generator, columnName, DataTypes.STRING, false, compressorName);
    byte[][] validateData = new byte[10][];
    int index = 0;
    try {
        for (int i = 1; i <= 5; i++) {
            byte[] data = ("vishal" + i).getBytes();
            ByteBuffer byteBuffer = ByteBuffer.allocate(data.length + 2);
            byteBuffer.putShort((short) data.length);
            byteBuffer.put(data);
            validateData[index] = data;
            pageLevelDictionary1.getDictionaryValue(byteBuffer.array());
            index++;
        }
        Assert.assertTrue(true);
    } catch (DictionaryThresholdReachedException e) {
        Assert.assertTrue(false);
    }
    PageLevelDictionary pageLevelDictionary2 = new PageLevelDictionary(generator, columnName, DataTypes.STRING, false, compressorName);
    try {
        for (int i = 1; i <= 5; i++) {
            byte[] data = ("vikas" + i).getBytes();
            ByteBuffer byteBuffer = ByteBuffer.allocate(data.length + 2);
            byteBuffer.putShort((short) data.length);
            byteBuffer.put(data);
            pageLevelDictionary2.getDictionaryValue(byteBuffer.array());
        }
        Assert.assertTrue(true);
    } catch (DictionaryThresholdReachedException e) {
        Assert.assertTrue(false);
    }
    try {
        for (int i = 6; i <= 10; i++) {
            byte[] data = ("vishal" + i).getBytes();
            ByteBuffer byteBuffer = ByteBuffer.allocate(data.length + 2);
            byteBuffer.putShort((short) data.length);
            byteBuffer.put(data);
            validateData[index] = data;
            pageLevelDictionary1.getDictionaryValue(byteBuffer.array());
            index++;
        }
        Assert.assertTrue(true);
    } catch (DictionaryThresholdReachedException e) {
        Assert.assertTrue(false);
    }
    try {
        for (int i = 6; i <= 10; i++) {
            byte[] data = ("vikas" + i).getBytes();
            ByteBuffer byteBuffer = ByteBuffer.allocate(data.length + 2);
            byteBuffer.putShort((short) data.length);
            byteBuffer.put(data);
            pageLevelDictionary2.getDictionaryValue(byteBuffer.array());
        }
        Assert.assertTrue(true);
    } catch (DictionaryThresholdReachedException e) {
        Assert.assertTrue(false);
    }
    try {
        LocalDictionaryChunk localDictionaryChunkForBlocklet = pageLevelDictionary1.getLocalDictionaryChunkForBlocklet();
        List<Encoding> encodings = localDictionaryChunkForBlocklet.getDictionary_meta().getEncoders();
        EncodingFactory encodingFactory = DefaultEncodingFactory.getInstance();
        List<ByteBuffer> encoderMetas = localDictionaryChunkForBlocklet.getDictionary_meta().getEncoder_meta();
        ColumnPageDecoder decoder = encodingFactory.createDecoder(encodings, encoderMetas, compressorName);
        ColumnPage decode = decoder.decode(localDictionaryChunkForBlocklet.getDictionary_data(), 0, localDictionaryChunkForBlocklet.getDictionary_data().length);
        BitSet bitSet = BitSet.valueOf(CompressorFactory.getInstance().getCompressor(compressorName).unCompressByte(localDictionaryChunkForBlocklet.getDictionary_values()));
        Assert.assertTrue(bitSet.cardinality() == validateData.length);
        for (int i = 0; i < validateData.length; i++) {
            Assert.assertTrue(Arrays.equals(decode.getBytes(i), validateData[i]));
        }
    } catch (IOException e) {
        Assert.assertTrue(false);
    }
}

Example 13 with ColumnPageDecoder

use of org.apache.carbondata.core.datastore.page.encoding.ColumnPageDecoder in project carbondata by apache.

the class TestPageLevelDictionary method testPageLevelDictionaryContainsOnlyUsedDictionaryValues.

@Test
public void testPageLevelDictionaryContainsOnlyUsedDictionaryValues() {
    LocalDictionaryGenerator generator = new ColumnLocalDictionaryGenerator(1000, 2);
    String columnName = "column1";
    PageLevelDictionary pageLevelDictionary1 = new PageLevelDictionary(generator, columnName, DataTypes.STRING, false, compressorName);
    byte[][] validateData = new byte[500][];
    try {
        for (int i = 1; i <= 500; i++) {
            byte[] data = ("vishal" + i).getBytes();
            ByteBuffer byteBuffer = ByteBuffer.allocate(data.length + 2);
            byteBuffer.putShort((short) data.length);
            byteBuffer.put(data);
            validateData[i - 1] = data;
            pageLevelDictionary1.getDictionaryValue(byteBuffer.array());
        }
        Assert.assertTrue(true);
    } catch (DictionaryThresholdReachedException e) {
        Assert.assertTrue(false);
    }
    PageLevelDictionary pageLevelDictionary2 = new PageLevelDictionary(generator, columnName, DataTypes.STRING, false, compressorName);
    try {
        for (int i = 1; i <= 500; i++) {
            byte[] data = ("vikas" + i).getBytes();
            ByteBuffer byteBuffer = ByteBuffer.allocate(data.length + 2);
            byteBuffer.putShort((short) data.length);
            byteBuffer.put(data);
            pageLevelDictionary2.getDictionaryValue(byteBuffer.array());
        }
        Assert.assertTrue(true);
    } catch (DictionaryThresholdReachedException e) {
        Assert.assertTrue(false);
    }
    try {
        LocalDictionaryChunk localDictionaryChunkForBlocklet = pageLevelDictionary1.getLocalDictionaryChunkForBlocklet();
        List<Encoding> encodings = localDictionaryChunkForBlocklet.getDictionary_meta().getEncoders();
        EncodingFactory encodingFactory = DefaultEncodingFactory.getInstance();
        List<ByteBuffer> encoderMetas = localDictionaryChunkForBlocklet.getDictionary_meta().getEncoder_meta();
        ColumnPageDecoder decoder = encodingFactory.createDecoder(encodings, encoderMetas, compressorName);
        ColumnPage decode = decoder.decode(localDictionaryChunkForBlocklet.getDictionary_data(), 0, localDictionaryChunkForBlocklet.getDictionary_data().length);
        for (int i = 0; i < 500; i++) {
            Arrays.equals(decode.getBytes(i), validateData[i]);
        }
    } catch (IOException e) {
        Assert.assertTrue(false);
    }
}

Example 14 with ColumnPageDecoder

use of org.apache.carbondata.core.datastore.page.encoding.ColumnPageDecoder in project carbondata by apache.

the class DimensionChunkReaderV3 method decodeDimensionByMeta.

private ColumnPage decodeDimensionByMeta(DataChunk2 pageMetadata, ByteBuffer pageData, int offset, boolean isLocalDictEncodedPage, ColumnVectorInfo vectorInfo, BitSet nullBitSet, ReusableDataBuffer reusableDataBuffer) throws IOException {
    List<Encoding> encodings = pageMetadata.getEncoders();
    List<ByteBuffer> encoderMetas = pageMetadata.getEncoder_meta();
    String compressorName = CarbonMetadataUtil.getCompressorNameFromChunkMeta(pageMetadata.getChunk_meta());
    ColumnPageDecoder decoder = encodingFactory.createDecoder(encodings, encoderMetas, compressorName, vectorInfo != null);
    if (vectorInfo != null) {
        // set encodings of current page in the vectorInfo, used for decoding the complex child page
        vectorInfo.encodings = encodings;
        vectorInfo.vector.setCarbonDataFileWrittenVersion(vectorInfo.carbonDataFileWrittenVersion);
        decoder.decodeAndFillVector(pageData.array(), offset, pageMetadata.data_page_length, vectorInfo, nullBitSet, isLocalDictEncodedPage, pageMetadata.numberOfRowsInpage, reusableDataBuffer);
        if (vectorInfo.vector.getType().isComplexType() && !vectorInfo.vectorStack.isEmpty()) {
            // For complex type, always top of the vector stack is processed in decodeAndFillVector.
            // so, pop() the top vector as its processing is finished.
            vectorInfo.vectorStack.pop();
        }
        return null;
    } else {
        return decoder.decode(pageData.array(), offset, pageMetadata.data_page_length, isLocalDictEncodedPage);
    }
}

Example 15 with ColumnPageDecoder

use of org.apache.carbondata.core.datastore.page.encoding.ColumnPageDecoder in project carbondata by apache.

the class CarbonTestUtil method validateDictionary.

public static Boolean validateDictionary(DimensionRawColumnChunk rawColumnPage, String[] data) throws IOException {
    LocalDictionaryChunk local_dictionary = rawColumnPage.getDataChunkV3().local_dictionary;
    if (null != local_dictionary) {
        String compressorName = CarbonMetadataUtil.getCompressorNameFromChunkMeta(rawColumnPage.getDataChunkV3().getData_chunk_list().get(0).getChunk_meta());
        List<org.apache.carbondata.format.Encoding> encodings = local_dictionary.getDictionary_meta().encoders;
        DefaultEncodingFactory encodingFactory = (DefaultEncodingFactory) DefaultEncodingFactory.getInstance();
        ColumnPageDecoder decoder = encodingFactory.createDecoder(encodings, local_dictionary.getDictionary_meta().getEncoder_meta(), compressorName);
        LazyColumnPage dictionaryPage = (LazyColumnPage) decoder.decode(local_dictionary.getDictionary_data(), 0, local_dictionary.getDictionary_data().length);
        HashMap<DictionaryByteArrayWrapper, Integer> dictionaryMap = new HashMap<>();
        BitSet usedDictionaryValues = BitSet.valueOf(CompressorFactory.getInstance().getCompressor(compressorName).unCompressByte(local_dictionary.getDictionary_values()));
        int index = 0;
        int i = usedDictionaryValues.nextSetBit(0);
        while (i >= 0) {
            dictionaryMap.put(new DictionaryByteArrayWrapper(dictionaryPage.getBytes(index)), i);
            i = usedDictionaryValues.nextSetBit(i + 1);
            index += 1;
        }
        for (i = 0; i < data.length; i++) {
            if (null == dictionaryMap.get(new DictionaryByteArrayWrapper(data[i].getBytes(Charset.forName(CarbonCommonConstants.DEFAULT_CHARSET))))) {
                return false;
            }
        }
        return true;
    }
    return false;
}