Examples with CompressionCodec org.apache.hadoop.io.compress.CompressionCodec used on opensource projects

Example 21 with CompressionCodec

use of org.apache.hadoop.io.compress.CompressionCodec in project hadoop by apache.

the class TextOutputFormat method getRecordWriter.

public RecordWriter<K, V> getRecordWriter(TaskAttemptContext job) throws IOException, InterruptedException {
    Configuration conf = job.getConfiguration();
    boolean isCompressed = getCompressOutput(job);
    String keyValueSeparator = conf.get(SEPERATOR, "\t");
    CompressionCodec codec = null;
    String extension = "";
    if (isCompressed) {
        Class<? extends CompressionCodec> codecClass = getOutputCompressorClass(job, GzipCodec.class);
        codec = ReflectionUtils.newInstance(codecClass, conf);
        extension = codec.getDefaultExtension();
    }
    Path file = getDefaultWorkFile(job, extension);
    FileSystem fs = file.getFileSystem(conf);
    FSDataOutputStream fileOut = fs.create(file, false);
    if (isCompressed) {
        return new LineRecordWriter<>(new DataOutputStream(codec.createOutputStream(fileOut)), keyValueSeparator);
    } else {
        return new LineRecordWriter<>(fileOut, keyValueSeparator);
    }
}

Example 22 with CompressionCodec

use of org.apache.hadoop.io.compress.CompressionCodec in project hadoop by apache.

the class FSImageCompression method createCompression.

/**
   * Create a compression instance using the codec specified by
   * <code>codecClassName</code>
   */
static FSImageCompression createCompression(Configuration conf, String codecClassName) throws IOException {
    CompressionCodecFactory factory = new CompressionCodecFactory(conf);
    CompressionCodec codec = factory.getCodecByClassName(codecClassName);
    if (codec == null) {
        throw new IOException("Not a supported codec: " + codecClassName);
    }
    return new FSImageCompression(codec);
}

Example 23 with CompressionCodec

use of org.apache.hadoop.io.compress.CompressionCodec in project hadoop-pcap by RIPE-NCC.

the class PcapInputFormat method initPcapRecordReader.

public static PcapRecordReader initPcapRecordReader(Path path, long start, long length, Reporter reporter, Configuration conf) throws IOException {
    FileSystem fs = path.getFileSystem(conf);
    FSDataInputStream baseStream = fs.open(path);
    DataInputStream stream = baseStream;
    CompressionCodecFactory compressionCodecs = new CompressionCodecFactory(conf);
    final CompressionCodec codec = compressionCodecs.getCodec(path);
    if (codec != null)
        stream = new DataInputStream(codec.createInputStream(stream));
    PcapReader reader = initPcapReader(stream, conf);
    return new PcapRecordReader(reader, start, length, baseStream, stream, reporter);
}

Example 24 with CompressionCodec

use of org.apache.hadoop.io.compress.CompressionCodec in project hadoop-pcap by RIPE-NCC.

the class PcapInputFormat method initPcapRecordReader.

public static PcapRecordReader initPcapRecordReader(Path path, long start, long length, TaskAttemptContext context) throws IOException {
    Configuration conf = context.getConfiguration();
    FileSystem fs = path.getFileSystem(conf);
    FSDataInputStream baseStream = fs.open(path);
    DataInputStream stream = baseStream;
    CompressionCodecFactory compressionCodecs = new CompressionCodecFactory(conf);
    final CompressionCodec codec = compressionCodecs.getCodec(path);
    if (codec != null)
        stream = new DataInputStream(codec.createInputStream(stream));
    PcapReader reader = initPcapReader(stream, conf);
    return new PcapRecordReader(reader, start, length, baseStream, stream, context);
}

Example 25 with CompressionCodec

use of org.apache.hadoop.io.compress.CompressionCodec in project jena by apache.

the class AbstractLineBasedNodeTupleReader method initialize.

@Override
public final void initialize(InputSplit genericSplit, TaskAttemptContext context) throws IOException {
    LOG.debug("initialize({}, {})", genericSplit, context);
    // Assuming file split
    if (!(genericSplit instanceof FileSplit))
        throw new IOException("This record reader only supports FileSplit inputs");
    FileSplit split = (FileSplit) genericSplit;
    // Intermediate : RDFParser but need to make a Iterator<Quad/Triple>
    LabelToNode labelToNode = RdfIOUtils.createLabelToNode(context, split.getPath());
    maker = new ParserProfileStd(RiotLib.factoryRDF(labelToNode), ErrorHandlerFactory.errorHandlerStd, IRIResolver.create(), PrefixMapFactory.createForInput(), null, true, false);
    Configuration config = context.getConfiguration();
    this.ignoreBadTuples = config.getBoolean(RdfIOConstants.INPUT_IGNORE_BAD_TUPLES, true);
    if (this.ignoreBadTuples)
        LOG.warn("Configured to ignore bad tuples, parsing errors will be logged and the bad line skipped but no errors will be thrownConsider setting {} to false to disable this behaviour", RdfIOConstants.INPUT_IGNORE_BAD_TUPLES);
    // Figure out what portion of the file to read
    this.maxLineLength = config.getInt(HadoopIOConstants.MAX_LINE_LENGTH, Integer.MAX_VALUE);
    start = split.getStart();
    end = start + split.getLength();
    final Path file = split.getPath();
    long totalLength = file.getFileSystem(context.getConfiguration()).getFileStatus(file).getLen();
    compressionCodecs = new CompressionCodecFactory(config);
    final CompressionCodec codec = compressionCodecs.getCodec(file);
    LOG.info(String.format("Got split with start %d and length %d for file with total length of %d", new Object[] { start, split.getLength(), totalLength }));
    // Open the file and seek to the start of the split
    FileSystem fs = file.getFileSystem(config);
    FSDataInputStream fileIn = fs.open(file);
    boolean skipFirstLine = false;
    if (codec != null) {
        // Add 1 and verify we got complete split
        if (totalLength > split.getLength() + 1)
            throw new IOException("This record reader can only be used with compressed input where the split covers the whole file");
        in = new LineReader(codec.createInputStream(fileIn), config);
        estLength = end;
        end = Long.MAX_VALUE;
    } else {
        // Uncompressed input
        if (start != 0) {
            skipFirstLine = true;
            --start;
            fileIn.seek(start);
        }
        in = new LineReader(fileIn, config);
    }
    // NLineInputFormat will provide the split information to use
    if (skipFirstLine) {
        start += in.readLine(new Text(), 0, (int) Math.min(Integer.MAX_VALUE, end - start));
    }
    this.pos = start;
}