org.apache.hadoop.io.compress.CompressionCodec#createOutputStream

Source File: IFile.java From tez with Apache License 2.0

6 votes

void setupOutputStream(CompressionCodec codec) throws IOException {
  this.checksumOut = new IFileOutputStream(this.rawOut);
  if (codec != null) {
    this.compressor = CodecPool.getCompressor(codec);
    if (this.compressor != null) {
      this.compressor.reset();
      this.compressedOut = codec.createOutputStream(checksumOut, compressor);
      this.out = new FSDataOutputStream(this.compressedOut,  null);
      this.compressOutput = true;
    } else {
      LOG.warn("Could not obtain compressor from CodecPool");
      this.out = new FSDataOutputStream(checksumOut,null);
    }
  } else {
    this.out = new FSDataOutputStream(checksumOut,null);
  }
}

Source File: Anonymizer.java From hadoop with Apache License 2.0

6 votes

private JsonGenerator createJsonGenerator(Configuration conf, Path path) 
throws IOException {
  FileSystem outFS = path.getFileSystem(conf);
  CompressionCodec codec =
    new CompressionCodecFactory(conf).getCodec(path);
  OutputStream output;
  Compressor compressor = null;
  if (codec != null) {
    compressor = CodecPool.getCompressor(codec);
    output = codec.createOutputStream(outFS.create(path), compressor);
  } else {
    output = outFS.create(path);
  }

  JsonGenerator outGen = outFactory.createJsonGenerator(output, 
                                                        JsonEncoding.UTF8);
  outGen.useDefaultPrettyPrinter();
  
  return outGen;
}

Source File: CompressionEmulationUtil.java From hadoop with Apache License 2.0

6 votes

/**
 * Returns a {@link OutputStream} for a file that might need 
 * compression.
 */
static OutputStream getPossiblyCompressedOutputStream(Path file, 
                                                      Configuration conf)
throws IOException {
  FileSystem fs = file.getFileSystem(conf);
  JobConf jConf = new JobConf(conf);
  if (org.apache.hadoop.mapred.FileOutputFormat.getCompressOutput(jConf)) {
    // get the codec class
    Class<? extends CompressionCodec> codecClass =
      org.apache.hadoop.mapred.FileOutputFormat
                              .getOutputCompressorClass(jConf, 
                                                        GzipCodec.class);
    // get the codec implementation
    CompressionCodec codec = ReflectionUtils.newInstance(codecClass, conf);

    // add the appropriate extension
    file = file.suffix(codec.getDefaultExtension());

    if (isCompressionEmulationEnabled(conf)) {
      FSDataOutputStream fileOut = fs.create(file, false);
      return new DataOutputStream(codec.createOutputStream(fileOut));
    }
  }
  return fs.create(file, false);
}

Source File: CompressedFileWriteFromLocal.java From hiped2 with Apache License 2.0

6 votes

public static void main(String... args) throws Exception {
  Configuration config = new Configuration();
  FileSystem hdfs = FileSystem.get(config);

  Class<?> codecClass = Class.forName(args[0]);
  CompressionCodec codec = (CompressionCodec)
      ReflectionUtils.newInstance(codecClass, config);

  InputStream is = FileSystem.getLocal(config).open(new Path(args[1]));
  OutputStream os = hdfs.create(
      new Path(args[2] + codec.getDefaultExtension()));


  OutputStream cos = codec.createOutputStream(os);

  IOUtils.copyBytes(is, cos, config, true);

  IOUtils.closeStream(os);
  IOUtils.closeStream(is);
}

Source File: LineBreakAccessor.java From pxf with Apache License 2.0

5 votes

private void createOutputStream(Path file, CompressionCodec codec)
        throws IOException {
    fsdos = fs.create(file, false);
    if (codec != null) {
        dos = new DataOutputStream(codec.createOutputStream(fsdos));
    } else {
        dos = fsdos;
    }

}

Source File: FsShell.java From RDFS with Apache License 2.0

5 votes

/**
 * Compress a file.
 */
private int compress(String argv[], Configuration conf) throws IOException {
  int i = 0;
  String cmd = argv[i++];
  String srcf = argv[i++];
  String dstf = argv[i++];

  Path srcPath = new Path(srcf);
  FileSystem srcFs = srcPath.getFileSystem(getConf());
  Path dstPath = new Path(dstf);
  FileSystem dstFs = dstPath.getFileSystem(getConf());

  // Create codec
  CompressionCodecFactory factory = new CompressionCodecFactory(conf);
  CompressionCodec codec = factory.getCodec(dstPath);
  if (codec == null) {
    System.err.println(cmd.substring(1) + ": cannot find compression codec for "
        + dstf);
    return 1;
  }

  // open input stream
  InputStream in = srcFs.open(srcPath);

  // Create compression stream
  OutputStream out = dstFs.create(dstPath);
  out = codec.createOutputStream(out);

  IOUtils.copyBytes(in, out, conf, true);

  return 0;
}

Source File: AbstractFileOutputOperatorTest.java From attic-apex-malhar with Apache License 2.0

5 votes

@Test
public void testSnappyCompressionSimple() throws IOException
{
  if (checkNativeSnappy()) {
    return;
  }

  File snappyFile = new File(testMeta.getDir(), "snappyTestFile.snappy");

  BufferedOutputStream os = new BufferedOutputStream(new FileOutputStream(snappyFile));
  Configuration conf = new Configuration();
  CompressionCodec codec = (CompressionCodec)ReflectionUtils.newInstance(SnappyCodec.class, conf);
  FilterStreamCodec.SnappyFilterStream filterStream = new FilterStreamCodec.SnappyFilterStream(
      codec.createOutputStream(os));

  int ONE_MB = 1024 * 1024;

  String testStr = "TestSnap-16bytes";
  for (int i = 0; i < ONE_MB; i++) { // write 16 MBs
    filterStream.write(testStr.getBytes());
  }
  filterStream.flush();
  filterStream.close();

  CompressionInputStream is = codec.createInputStream(new FileInputStream(snappyFile));

  byte[] recovered = new byte[testStr.length()];
  int bytesRead = is.read(recovered);
  is.close();
  assertEquals(testStr, new String(recovered));
}

Source File: IFile.java From incubator-tez with Apache License 2.0

5 votes

public Writer(Configuration conf, FSDataOutputStream outputStream,
    Class keyClass, Class valueClass,
    CompressionCodec codec, TezCounter writesCounter, TezCounter serializedBytesCounter,
    boolean rle) throws IOException {
  this.rawOut = outputStream;
  this.writtenRecordsCounter = writesCounter;
  this.serializedUncompressedBytes = serializedBytesCounter;
  this.checksumOut = new IFileOutputStream(outputStream);
  this.start = this.rawOut.getPos();
  this.rle = rle;
  if (codec != null) {
    this.compressor = CodecPool.getCompressor(codec);
    if (this.compressor != null) {
      this.compressor.reset();
      this.compressedOut = codec.createOutputStream(checksumOut, compressor);
      this.out = new FSDataOutputStream(this.compressedOut,  null);
      this.compressOutput = true;
    } else {
      LOG.warn("Could not obtain compressor from CodecPool");
      this.out = new FSDataOutputStream(checksumOut,null);
    }
  } else {
    this.out = new FSDataOutputStream(checksumOut,null);
  }
  writeHeader(outputStream);
  this.keyClass = keyClass;
  this.valueClass = valueClass;

  if (keyClass != null) {
    SerializationFactory serializationFactory =
      new SerializationFactory(conf);
    this.keySerializer = serializationFactory.getSerializer(keyClass);
    this.keySerializer.open(buffer);
    this.valueSerializer = serializationFactory.getSerializer(valueClass);
    this.valueSerializer.open(buffer);
  }
}

Source File: FreightStreamer.java From RDFS with Apache License 2.0

5 votes

/**
 * Compress a file.
 */
private int compress(String argv[], Configuration conf) throws IOException {
  int i = 0;
  String cmd = argv[i++];
  String srcf = argv[i++];
  String dstf = argv[i++];

  Path srcPath = new Path(srcf);
  FileSystem srcFs = srcPath.getFileSystem(getConf());
  Path dstPath = new Path(dstf);
  FileSystem dstFs = dstPath.getFileSystem(getConf());

  // Create codec
  CompressionCodecFactory factory = new CompressionCodecFactory(conf);
  CompressionCodec codec = factory.getCodec(dstPath);
  if (codec == null) {
    System.err.println(cmd.substring(1) + ": cannot find compression codec for "
        + dstf);
    return 1;
  }

  // open input stream
  InputStream in = srcFs.open(srcPath);

  // Create compression stream
  OutputStream out = dstFs.create(dstPath);
  out = codec.createOutputStream(out);

  IOUtils.copyBytes(in, out, conf, true);

  return 0;
}

Source File: IFile.java From RDFS with Apache License 2.0

5 votes

public Writer(Configuration conf, FSDataOutputStream out, 
    Class<K> keyClass, Class<V> valueClass,
    CompressionCodec codec, Counters.Counter writesCounter)
    throws IOException {
  this.writtenRecordsCounter = writesCounter;
  this.checksumOut = new IFileOutputStream(out);
  this.rawOut = out;
  this.start = this.rawOut.getPos();
  
  if (codec != null) {
    this.compressor = CodecPool.getCompressor(codec);
    this.compressor.reset();
    this.compressedOut = codec.createOutputStream(checksumOut, compressor);
    this.out = new FSDataOutputStream(this.compressedOut,  null);
    this.compressOutput = true;
  } else {
    this.out = new FSDataOutputStream(checksumOut,null);
  }
  
  this.keyClass = keyClass;
  this.valueClass = valueClass;
  SerializationFactory serializationFactory = new SerializationFactory(conf);
  this.keySerializer = serializationFactory.getSerializer(keyClass);
  this.keySerializer.open(buffer);
  this.valueSerializer = serializationFactory.getSerializer(valueClass);
  this.valueSerializer.open(buffer);
}

Source File: Compression.java From hbase with Apache License 2.0

5 votes

/**
 * Creates a compression stream without any additional wrapping into
 * buffering streams.
 */
public CompressionOutputStream createPlainCompressionStream(
    OutputStream downStream, Compressor compressor) throws IOException {
  CompressionCodec codec = getCodec(conf);
  ((Configurable)codec).getConf().setInt("io.file.buffer.size", 32 * 1024);
  return codec.createOutputStream(downStream, compressor);
}

Source File: IFile.java From big-c with Apache License 2.0

5 votes

public Writer(Configuration conf, FSDataOutputStream out, 
    Class<K> keyClass, Class<V> valueClass,
    CompressionCodec codec, Counters.Counter writesCounter,
    boolean ownOutputStream)
    throws IOException {
  this.writtenRecordsCounter = writesCounter;
  this.checksumOut = new IFileOutputStream(out);
  this.rawOut = out;
  this.start = this.rawOut.getPos();
  if (codec != null) {
    this.compressor = CodecPool.getCompressor(codec);
    if (this.compressor != null) {
      this.compressor.reset();
      this.compressedOut = codec.createOutputStream(checksumOut, compressor);
      this.out = new FSDataOutputStream(this.compressedOut,  null);
      this.compressOutput = true;
    } else {
      LOG.warn("Could not obtain compressor from CodecPool");
      this.out = new FSDataOutputStream(checksumOut,null);
    }
  } else {
    this.out = new FSDataOutputStream(checksumOut,null);
  }
  
  this.keyClass = keyClass;
  this.valueClass = valueClass;

  if (keyClass != null) {
    SerializationFactory serializationFactory = 
      new SerializationFactory(conf);
    this.keySerializer = serializationFactory.getSerializer(keyClass);
    this.keySerializer.open(buffer);
    this.valueSerializer = serializationFactory.getSerializer(valueClass);
    this.valueSerializer.open(buffer);
  }
  this.ownOutputStream = ownOutputStream;
}

Source File: TestCombineTextInputFormat.java From big-c with Apache License 2.0

5 votes

private static void writeFile(FileSystem fs, Path name,
                              CompressionCodec codec,
                              String contents) throws IOException {
  OutputStream stm;
  if (codec == null) {
    stm = fs.create(name);
  } else {
    stm = codec.createOutputStream(fs.create(name));
  }
  stm.write(contents.getBytes());
  stm.close();
}

Source File: DefaultOutputter.java From hadoop with Apache License 2.0

5 votes

@Override
public void init(Path path, Configuration conf) throws IOException {
  FileSystem fs = path.getFileSystem(conf);
  CompressionCodec codec = new CompressionCodecFactory(conf).getCodec(path);
  OutputStream output;
  if (codec != null) {
    compressor = CodecPool.getCompressor(codec);
    output = codec.createOutputStream(fs.create(path), compressor);
  } else {
    output = fs.create(path);
  }
  writer = new JsonObjectMapperWriter<T>(output, 
      conf.getBoolean("rumen.output.pretty.print", true));
}

Source File: IFile.java From hadoop with Apache License 2.0

5 votes

public Writer(Configuration conf, FSDataOutputStream out, 
    Class<K> keyClass, Class<V> valueClass,
    CompressionCodec codec, Counters.Counter writesCounter,
    boolean ownOutputStream)
    throws IOException {
  this.writtenRecordsCounter = writesCounter;
  this.checksumOut = new IFileOutputStream(out);
  this.rawOut = out;
  this.start = this.rawOut.getPos();
  if (codec != null) {
    this.compressor = CodecPool.getCompressor(codec);
    if (this.compressor != null) {
      this.compressor.reset();
      this.compressedOut = codec.createOutputStream(checksumOut, compressor);
      this.out = new FSDataOutputStream(this.compressedOut,  null);
      this.compressOutput = true;
    } else {
      LOG.warn("Could not obtain compressor from CodecPool");
      this.out = new FSDataOutputStream(checksumOut,null);
    }
  } else {
    this.out = new FSDataOutputStream(checksumOut,null);
  }
  
  this.keyClass = keyClass;
  this.valueClass = valueClass;

  if (keyClass != null) {
    SerializationFactory serializationFactory = 
      new SerializationFactory(conf);
    this.keySerializer = serializationFactory.getSerializer(keyClass);
    this.keySerializer.open(buffer);
    this.valueSerializer = serializationFactory.getSerializer(valueClass);
    this.valueSerializer.open(buffer);
  }
  this.ownOutputStream = ownOutputStream;
}

Source File: TestCombineTextInputFormat.java From hadoop with Apache License 2.0

5 votes

private static void writeFile(FileSystem fs, Path name,
                              CompressionCodec codec,
                              String contents) throws IOException {
  OutputStream stm;
  if (codec == null) {
    stm = fs.create(name);
  } else {
    stm = codec.createOutputStream(fs.create(name));
  }
  stm.write(contents.getBytes());
  stm.close();
}

Source File: TestCombineTextInputFormat.java From hadoop with Apache License 2.0

5 votes

private static void writeFile(FileSystem fs, Path name,
                              CompressionCodec codec,
                              String contents) throws IOException {
  OutputStream stm;
  if (codec == null) {
    stm = fs.create(name);
  } else {
    stm = codec.createOutputStream(fs.create(name));
  }
  stm.write(contents.getBytes());
  stm.close();
}

Source File: TestExport.java From aliyun-maxcompute-data-collectors with Apache License 2.0

5 votes

/**
 * Create a data file that gets exported to the db.
 * @param fileNum the number of the file (for multi-file export)
 * @param numRecords how many records to write to the file.
 * @param gzip is true if the file should be gzipped.
 */
protected void createTextFile(int fileNum, int numRecords, boolean gzip,
    ColumnGenerator... extraCols) throws IOException {
  int startId = fileNum * numRecords;

  String ext = ".txt";
  if (gzip) {
    ext = ext + ".gz";
  }
  Path tablePath = getTablePath();
  Path filePath = new Path(tablePath, "part" + fileNum + ext);

  Configuration conf = new Configuration();
  if (!BaseSqoopTestCase.isOnPhysicalCluster()) {
    conf.set(CommonArgs.FS_DEFAULT_NAME, CommonArgs.LOCAL_FS);
  }
  FileSystem fs = FileSystem.get(conf);
  fs.mkdirs(tablePath);
  OutputStream os = fs.create(filePath);
  if (gzip) {
    CompressionCodecFactory ccf = new CompressionCodecFactory(conf);
    CompressionCodec codec = ccf.getCodec(filePath);
    os = codec.createOutputStream(os);
  }
  BufferedWriter w = new BufferedWriter(new OutputStreamWriter(os));
  for (int i = 0; i < numRecords; i++) {
    w.write(getRecordLine(startId + i, extraCols));
  }
  w.close();
  os.close();

  if (gzip) {
    verifyCompressedFile(filePath, numRecords);
  }
}

Source File: PushdownLargeFieldedListsVisitor.java From datawave with Apache License 2.0

5 votes

protected URI createFst(SortedSet<String> values) throws IOException, ClassNotFoundException, InstantiationException, IllegalAccessException {
    FST fst = DatawaveFieldIndexListIteratorJexl.getFST(values);
    
    // now serialize to our file system
    CompressionCodec codec = null;
    String extension = "";
    if (config.getHdfsFileCompressionCodec() != null) {
        ClassLoader classLoader = Thread.currentThread().getContextClassLoader();
        if (classLoader == null) {
            classLoader = this.getClass().getClassLoader();
        }
        Class<? extends CompressionCodec> clazz = Class.forName(config.getHdfsFileCompressionCodec(), true, classLoader).asSubclass(CompressionCodec.class);
        codec = clazz.newInstance();
        extension = codec.getDefaultExtension();
    }
    int fstCount = config.getFstCount().incrementAndGet();
    Path fstFile = new Path(fstHdfsUri, "PushdownLargeFileFst." + fstCount + ".fst" + extension);
    
    OutputStream fstFileOut = new BufferedOutputStream(fs.create(fstFile, false));
    if (codec != null) {
        fstFileOut = codec.createOutputStream(fstFileOut);
    }
    
    OutputStreamDataOutput outStream = new OutputStreamDataOutput(fstFileOut);
    fst.save(outStream);
    outStream.close();
    
    return fstFile.toUri();
}

Source File: RubixRecordWriter.java From Cubert with Apache License 2.0

4 votes

@SuppressWarnings({ "unchecked", "rawtypes" })
public RubixRecordWriter(Configuration conf,
                         FSDataOutputStream out,
                         Class keyClass,
                         Class valueClass,
                         CompressionCodec codec) throws IOException
{
    this.out = out;

    final SerializationFactory serializationFactory = new SerializationFactory(conf);
    keySerializer = serializationFactory.getSerializer(keyClass);

    ObjectMapper mapper = new ObjectMapper();
    metadataJson =
            mapper.readValue(conf.get(CubertStrings.JSON_METADATA), JsonNode.class);
    ((ObjectNode) metadataJson).put("keyClass", keyClass.getCanonicalName());
    ((ObjectNode) metadataJson).put("valueClass", valueClass.getCanonicalName());
    BlockSchema schema = new BlockSchema(metadataJson.get("schema"));

    if (conf.getBoolean(CubertStrings.USE_COMPACT_SERIALIZATION, false)
            && schema.isFlatSchema())
    {
        valueSerializer = new CompactSerializer<V>(schema);
        ((ObjectNode) metadataJson).put("serializationType",
                                        BlockSerializationType.COMPACT.toString());
    }
    else
    {
        valueSerializer = serializationFactory.getSerializer(valueClass);
        ((ObjectNode) metadataJson).put("serializationType",
                                        BlockSerializationType.DEFAULT.toString());
    }

    keySerializer.open(keySectionStream);

    if (codec == null)
    {
        valueSerializer.open(out);
        compressedStream = null;
    }
    else
    {
        compressedStream = codec.createOutputStream(out);
        valueSerializer.open(compressedStream);
    }

}

Java Code Examples for org.apache.hadoop.io.compress.CompressionCodec#createOutputStream()