Java Code Examples for org.apache.hadoop.io.compress.SplitCompressionInputStream#getAdjustedEnd()

The following examples show how to use org.apache.hadoop.io.compress.SplitCompressionInputStream#getAdjustedEnd() . You can vote up the ones you like or vote down the ones you don't like, and go to the original project or source file by following the links above each example. You may check out the related API usage on the sidebar.
Example 1
Source File: AbstractEthereumRecordReader.java    From hadoopcryptoledger with Apache License 2.0 5 votes vote down vote up
/***
 * Initializes readers
 * 
 * @param split Split to be used (asssumed to be a file split)
 * ϟaram context context of the job
 * @throws java.io.IOException in case of errors reading from the filestream provided by Hadoop
 * @throws java.lang.InterruptedException in case of thread interruption
 * 
 */

@Override
public void initialize(InputSplit split, TaskAttemptContext context) throws IOException, InterruptedException {
	   FileSplit fSplit = (FileSplit)split;
	   // Initialize start and end of split
	      start = fSplit.getStart();
	      end = start + fSplit.getLength();
	      final Path file = fSplit.getPath();
	      codec = new CompressionCodecFactory(context.getConfiguration()).getCodec(file);
	      final FileSystem fs = file.getFileSystem(context.getConfiguration());
	      FSDataInputStream fileIn = fs.open(file);
	      // open stream
	        if (isCompressedInput()) { // decompress
	        	decompressor = CodecPool.getDecompressor(codec);
	        	if (codec instanceof SplittableCompressionCodec) {
	  		
	          	final SplitCompressionInputStream cIn =((SplittableCompressionCodec)codec).createInputStream(fileIn, decompressor, start, end,SplittableCompressionCodec.READ_MODE.CONTINUOUS);
	  				ebr = new EthereumBlockReader(cIn, this.maxSizeEthereumBlock,this.bufferSize,this.useDirectBuffer);
	  				start = cIn.getAdjustedStart();
	         		end = cIn.getAdjustedEnd();
	          	filePosition = cIn; // take pos from compressed stream
	        } else {
	        	ebr = new EthereumBlockReader(codec.createInputStream(fileIn,decompressor), this.maxSizeEthereumBlock,this.bufferSize,this.useDirectBuffer);
	        	filePosition = fileIn;
	        }
	      } else {
	        fileIn.seek(start);
	        ebr = new EthereumBlockReader(fileIn, this.maxSizeEthereumBlock,this.bufferSize,this.useDirectBuffer);
	        filePosition = fileIn;
	      }
}
 
Example 2
Source File: AbstractBitcoinRecordReader.java    From hadoopcryptoledger with Apache License 2.0 5 votes vote down vote up
/**
* Initializes reader
* @param split Split to use (assumed to be a file split)
* @param context context of the job
*
*
* @throws java.io.IOException in case of errors reading from the filestream provided by Hadoop
* @throws java.lang.InterruptedException in case of thread interruption
*
*/
@Override
public void initialize(InputSplit split, TaskAttemptContext context) throws IOException, InterruptedException {
   FileSplit fSplit = (FileSplit)split;
 // Initialize start and end of split
    start = fSplit.getStart();
    end = start + fSplit.getLength();
    final Path file = fSplit.getPath();
    codec = new CompressionCodecFactory(context.getConfiguration()).getCodec(file);
    final FileSystem fs = file.getFileSystem(context.getConfiguration());
    FSDataInputStream fileIn = fs.open(file);
    // open stream
      if (isCompressedInput()) { // decompress
      	decompressor = CodecPool.getDecompressor(codec);
      	if (codec instanceof SplittableCompressionCodec) {
		
        	final SplitCompressionInputStream cIn =((SplittableCompressionCodec)codec).createInputStream(fileIn, decompressor, start, end,SplittableCompressionCodec.READ_MODE.CONTINUOUS);
				bbr = new BitcoinBlockReader(cIn, this.maxSizeBitcoinBlock,this.bufferSize,this.specificMagicByteArray,this.useDirectBuffer,this.readAuxPOW);
		start = cIn.getAdjustedStart();
       		end = cIn.getAdjustedEnd();
        	filePosition = cIn; // take pos from compressed stream
      } else {
	bbr = new BitcoinBlockReader(codec.createInputStream(fileIn,decompressor), this.maxSizeBitcoinBlock,this.bufferSize,this.specificMagicByteArray,this.useDirectBuffer,readAuxPOW);
	filePosition = fileIn;
      }
    } else {
      fileIn.seek(start);
      bbr = new BitcoinBlockReader(fileIn, this.maxSizeBitcoinBlock,this.bufferSize,this.specificMagicByteArray,this.useDirectBuffer,readAuxPOW);  
      filePosition = fileIn;
    }
    // seek to block start (for the case a block overlaps a split)
    try {
    	bbr.seekBlockStart();
    } catch (BitcoinBlockReadException bbre) {
		LOG.error("Error reading Bitcoin blockchhain data");
		LOG.error(bbre);
    } 
}
 
Example 3
Source File: ChunkRecordReader.java    From pxf with Apache License 2.0 4 votes vote down vote up
/**
 * Constructs a ChunkRecordReader instance.
 *
 * @param job the job configuration
 * @param split contains the file name, begin byte of the split and the
 *            bytes length
 * @throws IOException if an I/O error occurs when accessing the file or
 *             creating input stream to read from it
 */
public ChunkRecordReader(Configuration job, FileSplit split)
        throws IOException, IncompatibleInputStreamException {
    maxLineLength = job.getInt(MAX_LINE_LENGTH, Integer.MAX_VALUE);
    validateLength(maxLineLength);
    start = split.getStart();
    end = start + split.getLength();
    final Path file = split.getPath();
    compressionCodecs = new CompressionCodecFactory(job);
    codec = compressionCodecs.getCodec(file);

    // openForWrite the file and seek to the start of the split
    final FileSystem fs = file.getFileSystem(job);
    fileIn = fs.open(file, ChunkReader.DEFAULT_BUFFER_SIZE);
    fileLength = getInputStream().getFileLength();
    if (isCompressedInput()) {
        decompressor = CodecPool.getDecompressor(codec);
        if (codec instanceof SplittableCompressionCodec) {
            final SplitCompressionInputStream cIn = ((SplittableCompressionCodec) codec).createInputStream(
                    fileIn, decompressor, start, end,
                    SplittableCompressionCodec.READ_MODE.BYBLOCK);
            in = new ChunkReader(cIn);
            start = cIn.getAdjustedStart();
            end = cIn.getAdjustedEnd();
            filePosition = cIn; // take pos from compressed stream
        } else {
            in = new ChunkReader(codec.createInputStream(fileIn,
                    decompressor));
            filePosition = fileIn;
        }
    } else {
        fileIn.seek(start);
        in = new ChunkReader(fileIn);
        filePosition = fileIn;
    }
    /*
     * If this is not the first split, we always throw away first record
     * because we always (except the last split) read one extra line in
     * next() method.
     */
    if (start != 0) {
        start += in.readLine(new ChunkWritable(), maxBytesToConsume(start));
    }
    this.pos = start;
}
 
Example 4
Source File: LineRecordReader.java    From hadoop with Apache License 2.0 4 votes vote down vote up
public void initialize(InputSplit genericSplit,
                       TaskAttemptContext context) throws IOException {
  FileSplit split = (FileSplit) genericSplit;
  Configuration job = context.getConfiguration();
  this.maxLineLength = job.getInt(MAX_LINE_LENGTH, Integer.MAX_VALUE);
  start = split.getStart();
  end = start + split.getLength();
  final Path file = split.getPath();

  // open the file and seek to the start of the split
  final FileSystem fs = file.getFileSystem(job);
  fileIn = fs.open(file);
  
  CompressionCodec codec = new CompressionCodecFactory(job).getCodec(file);
  if (null!=codec) {
    isCompressedInput = true;	
    decompressor = CodecPool.getDecompressor(codec);
    if (codec instanceof SplittableCompressionCodec) {
      final SplitCompressionInputStream cIn =
        ((SplittableCompressionCodec)codec).createInputStream(
          fileIn, decompressor, start, end,
          SplittableCompressionCodec.READ_MODE.BYBLOCK);
      in = new CompressedSplitLineReader(cIn, job,
          this.recordDelimiterBytes);
      start = cIn.getAdjustedStart();
      end = cIn.getAdjustedEnd();
      filePosition = cIn;
    } else {
      in = new SplitLineReader(codec.createInputStream(fileIn,
          decompressor), job, this.recordDelimiterBytes);
      filePosition = fileIn;
    }
  } else {
    fileIn.seek(start);
    in = new SplitLineReader(fileIn, job, this.recordDelimiterBytes);
    filePosition = fileIn;
  }
  // If this is not the first split, we always throw away first record
  // because we always (except the last split) read one extra line in
  // next() method.
  if (start != 0) {
    start += in.readLine(new Text(), 0, maxBytesToConsume(start));
  }
  this.pos = start;
}
 
Example 5
Source File: LineRecordReader.java    From big-c with Apache License 2.0 4 votes vote down vote up
public void initialize(InputSplit genericSplit,
                       TaskAttemptContext context) throws IOException {
  FileSplit split = (FileSplit) genericSplit;
  Configuration job = context.getConfiguration();
  this.maxLineLength = job.getInt(MAX_LINE_LENGTH, Integer.MAX_VALUE);
  start = split.getStart();
  end = start + split.getLength();
  final Path file = split.getPath();

  // open the file and seek to the start of the split
  final FileSystem fs = file.getFileSystem(job);
  fileIn = fs.open(file);
  
  CompressionCodec codec = new CompressionCodecFactory(job).getCodec(file);
  if (null!=codec) {
    isCompressedInput = true;	
    decompressor = CodecPool.getDecompressor(codec);
    if (codec instanceof SplittableCompressionCodec) {
      final SplitCompressionInputStream cIn =
        ((SplittableCompressionCodec)codec).createInputStream(
          fileIn, decompressor, start, end,
          SplittableCompressionCodec.READ_MODE.BYBLOCK);
      in = new CompressedSplitLineReader(cIn, job,
          this.recordDelimiterBytes);
      start = cIn.getAdjustedStart();
      end = cIn.getAdjustedEnd();
      filePosition = cIn;
    } else {
      in = new SplitLineReader(codec.createInputStream(fileIn,
          decompressor), job, this.recordDelimiterBytes);
      filePosition = fileIn;
    }
  } else {
    fileIn.seek(start);
    in = new SplitLineReader(fileIn, job, this.recordDelimiterBytes);
    filePosition = fileIn;
  }
  // If this is not the first split, we always throw away first record
  // because we always (except the last split) read one extra line in
  // next() method.
  if (start != 0) {
    start += in.readLine(new Text(), 0, maxBytesToConsume(start));
  }
  this.pos = start;
}
 
Example 6
Source File: AbstractEthereumRecordReader.java    From hadoopcryptoledger with Apache License 2.0 4 votes vote down vote up
/**
* Creates an Abstract Record Reader for Ethereum blocks
* @param split Split to use (assumed to be a file split)
* @param job Configuration:
 * io.file.buffer.size: Size of in-memory  specified in the given Configuration. If io.file.buffer.size is not specified the default buffersize will be used. Furthermore, one may specify hadoopcryptoledger.ethereumblockinputformat.maxblocksize, which defines the maximum size a Ethereum block may have. By default it is 1M). If you want to experiment with performance using DirectByteBuffer instead of HeapByteBuffer you can use "hadoopcryptoledeger.ethereumblockinputformat.usedirectbuffer" (default: false). Note that it might have some unwanted consequences such as circumwenting Yarn memory management. The option is experimental and might be removed in future versions. 
* @param reporter Reporter
*
*
* @throws java.io.IOException in case of errors reading from the filestream provided by Hadoop
*
*/
public AbstractEthereumRecordReader(FileSplit split,JobConf job, Reporter reporter) throws IOException {
    LOG.debug("Reading configuration");
    // parse configuration
     this.reporter=reporter;
     this.conf=job;	
	this.maxSizeEthereumBlock=conf.getInt(AbstractEthereumRecordReader.CONF_MAXBLOCKSIZE,AbstractEthereumRecordReader.DEFAULT_MAXSIZE_ETHEREUMBLOCK);
	this.bufferSize=conf.getInt(AbstractEthereumRecordReader.CONF_BUFFERSIZE,AbstractEthereumRecordReader.DEFAULT_BUFFERSIZE);

	this.useDirectBuffer=conf.getBoolean(AbstractEthereumRecordReader.CONF_USEDIRECTBUFFER,AbstractEthereumRecordReader.DEFAULT_USEDIRECTBUFFER);
    // Initialize start and end of split
    start = split.getStart();
    end = start + split.getLength();
    final Path file = split.getPath();
    codec = new CompressionCodecFactory(job).getCodec(file);
    final FileSystem fs = file.getFileSystem(job);
    fileIn = fs.open(file);
    // open stream
      if (isCompressedInput()) { // decompress
	LOG.debug("Decompressing file");
      	decompressor = CodecPool.getDecompressor(codec);
      	if (codec instanceof SplittableCompressionCodec) {
		LOG.debug("SplittableCompressionCodec");
        	final SplitCompressionInputStream cIn =((SplittableCompressionCodec)codec).createInputStream(fileIn, decompressor, start, end,SplittableCompressionCodec.READ_MODE.CONTINUOUS);
        	ebr = new EthereumBlockReader(cIn, this.maxSizeEthereumBlock,this.bufferSize,this.useDirectBuffer);
  			start = cIn.getAdjustedStart();
       		end = cIn.getAdjustedEnd();
        	filePosition = cIn; // take pos from compressed stream
      } else {
	LOG.debug("Not-splitable compression codec");
  	ebr = new EthereumBlockReader(codec.createInputStream(fileIn,decompressor), this.maxSizeEthereumBlock,this.bufferSize,this.useDirectBuffer);
            filePosition = fileIn;
      }
    } else {
      LOG.debug("Processing file without compression");
      fileIn.seek(start);
      ebr = new EthereumBlockReader(fileIn, this.maxSizeEthereumBlock,this.bufferSize,this.useDirectBuffer);
	     filePosition = fileIn;
    }
    // initialize reader

    this.reporter.setStatus("Ready to read");
}
 
Example 7
Source File: AbstractBitcoinRecordReader.java    From hadoopcryptoledger with Apache License 2.0 4 votes vote down vote up
/**
* Creates an Abstract Record Reader for Bitcoin blocks
* @param split Split to use (assumed to be a file split)
* @param job Configuration:
* io.file.buffer.size: Size of in-memory  specified in the given Configuration. If io.file.buffer.size is not specified the default buffersize (maximum size of a bitcoin block) will be used. The configuration hadoopcryptoledger.bitcoinblockinputformat.filter.magic allows specifying the magic identifier of the block. The magic is a comma-separated list of Hex-values (e.g. F9BEB4D9,FABFB5DA,0B110907,0B110907). The default magic is always F9BEB4D9. One needs to specify at least one magic, otherwise it will be difficult to find blocks in splits. Furthermore, one may specify hadoopcryptoledger.bitcoinblockinputformat.maxblocksize, which defines the maximum size a bitcoin block may have. By default it is 8M). If you want to experiment with performance using DirectByteBuffer instead of HeapByteBuffer you can use "hadoopcryptoledeger.bitcoinblockinputformat.usedirectbuffer" (default: false). Note that it might have some unwanted consequences such as circumwenting Yarn memory management. The option is experimental and might be removed in future versions. 
* @param reporter Reporter
*
*
* @throws java.io.IOException in case of errors reading from the filestream provided by Hadoop
* @throws org.zuinnote.hadoop.bitcoin.format.exception.HadoopCryptoLedgerConfigurationException in case of an invalid HadoopCryptoLedger-specific configuration of the inputformat
* @throws org.zuinnote.hadoop.bitcoin.format.exception.BitcoinBlockReadException in case the Bitcoin data contains invalid blocks (e.g. magic might be different)
*
*/
public AbstractBitcoinRecordReader(FileSplit split,JobConf job, Reporter reporter) throws IOException,HadoopCryptoLedgerConfigurationException,BitcoinBlockReadException {
    LOG.debug("Reading configuration");
    // parse configuration
     this.reporter=reporter;
     this.conf=job;	
	this.maxSizeBitcoinBlock=conf.getInt(AbstractBitcoinRecordReader.CONF_MAXBLOCKSIZE,AbstractBitcoinRecordReader.DEFAULT_MAXSIZE_BITCOINBLOCK);
	this.bufferSize=conf.getInt(AbstractBitcoinRecordReader.CONF_BUFFERSIZE,AbstractBitcoinRecordReader.DEFAULT_BUFFERSIZE);
	this.specificMagic=conf.get(AbstractBitcoinRecordReader.CONF_FILTERMAGIC);
	// we need to provide at least 
	if ((this.specificMagic==null) || (this.specificMagic.length()==0)) {
		 this.specificMagic=AbstractBitcoinRecordReader.DEFAULT_MAGIC;
	}
	if ((this.specificMagic!=null) && (this.specificMagic.length()>0)) {
		this.specificMagicStringArray=specificMagic.split(",");
		specificMagicByteArray=new byte[specificMagicStringArray.length][4]; // each magic is always 4 byte
		for (int i=0;i<specificMagicStringArray.length;i++) {
				byte[] currentMagicNo=BitcoinUtil.convertHexStringToByteArray(specificMagicStringArray[i]);
				if (currentMagicNo.length!=4) {
					throw new HadoopCryptoLedgerConfigurationException("Error: Configuration. Magic number has not a length of 4 bytes. Index: "+i);
				}
				specificMagicByteArray[i]=currentMagicNo;
		}
	}	
	this.useDirectBuffer=conf.getBoolean(AbstractBitcoinRecordReader.CONF_USEDIRECTBUFFER,AbstractBitcoinRecordReader.DEFAULT_USEDIRECTBUFFER);
	this.readAuxPOW=conf.getBoolean(AbstractBitcoinRecordReader.CONF_READAUXPOW,AbstractBitcoinRecordReader.DEFAULT_READAUXPOW);
    // Initialize start and end of split
    start = split.getStart();
    end = start + split.getLength();
    final Path file = split.getPath();
    codec = new CompressionCodecFactory(job).getCodec(file);
    final FileSystem fs = file.getFileSystem(job);
    fileIn = fs.open(file);
    // open stream
      if (isCompressedInput()) { // decompress
	LOG.debug("Decompressing file");
      	decompressor = CodecPool.getDecompressor(codec);
      	if (codec instanceof SplittableCompressionCodec) {
		LOG.debug("SplittableCompressionCodec");
        	final SplitCompressionInputStream cIn =((SplittableCompressionCodec)codec).createInputStream(fileIn, decompressor, start, end,SplittableCompressionCodec.READ_MODE.CONTINUOUS);
		bbr = new BitcoinBlockReader(cIn, this.maxSizeBitcoinBlock,this.bufferSize,this.specificMagicByteArray,this.useDirectBuffer,this.readAuxPOW);  
		start = cIn.getAdjustedStart();
       		end = cIn.getAdjustedEnd();
        	filePosition = cIn; // take pos from compressed stream
      } else {
	LOG.debug("Not-splitable compression codec");
	bbr = new BitcoinBlockReader(codec.createInputStream(fileIn,decompressor), this.maxSizeBitcoinBlock,this.bufferSize,this.specificMagicByteArray,this.useDirectBuffer,this.readAuxPOW);
        filePosition = fileIn;
      }
    } else {
      LOG.debug("Processing file without compression");
      fileIn.seek(start);
      bbr = new BitcoinBlockReader(fileIn, this.maxSizeBitcoinBlock,this.bufferSize,this.specificMagicByteArray,this.useDirectBuffer,this.readAuxPOW);  
      filePosition = fileIn;
    }
    // initialize reader
    // seek to block start (for the case a block overlaps a split)
    LOG.debug("Seeking to block start");
    this.reporter.setStatus("Seeking Block start");
    bbr.seekBlockStart();
    this.reporter.setStatus("Ready to read");
}