public void initialize(InputSplit genericSplit, TaskAttemptContext context) throws IOException {
FileSplit split = (FileSplit) genericSplit;
Configuration job = context.getConfiguration();
m_Sb.setLength(0);
m_Start = split.getStart();
m_End = m_Start + split.getLength();
final Path file = split.getPath();
compressionCodecs = new CompressionCodecFactory(job);
final CompressionCodec codec = compressionCodecs.getCodec(file);
// open the file and seek to the m_Start of the split
FileSystem fs = file.getFileSystem(job);
// getFileStatus fileStatus = fs.getFileStatus(split.getPath());
//noinspection deprecation
@SuppressWarnings(value = "deprecated")
long length = fs.getLength(file);
FSDataInputStream fileIn = fs.open(split.getPath());
if (m_Start > 0) fileIn.seek(m_Start);
if (codec != null) {
CompressionInputStream inputStream = codec.createInputStream(fileIn);
m_Input = new BufferedReader(new InputStreamReader(inputStream));
m_End = length;
} else {
m_Input = new BufferedReader(new InputStreamReader(fileIn));
}
m_Current = m_Start;
m_Key = split.getPath().getName();
}
HarFsInputStream(FileSystem fs, Path path, long start, long length, int bufferSize)
throws IOException {
if (length < 0) {
throw new IllegalArgumentException("Negative length [" + length + "]");
}
underLyingStream = fs.open(path, bufferSize);
underLyingStream.seek(start);
// the start of this file in the part file
this.start = start;
// the position pointer in the part file
this.position = start;
// the end pointer in the part file
this.end = start + length;
}
@Override
public synchronized long skip(long n) throws IOException {
long tmpN = n;
if (tmpN > 0) {
final long actualRemaining = end - position;
if (tmpN > actualRemaining) {
tmpN = actualRemaining;
}
underLyingStream.seek(tmpN + position);
position += tmpN;
return tmpN;
}
// NB: the contract is described in java.io.InputStream.skip(long):
// this method returns the number of bytes actually skipped, so,
// the return value should never be negative.
return 0;
}
@Override
public synchronized void seek(final long pos) throws IOException {
validatePosition(pos);
position = start + pos;
underLyingStream.seek(position);
}
private void parseMetaData() throws IOException {
Text line = new Text();
long read;
FSDataInputStream in = null;
LineReader lin = null;
try {
in = fs.open(masterIndexPath);
FileStatus masterStat = fs.getFileStatus(masterIndexPath);
masterIndexTimestamp = masterStat.getModificationTime();
lin = new LineReader(in, getConf());
read = lin.readLine(line);
// the first line contains the version of the index file
String versionLine = line.toString();
String[] arr = versionLine.split(" ");
version = Integer.parseInt(arr[0]);
// make it always backwards-compatible
if (this.version > HarFileSystem.VERSION) {
throw new IOException(
"Invalid version " + this.version + " expected " + HarFileSystem.VERSION);
}
// each line contains a hashcode range and the index file name
String[] readStr;
while (read < masterStat.getLen()) {
int b = lin.readLine(line);
read += b;
readStr = line.toString().split(" ");
int startHash = Integer.parseInt(readStr[0]);
int endHash = Integer.parseInt(readStr[1]);
stores.add(
new Store(
Long.parseLong(readStr[2]), Long.parseLong(readStr[3]), startHash, endHash));
line.clear();
}
} catch (IOException ioe) {
LOG.warn("Encountered exception ", ioe);
throw ioe;
} finally {
IOUtils.cleanup(LOG, lin, in);
}
FSDataInputStream aIn = fs.open(archiveIndexPath);
try {
FileStatus archiveStat = fs.getFileStatus(archiveIndexPath);
archiveIndexTimestamp = archiveStat.getModificationTime();
LineReader aLin;
// now start reading the real index file
for (Store s : stores) {
read = 0;
aIn.seek(s.begin);
aLin = new LineReader(aIn, getConf());
while (read + s.begin < s.end) {
int tmp = aLin.readLine(line);
read += tmp;
String lineFeed = line.toString();
String[] parsed = lineFeed.split(" ");
parsed[0] = decodeFileName(parsed[0]);
archive.put(new Path(parsed[0]), new HarStatus(lineFeed));
line.clear();
}
}
} finally {
IOUtils.cleanup(LOG, aIn);
}
}