in src/main/java/com/hadoop/mapreduce/LzoLineRecordReader.java [88:117]
public void initialize(InputSplit genericSplit, TaskAttemptContext context) throws IOException, InterruptedException {
FileSplit split = (FileSplit) genericSplit;
start = split.getStart();
end = start + split.getLength();
final Path file = split.getPath();
Configuration job = CompatibilityUtil.getConfiguration(context);
FileSystem fs = file.getFileSystem(job);
CompressionCodecFactory compressionCodecs = new CompressionCodecFactory(job);
final CompressionCodec codec = compressionCodecs.getCodec(file);
if (codec == null) {
throw new IOException("Codec for file " + file + " not found, cannot run");
}
// open the file and seek to the start of the split
fileIn = fs.open(split.getPath());
// creates input stream and also reads the file header
in = new LineReader(codec.createInputStream(fileIn), job);
if (start != 0) {
fileIn.seek(start);
// read and ignore the first line
in.readLine(new Text());
start = fileIn.getPos();
}
this.pos = start;
}