Java Code Examples for org.apache.hadoop.mapreduce.lib.input.CombineFileSplit#getPath()
The following examples show how to use
org.apache.hadoop.mapreduce.lib.input.CombineFileSplit#getPath() .
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: MultiFileWordCount.java From hadoop with Apache License 2.0 | 6 votes |
public CombineFileLineRecordReader(CombineFileSplit split, TaskAttemptContext context, Integer index) throws IOException { this.path = split.getPath(index); fs = this.path.getFileSystem(context.getConfiguration()); this.startOffset = split.getOffset(index); this.end = startOffset + split.getLength(index); boolean skipFirstLine = false; //open the file fileIn = fs.open(path); if (startOffset != 0) { skipFirstLine = true; --startOffset; fileIn.seek(startOffset); } reader = new LineReader(fileIn); if (skipFirstLine) { // skip first line and re-establish "startOffset". startOffset += reader.readLine(new Text(), 0, (int)Math.min((long)Integer.MAX_VALUE, end - startOffset)); } this.pos = startOffset; }
Example 2
Source File: MultiFileWordCount.java From big-c with Apache License 2.0 | 6 votes |
public CombineFileLineRecordReader(CombineFileSplit split, TaskAttemptContext context, Integer index) throws IOException { this.path = split.getPath(index); fs = this.path.getFileSystem(context.getConfiguration()); this.startOffset = split.getOffset(index); this.end = startOffset + split.getLength(index); boolean skipFirstLine = false; //open the file fileIn = fs.open(path); if (startOffset != 0) { skipFirstLine = true; --startOffset; fileIn.seek(startOffset); } reader = new LineReader(fileIn); if (skipFirstLine) { // skip first line and re-establish "startOffset". startOffset += reader.readLine(new Text(), 0, (int)Math.min((long)Integer.MAX_VALUE, end - startOffset)); } this.pos = startOffset; }
Example 3
Source File: HalyardBulkLoad.java From Halyard with Apache License 2.0 | 6 votes |
@Override public List<InputSplit> getSplits(JobContext job) throws IOException { List<InputSplit> splits = super.getSplits(job); long maxSize = MAX_SINGLE_FILE_MULTIPLIER * job.getConfiguration().getLong("mapreduce.input.fileinputformat.split.maxsize", 0); if (maxSize > 0) { List<InputSplit> newSplits = new ArrayList<>(); for (InputSplit spl : splits) { CombineFileSplit cfs = (CombineFileSplit)spl; for (int i=0; i<cfs.getNumPaths(); i++) { long length = cfs.getLength(); if (length > maxSize) { int replicas = (int)Math.ceil((double)length / (double)maxSize); Path path = cfs.getPath(i); for (int r=1; r<replicas; r++) { newSplits.add(new CombineFileSplit(new Path[]{path}, new long[]{r}, new long[]{length}, cfs.getLocations())); } } } } splits.addAll(newSplits); } return splits; }
Example 4
Source File: AbstractCombineFileRecordReader.java From kite with Apache License 2.0 | 6 votes |
@Override public void initialize(InputSplit split, TaskAttemptContext context) throws IOException, InterruptedException { if (delegate != null) { delegate.close(); } if (split instanceof CombineFileSplit) { CombineFileSplit combineSplit = (CombineFileSplit) split; FileSplit fileSplit = new FileSplit(combineSplit.getPath(idx), combineSplit.getOffset(idx), combineSplit.getLength(idx), combineSplit.getLocations()); delegate = getInputFormat().createRecordReader(fileSplit, context); delegate.initialize(fileSplit, context); } else { throw new DatasetOperationException( "Split is not a CombineFileSplit: %s:%s", split.getClass().getCanonicalName(), split); } }
Example 5
Source File: MDSCombineSpreadReader.java From multiple-dimension-spread with Apache License 2.0 | 5 votes |
public MDSCombineSpreadReader( final CombineFileSplit split , final TaskAttemptContext context , final Integer index ) throws IOException{ Configuration config = context.getConfiguration(); Path path = split.getPath( index ); FileSystem fs = path.getFileSystem( config ); long fileLength = fs.getLength( path ); InputStream in = fs.open( path ); innerReader = new MDSSpreadReader(); innerReader.setStream( in , fileLength , 0 , fileLength ); }
Example 6
Source File: AvroKeyCombineFileRecordReader.java From incubator-gobblin with Apache License 2.0 | 5 votes |
private static Schema getSchema(CombineFileSplit split, TaskAttemptContext cx, Integer idx) throws IOException { Schema schema = AvroJob.getInputKeySchema(cx.getConfiguration()); if (schema != null) { return schema; } Path path = split.getPath(idx); FileSystem fs = path.getFileSystem(cx.getConfiguration()); return AvroUtils.getSchemaFromDataFile(path, fs); }
Example 7
Source File: CsvBlurDriver.java From incubator-retired-blur with Apache License 2.0 | 5 votes |
@SuppressWarnings("unused") public SequenceFileRecordReaderWrapper(CombineFileSplit split, TaskAttemptContext context, Integer index) throws IOException { fileSplit = new FileSplit(split.getPath(index), split.getOffset(index), split.getLength(index), split.getLocations()); delegate = new SequenceFileInputFormat<Writable, Text>().createRecordReader(fileSplit, context); }
Example 8
Source File: CombinedSequenceFileRecordReader.java From datawave with Apache License 2.0 | 4 votes |
@Override public void initialize(InputSplit split, TaskAttemptContext context) throws IOException, InterruptedException { CombineFileSplit fSplit = (CombineFileSplit) split; super.initialize(new FileSplit(fSplit.getPath(index), 0, fSplit.getLength(index), fSplit.getLocations()), context); }
Example 9
Source File: OrcValueCombineFileRecordReader.java From incubator-gobblin with Apache License 2.0 | 4 votes |
private static TypeDescription getSchema(CombineFileSplit split, TaskAttemptContext context, Integer idx) throws IOException { Path path = split.getPath(idx); return OrcUtils.getTypeDescriptionFromFile(context.getConfiguration(), path); }
Example 10
Source File: OrcValueCombineFileRecordReader.java From incubator-gobblin with Apache License 2.0 | 4 votes |
private static RecordReader getRecordReaderFromFile(CombineFileSplit split, TaskAttemptContext context, Integer idx) throws IOException { Path path = split.getPath(idx); return OrcUtils.getRecordReaderFromFile(context.getConfiguration(), path).rows(); }