List of usage examples for org.apache.hadoop.mapreduce Mapper run
public void run(Context context) throws IOException, InterruptedException
From source file:org.apache.tez.mapreduce.processor.map.MapProcessor.java
License:Apache License
private void runNewMapper(final JobConf job, MRTaskReporter reporter, final MRInputLegacy in, KeyValueWriter out) throws IOException, InterruptedException { // Initialize input in-line since it sets parameters which may be used by the processor. // Done only for MRInput. // TODO use new method in MRInput to get required info //in.initialize(job, master); // make a task context so we can get the classes org.apache.hadoop.mapreduce.TaskAttemptContext taskContext = getTaskAttemptContext(); // make a mapper org.apache.hadoop.mapreduce.Mapper mapper; try {//from w ww . j ava 2 s. com mapper = (org.apache.hadoop.mapreduce.Mapper) ReflectionUtils.newInstance(taskContext.getMapperClass(), job); } catch (ClassNotFoundException cnfe) { throw new IOException(cnfe); } org.apache.hadoop.mapreduce.RecordReader input = new NewRecordReader(in); org.apache.hadoop.mapreduce.RecordWriter output = new NewOutputCollector(out); org.apache.hadoop.mapreduce.InputSplit split = in.getNewInputSplit(); updateJobWithSplit(job, split); org.apache.hadoop.mapreduce.MapContext mapContext = new MapContextImpl(job, taskAttemptId, input, output, committer, processorContext, split, reporter); org.apache.hadoop.mapreduce.Mapper.Context mapperContext = new WrappedMapper().getMapContext(mapContext); input.initialize(split, mapperContext); mapper.run(mapperContext); // Set progress to 1.0f if there was no exception, reporter.setProgress(1.0f); this.statusUpdate(); input.close(); output.close(mapperContext); }
From source file:org.gridgain.grid.kernal.processors.hadoop.v2.GridHadoopV2MapTask.java
License:Open Source License
/** {@inheritDoc} */ @SuppressWarnings({ "ConstantConditions", "unchecked" }) @Override// w w w. jav a2 s .c om public void run0(GridHadoopV2TaskContext taskCtx) throws GridException { GridHadoopInputSplit split = info().inputSplit(); InputSplit nativeSplit; if (split instanceof GridHadoopFileBlock) { GridHadoopFileBlock block = (GridHadoopFileBlock) split; nativeSplit = new FileSplit(new Path(block.file().toString()), block.start(), block.length(), null); } else nativeSplit = (InputSplit) taskCtx.getNativeSplit(split); assert nativeSplit != null; OutputFormat outputFormat = null; Exception err = null; JobContextImpl jobCtx = taskCtx.jobContext(); try { InputFormat inFormat = ReflectionUtils.newInstance(jobCtx.getInputFormatClass(), hadoopContext().getConfiguration()); RecordReader reader = inFormat.createRecordReader(nativeSplit, hadoopContext()); reader.initialize(nativeSplit, hadoopContext()); hadoopContext().reader(reader); GridHadoopJobInfo jobInfo = taskCtx.job().info(); outputFormat = jobInfo.hasCombiner() || jobInfo.hasReducer() ? null : prepareWriter(jobCtx); Mapper mapper = ReflectionUtils.newInstance(jobCtx.getMapperClass(), hadoopContext().getConfiguration()); try { mapper.run(new WrappedMapper().getMapContext(hadoopContext())); } finally { closeWriter(); } commit(outputFormat); } catch (InterruptedException e) { err = e; Thread.currentThread().interrupt(); throw new GridInterruptedException(e); } catch (Exception e) { err = e; throw new GridException(e); } finally { if (err != null) abort(outputFormat); } }
From source file:org.warcbase.mapreduce.lib.Chain.java
License:Apache License
@SuppressWarnings("unchecked") void runMapper(TaskInputOutputContext context, int index) throws IOException, InterruptedException { Mapper mapper = mappers.get(index); RecordReader rr = new ChainRecordReader(context); RecordWriter rw = new ChainRecordWriter(context); Mapper.Context mapperContext = createMapContext(rr, rw, context, getConf(index)); mapper.run(mapperContext); rr.close();/*from ww w. j ava2 s . c o m*/ rw.close(context); }