List of usage examples for org.apache.hadoop.mapreduce Job getOutputFormatClass
@SuppressWarnings("unchecked") public Class<? extends OutputFormat<?, ?>> getOutputFormatClass() throws ClassNotFoundException
From source file:com.moz.fiji.mapreduce.output.TestFileMapReduceJobOutput.java
License:Apache License
@Test public void testConfigure() throws ClassNotFoundException, IOException { final Path filePath = new Path("foo/bar"); final int numSplits = 42; final Class<? extends OutputFormat> outputFormatClass = TextOutputFormat.class; FileMapReduceJobOutput jobOutput = new ConcreteFileMapReduceJobOutput(filePath, numSplits, outputFormatClass);//w w w. j a v a2s . c o m Job job = new Job(); jobOutput.configure(job); // The output format class should be set in the job configuration. assertEquals(outputFormatClass, job.getOutputFormatClass()); // The file output path should be set in the job configuration. assert (FileOutputFormat.getOutputPath(job).toString().endsWith(filePath.toString())); // The number of reduce tasks should be set to the number of splits. assertEquals(numSplits, job.getNumReduceTasks()); }
From source file:com.moz.fiji.mapreduce.TestFijiBulkImportJobBuilder.java
License:Apache License
@Test public void testBuildWithHFileOutput() throws Exception { final FijiMapReduceJob mrjob = FijiBulkImportJobBuilder.create().withConf(getConf()) .withInput(MapReduceJobInputs.newTextMapReduceJobInput(new Path(mTempPath, "input"))) .withBulkImporter(NoopBulkImporter.class).withOutput(MapReduceJobOutputs .newHFileMapReduceJobOutput(mTable.getURI(), new Path(mTempPath, "output"), 10)) .build();// w ww . ja v a 2s . com final Job job = mrjob.getHadoopJob(); assertEquals(TextInputFormat.class, job.getInputFormatClass()); assertEquals(BulkImportMapper.class, job.getMapperClass()); assertEquals(NoopBulkImporter.class, job.getConfiguration().getClass(FijiConfKeys.FIJI_BULK_IMPORTER_CLASS, null)); assertEquals(IdentityReducer.class, job.getReducerClass()); assertEquals(10, job.getNumReduceTasks()); assertEquals(FijiHFileOutputFormat.class, job.getOutputFormatClass()); assertEquals(TotalOrderPartitioner.class, job.getPartitionerClass()); }
From source file:com.moz.fiji.mapreduce.TestFijiBulkImportJobBuilder.java
License:Apache License
@Test public void testBuildWithKeyValueStore() throws Exception { final FijiMapReduceJob mrjob = FijiBulkImportJobBuilder.create().withConf(getConf()) .withInput(MapReduceJobInputs.newTextMapReduceJobInput(new Path(mTempPath, "input"))) .withBulkImporter(KVStoreBulkImporter.class).withOutput(MapReduceJobOutputs .newHFileMapReduceJobOutput(mTable.getURI(), new Path(mTempPath, "output"), 10)) .build();/*from www . j a va 2 s.c o m*/ final Job job = mrjob.getHadoopJob(); // Verify that everything else is what we expected as in the previous test // (except the bulk importer class name)... assertEquals(TextInputFormat.class, job.getInputFormatClass()); assertEquals(BulkImportMapper.class, job.getMapperClass()); assertEquals(KVStoreBulkImporter.class, job.getConfiguration().getClass(FijiConfKeys.FIJI_BULK_IMPORTER_CLASS, null)); assertEquals(IdentityReducer.class, job.getReducerClass()); assertEquals(10, job.getNumReduceTasks()); assertEquals(FijiHFileOutputFormat.class, job.getOutputFormatClass()); assertEquals(TotalOrderPartitioner.class, job.getPartitionerClass()); // KeyValueStore-specific checks here. final Configuration confOut = job.getConfiguration(); assertEquals(1, confOut.getInt(KeyValueStoreConfigSerializer.CONF_KEY_VALUE_STORE_COUNT, 0)); assertEquals(EmptyKeyValueStore.class.getName(), confOut.get(KeyValueStoreConfiguration.KEY_VALUE_STORE_NAMESPACE + "0." + KeyValueStoreConfigSerializer.CONF_CLASS)); assertEquals("foostore", confOut.get(KeyValueStoreConfiguration.KEY_VALUE_STORE_NAMESPACE + "0." + KeyValueStoreConfigSerializer.CONF_NAME)); }
From source file:com.moz.fiji.mapreduce.TestFijiGatherJobBuilder.java
License:Apache License
@Test public void testGatherToHFile() throws Exception { final FijiMapReduceJob gatherJob = FijiGatherJobBuilder.create().withConf(getConf()) .withInputTable(mTable.getURI()).withGatherer(GatherToHFile.class).withOutput(MapReduceJobOutputs .newHFileMapReduceJobOutput(mTable.getURI(), getLocalTestPath("hfile"), 10)) .build();/*w w w. ja v a 2 s . com*/ final Job job = gatherJob.getHadoopJob(); final Configuration conf = job.getConfiguration(); assertEquals(GatherToHFile.class.getName(), conf.get(FijiConfKeys.FIJI_GATHERER_CLASS)); assertEquals(null, job.getCombinerClass()); assertEquals(IdentityReducer.class, job.getReducerClass()); assertEquals(10, job.getNumReduceTasks()); assertEquals(FijiHFileOutputFormat.class, job.getOutputFormatClass()); assertEquals(HFileKeyValue.class, job.getOutputKeyClass()); assertEquals(NullWritable.class, job.getOutputValueClass()); }
From source file:com.moz.fiji.mapreduce.TestFijiGatherJobBuilder.java
License:Apache License
@Test public void testGatherReducerToHFile() throws Exception { final FijiMapReduceJob gatherJob = FijiGatherJobBuilder.create().withConf(getConf()) .withInputTable(mTable.getURI()).withGatherer(SimpleGatherer.class) .withReducer(ReducerToHFile.class).withOutput(MapReduceJobOutputs .newHFileMapReduceJobOutput(mTable.getURI(), getLocalTestPath("hfile"), 10)) .build();/*from w ww.j a v a 2s . c o m*/ final Job job = gatherJob.getHadoopJob(); final Configuration conf = job.getConfiguration(); assertEquals(SimpleGatherer.class.getName(), conf.get(FijiConfKeys.FIJI_GATHERER_CLASS)); assertEquals(null, job.getCombinerClass()); assertEquals(ReducerToHFile.class, job.getReducerClass()); assertEquals(10, job.getNumReduceTasks()); assertEquals(SequenceFileOutputFormat.class, job.getOutputFormatClass()); assertEquals(HFileKeyValue.class, job.getOutputKeyClass()); assertEquals(NullWritable.class, job.getOutputValueClass()); }
From source file:com.moz.fiji.mapreduce.TestFijiMapReduceJobBuilder.java
License:Apache License
@Test public void testBuild() throws Exception { final FijiMapReduceJob job = FijiMapReduceJobBuilder.create().withConf(mConf) .withInput(MapReduceJobInputs.newTextMapReduceJobInput(new Path("/path/to/my/input"))) .withMapper(MyMapper.class).withReducer(MyReducer.class) .withOutput(MapReduceJobOutputs.newTextMapReduceJobOutput(new Path("/path/to/my/output"), 16)) .build();//from w w w .ja v a 2 s .c o m final Job hadoopJob = job.getHadoopJob(); assertEquals(TextInputFormat.class, hadoopJob.getInputFormatClass()); assertEquals(MyMapper.class, hadoopJob.getMapperClass()); assertEquals(MyReducer.class, hadoopJob.getReducerClass()); assertEquals(16, hadoopJob.getNumReduceTasks()); assertEquals(TextOutputFormat.class, hadoopJob.getOutputFormatClass()); // KeyValueStore-specific checks here. Configuration confOut = hadoopJob.getConfiguration(); assertEquals(2, confOut.getInt(KeyValueStoreConfigSerializer.CONF_KEY_VALUE_STORE_COUNT, 0)); assertEquals(EmptyKeyValueStore.class.getName(), confOut.get(KeyValueStoreConfiguration.KEY_VALUE_STORE_NAMESPACE + "0." + KeyValueStoreConfigSerializer.CONF_CLASS)); assertEquals("mapperMap", confOut.get(KeyValueStoreConfiguration.KEY_VALUE_STORE_NAMESPACE + "0." + KeyValueStoreConfigSerializer.CONF_NAME)); assertEquals(EmptyKeyValueStore.class.getName(), confOut.get(KeyValueStoreConfiguration.KEY_VALUE_STORE_NAMESPACE + "1." + KeyValueStoreConfigSerializer.CONF_CLASS)); assertEquals("reducerMap", confOut.get(KeyValueStoreConfiguration.KEY_VALUE_STORE_NAMESPACE + "1." + KeyValueStoreConfigSerializer.CONF_NAME)); }
From source file:com.moz.fiji.mapreduce.TestFijiMapReduceJobBuilder.java
License:Apache License
@Test public void testBuildWithXmlKVStores() throws Exception { // Test that we can override default configuration KeyValueStores from an XML file. final InputStream xmlStores = Resources.openSystemResource("com.moz.fiji/mapreduce/test-kvstores.xml"); // This file needs to exist before we build the job, or else // we can't build the job; it's referenced by a key-value store that checks // for its presence. final File tmpFile = new File("/tmp/foo.seq"); if (tmpFile.createNewFile()) { // We created this temp file, we're responsible for deleting it. tmpFile.deleteOnExit();/* www .ja va 2 s . com*/ } LOG.info("Building job..."); final FijiMapReduceJob job = FijiMapReduceJobBuilder.create().withConf(mConf) .withInput(MapReduceJobInputs.newTextMapReduceJobInput(new Path("/path/to/my/input"))) .withMapper(MyMapper.class).withReducer(MyReducer.class) .withOutput(MapReduceJobOutputs.newTextMapReduceJobOutput(new Path("/path/to/my/output"), 16)) .withStoreBindings(xmlStores).build(); xmlStores.close(); LOG.info("Verifying job configuration..."); final Job hadoopJob = job.getHadoopJob(); assertEquals(TextInputFormat.class, hadoopJob.getInputFormatClass()); assertEquals(MyMapper.class, hadoopJob.getMapperClass()); assertEquals(MyReducer.class, hadoopJob.getReducerClass()); assertEquals(16, hadoopJob.getNumReduceTasks()); assertEquals(TextOutputFormat.class, hadoopJob.getOutputFormatClass()); // KeyValueStore-specific checks here. // We override mapperMap with a SeqFileKeyValueStore. Configuration confOut = hadoopJob.getConfiguration(); assertEquals(2, confOut.getInt(KeyValueStoreConfigSerializer.CONF_KEY_VALUE_STORE_COUNT, 0)); assertEquals(SeqFileKeyValueStore.class.getName(), confOut.get(KeyValueStoreConfiguration.KEY_VALUE_STORE_NAMESPACE + "0." + KeyValueStoreConfigSerializer.CONF_CLASS)); assertEquals("mapperMap", confOut.get(KeyValueStoreConfiguration.KEY_VALUE_STORE_NAMESPACE + "0." + KeyValueStoreConfigSerializer.CONF_NAME)); assertEquals(EmptyKeyValueStore.class.getName(), confOut.get(KeyValueStoreConfiguration.KEY_VALUE_STORE_NAMESPACE + "1." + KeyValueStoreConfigSerializer.CONF_CLASS)); assertEquals("reducerMap", confOut.get(KeyValueStoreConfiguration.KEY_VALUE_STORE_NAMESPACE + "1." + KeyValueStoreConfigSerializer.CONF_NAME)); }
From source file:com.moz.fiji.mapreduce.TestFijiProduceJobBuilder.java
License:Apache License
@Test public void testBuildWithHFileOutput() throws ClassNotFoundException, IOException { final FijiMapReduceJob produceJob = FijiProduceJobBuilder.create().withConf(getConf()) .withInputTable(mTable.getURI()).withProducer(MyProducer.class) .withOutput(// www. j a va2s . c o m MapReduceJobOutputs.newHFileMapReduceJobOutput(mTable.getURI(), new Path("foo/bar"), 10)) .build(); // Verify that the MR Job was configured correctly. final Job job = produceJob.getHadoopJob(); assertEquals(HBaseFijiTableInputFormat.class, job.getInputFormatClass()); assertEquals(ProduceMapper.class, job.getMapperClass()); assertEquals(MyProducer.class, job.getConfiguration().getClass(FijiConfKeys.FIJI_PRODUCER_CLASS, null)); assertEquals(10, job.getNumReduceTasks()); assertEquals(FijiHFileOutputFormat.class, job.getOutputFormatClass()); }
From source file:com.splicemachine.mrio.api.SpliceTableMapReduceUtil.java
License:Apache License
/** * Add the dependency jars as well as jars for any of the configured * job classes to the job configuration, so that JobClient will ship them * to the cluster and add them to the DistributedCache. */// w w w . j av a2s .c om public static void addDependencyJars(Job job) throws IOException { try { addDependencyJars(job.getConfiguration(), org.apache.zookeeper.ZooKeeper.class, com.google.protobuf.Message.class, com.google.common.collect.ImmutableSet.class, job.getMapOutputKeyClass(), job.getMapOutputValueClass(), job.getInputFormatClass(), job.getOutputKeyClass(), job.getOutputValueClass(), job.getOutputFormatClass(), job.getPartitionerClass(), job.getCombinerClass()); } catch (ClassNotFoundException e) { throw new IOException(e); } }
From source file:com.xiaomi.linden.hadoop.indexing.job.LindenJob.java
License:Apache License
@Override public int run(String[] strings) throws Exception { Configuration conf = getConf(); String dir = conf.get(LindenJobConfig.INPUT_DIR, null); logger.info("input dir:" + dir); Path inputPath = new Path(StringUtils.unEscapeString(dir)); Path outputPath = new Path(conf.get(LindenJobConfig.OUTPUT_DIR)); String indexPath = conf.get(LindenJobConfig.INDEX_PATH); FileSystem fs = FileSystem.get(conf); if (fs.exists(outputPath)) { fs.delete(outputPath, true);/*from www .ja v a 2 s . c o m*/ } if (fs.exists(new Path(indexPath))) { fs.delete(new Path(indexPath), true); } int numShards = conf.getInt(LindenJobConfig.NUM_SHARDS, 1); Shard[] shards = createShards(indexPath, numShards); Shard.setIndexShards(conf, shards); //empty trash; (new Trash(conf)).expunge(); Job job = Job.getInstance(conf, "linden-hadoop-indexing"); job.setJarByClass(LindenJob.class); job.setMapperClass(LindenMapper.class); job.setCombinerClass(LindenCombiner.class); job.setReducerClass(LindenReducer.class); job.setMapOutputKeyClass(Shard.class); job.setMapOutputValueClass(IntermediateForm.class); job.setOutputKeyClass(Shard.class); job.setOutputValueClass(Text.class); job.setInputFormatClass(TextInputFormat.class); job.setOutputFormatClass(IndexUpdateOutputFormat.class); job.setReduceSpeculativeExecution(false); job.setNumReduceTasks(numShards); String lindenSchemaFile = conf.get(LindenJobConfig.SCHEMA_FILE_URL); if (lindenSchemaFile == null) { throw new IOException("no schema file is found"); } logger.info("Adding schema file: " + lindenSchemaFile); job.addCacheFile(new URI(lindenSchemaFile + "#lindenSchema")); String lindenPropertiesFile = conf.get(LindenJobConfig.LINDEN_PROPERTIES_FILE_URL); if (lindenPropertiesFile == null) { throw new IOException("no linden properties file is found"); } logger.info("Adding linden properties file: " + lindenPropertiesFile); job.addCacheFile(new URI(lindenPropertiesFile + "#lindenProperties")); FileInputFormat.setInputPaths(job, inputPath); FileOutputFormat.setOutputPath(job, outputPath); Path[] inputs = FileInputFormat.getInputPaths(job); StringBuilder buffer = new StringBuilder(inputs[0].toString()); for (int i = 1; i < inputs.length; i++) { buffer.append(","); buffer.append(inputs[i].toString()); } logger.info("mapreduce.input.dir = " + buffer.toString()); logger.info("mapreduce.output.dir = " + FileOutputFormat.getOutputPath(job).toString()); logger.info("mapreduce.job.num.reduce.tasks = " + job.getNumReduceTasks()); logger.info(shards.length + " shards = " + conf.get(LindenJobConfig.INDEX_SHARDS)); logger.info("mapreduce.input.format.class = " + job.getInputFormatClass()); logger.info("mapreduce.output.format.class = " + job.getOutputFormatClass()); logger.info("mapreduce.cluster.temp.dir = " + conf.get(MRJobConfig.TEMP_DIR)); job.waitForCompletion(true); if (!job.isSuccessful()) { throw new RuntimeException("Job failed"); } return 0; }