List of usage examples for org.apache.hadoop.mapreduce TaskAttemptContext getConfiguration
public Configuration getConfiguration();
From source file:com.conversantmedia.mapreduce.output.BloomFilterOutputFormat.java
License:Apache License
protected int getExpectedInsertions(TaskAttemptContext job) { return this.insertionSize > 0 ? this.insertionSize : job.getConfiguration().getInt(CONF_KEY_EXPECTED_INSERTIONS, -1); }
From source file:com.conversantmedia.mapreduce.tool.AnnotatedDelegatingCombiner.java
License:Apache License
@Override public Reducer<K1, V1, K2, V2> getDelegate(TaskAttemptContext context) { if (this.reducer == null) { Class<? extends Reducer> reducerClass = (Class<? extends Reducer>) context.getConfiguration() .getClass(CONFKEY_DELEGATE_COMBINER_CLASS, Reducer.class); reducer = (Reducer<K1, V1, K2, V2>) ReflectionUtils.newInstance(reducerClass, context.getConfiguration()); }//from www . j a v a2s . c o m return this.reducer; }
From source file:com.conversantmedia.mapreduce.tool.AnnotatedDelegatingMapper.java
License:Apache License
@Override public Mapper<K1, V1, K2, V2> getDelegate(TaskAttemptContext context) { if (mapper == null) { Class<? extends Mapper> mapperClass = (Class<? extends Mapper>) context.getConfiguration() .getClass(CONFKEY_DELEGATE_MAPPER_CLASS, Mapper.class); mapper = (Mapper<K1, V1, K2, V2>) ReflectionUtils.newInstance(mapperClass, context.getConfiguration()); }/*from www .j a va 2 s. co m*/ return this.mapper; }
From source file:com.conversantmedia.mapreduce.tool.AnnotatedDelegatingReducer.java
License:Apache License
@Override public Reducer<K1, V1, K2, V2> getDelegate(TaskAttemptContext context) { if (this.reducer == null) { Class<? extends Reducer> reducerClass = (Class<? extends Reducer>) context.getConfiguration() .getClass(CONFKEY_DELEGATE_REDUCER_CLASS, Reducer.class); reducer = (Reducer<K1, V1, K2, V2>) ReflectionUtils.newInstance(reducerClass, context.getConfiguration()); }/* w ww. j a va 2s . co m*/ return this.reducer; }
From source file:com.cotdp.hadoop.BrotliFileRecordReader.java
License:Apache License
/** * Initialize and open the ZIP file from the FileSystem *//*from w w w . j a v a 2s . c om*/ @Override public void initialize(InputSplit inputSplit, TaskAttemptContext taskAttemptContext) throws IOException, InterruptedException { FileSplit split = (FileSplit) inputSplit; Configuration conf = taskAttemptContext.getConfiguration(); Path path = split.getPath(); FileSystem fs = path.getFileSystem(conf); // Set the file path as the key currentKey.set(path.getName()); // Open the stream fsin = fs.open(path); String cmd = "/bin/cat"; ProcessBuilder pb = new ProcessBuilder(); pb.redirectOutput(); pb.command(cmd); try { decompressor = pb.start(); } catch (IOException e) { // TODO Auto-generated catch block e.printStackTrace(); } }
From source file:com.cotdp.hadoop.ZipFileRecordReader.java
License:Apache License
/** * Initialise and open the ZIP file from the FileSystem *//*from w w w. j a v a 2 s .com*/ public void initialize(InputSplit inputSplit, TaskAttemptContext taskAttemptContext) throws IOException, InterruptedException { FileSplit split = (FileSplit) inputSplit; Configuration conf = taskAttemptContext.getConfiguration(); Path path = split.getPath(); FileSystem fs = path.getFileSystem(conf); // Open the stream fsin = fs.open(path); zip = new ZipInputStream(fsin); try { entry = zip.getNextEntry(); } catch (ZipException e) { if (ZipFileInputFormat.getLenient() == false) throw e; } }
From source file:com.couchbase.sqoop.mapreduce.db.CouchbaseInputFormat.java
License:Apache License
@Override /** {@inheritDoc} */ public RecordReader<Text, T> createRecordReader(final InputSplit split, final TaskAttemptContext context) throws IOException, InterruptedException { System.out.printf("Creating Record Reader %s", split.getLocations().toString()); return createRecordReader(split, context.getConfiguration()); }
From source file:com.couchbase.sqoop.mapreduce.db.CouchbaseOutputFormat.java
License:Apache License
@Override public RecordWriter<K, V> getRecordWriter(final TaskAttemptContext context) throws IOException, InterruptedException { return new CouchbaseRecordWriter(new CouchbaseConfiguration(context.getConfiguration())); }
From source file:com.datasalt.pangool.solr.SolrRecordWriter.java
License:Apache License
public SolrRecordWriter(int batchSize, boolean outputZipFile, int threadCount, int queueSize, String localSolrHome, String zipName, TupleDocumentConverter converter, TaskAttemptContext context) { this.localSolrHome = localSolrHome; this.zipName = zipName; conf = context.getConfiguration(); this.batchSize = batchSize; setLogLevel("org.apache.solr.core", "WARN"); setLogLevel("org.apache.solr.update", "WARN"); Logger.getLogger("org.apache.solr.core").setLevel(Level.WARN); Logger.getLogger("org.apache.solr.update").setLevel(Level.WARN); java.util.logging.Logger.getLogger("org.apache.solr.core").setLevel(java.util.logging.Level.WARNING); java.util.logging.Logger.getLogger("org.apache.solr.update").setLevel(java.util.logging.Level.WARNING); setLogLevel("org.apache.solr", "WARN"); Logger.getLogger("org.apache.solr").setLevel(Level.WARN); java.util.logging.Logger.getLogger("org.apache.solr").setLevel(java.util.logging.Level.WARNING); heartBeater = new HeartBeater(context); try {// www . java2 s .c o m heartBeater.needHeartBeat(); /** The actual file in hdfs that holds the configuration. */ this.outputZipFile = outputZipFile; this.fs = FileSystem.get(conf); perm = new Path(FileOutputFormat.getOutputPath(context), getOutFileName(context, "part")); // Make a task unique name that contains the actual index output name to // make debugging simpler // Note: if using JVM reuse, the sequence number will not be reset for a // new task using the jvm Path temp = conf.getLocalPath("mapred.local.dir", "solr_" + conf.get("mapred.task.id") + '.' + sequence.incrementAndGet()); if (outputZipFile && !perm.getName().endsWith(".zip")) { perm = perm.suffix(".zip"); } fs.delete(temp, true); // delete old, if any fs.delete(perm, true); // delete old, if any local = fs.startLocalOutput(perm, temp); solrHome = findSolrConfig(conf); // } // Verify that the solr home has a conf and lib directory if (solrHome == null) { throw new IOException("Unable to find solr home setting"); } // Setup a solr instance that we can batch writes to LOG.info("SolrHome: " + solrHome.toUri()); String dataDir = new File(local.toString(), "data").getAbsoluteFile().toString(); // copy the schema to the conf dir File confDir = new File(local.toString(), "conf"); confDir.mkdirs(); File unpackedSolrHome = new File(solrHome.toString()); FileUtils.copyDirectory(new File(unpackedSolrHome, "conf"), confDir); Properties props = new Properties(); props.setProperty("solr.data.dir", dataDir); props.setProperty("solr.home", solrHome.toString()); SolrResourceLoader loader = new SolrResourceLoader(solrHome.toString(), null, props); LOG.info(String.format( "Constructed instance information solr.home %s (%s), instance dir %s, conf dir %s, writing index to temporary directory %s, with permdir %s", solrHome, solrHome.toUri(), loader.getInstanceDir(), loader.getConfigDir(), dataDir, perm)); CoreContainer container = new CoreContainer(loader); CoreDescriptor descr = new CoreDescriptor(container, "core1", solrHome.toString()); descr.setDataDir(dataDir); descr.setCoreProperties(props); core = container.create(descr); container.register(core, false); solr = new EmbeddedSolrServer(container, "core1"); batchWriter = new BatchWriter(solr, batchSize, context.getTaskAttemptID().getTaskID(), threadCount, queueSize); this.converter = converter; } catch (Exception e) { e.printStackTrace(); LOG.error(e); throw new IllegalStateException(String.format("Failed to initialize record writer for %s, %s", context.getJobName(), conf.get("mapred.task.id")), e); } finally { heartBeater.cancelHeartBeat(); } }
From source file:com.datasalt.pangool.tuplemr.avro.AvroInputFormat.java
License:Apache License
@Override public RecordReader<AvroWrapper<T>, NullWritable> createRecordReader(InputSplit inputSplit, TaskAttemptContext context) throws IOException, InterruptedException { context.setStatus(inputSplit.toString()); return new AvroRecordReader<T>(getSchema(), isReflect, context.getConfiguration(), (FileSplit) inputSplit); }