List of usage examples for org.apache.hadoop.conf Configuration clear
public void clear()
From source file:at.ac.tuwien.infosys.jcloudscale.datastore.driver.hbase.HbaseConfig.java
License:Apache License
/** * Create the HBase Configuration for a given datastore * * @param datastore the given datastore/* ww w . j av a2 s . co m*/ * @return the HBase Configuration */ public static Configuration getConfig(Datastore datastore) { Configuration configuration = HBaseConfiguration.create(); configuration.clear(); configuration.set("hbase.zookeeper.quorum", datastore.getHost()); return configuration; }
From source file:co.cask.cdap.app.runtime.spark.SparkRuntimeContextProvider.java
License:Apache License
private static Configuration createHConf() throws MalformedURLException { Configuration hConf = new Configuration(); hConf.clear(); hConf.addResource(new File(HCONF_FILE_NAME).toURI().toURL()); return hConf; }
From source file:co.cask.cdap.data.stream.StreamInputFormatProvider.java
License:Apache License
/** * Sets the {@link StreamEventDecoder} to be used by the InputFormat for the given type. If the * {@link StreamBatchReadable} already defined a {@link StreamEventDecoder} or {@link FormatSpecification}, * this method is a no-op.//from w w w.j av a 2 s . c om * * @param configuration configuration to update * @param type type for {@link StreamEventData} to decode to * @return the same configuration map as in the argument. */ public Map<String, String> setDecoderType(Map<String, String> configuration, Type type) { if (streamBatchReadable.getFormatSpecification() == null && streamBatchReadable.getDecoderType() == null) { Configuration hConf = new Configuration(); hConf.clear(); StreamInputFormat.inferDecoderClass(hConf, type); configuration.putAll(ConfigurationUtil.toMap(hConf)); } return configuration; }
From source file:co.cask.cdap.data.stream.StreamInputFormatProvider.java
License:Apache License
@Override public Map<String, String> getInputFormatConfiguration() { Id.Stream streamId = Id.Stream.from(namespaceId, streamBatchReadable.getStreamName()); try {// ww w . j a va 2s . c o m StreamConfig streamConfig = streamAdmin.getConfig(streamId); Location streamPath = StreamUtils.createGenerationLocation(streamConfig.getLocation(), StreamUtils.getGeneration(streamConfig)); Configuration hConf = new Configuration(); hConf.clear(); StreamInputFormat.setTTL(hConf, streamConfig.getTTL()); StreamInputFormat.setStreamPath(hConf, streamPath.toURI()); StreamInputFormat.setTimeRange(hConf, streamBatchReadable.getStartTime(), streamBatchReadable.getEndTime()); FormatSpecification formatSpec = streamBatchReadable.getFormatSpecification(); if (formatSpec != null) { StreamInputFormat.setBodyFormatSpecification(hConf, formatSpec); } else { String decoderType = streamBatchReadable.getDecoderType(); if (decoderType != null) { StreamInputFormat.setDecoderClassName(hConf, decoderType); } } return ConfigurationUtil.toMap(hConf); } catch (IOException e) { throw Throwables.propagate(e); } }
From source file:co.cask.cdap.etl.batch.spark.SparkBatchSinkFactory.java
License:Apache License
<K, V> void writeFromRDD(JavaPairRDD<K, V> rdd, JavaSparkExecutionContext sec, String sinkName, Class<K> keyClass, Class<V> valueClass) { Set<String> outputNames = sinkOutputs.get(sinkName); if (outputNames == null || outputNames.size() == 0) { // should never happen if validation happened correctly at pipeline configure time throw new IllegalArgumentException( sinkName + " has no outputs. " + "Please check that the sink calls addOutput at some point."); }/*from w w w .j a v a 2 s .c o m*/ for (String outputName : outputNames) { OutputFormatProvider outputFormatProvider = outputFormatProviders.get(outputName); if (outputFormatProvider != null) { Configuration hConf = new Configuration(); hConf.clear(); for (Map.Entry<String, String> entry : outputFormatProvider.getOutputFormatConfiguration() .entrySet()) { hConf.set(entry.getKey(), entry.getValue()); } hConf.set(MRJobConfig.OUTPUT_FORMAT_CLASS_ATTR, outputFormatProvider.getOutputFormatClassName()); rdd.saveAsNewAPIHadoopDataset(hConf); } DatasetInfo datasetInfo = datasetInfos.get(outputName); if (datasetInfo != null) { sec.saveAsDataset(rdd, datasetInfo.getDatasetName(), datasetInfo.getDatasetArgs()); } } }
From source file:co.cask.cdap.etl.batch.spark.SparkBatchSourceFactory.java
License:Apache License
@SuppressWarnings("unchecked") public <K, V> JavaPairRDD<K, V> createRDD(JavaSparkExecutionContext sec, JavaSparkContext jsc, Class<K> keyClass, Class<V> valueClass) { if (streamBatchReadable != null) { FormatSpecification formatSpec = streamBatchReadable.getFormatSpecification(); if (formatSpec != null) { return (JavaPairRDD<K, V>) sec.fromStream(streamBatchReadable.getStreamName(), formatSpec, streamBatchReadable.getStartTime(), streamBatchReadable.getEndTime(), StructuredRecord.class); }//from w w w .j a va 2 s .c o m String decoderType = streamBatchReadable.getDecoderType(); if (decoderType == null) { return (JavaPairRDD<K, V>) sec.fromStream(streamBatchReadable.getStreamName(), streamBatchReadable.getStartTime(), streamBatchReadable.getEndTime(), valueClass); } else { try { Class<StreamEventDecoder<K, V>> decoderClass = (Class<StreamEventDecoder<K, V>>) Thread .currentThread().getContextClassLoader().loadClass(decoderType); return sec.fromStream(streamBatchReadable.getStreamName(), streamBatchReadable.getStartTime(), streamBatchReadable.getEndTime(), decoderClass, keyClass, valueClass); } catch (Exception e) { throw Throwables.propagate(e); } } } if (inputFormatProvider != null) { Configuration hConf = new Configuration(); hConf.clear(); for (Map.Entry<String, String> entry : inputFormatProvider.getInputFormatConfiguration().entrySet()) { hConf.set(entry.getKey(), entry.getValue()); } ClassLoader classLoader = Objects.firstNonNull(currentThread().getContextClassLoader(), getClass().getClassLoader()); try { @SuppressWarnings("unchecked") Class<InputFormat> inputFormatClass = (Class<InputFormat>) classLoader .loadClass(inputFormatProvider.getInputFormatClassName()); return jsc.newAPIHadoopRDD(hConf, inputFormatClass, keyClass, valueClass); } catch (ClassNotFoundException e) { throw Throwables.propagate(e); } } if (datasetInfo != null) { return sec.fromDataset(datasetInfo.getDatasetName(), datasetInfo.getDatasetArgs()); } // This should never happen since the constructor is private and it only get calls from static create() methods // which make sure one and only one of those source type will be specified. throw new IllegalStateException("Unknown source type"); }
From source file:co.cask.cdap.etl.spark.batch.SparkBatchSinkFactory.java
License:Apache License
public <K, V> void writeFromRDD(JavaPairRDD<K, V> rdd, JavaSparkExecutionContext sec, String sinkName, Class<K> keyClass, Class<V> valueClass) { Set<String> outputNames = sinkOutputs.get(sinkName); if (outputNames == null || outputNames.isEmpty()) { // should never happen if validation happened correctly at pipeline configure time throw new IllegalArgumentException( sinkName + " has no outputs. " + "Please check that the sink calls addOutput at some point."); }/* w w w . j a v a 2 s . c o m*/ for (String outputName : outputNames) { OutputFormatProvider outputFormatProvider = outputFormatProviders.get(outputName); if (outputFormatProvider != null) { Configuration hConf = new Configuration(); hConf.clear(); for (Map.Entry<String, String> entry : outputFormatProvider.getOutputFormatConfiguration() .entrySet()) { hConf.set(entry.getKey(), entry.getValue()); } hConf.set(MRJobConfig.OUTPUT_FORMAT_CLASS_ATTR, outputFormatProvider.getOutputFormatClassName()); rdd.saveAsNewAPIHadoopDataset(hConf); } DatasetInfo datasetInfo = datasetInfos.get(outputName); if (datasetInfo != null) { sec.saveAsDataset(rdd, datasetInfo.getDatasetName(), datasetInfo.getDatasetArgs()); } } }
From source file:co.cask.cdap.etl.spark.batch.SparkBatchSourceFactory.java
License:Apache License
@SuppressWarnings("unchecked") private <K, V> JavaPairRDD<K, V> createInputRDD(JavaSparkExecutionContext sec, JavaSparkContext jsc, String inputName, Class<K> keyClass, Class<V> valueClass) { if (streams.containsKey(inputName)) { Input.StreamInput streamInput = streams.get(inputName); FormatSpecification formatSpec = streamInput.getBodyFormatSpec(); if (formatSpec != null) { return (JavaPairRDD<K, V>) sec.fromStream(streamInput.getName(), formatSpec, streamInput.getStartTime(), streamInput.getEndTime(), StructuredRecord.class); }//www . ja v a2 s . c o m String decoderType = streamInput.getDecoderType(); if (decoderType == null) { return (JavaPairRDD<K, V>) sec.fromStream(streamInput.getName(), streamInput.getStartTime(), streamInput.getEndTime(), valueClass); } else { try { Class<StreamEventDecoder<K, V>> decoderClass = (Class<StreamEventDecoder<K, V>>) Thread .currentThread().getContextClassLoader().loadClass(decoderType); return sec.fromStream(streamInput.getName(), streamInput.getStartTime(), streamInput.getEndTime(), decoderClass, keyClass, valueClass); } catch (Exception e) { throw Throwables.propagate(e); } } } if (inputFormatProviders.containsKey(inputName)) { InputFormatProvider inputFormatProvider = inputFormatProviders.get(inputName); Configuration hConf = new Configuration(); hConf.clear(); for (Map.Entry<String, String> entry : inputFormatProvider.getInputFormatConfiguration().entrySet()) { hConf.set(entry.getKey(), entry.getValue()); } ClassLoader classLoader = Objects.firstNonNull(currentThread().getContextClassLoader(), getClass().getClassLoader()); try { @SuppressWarnings("unchecked") Class<InputFormat> inputFormatClass = (Class<InputFormat>) classLoader .loadClass(inputFormatProvider.getInputFormatClassName()); return jsc.newAPIHadoopRDD(hConf, inputFormatClass, keyClass, valueClass); } catch (ClassNotFoundException e) { throw Throwables.propagate(e); } } if (datasetInfos.containsKey(inputName)) { DatasetInfo datasetInfo = datasetInfos.get(inputName); return sec.fromDataset(datasetInfo.getDatasetName(), datasetInfo.getDatasetArgs()); } // This should never happen since the constructor is private and it only get calls from static create() methods // which make sure one and only one of those source type will be specified. throw new IllegalStateException("Unknown source type"); }
From source file:co.cask.cdap.internal.app.runtime.batch.dataset.DatasetInputFormatProvider.java
License:Apache License
private Map<String, String> createBatchReadableConfiguration() { List<Split> splits = this.splits; if (splits == null) { splits = ((BatchReadable<?, ?>) dataset).getSplits(); }/*from w w w . j a v a 2 s . co m*/ Configuration hConf = new Configuration(); hConf.clear(); try { AbstractBatchReadableInputFormat.setDatasetSplits(hConf, datasetName, datasetArgs, splits); return ConfigurationUtil.toMap(hConf); } catch (IOException e) { throw new IllegalArgumentException(e); } }
From source file:co.cask.cdap.internal.app.runtime.batch.dataset.DatasetOutputFormatProvider.java
License:Apache License
private Map<String, String> createDatasetConfiguration(String datasetName, Map<String, String> datasetArgs) { Configuration hConf = new Configuration(); hConf.clear(); AbstractBatchWritableOutputFormat.setDataset(hConf, datasetName, datasetArgs); return ConfigurationUtil.toMap(hConf); }