List of usage examples for org.apache.hadoop.conf Configuration getStrings
public String[] getStrings(String name)
name
property as an array of String
s. From source file:FormatStorage1.IHead.java
License:Open Source License
public void fromJobConf(Configuration conf) { this.magic = conf.getInt(ConstVar.HD_magic, ConstVar.NewFormatMagic); this.var = (byte) conf.getInt(ConstVar.HD_var, 0); this.ver = (byte) conf.getInt(ConstVar.HD_ver, 0); this.lineindex = (byte) conf.getInt(ConstVar.HD_lineindex, 1); this.primaryIndex = (short) conf.getInt(ConstVar.HD_primaryIndex, -1); this.compress = (byte) conf.getInt(ConstVar.HD_compress, 0); this.compressStyle = (byte) conf.getInt(ConstVar.HD_compressStyle, 0); this.encode = (byte) conf.getInt(ConstVar.HD_encode, 0); this.encodeStyle = (byte) conf.getInt(ConstVar.HD_encodeStyle, 0); this.fieldMap = new IFieldMap(); String[] fieldStrings = conf.getStrings(ConstVar.HD_fieldMap); if (fieldStrings != null) for (int i = 0; i < fieldStrings.length; i++) { String[] def = fieldStrings[i].split(ConstVar.RecordSplit); byte type = Byte.valueOf(def[0]); int index = Integer.valueOf(def[2]); fieldMap.addFieldType(new IRecord.IFType(type, index)); }/*from w w w. j a va 2 s . c o m*/ this.udi = new IUserDefinedHeadInfo(); String[] udistrs = conf.getStrings(ConstVar.HD_udi); if (udistrs != null) for (int i = 0; i < udistrs.length; i++) { String[] def = udistrs[i].split(ConstVar.RecordSplit); udi.addInfo(Integer.valueOf(def[0]), def[1]); } }
From source file:full_MapReduce.AttributeInfoMapper.java
License:Open Source License
public void map(TextArrayWritable key, IntWritable value, Context context) throws IOException, InterruptedException { Configuration conf = context.getConfiguration(); Writable[] key_values = key.get();/* w w w.j a v a 2 s .c om*/ Text[] attributes_and_class = new Text[key_values.length]; for (Integer i = 0; i < key_values.length; ++i) { String[] value_needed = conf.getStrings(i.toString()); if (value_needed == null) { attributes_and_class[i] = (Text) key_values[i]; } else if (!value_needed[0].equals(((Text) key_values[i]).toString())) { return; } } Text classification = attributes_and_class[attributes_and_class.length - 1]; for (Integer i = 0; i < attributes_and_class.length - 1; ++i) { if (attributes_and_class[i] != null) { context.write(new Text(i.toString()), new AttributeCounterWritable(attributes_and_class[i], classification, value)); } } }
From source file:IndexService.IColumnRecordReader.java
License:Open Source License
IColumnRecordReader(Configuration conf, IColumnInputSplit split) throws IOException { this.conf = conf; String[] indexfilemap = conf.getStrings(ConstVar.HD_index_filemap); String splitfilename = split.getPath().toString(); for (int i = 0; i < indexfilemap.length; i++) { if (splitfilename.endsWith(indexfilemap[i])) { this.fileindex = i; break; }/* ww w. jav a 2 s .c om*/ } indexkeys = new ArrayList<Integer>(); String idss = conf.get("index.ids"); String[] strs = idss.split(","); for (int i = 0; i < strs.length; i++) { indexkeys.add(Integer.parseInt(strs[i])); } icdf = new IColumnDataFile(conf); icdf.open(split.getPath().toString()); if (split.wholefileASasplit) { this.beginLine = 0; this.endLine = icdf.recnum() - 1; } else { beginLine = split.beginline; endLine = split.beginline + split.recnum - 1; } currentline = beginLine; icdf.seek(currentline); record = icdf.getIRecordObj(); }
From source file:IndexService.IFormatRecordReader.java
License:Open Source License
public IFormatRecordReader(Configuration conf, IFormatInputSplit split) throws IOException { this.conf = conf; String[] indexfilemap = conf.getStrings(ConstVar.HD_index_filemap); String splitfilename = split.getPath().toString(); for (int i = 0; i < indexfilemap.length; i++) { if (splitfilename.endsWith(indexfilemap[i])) { this.fileindex = i; break; }/*w ww . j a v a2 s . c o m*/ } indexkeys = new ArrayList<Integer>(); String idss = conf.get("index.ids"); String[] strs = idss.split(","); for (int i = 0; i < strs.length; i++) { indexkeys.add(Integer.parseInt(strs[i])); } ifdf = new IFormatDataFile(conf); ifdf.open(split.getPath().toString()); if (split.wholefileASasplit) { this.beginLine = 0; this.endLine = ifdf.segIndex().recnum() - 1; } else { beginLine = split.beginline; endLine = split.beginline + split.recnum - 1; } currentline = beginLine; ifdf.seek(currentline); record = ifdf.getIRecordObj(); }
From source file:IndexService.IndexMergeIFormatRecordReader.java
License:Open Source License
public IndexMergeIFormatRecordReader(Configuration conf, IndexMergeIFormatSplit split) throws IOException { this.conf = conf; this.filesmap = new HashMap<String, Integer>(); String[] strs = conf.getStrings(ConstVar.HD_index_filemap); for (int i = 0; i < strs.length; i++) { filesmap.put(strs[i], i);//from ww w . j a v a 2 s . c o m } ifdf = new IFormatDataFile(conf); ifdf.open(split.getPath().toString()); this.splitinfos = ifdf.fileInfo().head().getUdi().infos(); if (split.wholefileASasplit) { this.recnum = ifdf.segIndex().recnum(); } else { this.recnum = split.recnum; } ifdf.seek(split.beginline); record = ifdf.getIRecordObj(); }
From source file:io.apigee.lembos.mapreduce.LembosMapReduceEnvironment.java
License:Apache License
/** * Retrieves an initialized MapReduce environment. * * @param conf the Hadoop configuration//from w ww. j a v a2 s.c o m * * @return the initialized MapReduce environment * * @throws ExecutionException if something goes wrong executing the Node.js module * @throws IOException if something goes wrong setting up the environment * @throws InterruptedException if the execution of the Node.js module gets interrupted * @throws NodeException if there is an issue with the Node.js module */ public static LembosMapReduceEnvironment fromConf(final Configuration conf) throws ExecutionException, InterruptedException, IOException, NodeException { File modulePath = new File(LembosConstants.MR_DISTRIBUTED_CACHE_SYMLINK); if (modulePath.exists()) { // Locate module in DistributedCache final File[] children = modulePath.listFiles(); if (children == null) { throw new RuntimeException("Unable to find Node.js module in DistributedCache"); } for (final File child : children) { final String moduleName = conf.get(LembosConstants.MR_MODULE_NAME); final String[] possibleNames = new String[] { moduleName, moduleName + ".js", moduleName + ".json", moduleName + ".node" }; if (Arrays.asList(possibleNames).contains(child.getName())) { modulePath = child; break; } } } else if (!modulePath.exists() && conf.get(LembosConstants.MR_MODULE_PATH) != null) { // Default to module path for testing and for LocalJobRunner (DistributedCache does not work in local mode) modulePath = new File(conf.get(LembosConstants.MR_MODULE_PATH)); } final LembosMapReduceEnvironment mrEnv = new LembosMapReduceEnvironment( conf.get(LembosConstants.MR_MODULE_NAME), modulePath, conf.getStrings(LembosConstants.MR_MODULE_ARGS)); mrEnv.initialize(); return mrEnv; }
From source file:mvm.rya.accumulo.pig.IndexWritingTool.java
License:Apache License
@Override public int run(final String[] args) throws Exception { Preconditions.checkArgument(args.length == 7, "java " + IndexWritingTool.class.getCanonicalName() + " hdfsSaveLocation sparqlFile cbinstance cbzk cbuser cbpassword rdfTablePrefix."); final String inputDir = args[0]; final String sparqlFile = args[1]; final String instStr = args[2]; final String zooStr = args[3]; final String userStr = args[4]; final String passStr = args[5]; final String tablePrefix = args[6]; String sparql = FileUtils.readFileToString(new File(sparqlFile)); Job job = new Job(getConf(), "Write HDFS Index to Accumulo"); job.setJarByClass(this.getClass()); Configuration jobConf = job.getConfiguration(); jobConf.setBoolean("mapred.map.tasks.speculative.execution", false); setVarOrders(sparql, jobConf);//from w w w . ja v a2 s . c o m TextInputFormat.setInputPaths(job, inputDir); job.setInputFormatClass(TextInputFormat.class); job.setMapperClass(MyMapper.class); job.setMapOutputKeyClass(Text.class); job.setMapOutputValueClass(Mutation.class); job.setOutputKeyClass(Text.class); job.setOutputValueClass(Mutation.class); job.setNumReduceTasks(0); String tableName; if (zooStr.equals("mock")) { tableName = tablePrefix; } else { tableName = tablePrefix + "INDEX_" + UUID.randomUUID().toString().replace("-", "").toUpperCase(); } setAccumuloOutput(instStr, zooStr, userStr, passStr, job, tableName); jobConf.set(sparql_key, sparql); int complete = job.waitForCompletion(true) ? 0 : -1; if (complete == 0) { String[] varOrders = jobConf.getStrings("varOrders"); String orders = Joiner.on("\u0000").join(varOrders); Instance inst; if (zooStr.equals("mock")) { inst = new MockInstance(instStr); } else { inst = new ZooKeeperInstance(instStr, zooStr); } Connector conn = inst.getConnector(userStr, passStr.getBytes()); BatchWriter bw = conn.createBatchWriter(tableName, 10, 5000, 1); Counters counters = job.getCounters(); Counter c1 = counters.findCounter(cardCounter, cardCounter); Mutation m = new Mutation("~SPARQL"); Value v = new Value(sparql.getBytes()); m.put(new Text("" + c1.getValue()), new Text(orders), v); bw.addMutation(m); bw.close(); return complete; } else { return complete; } }
From source file:org.apache.accumulo.examples.wikisearch.ingest.WikipediaConfiguration.java
License:Apache License
/** * Helper method to get properties from Hadoop configuration * * @throws IllegalArgumentException/*from ww w . j av a 2s. c om*/ * if property is not defined, null, or empty. Or if resultClass is not handled. * @return value of property */ @SuppressWarnings("unchecked") public static <T> T isNull(Configuration conf, String propertyName, Class<T> resultClass) { String p = conf.get(propertyName); if (StringUtils.isEmpty(p)) { throw new IllegalArgumentException(propertyName + " must be specified"); } if (resultClass.equals(String.class)) { return (T) p; } else if (resultClass.equals(String[].class)) { return (T) conf.getStrings(propertyName); } else if (resultClass.equals(Boolean.class)) { return (T) Boolean.valueOf(p); } else if (resultClass.equals(Long.class)) { return (T) Long.valueOf(p); } else if (resultClass.equals(Integer.class)) { return (T) Integer.valueOf(p); } else if (resultClass.equals(Float.class)) { return (T) Float.valueOf(p); } else if (resultClass.equals(Double.class)) { return (T) Double.valueOf(p); } else { throw new IllegalArgumentException(resultClass.getSimpleName() + " is unhandled."); } }
From source file:org.apache.blur.mapreduce.lib.CsvBlurMapper.java
License:Apache License
public static Map<String, List<String>> getFamilyAndColumnNameMap(Configuration configuration) { Map<String, List<String>> columnNameMap = new HashMap<String, List<String>>(); for (String family : getFamilyNames(configuration)) { String[] columnsNames = configuration.getStrings(BLUR_CSV_FAMILY_COLUMN_PREFIX + family); columnNameMap.put(family, Arrays.asList(columnsNames)); }//w w w.j a v a 2 s .co m return columnNameMap; }
From source file:org.apache.gora.mapreduce.GoraMapReduceUtils.java
License:Apache License
/** * Add our own serializer (obtained via the {@link PersistentSerialization} * wrapper) to any other <code>io.serializations</code> which may be specified * within existing Hadoop configuration. * //from w ww . j ava 2 s . c o m * @param conf the Hadoop configuration object * @param reuseObjects boolean parameter to reuse objects */ public static void setIOSerializations(Configuration conf, boolean reuseObjects) { String serializationClass = PersistentSerialization.class.getCanonicalName(); String[] serializations = StringUtils.joinStringArrays(conf.getStrings("io.serializations"), "org.apache.hadoop.io.serializer.WritableSerialization", StringSerialization.class.getCanonicalName(), serializationClass); conf.setStrings("io.serializations", serializations); }