List of usage examples for org.apache.hadoop.mapred TaskID forName
public static TaskID forName(String str) throws IllegalArgumentException
From source file:de.l3s.streamcorpus.mapreduce.TerrierIndexing.java
License:Mozilla Public License
/** Performs cleanup of an index path removing temporary files */ public static void deleteTaskFiles(String path, JobID job) { String[] fileNames = Files.list(path); if (fileNames == null) return;/*from w ww. java 2 s. c om*/ for (String filename : fileNames) { String periodParts[] = filename.split("\\."); try { TaskID tid = TaskID.forName(periodParts[0]); if (tid.getJobID().equals(job)) { if (!Files.delete(path + "/" + filename)) logger.warn("Could not delete temporary map side-effect file " + path + "/" + filename); } } catch (Exception e) { } } }
From source file:hivemall.utils.hadoop.HadoopUtils.java
License:Open Source License
@Nonnull public static String getJobIdFromTaskId(@Nonnull String taskidStr) { if (!taskidStr.startsWith("task_")) {// workaround for Tez taskidStr = taskidStr.replace("task", "task_"); taskidStr = taskidStr.substring(0, taskidStr.lastIndexOf('_')); }/*from www. j a v a2 s . c o m*/ TaskID taskId = TaskID.forName(taskidStr); JobID jobId = taskId.getJobID(); return jobId.toString(); }
From source file:org.elasticsearch.hadoop.mr.HadoopCfgUtils.java
License:Apache License
public static TaskID getTaskID(Configuration cfg) { // first try with the attempt since some Hadoop versions mix the two String taskAttemptId = HadoopCfgUtils.getTaskAttemptId(cfg); if (StringUtils.hasText(taskAttemptId)) { try {/*from ww w .j a va 2s . com*/ return TaskAttemptID.forName(taskAttemptId).getTaskID(); } catch (IllegalArgumentException ex) { // the task attempt is invalid (Tez in particular uses the wrong string - see #346) // try to fallback to task id return parseTaskIdFromTaskAttemptId(taskAttemptId); } } String taskIdProp = HadoopCfgUtils.getTaskId(cfg); // double-check task id bug in Hadoop 2.5.x if (StringUtils.hasText(taskIdProp) && !taskIdProp.contains("attempt")) { return TaskID.forName(taskIdProp); } return null; }
From source file:org.terrier.structures.indexing.singlepass.hadoop.MapData.java
License:Mozilla Public License
/** * Constructor - Loads the Map Information from the DataInputStream Provided * @param in - Stream of the Map data file *//* w ww. ja va 2 s . c om*/ public MapData(DataInputStream in) throws IOException { super(); mapTaskID = in.readUTF(); int_mapTaskId = TaskID.forName(mapTaskID).getId(); int flushSize; while ((flushSize = in.readInt()) != -1) { flushDocSizes.add(flushSize); } numMapDocs = in.readInt(); splitnum = in.readInt(); logger.info("map " + mapTaskID + " processed split " + splitnum + " which had " + numMapDocs + " docs, with " + flushDocSizes.size() + " flushes\n"); }