Examples of mapredWork

Examples of org.apache.hadoop.hive.ql.plan.MapredWork

229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
    } else if (mvWork.getLoadFileWork() != null) {
      statsWork = new StatsWork(mvWork.getLoadFileWork());
    }
    assert statsWork != null : "Error when genereting StatsTask";
    statsWork.setStatsReliable(hconf.getBoolVar(ConfVars.HIVE_STATS_RELIABLE));
    MapredWork mrWork = (MapredWork) currTask.getWork();

    // AggKey in StatsWork is used for stats aggregation while StatsAggPrefix
    // in FileSinkDesc is used for stats publishing. They should be consistent.
    statsWork.setAggKey(((FileSinkOperator) nd).getConf().getStatsAggPrefix());
    Task<? extends Serializable> statsTask = TaskFactory.get(statsWork, hconf);

    // mark the MapredWork and FileSinkOperator for gathering stats
    nd.getConf().setGatherStats(true);
    mrWork.getMapWork().setGatheringStats(true);
    if (mrWork.getReduceWork() != null) {
      mrWork.getReduceWork().setGatheringStats(true);
    }
    nd.getConf().setStatsReliable(hconf.getBoolVar(ConfVars.HIVE_STATS_RELIABLE));
    nd.getConf().setMaxStatsKeyPrefixLength(
        hconf.getIntVar(ConfVars.HIVE_STATS_KEY_PREFIX_MAX_LENGTH));
    // mrWork.addDestinationTable(nd.getConf().getTableInfo().getTableName());
View Full Code Here

Examples of org.apache.hadoop.hive.ql.plan.MapredWork

369
370
371
372
373
374
375
376
377
378
379
        throw new SemanticException(msg);
      }

    } else {
      cplan = createMRWorkForMergingFiles(conf, tsMerge, fsInputDesc);
      work = new MapredWork();
      ((MapredWork)work).setMapWork(cplan);
      // use CombineHiveInputFormat for map-only merging
    }
    cplan.setInputformat("org.apache.hadoop.hive.ql.io.CombineHiveInputFormat");
    // NOTE: we should gather stats in MR1 rather than MR2 at merge job since we don't
View Full Code Here

Examples of org.apache.hadoop.hive.ql.plan.MapredWork

486
487
488
489
490
491
492
493
494
495
496
497
    String inputDir = fsDesc.getFinalDirName();
    TableDesc tblDesc = fsDesc.getTableInfo();
    aliases.add(inputDir); // dummy alias: just use the input path

    // constructing the default MapredWork
    MapredWork cMrPlan = GenMapRedUtils.getMapRedWorkFromConf(conf);
    MapWork cplan = cMrPlan.getMapWork();
    cplan.getPathToAliases().put(inputDir, aliases);
    cplan.getPathToPartitionInfo().put(inputDir, new PartitionDesc(tblDesc, null));
    cplan.getAliasToWork().put(inputDir, topOp);
    cplan.setMapperCannotSpanPartns(true);
View Full Code Here

Examples of org.apache.hadoop.hive.ql.plan.MapredWork

102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
    Operator<? extends OperatorDesc> reducer = op.getChildOperators().get(0);
    Map<Operator<? extends OperatorDesc>, GenMapRedCtx> mapCurrCtx =
        opProcCtx.getMapCurrCtx();
    GenMapRedCtx mapredCtx = mapCurrCtx.get(op.getParentOperators().get(0));
    Task<? extends Serializable> currTask = mapredCtx.getCurrTask();
    MapredWork plan = (MapredWork) currTask.getWork();
    HashMap<Operator<? extends OperatorDesc>, Task<? extends Serializable>> opTaskMap =
        opProcCtx.getOpTaskMap();
    Operator<? extends OperatorDesc> currTopOp = opProcCtx.getCurrTopOp();

    opTaskMap.put(reducer, currTask);
    plan.setReduceWork(new ReduceWork());
    plan.getReduceWork().setReducer(reducer);
    ReduceSinkDesc desc = op.getConf();

    plan.getReduceWork().setNumReduceTasks(desc.getNumReducers());

    if (needsTagging(plan.getReduceWork())) {
      plan.getReduceWork().setNeedsTagging(true);
    }

    assert currTopOp != null;
    String currAliasId = opProcCtx.getCurrAliasId();
View Full Code Here

Examples of org.apache.hadoop.hive.ql.plan.mapredWork

52
53
54
55
56
57
58
59
60
61
62
        auxJars = " -libjars " + auxJars + " ";
      } else {
        auxJars = " ";
      }

      mapredWork plan = getWork();

      File planFile = File.createTempFile("plan", ".xml");
      LOG.info("Generating plan file " + planFile.toString());
      FileOutputStream out = new FileOutputStream(planFile);
      Utilities.serializeMapRedWork(plan, out);
View Full Code Here

Examples of org.apache.hadoop.hive.ql.plan.mapredWork

121
122
123
124
125
126
127
128
    return true;
  }

  @Override
  public boolean hasReduce() {
    mapredWork w = getWork();
    return w.getReducer() != null;
  }
View Full Code Here

Examples of org.apache.hadoop.hive.ql.plan.mapredWork

64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
    fieldNames = fieldNameArray.toArray(new String [0]);
  }

  public void configure(JobConf job) {
    jc = job;
    mapredWork gWork = Utilities.getMapRedWork(job);
    reducer = gWork.getReducer();
    reducer.setMapredWork(gWork);
    isTagged = gWork.getNeedsTagging();
    try {
      tableDesc keyTableDesc = gWork.getKeyDesc();
      inputKeyDeserializer = (SerDe)ReflectionUtils.newInstance(keyTableDesc.getDeserializerClass(), null);
      inputKeyDeserializer.initialize(null, keyTableDesc.getProperties());
      keyObjectInspector = inputKeyDeserializer.getObjectInspector();
      for(int tag=0; tag<gWork.getTagToValueDesc().size(); tag++) {
        // We should initialize the SerDe with the TypeInfo when available.
        tableDesc valueTableDesc = gWork.getTagToValueDesc().get(tag);
        inputValueDeserializer[tag] = (SerDe)ReflectionUtils.newInstance(valueTableDesc.getDeserializerClass(), null);
        inputValueDeserializer[tag].initialize(null, valueTableDesc.getProperties());
        valueObjectInspector[tag] = inputValueDeserializer[tag].getObjectInspector();
       
        ArrayList<ObjectInspector> ois = new ArrayList<ObjectInspector>();
View Full Code Here

Examples of org.apache.hadoop.hive.ql.plan.mapredWork

39
40
41
42
43
44
45
46
47
48
  public static final Log l4j = LogFactory.getLog("ExecMapper");
  private static boolean done;
 
  public void configure(JobConf job) {
    jc = job;
    mapredWork mrwork = Utilities.getMapRedWork(job);
    mo = new MapOperator ();
    mo.setConf(mrwork);
    // we don't initialize the operator until we have set the output collector
  }
View Full Code Here

Examples of org.apache.hadoop.hive.ql.plan.mapredWork

556
557
558
559
560
561
562
563
564
565
566
          throw new HiveException(e.getMessage(), e);
        }
      }
    }

    mapredWork plan = Utilities.deserializeMapRedWork(pathData);
    ExecDriver ed = new ExecDriver(plan, conf, isSilent);
    int ret = ed.execute();
    if (ret != 0) {
      System.out.println("Job Failed");
      System.exit(2);
View Full Code Here

Examples of org.apache.hadoop.hive.ql.plan.mapredWork

626
627
628
629
630
631
632
633
    return true;
  }

  @Override
  public boolean hasReduce() {
    mapredWork w = getWork();
    return w.getReducer() != null;
  }
View Full Code Here
TOP
Copyright © 2018 www.massapi.com. All rights reserved.
All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.