Examples of GenMapRedCtx

Examples of org.apache.hadoop.hive.ql.optimizer.GenMRProcContext.GenMapRedCtx

75
76
77
78
79
80
81
82
83
84
85
86
   * @param opProcCtx processing context
   */
  public static void initPlan(ReduceSinkOperator op, GenMRProcContext opProcCtx) throws SemanticException {
    Operator<? extends Serializable> reducer = op.getChildOperators().get(0);
    Map<Operator<? extends Serializable>, 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 Serializable>, Task<? extends Serializable>> opTaskMap = opProcCtx.getOpTaskMap();
    Operator<? extends Serializable> currTopOp = opProcCtx.getCurrTopOp();

    opTaskMap.put(reducer, currTask);
View Full Code Here

Examples of org.apache.hadoop.hive.ql.optimizer.GenMRProcContext.GenMapRedCtx

120
121
122
123
124
125
126
127
128
129
130
131
      boolean setReducer, int pos)
    throws SemanticException {
    Map<Operator<? extends Serializable>, GenMapRedCtx> mapCurrCtx = opProcCtx.getMapCurrCtx();
    assert (((pos == -1) && (readInputMapJoin)) || (pos != -1));
    int parentPos = (pos == -1) ? 0 : pos;
    GenMapRedCtx mapredCtx = mapCurrCtx.get(op.getParentOperators().get(parentPos));
    Task<? extends Serializable> currTask    = mapredCtx.getCurrTask();
    mapredWork plan = (mapredWork) currTask.getWork();
    HashMap<Operator<? extends Serializable>, Task<? extends Serializable>> opTaskMap = opProcCtx.getOpTaskMap();
    Operator<? extends Serializable> currTopOp = opProcCtx.getCurrTopOp();

    // The mapjoin has already been encountered. Some context must be stored about that
View Full Code Here

Examples of org.apache.hadoop.hive.ql.optimizer.GenMRProcContext.GenMapRedCtx

202
203
204
205
206
207
208
209
210
211
212
213
   * @param opProcCtx processing context
   */
  public static void initUnionPlan(ReduceSinkOperator op, GenMRProcContext opProcCtx) throws SemanticException {
    Operator<? extends Serializable> reducer = op.getChildOperators().get(0);
    Map<Operator<? extends Serializable>, 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 Serializable>, Task<? extends Serializable>> opTaskMap = opProcCtx.getOpTaskMap();

    opTaskMap.put(reducer, currTask);
    plan.setReducer(reducer);
View Full Code Here

Examples of org.apache.hadoop.hive.ql.optimizer.GenMRProcContext.GenMapRedCtx

658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
    childOpList.add(op);
    ts_op.setChildOperators(childOpList);
    op.getParentOperators().set(posn, ts_op);
  
    Map<Operator<? extends Serializable>, GenMapRedCtx> mapCurrCtx = opProcCtx.getMapCurrCtx();
    mapCurrCtx.put(ts_op, new GenMapRedCtx(childTask, null, null));
   
    String streamDesc = taskTmpDir;
    mapredWork cplan = (mapredWork) childTask.getWork();

    if (setReducer) {
      Operator<? extends Serializable> reducer = op.getChildOperators().get(0);
   
      if (reducer.getClass() == JoinOperator.class) {
        String origStreamDesc;
        streamDesc = "$INTNAME";
        origStreamDesc = streamDesc;
        int pos = 0;
        while (cplan.getAliasToWork().get(streamDesc) != null)
          streamDesc = origStreamDesc.concat(String.valueOf(++pos));
      }

      // TODO: Allocate work to remove the temporary files and make that
      // dependent on the redTask
      if (reducer.getClass() == JoinOperator.class)
        cplan.setNeedsTagging(true);
    }
       
    // Add the path to alias mapping
    setTaskPlan(taskTmpDir, streamDesc, ts_op, cplan, local, tt_desc);

    // This can be cleaned up as a function table in future
    if (op instanceof MapJoinOperator) {
      MapJoinOperator mjOp = (MapJoinOperator)op;
      opProcCtx.setCurrMapJoinOp(mjOp);
      GenMRMapJoinCtx mjCtx = opProcCtx.getMapJoinCtx(mjOp);
      if (mjCtx == null)
        mjCtx = new GenMRMapJoinCtx(taskTmpDir, tt_desc, ts_op, null);
      else {
        mjCtx.setTaskTmpDir(taskTmpDir);
        mjCtx.setTTDesc(tt_desc);
        mjCtx.setRootMapJoinOp(ts_op);
      }
      opProcCtx.setMapJoinCtx(mjOp, mjCtx);
      opProcCtx.getMapCurrCtx().put(parent, new GenMapRedCtx(childTask, null, null));
    }
   
    currTopOp = null;
    String currAliasId = null;
   
View Full Code Here

Examples of org.apache.hadoop.hive.ql.optimizer.GenMRProcContext.GenMapRedCtx

44
45
46
47
48
49
50
51
52
53
   */
  public Object process(Node nd, Stack<Node> stack, NodeProcessorCtx procCtx, Object... nodeOutputs) throws SemanticException {
    GenMRProcContext ctx = (GenMRProcContext)procCtx;

    Map<Operator<? extends Serializable>, GenMapRedCtx> mapCurrCtx = ctx.getMapCurrCtx();
    GenMapRedCtx mapredCtx = mapCurrCtx.get((Operator<? extends Serializable>)stack.get(stack.size()-2));
    mapCurrCtx.put((Operator<? extends Serializable>)nd,
        new GenMapRedCtx(mapredCtx.getCurrTask(), mapredCtx.getCurrTopOp(), mapredCtx.getCurrAliasId()));
    return null;
  }
View Full Code Here

Examples of org.apache.hadoop.hive.ql.optimizer.GenMRProcContext.GenMapRedCtx

49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
  public Object process(Node nd, Stack<Node> stack, NodeProcessorCtx opProcCtx, Object... nodeOutputs) throws SemanticException {
    ReduceSinkOperator op = (ReduceSinkOperator)nd;
    GenMRProcContext ctx = (GenMRProcContext)opProcCtx;

    Map<Operator<? extends Serializable>, GenMapRedCtx> mapCurrCtx = ctx.getMapCurrCtx();
    GenMapRedCtx mapredCtx = mapCurrCtx.get((Operator<? extends Serializable>)stack.get(stack.size()-2));
    Task<? extends Serializable> currTask    = mapredCtx.getCurrTask();
    mapredWork currPlan = (mapredWork) currTask.getWork();
    Operator<? extends Serializable> currTopOp   = mapredCtx.getCurrTopOp();
    String currAliasId = mapredCtx.getCurrAliasId();
    Operator<? extends Serializable> reducer = op.getChildOperators().get(0);
    HashMap<Operator<? extends Serializable>, Task<? extends Serializable>> opTaskMap = ctx.getOpTaskMap();
    Task<? extends Serializable> opMapTask = opTaskMap.get(reducer);

    ctx.setCurrTopOp(currTopOp);
    ctx.setCurrAliasId(currAliasId);
    ctx.setCurrTask(currTask);

    // If the plan for this reducer does not exist, initialize the plan
    if (opMapTask == null) {
      if (currPlan.getReducer() == null)
        GenMapRedUtils.initPlan(op, ctx);
      else
        GenMapRedUtils.splitPlan(op, ctx);
    }
    // This will happen in case of joins. The current plan can be thrown away after being merged with the
    // original plan
    else {
      GenMapRedUtils.joinPlan(op, null, opMapTask, ctx, -1, false, false, false);
      currTask = opMapTask;
      ctx.setCurrTask(currTask);
    }

    mapCurrCtx.put(op, new GenMapRedCtx(ctx.getCurrTask(), ctx.getCurrTopOp(), ctx.getCurrAliasId()));
    return null;
  }
View Full Code Here

Examples of org.apache.hadoop.hive.ql.optimizer.GenMRProcContext.GenMapRedCtx

78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
      // find the branch on which this processor was invoked
      int pos = getPositionParent(mapJoin, stack);

      Map<Operator<? extends Serializable>, GenMapRedCtx> mapCurrCtx = ctx.getMapCurrCtx();
      GenMapRedCtx mapredCtx = mapCurrCtx.get(mapJoin.getParentOperators().get(pos));
      Task<? extends Serializable> currTask    = mapredCtx.getCurrTask();
      mapredWork currPlan = (mapredWork) currTask.getWork();
      Operator<? extends Serializable> currTopOp   = mapredCtx.getCurrTopOp();
      String currAliasId = mapredCtx.getCurrAliasId();
      Operator<? extends Serializable> reducer = mapJoin;
      HashMap<Operator<? extends Serializable>, Task<? extends Serializable>> opTaskMap = ctx.getOpTaskMap();
      Task<? extends Serializable> opMapTask = opTaskMap.get(reducer);
     
      ctx.setCurrTopOp(currTopOp);
      ctx.setCurrAliasId(currAliasId);
      ctx.setCurrTask(currTask);
     
      // If the plan for this reducer does not exist, initialize the plan
      if (opMapTask == null) {
        assert currPlan.getReducer() == null;
        GenMapRedUtils.initMapJoinPlan(mapJoin, ctx, false, false, false, pos);
      }
      // The current plan can be thrown away after being merged with the original plan
      else {
        GenMapRedUtils.joinPlan(mapJoin, null, opMapTask, ctx, pos, false, false, false);
        currTask = opMapTask;
        ctx.setCurrTask(currTask);
      }
     
      mapCurrCtx.put(mapJoin, new GenMapRedCtx(ctx.getCurrTask(), ctx.getCurrTopOp(), ctx.getCurrAliasId()));
      return null;
    }
View Full Code Here

Examples of org.apache.hadoop.hive.ql.optimizer.GenMRProcContext.GenMapRedCtx

180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
     
      if (listMapJoinOps.contains(mapJoin)) {
        ctx.setCurrAliasId(null);
        ctx.setCurrTopOp(null);
        Map<Operator<? extends Serializable>, GenMapRedCtx> mapCurrCtx = ctx.getMapCurrCtx();
        mapCurrCtx.put((Operator<? extends Serializable>)nd, new GenMapRedCtx(ctx.getCurrTask(), null, null));
        return null;
      }

      ctx.setCurrMapJoinOp(mapJoin);
     
      Task<? extends Serializable> currTask = ctx.getCurrTask();
      GenMRMapJoinCtx mjCtx = ctx.getMapJoinCtx(mapJoin);
      if (mjCtx == null) {
        mjCtx = new GenMRMapJoinCtx();
        ctx.setMapJoinCtx(mapJoin, mjCtx);
      }
     
      mapredWork mjPlan = GenMapRedUtils.getMapRedWork();
      Task<? extends Serializable> mjTask = TaskFactory.get(mjPlan, parseCtx.getConf());
     
      tableDesc tt_desc =
        PlanUtils.getIntermediateFileTableDesc(PlanUtils.sortFieldSchemas(
            PlanUtils.getFieldSchemasFromRowSchema(mapJoin.getSchema(), "temporarycol")));
     
      // generate the temporary file
      Context baseCtx = parseCtx.getContext();
      String taskTmpDir = baseCtx.getMRTmpFileURI();
     
      // Add the path to alias mapping
      mjCtx.setTaskTmpDir(taskTmpDir);
      mjCtx.setTTDesc(tt_desc);
      mjCtx.setRootMapJoinOp(sel);
     
      sel.setParentOperators(null);
     
      // Create a file sink operator for this file name
      Operator<? extends Serializable> fs_op =
        OperatorFactory.get
        (new fileSinkDesc(taskTmpDir, tt_desc,
                          parseCtx.getConf().getBoolVar(HiveConf.ConfVars.COMPRESSINTERMEDIATE)),
         mapJoin.getSchema());
     
      assert mapJoin.getChildOperators().size() == 1;
      mapJoin.getChildOperators().set(0, fs_op);
     
      List<Operator<? extends Serializable>> parentOpList = new ArrayList<Operator<? extends Serializable>>();
      parentOpList.add(mapJoin);
      fs_op.setParentOperators(parentOpList);
     
      currTask.addDependentTask(mjTask);
     
      ctx.setCurrTask(mjTask);
      ctx.setCurrAliasId(null);
      ctx.setCurrTopOp(null);
     
      Map<Operator<? extends Serializable>, GenMapRedCtx> mapCurrCtx = ctx.getMapCurrCtx();
      mapCurrCtx.put((Operator<? extends Serializable>)nd, new GenMapRedCtx(ctx.getCurrTask(), null, null));
     
      return null;
    }
View Full Code Here

Examples of org.apache.hadoop.hive.ql.optimizer.GenMRProcContext.GenMapRedCtx

263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
      // find the branch on which this processor was invoked
      int pos = getPositionParent(mapJoin, stack);

      Map<Operator<? extends Serializable>, GenMapRedCtx> mapCurrCtx = ctx.getMapCurrCtx();
      GenMapRedCtx mapredCtx = mapCurrCtx.get(mapJoin.getParentOperators().get(pos));
      Task<? extends Serializable> currTask    = mapredCtx.getCurrTask();
      mapredWork currPlan = (mapredWork) currTask.getWork();
      String currAliasId = mapredCtx.getCurrAliasId();
      Operator<? extends Serializable> reducer = mapJoin;
      HashMap<Operator<? extends Serializable>, Task<? extends Serializable>> opTaskMap = ctx.getOpTaskMap();
      Task<? extends Serializable> opMapTask = opTaskMap.get(reducer);
     
      ctx.setCurrTask(currTask);
     
      // If the plan for this reducer does not exist, initialize the plan
      if (opMapTask == null) {
        assert currPlan.getReducer() == null;
        GenMapRedUtils.initMapJoinPlan(mapJoin, ctx, true, false, false, pos);
      }
      // The current plan can be thrown away after being merged with the original plan
      else {
        GenMapRedUtils.joinPlan(mapJoin, currTask, opMapTask, ctx, pos, false, true, false);
        currTask = opMapTask;
        ctx.setCurrTask(currTask);
      }
     
      mapCurrCtx.put(mapJoin, new GenMapRedCtx(ctx.getCurrTask(), null, null));
      return null;
    }
View Full Code Here

Examples of org.apache.hadoop.hive.ql.optimizer.GenMRProcContext.GenMapRedCtx

316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
      // find the branch on which this processor was invoked
      int pos = getPositionParent(mapJoin, stack);

      Map<Operator<? extends Serializable>, GenMapRedCtx> mapCurrCtx = ctx.getMapCurrCtx();
      GenMapRedCtx mapredCtx = mapCurrCtx.get(mapJoin.getParentOperators().get(pos));
      Task<? extends Serializable> currTask    = mapredCtx.getCurrTask();
      mapredWork currPlan = (mapredWork) currTask.getWork();
      Operator<? extends Serializable> reducer = mapJoin;
      HashMap<Operator<? extends Serializable>, Task<? extends Serializable>> opTaskMap = ctx.getOpTaskMap();
      Task<? extends Serializable> opMapTask = opTaskMap.get(reducer);
     
      // union result cannot be a map table
      boolean local = (pos == ((mapJoinDesc)mapJoin.getConf()).getPosBigTable()) ? false : true;
      if (local)
        throw new SemanticException(ErrorMsg.INVALID_MAPJOIN_TABLE.getMsg());
     
      // If the plan for this reducer does not exist, initialize the plan
      if (opMapTask == null) {
        assert currPlan.getReducer() == null;
        ctx.setCurrMapJoinOp(mapJoin);
        GenMapRedUtils.initMapJoinPlan(mapJoin, ctx, true, true, false, pos);
        ctx.setCurrUnionOp(null);
      }
      // The current plan can be thrown away after being merged with the original plan
      else {
        Task<? extends Serializable> uTask = ctx.getUnionTask(ctx.getCurrUnionOp()).getUTask();
        if (uTask.getId().equals(opMapTask.getId()))
          GenMapRedUtils.joinPlan(mapJoin, null, opMapTask, ctx, pos, false, false, true);
        else
          GenMapRedUtils.joinPlan(mapJoin, uTask, opMapTask, ctx, pos, false, false, true);
        currTask = opMapTask;
        ctx.setCurrTask(currTask);
      }
     
      mapCurrCtx.put(mapJoin, new GenMapRedCtx(ctx.getCurrTask(), ctx.getCurrTopOp(), ctx.getCurrAliasId()));
      return 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.