Package org.apache.pig.backend.hadoop.executionengine.physicalLayer

Examples of org.apache.pig.backend.hadoop.executionengine.physicalLayer.Result

125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
     * format, i.e, (key,indexedTuple(value))
     */
    @Override
    public Result getNext(Tuple t) throws ExecException {
       
        Result inp = null;
        Result res = null;
       
        // Load the skewed join key partitioning file
        if (!loaded) {
          try {
            loadPartitionFile();
          } catch (Exception e) {
            throw new RuntimeException(e);
          }
        }
   
        while (true) {
            inp = processInput();
            if (inp.returnStatus == POStatus.STATUS_EOP || inp.returnStatus == POStatus.STATUS_ERR)
                break;
            if (inp.returnStatus == POStatus.STATUS_NULL)
                continue;
           
            for (PhysicalPlan ep : plans) {
                ep.attachInput((Tuple)inp.result);
            }
            List<Result> resLst = new ArrayList<Result>();
            for (ExpressionOperator op : leafOps){
               
                switch(op.getResultType()){
                case DataType.BAG:
                    res = op.getNext(dummyBag);
                    break;
                case DataType.BOOLEAN:
                    res = op.getNext(dummyBool);
                    break;
                case DataType.BYTEARRAY:
                    res = op.getNext(dummyDBA);
                    break;
                case DataType.CHARARRAY:
                    res = op.getNext(dummyString);
                    break;
                case DataType.DOUBLE:
                    res = op.getNext(dummyDouble);
                    break;
                case DataType.FLOAT:
                    res = op.getNext(dummyFloat);
                    break;
                case DataType.INTEGER:
                    res = op.getNext(dummyInt);
                    break;
                case DataType.LONG:
                    res = op.getNext(dummyLong);
                    break;
                case DataType.MAP:
                    res = op.getNext(dummyMap);
                    break;
                case DataType.TUPLE:
                    res = op.getNext(dummyTuple);
                    break;
                }
                if(res.returnStatus!=POStatus.STATUS_OK)
                    return new Result();
                resLst.add(res);
            }
            res.result = constructPROutput(resLst,(Tuple)inp.result);
           
            return res;
View Full Code Here
154
155
156
157
158
159
160
161
162
163
164
                    return eopResult ;
                }
                if(lastInd >= inputs.size() || done.nextClearBit(lastInd) >= inputs.size())
                    lastInd = 0;
                int ind = done.nextClearBit(lastInd);
                Result res;

                while(true){
                    if(reporter!=null) reporter.progress();
                    res = inputs.get(ind).getNext(t);
                    lastInd = ind + 1;
View Full Code Here
197
198
199
200
201
202
203
204
205
206
207
208
     * physical plan and returns it maintaining an additional state
     * to denote the begin and end of the nested plan processing.
     */
    @Override
    public Result getNext(Tuple t) throws ExecException {     
        Result res = null;
        Result inp = null;
        //The nested plan is under processing
        //So return tuples that the generate oper
        //returns
        if(processingPlan){
            while(true) {
View Full Code Here
306
307
308
309
310
311
312
313
314
315
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
            return res;
        }
    }

    protected Result processPlan() throws ExecException{     
        Result res = new Result();

        //We check if all the databags have exhausted the tuples. If so we enforce the reading of new data by setting data and its to null
        if(its != null) {
            boolean restartIts = true;
            for(int i = 0; i < noItems; ++i) {
                if(its[i] != null && isToBeFlattenedArray[i] == true)
                    restartIts &= !its[i].hasNext();
            }
            //this means that all the databags have reached their last elements. so we need to force reading of fresh databags
            if(restartIts) {
                its = null;
                data = null;
            }
        }
       
        if(its == null) {
            //getNext being called for the first time OR starting with a set of new data from inputs
            its = new Iterator[noItems];
            bags = new Object[noItems];
           
            for(int i = 0; i < noItems; ++i) {
                //Getting the iterators
                //populate the input data
                Result inputData = null;
                switch(resultTypes[i]) {
                case DataType.BAG:
                    inputData = planLeafOps[i].getNext(dummyBag);
                    break;
View Full Code Here
57
58
59
60
61
62
63
64
65
66
67
    }
   

    public Result getNext(Tuple t) throws ExecException {
      while(true) {
        Result in = processInput();
        if (in.returnStatus == POStatus.STATUS_NULL) {
          continue;
        }
       
        if (in.returnStatus == POStatus.STATUS_OK) {
View Full Code Here
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
     * format, i.e, (key,indexedTuple(value))
     */
    @Override
    public Result getNext(Tuple t) throws ExecException {
       
        Result inp = null;
        Result res = ERR_RESULT;
        while (true) {
            inp = processInput();
            if (inp.returnStatus == POStatus.STATUS_EOP || inp.returnStatus == POStatus.STATUS_ERR)
                break;
            if (inp.returnStatus == POStatus.STATUS_NULL)
                continue;
           
            for (PhysicalPlan ep : plans) {
                ep.attachInput((Tuple)inp.result);
            }
            List<Result> resLst = new ArrayList<Result>();
            for (ExpressionOperator op : leafOps){
               
                switch(op.getResultType()){
                case DataType.BAG:
                    res = op.getNext(dummyBag);
                    break;
                case DataType.BOOLEAN:
                    res = op.getNext(dummyBool);
                    break;
                case DataType.BYTEARRAY:
                    res = op.getNext(dummyDBA);
                    break;
                case DataType.CHARARRAY:
                    res = op.getNext(dummyString);
                    break;
                case DataType.DOUBLE:
                    res = op.getNext(dummyDouble);
                    break;
                case DataType.FLOAT:
                    res = op.getNext(dummyFloat);
                    break;
                case DataType.INTEGER:
                    res = op.getNext(dummyInt);
                    break;
                case DataType.LONG:
                    res = op.getNext(dummyLong);
                    break;
                case DataType.MAP:
                    res = op.getNext(dummyMap);
                    break;
                case DataType.TUPLE:
                    res = op.getNext(dummyTuple);
                    break;
                default:
                    log.error("Invalid result type: "
                            + DataType.findType(op.getResultType()));
                    break;
                }
               
                // allow null as group by key
                if (res.returnStatus != POStatus.STATUS_OK
                        && res.returnStatus != POStatus.STATUS_NULL) {
                    return new Result();
                }
               
                resLst.add(res);
            }
            res.result = constructLROutput(resLst,(Tuple)inp.result);
View Full Code Here
126
127
128
129
130
131
132
133
134
135
136
        }
    }
   
    @Override
    public Result getNext(Tuple t) throws ExecException {
        Result res = processInput();
        try {
            switch (res.returnStatus) {
            case POStatus.STATUS_OK:
                storer.putNext((Tuple)res.result);
                res = empty;
View Full Code Here
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
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
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
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
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
     * format, i.e, (key,indexedTuple(value))
     */
    @Override
    public Result getNext(Tuple t) throws ExecException {
       
        Result inp = null;
        Result res = ERR_RESULT;
        while (true) {
            inp = processInput();
            if (inp.returnStatus == POStatus.STATUS_EOP || inp.returnStatus == POStatus.STATUS_ERR)
                break;
            if (inp.returnStatus == POStatus.STATUS_NULL)
                continue;
           
            for (PhysicalPlan ep : plans) {
                ep.attachInput((Tuple)inp.result);
            }
           
            List<Result> resLst = new ArrayList<Result>();
           
            if (secondaryPlans!=null) {
                for (PhysicalPlan ep : secondaryPlans) {
                    ep.attachInput((Tuple)inp.result);
                }
            }
           
            List<Result> secondaryResLst = null;
            if (secondaryLeafOps!=null)
                secondaryResLst = new ArrayList<Result>();
           
            for (ExpressionOperator op : leafOps){
               
                switch(op.getResultType()){
                case DataType.BAG:
                    res = op.getNext(dummyBag);
                    break;
                case DataType.BOOLEAN:
                    res = op.getNext(dummyBool);
                    break;
                case DataType.BYTEARRAY:
                    res = op.getNext(dummyDBA);
                    break;
                case DataType.CHARARRAY:
                    res = op.getNext(dummyString);
                    break;
                case DataType.DOUBLE:
                    res = op.getNext(dummyDouble);
                    break;
                case DataType.FLOAT:
                    res = op.getNext(dummyFloat);
                    break;
                case DataType.INTEGER:
                    res = op.getNext(dummyInt);
                    break;
                case DataType.LONG:
                    res = op.getNext(dummyLong);
                    break;
                case DataType.MAP:
                    res = op.getNext(dummyMap);
                    break;
                case DataType.TUPLE:
                    res = op.getNext(dummyTuple);
                    break;
                default:
                    log.error("Invalid result type: " + DataType.findType(op.getResultType()));
                    break;
                }
               
                // allow null as group by key
                if (res.returnStatus != POStatus.STATUS_OK && res.returnStatus != POStatus.STATUS_NULL) {
                    return new Result();
                }
             
                resLst.add(res);
            }
           
            if (secondaryLeafOps!=null)
            {
                for (ExpressionOperator op : secondaryLeafOps){
                   
                    switch(op.getResultType()){
                    case DataType.BAG:
                        res = op.getNext(dummyBag);
                        break;
                    case DataType.BOOLEAN:
                        res = op.getNext(dummyBool);
                        break;
                    case DataType.BYTEARRAY:
                        res = op.getNext(dummyDBA);
                        break;
                    case DataType.CHARARRAY:
                        res = op.getNext(dummyString);
                        break;
                    case DataType.DOUBLE:
                        res = op.getNext(dummyDouble);
                        break;
                    case DataType.FLOAT:
                        res = op.getNext(dummyFloat);
                        break;
                    case DataType.INTEGER:
                        res = op.getNext(dummyInt);
                        break;
                    case DataType.LONG:
                        res = op.getNext(dummyLong);
                        break;
                    case DataType.MAP:
                        res = op.getNext(dummyMap);
                        break;
                    case DataType.TUPLE:
                        res = op.getNext(dummyTuple);
                        break;
                    default:
                        log.error("Invalid result type: " + DataType.findType(op.getResultType()));
                        break;
                    }
                   
                    // allow null as group by key
                    if (res.returnStatus != POStatus.STATUS_OK && res.returnStatus != POStatus.STATUS_NULL) {
                        return new Result();
                    }
                   
                    secondaryResLst.add(res);
                }
            }
View Full Code Here
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
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
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
293
294
295
296
297
298
299
300
301
        return false;
    }

    @Override
    public Result getNext(Tuple t) throws ExecException {
        Result res = null;
        Result inp = null;
        if (!setUp) {
            setUpHashMap();
            setUp = true;
        }
        if (processingPlan) {
            // Return tuples from the for each operator
            // Assumes that it is configured appropriately with
            // the bags for the current key.
            while (true) {
                res = fe.getNext(dummyTuple);

                if (res.returnStatus == POStatus.STATUS_OK) {
                    return res;
                }
                if (res.returnStatus == POStatus.STATUS_EOP) {
                    // We have completed all cross-products now its time to move
                    // to next tuple of left side
                    processingPlan = false;                   
                    break;
                }
                if (res.returnStatus == POStatus.STATUS_ERR) {
                    return res;
                }
                if (res.returnStatus == POStatus.STATUS_NULL) {
                    continue;
                }
            }
        }
        while (true) {
            // Process the current input
            inp = processInput();
            if (inp.returnStatus == POStatus.STATUS_EOP
                    || inp.returnStatus == POStatus.STATUS_ERR)
                return inp;
            if (inp.returnStatus == POStatus.STATUS_NULL) {
                continue;
            }

            // Separate Key & Value using the fragment's LR operator
            POLocalRearrange lr = LRs[fragment];
            lr.attachInput((Tuple) inp.result);
            Result lrOut = lr.getNext(dummyTuple);
            if (lrOut.returnStatus != POStatus.STATUS_OK) {
                log
                        .error("LocalRearrange isn't configured right or is not working");
                return new Result();
            }
            Tuple lrOutTuple = (Tuple) lrOut.result;
            Tuple key = TupleFactory.getInstance().newTuple(1);
            key.set(0, lrOutTuple.get(1));
            Tuple value = getValueTuple(lr, lrOutTuple);
            lr.detachInput();
            // Configure the for each operator with the relevant bags
            int i = -1;
            boolean noMatch = false;
            for (ConstantExpression ce : constExps) {
                ++i;
                if (i == fragment) {
                    // We set the first CE as the tuple from fragmented Left
                    ce.setValue(value);
                    continue;
                }
                Map<Tuple, List<Tuple>> replicate = replicates[i];
                if (!replicate.containsKey(key)) {
                    if (isLeftOuterJoin) {
                        ce.setValue(nullBag);
                    }
                    noMatch = true;
                    break;
                }
                ce.setValue(new NonSpillableDataBag(replicate.get(key)));
            }

            // If this is not LeftOuter Join and there was no match we
            // skip the processing of this left tuple and move ahead
            if (!isLeftOuterJoin && noMatch)
                continue;
            fe.attachInput(dumTup);
            processingPlan = true;

            // We are all set, we call getNext (this function) which will call
            // getNext on ForEach
            // And that will return one tuple of Cross-Product between set
            // constant Expressions
            // All subsequent calls ( by parent ) to this function will return
            // next tuple of crossproduct
            Result gn = getNext(dummyTuple);

            return gn;
        }
    }
View Full Code Here
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
        public void run() {
            try {
                // Read tuples from the previous operator in the pipeline
                // and pass it to the executable
                while (true) {
                    Result inp = null;
                    inp = binaryInputQueue.take();
                    synchronized (poStream) {
                        // notify waiting producer
                        // the if check is to keep "findbugs"
                        // happy
                        if(inp != null)
                            poStream.notifyAll();
                    }
                    // We should receive an EOP only when *ALL* input
                    // for this process has already been sent and no
                    // more input is expected
                    if (inp != null && inp.returnStatus == POStatus.STATUS_EOP) {
                        // signal cleanup in ExecutableManager
                        close();
                        return;
                    }
                    if (inp != null && inp.returnStatus == POStatus.STATUS_OK) {
                        // Check if there was a problem with the managed process
                        if (outerrThreadsError != null) {
                            throw new IOException(
                                    "Output/Error thread failed with: "
                                            + outerrThreadsError);
                        }

                        // Pass the serialized tuple to the executable via the
                        // InputHandler
                        Tuple t = null;
                        try {
                            t = (Tuple) inp.result;
                            inputHandler.putNext(t);                           
                        } catch (IOException e) {
                            // if input type is synchronous then it could
                            // be related to the process terminating
                            if(inputHandler.getInputType() == InputType.SYNCHRONOUS) {
                                LOG.warn("Exception while trying to write to stream binary's input", e);
                                // could be because the process
                                // died OR closed the input stream
                                // we will only call close() here and not
                                // worry about deducing whether the process died
                                // normally or abnormally - if there was any real
                                // issue the ProcessOutputThread should see
                                // a non zero exit code from the process and send
                                // a POStatus.STATUS_ERR back - what if we got
                                // an IOException because there was only an issue with
                                // writing to input of the binary - hmm..hope that means
                                // the process died abnormally!!
                                close();
                                return;
                            } else {
                                // asynchronous case - then this is a real exception
                                LOG.error("Exception while trying to write to stream binary's input", e);
                                // send POStatus.STATUS_ERR to POStream to signal the error
                                // Generally the ProcessOutputThread would do this but now
                                // we should do it here since neither the process nor the
                                // ProcessOutputThread will ever be spawned
                                Result res = new Result(POStatus.STATUS_ERR,
                                        "Exception while trying to write to stream binary's input" + e.getMessage());
                                sendOutput(poStream.getBinaryOutputQueue(), res);
                                throw e;
                            }
                        }
View Full Code Here
TOP

Related Classes of org.apache.pig.backend.hadoop.executionengine.physicalLayer.Result

  • org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigCombiner$Combine
  • org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigGenericMapBase
  • org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigGenericMapReduce$MapWithPartitionIndex
  • org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigGenericMapReduce$Reduce
  • org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigGenericMapReduce$ReduceWithComparator
  • org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigMapBase
  • org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigMapReduce$MapWithPartitionIndex
  • org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigMapReduce$Reduce
  • org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigMapReduce$ReduceWithComparator
  • org.apache.pig.backend.hadoop.executionengine.physicalLayer.expressionOperators.Add
    • Copyright © 2018 www.massapicom. 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.