Examples of PigSplit

Examples of org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigSplit

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
        List<InputSplit> result = pigInputFormat.getPigSplits(rawSplits, 0, ok,
                        null, true, conf);
        Assert.assertEquals(2, result.size());
        int idx = 0;
        for (InputSplit split : result) {
            PigSplit pigSplit = (PigSplit) split;
            int len = pigSplit.getNumPaths();
            if (idx == 0) {
                Assert.assertEquals(2, len);
                checkLocations(pigSplit.getLocations(), new String[] {
                                "l1", "l4", "l5"
                });
                Assert.assertEquals(500, pigSplit.getLength(0));
                Assert.assertEquals(100, pigSplit.getLength(1));
            }
            else {
                Assert.assertEquals(4, len);
                Assert.assertEquals(500, pigSplit.getLength(0));
                checkLocations(pigSplit.getLocations(), new String[] {
                                "l1", "l2", "l3"
                });
                Assert.assertEquals(200, pigSplit.getLength(1));
                Assert.assertEquals(200, pigSplit.getLength(2));
                Assert.assertEquals(100, pigSplit.getLength(3));
            }
            idx++;
        }
    }
View Full Code Here

Examples of org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigSplit

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
        List<InputSplit> result = pigInputFormat.getPigSplits(rawSplits, 0, ok,
                        null, false, conf);
        Assert.assertEquals(3, result.size());
        int index = 0;
        for (InputSplit split : result) {
            PigSplit pigSplit = (PigSplit) split;
            int len = pigSplit.getNumPaths();
            if (index == 0) {
                Assert.assertEquals(1, len);
                checkLocations(pigSplit.getLocations(), new String[] {
                                "l1", "l2", "l3"
                });
                Assert.assertEquals(500, pigSplit.getLength(0));
            }
            else if (index == 1) {
                Assert.assertEquals(1, len);
                checkLocations(pigSplit.getLocations(), new String[] {
                                "l1", "l2", "l3"
                });
                Assert.assertEquals(400, pigSplit.getLength(0));
            }
            else {
                Assert.assertEquals(1, len);
                checkLocations(pigSplit.getLocations(), new String[] {
                                "l1", "l4", "l5"
                });
                Assert.assertEquals(400, pigSplit.getLength(0));
            }
            index++;
        }
    }
View Full Code Here

Examples of org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigSplit

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
        List<InputSplit> result = pigInputFormat.getPigSplits(rawSplits, 0, ok,
                        null, true, conf);
        Assert.assertEquals(3, result.size());
        int idx = 0;
        for (InputSplit split : result) {
            PigSplit pigSplit = (PigSplit) split;
            int len = pigSplit.getNumPaths();
            if (idx == 0) {
                Assert.assertEquals(2, len);
                checkLocations(pigSplit.getLocations(), new String[] {
                                "l1", "l2", "l3"
                });
                Assert.assertEquals(600, pigSplit.getLength(0));
                Assert.assertEquals(400, pigSplit.getLength(1));
            }
            else if (idx == 1) {
                Assert.assertEquals(3, len);
                checkLocations(pigSplit.getLocations(), new String[] {
                                "l1", "l2", "l3"
                });
                Assert.assertEquals(500, pigSplit.getLength(0));
                Assert.assertEquals(300, pigSplit.getLength(1));
                Assert.assertEquals(200, pigSplit.getLength(2));
            }
            else {
                Assert.assertEquals(1, len);
                checkLocations(pigSplit.getLocations(), new String[] {
                                "l1", "l2", "l3"
                });
                Assert.assertEquals(100, pigSplit.getLength(0));
            }
            idx++;
        }
    }
View Full Code Here

Examples of org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigSplit

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
        List<InputSplit> result = pigInputFormat.getPigSplits(rawSplits, 0, ok,
                        null, true, conf);
        Assert.assertEquals(3, result.size());
        int idx = 0;
        for (InputSplit split : result) {
            PigSplit pigSplit = (PigSplit) split;
            int len = pigSplit.getNumPaths();
            if (idx == 0) {
                Assert.assertEquals(2, len);
                checkLocations(pigSplit.getLocations(), new String[] {
                                "l1", "l2", "l3"
                });
                Assert.assertEquals(600, pigSplit.getLength(0));
                Assert.assertEquals(400, pigSplit.getLength(1));
            }
            else if (idx == 1) {
                Assert.assertEquals(3, len);
                checkLocations(pigSplit.getLocations(), new String[] {
                                "l1", "l2", "l3"
                });
                Assert.assertEquals(500, pigSplit.getLength(0));
                Assert.assertEquals(300, pigSplit.getLength(1));
                Assert.assertEquals(200, pigSplit.getLength(2));
            }
            else {
                Assert.assertEquals(1, len);
                checkLocations(pigSplit.getLocations(), new String[] {
                                "l1", "l2", "l3"
                });
                Assert.assertEquals(100, pigSplit.getLength(0));
            }
            idx++;
        }
    }
View Full Code Here

Examples of org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigSplit

412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
        List<InputSplit> result = pigInputFormat.getPigSplits(rawSplits, 0, ok,
                        null, true, conf);
        Assert.assertEquals(result.size(), 1);
        int index = 0;
        for (InputSplit split : result) {
            PigSplit pigSplit = (PigSplit) split;
            int len = pigSplit.getNumPaths();
            Assert.assertEquals(3, len);
            checkLocations(pigSplit.getLocations(), new String[] {
                            "l1", "l2", "l3", "l4", "l5"
            });
            Assert.assertEquals(200, pigSplit.getLength(0));
            Assert.assertEquals(100, pigSplit.getLength(1));
            Assert.assertEquals(100, pigSplit.getLength(2));
            index++;
        }
    }
View Full Code Here

Examples of org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigSplit

443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
        List<InputSplit> result = pigInputFormat.getPigSplits(rawSplits, 0, ok,
                        null, true, conf);
        Assert.assertEquals(result.size(), 1);
        int index = 0;
        for (InputSplit split : result) {
            PigSplit pigSplit = (PigSplit) split;
            int len = pigSplit.getNumPaths();
            Assert.assertEquals(3, len);
            // only 5 locations are in list: refer to PIG-1648 for more details
            checkLocationOrdering(pigSplit.getLocations(), new String[] {
                            "l5", "l1", "l6", "l3", "l4"
            });
            Assert.assertEquals(400, pigSplit.getLength(0));
            Assert.assertEquals(200, pigSplit.getLength(1));
            Assert.assertEquals(100, pigSplit.getLength(2));
            index++;
        }
    }
View Full Code Here

Examples of org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigSplit

118
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
185
186
187
188
189
190
191
                ArrayList<ArrayList<OperatorKey>> inpTargets =
                    (ArrayList<ArrayList<OperatorKey>>)
                      ObjectSerializer.deserialize(jobConf.get("pig.inpTargets"));
                intermediateData.clear();
                MapReduceOper mro = jobToMroMap.get(job);
                PigSplit split = null;
                List<POStore> stores = null;
                PhysicalOperator pack = null;
                // revisit as there are new physical operators from MR compilation
                if (!mro.mapPlan.isEmpty())
                    attacher.revisit(mro.mapPlan);
                if (!mro.reducePlan.isEmpty()) {
                    attacher.revisit(mro.reducePlan);
                    pack = mro.reducePlan.getRoots().get(0);
                }
               
                List<POLoad> lds = PlanHelper.getPhysicalOperators(mro.mapPlan, POLoad.class);
                if (!mro.mapPlan.isEmpty()) {
                    stores = PlanHelper.getPhysicalOperators(mro.mapPlan, POStore.class);
                }
                if (!mro.reducePlan.isEmpty()) {
                    if (stores == null)
                        stores = PlanHelper.getPhysicalOperators(mro.reducePlan, POStore.class);
                    else
                        stores.addAll(PlanHelper.getPhysicalOperators(mro.reducePlan, POStore.class));
                }

                for (POStore store : stores) {
                    output.put(store.getSFile().getFileName(), attacher.getDataMap().get(store));
                }
              
                OutputAttacher oa = new OutputAttacher(mro.mapPlan, output);
                oa.visit();
               
                if (!mro.reducePlan.isEmpty()) {
                    oa = new OutputAttacher(mro.reducePlan, output);
                    oa.visit();
                }
                int index = 0;
                for (POLoad ld : lds) {
                    input = output.get(ld.getLFile().getFileName());
                    if (input == null && baseData != null) {
                        for (LogicalRelationalOperator lo : baseData.keySet()) {
                            if (((LOLoad) lo).getSchemaFile().equals(ld.getLFile().getFileName()))
                            {
                                 input = baseData.get(lo);
                                 break;
                            }
                        }
                    }
                    if (input != null)
                        mro.mapPlan.remove(ld);
                }
                for (POLoad ld : lds) {
                    // check newly generated data first
                    input = output.get(ld.getLFile().getFileName());
                    if (input == null && baseData != null) {
                        if (input == null && baseData != null) {
                            for (LogicalRelationalOperator lo : baseData.keySet()) {
                                if (((LOLoad) lo).getSchemaFile().equals(ld.getLFile().getFileName()))
                                {
                                     input = baseData.get(lo);
                                     break;
                                }
                            }
                        }
                    }
                    needFileInput = (input == null);
                    split = new PigSplit(null, index, needFileInput ? emptyInpTargets : inpTargets.get(index), 0);
                    ++index;
                    Mapper<Text, Tuple, PigNullableWritable, Writable> map;

                    if (mro.reducePlan.isEmpty()) {
                        // map-only
View Full Code Here

Examples of org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigSplit

117
118
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
185
186
187
188
189
190
                ArrayList<ArrayList<OperatorKey>> inpTargets =
                    (ArrayList<ArrayList<OperatorKey>>)
                      ObjectSerializer.deserialize(jobConf.get("pig.inpTargets"));
                intermediateData.clear();
                MapReduceOper mro = jobToMroMap.get(job);
                PigSplit split = null;
                List<POStore> stores = null;
                PhysicalOperator pack = null;
                // revisit as there are new physical operators from MR compilation
                if (!mro.mapPlan.isEmpty())
                    attacher.revisit(mro.mapPlan);
                if (!mro.reducePlan.isEmpty()) {
                    attacher.revisit(mro.reducePlan);
                    pack = mro.reducePlan.getRoots().get(0);
                }
               
                List<POLoad> lds = PlanHelper.getPhysicalOperators(mro.mapPlan, POLoad.class);
                if (!mro.mapPlan.isEmpty()) {
                    stores = PlanHelper.getPhysicalOperators(mro.mapPlan, POStore.class);
                }
                if (!mro.reducePlan.isEmpty()) {
                    if (stores == null)
                        stores = PlanHelper.getPhysicalOperators(mro.reducePlan, POStore.class);
                    else
                        stores.addAll(PlanHelper.getPhysicalOperators(mro.reducePlan, POStore.class));
                }

                for (POStore store : stores) {
                    output.put(store.getSFile().getFileName(), attacher.getDataMap().get(store));
                }
              
                OutputAttacher oa = new OutputAttacher(mro.mapPlan, output);
                oa.visit();
               
                if (!mro.reducePlan.isEmpty()) {
                    oa = new OutputAttacher(mro.reducePlan, output);
                    oa.visit();
                }
                int index = 0;
                for (POLoad ld : lds) {
                    input = output.get(ld.getLFile().getFileName());
                    if (input == null && baseData != null) {
                        for (LogicalRelationalOperator lo : baseData.keySet()) {
                            if (((LOLoad) lo).getSchemaFile().equals(ld.getLFile().getFileName()))
                            {
                                 input = baseData.get(lo);
                                 break;
                            }
                        }
                    }
                    if (input != null)
                        mro.mapPlan.remove(ld);
                }
                for (POLoad ld : lds) {
                    // check newly generated data first
                    input = output.get(ld.getLFile().getFileName());
                    if (input == null && baseData != null) {
                        if (input == null && baseData != null) {
                            for (LogicalRelationalOperator lo : baseData.keySet()) {
                                if (((LOLoad) lo).getSchemaFile().equals(ld.getLFile().getFileName()))
                                {
                                     input = baseData.get(lo);
                                     break;
                                }
                            }
                        }
                    }
                    needFileInput = (input == null);
                    split = new PigSplit(null, index, needFileInput ? emptyInpTargets : inpTargets.get(index), 0);
                    ++index;
                    Mapper<Text, Tuple, PigNullableWritable, Writable> map;

                    if (mro.reducePlan.isEmpty()) {
                        // map-only
View Full Code Here

Examples of org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigSplit

196
197
198
199
200
201
202
203
204
205
        reader = inputFormat.createRecordReader(curSplit, tAContext);
        reader.initialize(curSplit, tAContext);
        // create a dummy pigsplit - other than the actual split, the other
        // params are really not needed here where we are just reading the
        // input completely
        PigSplit pigSplit = new PigSplit(new InputSplit[] {curSplit}, -1,
                new ArrayList<OperatorKey>(), -1);
        wrappedLoadFunc.prepareToRead(reader, pigSplit);
        return true;
    }
View Full Code Here

Examples of org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.PigSplit

105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
        List<InputSplit> result = pigInputFormat.getPigSplits(rawSplits, 0, ok,
                        null, true, conf);
        Assert.assertEquals(result.size(), 2);
        int index = 0;
        for (InputSplit split : result) {
            PigSplit pigSplit = (PigSplit) split;
            int len = pigSplit.getNumPaths();
            if (index == 0) {
                Assert.assertEquals(2, len);
                checkLocations(pigSplit.getLocations(), new String[] {
                                "l1", "l2", "l3"
                });
                Assert.assertEquals(500, pigSplit.getLength(0));
                Assert.assertEquals(400, pigSplit.getLength(1));
            }
            else {
                Assert.assertEquals(1, len);
                checkLocations(pigSplit.getLocations(), new String[] {
                                "l1", "l4", "l5"
                });
                Assert.assertEquals(400, pigSplit.getLength(0));
            }
            index++;
        }
    }
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.