Package org.apache.hadoop.mapred

Examples of org.apache.hadoop.mapred.InputSplit

112
113
114
115
116
117
118
119
120
121
122
123
124
      LOG.info("splitting: got =        " + splits.length);

      // we should have a single split as the length is comfortably smaller than
      // the block size
      Assert.assertEquals("We got more than one splits!", 1, splits.length);
      InputSplit split = splits[0];
      Assert.assertEquals("It should be TezGroupedSplit",
        TezGroupedSplit.class, split.getClass());

      // check the split
      BitSet bits = new BitSet(length);
      LOG.debug("split= " + split);
      RecordReader<LongWritable, Text> reader =
View Full Code Here
300
301
302
303
304
305
306
307
308
309
310
311
312
    format.setInputFormat(mockWrappedFormat);
   
    job = (JobConf) TezMapReduceSplitsGrouper.createConfigBuilder(job)
        .setGroupingSplitSize(50*1000*1000l, 500*1000*1000l)
        .build();
    InputSplit mockSplit1 = mock(InputSplit.class);
    when(mockSplit1.getLength()).thenReturn(10*1000*1000l);
    when(mockSplit1.getLocations()).thenReturn(null);
    int numSplits = 100;
    InputSplit[] mockSplits = new InputSplit[numSplits];
    for (int i=0; i<numSplits; i++) {
      mockSplits[i] = mockSplit1;
    }
View Full Code Here
342
343
344
345
346
347
348
349
350
351
352
353
354
    // put multiple splits with multiple copies in the same location
    String[] locations = {"common", "common", "common"};
    int numSplits = 3;
    InputSplit[] mockSplits = new InputSplit[numSplits];
    for (int i=0; i<numSplits; i++) {
      InputSplit mockSplit = mock(InputSplit.class);
      when(mockSplit.getLength()).thenReturn(10*1000*1000l);
      when(mockSplit.getLocations()).thenReturn(locations);
      mockSplits[i] = mockSplit;
    }
    when(mockWrappedFormat.getSplits((JobConf)anyObject(), anyInt())).thenReturn(mockSplits);
   
    format.setDesiredNumberOfSplits(1);
View Full Code Here
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
    format.setInputFormat(mockWrappedFormat);
   
    // put multiple splits with multiple copies in the same location
    int numSplits = 3;
    InputSplit[] mockSplits = new InputSplit[numSplits];
    InputSplit mockSplit1 = mock(InputSplit.class);
    when(mockSplit1.getLength()).thenReturn(10*1000*1000l);
    when(mockSplit1.getLocations()).thenReturn(null);
    mockSplits[0] = mockSplit1;
    InputSplit mockSplit2 = mock(InputSplit.class);
    when(mockSplit2.getLength()).thenReturn(10*1000*1000l);
    when(mockSplit2.getLocations()).thenReturn(new String[] {null});
    mockSplits[1] = mockSplit2;
    InputSplit mockSplit3 = mock(InputSplit.class);
    when(mockSplit3.getLength()).thenReturn(10*1000*1000l);
    when(mockSplit3.getLocations()).thenReturn(new String[] {null, null});
    mockSplits[2] = mockSplit3;

    when(mockWrappedFormat.getSplits((JobConf)anyObject(), anyInt())).thenReturn(mockSplits);
   
    format.setDesiredNumberOfSplits(1);
View Full Code Here
135
136
137
138
139
140
141
142
143
144
145
146
147
148
    split.write(out);
  }
 
  InputSplit readWrappedSplit(DataInput in, Class<? extends InputSplit> clazz)
      throws IOException {
    InputSplit split;
    try {
      split = ReflectionUtils.newInstance(clazz, conf);
    } catch (Exception e) {
      throw new TezUncheckedException(e);
    }
    split.readFields(in);
    return split;
  }
View Full Code Here
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
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
  public void testSerializedPayload() throws IOException {

    Configuration conf = new Configuration(false);
    conf.setBoolean(MRJobConfig.MR_TEZ_INPUT_INITIALIZER_SERIALIZE_EVENT_PAYLOAD, true);
    ByteString confByteString = TezUtils.createByteStringFromConf(conf);
    InputSplit split1 = new InputSplitForTest(1);
    InputSplit split2 = new InputSplitForTest(2);
    MRSplitProto proto1 = MRInputHelpers.createSplitProto(split1);
    MRSplitProto proto2 = MRInputHelpers.createSplitProto(split2);
    MRSplitsProto.Builder splitsProtoBuilder = MRSplitsProto.newBuilder();
    splitsProtoBuilder.addSplits(proto1);
    splitsProtoBuilder.addSplits(proto2);
    MRInputUserPayloadProto.Builder payloadProto = MRInputUserPayloadProto.newBuilder();
    payloadProto.setSplits(splitsProtoBuilder.build());
    payloadProto.setConfigurationBytes(confByteString);
    UserPayload userPayload =
        UserPayload.create(payloadProto.build().toByteString().asReadOnlyByteBuffer());

    InputInitializerContext context = new TezRootInputInitializerContextForTest(userPayload);
    MRInputSplitDistributor splitDist = new MRInputSplitDistributor(context);

    List<Event> events = splitDist.initialize();

    assertEquals(3, events.size());
    assertTrue(events.get(0) instanceof InputUpdatePayloadEvent);
    assertTrue(events.get(1) instanceof InputDataInformationEvent);
    assertTrue(events.get(2) instanceof InputDataInformationEvent);

    InputDataInformationEvent diEvent1 = (InputDataInformationEvent) (events.get(1));
    InputDataInformationEvent diEvent2 = (InputDataInformationEvent) (events.get(2));

    assertNull(diEvent1.getDeserializedUserPayload());
    assertNull(diEvent2.getDeserializedUserPayload());

    assertNotNull(diEvent1.getUserPayload());
    assertNotNull(diEvent2.getUserPayload());

    MRSplitProto event1Proto = MRSplitProto.parseFrom(ByteString.copyFrom(diEvent1.getUserPayload()));
    InputSplit is1 = MRInputUtils.getOldSplitDetailsFromEvent(event1Proto, new Configuration());
    assertTrue(is1 instanceof InputSplitForTest);
    assertEquals(1, ((InputSplitForTest) is1).identifier);

    MRSplitProto event2Proto = MRSplitProto.parseFrom(ByteString.copyFrom(diEvent2.getUserPayload()));
    InputSplit is2 = MRInputUtils.getOldSplitDetailsFromEvent(event2Proto, new Configuration());
    assertTrue(is2 instanceof InputSplitForTest);
    assertEquals(2, ((InputSplitForTest) is2).identifier);
  }
View Full Code Here
105
106
107
108
109
110
111
112
113
114
115
116
  public void testDeserializedPayload() throws IOException {

    Configuration conf = new Configuration(false);
    conf.setBoolean(MRJobConfig.MR_TEZ_INPUT_INITIALIZER_SERIALIZE_EVENT_PAYLOAD, false);
    ByteString confByteString = TezUtils.createByteStringFromConf(conf);
    InputSplit split1 = new InputSplitForTest(1);
    InputSplit split2 = new InputSplitForTest(2);
    MRSplitProto proto1 = MRInputHelpers.createSplitProto(split1);
    MRSplitProto proto2 = MRInputHelpers.createSplitProto(split2);
    MRSplitsProto.Builder splitsProtoBuilder = MRSplitsProto.newBuilder();
    splitsProtoBuilder.addSplits(proto1);
    splitsProtoBuilder.addSplits(proto2);
View Full Code Here
167
168
169
170
171
172
173
174
175
176
177
    // Initialize input in-line since it sets parameters which may be used by the processor.
    // Done only for MRInput.
    // TODO use new method in MRInput to get required info
    //input.initialize(job, master);
   
    InputSplit inputSplit = input.getOldInputSplit();
   
    updateJobWithSplit(job, inputSplit);

    RecordReader in = new OldRecordReader(input);
View Full Code Here
172
173
174
175
176
177
178
179
180
    }

    Deserializer<InputSplit> deserializer = serializationFactory
        .getDeserializer(clazz);
    deserializer.open(splitProto.getSplitBytes().newInput());
    InputSplit inputSplit = deserializer.deserialize(null);
    deserializer.close();
    return inputSplit;
  }
View Full Code Here
166
167
168
169
170
171
172
173
174
175
176
  public RecordReader getRecordReader(InputSplit split, JobConf job,
      Reporter reporter) throws IOException {

    HiveInputSplit hsplit = (HiveInputSplit)split;

    InputSplit inputSplit = hsplit.getInputSplit();
    String inputFormatClassName = null;
    Class inputFormatClass = null;
    try {
      inputFormatClassName = hsplit.inputFormatClassName();
      inputFormatClass = Class.forName(inputFormatClassName);
View Full Code Here
TOP

Related Classes of org.apache.hadoop.mapred.InputSplit

  • com.ebay.erl.mobius.core.mapred.MobiusInputSampler
  • com.sap.hadoop.windowing.io.TableWindowingInput
  • com.taobao.zeus.jobs.sub.tool.DataPreviewJob
  • org.apache.accumulo.core.client.mapred.AccumuloInputFormatTest
  • org.apache.drill.exec.store.hive.HiveSubScan
  • org.apache.hadoop.contrib.mongoreduce.MongoStreamInputFormat
  • org.apache.hadoop.hbase.mapred.TestTableSnapshotInputFormat
  • org.apache.hadoop.hive.accumulo.mr.TestHiveAccumuloTableInputFormat
  • org.apache.hadoop.hive.ql.exec.StatsNoJobTask
  • org.apache.hadoop.hive.ql.exec.StatsNoJobTask$StatsCollection
    • 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.