Package org.apache.hadoop.mapred

Examples of org.apache.hadoop.mapred.FileSplit

250
251
252
253
254
255
256
257
258
259
260
    AvroJob.setInputSchema(job, readerSchema);

    String dir = System.getProperty("test.dir", ".") + "/mapred";
    Path inputPath = new Path(dir + "/out" + "/myavro1-r-00000.avro");
    FileStatus fileStatus = FileSystem.get(job).getFileStatus(inputPath);
    FileSplit fileSplit = new FileSplit(inputPath, 0, fileStatus.getLen(), job);
    AvroWrapper<Utf8> inputPair = new AvroWrapper<Utf8>(null);
    NullWritable ignore = NullWritable.get();
    AvroRecordReader<Utf8> recordReader = new AvroRecordReader<Utf8>(job, fileSplit);
    long sumOfCounts = 0;
    long numOfCounts = 0;
View Full Code Here
278
279
280
281
282
283
284
285
286
287
288
    AvroJob.setInputSchema(job, readerSchema);

    String dir = System.getProperty("test.dir", ".") + "/mapred";
    Path inputPath = new Path(dir + "/out" + "/testavrofile1-r-00000.avro");
    FileStatus fileStatus = FileSystem.get(job).getFileStatus(inputPath);
    FileSplit fileSplit = new FileSplit(inputPath, 0, fileStatus.getLen(), job);
    AvroWrapper<Utf8> inputPair = new AvroWrapper<Utf8>(null);
    NullWritable ignore = NullWritable.get();
    AvroRecordReader<Utf8> recordReader = new AvroRecordReader<Utf8>(job, fileSplit);
    long sumOfCounts = 0;
    long numOfCounts = 0;
View Full Code Here
330
331
332
333
334
335
336
337
338
339
340
    Schema readerSchema = Schema.create(Schema.Type.STRING);
    AvroJob.setInputSchema(job, readerSchema);
    String dir= System.getProperty("test.dir", ".") + "/mapred";
    Path inputPath = new Path(dir + "/out" + "/myavro2-m-00000.avro");
    FileStatus fileStatus = FileSystem.get(job).getFileStatus(inputPath);
    FileSplit fileSplit = new FileSplit(inputPath, 0, fileStatus.getLen(), job);
    AvroRecordReader<Utf8> recordReader_new = new AvroRecordReader<Utf8>(job, fileSplit);
    AvroWrapper<Utf8> inputPair_new = new AvroWrapper<Utf8>(null);
    NullWritable ignore = NullWritable.get();
    long testl=0;
     while(recordReader_new.next(inputPair_new, ignore)) {
View Full Code Here
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
        for (in = new SequenceFile.Reader(fs, srcs, job); in.next(key, value);) {
          long curr = in.getPosition();
          long delta = curr - prev;
          if (++count > targetcount) {
            count = 0;
            splits.add(new FileSplit(srcs, prev, delta, (String[]) null));
            prev = curr;
          }
        }
      } finally {
        in.close();
      }
      long remaining = fs.getFileStatus(srcs).getLen() - prev;
      if (remaining != 0) {
        splits.add(new FileSplit(srcs, prev, remaining, (String[]) null));
      }
      LOG.info("jobname= " + jobName + " numSplits=" + numSplits +
               ", splits.size()=" + splits.size());
      return splits.toArray(new FileSplit[splits.size()]);
    }
View Full Code Here
147
148
149
150
151
152
153
154
155
156
157
          // we may miss a sync here
          HiveInputSplit newSplit = split;
          if (split.inputFormatClassName().contains("RCFile")
              || split.inputFormatClassName().contains("SequenceFile")) {
            if (split.getStart() > SequenceFile.SYNC_INTERVAL) {
              newSplit = new HiveInputSplit(new FileSplit(split.getPath(),
                  split.getStart() - SequenceFile.SYNC_INTERVAL,
                  split.getLength() + SequenceFile.SYNC_INTERVAL,
                  split.getLocations()),
                  split.inputFormatClassName());
            }
View Full Code Here
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
  public void initIOContext(FileSplit split, JobConf job,
      Class inputFormatClass, RecordReader recordReader) throws IOException {
    boolean blockPointer = false;
    long blockStart = -1;
    FileSplit fileSplit = (FileSplit) split;
    Path path = fileSplit.getPath();
    FileSystem fs = path.getFileSystem(job);
    if (inputFormatClass.getName().contains("SequenceFile")) {
      SequenceFile.Reader in = new SequenceFile.Reader(fs, path, job);
      blockPointer = in.isBlockCompressed();
      in.sync(fileSplit.getStart());
      blockStart = in.getPosition();
      in.close();
    } else if (recordReader instanceof RCFileRecordReader) {
      blockPointer = true;
      blockStart = ((RCFileRecordReader) recordReader).getStart();
    } else if (inputFormatClass.getName().contains("RCFile")) {
      blockPointer = true;
      RCFile.Reader in = new RCFile.Reader(fs, path, job);
      in.sync(fileSplit.getStart());
      blockStart = in.getPosition();
      in.close();
    }
    this.initIOContext(blockStart, blockPointer, split.getPath().toString());
  }
View Full Code Here
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
      reporter.setStatus(split.toString());
     
      // find part name
      SegmentPart segmentPart;
      final String spString;
      final FileSplit fSplit = (FileSplit) split;
      try {
        segmentPart = SegmentPart.get(fSplit);
        spString = segmentPart.toString();
      } catch (IOException e) {
        throw new RuntimeException("Cannot identify segment:", e);
      }
     
      SequenceFile.Reader reader =
        new SequenceFile.Reader(FileSystem.get(job), fSplit.getPath(), job);
     
      final Writable w;
      try {
        w = (Writable) reader.getValueClass().newInstance();
      } catch (Exception e) {
View Full Code Here
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
        for (; sl.next(key, value); last = sl.getPosition()) {
          // if adding this split would put this split past the target size,
          // cut the last split and put this next file in the next split.
          if (acc + key.get() > targetsize && acc != 0) {
            long splitsize = last - pos;
            splits.add(new FileSplit(src, pos, splitsize, (String[])null));
            cbrem -= splitsize;
            pos = last;
            acc = 0L;
          }
          acc += key.get();
        }
      }
      finally {
        checkAndClose(sl);
      }
      if (cbrem != 0) {
        splits.add(new FileSplit(src, pos, cbrem, (String[])null));
      }

      return splits.toArray(new FileSplit[splits.size()]);
    }
View Full Code Here
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
  public Row first() throws HiveException {
    Row r = super.first();

    if ( blockInfos.size() > 0 ) {
      InputSplit[] inputSplits = getInputSplits();
      FileSplit fS = null;
      BlockInfo bI = blockInfos.get(0);
      bI.startingSplit = 0;
      int i = 1;
      bI = i < blockInfos.size() ? blockInfos.get(i) : null;
      for(int j=1; j < inputSplits.length && bI != null; j++) {
        fS = (FileSplit) inputSplits[j];
        while (bI != null && bI.startOffset < fS.getStart() ) {
          bI.startingSplit = j - 1;
          i++;
          bI = i < blockInfos.size() ? blockInfos.get(i) : null;
        }
      }
View Full Code Here
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
        Preconditions.checkState(false);
      } else if (event instanceof RootInputDataInformationEvent) {
        dataInformationEventSeen = true;
        RootInputDataInformationEvent diEvent = (RootInputDataInformationEvent) event;
        dataInformationEvents.add(diEvent);
        FileSplit fileSplit;
        try {
          fileSplit = getFileSplitFromEvent(diEvent);
        } catch (IOException e) {
          throw new RuntimeException("Failed to get file split for event: " + diEvent);
        }
        List<FileSplit> fsList = pathFileSplitsMap.get(fileSplit.getPath());
        if (fsList == null) {
          fsList = new ArrayList<FileSplit>();
          pathFileSplitsMap.put(fileSplit.getPath(), fsList);
        }
        fsList.add(fileSplit);
      }
    }
View Full Code Here
TOP

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

  • ar.ext.spark.hbin.HBINInputFormat
  • cascading.tap.hadoop.io.CombineFileRecordReaderWrapper
  • com.alexholmes.hadooputils.combine.common.mapred.CommonCombineRecordReader
  • com.cloudera.iterativereduce.io.TextRecordParser
  • com.ebay.erl.mobius.core.mapred.FileInputFormatHelper
  • com.esri.json.hadoop.TestUnenclosedJsonRecordReader
  • com.facebook.hive.orc.OrcInputFormat
  • com.facebook.presto.hive.AbstractTestHiveFileFormats
  • com.facebook.presto.hive.BenchmarkHiveFileFormats$BenchmarkFile
  • com.facebook.presto.hive.HiveRecordSet
    • 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.