Package org.apache.hadoop.mapred

Examples of org.apache.hadoop.mapred.FileSplit

59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
        RCFileInputFormat inputFormat = new RCFileInputFormat();
        @SuppressWarnings("deprecation")
        SerDe serde = new ColumnarSerDe();
        File file = File.createTempFile("presto_test", "rc-text");
        try {
            FileSplit split = createTestFile(file.getAbsolutePath(), outputFormat, serde, null);
            @SuppressWarnings("unchecked")
            RecordReader<?, BytesRefArrayWritable> recordReader = (RecordReader<?, BytesRefArrayWritable>) inputFormat.getRecordReader(split, jobConf, Reporter.NULL);
            Properties splitProperties = new Properties();
            splitProperties.setProperty("serialization.lib", "org.apache.hadoop.hive.serde2.columnar.ColumnarSerDe");
            splitProperties.setProperty("columns", COLUMN_NAMES_STRING);
            splitProperties.setProperty("columns.types", COLUMN_TYPES);
            RecordCursor cursor = new ColumnarTextHiveRecordCursor<>(recordReader,
                    split.getLength(),
                    splitProperties,
                    new ArrayList<HivePartitionKey>(),
                    getColumns(),
                    DateTimeZone.getDefault(),
                    DateTimeZone.getDefault(),
View Full Code Here
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
        RCFileInputFormat inputFormat = new RCFileInputFormat();
        @SuppressWarnings("deprecation")
        SerDe serde = new LazyBinaryColumnarSerDe();
        File file = File.createTempFile("presto_test", "rc-binary");
        try {
            FileSplit split = createTestFile(file.getAbsolutePath(), outputFormat, serde, null);
            @SuppressWarnings("unchecked")
            RecordReader<?, BytesRefArrayWritable> recordReader = (RecordReader<?, BytesRefArrayWritable>) inputFormat.getRecordReader(split, jobConf, Reporter.NULL);
            Properties splitProperties = new Properties();
            splitProperties.setProperty("serialization.lib", "org.apache.hadoop.hive.serde2.columnar.LazyBinaryColumnarSerDe");
            splitProperties.setProperty("columns", COLUMN_NAMES_STRING);
            splitProperties.setProperty("columns.types", COLUMN_TYPES);
            RecordCursor cursor = new ColumnarBinaryHiveRecordCursor<>(recordReader,
                    split.getLength(),
                    splitProperties,
                    new ArrayList<HivePartitionKey>(),
                    getColumns(),
                    DateTimeZone.getDefault(),
                    new TypeRegistry());
View Full Code Here
164
165
166
167
168
169
170
171
172
  }


  private void checkSplit(InputSplit split, long start, long length) {
    assertThat(split, instanceOf(FileSplit.class));
    FileSplit fileSplit = (FileSplit) split;
    assertThat(fileSplit.getStart(), is(start));
    assertThat(fileSplit.getLength(), is(length));
  }
View Full Code Here
114
115
116
117
118
119
120
121
122
123
124
    AvroJob.setInputSchema(job, readerSchema);
   
    String dir = System.getProperty("test.dir", ".") + "/mapred";
    Path inputPath = new Path(dir + "/out" + "/part-00000" + AvroOutputFormat.EXT);
    FileStatus fileStatus = FileSystem.get(job).getFileStatus(inputPath);
    FileSplit fileSplit = new FileSplit(inputPath, 0, fileStatus.getLen(), job);
   
    AvroRecordReader<Pair<Integer, Long>> recordReader = new AvroRecordReader<Pair<Integer, Long>>(job, fileSplit);
   
    AvroWrapper<Pair<Integer, Long>> inputPair = new AvroWrapper<Pair<Integer, Long>>(null);
    NullWritable ignore = NullWritable.get();
View Full Code Here
182
183
184
185
186
187
188
189
190
191
192
193
194
195
      return sideFs_.create(new Path(uri.getSchemeSpecificPart()));
    }
  }

  String getSideEffectFileName() {
    FileSplit split = StreamUtil.getCurrentSplit(job_);
    String leaf = split.getPath().getName();
    if (split.getStart() == 0) {
      return leaf;
    } else {
      return new FileSplit(new Path(leaf), split.getStart(),
                           split.getLength(), job_).toString();
    }
  }
View Full Code Here
416
417
418
419
420
421
422
      return null;
    }
    Path p = new Path(path);
    long start = Long.parseLong(job.get("map.input.start"));
    long length = Long.parseLong(job.get("map.input.length"));
    return new FileSplit(p, start, length, job);
  }
View Full Code Here
107
108
109
110
111
112
113
114
115
116
117
    if (length < 0 || (start + length) > fileLen) {
      length = fileLen - start;
    }

    //share the code with RecordReader.
    FileSplit split = new FileSplit(fileName,start, length, new JobConf(conf));
    RCFileRecordReader recordReader = new RCFileRecordReader(conf, split);
    LongWritable key = new LongWritable();
    BytesRefArrayWritable value = new BytesRefArrayWritable();
    StringBuilder buf = new StringBuilder(STRING_BUFFER_SIZE); // extra capacity in case we overrun, to avoid resizing
    while (recordReader.next(key, value)) {
View Full Code Here
56
57
58
59
60
61
62
63
64
65
66
    }
    InputFormat inputFormat = HiveInputFormat.getInputFormatFromCache(
        inputFormatClass, jobConf);

    // create a split for the given partition
    FileSplit fsplit = new FileSplit(hsplit.getPaths()[partition], hsplit
        .getStartOffsets()[partition], hsplit.getLengths()[partition], hsplit
        .getLocations());

    this.setRecordReader(inputFormat.getRecordReader(fsplit, jobConf, reporter));
View Full Code Here
68
69
70
71
72
73
74
  public static class SymlinkTextInputSplit extends FileSplit {
    private final FileSplit split;

    public SymlinkTextInputSplit() {
      super((Path)null, 0, 0, (String[])null);
      split = new FileSplit((Path)null, 0, 0, (String[])null);
    }
View Full Code Here
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
  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
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.