Package co.cask.cdap.api.data.batch

Examples of co.cask.cdap.api.data.batch.RecordScannable

59
60
61
62
63
64
65
66
67
68
69
      throw new IOException(
        String.format("Dataset %s does not implement RecordScannable, and hence cannot be queried in Hive.",
                      conf.get(Constants.Explore.DATASET_NAME)));
    }

    RecordScannable recordScannable = (RecordScannable) dataset;

    if (recordScannable instanceof TransactionAware) {
      startTransaction(conf, (TransactionAware) recordScannable);
    }
View Full Code Here
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
  private static final Gson GSON = new Gson();

  @Override
  public InputSplit[] getSplits(JobConf jobConf, int numSplits) throws IOException {

    RecordScannable recordScannable = DatasetAccessor.getRecordScannable(jobConf);

    try {
      Job job = new Job(jobConf);
      JobContext jobContext = ShimLoader.getHadoopShims().newJobContext(job);
      Path[] tablePaths = FileInputFormat.getInputPaths(jobContext);

      List<Split> dsSplits = recordScannable.getSplits();

      InputSplit[] inputSplits = new InputSplit[dsSplits.size()];
      for (int i = 0; i < dsSplits.size(); i++) {
        inputSplits[i] = new DatasetInputSplit(dsSplits.get(i), tablePaths[0]);
      }
      return inputSplits;
    } finally {
      recordScannable.close();
    }
  }
View Full Code Here
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
103
104
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
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
  @Override
  public RecordReader<Void, ObjectWritable> getRecordReader(final InputSplit split, JobConf jobConf, Reporter reporter)
    throws IOException {

    final RecordScannable recordScannable = DatasetAccessor.getRecordScannable(jobConf);

    if (!(split instanceof DatasetInputSplit)) {
      throw new IOException("Invalid type for InputSplit: " + split.getClass().getName());
    }
    final DatasetInputSplit datasetInputSplit = (DatasetInputSplit) split;

    final RecordScanner recordScanner = recordScannable.createSplitRecordScanner(
        new Split() {
          @Override
          public long getLength() {
            try {
              return split.getLength();
            } catch (IOException e) {
              throw new RuntimeException(e);
            }
          }
        }
    );

    return new RecordReader<Void, ObjectWritable>() {
      private final AtomicBoolean initialized = new AtomicBoolean(false);

      private void initialize() throws IOException {
        try {
          recordScanner.initialize(datasetInputSplit.getDataSetSplit());
        } catch (InterruptedException ie) {
          Thread.currentThread().interrupt();
          throw new IOException("Interrupted while initializing reader", ie);
        }
        initialized.set(true);
      }

      @Override
      public boolean next(Void key, ObjectWritable value) throws IOException {
        if (!initialized.get()) {
          initialize();
        }

        try {
          boolean retVal = recordScanner.nextRecord();
          if (retVal) {
            value.set(recordScanner.getCurrentRecord());
          }
          return retVal;
        } catch (InterruptedException e) {
          Thread.currentThread().interrupt();
          throw new IOException(e);
        }
      }

      @Override
      public Void createKey() {
        return null;
      }

      @Override
      public ObjectWritable createValue() {
        return new ObjectWritable();
      }

      @Override
      public long getPos() throws IOException {
        // Not required.
        return 0;
      }

      @Override
      public void close() throws IOException {
        try {
          recordScanner.close();
        } finally {
          recordScannable.close();
        }
      }

      @Override
      public float getProgress() throws IOException {
View Full Code Here
TOP

Related Classes of co.cask.cdap.api.data.batch.RecordScannable

  • co.cask.cdap.hive.datasets.DatasetAccessor
  • co.cask.cdap.hive.datasets.DatasetInputFormat
    • 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.