Package org.apache.flink.api.common.io.FileInputFormat

Examples of org.apache.flink.api.common.io.FileInputFormat.FileBaseStatistics

90
91
92
93
94
95
96
  public OptimizedPlan compileNoStats(Plan p) {
    return this.noStatsCompiler.compile(p);
  }
 
  public void setSourceStatistics(GenericDataSourceBase<?, ?> source, long size, float recordWidth) {
    setSourceStatistics(source, new FileBaseStatistics(Long.MAX_VALUE, size, recordWidth));
  }
View Full Code Here
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
      DummyFileInputFormat format = new DummyFileInputFormat();
      format.setFilePath(tempFile);
      format.configure(new Configuration());
     
     
      FileBaseStatistics stats = format.getStatistics(null);
      Assert.assertEquals("The file size from the statistics is wrong.", SIZE, stats.getTotalInputSize());
     
      format = new DummyFileInputFormat();
      format.setFilePath(tempFile);
      format.configure(new Configuration());
     
      FileBaseStatistics newStats = format.getStatistics(stats);
      Assert.assertTrue("Statistics object was changed", newStats == stats);

      // insert fake stats with the correct modification time. the call should return the fake stats
      format = new DummyFileInputFormat();
      format.setFilePath(tempFile);
      format.configure(new Configuration());
     
      FileBaseStatistics fakeStats = new FileBaseStatistics(stats.getLastModificationTime(), FAKE_SIZE, BaseStatistics.AVG_RECORD_BYTES_UNKNOWN);
      BaseStatistics latest = format.getStatistics(fakeStats);
      Assert.assertEquals("The file size from the statistics is wrong.", FAKE_SIZE, latest.getTotalInputSize());
     
      // insert fake stats with the expired modification time. the call should return new accurate stats
      format = new DummyFileInputFormat();
      format.setFilePath(tempFile);
      format.configure(new Configuration());
     
      FileBaseStatistics outDatedFakeStats = new FileBaseStatistics(stats.getLastModificationTime()-1, FAKE_SIZE, BaseStatistics.AVG_RECORD_BYTES_UNKNOWN);
      BaseStatistics reGathered = format.getStatistics(outDatedFakeStats);
      Assert.assertEquals("The file size from the statistics is wrong.", SIZE, reGathered.getTotalInputSize());
     
    } catch (Exception ex) {
      ex.printStackTrace();
View Full Code Here
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
     
      DummyFileInputFormat format = new DummyFileInputFormat();
      format.setFilePath(tempDir);
      format.configure(new Configuration());
     
      FileBaseStatistics stats = format.getStatistics(null);
      Assert.assertEquals("The file size from the statistics is wrong.", TOTAL, stats.getTotalInputSize());
     
      format = new DummyFileInputFormat();
      format.setFilePath(tempDir);
      format.configure(new Configuration());
     
      FileBaseStatistics newStats = format.getStatistics(stats);
      Assert.assertTrue("Statistics object was changed", newStats == stats);

      // insert fake stats with the correct modification time. the call should return the fake stats
      format = new DummyFileInputFormat();
      format.setFilePath(tempDir);
      format.configure(new Configuration());
     
      FileBaseStatistics fakeStats = new FileBaseStatistics(stats.getLastModificationTime(), FAKE_SIZE, BaseStatistics.AVG_RECORD_BYTES_UNKNOWN);
      BaseStatistics latest = format.getStatistics(fakeStats);
      Assert.assertEquals("The file size from the statistics is wrong.", FAKE_SIZE, latest.getTotalInputSize());
     
      // insert fake stats with the correct modification time. the call should return the fake stats
      format = new DummyFileInputFormat();
      format.setFilePath(tempDir);
      format.configure(new Configuration());
     
      FileBaseStatistics outDatedFakeStats = new FileBaseStatistics(stats.getLastModificationTime()-1, FAKE_SIZE, BaseStatistics.AVG_RECORD_BYTES_UNKNOWN);
      BaseStatistics reGathered = format.getStatistics(outDatedFakeStats);
      Assert.assertEquals("The file size from the statistics is wrong.", TOTAL, reGathered.getTotalInputSize());
     
    } catch (Exception ex) {
      ex.printStackTrace();
View Full Code Here
112
113
114
115
116
117
118
119
120
121
122
    // only gather base statistics for FileInputFormats
    if(!(mapredInputFormat instanceof FileInputFormat)) {
      return null;
    }
   
    final FileBaseStatistics cachedFileStats = (cachedStats != null && cachedStats instanceof FileBaseStatistics) ?
        (FileBaseStatistics) cachedStats : null;
   
    try {
      final org.apache.hadoop.fs.Path[] paths = FileInputFormat.getInputPaths(this.jobConf);
     
View Full Code Here
244
245
246
247
248
249
250
    // sanity check
    if (len <= 0) {
      len = BaseStatistics.SIZE_UNKNOWN;
    }
   
    return new FileBaseStatistics(latestModTime, len, BaseStatistics.AVG_RECORD_BYTES_UNKNOWN);
  }
View Full Code Here
117
118
119
120
121
122
123
124
125
126
127
      jobContext = HadoopUtils.instantiateJobContext(configuration, null);
    } catch (Exception e) {
      throw new RuntimeException(e);
    }
   
    final FileBaseStatistics cachedFileStats = (cachedStats != null && cachedStats instanceof FileBaseStatistics) ?
        (FileBaseStatistics) cachedStats : null;
       
        try {
          final org.apache.hadoop.fs.Path[] paths = FileInputFormat.getInputPaths(jobContext);
          return getFileStats(cachedFileStats, paths, new ArrayList<FileStatus>(1));
View Full Code Here
282
283
284
285
286
287
288
    // sanity check
    if (len <= 0) {
      len = BaseStatistics.SIZE_UNKNOWN;
    }
   
    return new FileBaseStatistics(latestModTime, len, BaseStatistics.AVG_RECORD_BYTES_UNKNOWN);
  }
View Full Code Here
84
85
86
87
88
89
90
  public OptimizedPlan compileNoStats(Plan p) {
    return this.noStatsCompiler.compile(p);
  }
 
  public void setSourceStatistics(GenericDataSourceBase<?, ?> source, long size, float recordWidth) {
    setSourceStatistics(source, new FileBaseStatistics(Long.MAX_VALUE, size, recordWidth));
  }
View Full Code Here
TOP

Related Classes of org.apache.flink.api.common.io.FileInputFormat.FileBaseStatistics

  • org.apache.flink.api.common.io.FileInputFormatTest
  • org.apache.flink.compiler.CompilerTestBase
  • org.apache.flink.hadoopcompatibility.mapred.HadoopInputFormat
  • org.apache.flink.hadoopcompatibility.mapreduce.HadoopInputFormat
  • org.apache.flink.test.compiler.util.CompilerTestBase
    • 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.