Examples of writeTextFile()

Examples of com.cloudera.crunch.impl.mr.MRPipeline.writeTextFile()

33
34
35
36
37
38
39
40
41
42
43
    PCollection<String> resources = extractFilterResources(logs);

    PTable<String, Long> counts = Aggregate.count(resources);

    // Instruct the pipeline to write the resulting counts to a text file.
    pipeline.writeTextFile(counts, args[1]);
    // Execute the pipeline as a MapReduce.
    pipeline.done();
  }

  public static PCollection<String> extractFilterResources(PCollection<CommonLogEntry> logs) {
View Full Code Here

Examples of org.apache.crunch.Pipeline.writeTextFile()

76
77
78
79
80
81
82
83
84
85
86
    // Calculate average response size by ip address
    PTable<String, Double> avgs = remoteAddrResponseSize.parallelDo(calulateAverage,
        Writables.tableOf(Writables.strings(), Writables.doubles()));

    // write the result to a text file
    pipeline.writeTextFile(avgs, args[1]);
    // Execute the pipeline as a MapReduce.
    PipelineResult result = pipeline.done();

    return result.succeeded() ? 0 : 1;
  }
View Full Code Here

Examples of org.apache.crunch.Pipeline.writeTextFile()

74
75
76
77
78
79
80
81
82
83
84
    // Calculate average response size by ip address
    PTable<String, Double> avgs = remoteAddrResponseSize.parallelDo(calulateAverage,
        Writables.tableOf(Writables.strings(), Writables.doubles()));

    // write the result to a text file
    pipeline.writeTextFile(avgs, args[1]);
    // Execute the pipeline as a MapReduce.
    PipelineResult result = pipeline.done();

    return result.succeeded() ? 0 : 1;
  }
View Full Code Here

Examples of org.apache.crunch.Pipeline.writeTextFile()

66
67
68
69
70
71
72
73
74
75
76
    // Table of (ip, sum(response size))
    PTable<String, Long> ipAddrResponseSize = lines
        .parallelDo(extractIPResponseSize, Writables.tableOf(Writables.strings(), Writables.longs())).groupByKey()
        .combineValues(longSumCombiner);

    pipeline.writeTextFile(ipAddrResponseSize, args[1]);
    // Execute the pipeline as a MapReduce.
    PipelineResult result = pipeline.done();

    return result.succeeded() ? 0 : 1;
  }
View Full Code Here

Examples of org.apache.crunch.Pipeline.writeTextFile()

98
99
100
101
102
103
104
105
106
107
108
  @Test
  public void testMemPipelineFileWriter() throws Exception {
    File outputDir = baseTmpDir.getFile("mempipe");
    Pipeline p = MemPipeline.getInstance();
    PCollection<String> lines = MemPipeline.collectionOf("hello", "world");
    p.writeTextFile(lines, outputDir.toString());
    p.done();
    File outputFile = getOutputFile(outputDir, "*.txt");

    List<String> txt = Files.readLines(outputFile, Charsets.UTF_8);
    assertEquals(ImmutableList.of("hello", "world"), txt);
View Full Code Here

Examples of org.apache.crunch.Pipeline.writeTextFile()

136
137
138
139
140
141
142
143
144
              }
            }
          }
        });

    pipeline.writeTextFile(result, tmpDir.getFileName("unused"));
    Assert.assertTrue("Should succeed", pipeline.done().succeeded());
  }
}
View Full Code Here

Examples of org.apache.crunch.Pipeline.writeTextFile()

77
78
79
80
81
82
83
84
85
86
87
  @Test
  public void testMemPipelineFileWriter() throws Exception {
    File tmpDir = baseTmpDir.getFile("mempipe");
    Pipeline p = MemPipeline.getInstance();
    PCollection<String> lines = MemPipeline.collectionOf("hello", "world");
    p.writeTextFile(lines, tmpDir.toString());
    p.done();
    assertTrue(tmpDir.exists());
    File[] files = tmpDir.listFiles();
    assertTrue(files != null && files.length > 0);
    for (File f : files) {
View Full Code Here

Examples of org.apache.crunch.Pipeline.writeTextFile()

73
74
75
76
77
78
79
80
81
82
83
    // Calculate average response size by ip address
    PTable<String, Double> avgs = remoteAddrResponseSize.parallelDo(calulateAverage,
        Writables.tableOf(Writables.strings(), Writables.doubles()));

    // write the result to a text file
    pipeline.writeTextFile(avgs, args[1]);
    // Execute the pipeline as a MapReduce.
    pipeline.done();
    return 0;
  }
View Full Code Here

Examples of org.apache.crunch.Pipeline.writeTextFile()

65
66
67
68
69
70
71
72
73
74
75
    // Table of (ip, sum(response size))
    PTable<String, Long> ipAddrResponseSize = lines
        .parallelDo(extractIPResponseSize, Writables.tableOf(Writables.strings(), Writables.longs())).groupByKey()
        .combineValues(longSumCombiner);

    pipeline.writeTextFile(ipAddrResponseSize, args[1]);
    // Execute the pipeline as a MapReduce.
    pipeline.done();
    return 0;
  }
View Full Code Here

Examples of org.apache.crunch.Pipeline.writeTextFile()

62
63
64
65
66
67
68
69
70
71
72
    // Best of all, the count() function doesn't need to know anything about
    // the kind of data stored in the input PCollection.
    PTable<String, Long> counts = words.count();

    // Instruct the pipeline to write the resulting counts to a text file.
    pipeline.writeTextFile(counts, args[2]);
    // Execute the pipeline as a MapReduce.
    pipeline.done();
    return 0;
  }
View Full Code Here
TOP
Copyright © 2018 www.massapi.com. 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.