Package org.apache.crunch.impl.mr

Examples of org.apache.crunch.impl.mr.MRPipeline.readTextFile()

56
57
58
59
60
61
62
63
64
65
66
  @Test
  public void testWritables() throws Exception {
    Pipeline pipeline = new MRPipeline(AggregateIT.class, tmpDir.getDefaultConfiguration());
    String shakesInputPath = tmpDir.copyResourceFileName("shakes.txt");
    PCollection<String> shakes = pipeline.readTextFile(shakesInputPath);
    runMinMax(shakes, WritableTypeFamily.getInstance());
    pipeline.done();
  }

  @Test
View Full Code Here
65
66
67
68
69
70
71
72
73
74
75
  @Test
  public void testAvro() throws Exception {
    Pipeline pipeline = new MRPipeline(AggregateIT.class, tmpDir.getDefaultConfiguration());
    String shakesInputPath = tmpDir.copyResourceFileName("shakes.txt");
    PCollection<String> shakes = pipeline.readTextFile(shakesInputPath);
    runMinMax(shakes, AvroTypeFamily.getInstance());
    pipeline.done();
  }

  @Test
View Full Code Here
108
109
110
111
112
113
114
115
116
117
118
  @Test
  public void testCollectUrls() throws Exception {
    Pipeline p = new MRPipeline(AggregateIT.class, tmpDir.getDefaultConfiguration());
    String urlsInputPath = tmpDir.copyResourceFileName("urls.txt");
    PTable<String, Collection<String>> urls = Aggregate.collectValues(p.readTextFile(urlsInputPath).parallelDo(
        new SplitFn(), tableOf(strings(), strings())));
    for (Pair<String, Collection<String>> e : urls.materialize()) {
      String key = e.first();
      int expectedSize = 0;
      if ("www.A.com".equals(key)) {
View Full Code Here
140
141
142
143
144
145
146
147
148
149
150
  }

  @Test
  public void testCollectValues_Writables() throws IOException {
    Pipeline pipeline = new MRPipeline(AggregateIT.class, tmpDir.getDefaultConfiguration());
    Map<Integer, Collection<Text>> collectionMap = pipeline.readTextFile(tmpDir.copyResourceFileName("set2.txt"))
        .parallelDo(new MapStringToTextPair(), Writables.tableOf(Writables.ints(), Writables.writables(Text.class)))
        .collectValues().materializeToMap();

    assertEquals(1, collectionMap.size());
View Full Code Here
154
155
156
157
158
159
160
161
162
163
164
  @Test
  public void testCollectValues_Avro() throws IOException {

    MapStringToEmployeePair mapFn = new MapStringToEmployeePair();
    Pipeline pipeline = new MRPipeline(AggregateIT.class, tmpDir.getDefaultConfiguration());
    Map<Integer, Collection<Employee>> collectionMap = pipeline.readTextFile(tmpDir.copyResourceFileName("set2.txt"))
        .parallelDo(mapFn, Avros.tableOf(Avros.ints(), Avros.records(Employee.class))).collectValues()
        .materializeToMap();

    assertEquals(1, collectionMap.size());
View Full Code Here
70
71
72
73
74
75
76
77
78
79
80
  @Test
  public void testMRMaterializeToMap() throws IOException {
    Pipeline p = new MRPipeline(MaterializeToMapIT.class, tmpDir.getDefaultConfiguration());
    String inputFile = tmpDir.copyResourceFileName("set1.txt");
    PCollection<String> c = p.readTextFile(inputFile);
    PTypeFamily tf = c.getTypeFamily();
    PTable<Integer, String> t = c.parallelDo(new Set1Mapper(), tf.tableOf(tf.ints(), tf.strings()));
    Map<Integer, String> m = t.materializeToMap();
    assertMatches(m);
  }
View Full Code Here
46
47
48
49
50
51
52
53
54
55
56
  }

  public static void run(PTypeFamily typeFamily, TemporaryPath tmpDir) throws Exception {
    Pipeline pipeline = new MRPipeline(MapsIT.class, tmpDir.getDefaultConfiguration());
    String shakesInputPath = tmpDir.copyResourceFileName("shakes.txt");
    PCollection<String> shakespeare = pipeline.readTextFile(shakesInputPath);
    Iterable<Pair<String, Map<String, Long>>> output = shakespeare
        .parallelDo(new DoFn<String, Pair<String, Map<String, Long>>>() {
          @Override
          public void process(String input, Emitter<Pair<String, Map<String, Long>>> emitter) {
            String last = null;
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.