Package org.apache.crunch.types

Examples of org.apache.crunch.types.PTypeFamily.pairs()

79
80
81
82
83
84
85
86
87
88
89
  static <K, U, V> PGroupedTable<Pair<K, Integer>, Pair<U, V>> preJoin(PTable<K, U> left, PTable<K, V> right,
                                                                       int numReducers) {
    PTypeFamily ptf = left.getTypeFamily();
    PTableType<Pair<K, Integer>, Pair<U, V>> ptt = ptf.tableOf(ptf.pairs(left.getKeyType(), ptf.ints()),
        ptf.pairs(left.getValueType(), right.getValueType()));

    PTable<Pair<K, Integer>, Pair<U, V>> tag1 = left.parallelDo("joinTagLeft",
        new MapFn<Pair<K, U>, Pair<Pair<K, Integer>, Pair<U, V>>>() {
          @Override
          public Pair<Pair<K, Integer>, Pair<U, V>> map(Pair<K, U> input) {
View Full Code Here
96
97
98
99
100
101
102
103
104
105
106
  private static <K, V1, V2> PGroupedTable<Pair<K, V1>, Pair<V1, V2>> prepare(
      PTable<K, Pair<V1, V2>> input, int numReducers) {
    PTypeFamily ptf = input.getTypeFamily();
    PType<Pair<V1, V2>> valueType = input.getValueType();
    PTableType<Pair<K, V1>, Pair<V1, V2>> inter = ptf.tableOf(
        ptf.pairs(input.getKeyType(), valueType.getSubTypes().get(0)),
        valueType);
    PTableType<K, Collection<Pair<V1, V2>>> out = ptf.tableOf(input.getKeyType(),
        ptf.collections(input.getValueType()));
    GroupingOptions.Builder gob = GroupingOptions.builder()
        .groupingComparatorClass(JoinUtils.getGroupingComparator(ptf))
View Full Code Here
87
88
89
90
91
92
93
94
95
96
97
    MapsideJoinDoFn<K, U, V> mapJoinDoFn = new MapsideJoinDoFn<K, U, V>(rightReadable, includeUnmatchedLeftValues);
    ParallelDoOptions options = ParallelDoOptions.builder()
        .sourceTargets(rightReadable.getSourceTargets())
        .build();
    return left.parallelDo("mapjoin", mapJoinDoFn,
        tf.tableOf(left.getKeyType(), tf.pairs(left.getValueType(), right.getValueType())),
        options);
  }

  static class MapsideJoinDoFn<K, U, V> extends DoFn<Pair<K, U>, Pair<K, Pair<U, V>>> {
View Full Code Here
165
166
167
168
169
170
171
172
173
174
175
  }

  public static <K, V> PTable<K, V> top(PTable<K, V> ptable, int limit, boolean maximize) {
    PTypeFamily ptf = ptable.getTypeFamily();
    PTableType<K, V> base = ptable.getPTableType();
    PType<Pair<K, V>> pairType = ptf.pairs(base.getKeyType(), base.getValueType());
    PTableType<Integer, Pair<K, V>> inter = ptf.tableOf(ptf.ints(), pairType);
    return ptable.parallelDo("top" + limit + "map", new TopKFn<K, V>(limit, maximize), inter)
        .groupByKey(1).combineValues(new TopKCombineFn<K, V>(limit, maximize))
        .parallelDo("top" + limit + "reduce", new DoFn<Pair<Integer, Pair<K, V>>, Pair<K, V>>() {
          public void process(Pair<Integer, Pair<K, V>> input, Emitter<Pair<K, V>> emitter) {
View Full Code Here
90
91
92
93
94
95
96
97
98
99
100
                Pair<String, String> pair = Pair.of(word.toLowerCase(Locale.ENGLISH), title);
                emitter.emit(pair);
              }
            }
          }
        }, ptf.pairs(ptf.strings(), ptf.strings())));

    if (transformTF) {
      /*
       * Input: Pair<Pair<String, String>, Long> Pair<Pair<word, title>, count
       * in title>
View Full Code Here
107
108
109
110
111
112
113
114
115
116
117
            @Override
            public Pair<String, Pair<String, Long>> map(Pair<Pair<String, String>, Long> input) {
              Pair<String, String> wordDocumentPair = input.first();
              return Pair.of(wordDocumentPair.first(), Pair.of(wordDocumentPair.second(), input.second()));
            }
          }, ptf.tableOf(ptf.strings(), ptf.pairs(ptf.strings(), ptf.longs())));

      pipeline.writeTextFile(wordDocumentCountPair, transformedOutput.getAbsolutePath());
    }

    SourceTarget<String> st = At.textFile(tfOutput.getAbsolutePath());
View Full Code Here
59
60
61
62
63
64
65
66
67
68
69
      int numReducers,
      PTable<K, U> left,
      PTable<K, V> right) {
    PTypeFamily tf = left.getTypeFamily();
    return cogroup(
        tf.pairs(tf.collections(left.getValueType()),
                 tf.collections(right.getValueType())),
        TupleFactory.PAIR,
        numReducers,
        left, right);
  }
View Full Code Here
66
67
68
69
70
71
72
73
74
75
76
  private static <K, V1, V2> PGroupedTable<Pair<K, V1>, Pair<V1, V2>> prepare(
      PTable<K, Pair<V1, V2>> input) {
    PTypeFamily ptf = input.getTypeFamily();
    PType<Pair<V1, V2>> valueType = input.getValueType();
    PTableType<Pair<K, V1>, Pair<V1, V2>> inter = ptf.tableOf(
        ptf.pairs(input.getKeyType(), valueType.getSubTypes().get(0)),
        valueType);
    PTableType<K, Collection<Pair<V1, V2>>> out = ptf.tableOf(input.getKeyType(),
        ptf.collections(input.getValueType()));
    return input.parallelDo("SecondarySort.format", new SSFormatFn<K, V1, V2>(), inter)
        .groupByKey(
View Full Code Here
153
154
155
156
157
158
159
160
161
162
163
  }

  public static <K, V> PTable<K, V> top(PTable<K, V> ptable, int limit, boolean maximize) {
    PTypeFamily ptf = ptable.getTypeFamily();
    PTableType<K, V> base = ptable.getPTableType();
    PType<Pair<K, V>> pairType = ptf.pairs(base.getKeyType(), base.getValueType());
    PTableType<Integer, Pair<K, V>> inter = ptf.tableOf(ptf.ints(), pairType);
    return ptable.parallelDo("top" + limit + "map", new TopKFn<K, V>(limit, maximize), inter)
        .groupByKey(1).combineValues(new TopKCombineFn<K, V>(limit, maximize))
        .parallelDo("top" + limit + "reduce", new DoFn<Pair<Integer, Pair<K, V>>, Pair<K, V>>() {
          public void process(Pair<Integer, Pair<K, V>> input, Emitter<Pair<K, V>> emitter) {
View Full Code Here
76
77
78
79
80
81
82
83
84
85
86
      ParallelDoOptions.Builder optionsBuilder = ParallelDoOptions.builder();
      if (mi.isSourceTarget()) {
        optionsBuilder.sourceTargets((SourceTarget) mi.getSource());
      }
      return left.parallelDo("mapjoin", mapJoinDoFn,
          tf.tableOf(left.getKeyType(), tf.pairs(left.getValueType(), right.getValueType())),
          optionsBuilder.build());
    } else { // in-memory pipeline
      return left.parallelDo(new InMemoryJoinFn<K, U, V>(iterable),
          tf.tableOf(left.getKeyType(), tf.pairs(left.getValueType(), right.getValueType())));
    }
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.