Package org.apache.crunch.types

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

78
79
80
81
82
83
84
85
86
87
88
  }

  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
View Full Code Here
95
96
97
98
99
100
101
102
103
104
105
 
  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()
View Full Code Here
98
99
100
101
102
103
104
105
106
107
108
    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))
        .partitionerClass(JoinUtils.getPartitionerClass(ptf));
    if (numReducers > 0) {
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
63
64
65
66
67
68
69
70
71
72
73
    PTypeFamily tf = collect.getTypeFamily();
    return collect.parallelDo("Aggregate.count", new MapFn<S, Pair<S, Long>>() {
      public Pair<S, Long> map(S input) {
        return Pair.of(input, 1L);
      }
    }, tf.tableOf(collect.getPType(), tf.longs()))
        .groupByKey(numPartitions)
        .combineValues(Aggregators.SUM_LONGS());
  }
 
  /**
 
View Full Code Here
82
83
84
85
86
87
88
89
90
91
92
    PTable<Integer, Long> countTable = collect
        .parallelDo("Aggregate.count", new MapFn<S, Pair<Integer, Long>>() {
          public Pair<Integer, Long> map(S input) {
            return Pair.of(1, 1L);
          }
        }, tf.tableOf(tf.ints(), tf.longs()))
        .groupByKey(GroupingOptions.builder().numReducers(1).build())
        .combineValues(Aggregators.SUM_LONGS());
    PCollection<Long> count = countTable.values();
    return new FirstElementPObject<Long>(count);
  }
View Full Code Here
166
167
168
169
170
171
172
173
174
175
176
  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) {
            emitter.emit(input.second());
View Full Code Here
201
202
203
204
205
206
207
208
209
210
211
          public void cleanup(Emitter<Pair<Boolean, S>> emitter) {
            if (max != null) {
              emitter.emit(Pair.of(true, max));
            }
          }
        }, tf.tableOf(tf.booleans(), collect.getPType())).groupByKey(1)
        .combineValues(new CombineFn<Boolean, S>() {
          public void process(Pair<Boolean, Iterable<S>> input, Emitter<Pair<Boolean, S>> emitter) {
            S max = null;
            for (S v : input.second()) {
              if (max == null || ((Comparable<S>) max).compareTo(v) < 0) {
View Full Code Here
241
242
243
244
245
246
247
248
249
250
251
          public void cleanup(Emitter<Pair<Boolean, S>> emitter) {
            if (min != null) {
              emitter.emit(Pair.of(false, min));
            }
          }
        }, tf.tableOf(tf.booleans(), collect.getPType())).groupByKey(1)
        .combineValues(new CombineFn<Boolean, S>() {
          public void process(Pair<Boolean, Iterable<S>> input, Emitter<Pair<Boolean, S>> emitter) {
            S min = null;
            for (S v : input.second()) {
              if (min == null || ((Comparable<S>) min).compareTo(v) > 0) {
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
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.