Package org.apache.crunch.types

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

158
159
160
161
162
163
164
165
166
167
168
        new MapFn<Pair<T, N>, Pair<Integer, Pair<T, N>>>() {
          @Override
          public Pair<Integer, Pair<T, N>> map(Pair<T, N> p) {
            return Pair.of(0, p);
          }
        }, ptf.tableOf(ptf.ints(), input.getPType()));
    int[] ss = { sampleSize };
    return groupedWeightedReservoirSample(groupedIn, ss, seed)
        .parallelDo("Extract sampled value from pair", new MapFn<Pair<Integer, T>, T>() {
          @Override
          public T map(Pair<Integer, T> p) {
View Full Code Here
199
200
201
202
203
204
205
206
207
208
209
      PTable<Integer, Pair<T, N>> input,
      int[] sampleSizes,
      Long seed) {
    PTypeFamily ptf = input.getTypeFamily();
    PType<T> ttype = (PType<T>) input.getPTableType().getValueType().getSubTypes().get(0);
    PTableType<Integer, Pair<Double, T>> ptt = ptf.tableOf(ptf.ints(),
        ptf.pairs(ptf.doubles(), ttype));
   
    return input.parallelDo("Initial reservoir sampling", new ReservoirSampleFn<T, N>(sampleSizes, seed, ttype), ptt)
        .groupByKey(1)
        .combineValues(new WRSCombineFn<T>(sampleSizes, ttype))
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()
        .requireSortedKeys()
        .groupingComparatorClass(JoinUtils.getGroupingComparator(ptf))
        .partitionerClass(JoinUtils.getPartitionerClass(ptf));
View Full Code Here
69
70
71
72
73
74
75
76
77
78
79
   * @return joined tables
   */
  public PTable<K, Pair<U, V>> join(PTable<K, U> left, PTable<K, V> right, JoinFn<K, U, V> joinFn) {
    PTypeFamily ptf = left.getTypeFamily();
    PGroupedTable<Pair<K, Integer>, Pair<U, V>> grouped = preJoin(left, right, numReducers);
    PTableType<K, Pair<U, V>> ret = ptf
        .tableOf(left.getKeyType(), ptf.pairs(left.getValueType(), right.getValueType()));

    return grouped.parallelDo(joinFn.getJoinType() + grouped.getName(), joinFn, ret);
  }

View Full Code Here
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
115
116
117
118
119
120
121
122
123
124
125
   *
   * @return a {@code PCollection} representing the sorted collection.
   */
  public static <T> PCollection<T> sort(PCollection<T> collection, int numReducers, Order order) {
    PTypeFamily tf = collection.getTypeFamily();
    PTableType<T, Void> type = tf.tableOf(collection.getPType(), tf.nulls());
    Configuration conf = collection.getPipeline().getConfiguration();
    PTable<T, Void> pt = collection.parallelDo("sort-pre", new DoFn<T, Pair<T, Void>>() {
      @Override
      public void process(T input, Emitter<Pair<T, Void>> emitter) {
        emitter.emit(Pair.of(input, (Void) null));
View Full Code Here
158
159
160
161
162
163
164
165
166
167
168
        new MapFn<Pair<T, N>, Pair<Integer, Pair<T, N>>>() {
          @Override
          public Pair<Integer, Pair<T, N>> map(Pair<T, N> p) {
            return Pair.of(0, p);
          }
        }, ptf.tableOf(ptf.ints(), input.getPType()));
    int[] ss = { sampleSize };
    return groupedWeightedReservoirSample(groupedIn, ss, seed)
        .parallelDo(new MapFn<Pair<Integer, T>, T>() {
          @Override
          public T map(Pair<Integer, T> p) {
View Full Code Here
199
200
201
202
203
204
205
206
207
208
209
      PTable<Integer, Pair<T, N>> input,
      int[] sampleSizes,
      Long seed) {
    PTypeFamily ptf = input.getTypeFamily();
    PType<T> ttype = (PType<T>) input.getPTableType().getValueType().getSubTypes().get(0);
    PTableType<Integer, Pair<Double, T>> ptt = ptf.tableOf(ptf.ints(),
        ptf.pairs(ptf.doubles(), ttype));
   
    return input.parallelDo(new ReservoirSampleFn<T, N>(sampleSizes, seed, ttype), ptt)
        .groupByKey(1)
        .combineValues(new WRSCombineFn<T>(sampleSizes, ttype))
View Full Code Here
69
70
71
72
73
74
75
76
77
78
79
   * @return joined tables
   */
  public PTable<K, Pair<U, V>> join(PTable<K, U> left, PTable<K, V> right, JoinFn<K, U, V> joinFn) {
    PTypeFamily ptf = left.getTypeFamily();
    PGroupedTable<Pair<K, Integer>, Pair<U, V>> grouped = preJoin(left, right, numReducers);
    PTableType<K, Pair<U, V>> ret = ptf
        .tableOf(left.getKeyType(), ptf.pairs(left.getValueType(), right.getValueType()));

    return grouped.parallelDo(joinFn.getJoinType() + grouped.getName(), joinFn, ret);
  }

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.