compiler/frontend/spark/src/main/java/org/apache/nemo/compiler/frontend/spark/transform/ReduceByKeyTransform.java

/*
 * Licensed to the Apache Software Foundation (ASF) under one
 * or more contributor license agreements.  See the NOTICE file
 * distributed with this work for additional information
 * regarding copyright ownership.  The ASF licenses this file
 * to you under the Apache License, Version 2.0 (the
 * "License"); you may not use this file except in compliance
 * with the License.  You may obtain a copy of the License at
 *
 *   http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing,
 * software distributed under the License is distributed on an
 * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
 * KIND, either express or implied.  See the License for the
 * specific language governing permissions and limitations
 * under the License.
 */
package org.apache.nemo.compiler.frontend.spark.transform;

import org.apache.nemo.common.ir.OutputCollector;
import org.apache.nemo.common.ir.vertex.transform.NoWatermarkEmitTransform;
import org.apache.spark.api.java.function.Function2;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import scala.Tuple2;

import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;

/**
 * Transform for reduce by key transformation.
 *
 * @param <K> key type.
 * @param <V> value type.
 */
public final class ReduceByKeyTransform<K, V> extends NoWatermarkEmitTransform<Tuple2<K, V>, Tuple2<K, V>> {
  private static final Logger LOG = LoggerFactory.getLogger(ReduceByKeyTransform.class.getName());

  private final Map<K, List<V>> keyToValues;
  private final Function2<V, V, V> func;
  private OutputCollector<Tuple2<K, V>> outputCollector;

  /**
   * Constructor.
   *
   * @param func reduce function.
   */
  public ReduceByKeyTransform(final Function2<V, V, V> func) {
    this.func = func;
    this.keyToValues = new HashMap<>();
  }

  @Override
  public void prepare(final Context context, final OutputCollector<Tuple2<K, V>> oc) {
    this.outputCollector = oc;
  }

  @Override
  public void onData(final Tuple2<K, V> element) {
    final K key = element._1;
    final V value = element._2;

    keyToValues.putIfAbsent(key, new ArrayList<>());
    keyToValues.get(key).add(value);
  }

  @Override
  public void close() {
    if (keyToValues.isEmpty()) {
      LOG.warn("Spark ReduceByKeyTransform received no data!");
    } else {
      keyToValues.entrySet().stream().map(entry -> {
        final V value = ReduceTransform.reduceIterator(entry.getValue().iterator(), func);
        return new Tuple2<>(entry.getKey(), value);
      }).forEach(outputCollector::emit);
      keyToValues.clear();
    }
  }
}