com.datatorrent.lib.algo.UniqueCounter Maven / Gradle / Ivy

Go to download
/**
 * 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 com.datatorrent.lib.algo;

import java.util.HashMap;
import java.util.Map;

import org.apache.commons.lang.mutable.MutableInt;

import com.datatorrent.api.DefaultInputPort;
import com.datatorrent.api.DefaultOutputPort;
import com.datatorrent.api.annotation.OperatorAnnotation;
import com.datatorrent.lib.util.BaseUniqueKeyCounter;
import com.datatorrent.lib.util.UnifierHashMapSumKeys;

/**
 * This operator counts the number of times a tuple exists in a window. A map from tuples to counts is emitted at the end of each window.
 * 
 * Counts the number of times a key exists in a window; Count is emitted at end of window in a single HashMap.
 * 
 * 
 * This is an end of window operator

 * 

 * StateFull : yes,  Tuples are aggregated over application window(s). 

 * Partitions : Yes,  Unique count is unified at output port. 

 * 

 * Ports:

 * data: expects K

 * count: emits HashMap<K,Integer>

 * Properties: None

 * 

 * 
 *
 * @displayName Count Unique Tuples
 * @category Stats and Aggregations
 * @tags count
 *
 * @since 0.3.2
 */

@OperatorAnnotation(partitionable = true)
public class UniqueCounter extends BaseUniqueKeyCounter
{
  private boolean cumulative;

  /**
   * The input port which receives incoming tuples.
   */
  public final transient DefaultInputPort data = new DefaultInputPort()
  {
    /**
     * Reference counts tuples
     */
    @Override
    public void process(K tuple)
    {
      processTuple(tuple);
    }

  };

  /**
   * The output port which emits a map from keys to the number of times they occurred within an application window.
   */
  public final transient DefaultOutputPort> count = new DefaultOutputPort>()
  {
    @Override
    public Unifier> getUnifier()
    {
      UnifierHashMapSumKeys unifierHashMapSumKeys =  new UnifierHashMapSumKeys<>();
      unifierHashMapSumKeys.setType(Integer.class);
      return unifierHashMapSumKeys;
    }
  };

  /**
   * Emits one HashMap as tuple
   */
  @Override
  public void endWindow()
  {
    HashMap tuple = null;
    for (Map.Entry e: map.entrySet()) {
      if (tuple == null) {
        tuple = new HashMap();
      }
      tuple.put(e.getKey(), e.getValue().toInteger());
    }
    if (tuple != null) {
      count.emit(tuple);
    }
    if (!cumulative) {
      map.clear();
    }
  }

  /**
   * Gets the cumulative mode.
   * @return The cumulative mode.
   */
  public boolean isCumulative()
  {
    return cumulative;
  }

  /**
   * If enabled then the unique keys is counted and maintained in memory for the life of the operator. If not enabled
   * keys are counted a per window bases.

   * Note: If cumulative mode is enabled and the operator receives many unique keys, then this operator
   * could eventually run out of memory.
   * @param cumulative
   */
  public void setCumulative(boolean cumulative)
  {
    this.cumulative = cumulative;
  }
}