How to use

getSize

method

in

org.apache.crunch.PCollection

 public static <T> int getRecommendedPartitions(PCollection<T> pcollection, Configuration conf) {
  long bytesPerTask = conf.getLong(BYTES_PER_REDUCE_TASK, DEFAULT_BYTES_PER_REDUCE_TASK);
  Preconditions.checkArgument(bytesPerTask > 0);
  int recommended = 1 + (int) (pcollection.getSize() / bytesPerTask);
  int maxRecommended = conf.getInt(MAX_REDUCERS, DEFAULT_MAX_REDUCERS);
  if (maxRecommended > 0 && recommended > maxRecommended) {
   return maxRecommended;
  } else {
   return recommended;
  }
 }
}