Spark1.4源码走读笔记之隐式转换

Posted 2020-08-09

 RDD源码中隐式转换

 def distinct(numPartitions: Int)(implicit ord: Ordering[T] = null): RDD[T] = withScope {

    map(x => (x, null)).reduceByKey((x, y) => x, numPartitions).map(_._1)

  }

使用了隐式参数，运行时不需要手动提供参数ord，运行时会根据上下文注入参数

  def groupBy[K](f: T => K)(implicit kt: ClassTag[K]): RDD[(K, Iterable[T])] = withScope {

    groupBy[K](f, defaultPartitioner(this))

  }

  def groupBy[K](

      f: T => K,

      numPartitions: Int)(implicit kt: ClassTag[K]): RDD[(K, Iterable[T])] = withScope {

    groupBy(f, new HashPartitioner(numPartitions))

  }