RDD编程-RDD算子的使用（三）-Action算子

    @Test def reduceTest(): Unit = {
        // 1. 通过集合，创建一个RDD
        val rdd: RDD[Int] = sc.parallelize(1 to 100, 2)

        // 2. 聚合运算
        val res: Int = rdd.reduce(_ + _)

        // 3. 输出结果
        println(res)
    }

    @Test def foldTest(): Unit = {
        val rdd: RDD[Int] = sc.parallelize(1 to 100, 2)
        val res: Int = rdd.fold(10)(_ + _)
        println(res)
    }

    @Test def aggregateTest(): Unit = {
        // 1. 准备一个集合
        val rdd1: RDD[Int] = sc.parallelize(1 to 100, 2)
        // 2. 聚合
        val res: Int = rdd1.aggregate(80)(Math.max, _ + _)
        println(res)
    }

    
    @Test def aggregateTest(): Unit = {
        // 1. 准备一个集合
        val rdd1: RDD[Int] = sc.parallelize(1 to 100, 2)
        // 2. 聚合
        val res: Int = rdd1.aggregate(80)(Math.max, _ + _)
        println(res)
    }

    
    @Test def collectAsMap(): Unit = {
        val rdd: RDD[(Int, String)] = sc.parallelize(Array("Lily", "Uncle Wang", "Polly")).keyBy(_.length)
        val res: collection.Map[Int, String] = rdd.collectAsMap()

    
    @Test def countTest(): Unit = {
        val rdd: RDD[Int] = sc.parallelize(1 to 100)

        val count: Long = rdd.count()
        println(count)
    }

    
    @Test def countByKeyTest(): Unit = {
        val rdd: RDD[(Int, String)] = sc.parallelize(Array("Jim", "Tom", "Lily", "Lucy", "Polly", "Snoppy")).keyBy(_.length)
        val res: collection.Map[Int, Long] = rdd.countByKey()
        println(res)
    }

    
    @Test def takeTest(): Unit = {
        val rdd: RDD[Int] = sc.parallelize(Array(1, 2, 3, 4, 5, 6, 7))
        val res: Array[Int] = rdd.take(3)
        res.foreach(println)
    }

    
    @Test def takeSample(): Unit = {
        val rdd: RDD[Int] = sc.parallelize(1 to 100)
        val res: Array[Int] = rdd.takeSample(withReplacement = true, 10)
        res.foreach(println)
    }

    
    @Test def takeOrdered(): Unit = {
        val rdd: RDD[Int] = sc.parallelize(Array(1, 3, 5, 7, 9, 0, 8, 6, 4, 2))
        val res: Array[Int] = rdd.takeOrdered(5)
        res.foreach(println)
    }

    
    @Test def top(): Unit = {
        val rdd: RDD[Int] = sc.parallelize(Array(1, 3, 5, 7, 9, 0, 8, 6, 4, 2))
        val res: Array[Int] = rdd.top(5)
        res.foreach(println)
    }

    
    @Test def first(): Unit = {
        val res: Int = sc.parallelize(Array(1, 3, 5, 7, 9, 0, 8, 6, 4, 2)).first()
        println(res)
    }

    @Test def top(): Unit = {
        val rdd: RDD[Int] = sc.parallelize(Array(1, 3, 5, 7, 9, 0, 8, 6, 4, 2))
        //val res: Array[Int] = rdd.top(5)
        rdd.foreach(println)
    }

    @Test def saveTest(): Unit = {
        val rdd: RDD[Int] = sc.parallelize(1 to 100, 4)
        rdd.saveAsTextFile("file/spark/out")
    }