SparkStreaming中的转换算子2--有状态的转换算子updateStateByKey

• 将之前批次的状态保存，

package SparkStreaming.trans

import org.apache.spark.SparkConf
import org.apache.spark.storage.StorageLevel
import org.apache.spark.streaming.dstream.DStream
import org.apache.spark.streaming.{Seconds, StreamingContext}

object ByUpdateByKey {
  def main(args: Array[String]): Unit = {
    val conf = new SparkConf().setMaster("local[3]").setAppName("transform3")
    val ssc = new StreamingContext(conf, Seconds(3))
    ssc.checkpoint("hdfs://node1:9000/sparkstreaming")

    val ds: DStream[String] = ssc.socketTextStream("node1", 44444, StorageLevel.MEMORY_ONLY)
    val ds1: DStream[(String, Int)] = ds.flatMap(_.split(" ")).map((_, 1))
    /*
     (A, B)
     A:之前批次处理得到的结果
     B:当前批次处理得到的结果
     */
    val ds2 = ds1.updateStateByKey((array: Seq[Int], state: Option[Int]) => {
      var num: Int = state.getOrElse(0)
      for (elem <- array) {
        num += elem
      }
      Option(num)
    })

    ds2.print()

    ssc.start()
    ssc.awaitTermination()
  }
}