spark streaming 5 read kafka

package com.shujia.spark.streaming

import org.apache.spark.SparkConf
import org.apache.spark.rdd.RDD
import org.apache.spark.sql.{DataFrame, SaveMode, SparkSession}
import org.apache.spark.streaming.dstream.{DStream, ReceiverInputDStream}
import org.apache.spark.streaming.{Durations, StreamingContext}

object Demo4SSCToMysql {
  def main(args: Array[String]): Unit = {


    val spark: SparkSession = SparkSession.builder()
      .master("local[2]")
      .appName("mysql")
      .getOrCreate()

    import spark.implicits._

    /**
      * 创建streaming 上下文对象，指定batch的间隔时间,多久计算一次
      *
      */
    val ssc = new StreamingContext(spark.sparkContext, Durations.seconds(5))


    //设置checkpoint
    ssc.checkpoint("data/checkpoint")


    /**
      * ReceiverInputDStream: 被动接收数据，将接收过来的数据放在内存或者磁盘上
      * 接收数据会一直占用资源
      *
      * nc -lk 9888
      *
      *
      * yum install nc
      */

    //读取数据
    val linesDS: ReceiverInputDStream[String] = ssc.socketTextStream("master", 8888)


    val wordsDS: DStream[String] = linesDS.flatMap(_.split(","))

    val kvDS: DStream[(String, Int)] = wordsDS.map((_, 1))

    def updateFun(seq: Seq[Int], option: Option[Int]): Option[Int] = {

      //计算当前batch单词的数量
      val currCount: Int = seq.sum

      //获取之前单词的数量
      val lastCount: Int = option.getOrElse(0)

      //返回最新单词的数量
      Some(currCount + lastCount)

    }

    val countDS: DStream[(String, Int)] = kvDS.updateStateByKey(updateFun)


    //将结果保存到hdfs
    //countDS.saveAsTextFiles("data/sparkstream", "txt")


    countDS.foreachRDD((rdd: RDD[(String, Int)]) => {
      val countDF: DataFrame = rdd.toDF("word", "c")

      //将数据保存到mysql中

      countDF
        .write
        .format("jdbc")
        .mode(SaveMode.Overwrite)
        .option("url", "jdbc:mysql://master:3306?useUnicode=true&characterEncoding=utf-8")
        .option("dbtable", "student.wordcount")
        .option("user", "root")
        .option("password", "123456")
        .save()

    })

    //启动spark streaming
    ssc.start()
    ssc.awaitTermination()
    ssc.stop()

  }

}