模拟区道路监控管理(数据kafka消费者)

object Test2 {
  def main(args: Array[String]): Unit = {
    val conf = new SparkConf().setMaster("local[2]").setAppName("monitor_action_test2")
    val sc = new SparkContext(conf)
    sc.setLogLevel("WARN")
    val ssc = new StreamingContext(sc, Seconds(5))

    val kafkaParams = Map[String, Object](
      "bootstrap.servers" -> "hadoop-senior.test.com:9092",
      "key.deserializer" -> classOf[StringDeserializer],
      "value.deserializer" -> classOf[StringDeserializer],
      "group.id" -> "use_a_separate_group_id_for_each_stream",
      "auto.offset.reset" -> "latest",
      "enable.auto.commit" -> (false: java.lang.Boolean)
    )

    val topics = Array("RoadRealTimeLog")
    val stream = KafkaUtils.createDirectStream[String, String](
      ssc,
      PreferConsistent,
      Subscribe[String, String](topics, kafkaParams)
    )

    val cars = stream.map(line => line.value().split("\t")(3))
    val result1 = cars.transform(rdd => rdd.distinct()).count()
    result1.print()

    val result2 = cars.map((_,1)).reduceByKey(_+_).filter(_._2 >= 2)
    result2.print()

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