SparkStreaming整合Kafka实现wordcount案例（spark-streaming-kafka-0-10

      org.apache.spark
      spark-streaming-kafka-0-10_2.11
      2.4.5

def main(args: Array[String]): Unit = {
    val conf: SparkConf = new SparkConf().setMaster("local[2]").setAppName("SparkStreamDemo1")

    //定义流处理文件，采集周期为3秒
    val streamingContext = new StreamingContext(conf, Seconds(3))

    //数据源
    val socketLineStream: ReceiverInputDStream[String] = streamingContext.socketTextStream("192.168.91.180",7777)

    //处理每三秒钟采集到的数据
    val wordStream: DStream[String] = socketLineStream.flatMap(line => line.split("\s+"))

    val wordCountStream: DStream[(String, Int)] = wordStream.map(x => (x, 1)).reduceByKey(_ + _)
    wordCountStream.print()

    //启动采集器
    streamingContext.start()
    streamingContext.awaitTermination()
  }

启动nc监听端口7777 nc -lk 7777