how to implement Exactly-once when Spark Streaming + Kafka Integration

I have a question.There is a guide how to implement exactly-one，here is the code: https://spark.apache.org/docs/latest/streaming-kafka-0-10-integration.html#storing-offsets

val stream = KafkaUtils.createDirectStream[String, String](
  streamingContext,
  PreferConsistent,
  Subscribe[String, String](topics, kafkaParams)
)

stream.map(record => (record.key, record.value))
 //=====================================================
 //separate line
 //=====================================================
stream.foreachRDD { rdd =>
  val offsetRanges = rdd.asInstanceOf[HasOffsetRanges].offsetRanges

  // some time later, after outputs have completed
  stream.asInstanceOf[CanCommitOffsets].commitAsync(offsetRanges)
}

But what if I want to using the 'reduceByKeyAndWindow' int the separate line,just like this:

 val stream = KafkaUtils.createDirectStream[String, String](
  streamingContext,
  PreferConsistent,
  Subscribe[String, String](topics, kafkaParams)
)


val lines: DStream[String] = stream.map(record => record.value)
lines.map(row => {
  (row.split(",")(1), 1)
}).reduceByKeyAndWindow((a: Int, b: Int) => (a + b), Seconds(30), Seconds(5))
  .foreachRDD(rdd => {
    val offsetRanges = rdd.asInstanceOf[HasOffsetRanges].offsetRanges
    //mycode start
    rdd.foreach(println)
    //mycaode end
    stream.asInstanceOf[CanCommitOffsets].commitAsync(offsetRanges)
  })

I am trying this ,but I got a error:

Exception in thread "main" java.lang.ClassCastException: org.apache.spark.rdd.MapPartitionsRDD cannot be cast to org.apache.spark.streaming.kafka010.HasOffsetRanges

AnyHelp?Thanks in advance！