spark-streaming-kafka怎样通过KafkaUtils.createDirectStream的方式处理数据，相信很多没有经验的人对此束手无策，为此本文总结了问题出现的原因和解决方法，通过这篇文章希望你能解决这个问题。
10年积累的成都做网站、成都网站建设经验，可以快速应对客户对网站的新想法和需求。提供各种问题对应的解决方案。让选择我们的客户得到更好、更有力的网络服务。我虽然不认识你，你也不认识我。但先网站制作后付款的网站建设流程，更有松桃免费网站建设让你可以放心的选择与我们合作。
package hgs.spark.streaming
import org.apache.spark.SparkConf
import org.apache.spark.SparkContext
import org.apache.spark.streaming.StreamingContext
import org.apache.spark.streaming.Seconds
import org.apache.spark.streaming.kafka.KafkaUtils
import org.apache.spark.streaming.kafka.KafkaCluster
import scala.collection.immutable.Map
import java.util.NoSuchElementException
import org.apache.spark.SparkException
import kafka.common.TopicAndPartition
import kafka.message.MessageAndMetadata
import org.codehaus.jackson.map.deser.std.PrimitiveArrayDeserializers.StringDeser
import kafka.serializer.StringDecoder
import org.apache.spark.streaming.kafka.DirectKafkaInputDStream
import org.apache.spark.rdd.RDD
import org.apache.spark.streaming.kafka.HasOffsetRanges
import org.apache.spark.HashPartitioner
object SparkStreamingKafkaDirectWordCount {
def main(args: Array[String]): Unit = {
val conf = new SparkConf().setAppName("KafkaWordCount").setMaster("local[5]")
conf.set("spark.streaming.kafka.maxRatePerPartition", "1")
val sc = new SparkContext(conf)
val ssc = new StreamingContext(sc,Seconds(1))
ssc.checkpoint("d:\\checkpoint")
val kafkaParams = Map[String,String](
"metadata.broker.list"->"bigdata01:9092,bigdata02:9092,bigdata03:9092",
"group.id"->"group_hgs",
"zookeeper.connect"->"bigdata01:2181,bigdata02:2181,bigdata03:2181")
val kc = new KafkaCluster(kafkaParams)
val topics = Set[String]("test")
//每个rdd返回的数据是(K,V)类型的，该函数规定了函数返回数据的类型
val mmdFunct = (mmd: MessageAndMetadata[String, String])=>(mmd.topic+" "+mmd.partition,mmd.message())
val rds = KafkaUtils.createDirectStream[String,String,StringDecoder,StringDecoder,(String,String)](ssc, kafkaParams, getOffsets(topics,kc,kafkaParams),mmdFunct)
val updateFunc=(iter:Iterator[(String,Seq[Int],Option[Int])])=>{
//iter.flatMap(it=>Some(it._2.sum+it._3.getOrElse(0)).map((it._1,_)))//方式一
//iter.flatMap{case(x,y,z)=>{Some(y.sum+z.getOrElse(0)).map((x,_))}}//方式二
iter.flatMap(it=>Some(it._1,(it._2.sum.toInt+it._3.getOrElse(0))))//方式三
}
val words = rds.flatMap(x=>x._2.split(" ")).map((_,1))
//val wordscount = words.map((_,1)).updateStateByKey(updateFunc, new HashPartitioner(sc.defaultMinPartitions), true)
//println(getOffsets(topics,kc,kafkaParams))
rds.foreachRDD(rdd=>{
if(!rdd.isEmpty()){
//对每个dataStreamoffset进行更新
upateOffsets(topics,kc,rdd,kafkaParams)
}
}
)
words.print()
ssc.start()
ssc.awaitTermination()
}
def getOffsets(topics : Set[String],kc:KafkaCluster,kafkaParams:Map[String,String]):Map[TopicAndPartition, Long]={
val topicAndPartitionsOrNull = kc.getPartitions(topics)
if(topicAndPartitionsOrNull.isLeft){
throw new SparkException(s"$topics in the set may not found")
}
else{
val topicAndPartitions = topicAndPartitionsOrNull.right.get
val groups = kafkaParams.get("group.id").get
val offsetOrNull = kc.getConsumerOffsets(groups, topicAndPartitions)
if(offsetOrNull.isLeft){
println(s"$groups you assignment may not exists!now redirect to zero!")
//如果没有消费过，则从最开始的位置消费
val erliestOffset = kc.getEarliestLeaderOffsets(topicAndPartitions)
if(erliestOffset.isLeft)
throw new SparkException(s"Topics and Partions not definded not found!")
else
erliestOffset.right.get.map(x=>(x._1,x._2.offset))
}
else{
//如果消费组已经存在则从记录的地方开始消费
offsetOrNull.right.get
}
}
}
//每次拉取数据后存储offset到ZK
def upateOffsets(topics : Set[String],kc:KafkaCluster,directRDD:RDD[(String,String)],kafkaParams:Map[String,String]){
val offsetRanges = directRDD.asInstanceOf[HasOffsetRanges].offsetRanges
for(offr <-offsetRanges){
val topicAndPartitions = TopicAndPartition(offr.topic,offr.partition)
val yesOrNo = kc.setConsumerOffsets(kafkaParams.get("group.id").get, Map(topicAndPartitions->offr.untilOffset))
if(yesOrNo.isLeft){
println(s"Error when update offset of $topicAndPartitions")
}
}
}
}
/* val conf = new SparkConf().setAppName("KafkaWordCount").setMaster("local[2]")
val sc = new SparkContext(conf)
val ssc = new StreamingContext(sc,Seconds(4))
val kafkaParams = Map[String,String](
"metadata.broker.list"->"bigdata01:9092,bigdata02:9092,bigdata03:9092")
val kc = new KafkaCluster(kafkaParams)
//获取topic与paritions的信息
//val tmp = kc.getPartitions(Set[String]("test7"))
//结果：topicAndPartitons=Set([test7,0], [test7,1], [test7,2])
//val topicAndPartitons = tmp.right.get
//println(topicAndPartitons)
//每个分区对应的leader信息
//val tmp = kc.getPartitions(Set[String]("test7"))
//val topicAndPartitons = tmp.right.get
//结果：leadersPerPartitions= Right(Map([test7,0] -> (bigdata03,9092), [test7,1] -> (bigdata01,9092), [test7,2] -> (bigdata02,9092)))
//val leadersPerPartitions = kc.findLeaders(topicAndPartitons)
//println(leadersPerPartitions)
//每增加一条消息，对应的partition的offset都会加1,即LeaderOffset(bigdata02,9092,23576)第三个参数会加一
//val tmp = kc.getPartitions(Set[String]("test"))
//val topicAndPartitons = tmp.right.get
//结果t= Right(Map([test7,0] -> LeaderOffset(bigdata03,9092,23568), [test7,2] -> LeaderOffset(bigdata02,9092,23576), [test7,1] -> LeaderOffset(bigdata01,9092,23571)))
//val t = kc.getLatestLeaderOffsets(topicAndPartitons)
// println(t)
//findLeader需要两个参数 topic 分区编号
//val tmp = kc.findLeader("test7",0)
//结果leader=RightProjection(Right((bigdata03,9092)))
//val leader = tmp.right
//val tp = leader.flatMap(x=>{Either.cond(false, None,(x._1,x._2))})
val tmp = kc.getPartitions(Set[String]("test"))
val ttp = tmp.right.get
while(true){
try{
val tp = kc.getConsumerOffsets("group_test1", ttp)
val maps = tp.right.get
println(maps)
Thread.sleep(2000)
}
catch{
case ex:NoSuchElementException=>{println("test")}
}
}*/看完上述内容，你们掌握spark-streaming-kafka怎样通过KafkaUtils.createDirectStream的方式处理数据的方法了吗？如果还想学到更多技能或想了解更多相关内容，欢迎关注创新互联行业资讯频道，感谢各位的阅读！
