6.mapPartitionsWithIndex操作
将RDD中所有考试分数高于95分的“学生准考证号”,“对应分数” 和“当前数据所在分区”拼接后输出。
scala> val rddData =sc.parallelize(Array(("20180001",83),("201800002",97),("20180003",100),("20180004",95),("20180005",87)),2)
rddData: org.apache.spark.rdd.RDD[(String, Int)] = ParallelCollectionRDD[0] at parallelize at <console>:24
scala> val rddData2 = rddData.mapPartitionsWithIndex((index,iter) => {var result = List[String]()
| while(iter.hasNext){result = iter.next() match{case (id,grade) if grade >= 95 => id + "_" + grade + "[" + index + "]" :: result case _ => result}}
| result.iterator
| })
rddData2: org.apache.spark.rdd.RDD[String] = MapPartitionsRDD[1] at mapPartitionsWithIndex at <console>:26
scala> rddData2.collect
res0: Array[String] = Array(201800002_97[0], 20180004_95[1], 20180003_100[1])
说明:
mapPartitionsWithIndex与mapPartitions功能类似,但是它可以知道数据的分区位置。