Map vs FlatMap(Scala)

Map vs FlatMap(Scala)

val data = Seq("his is sample string",
              "one more sample string" ,
              "third line",
              "line",
              "this is one more line")

data: Seq[String] = List(his is sample string, one more sample string, third line, line, this is one more line)

val df = spark.sparkContext.parallelize(data)

df: org.apache.spark.rdd.RDD[String] = ParallelCollectionRDD[7] at parallelize at command-2343346390787948:1

df.map(x => (x , x.length)).collect()

res0: Array[(String, Int)] = Array((his is sample string,20), (one more sample string,22), (third line,10), (line,4), (this is one more line,21))

df.map(x => x.split(" ")).collect()

res1: Array[Array[String]] = Array(Array(his, is, sample, string), Array(one, more, sample, string), Array(third, line), Array(line), Array(this, is, one, more, line))

df.flatMap( x => x.split(" ")).collect()

res2: Array[String] = Array(his, is, sample, string, one, more, sample, string, third, line, line, this, is, one, more, line)