How to use RDD filter with other function

Question

I am working on Spark RDD. I know how to filter a RDD like val y = rdd.filter(e => e%2==0), but I do not know how to combine filter with other function like Row.

In val rst = rdd.map(ab => Row(ab.a, ab.b)), I want to filter out ab.b > 0, but I tried put filter at multiple place and they do not work.

Can someone help.

Answer 1

I'm not sure about the "out" part in "filter out": do you want to keep those entries, or do you want to get rid of them? If you want to drop all entries with ab.b > 0, then you need

val result = rdd.filterNot(_.b > 0).map(ab => Row(ab.a, ab.b))
If you want to retain only the entries with ab.b > 0, then try

val result = rdd.filter(_.b > 0).map(ab => Row(ab.a, ab.b))
The underscore _ is simply the shorter form of

val result = rdd.filter(ab => ab.b > 0).map(ab => Row(ab.a, ab.b))

Hope this will help.

Answer 2

val x = sc.parallelize(1 to 10, 2)
 
// filter operation 
val y = x.filter(e => e%2==0) 
y.collect
// res0: Array[Int] = Array(2, 4, 6, 8, 10)
 
// RDD y can be re written with shorter syntax in scala as 
val y = x.filter(_ % 2 == 0)
y.collect
// res1: Array[Int] = Array(2, 4, 6, 8, 10)