spark_rdd_算子

发表于

rdd

rdd

creattion

tranformation

reduceByKey

1
2
3
4
5
6
7
8
9
10
11
12
13
14
// 
val rdd = sc.parallelize(Seq(
  ("123", State("123", Set("1", "2", "3"))),
  ("123", State("123", Set("1", "2", "3", "4"))),
  ("123", State("123", Set("1", "2", "3", "4", "5"))),
  ("aaa", State("aaa", Set("1", "2", "3")))
))
rdd.reduceByKey((x, y) => {
  if (x.objs.size > y.objs.size) {
    x
  } else {
    y
  }
}).foreach(println)

cache

1
2
3
4
5
def invertTree(root):
    if root is None:
        return None
    root.left, root.right = invertTree(root.right), invertTree(root.left)
    return root
1
2
3
4
5
def invert(node):
    if node is None:
        return None
    else
        return Tree(node.value, invert(node.right), invert(node.left))

action