Merge pull request #351 from stephenh/values

Add PairRDDFunctions.keys and values.

core/src/main/scala/spark/PairRDDFunctions.scala

@@ -615,6 +615,16 @@ class PairRDDFunctions[K: ClassManifest, V: ClassManifest](
     writer.cleanup()
   }
 
+  /**
+   * Return an RDD with the keys of each tuple.
+   */
+  def keys: RDD[K] = self.map(_._1)
+  
+  /**
+   * Return an RDD with the values of each tuple.
+   */
+  def values: RDD[V] = self.map(_._2)
+
   private[spark] def getKeyClass() = implicitly[ClassManifest[K]].erasure
 
   private[spark] def getValueClass() = implicitly[ClassManifest[V]].erasure

core/src/test/scala/spark/ShuffleSuite.scala

@@ -216,6 +216,13 @@ class ShuffleSuite extends FunSuite with ShouldMatchers with BeforeAndAfter {
     // Test that a shuffle on the file works, because this used to be a bug
     assert(file.map(line => (line, 1)).reduceByKey(_ + _).collect().toList === Nil)
   }
+
+  test("keys and values") {
+    sc = new SparkContext("local", "test")
+    val rdd = sc.parallelize(Array((1, "a"), (2, "b")))
+    assert(rdd.keys.collect().toList === List(1, 2))
+    assert(rdd.values.collect().toList === List("a", "b"))
+  }
 }
 
 object ShuffleSuite {