SPARK-1630: Make PythonRDD handle NULL elements and strings gracefully

core/src/main/scala/org/apache/spark/api/python/PythonRDD.scala

@@ -245,7 +245,7 @@ private object SpecialLengths {
   val TIMING_DATA = -3
 }
 
-private[spark] object PythonRDD {
+private[spark] object PythonRDD extends Logging {
   val UTF8 = Charset.forName("UTF-8")
 
   def readRDDFromFile(sc: JavaSparkContext, filename: String, parallelism: Int):
@@ -301,15 +301,23 @@ private[spark] object PythonRDD {
               throw new SparkException("Unexpected Tuple2 element type " + pair._1.getClass)
           }
         case other =>
-          throw new SparkException("Unexpected element type " + first.getClass)
+          if (other == null) {
+            logDebug("Encountered NULL element from iterator. We skip writing NULL to stream.")
+          } else {
+            throw new SparkException("Unexpected element type " + first.getClass)
+          }
       }
     }
   }
 
   def writeUTF(str: String, dataOut: DataOutputStream) {
-    val bytes = str.getBytes(UTF8)
-    dataOut.writeInt(bytes.length)
-    dataOut.write(bytes)
+    if (str == null) {
+      logDebug("Encountered NULL string. We skip writing NULL to stream.")
+    } else {
+        val bytes = str.getBytes(UTF8)
+        dataOut.writeInt(bytes.length)
+        dataOut.write(bytes)
+    }
   }
 
   def writeToFile[T](items: java.util.Iterator[T], filename: String) {

core/src/test/scala/org/apache/spark/api/python/PythonRDDSuite.scala

@@ -29,5 +29,10 @@ class PythonRDDSuite extends FunSuite {
         PythonRDD.writeIteratorToStream(input.iterator, buffer)
     }
 
+    test("Handle nulls gracefully") {
+        val input: List[String] = List("a", null)
+        val buffer = new DataOutputStream(new ByteArrayOutputStream)
+        PythonRDD.writeIteratorToStream(input.iterator, buffer)
+    }
 }