add comments

Author: giwa

python/pyspark/java_gateway.py

@@ -111,7 +111,7 @@ def run(self):
     java_import(gateway.jvm, "org.apache.spark.streaming.*")  # do we need this?
     java_import(gateway.jvm, "org.apache.spark.streaming.api.java.*")
     java_import(gateway.jvm, "org.apache.spark.streaming.api.python.*")
-    java_import(gateway.jvm, "org.apache.spark.streaming.dstream.*")
+    java_import(gateway.jvm, "org.apache.spark.streaming.dstream.*")  # do we need this?
     java_import(gateway.jvm, "org.apache.spark.mllib.api.python.*")
     java_import(gateway.jvm, "org.apache.spark.sql.SQLContext")
     java_import(gateway.jvm, "org.apache.spark.sql.hive.HiveContext")

python/pyspark/streaming/context.py

@@ -152,9 +152,6 @@ def _testInputStream(self, test_inputs, numSlices=None):
             test_rdds.append(test_rdd._jrdd)
             test_rdd_deserializers.append(test_rdd._jrdd_deserializer)
 
-#        if len(set(test_rdd_deserializers)) > 1:
-#            raise IOError("Deserializer should be one type to run test case. "
-#                          "See the SparkContext.parallelize to understand how to decide deserializer")
         jtest_rdds = ListConverter().convert(test_rdds, SparkContext._gateway._gateway_client)
         jinput_stream = self._jvm.PythonTestInputStream(self._jssc, jtest_rdds).asJavaDStream()
 

python/pyspark/streaming/dstream.py

@@ -425,6 +425,7 @@ def saveAsTextFile(rdd, time):
 # TODO: implement leftOuterJoin
 # TODO: implemtnt rightOuterJoin
 
+
 class PipelinedDStream(DStream):
     def __init__(self, prev, func, preservesPartitioning=False):
         if not isinstance(prev, PipelinedDStream) or not prev._is_pipelinable():