Fix UnicodeEncodeError in PySpark saveAsTextFile().

Fixes SPARK-970.

python/pyspark/rdd.py

@@ -605,7 +605,10 @@ class RDD(object):
         '0\\n1\\n2\\n3\\n4\\n5\\n6\\n7\\n8\\n9\\n'
         """
         def func(split, iterator):
-            return (str(x).encode("utf-8") for x in iterator)
+            for x in iterator:
+                if not isinstance(x, basestring):
+                    x = unicode(x)
+                yield x.encode("utf-8")
         keyed = PipelinedRDD(self, func)
         keyed._bypass_serializer = True
         keyed._jrdd.map(self.ctx._jvm.BytesToString()).saveAsTextFile(path)

python/pyspark/tests.py

@@ -19,6 +19,8 @@
 Unit tests for PySpark; additional tests are implemented as doctests in
 individual modules.
 """
+from fileinput import input
+from glob import glob
 import os
 import shutil
 import sys
@@ -138,6 +140,19 @@ class TestAddFile(PySparkTestCase):
         self.assertEqual("Hello World from inside a package!", UserClass().hello())
 
 
+class TestRDDFunctions(PySparkTestCase):
+
+    def test_save_as_textfile_with_unicode(self):
+        # Regression test for SPARK-970
+        x = u"\u00A1Hola, mundo!"
+        data = self.sc.parallelize([x])
+        tempFile = NamedTemporaryFile(delete=True)
+        tempFile.close()
+        data.saveAsTextFile(tempFile.name)
+        raw_contents = ''.join(input(glob(tempFile.name + "/part-0000*")))
+        self.assertEqual(x, unicode(raw_contents.strip(), "utf-8"))
+
+
 class TestIO(PySparkTestCase):
 
     def test_stdout_redirection(self):