SPARK-1917: fix PySpark import of scipy.special functions

https://issues.apache.org/jira/browse/SPARK-1917

Author: Uri Laserson <[email protected]>

Closes apache#866 from laserson/SPARK-1917 and squashes the following commits:

d947e8c [Uri Laserson] Added test for scipy.special importing
1798bbd [Uri Laserson] SPARK-1917: fix PySpark import of scipy.special

Conflicts:
	python/pyspark/tests.py

Author: laserson

python/pyspark/cloudpickle.py

@@ -933,7 +933,7 @@ def _change_cell_value(cell, newval):
 Note: These can never be renamed due to client compatibility issues"""
 
 def _getobject(modname, attribute):
-    mod = __import__(modname)
+    mod = __import__(modname, fromlist=[attribute])
     return mod.__dict__[attribute]
 
 def _generateImage(size, mode, str_rep):

python/pyspark/tests.py

@@ -33,6 +33,14 @@
 from pyspark.java_gateway import SPARK_HOME
 from pyspark.serializers import read_int
 
+_have_scipy = False
+try:
+    import scipy.sparse
+    _have_scipy = True
+except:
+    # No SciPy, but that's okay, we'll skip those tests
+    pass
+
 
 class PySparkTestCase(unittest.TestCase):
 
@@ -234,5 +242,22 @@ def test_termination_sigterm(self):
         from signal import SIGTERM
         self.do_termination_test(lambda daemon: os.kill(daemon.pid, SIGTERM))
 
+
+@unittest.skipIf(not _have_scipy, "SciPy not installed")
+class SciPyTests(PySparkTestCase):
+    """General PySpark tests that depend on scipy """
+
+    def test_serialize(self):
+        from scipy.special import gammaln
+        x = range(1, 5)
+        expected = map(gammaln, x)
+        observed = self.sc.parallelize(x).map(gammaln).collect()
+        self.assertEqual(expected, observed)
+
+
 if __name__ == "__main__":
+    if not _have_scipy:
+        print "NOTE: Skipping SciPy tests as it does not seem to be installed"
     unittest.main()
+    if not _have_scipy:
+        print "NOTE: SciPy tests were skipped as it does not seem to be installed"