erp5_data_notebook += test un/pickle pd.DF=>np.rec

Add test which verifies that we can un/pickle 'pd.DataFrame' which were
converted to 'np.recarray' via the '.to_records()' method.

This test is necessary because it turns out that the combination of
pandas >= 0.20.x + python2 doesn't support this functionality by
default if the data frames column names are unicodes.

Please see nexedi/erp5!1738 (comment 179298) for
additional context.

/reviewed-on nexedi/erp5!1738
/reviewed-by @jerome @klaus

bt5/erp5_data_notebook/TestTemplateItem/portal_components/test.erp5.testExecuteJupyter.py

@@ -35,6 +35,8 @@ import jupyter_client
 import base64
 import random
 import string
+import pickle
+import pandas as pd
 
 
 class TestExecuteJupyter(ERP5TypeTestCase):
@@ -1067,3 +1069,16 @@ print os.path
     else:
       has_access_to_erp5_kernel = True
     self.assertTrue(has_access_to_erp5_kernel)
+
+  def testConvertedDataFramesArePicklableAndUnpicklable(self):
+    '''
+      Verify we can pickle + unpickle pd.DataFrames which were converted to np.recarrays
+    '''
+    self.assertEqual(
+      pickle.loads(
+        pickle.dumps(
+          pd.DataFrame(columns=[u"test"]).to_records()
+        )
+      ).dtype.descr[1][0],
+      "test"
+    )