IntColumn Stats Diff (#279)

* Welcome to the best commit you've ever seen

* This is almost ready. The code is still cooking

* Some NICE code.

dataprofiler/profilers/numerical_column_stats.py

@@ -247,7 +247,7 @@ def _add_helper(self, other1, other2):
         if "num_negatives" in self.__calculations.keys():
             self.num_negatives = other1.num_negatives + other2.num_negatives
 
-    def _diff_helper(self, other_profile):
+    def diff(self, other_profile, options=None):
         """
         Finds the differences for several numerical stats.
 
@@ -256,6 +256,12 @@ def _diff_helper(self, other_profile):
         :return: the numerical stats differences
         :rtype: dict
         """
+        cls = self.__class__
+        if not isinstance(other_profile, cls):
+            raise TypeError("Unsupported operand type(s) for diff: '{}' "
+                            "and '{}'".format(cls.__name__,
+                                              other_profile.__class__.__name__))
+
         differences = {
             "min": utils.find_diff_of_numbers(self.min, other_profile.min),
             "max": utils.find_diff_of_numbers(self.max, other_profile.max),

dataprofiler/tests/profilers/test_int_column_profile.py

@@ -788,3 +788,35 @@ def test_profile_merge_bin_edges_indices(self):
         profile_2.update(data_2)
 
         profile_1 + profile_2
+
+    def test_diff(self):
+        """
+        Makes sure the IntColumn Diff() works appropriately.
+        """
+        data = [2, 'not an int', 6, 4]
+        df = pd.Series(data).apply(str)
+        profiler1 = IntColumn("Int")
+        profiler1.update(df)
+
+        data = [1, 15]
+        df = pd.Series(data).apply(str)
+        profiler2 = IntColumn("Int")
+        profiler2.update(df)
+
+        # Assert the difference report is correct
+        expected_diff = {
+            'max': -9.0,
+            'mean': -4.0,
+            'min': 1.0,
+            'stddev': -7.899494936611665,
+            'sum': -4.0,
+            'variance': -94.0
+        }
+        self.assertDictEqual(expected_diff, profiler1.diff(profiler2))
+
+        # Assert type error is properly called
+        with self.assertRaises(TypeError) as exc:
+            profiler1.diff("Inproper input")
+        self.assertEqual(str(exc.exception),
+                         "Unsupported operand type(s) for diff: 'IntColumn' and"
+                         " 'str'")

dataprofiler/tests/profilers/test_numeric_stats_mixin_profile.py

@@ -503,7 +503,7 @@ def test_merge_num_zeros_and_negatives(self):
         self.assertEqual(num_profiler.num_zeros, 0)
         self.assertEqual(num_profiler.num_negatives, 0)
 
-    def test_diff_helper(self):
+    def test_diff(self):
         """
         Checks _diff_helper() works appropriately.
         """
@@ -528,4 +528,11 @@ def test_diff_helper(self):
             'variance': -8.362573099415204,
             'stddev': -2.0238425028660023
         }
-        self.assertDictEqual(expected_diff, other1._diff_helper(other2))
+        self.assertDictEqual(expected_diff, other1.diff(other2))
+
+        # Assert type error is properly called
+        with self.assertRaises(TypeError) as exc:
+            other1.diff("Inproper input")
+        self.assertEqual(str(exc.exception),
+                         "Unsupported operand type(s) for diff: 'TestColumn' and"
+                         " 'str'")