OrderColumn Differences

dataprofiler/profilers/order_column_profile.py

@@ -1,4 +1,5 @@
 from . import BaseColumnProfiler
+from . import utils
 from .profiler_options import OrderOptions
 
 
@@ -237,6 +238,25 @@ def profile(self):
         """
         return dict(order=self.order, times=self.times)
 
+    def diff(self, other_profile, options=None):
+        """
+        Generates the differences between the orders of two OrderColumns
+
+        :return: Dict containing the differences between orders in their
+        appropriate output formats
+        :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 = {
+            "order": utils.find_diff_of_strings(self.order, other_profile.order)
+        }
+        return differences
+
     @BaseColumnProfiler._timeit(name="order")
     def _get_data_order(self, df_series):
         """

dataprofiler/tests/profilers/test_order_column_profile.py

@@ -325,3 +325,25 @@ def test_order_column_with_wrong_options(self):
                                    "OrderColumn parameter 'options' must be of"
                                    " type OrderOptions."):
             profiler = OrderColumn("Order", options="wrong_data_type")
+
+    def test_diff(self):
+        data = [1, 2, 3, 4, 5, 6]
+        df = pd.Series(data).apply(str)
+        profiler = OrderColumn("placeholder_name")
+        profiler.update(df)
+
+        data2 = [7, 8, 9, 10]
+        df2 = pd.Series(data2).apply(str)
+        profiler2 = OrderColumn("placeholder_name")
+        profiler2.update(df2)
+
+        diff = profiler.diff(profiler2)
+        self.assertEqual("unchanged", diff["order"])
+
+        data3 = [4, 2, 3, 1, 5]
+        df3 = pd.Series(data3).apply(str)
+        profiler3 = OrderColumn("placeholder_name")
+        profiler3.update(df3)
+
+        diff = profiler.diff(profiler3)
+        self.assertEqual(["ascending", "random"], diff["order"])