Report functionality added to BaseColumnProfiler (#497)

* Fix: updated test_(int/float)_column_profile.py test_diff to cover json.dumps() functionality. Updated numerical_column_stats.py _perform_t_test to cast conservative and welch values to floats before assignment.

* Fix: addressed PR change requests.

* Fix: Reformatting

* Bumped version

* added report function to BaseColumnProfiler, added report to profile classes, added test coverage

* fixing formatting and naming

dataprofiler/profilers/base_column_profilers.py

@@ -230,6 +230,17 @@ def profile(self):
         """
         raise NotImplementedError()
 
+    @abc.abstractmethod
+    def report(self, remove_disabled_flag=False):
+        """
+        Private abstract method for returning report.
+
+        :param remove_disabled_flag: flag to determine if disabled options should be excluded in the report.
+        :type remove_disabled_flag: boolean
+        """
+
+        raise NotImplementedError()
+
 
 class BaseColumnPrimitiveTypeProfiler(with_metaclass(abc.ABCMeta,
                                                      BaseColumnProfiler)):

dataprofiler/profilers/categorical_column_profile.py

@@ -117,6 +117,15 @@ def diff(self, other_profile, options=None):
 
         return differences
 
+    def report(self, remove_disabled_flag=False):
+        """
+        Private abstract method for returning report.
+
+        :param remove_disabled_flag: flag to determine if disabled options should be excluded in the report.
+        :type remove_disabled_flag: boolean
+        """
+        return self.profile
+
     @property
     def profile(self):
         """

dataprofiler/profilers/column_profile_compilers.py

@@ -194,6 +194,15 @@ class ColumnPrimitiveTypeProfileCompiler(BaseCompiler):
     ]
     _option_class = StructuredOptions
 
+    def report(self, remove_disabled_flag=False):
+        """
+        Private abstract method for returning report.
+
+        :param remove_disabled_flag: flag to determine if disabled options should be excluded in the report.
+        :type remove_disabled_flag: boolean
+        """
+        return self.profile
+
     @property
     def profile(self):
         profile = {
@@ -287,6 +296,15 @@ class ColumnStatsProfileCompiler(BaseCompiler):
     ]
     _option_class = StructuredOptions
 
+    def report(self, remove_disabled_flag=False):
+        """
+        Private abstract method for returning report.
+
+        :param remove_disabled_flag: flag to determine if disabled options should be excluded in the report.
+        :type remove_disabled_flag: boolean
+        """
+        return self.profile
+
     @property
     def profile(self):
         profile = dict()
@@ -325,6 +343,15 @@ class ColumnDataLabelerCompiler(BaseCompiler):
     ]
     _option_class = StructuredOptions
 
+    def report(self, remove_disabled_flag=False):
+        """
+        Private abstract method for returning report.
+
+        :param remove_disabled_flag: flag to determine if disabled options should be excluded in the report.
+        :type remove_disabled_flag: boolean
+        """
+        return self.profile
+
     @property
     def profile(self):
         profile = {
@@ -376,6 +403,15 @@ class UnstructuredCompiler(BaseCompiler):
 
     _option_class = UnstructuredOptions
 
+    def report(self, remove_disabled_flag=False):
+        """
+        Private abstract method for returning report.
+
+        :param remove_disabled_flag: flag to determine if disabled options should be excluded in the report.
+        :type remove_disabled_flag: boolean
+        """
+        return self.profile
+
     @property
     def profile(self):
         profile = {

dataprofiler/profilers/data_labeler_column_profile.py

@@ -239,6 +239,15 @@ def profile(self):
         }
         return profile
 
+    def report(self, remove_disabled_flag=False):
+        """
+        Private abstract method for returning report.
+
+        :param remove_disabled_flag: flag to determine if disabled options should be excluded in the report.
+        :type remove_disabled_flag: boolean
+        """
+        return self.profile
+
     def diff(self, other_profile, options=None):
         """
         Generates the differences between the orders of two DataLabeler columns

dataprofiler/profilers/datetime_column_profile.py

@@ -111,6 +111,15 @@ def __add__(self, other):
             self.date_formats, other.date_formats)
         return merged_profile
 
+    def report(self, remove_disabled_flag=False):
+        """
+        Private abstract method for returning report.
+
+        :param remove_disabled_flag: flag to determine if disabled options should be excluded in the report.
+        :type remove_disabled_flag: boolean
+        """
+        return self.profile
+
     @property
     def profile(self):
         """

dataprofiler/profilers/int_column_profile.py

@@ -53,6 +53,15 @@ def __add__(self, other):
                                  other.__calculations)
         return merged_profile
 
+    def report(self, remove_disabled_flag=False):
+        """
+        Private abstract method for returning report.
+
+        :param remove_disabled_flag: flag to determine if disabled options should be excluded in the report.
+        :type remove_disabled_flag: boolean
+        """
+        return self.profile
+
     @property
     def profile(self):
         """

dataprofiler/profilers/order_column_profile.py

@@ -228,6 +228,15 @@ def __add__(self, other):
                                  other.__calculations)
         return merged_profile
 
+    def report(self, remove_disabled_flag=False):
+        """
+        Private abstract method for returning report.
+
+        :param remove_disabled_flag: flag to determine if disabled options should be excluded in the report.
+        :type remove_disabled_flag: boolean
+        """
+        return self.profile
+
     @property
     def profile(self):
         """

dataprofiler/profilers/unstructured_labeler_profile.py

@@ -89,6 +89,15 @@ def __add__(self, other):
 
         return merged_profile
 
+    def     report(self, remove_disabled_flag=False):
+        """
+        Private abstract method for returning report.
+
+        :param remove_disabled_flag: flag to determine if disabled options should be excluded in the report.
+        :type remove_disabled_flag: boolean
+        """
+        return self.profile
+
     def diff(self, other_profile, options=None):
         """
         Finds the differences for two unstructured labeler profiles

dataprofiler/tests/profilers/test_base_column_profilers.py

@@ -152,7 +152,7 @@ def test_cannot_instantiate(self):
             BaseColumnPrimitiveTypeProfiler()
         self.assertEqual(
             "Can't instantiate abstract class BaseColumnPrimitiveTypeProfiler "
-            "with abstract methods _update_helper, profile, update",
+            "with abstract methods _update_helper, profile, report, update",
             str(e.exception)
         )
 
@@ -181,7 +181,7 @@ def test_update_column_base_properties(self):
     def test_update_match_are_abstract(self):
         six.assertCountEqual(
             self,
-            {'_update_helper', 'update', 'profile'},
+            {'_update_helper', 'update', 'report', 'profile'},
             BaseColumnPrimitiveTypeProfiler.__abstractmethods__
         )
 
@@ -284,6 +284,16 @@ def test_updated_profile(self):
             self.assertEqual(1, profiler_mock.call_count)
         self._delete_profiler_mocks()
 
+    def test_report(self):
+        self._create_profiler_mocks()
+        profile = self.column_profiler(self.aws_dataset["datetime"])
+
+        report1 = profile.profile
+        report2 = profile.report(remove_disabled_flag=False)
+        report3 = profile.report(remove_disabled_flag=True)
+        self.assertDictEqual(report1, report2)
+        self.assertDictEqual(report1, report3)
+
     def test_profile(self):
         self._create_profiler_mocks()
         profile = self.column_profiler(self.aws_dataset["datetime"])

dataprofiler/tests/profilers/test_categorical_column_profile.py

@@ -167,6 +167,7 @@ def test_true_categorical_report(self):
                 ('unalikeability', 2*(12 + 15 + 20)/132)
             ]),
         )
+
         # We have to pop these values because sometimes the order changes
         self.assertCountEqual(expected_profile['statistics'].pop('categories'),
                               report["statistics"].pop('categories'))
@@ -190,6 +191,17 @@ def test_false_categorical_report(self):
         )
         self.assertEqual(report, expected_profile)
 
+    def test_report(self):
+        df_non_categorical = pd.Series(list(map(str, range(0, 20))))
+        profile = CategoricalColumn(df_non_categorical.name)
+        profile.update(df_non_categorical)
+
+        report1 = profile.profile
+        report2 = profile.report(remove_disabled_flag=False)
+        report3 = profile.report(remove_disabled_flag=True)
+        self.assertDictEqual(report1, report2)
+        self.assertDictEqual(report1, report3)
+
     def test_categorical_merge(self):
         df1 = pd.Series(["abcd", "aa", "abcd", "aa", "b", "4", "3", "2",
                          "dfd", "2", np.nan])

dataprofiler/tests/profilers/test_data_labeler_column_profile.py

@@ -164,6 +164,18 @@ def test_profile(self, mock_instance):
             expected = defaultdict(float, {'data_labeler_predict': 2.0})
             self.assertEqual(expected, profiler.profile['times'])
 
+    def test_report(self, mock_instance):
+        self._setup_data_labeler_mock(mock_instance)
+
+        data = pd.Series(['1', '2', '3'])
+        profile = DataLabelerColumn(data.name)
+
+        report1 = profile.profile
+        report2 = profile.report(remove_disabled_flag=False)
+        report3 = profile.report(remove_disabled_flag=True)
+        self.assertDictEqual(report1, report2)
+        self.assertDictEqual(report1, report3)
+
     def test_label_match(self, mock_instance):
         """
         Test label match between avg_prediction and data_label_representation

dataprofiler/tests/profilers/test_datetime_column_profile.py

@@ -261,14 +261,28 @@ def test_profile(self):
             self.assertEqual(expected, profiler.profile['times'])
             profile = profiler.profile
             self.assertCountEqual(expected_profile, profile)
-            
+
             # Validate time in datetime class has expected time after second
             # update
             profiler.update(df)
             expected = defaultdict(float, {'datetime': 2.0})
             self.assertEqual(expected, profiler.profile['times'])
             self.assertEqual(expected_profile.pop('max'), profiler.profile['max'])
 
+    def test_report(self):
+        data = [
+            2.5, 12.5, '2013-03-10 15:43:30', 5, '03/10/13 15:43',
+            'Mar 11, 2013'
+        ]
+        df = pd.Series(data).apply(str)
+        profile = DateTimeColumn(df.name)
+
+        report1 = profile.profile
+        report2 = profile.report(remove_disabled_flag=False)
+        report3 = profile.report(remove_disabled_flag=True)
+        self.assertDictEqual(report1, report2)
+        self.assertDictEqual(report1, report3)
+
     def test_warning_for_bad_dates(self):
 
         df = pd.Series(['03/10/2013 15:43'])

dataprofiler/tests/profilers/test_order_column_profile.py

@@ -110,6 +110,18 @@ def test_profile(self):
             # key and value populated correctly
             self.assertDictEqual(expected_profile, profile)
 
+    def test_report(self):
+        data = [1]
+        df = pd.Series(data).apply(str)
+
+        profile  = OrderColumn(df.name)
+
+        report1 = profile.profile
+        report2 = profile.report(remove_disabled_flag=False)
+        report3 = profile.report(remove_disabled_flag=True)
+        self.assertDictEqual(report1, report2)
+        self.assertDictEqual(report1, report3)
+
     def test_profile_merge(self):
         data = [1, 2, 3, 4, 5, 6]
         df = pd.Series(data).apply(str)

dataprofiler/tests/profilers/test_profile_builder.py

@@ -1836,7 +1836,7 @@ def test_column_names(self):
     def test_update_match_are_abstract(self):
         six.assertCountEqual(
             self,
-            {'profile', '_update_helper', 'update'},
+            {'profile', '_update_helper', 'report', 'update'},
             dp.profilers.BaseColumnProfiler.__abstractmethods__
         )
 

dataprofiler/tests/profilers/test_unstructured_labeler_profile.py

@@ -138,6 +138,36 @@ def test_profile(self, processor_class_mock, model_class_mock):
         # key and value populated correctly
         self.assertDictEqual(expected_profile, profile)
 
+    @mock.patch('dataprofiler.profilers.'
+                'unstructured_labeler_profile.DataLabeler')
+    @mock.patch('dataprofiler.profilers.'
+                'unstructured_labeler_profile.'
+                'CharPostprocessor')
+    def test_report(self, processor_class_mock, model_class_mock):
+        # setup mocks
+        model_mock = mock.Mock()
+        model_mock.reverse_label_mapping = {1: 'UNKNOWN'}
+        model_mock.predict.return_value = dict(pred=[[1]])
+        model_class_mock.return_value = model_mock
+        processor_mock = mock.Mock()
+        processor_mock.process.return_value = dict(pred=[[]])
+        processor_class_mock.return_value = processor_mock
+
+        # initialize labeler profile
+        profile = UnstructuredLabelerProfile()
+
+        sample = pd.Series(["a"])
+
+        time_array = [float(i) for i in range(4, 0, -1)]
+        with mock.patch('time.time', side_effect=lambda: time_array.pop()):
+            profile.update(sample)
+
+        report1 = profile.profile
+        report2 = profile.report(remove_disabled_flag=False)
+        report3 = profile.report(remove_disabled_flag=True)
+        self.assertDictEqual(report1, report2)
+        self.assertDictEqual(report1, report3)
+
     @mock.patch('dataprofiler.profilers.'
                 'unstructured_labeler_profile.DataLabeler')
     @mock.patch('dataprofiler.profilers.'