adding use_enum_value and threshold_to_diff_deeper

Author: seperman

deepdiff/deephash.py

@@ -139,6 +139,7 @@ def __init__(self,
                  ignore_numeric_type_changes=False,
                  ignore_type_subclasses=False,
                  ignore_string_case=False,
+                 use_enum_value=False,
                  exclude_obj_callback=None,
                  number_to_string_func=None,
                  ignore_private_variables=True,
@@ -154,7 +155,7 @@ def __init__(self,
                  "exclude_paths, include_paths, exclude_regex_paths, hasher, ignore_repetition, "
                  "number_format_notation, apply_hash, ignore_type_in_groups, ignore_string_type_changes, "
                  "ignore_numeric_type_changes, ignore_type_subclasses, ignore_string_case "
-                 "number_to_string_func, ignore_private_variables, parent "
+                 "number_to_string_func, ignore_private_variables, parent, use_enum_value "
                  "encodings, ignore_encoding_errors") % ', '.join(kwargs.keys()))
         if isinstance(hashes, MutableMapping):
             self.hashes = hashes
@@ -170,6 +171,7 @@ def __init__(self,
         self.exclude_regex_paths = convert_item_or_items_into_compiled_regexes_else_none(exclude_regex_paths)
         self.hasher = default_hasher if hasher is None else hasher
         self.hashes[UNPROCESSED_KEY] = []
+        self.use_enum_value = use_enum_value
 
         self.significant_digits = self.get_significant_digits(significant_digits, ignore_numeric_type_changes)
         self.truncate_datetime = get_truncate_datetime(truncate_datetime)
@@ -206,10 +208,10 @@ def __init__(self,
     sha1hex = sha1hex
 
     def __getitem__(self, obj, extract_index=0):
-        return self._getitem(self.hashes, obj, extract_index=extract_index)
+        return self._getitem(self.hashes, obj, extract_index=extract_index, use_enum_value=self.use_enum_value)
 
     @staticmethod
-    def _getitem(hashes, obj, extract_index=0):
+    def _getitem(hashes, obj, extract_index=0, use_enum_value=False):
         """
         extract_index is zero for hash and 1 for count and None to get them both.
         To keep it backward compatible, we only get the hash by default so it is set to zero by default.
@@ -220,6 +222,8 @@ def _getitem(hashes, obj, extract_index=0):
             key = BoolObj.TRUE
         elif obj is False:
             key = BoolObj.FALSE
+        elif use_enum_value and isinstance(obj, Enum):
+            key = obj.value
 
         result_n_count = (None, 0)
 
@@ -256,14 +260,14 @@ def get(self, key, default=None, extract_index=0):
         return self.get_key(self.hashes, key, default=default, extract_index=extract_index)
 
     @staticmethod
-    def get_key(hashes, key, default=None, extract_index=0):
+    def get_key(hashes, key, default=None, extract_index=0, use_enum_value=False):
         """
         get_key method for the hashes dictionary.
         It can extract the hash for a given key that is already calculated when extract_index=0
         or the count of items that went to building the object whenextract_index=1.
         """
         try:
-            result = DeepHash._getitem(hashes, key, extract_index=extract_index)
+            result = DeepHash._getitem(hashes, key, extract_index=extract_index, use_enum_value=use_enum_value)
         except KeyError:
             result = default
         return result
@@ -481,6 +485,8 @@ def _hash(self, obj, parent, parents_ids=EMPTY_FROZENSET):
         if isinstance(obj, bool):
             obj = self._prep_bool(obj)
             result = None
+        elif self.use_enum_value and isinstance(obj, Enum):
+            obj = obj.value
         else:
             result = not_hashed
         try:

deepdiff/diff.py

@@ -98,6 +98,7 @@ def _report_progress(_stats, progress_logger, duration):
     'number_format_notation',
     'ignore_string_type_changes',
     'ignore_numeric_type_changes',
+    'use_enum_value',
     'ignore_type_in_groups',
     'ignore_type_subclasses',
     'ignore_string_case',
@@ -116,6 +117,7 @@ class DeepDiff(ResultDict, SerializationMixin, DistanceMixin, Base):
     def __init__(self,
                  t1: Any,
                  t2: Any,
+                 _original_type=None,
                  cache_purge_level: int=1,
                  cache_size: int=0,
                  cache_tuning_sample_size: int=0,
@@ -126,9 +128,6 @@ def __init__(self,
                  exclude_obj_callback: Optional[Callable]=None,
                  exclude_obj_callback_strict: Optional[Callable]=None,
                  exclude_paths: Union[str, List[str]]=None,
-                 include_obj_callback: Optional[Callable]=None,
-                 include_obj_callback_strict: Optional[Callable]=None,
-                 include_paths: Union[str, List[str]]=None,
                  exclude_regex_paths: Union[str, List[str], Pattern[str], List[Pattern[str]], None]=None,
                  exclude_types: Optional[List[Any]]=None,
                  get_deep_distance: bool=False,
@@ -146,8 +145,10 @@ def __init__(self,
                  ignore_string_type_changes: bool=False,
                  ignore_type_in_groups: Optional[List[Tuple]]=None,
                  ignore_type_subclasses: bool=False,
+                 include_obj_callback: Optional[Callable]=None,
+                 include_obj_callback_strict: Optional[Callable]=None,
+                 include_paths: Union[str, List[str]]=None,
                  iterable_compare_func: Optional[Callable]=None,
-                 zip_ordered_iterables: bool=False,
                  log_frequency_in_sec: int=0,
                  math_epsilon: Optional[float]=None,
                  max_diffs: Optional[int]=None,
@@ -157,10 +158,12 @@ def __init__(self,
                  progress_logger: Callable=logger.info,
                  report_repetition: bool=False,
                  significant_digits: Optional[int]=None,
+                 threshold_to_diff_deeper: float = 0,
                  truncate_datetime: Optional[str]=None,
+                 use_enum_value: bool=False,
                  verbose_level: int=1,
                  view: str=TEXT_VIEW,
-                 _original_type=None,
+                 zip_ordered_iterables: bool=False,
                  _parameters=None,
                  _shared_parameters=None,
                  **kwargs):
@@ -175,7 +178,7 @@ def __init__(self,
                 "view, hasher, hashes, max_passes, max_diffs, zip_ordered_iterables, "
                 "cutoff_distance_for_pairs, cutoff_intersection_for_pairs, log_frequency_in_sec, cache_size, "
                 "cache_tuning_sample_size, get_deep_distance, group_by, group_by_sort_key, cache_purge_level, "
-                "math_epsilon, iterable_compare_func, _original_type, "
+                "math_epsilon, iterable_compare_func, use_enum_value, _original_type, threshold_to_diff_deeper, "
                 "ignore_order_func, custom_operators, encodings, ignore_encoding_errors, "
                 "_parameters and _shared_parameters.") % ', '.join(kwargs.keys()))
 
@@ -193,6 +196,8 @@ def __init__(self,
             self.ignore_numeric_type_changes = ignore_numeric_type_changes
             if strings == ignore_type_in_groups or strings in ignore_type_in_groups:
                 ignore_string_type_changes = True
+            self.use_enum_value = use_enum_value
+            self.threshold_to_diff_deeper = threshold_to_diff_deeper
             self.ignore_string_type_changes = ignore_string_type_changes
             self.ignore_type_in_groups = self.get_ignore_types_in_groups(
                 ignore_type_in_groups=ignore_type_in_groups,
@@ -513,6 +518,8 @@ def _get_clean_to_keys_mapping(self, keys, level):
         for key in keys:
             if self.ignore_string_type_changes and isinstance(key, bytes):
                 clean_key = key.decode('utf-8')
+            elif self.use_enum_value and isinstance(key, Enum):
+                clean_key = key.value
             elif isinstance(key, numbers):
                 type_ = "number" if self.ignore_numeric_type_changes else key.__class__.__name__
                 clean_key = self.number_to_string(key, significant_digits=self.significant_digits,
@@ -578,6 +585,12 @@ def _diff_dict(
         t_keys_added = t2_keys - t_keys_intersect
         t_keys_removed = t1_keys - t_keys_intersect
 
+        if self.threshold_to_diff_deeper:
+            len_keys_changed = (len(t_keys_added) + len(t_keys_removed))
+            if len_keys_changed and len(t_keys_intersect) / len_keys_changed < self.threshold_to_diff_deeper:
+                self._report_result('values_changed', level, local_tree=local_tree)
+                return
+
         for key in t_keys_added:
             if self._count_diff() is StopIteration:
                 return
@@ -861,31 +874,6 @@ def _diff_by_forming_pairs_and_comparing_one_by_one(
                 self._report_result('iterable_item_added', change_level, local_tree=local_tree)
 
             else:  # check if item value has changed
-
-                # if (i != j):
-                #     # Item moved
-                #     change_level = level.branch_deeper(
-                #         x,
-                #         y,
-                #         child_relationship_class=child_relationship_class,
-                #         child_relationship_param=i,
-                #         child_relationship_param2=j
-                #     )
-                #     self._report_result('iterable_item_moved', change_level)
-
-                # item_id = id(x)
-                # if parents_ids and item_id in parents_ids:
-                #     continue
-                # parents_ids_added = add_to_frozen_set(parents_ids, item_id)
-
-                # # Go one level deeper
-                # next_level = level.branch_deeper(
-                #     x,
-                #     y,
-                #     child_relationship_class=child_relationship_class,
-                #     child_relationship_param=j)
-                # self._diff(next_level, parents_ids_added)
-
                 if (i != j and ((x == y) or self.iterable_compare_func)):
                     # Item moved
                     change_level = level.branch_deeper(
@@ -1604,6 +1592,12 @@ def _diff(self, level, parents_ids=frozenset(), _original_type=None, local_tree=
                 if self.type_check_func(level.t1, type_group) and self.type_check_func(level.t2, type_group):
                     report_type_change = False
                     break
+            if self.use_enum_value and isinstance(level.t1, Enum):
+                level.t1 = level.t1.value
+                report_type_change = False
+            if self.use_enum_value and isinstance(level.t2, Enum):
+                level.t2 = level.t2.value
+                report_type_change = False
             if report_type_change:
                 self._diff_types(level, local_tree=local_tree)
                 return

tests/__init__.py

@@ -65,5 +65,13 @@ def __reduce__(self):
         return (self.__class__, (item, ))
 
     def __eq__(self, other):
-        both_no_items_attr = (not hasattr(self, 'item')) and (not hasattr(other, 'item'))
-        return both_no_items_attr or self.item == other.item
+        if hasattr(self, 'item') and hasattr(other, 'item'):
+            return self.item == other.item
+        if not hasattr(self, 'item') and not hasattr(other, 'item'):
+            return True
+        return False
+
+    def __str__(self):
+        return f"<Picklable: {self.item if hasattr(self, 'item') else 'delete'}>"
+
+    __repr__ = __str__

tests/test_delta.py

@@ -463,6 +463,14 @@ def test_delta_dict_items_added_retain_order(self):
         delta2 = Delta(diff=diff, bidirectional=True)
         assert t1 == t2 - delta2
 
+        delta3 = Delta(diff, always_include_values=True, bidirectional=True, raise_errors=True)
+        flat_rows_list = delta3.to_flat_rows()
+        delta4 = Delta(flat_rows_list=flat_rows_list,
+                      always_include_values=True, bidirectional=True, raise_errors=True)
+        assert t1 == t2 - delta4
+        assert t1 + delta4 == t2
+
+
     def test_delta_constr_flat_dict_list_param_preserve(self):
         """
         Issue: https://github.com/seperman/deepdiff/issues/457
@@ -818,6 +826,13 @@ def compare_func(item1, item2, level=None):
             }
         }
     },
+    'delta_case14b_threshold_to_diff_deeper': {
+        't1': picklalbe_obj_without_item,
+        't2': PicklableClass(11),
+        'deepdiff_kwargs': {'threshold_to_diff_deeper': 0.33},
+        'to_delta_kwargs': {},
+        'expected_delta_dict': {'values_changed': {'root': {'new_value': PicklableClass(11)}}}
+    },
     'delta_case15_diffing_simple_numbers': {
         't1': 1,
         't2': 2,
@@ -1451,6 +1466,7 @@ def test_delta_view_and_to_delta_dict_are_equal_when_parameteres_passed(self):
             'ignore_string_type_changes': False,
             'ignore_type_in_groups': [],
             'report_repetition': True,
+            'use_enum_value': False,
             'exclude_paths': None,
             'include_paths': None,
             'exclude_regex_paths': None,

tests/test_diff_text.py

@@ -16,6 +16,16 @@
 logging.disable(logging.CRITICAL)
 
 
+class MyEnum1(Enum):
+    book = "book"
+    cake = "cake"
+
+class MyEnum2(str, Enum):
+    book = "book"
+    cake = "cake"
+
+
+
 class TestDeepDiffText:
     """DeepDiff Tests."""
 
@@ -649,14 +659,6 @@ class MyEnum(Enum):
 
     def test_enum_ignore_type_change(self):
 
-        class MyEnum1(Enum):
-            book = "book"
-            cake = "cake"
-
-        class MyEnum2(str, Enum):
-            book = "book"
-            cake = "cake"
-
         diff = DeepDiff("book", MyEnum1.book)
         expected = {
             'type_changes': {'root': {'old_type': str, 'new_type': MyEnum1, 'old_value': 'book', 'new_value': MyEnum1.book}}}
@@ -668,6 +670,14 @@ class MyEnum2(str, Enum):
         diff3 = DeepDiff("book", MyEnum2.book, ignore_type_in_groups=[(Enum, str)])
         assert not diff3
 
+    def test_enum_use_enum_value1(self):
+        diff = DeepDiff("book", MyEnum2.book, use_enum_value=True)
+        assert not diff
+
+    def test_enum_use_enum_value_in_dict_key(self):
+        diff = DeepDiff({"book": 2}, {MyEnum2.book: 2}, use_enum_value=True)
+        assert not diff
+
     def test_precompiled_regex(self):
 
         pattern_1 = re.compile('foo')
@@ -950,6 +960,9 @@ def test_custom_objects_add_and_remove_verbose(self):
 
     def get_custom_object_with_added_removed_methods(self):
         class ClassA:
+            VAL = 1
+            VAL2 = 2
+
             def method_a(self):
                 pass
 
@@ -1000,14 +1013,21 @@ def test_dictionary_of_custom_objects(self):
         result = {}
         assert result == ddiff
 
-    def test_dictionary_with_string_keys(self):
+    def test_dictionary_with_string_keys1(self):
         t1 = {"veggie": "carrots"}
         t2 = {"meat": "carrots"}
 
         diff = DeepDiff(t1, t2)
         assert {'dictionary_item_added': ["root['meat']"],
                 'dictionary_item_removed': ["root['veggie']"]} == diff
 
+    def test_dictionary_with_string_keys_threshold_to_diff_deeper(self):
+        t1 = {"veggie": "carrots"}
+        t2 = {"meat": "carrots"}
+
+        diff = DeepDiff(t1, t2, threshold_to_diff_deeper=0.33)
+        assert {'values_changed': {'root': {'new_value': {'meat': 'carrots'}, 'old_value': {'veggie': 'carrots'}}}} == diff
+
     def test_dictionary_with_numeric_keys(self):
         t1 = {Decimal('10.01'): "carrots"}
         t2 = {10.01: "carrots"}

tests/test_hash.py

@@ -310,6 +310,8 @@ class MyEnum(Enum):
         assert DeepHashPrep(MyEnum.A) != DeepHashPrep(MyEnum.A.value)
         assert DeepHashPrep(MyEnum.A) != DeepHashPrep(MyEnum.B)
 
+        assert DeepHashPrep(MyEnum.A, use_enum_value=True)[MyEnum.A] == 'int:1'
+
     def test_dict_hash(self):
         string1 = "a"
         string1_prepped = prep_str(string1)