Merge branch 'dev' of github.com:seperman/deepdiff into dev

Author: seperman

deepdiff/deephash.py

@@ -144,6 +144,7 @@ def __init__(self,
                  parent="root",
                  encodings=None,
                  ignore_encoding_errors=False,
+                 ignore_iterable_order=True,
                  **kwargs):
         if kwargs:
             raise ValueError(
@@ -190,6 +191,7 @@ def __init__(self,
         self.ignore_private_variables = ignore_private_variables
         self.encodings = encodings
         self.ignore_encoding_errors = ignore_encoding_errors
+        self.ignore_iterable_order = ignore_iterable_order
 
         self._hash(obj, parent=parent, parents_ids=frozenset({get_id(obj)}))
 
@@ -424,7 +426,9 @@ def _prep_iterable(self, obj, parent, parents_ids=EMPTY_FROZENSET):
                 '{}|{}'.format(i, v) for i, v in result.items()
             ]
 
-        result = sorted(map(str, result))  # making sure the result items are string and sorted so join command works.
+        result = map(str, result) # making sure the result items are string so join command works.
+        if self.ignore_iterable_order:
+            result = sorted(result)  
         result = ','.join(result)
         result = KEY_TO_VAL_STR.format(type(obj).__name__, result)
 

deepdiff/diff.py

@@ -13,6 +13,7 @@
 from math import isclose as is_close
 from collections.abc import Mapping, Iterable, Sequence
 from collections import defaultdict
+from inspect import getmembers
 from itertools import zip_longest
 from ordered_set import OrderedSet
 from deepdiff.helper import (strings, bytes_type, numbers, uuids, datetimes, ListItemRemovedOrAdded, notpresent,
@@ -417,20 +418,25 @@ def _diff_enum(self, level, parents_ids=frozenset(), local_tree=None):
 
     def _diff_obj(self, level, parents_ids=frozenset(), is_namedtuple=False, local_tree=None):
         """Difference of 2 objects"""
+        processing_error = False
         try:
             if is_namedtuple:
                 t1 = level.t1._asdict()
                 t2 = level.t2._asdict()
-            else:
+            elif all('__dict__' in dir(t) for t in level):
                 t1 = detailed__dict__(level.t1, ignore_private_variables=self.ignore_private_variables)
                 t2 = detailed__dict__(level.t2, ignore_private_variables=self.ignore_private_variables)
-        except AttributeError:
-            try:
+            elif all('__slots__' in dir(t) for t in level):
                 t1 = self._dict_from_slots(level.t1)
                 t2 = self._dict_from_slots(level.t2)
-            except AttributeError:
-                self._report_result('unprocessed', level, local_tree=local_tree)
-                return
+            else:
+                t1 = {k: v for k, v in getmembers(level.t1) if not callable(v)}
+                t2 = {k: v for k, v in getmembers(level.t2) if not callable(v)}
+        except AttributeError:
+            processing_error = True
+        if processing_error is True:
+            self._report_result('unprocessed', level, local_tree=local_tree)
+            return
 
         self._diff_dict(
             level,
@@ -876,7 +882,8 @@ def _diff_by_forming_pairs_and_comparing_one_by_one(
                     x,
                     y,
                     child_relationship_class=child_relationship_class,
-                    child_relationship_param=j)
+                    child_relationship_param=j
+                )
                 self._diff(next_level, parents_ids_added, local_tree=local_tree)
 
     def _diff_ordered_iterable_by_difflib(
@@ -1529,7 +1536,7 @@ def _diff(self, level, parents_ids=frozenset(), _original_type=None, local_tree=
         if isinstance(level.t1, booleans):
             self._diff_booleans(level, local_tree=local_tree)
 
-        if isinstance(level.t1, strings):
+        elif isinstance(level.t1, strings):
             self._diff_str(level, local_tree=local_tree)
 
         elif isinstance(level.t1, datetimes):

deepdiff/model.py

@@ -577,6 +577,10 @@ def __setattr__(self, key, value):
         else:
             self.__dict__[key] = value
 
+    def __iter__(self):
+        yield self.t1
+        yield self.t2
+
     @property
     def repetition(self):
         return self.additional['repetition']

docs/deephash_doc.rst

@@ -123,6 +123,8 @@ ignore_private_variables: Boolean, default = True
 ignore_encoding_errors: Boolean, default = False
     If you want to get away with UnicodeDecodeError without passing explicit character encodings, set this option to True. If you want to make sure the encoding is done properly, keep this as False and instead pass an explicit list of character encodings to be considered via the encodings parameter.
 
+ignore_iterable_order: Boolean, default = True
+    If order of items in an iterable should not cause the hash of the iterable to be different.
 
 number_format_notation : string, default="f"
     number_format_notation is what defines the meaning of significant digits. The default value of "f" means the digits AFTER the decimal point. "f" stands for fixed point. The other option is "e" which stands for exponent notation or scientific notation.

requirements-cli.txt

@@ -1,2 +1,2 @@
 click==8.1.3
-pyyaml==6.0
+pyyaml==6.0.1

tests/test_diff_text.py

@@ -2,6 +2,7 @@
 import datetime
 import pytest
 import logging
+import re
 import uuid
 from enum import Enum
 from typing import List
@@ -605,6 +606,64 @@ class MyEnum(Enum):
         }
         assert ddiff == result
 
+    def test_precompiled_regex(self):
+
+        pattern_1 = re.compile('foo')
+        pattern_2 = re.compile('foo')
+        pattern_3 = re.compile('foo', flags=re.I)
+        pattern_4 = re.compile('(foo)')
+        pattern_5 = re.compile('bar')
+
+        # same object
+        ddiff = DeepDiff(pattern_1, pattern_1)
+        result = {}
+        assert ddiff == result
+
+        # same pattern, different object
+        ddiff = DeepDiff(pattern_1, pattern_2)
+        result = {}
+        assert ddiff == result
+
+        # same pattern, different flags
+        ddiff = DeepDiff(pattern_1, pattern_3)
+        result = {
+            'values_changed': {
+                'root.flags': {
+                    'new_value': 34,
+                    'old_value': 32,
+                },
+            }
+        }
+        assert ddiff == result
+
+        # same pattern, different groups
+        ddiff = DeepDiff(pattern_1, pattern_4)
+        result = {
+            'values_changed': {
+                'root.pattern': {
+                    'new_value': '(foo)',
+                    'old_value': 'foo',
+                },
+                'root.groups': {
+                    'new_value': 1,
+                    'old_value': 0,
+                },
+            }
+        }
+        assert ddiff == result
+
+        # different pattern
+        ddiff = DeepDiff(pattern_1, pattern_5)
+        result = {
+            'values_changed': {
+                'root.pattern': {
+                    'new_value': 'bar',
+                    'old_value': 'foo',
+                },
+            }
+        }
+        assert ddiff == result
+
     def test_custom_objects_change(self):
         t1 = CustomClass(1)
         t2 = CustomClass(2)

tests/test_hash.py

@@ -368,6 +368,21 @@ def test_same_sets_same_hash(self):
         t2_hash = DeepHashPrep(t2)
 
         assert t1_hash[get_id(t1)] == t2_hash[get_id(t2)]
+        
+    @pytest.mark.parametrize("list1, list2, ignore_iterable_order, is_equal", [
+        ([1, 2], [2, 1], False, False),
+        ([1, 2], [2, 1], True, True),
+        ([1, 2, 3], [1, 3, 2], False, False),
+        ([1, [1, 2, 3]], [1, [3, 2, 1]], False, False),
+        ([1, [1, 2, 3]], [1, [3, 2, 1]], True, True),
+        ((1, 2), (2, 1), False, False),
+        ((1, 2), (2, 1), True, True),
+    ])
+    def test_ignore_iterable_order(self, list1, list2, ignore_iterable_order, is_equal):
+        list1_hash = DeepHash(list1, ignore_iterable_order=ignore_iterable_order)
+        list2_hash = DeepHash(list2, ignore_iterable_order=ignore_iterable_order)
+        
+        assert is_equal == (list1_hash[list1] == list2_hash[list2])
 
     @pytest.mark.parametrize("t1, t2, significant_digits, number_format_notation, result", [
         ({0.012, 0.98}, {0.013, 0.99}, 1, "f", 'set:float:0.0,float:1.0'),