making summary function switch to str from json_dumps if it can't do

json_dumps. Adding support for ipaddress type in DeepDiff and DeepHash
and Search

Author: seperman

deepdiff/base.py

@@ -1,21 +1,12 @@
-from typing import Protocol, Any
+from typing import Any
 from deepdiff.helper import strings, numbers, SetOrdered
 
 
 DEFAULT_SIGNIFICANT_DIGITS_WHEN_IGNORE_NUMERIC_TYPES = 12
 TYPE_STABILIZATION_MSG = 'Unable to stabilize the Numpy array {} due to {}. Please set ignore_order=False.'
 
 
-class BaseProtocol(Protocol):
-    t1: Any
-    t2: Any
-    cutoff_distance_for_pairs: float
-    use_log_scale: bool
-    log_scale_similarity_threshold: float
-    view: str
-
-
-class Base(BaseProtocol):
+class Base:
     numbers = numbers
     strings = strings
 

deepdiff/deephash.py

@@ -8,7 +8,7 @@
 from pathlib import Path
 from enum import Enum
 from deepdiff.helper import (strings, numbers, times, unprocessed, not_hashed, add_to_frozen_set,
-                             convert_item_or_items_into_set_else_none, get_doc,
+                             convert_item_or_items_into_set_else_none, get_doc, ipranges,
                              convert_item_or_items_into_compiled_regexes_else_none,
                              get_id, type_is_subclass_of_type_group, type_in_type_group,
                              number_to_string, datetime_normalize, KEY_TO_VAL_STR,
@@ -142,7 +142,7 @@ class DeepHash(Base):
     __doc__ = doc
 
     def __init__(self,
-                 obj,
+                 obj: Any,
                  *,
                  apply_hash=True,
                  custom_operators: Optional[List[Any]] =None,
@@ -484,6 +484,11 @@ def _prep_number(self, obj):
                                         number_format_notation=self.number_format_notation)
         return KEY_TO_VAL_STR.format(type_, obj)
 
+    def _prep_ipranges(self, obj):
+        type_ = 'iprange'
+        obj = str(obj)
+        return KEY_TO_VAL_STR.format(type_, obj)
+
     def _prep_datetime(self, obj):
         type_ = 'datetime'
         obj = datetime_normalize(self.truncate_datetime, obj, default_timezone=self.default_timezone)
@@ -558,6 +563,9 @@ def _hash(self, obj, parent, parents_ids=EMPTY_FROZENSET):
         elif isinstance(obj, numbers):  # type: ignore
             result = self._prep_number(obj)
 
+        elif isinstance(obj, ipranges):
+            result = self._prep_ipranges(obj)
+
         elif isinstance(obj, MutableMapping):
             result, counts = self._prep_dict(obj=obj, parent=parent, parents_ids=parents_ids)
 

deepdiff/diff.py

@@ -27,7 +27,7 @@
                              np_ndarray, np_floating, get_numpy_ndarray_rows, RepeatedTimer,
                              TEXT_VIEW, TREE_VIEW, DELTA_VIEW, detailed__dict__, add_root_to_paths,
                              np, get_truncate_datetime, dict_, CannotCompare, ENUM_INCLUDE_KEYS,
-                             PydanticBaseModel, Opcode, SetOrdered)
+                             PydanticBaseModel, Opcode, SetOrdered, ipranges)
 from deepdiff.serialization import SerializationMixin
 from deepdiff.distance import DistanceMixin, logarithmic_similarity
 from deepdiff.model import (
@@ -1511,6 +1511,11 @@ def _diff_numbers(self, level, local_tree=None, report_type_change=True):
             if t1_s != t2_s:
                 self._report_result('values_changed', level, local_tree=local_tree)
 
+    def _diff_ipranges(self, level, local_tree=None):
+        """Diff IP ranges"""
+        if str(level.t1) != str(level.t2):
+            self._report_result('values_changed', level, local_tree=local_tree)
+
     def _diff_datetime(self, level, local_tree=None):
         """Diff DateTimes"""
         level.t1 = datetime_normalize(self.truncate_datetime, level.t1, default_timezone=self.default_timezone)
@@ -1705,6 +1710,9 @@ def _diff(self, level, parents_ids=frozenset(), _original_type=None, local_tree=
         elif isinstance(level.t1, datetime.datetime):
             self._diff_datetime(level, local_tree=local_tree)
 
+        elif isinstance(level.t1, ipranges):
+            self._diff_ipranges(level, local_tree=local_tree)
+
         elif isinstance(level.t1, (datetime.date, datetime.timedelta, datetime.time)):
             self._diff_time(level, local_tree=local_tree)
 

deepdiff/distance.py

@@ -1,22 +1,45 @@
 import math
 import datetime
-from deepdiff.base import BaseProtocol
+from typing import TYPE_CHECKING, Callable, Protocol, Any
 from deepdiff.deephash import DeepHash
 from deepdiff.helper import (
     DELTA_VIEW, numbers, strings, add_to_frozen_set, not_found, only_numbers, np, np_float64, time_to_seconds,
     cartesian_product_numpy, np_ndarray, np_array_factory, get_homogeneous_numpy_compatible_type_of_seq, dict_,
     CannotCompare)
 from collections.abc import Mapping, Iterable
 
+if TYPE_CHECKING:
+    from deepdiff.diff import DeepDiffProtocol
 
-DISTANCE_CALCS_NEEDS_CACHE = "Distance calculation can not happen once the cache is purged. Try with _cache='keep'"
+    class DistanceProtocol(DeepDiffProtocol, Protocol):
+        hashes: dict
+        deephash_parameters: dict
+        iterable_compare_func: Callable | None
+        math_epsilon: float
+        cutoff_distance_for_pairs: float
+
+        def __get_item_rough_length(self, item, parent:str="root") -> float:
+            ...
 
+        def _to_delta_dict(
+            self,
+            directed: bool = True,
+            report_repetition_required: bool = True,
+            always_include_values: bool = False,
+        ) -> dict:
+            ...
 
+        def __calculate_item_deephash(self, item: Any) -> None:
+            ...
 
 
-class DistanceMixin(BaseProtocol):
 
-    def _get_rough_distance(self):
+DISTANCE_CALCS_NEEDS_CACHE = "Distance calculation can not happen once the cache is purged. Try with _cache='keep'"
+
+
+class DistanceMixin:
+
+    def _get_rough_distance(self: "DistanceProtocol"):
         """
         Gives a numeric value for the distance of t1 and t2 based on how many operations are needed to convert
         one to the other.
@@ -51,7 +74,7 @@ def _get_rough_distance(self):
 
         return diff_length / (t1_len + t2_len)
 
-    def __get_item_rough_length(self, item, parent='root'):
+    def __get_item_rough_length(self: "DistanceProtocol", item, parent='root'):
         """
         Get the rough length of an item.
         It is used as a part of calculating the rough distance between objects.
@@ -69,7 +92,7 @@ def __get_item_rough_length(self, item, parent='root'):
             length = DeepHash.get_key(self.hashes, key=item, default=None, extract_index=1)
         return length
 
-    def __calculate_item_deephash(self, item):
+    def __calculate_item_deephash(self: "DistanceProtocol", item: Any) -> None:
         DeepHash(
             item,
             hashes=self.hashes,
@@ -79,8 +102,7 @@ def __calculate_item_deephash(self, item):
         )
 
     def _precalculate_distance_by_custom_compare_func(
-            self, hashes_added, hashes_removed, t1_hashtable, t2_hashtable, _original_type):
-
+            self: "DistanceProtocol", hashes_added, hashes_removed, t1_hashtable, t2_hashtable, _original_type):
         pre_calced_distances = dict_()
         for added_hash in hashes_added:
             for removed_hash in hashes_removed:
@@ -99,7 +121,7 @@ def _precalculate_distance_by_custom_compare_func(
         return pre_calced_distances
 
     def _precalculate_numpy_arrays_distance(
-            self, hashes_added, hashes_removed, t1_hashtable, t2_hashtable, _original_type):
+            self: "DistanceProtocol", hashes_added, hashes_removed, t1_hashtable, t2_hashtable, _original_type):
 
         # We only want to deal with 1D arrays.
         if isinstance(t2_hashtable[next(iter(hashes_added))].item, (np_ndarray, list)):
@@ -203,7 +225,7 @@ def _get_numbers_distance(num1, num2, max_=1, use_log_scale=False, log_scale_sim
         return 0
     if use_log_scale:
         distance = logarithmic_distance(num1, num2)
-        if distance < logarithmic_distance:
+        if distance < 0:
             return 0
         return distance
     if not isinstance(num1, float):
@@ -246,7 +268,7 @@ def numpy_apply_log_keep_sign(array, offset=MATH_LOG_OFFSET):
     return signed_log_values
 
 
-def logarithmic_similarity(a: numbers, b: numbers, threshold: float=0.1):
+def logarithmic_similarity(a: numbers, b: numbers, threshold: float=0.1) -> float:
     """
     A threshold of 0.1 translates to about 10.5% difference.
     A threshold of 0.5 translates to about 65% difference.
@@ -255,7 +277,7 @@ def logarithmic_similarity(a: numbers, b: numbers, threshold: float=0.1):
     return logarithmic_distance(a, b) < threshold
 
 
-def logarithmic_distance(a: numbers, b: numbers):
+def logarithmic_distance(a: numbers, b: numbers) -> float:
     # Apply logarithm to the absolute values and consider the sign
     a = float(a)
     b = float(b)

deepdiff/helper.py

@@ -8,7 +8,8 @@
 import string
 import time
 import enum
-from typing import NamedTuple, Any, List, Optional, Dict, Union, TYPE_CHECKING
+import ipaddress
+from typing import NamedTuple, Any, List, Optional, Dict, Union, TYPE_CHECKING, Tuple
 from ast import literal_eval
 from decimal import Decimal, localcontext, InvalidOperation as InvalidDecimalOperation
 from itertools import repeat
@@ -184,9 +185,10 @@ def get_semvar_as_integer(version):
 only_complex_number = (complex,) + numpy_complex_numbers
 only_numbers = (int, float, complex, Decimal) + numpy_numbers
 datetimes = (datetime.datetime, datetime.date, datetime.timedelta, datetime.time)
+ipranges = (ipaddress.IPv4Interface, ipaddress.IPv6Interface, ipaddress.IPv4Network, ipaddress.IPv6Network)
 uuids = (uuid.UUID, )
 times = (datetime.datetime, datetime.time)
-numbers = only_numbers + datetimes
+numbers: Tuple = only_numbers + datetimes
 booleans = (bool, np_bool_)
 
 basic_types = strings + numbers + uuids + booleans + (type(None), )

deepdiff/search.py

@@ -5,7 +5,7 @@
 import logging
 
 from deepdiff.helper import (
-    strings, numbers, add_to_frozen_set, get_doc, dict_, RE_COMPILED_TYPE
+    strings, numbers, add_to_frozen_set, get_doc, dict_, RE_COMPILED_TYPE, ipranges
 )
 
 logger = logging.getLogger(__name__)
@@ -115,7 +115,7 @@ def __init__(self,
             matched_values=self.__set_or_dict(),
             unprocessed=[])
         self.use_regexp = use_regexp
-        if not strict_checking and isinstance(item, numbers):
+        if not strict_checking and (isinstance(item, numbers) or isinstance(item, ipranges)):
             item = str(item)
         if self.use_regexp:
             try:
@@ -306,7 +306,7 @@ def __search(self, obj, item, parent="root", parents_ids=frozenset()):
         if self.__skip_this(item, parent):
             return
 
-        elif isinstance(obj, strings) and isinstance(item, (strings, RE_COMPILED_TYPE)):
+        elif isinstance(obj, strings) and isinstance(item, (strings, RE_COMPILED_TYPE, ipranges)):
             self.__search_str(obj, item, parent)
 
         elif isinstance(obj, strings) and isinstance(item, numbers):

deepdiff/serialization.py

@@ -32,6 +32,7 @@
     pydantic_base_model_type,
     PydanticBaseModel,
     NotPresent,
+    ipranges,
 )
 from deepdiff.model import DeltaResult
 
@@ -112,7 +113,8 @@ class UnsupportedFormatErr(TypeError):
     'SetOrdered': SetOrdered,
     'namedtuple': collections.namedtuple,
     'OrderedDict': collections.OrderedDict,
-    'Pattern': re.Pattern,    
+    'Pattern': re.Pattern,
+    'iprange': str,
 }
 
 

deepdiff/summarize.py

@@ -22,7 +22,10 @@ def calculate_weights(node):
         weight = 0
         children_weights = {}
         for k, v in node.items():
-            edge_weight = len(k)
+            try:
+                edge_weight = len(k)
+            except TypeError:
+                edge_weight = 1
             child_weight, child_structure = calculate_weights(v)
             total_weight = edge_weight + child_weight
             weight += total_weight
@@ -133,6 +136,9 @@ def greedy_tree_summarization_balanced(json_data: JSON, max_weight: int, balance
 
 
 def summarize(data: JSON, max_length:int=200, balance_threshold:float=0.6) -> str:
-    return json_dumps(
-        greedy_tree_summarization_balanced(data, max_length, balance_threshold)
-    )
+    try:
+        return json_dumps(
+            greedy_tree_summarization_balanced(data, max_length, balance_threshold)
+        )
+    except Exception:
+        return str(data)

tests/test_hash.py

@@ -4,11 +4,13 @@
 import pytz
 import logging
 import datetime
+import ipaddress
+from typing import Union
 from pathlib import Path
 from collections import namedtuple
 from functools import partial
 from enum import Enum
-from deepdiff import DeepHash
+from deepdiff import DeepDiff, DeepHash
 from deepdiff.deephash import (
     prepare_string_for_hashing, unprocessed,
     UNPROCESSED_KEY, BoolObj, HASH_LOOKUP_ERR_MSG, combine_hashes_lists)
@@ -999,10 +1001,39 @@ def test_combine_hashes_lists(self, items, prefix, expected):
         (7, b"First have a cup of potatos. Then \xc3\x28 cup of flour", None, False, UnicodeDecodeError, EXPECTED_MESSAGE3),
     ])
     def test_hash_encodings(self, test_num, item, encodings, ignore_encoding_errors, expected_result, expected_message):
-        if UnicodeDecodeError == expected_result:
+        if UnicodeDecodeError == expected_result:  # NOQA
             with pytest.raises(expected_result) as exc_info:
                 DeepHash(item, encodings=encodings, ignore_encoding_errors=ignore_encoding_errors)
             assert expected_message == str(exc_info.value), f"test_encodings test #{test_num} failed."
         else:
             result = DeepHash(item, encodings=encodings, ignore_encoding_errors=ignore_encoding_errors)
             assert expected_result == result, f"test_encodings test #{test_num} failed."
+
+    def test_ip_addresses(self):
+
+        class ClassWithIp:
+            """Class containing single data member to demonstrate deepdiff infinite iterate over IPv6Interface"""
+
+            def __init__(self, addr: str):
+                self.field: Union[
+                    ipaddress.IPv4Network,
+                    ipaddress.IPv6Network,
+                    ipaddress.IPv4Interface,
+                    ipaddress.IPv6Interface,
+                ] = ipaddress.IPv6Network(addr)
+
+
+        obj1 = ClassWithIp("2002:db8::/30")
+        obj1_hash = DeepHashPrep(obj1)
+        repr(obj1_hash)  # shouldn't raise error
+        assert r"objClassWithIp:{str:field:iprange:2002:db8::/30}" == obj1_hash[obj1]
+        obj2 = ClassWithIp("2001:db8::/32")
+        diff = DeepDiff(obj1, obj2)
+        assert {
+            "values_changed": {
+                "root.field": {
+                    "new_value": ipaddress.IPv6Network("2001:db8::/32"),
+                    "old_value": ipaddress.IPv6Network("2002:db8::/30"),
+                }
+            }
+        } == diff