Merge pull request #459 from seperman/dev

7.0.1

Author: seperman

CHANGELOG.md

@@ -1,5 +1,8 @@
 # DeepDiff Change log
 
+
+- v7-0-1
+    - Fixes the translation between Difflib opcodes and Delta flat rows.
 - v7-0-0
     - When verbose=2, return `new_path` when the `path` and `new_path` are different (for example when ignore_order=True and the index of items have changed).
     - Dropping support for Python 3.7

CITATION.cff

@@ -5,6 +5,6 @@ authors:
   given-names: "Sep"
   orcid: "https://orcid.org/0009-0009-5828-4345"
 title: "DeepDiff"
-version: 7.0.0
+version: 7.0.1
 date-released: 2024
 url: "https://github.com/seperman/deepdiff"

README.md

@@ -1,4 +1,4 @@
-# DeepDiff v 7.0.0
+# DeepDiff v 7.0.1
 
 ![Downloads](https://img.shields.io/pypi/dm/deepdiff.svg?style=flat)
 ![Python Versions](https://img.shields.io/pypi/pyversions/deepdiff.svg?style=flat)
@@ -17,12 +17,16 @@
 
 Tested on Python 3.8+ and PyPy3.
 
-- **[Documentation](https://zepworks.com/deepdiff/7.0.0/)**
+- **[Documentation](https://zepworks.com/deepdiff/7.0.1/)**
 
 ## What is new?
 
 Please check the [ChangeLog](CHANGELOG.md) file for the detailed information.
 
+DeepDiff 7-0-1
+
+- Fixes the translation between Difflib opcodes and Delta flat rows.
+
 DeepDiff 7-0-0
 
 - DeepDiff 7 comes with an improved delta object. [Delta to flat dictionaries](https://zepworks.com/deepdiff/current/serialization.html#delta-serialize-to-flat-dictionaries) have undergone a major change. We have also introduced [Delta serialize to flat rows](https://zepworks.com/deepdiff/current/serialization.html#delta-serialize-to-flat-rows).

deepdiff/__init__.py

@@ -1,6 +1,6 @@
 """This module offers the DeepDiff, DeepSearch, grep, Delta and DeepHash classes."""
 # flake8: noqa
-__version__ = '7.0.0'
+__version__ = '7.0.1'
 import logging
 
 if __name__ == '__main__':

deepdiff/delta.py

@@ -11,7 +11,9 @@
     strings, short_repr, numbers,
     np_ndarray, np_array_factory, numpy_dtypes, get_doc,
     not_found, numpy_dtype_string_to_type, dict_,
-    Opcode, FlatDeltaRow, UnkownValueCode,
+    Opcode, FlatDeltaRow, UnkownValueCode, FlatDataAction,
+    OPCODE_TAG_TO_FLAT_DATA_ACTION,
+    FLAT_DATA_ACTION_TO_OPCODE_TAG,
 )
 from deepdiff.path import (
     _path_to_elements, _get_nested_obj, _get_nested_obj_and_force,
@@ -877,8 +879,33 @@ def dumps(self):
     def to_dict(self):
         return dict(self.diff)
 
+    def _flatten_iterable_opcodes(self, _parse_path):
+        """
+        Converts op_codes to FlatDeltaRows
+        """
+        result = []
+        for path, op_codes in self.diff['_iterable_opcodes'].items():
+            for op_code in op_codes:
+                result.append(
+                    FlatDeltaRow(
+                        path=_parse_path(path),
+                        action=OPCODE_TAG_TO_FLAT_DATA_ACTION[op_code.tag],
+                        value=op_code.new_values,
+                        old_value=op_code.old_values,
+                        type=type(op_code.new_values),
+                        old_type=type(op_code.old_values),
+                        new_path=None,
+                        t1_from_index=op_code.t1_from_index,
+                        t1_to_index=op_code.t1_to_index,
+                        t2_from_index=op_code.t2_from_index,
+                        t2_to_index=op_code.t2_to_index,
+
+                    )
+                )
+        return result
+
     @staticmethod
-    def _get_flat_row(action, info, _parse_path, keys_and_funcs):
+    def _get_flat_row(action, info, _parse_path, keys_and_funcs, report_type_changes=True):
         for path, details in info.items():
             row = {'path': _parse_path(path), 'action': action}
             for key, new_key, func in keys_and_funcs:
@@ -887,6 +914,11 @@ def _get_flat_row(action, info, _parse_path, keys_and_funcs):
                         row[new_key] = func(details[key])
                     else:
                         row[new_key] = details[key]
+            if report_type_changes:
+                if 'value' in row and 'type' not in row:
+                    row['type'] = type(row['value'])
+                if 'old_value' in row and 'old_type' not in row:
+                    row['old_type'] = type(row['old_value'])
             yield FlatDeltaRow(**row)
 
     @staticmethod
@@ -918,28 +950,44 @@ def _from_flat_dicts(flat_dict_list):
             if action in FLATTENING_NEW_ACTION_MAP:
                 action = FLATTENING_NEW_ACTION_MAP[action]
                 index = path.pop()
-            if action in {'attribute_added', 'attribute_removed'}:
+            if action in {
+                FlatDataAction.attribute_added,
+                FlatDataAction.attribute_removed,
+            }:
                 root_element = ('root', GETATTR)
             else:
                 root_element = ('root', GET)
-            path_str = stringify_path(path, root_element=root_element)  # We need the string path
+            if isinstance(path, str):
+                path_str = path
+            else:
+                path_str = stringify_path(path, root_element=root_element)  # We need the string path
             if new_path and new_path != path:
                 new_path = stringify_path(new_path, root_element=root_element)
             else:
                 new_path = None
             if action not in result:
                 result[action] = {}
-            if action in {'iterable_items_added_at_indexes', 'iterable_items_removed_at_indexes'}:
+            if action in {
+                'iterable_items_added_at_indexes',
+                'iterable_items_removed_at_indexes',
+            }:
                 if path_str not in result[action]:
                     result[action][path_str] = {}
                 result[action][path_str][index] = value
-            elif action in {'set_item_added', 'set_item_removed'}:
+            elif action in {
+                FlatDataAction.set_item_added,
+                FlatDataAction.set_item_removed
+            }:
                 if path_str not in result[action]:
                     result[action][path_str] = set()
                 result[action][path_str].add(value)
             elif action in {
-                'dictionary_item_added', 'dictionary_item_removed',
-                'attribute_removed', 'attribute_added', 'iterable_item_added', 'iterable_item_removed',
+                FlatDataAction.dictionary_item_added,
+                FlatDataAction.dictionary_item_removed,
+                FlatDataAction.attribute_removed,
+                FlatDataAction.attribute_added,
+                FlatDataAction.iterable_item_added,
+                FlatDataAction.iterable_item_removed,
             }:
                 result[action][path_str] = value
             elif action == 'values_changed':
@@ -959,8 +1007,29 @@ def _from_flat_dicts(flat_dict_list):
                 ]:
                     if elem_value != UnkownValueCode:
                         result[action][path_str][elem] = elem_value
-            elif action == 'iterable_item_moved':
+            elif action == FlatDataAction.iterable_item_moved:
                 result[action][path_str] = {'value': value}
+            elif action in {
+                FlatDataAction.iterable_items_inserted,
+                FlatDataAction.iterable_items_deleted,
+                FlatDataAction.iterable_items_replaced,
+                FlatDataAction.iterable_items_equal,
+            }:
+                if '_iterable_opcodes' not in result:
+                    result['_iterable_opcodes'] = {}
+                if path_str not in result['_iterable_opcodes']:
+                    result['_iterable_opcodes'][path_str] = []
+                result['_iterable_opcodes'][path_str].append(
+                    Opcode(
+                        tag=FLAT_DATA_ACTION_TO_OPCODE_TAG[action],
+                        t1_from_index=flat_dict.get('t1_from_index'),
+                        t1_to_index=flat_dict.get('t1_to_index'),
+                        t2_from_index=flat_dict.get('t2_from_index'),
+                        t2_to_index=flat_dict.get('t2_to_index'),
+                        new_values=flat_dict.get('value'),
+                        old_values=flat_dict.get('old_value'),
+                    )
+                )
             if new_path:
                 result[action][path_str]['new_path'] = new_path
 
@@ -1060,6 +1129,9 @@ def to_flat_rows(self, include_action_in_path=False, report_type_changes=True) -
             'iterable_items_removed_at_indexes': 'unordered_iterable_item_removed',
         }
         for action, info in self.diff.items():
+            if action == '_iterable_opcodes':
+                result.extend(self._flatten_iterable_opcodes(_parse_path=_parse_path))
+                continue
             if action.startswith('_'):
                 continue
             if action in FLATTENING_NEW_ACTION_MAP:
@@ -1072,12 +1144,20 @@ def to_flat_rows(self, include_action_in_path=False, report_type_changes=True) -
                             path2.append((index, 'GET'))
                         else:
                             path2.append(index)
-                        result.append(FlatDeltaRow(path=path2, value=value, action=new_action))
+                        if report_type_changes:
+                            row = FlatDeltaRow(path=path2, value=value, action=new_action, type=type(value))
+                        else:
+                            row = FlatDeltaRow(path=path2, value=value, action=new_action)
+                        result.append(row)
             elif action in {'set_item_added', 'set_item_removed'}:
                 for path, values in info.items():
                     path = _parse_path(path)
                     for value in values:
-                        result.append(FlatDeltaRow(path=path, value=value, action=action))
+                        if report_type_changes:
+                            row = FlatDeltaRow(path=path, value=value, action=action, type=type(value))
+                        else:
+                            row = FlatDeltaRow(path=path, value=value, action=action)
+                        result.append(row)
             elif action == 'dictionary_item_added':
                 for path, value in info.items():
                     path = _parse_path(path)
@@ -1092,14 +1172,22 @@ def to_flat_rows(self, include_action_in_path=False, report_type_changes=True) -
                     elif isinstance(value, set) and len(value) == 1:
                         value = value.pop()
                         action = 'set_item_added'
-                    result.append(FlatDeltaRow(path=path, value=value, action=action))
+                    if report_type_changes:
+                        row = FlatDeltaRow(path=path, value=value, action=action, type=type(value))
+                    else:
+                        row = FlatDeltaRow(path=path, value=value, action=action)
+                    result.append(row)
             elif action in {
                 'dictionary_item_removed', 'iterable_item_added',
                 'iterable_item_removed', 'attribute_removed', 'attribute_added'
             }:
                 for path, value in info.items():
                     path = _parse_path(path)
-                    result.append(FlatDeltaRow(path=path, value=value, action=action))
+                    if report_type_changes:
+                        row = FlatDeltaRow(path=path, value=value, action=action, type=type(value))
+                    else:
+                        row = FlatDeltaRow(path=path, value=value, action=action)
+                    result.append(row)
             elif action == 'type_changes':
                 if not report_type_changes:
                     action = 'values_changed'
@@ -1109,16 +1197,16 @@ def to_flat_rows(self, include_action_in_path=False, report_type_changes=True) -
                     info=info,
                     _parse_path=_parse_path,
                     keys_and_funcs=keys_and_funcs,
+                    report_type_changes=report_type_changes,
                 ):
                     result.append(row)
-            elif action == '_iterable_opcodes':
-                result.extend(self._flatten_iterable_opcodes())
             else:
                 for row in self._get_flat_row(
                     action=action,
                     info=info,
                     _parse_path=_parse_path,
                     keys_and_funcs=keys_and_funcs,
+                    report_type_changes=report_type_changes,
                 ):
                     result.append(row)
         return result

deepdiff/helper.py

@@ -110,6 +110,17 @@ class pydantic_base_model_type:
 NUMERICS = frozenset(string.digits)
 
 
+class EnumBase(str, enum.Enum):
+    def __repr__(self):
+        """
+        We need to add a single quotes so we can easily copy the value when we do ipdb.
+        """
+        return f"'{self.name}'"
+
+    def __str__(self):
+        return self.name
+
+
 def _int_or_zero(value):
     """
     Tries to extract some number from a string.
@@ -739,6 +750,13 @@ def named_tuple_repr(self):
     return f"{self.__class__.__name__}({', '.join(fields)})"
 
 
+class OpcodeTag(EnumBase):
+    insert = 'insert'
+    delete = 'delete'
+    equal = 'equal'
+    replace = 'replace'
+
+
 class Opcode(NamedTuple):
     tag: str
     t1_from_index: int
@@ -751,7 +769,7 @@ class Opcode(NamedTuple):
     __repr__ = __str__ = named_tuple_repr
 
 
-class FlatDataAction(str, enum.Enum):
+class FlatDataAction(EnumBase):
     values_changed = 'values_changed'
     type_changes = 'type_changes'
     set_item_added = 'set_item_added'
@@ -771,7 +789,17 @@ class FlatDataAction(str, enum.Enum):
     unordered_iterable_item_removed = 'unordered_iterable_item_removed'
 
 
-UnkownValueCode = '*-UNKNOWN-*'
+OPCODE_TAG_TO_FLAT_DATA_ACTION = {
+    OpcodeTag.insert: FlatDataAction.iterable_items_inserted,
+    OpcodeTag.delete: FlatDataAction.iterable_items_deleted,
+    OpcodeTag.replace: FlatDataAction.iterable_items_replaced,
+    OpcodeTag.equal: FlatDataAction.iterable_items_equal,
+}
+
+FLAT_DATA_ACTION_TO_OPCODE_TAG = {v: i for i, v in OPCODE_TAG_TO_FLAT_DATA_ACTION.items()}
+
+
+UnkownValueCode = 'unknown___'
 
 
 class FlatDeltaRow(NamedTuple):

deepdiff/path.py

@@ -261,7 +261,8 @@ def parse_path(path, root_element=DEFAULT_FIRST_ELEMENT, include_actions=False):
 
     result = _path_to_elements(path, root_element=root_element)
     result = iter(result)
-    next(result)  # We don't want the root item
+    if root_element:
+        next(result)  # We don't want the root item
     if include_actions is False:
         return [i[0] for i in result]
     return [{'element': i[0], 'action': i[1]} for i in result]

docs/changelog.rst

@@ -5,6 +5,11 @@ Changelog
 
 DeepDiff Changelog
 
+
+-  v7-0-1
+
+   - Fixes the translation between Difflib opcodes and Delta flat rows.
+
 -  v7-0-0
 
    -  When verbose=2, return ``new_path`` when the ``path`` and

docs/conf.py

@@ -61,9 +61,9 @@
 # built documents.
 #
 # The short X.Y version.
-version = '7.0.0'
+version = '7.0.1'
 # The full version, including alpha/beta/rc tags.
-release = '7.0.0'
+release = '7.0.1'
 
 load_dotenv(override=True)
 DOC_VERSION = os.environ.get('DOC_VERSION', version)

docs/index.rst

@@ -4,7 +4,7 @@
    contain the root `toctree` directive.
 
 
-DeepDiff 7.0.0 documentation!
+DeepDiff 7.0.1 documentation!
 =============================
 
 *******