Better-support mypy strict equality flag (#8799)

pydantic/main.py

@@ -867,60 +867,62 @@ def __setstate__(self, state: dict[Any, Any]) -> None:
         _object_setattr(self, '__pydantic_private__', state['__pydantic_private__'])
         _object_setattr(self, '__dict__', state['__dict__'])
 
-    def __eq__(self, other: Any) -> bool:
-        if isinstance(other, BaseModel):
-            # When comparing instances of generic types for equality, as long as all field values are equal,
-            # only require their generic origin types to be equal, rather than exact type equality.
-            # This prevents headaches like MyGeneric(x=1) != MyGeneric[Any](x=1).
-            self_type = self.__pydantic_generic_metadata__['origin'] or self.__class__
-            other_type = other.__pydantic_generic_metadata__['origin'] or other.__class__
-
-            # Perform common checks first
-            if not (
-                self_type == other_type
-                and self.__pydantic_private__ == other.__pydantic_private__
-                and self.__pydantic_extra__ == other.__pydantic_extra__
-            ):
-                return False
-
-            # We only want to compare pydantic fields but ignoring fields is costly.
-            # We'll perform a fast check first, and fallback only when needed
-            # See GH-7444 and GH-7825 for rationale and a performance benchmark
-
-            # First, do the fast (and sometimes faulty) __dict__ comparison
-            if self.__dict__ == other.__dict__:
-                # If the check above passes, then pydantic fields are equal, we can return early
-                return True
-
-            # We don't want to trigger unnecessary costly filtering of __dict__ on all unequal objects, so we return
-            # early if there are no keys to ignore (we would just return False later on anyway)
-            model_fields = type(self).model_fields.keys()
-            if self.__dict__.keys() <= model_fields and other.__dict__.keys() <= model_fields:
-                return False
-
-            # If we reach here, there are non-pydantic-fields keys, mapped to unequal values, that we need to ignore
-            # Resort to costly filtering of the __dict__ objects
-            # We use operator.itemgetter because it is much faster than dict comprehensions
-            # NOTE: Contrary to standard python class and instances, when the Model class has a default value for an
-            # attribute and the model instance doesn't have a corresponding attribute, accessing the missing attribute
-            # raises an error in BaseModel.__getattr__ instead of returning the class attribute
-            # So we can use operator.itemgetter() instead of operator.attrgetter()
-            getter = operator.itemgetter(*model_fields) if model_fields else lambda _: _utils._SENTINEL
-            try:
-                return getter(self.__dict__) == getter(other.__dict__)
-            except KeyError:
-                # In rare cases (such as when using the deprecated BaseModel.copy() method),
-                # the __dict__ may not contain all model fields, which is how we can get here.
-                # getter(self.__dict__) is much faster than any 'safe' method that accounts
-                # for missing keys, and wrapping it in a `try` doesn't slow things down much
-                # in the common case.
-                self_fields_proxy = _utils.SafeGetItemProxy(self.__dict__)
-                other_fields_proxy = _utils.SafeGetItemProxy(other.__dict__)
-                return getter(self_fields_proxy) == getter(other_fields_proxy)
-
-        # other instance is not a BaseModel
-        else:
-            return NotImplemented  # delegate to the other item in the comparison
+    if not typing.TYPE_CHECKING:
+
+        def __eq__(self, other: Any) -> bool:
+            if isinstance(other, BaseModel):
+                # When comparing instances of generic types for equality, as long as all field values are equal,
+                # only require their generic origin types to be equal, rather than exact type equality.
+                # This prevents headaches like MyGeneric(x=1) != MyGeneric[Any](x=1).
+                self_type = self.__pydantic_generic_metadata__['origin'] or self.__class__
+                other_type = other.__pydantic_generic_metadata__['origin'] or other.__class__
+
+                # Perform common checks first
+                if not (
+                    self_type == other_type
+                    and self.__pydantic_private__ == other.__pydantic_private__
+                    and self.__pydantic_extra__ == other.__pydantic_extra__
+                ):
+                    return False
+
+                # We only want to compare pydantic fields but ignoring fields is costly.
+                # We'll perform a fast check first, and fallback only when needed
+                # See GH-7444 and GH-7825 for rationale and a performance benchmark
+
+                # First, do the fast (and sometimes faulty) __dict__ comparison
+                if self.__dict__ == other.__dict__:
+                    # If the check above passes, then pydantic fields are equal, we can return early
+                    return True
+
+                # We don't want to trigger unnecessary costly filtering of __dict__ on all unequal objects, so we return
+                # early if there are no keys to ignore (we would just return False later on anyway)
+                model_fields = type(self).model_fields.keys()
+                if self.__dict__.keys() <= model_fields and other.__dict__.keys() <= model_fields:
+                    return False
+
+                # If we reach here, there are non-pydantic-fields keys, mapped to unequal values, that we need to ignore
+                # Resort to costly filtering of the __dict__ objects
+                # We use operator.itemgetter because it is much faster than dict comprehensions
+                # NOTE: Contrary to standard python class and instances, when the Model class has a default value for an
+                # attribute and the model instance doesn't have a corresponding attribute, accessing the missing attribute
+                # raises an error in BaseModel.__getattr__ instead of returning the class attribute
+                # So we can use operator.itemgetter() instead of operator.attrgetter()
+                getter = operator.itemgetter(*model_fields) if model_fields else lambda _: _utils._SENTINEL
+                try:
+                    return getter(self.__dict__) == getter(other.__dict__)
+                except KeyError:
+                    # In rare cases (such as when using the deprecated BaseModel.copy() method),
+                    # the __dict__ may not contain all model fields, which is how we can get here.
+                    # getter(self.__dict__) is much faster than any 'safe' method that accounts
+                    # for missing keys, and wrapping it in a `try` doesn't slow things down much
+                    # in the common case.
+                    self_fields_proxy = _utils.SafeGetItemProxy(self.__dict__)
+                    other_fields_proxy = _utils.SafeGetItemProxy(other.__dict__)
+                    return getter(self_fields_proxy) == getter(other_fields_proxy)
+
+            # other instance is not a BaseModel
+            else:
+                return NotImplemented  # delegate to the other item in the comparison
 
     if typing.TYPE_CHECKING:
         # We put `__init_subclass__` in a TYPE_CHECKING block because, even though we want the type-checking benefits

tests/mypy/configs/pyproject-plugin-strict-equality.toml

@@ -0,0 +1,48 @@
+[build-system]
+requires = ["poetry>=0.12"]
+build_backend = "poetry.masonry.api"
+
+[tool.poetry]
+name = "test"
+version = "0.0.1"
+readme = "README.md"
+authors = [
+    "author@example.com"
+]
+
+[tool.poetry.dependencies]
+python = "*"
+
+[tool.pytest.ini_options]
+addopts = "-v -p no:warnings"
+
+[tool.mypy]
+plugins = "pydantic.mypy"
+ignore_missing_imports = true
+warn_return_any = true
+warn_unreachable = true
+warn_unused_configs = true
+follow_imports = "normal"
+show_column_numbers = true
+strict_optional = true
+warn_redundant_casts = true
+pretty = false
+strict = true
+warn_unused_ignores = true
+check_untyped_defs = true
+disallow_untyped_calls = true
+disallow_untyped_defs = true
+disallow_untyped_decorators = false
+strict_equality = true
+
+[tool.pydantic-mypy]
+init_forbid_extra = true
+init_typed = true
+warn_required_dynamic_aliases = true
+
+
+[[tool.mypy.overrides]]
+module = [
+    'pydantic_core.*',
+]
+follow_imports = "skip"

tests/mypy/modules/strict_equality.py

@@ -0,0 +1,11 @@
+from pydantic import BaseModel
+
+
+class User(BaseModel):
+    username: str
+
+
+user = User(username='test')
+print(user == 'test')
+print(user.username == int('1'))
+print(user.username == 'test')

tests/mypy/outputs/1.0.1/pyproject-plugin-strict-equality_toml/strict_equality.py

@@ -0,0 +1,13 @@
+from pydantic import BaseModel
+
+
+class User(BaseModel):
+    username: str
+
+
+user = User(username='test')
+print(user == 'test')
+# MYPY: error: Non-overlapping equality check (left operand type: "User", right operand type: "Literal['test']")  [comparison-overlap]
+print(user.username == int('1'))
+# MYPY: error: Non-overlapping equality check (left operand type: "str", right operand type: "int")  [comparison-overlap]
+print(user.username == 'test')

tests/mypy/test_mypy.py

@@ -112,6 +112,7 @@ def build(self) -> List[Union[Tuple[str, str], Any]]:
         ('pyproject-default.toml', 'computed_fields.py'),
         ('pyproject-default.toml', 'with_config_decorator.py'),
         ('pyproject-plugin-no-strict-optional.toml', 'no_strict_optional.py'),
+        ('pyproject-plugin-strict-equality.toml', 'strict_equality.py'),
     ]
 )
 
@@ -192,6 +193,8 @@ def test_mypy_results(config_filename: str, python_filename: str, request: pytes
     if test_config.existing is not None:
         existing_output_code = test_config.existing.output_path.read_text()
         print(f'Comparing output with {test_config.existing.output_path}')
+    else:
+        print('Expecting no mypy errors')
 
     merged_output = merge_python_and_mypy_output(input_code, mypy_out)
 
@@ -202,6 +205,9 @@ def test_mypy_results(config_filename: str, python_filename: str, request: pytes
         test_config.current.output_path.parent.mkdir(parents=True, exist_ok=True)
         test_config.current.output_path.write_text(merged_output)
     else:
+        print('**** Merged Output ****')
+        print(merged_output)
+        print('***********************')
         assert existing_output_code is not None, 'No output file found, run `make test-mypy-update` to create it'
         assert merged_output == existing_output_code
         expected_returncode = get_expected_return_code(existing_output_code)
@@ -279,7 +285,7 @@ def merge_python_and_mypy_output(source_code: str, mypy_output: str) -> str:
         if not line:
             continue
         try:
-            line_number, message = line.split(':', maxsplit=1)
+            line_number, message = re.split(r':(?:\d+:)?', line, maxsplit=1)
             merged_lines.insert(int(line_number), (f'# MYPY: {message.strip()}', True))
         except ValueError:
             # This could happen due to lack of a ':' in `line`, or the pre-':' contents not being a number