chore(core): add mypy pydantic plugin (#32604)

This helps to remove a bunch of mypy false positives.

libs/core/langchain_core/beta/runnables/context.py

@@ -253,7 +253,7 @@ class ContextSet(RunnableSerializable):
         """
         if key is not None:
             kwargs[key] = value
-        super().__init__(  # type: ignore[call-arg]
+        super().__init__(
             keys={
                 k: _coerce_set_value(v) if v is not None else None
                 for k, v in kwargs.items()

libs/core/langchain_core/documents/base.py

@@ -277,7 +277,7 @@ class Document(BaseMedia):
         """Pass page_content in as positional or named arg."""
         # my-py is complaining that page_content is not defined on the base class.
         # Here, we're relying on pydantic base class to handle the validation.
-        super().__init__(page_content=page_content, **kwargs)  # type: ignore[call-arg]
+        super().__init__(page_content=page_content, **kwargs)
 
     @classmethod
     def is_lc_serializable(cls) -> bool:

libs/core/langchain_core/language_models/chat_models.py

@@ -533,7 +533,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                     generations = [generations_with_error_metadata]
                 run_manager.on_llm_error(
                     e,
-                    response=LLMResult(generations=generations),  # type: ignore[arg-type]
+                    response=LLMResult(generations=generations),
                 )
                 raise
 
@@ -627,7 +627,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                 generations = [generations_with_error_metadata]
             await run_manager.on_llm_error(
                 e,
-                response=LLMResult(generations=generations),  # type: ignore[arg-type]
+                response=LLMResult(generations=generations),
             )
             raise
 
@@ -842,17 +842,17 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                     run_managers[i].on_llm_error(
                         e,
                         response=LLMResult(
-                            generations=[generations_with_error_metadata]  # type: ignore[list-item]
+                            generations=[generations_with_error_metadata]
                         ),
                     )
                 raise
         flattened_outputs = [
-            LLMResult(generations=[res.generations], llm_output=res.llm_output)  # type: ignore[list-item]
+            LLMResult(generations=[res.generations], llm_output=res.llm_output)
             for res in results
         ]
         llm_output = self._combine_llm_outputs([res.llm_output for res in results])
         generations = [res.generations for res in results]
-        output = LLMResult(generations=generations, llm_output=llm_output)  # type: ignore[arg-type]
+        output = LLMResult(generations=generations, llm_output=llm_output)
         if run_managers:
             run_infos = []
             for manager, flattened_output in zip(run_managers, flattened_outputs):
@@ -962,7 +962,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                     await run_managers[i].on_llm_error(
                         res,
                         response=LLMResult(
-                            generations=[generations_with_error_metadata]  # type: ignore[list-item]
+                            generations=[generations_with_error_metadata]
                         ),
                     )
                 exceptions.append(res)
@@ -972,7 +972,7 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                     *[
                         run_manager.on_llm_end(
                             LLMResult(
-                                generations=[res.generations],  # type: ignore[list-item, union-attr]
+                                generations=[res.generations],  # type: ignore[union-attr]
                                 llm_output=res.llm_output,  # type: ignore[union-attr]
                             )
                         )
@@ -982,12 +982,12 @@ class BaseChatModel(BaseLanguageModel[BaseMessage], ABC):
                 )
             raise exceptions[0]
         flattened_outputs = [
-            LLMResult(generations=[res.generations], llm_output=res.llm_output)  # type: ignore[list-item, union-attr]
+            LLMResult(generations=[res.generations], llm_output=res.llm_output)  # type: ignore[union-attr]
             for res in results
         ]
         llm_output = self._combine_llm_outputs([res.llm_output for res in results])  # type: ignore[union-attr]
         generations = [res.generations for res in results]  # type: ignore[union-attr]
-        output = LLMResult(generations=generations, llm_output=llm_output)  # type: ignore[arg-type]
+        output = LLMResult(generations=generations, llm_output=llm_output)
         await asyncio.gather(
             *[
                 run_manager.on_llm_end(flattened_output)

libs/core/langchain_core/prompts/chat.py

@@ -155,9 +155,7 @@ class MessagesPlaceholder(BaseMessagePromptTemplate):
         """
         # mypy can't detect the init which is defined in the parent class
         # b/c these are BaseModel classes.
-        super().__init__(  # type: ignore[call-arg]
-            variable_name=variable_name, optional=optional, **kwargs
-        )
+        super().__init__(variable_name=variable_name, optional=optional, **kwargs)
 
     def format_messages(self, **kwargs: Any) -> list[BaseMessage]:
         """Format messages from kwargs.

libs/core/langchain_core/runnables/base.py

@@ -2819,7 +2819,7 @@ class RunnableSequence(RunnableSerializable[Input, Output]):
         if len(steps_flat) < 2:
             msg = f"RunnableSequence must have at least 2 steps, got {len(steps_flat)}"
             raise ValueError(msg)
-        super().__init__(  # type: ignore[call-arg]
+        super().__init__(
             first=steps_flat[0],
             middle=list(steps_flat[1:-1]),
             last=steps_flat[-1],
@@ -3612,7 +3612,7 @@ class RunnableParallel(RunnableSerializable[Input, dict[str, Any]]):
         """
         merged = {**steps__} if steps__ is not None else {}
         merged.update(kwargs)
-        super().__init__(  # type: ignore[call-arg]
+        super().__init__(
             steps__={key: coerce_to_runnable(r) for key, r in merged.items()}
         )
 
@@ -5325,7 +5325,7 @@ class RunnableEach(RunnableEachBase[Input, Output]):
         )
 
 
-class RunnableBindingBase(RunnableSerializable[Input, Output]):
+class RunnableBindingBase(RunnableSerializable[Input, Output]):  # type: ignore[no-redef]
     """``Runnable`` that delegates calls to another ``Runnable`` with a set of kwargs.
 
     Use only if creating a new ``RunnableBinding`` subclass with different ``__init__``
@@ -5404,7 +5404,7 @@ class RunnableBindingBase(RunnableSerializable[Input, Output]):
                     ``Runnable`` with a custom type. Defaults to None.
             **other_kwargs: Unpacked into the base class.
         """  # noqa: E501
-        super().__init__(  # type: ignore[call-arg]
+        super().__init__(
             bound=bound,
             kwargs=kwargs or {},
             config=config or {},
@@ -5729,7 +5729,7 @@ class RunnableBindingBase(RunnableSerializable[Input, Output]):
             yield item
 
 
-class RunnableBinding(RunnableBindingBase[Input, Output]):
+class RunnableBinding(RunnableBindingBase[Input, Output]):  # type: ignore[no-redef]
     """Wrap a ``Runnable`` with additional functionality.
 
     A ``RunnableBinding`` can be thought of as a "runnable decorator" that

libs/core/langchain_core/runnables/branch.py

@@ -136,7 +136,7 @@ class RunnableBranch(RunnableSerializable[Input, Output]):
         super().__init__(
             branches=branches_,
             default=default_,
-        )  # type: ignore[call-arg]
+        )
 
     model_config = ConfigDict(
         arbitrary_types_allowed=True,

libs/core/langchain_core/runnables/history.py

@@ -38,7 +38,7 @@ MessagesOrDictWithMessages = Union[Sequence["BaseMessage"], dict[str, Any]]
 GetSessionHistoryCallable = Callable[..., BaseChatMessageHistory]
 
 
-class RunnableWithMessageHistory(RunnableBindingBase):
+class RunnableWithMessageHistory(RunnableBindingBase):  # type: ignore[no-redef]
     """Runnable that manages chat message history for another Runnable.
 
     A chat message history is a sequence of messages that represent a conversation.

libs/core/langchain_core/runnables/passthrough.py

@@ -186,7 +186,7 @@ class RunnablePassthrough(RunnableSerializable[Other, Other]):
             afunc = func
             func = None
 
-        super().__init__(func=func, afunc=afunc, input_type=input_type, **kwargs)  # type: ignore[call-arg]
+        super().__init__(func=func, afunc=afunc, input_type=input_type, **kwargs)
 
     @classmethod
     @override
@@ -406,7 +406,7 @@ class RunnableAssign(RunnableSerializable[dict[str, Any], dict[str, Any]]):
             mapper: A ``RunnableParallel`` instance that will be used to transform the
                 input dictionary.
         """
-        super().__init__(mapper=mapper, **kwargs)  # type: ignore[call-arg]
+        super().__init__(mapper=mapper, **kwargs)
 
     @classmethod
     @override
@@ -710,7 +710,7 @@ class RunnablePick(RunnableSerializable[dict[str, Any], dict[str, Any]]):
         Args:
             keys: A single key or a list of keys to pick from the input dictionary.
         """
-        super().__init__(keys=keys, **kwargs)  # type: ignore[call-arg]
+        super().__init__(keys=keys, **kwargs)
 
     @classmethod
     @override

libs/core/langchain_core/runnables/retry.py

@@ -47,7 +47,7 @@ class ExponentialJitterParams(TypedDict, total=False):
     """Random additional wait sampled from random.uniform(0, jitter)."""
 
 
-class RunnableRetry(RunnableBindingBase[Input, Output]):
+class RunnableRetry(RunnableBindingBase[Input, Output]):  # type: ignore[no-redef]
     """Retry a Runnable if it fails.
 
     RunnableRetry can be used to add retry logic to any object

libs/core/langchain_core/runnables/router.py

@@ -87,7 +87,7 @@ class RouterRunnable(RunnableSerializable[RouterInput, Output]):
         Args:
             runnables: A mapping of keys to Runnables.
         """
-        super().__init__(  # type: ignore[call-arg]
+        super().__init__(
             runnables={key: coerce_to_runnable(r) for key, r in runnables.items()}
         )
 

libs/core/langchain_core/structured_query.py

@@ -143,7 +143,7 @@ class Comparison(FilterDirective):
             value: The value to compare to.
         """
         # super exists from BaseModel
-        super().__init__(  # type: ignore[call-arg]
+        super().__init__(
             comparator=comparator, attribute=attribute, value=value, **kwargs
         )
 
@@ -166,9 +166,7 @@ class Operation(FilterDirective):
             arguments: The arguments to the operator.
         """
         # super exists from BaseModel
-        super().__init__(  # type: ignore[call-arg]
-            operator=operator, arguments=arguments, **kwargs
-        )
+        super().__init__(operator=operator, arguments=arguments, **kwargs)
 
 
 class StructuredQuery(Expr):
@@ -196,6 +194,4 @@ class StructuredQuery(Expr):
             limit: The limit on the number of results.
         """
         # super exists from BaseModel
-        super().__init__(  # type: ignore[call-arg]
-            query=query, filter=filter, limit=limit, **kwargs
-        )
+        super().__init__(query=query, filter=filter, limit=limit, **kwargs)

libs/core/langchain_core/tools/structured.py

@@ -228,7 +228,7 @@ class StructuredTool(BaseTool):
             name=name,
             func=func,
             coroutine=coroutine,
-            args_schema=args_schema,  # type: ignore[arg-type]
+            args_schema=args_schema,
             description=description_,
             return_direct=return_direct,
             response_format=response_format,

libs/core/pyproject.toml

@@ -64,6 +64,7 @@ langchain-text-splitters = { path = "../text-splitters" }
 
 
 [tool.mypy]
+plugins = ["pydantic.mypy"]
 strict = "True"
 strict_bytes = "True"
 enable_error_code = "deprecated"

libs/core/tests/unit_tests/output_parsers/test_openai_tools.py

@@ -810,7 +810,7 @@ def test_parse_with_different_pydantic_2_v1() -> None:
 
     # Can't get pydantic to work here due to the odd typing of tryig to support
     # both v1 and v2 in the same codebase.
-    parser = PydanticToolsParser(tools=[Forecast])  # type: ignore[list-item]
+    parser = PydanticToolsParser(tools=[Forecast])
     message = AIMessage(
         content="",
         tool_calls=[

libs/core/tests/unit_tests/output_parsers/test_pydantic_parser.py

@@ -13,7 +13,7 @@ from langchain_core.language_models import ParrotFakeChatModel
 from langchain_core.output_parsers import PydanticOutputParser
 from langchain_core.output_parsers.json import JsonOutputParser
 from langchain_core.prompts.prompt import PromptTemplate
-from langchain_core.utils.pydantic import TBaseModel
+from langchain_core.utils.pydantic import PydanticBaseModel, TBaseModel
 
 
 class ForecastV2(pydantic.BaseModel):
@@ -43,7 +43,7 @@ def test_pydantic_parser_chaining(
 
     model = ParrotFakeChatModel()
 
-    parser = PydanticOutputParser(pydantic_object=pydantic_object)  # type: ignore[type-var]
+    parser = PydanticOutputParser[PydanticBaseModel](pydantic_object=pydantic_object)
     chain = prompt | model | parser
 
     res = chain.invoke({})
@@ -66,7 +66,9 @@ def test_pydantic_parser_validation(pydantic_object: TBaseModel) -> None:
 
     model = ParrotFakeChatModel()
 
-    parser = PydanticOutputParser(pydantic_object=pydantic_object)  # type: ignore[arg-type,var-annotated]
+    parser: PydanticOutputParser[PydanticBaseModel] = PydanticOutputParser(
+        pydantic_object=pydantic_object
+    )
     chain = bad_prompt | model | parser
     with pytest.raises(OutputParserException):
         chain.invoke({})
@@ -88,7 +90,7 @@ def test_json_parser_chaining(
 
     model = ParrotFakeChatModel()
 
-    parser = JsonOutputParser(pydantic_object=pydantic_object)  # type: ignore[arg-type]
+    parser = JsonOutputParser(pydantic_object=pydantic_object)
     chain = prompt | model | parser
 
     res = chain.invoke({})
@@ -171,7 +173,7 @@ def test_pydantic_output_parser_type_inference() -> None:
 
     # Ignoring mypy error that appears in python 3.8, but not 3.11.
     # This seems to be functionally correct, so we'll ignore the error.
-    pydantic_parser = PydanticOutputParser(pydantic_object=SampleModel)
+    pydantic_parser = PydanticOutputParser[SampleModel](pydantic_object=SampleModel)
     schema = pydantic_parser.get_output_schema().model_json_schema()
 
     assert schema == {
@@ -202,5 +204,5 @@ def test_format_instructions_preserves_language() -> None:
             )
         )
 
-    parser = PydanticOutputParser(pydantic_object=Foo)
+    parser = PydanticOutputParser[Foo](pydantic_object=Foo)
     assert description in parser.get_format_instructions()

libs/core/tests/unit_tests/prompts/test_prompt.py

@@ -348,7 +348,7 @@ def test_prompt_invalid_template_format() -> None:
         PromptTemplate(
             input_variables=input_variables,
             template=template,
-            template_format="bar",  # type: ignore[arg-type]
+            template_format="bar",
         )
 
 

libs/core/tests/unit_tests/runnables/test_configurable.py

@@ -73,7 +73,7 @@ class MyOtherRunnable(RunnableSerializable[str, str]):
 
 def test_doubly_set_configurable() -> None:
     """Test that setting a configurable field with a default value works."""
-    runnable = MyRunnable(my_property="a")  # type: ignore[call-arg]
+    runnable = MyRunnable(my_property="a")
     configurable_runnable = runnable.configurable_fields(
         my_property=ConfigurableField(
             id="my_property",
@@ -86,7 +86,7 @@ def test_doubly_set_configurable() -> None:
 
 
 def test_alias_set_configurable() -> None:
-    runnable = MyRunnable(my_property="a")  # type: ignore[call-arg]
+    runnable = MyRunnable(my_property="a")
     configurable_runnable = runnable.configurable_fields(
         my_property=ConfigurableField(
             id="my_property_alias",
@@ -104,7 +104,7 @@ def test_alias_set_configurable() -> None:
 
 
 def test_field_alias_set_configurable() -> None:
-    runnable = MyRunnable(my_property_alias="a")
+    runnable = MyRunnable(my_property_alias="a")  # type: ignore[call-arg]
     configurable_runnable = runnable.configurable_fields(
         my_property=ConfigurableField(
             id="my_property",
@@ -122,7 +122,7 @@ def test_field_alias_set_configurable() -> None:
 
 
 def test_config_passthrough() -> None:
-    runnable = MyRunnable(my_property="a")  # type: ignore[call-arg]
+    runnable = MyRunnable(my_property="a")
     configurable_runnable = runnable.configurable_fields(
         my_property=ConfigurableField(
             id="my_property",
@@ -158,7 +158,7 @@ def test_config_passthrough() -> None:
 
 
 def test_config_passthrough_nested() -> None:
-    runnable = MyRunnable(my_property="a")  # type: ignore[call-arg]
+    runnable = MyRunnable(my_property="a")
     configurable_runnable = runnable.configurable_fields(
         my_property=ConfigurableField(
             id="my_property",

libs/core/tests/unit_tests/utils/test_utils.py

@@ -347,7 +347,7 @@ def test_using_secret_from_env_as_default_factory(
         secret: SecretStr = Field(default_factory=secret_from_env("TEST_KEY"))
 
     # Pass the secret as a parameter
-    foo = Foo(secret="super_secret")  # type: ignore[arg-type]
+    foo = Foo(secret="super_secret")
     assert foo.secret.get_secret_value() == "super_secret"
 
     # Set the environment variable