core: Improve mypy config

Author: cbornet

libs/core/langchain_core/document_loaders/base.py

@@ -80,7 +80,7 @@ async def alazy_load(self) -> AsyncIterator[Document]:
         iterator = await run_in_executor(None, self.lazy_load)
         done = object()
         while True:
-            doc = await run_in_executor(None, next, iterator, done)  # type: ignore[call-arg, arg-type]
+            doc = await run_in_executor(None, next, iterator, done)
             if doc is done:
                 break
             yield doc  # type: ignore[misc]

libs/core/langchain_core/embeddings/fake.py

@@ -52,7 +52,7 @@ class FakeEmbeddings(Embeddings, BaseModel):
     """The size of the embedding vector."""
 
     def _get_embedding(self) -> list[float]:
-        import numpy as np  # type: ignore[import-not-found, import-untyped]
+        import numpy as np
 
         return list(np.random.default_rng().normal(size=self.size))
 
@@ -109,7 +109,7 @@ class DeterministicFakeEmbedding(Embeddings, BaseModel):
     """The size of the embedding vector."""
 
     def _get_embedding(self, seed: int) -> list[float]:
-        import numpy as np  # type: ignore[import-not-found, import-untyped]
+        import numpy as np
 
         # set the seed for the random generator
         rng = np.random.default_rng(seed)

libs/core/langchain_core/globals.py

@@ -23,7 +23,7 @@ def set_verbose(value: bool) -> None:
         value: The new value for the `verbose` global setting.
     """
     try:
-        import langchain  # type: ignore[import]
+        import langchain  # type: ignore[import-not-found]
 
         # We're about to run some deprecated code, don't report warnings from it.
         # The user called the correct (non-deprecated) code path and shouldn't get
@@ -57,7 +57,7 @@ def get_verbose() -> bool:
         The value of the `verbose` global setting.
     """
     try:
-        import langchain  # type: ignore[import]
+        import langchain
 
         # We're about to run some deprecated code, don't report warnings from it.
         # The user called the correct (non-deprecated) code path and shouldn't get
@@ -96,7 +96,7 @@ def set_debug(value: bool) -> None:
         value: The new value for the `debug` global setting.
     """
     try:
-        import langchain  # type: ignore[import]
+        import langchain
 
         # We're about to run some deprecated code, don't report warnings from it.
         # The user called the correct (non-deprecated) code path and shouldn't get
@@ -128,7 +128,7 @@ def get_debug() -> bool:
         The value of the `debug` global setting.
     """
     try:
-        import langchain  # type: ignore[import]
+        import langchain
 
         # We're about to run some deprecated code, don't report warnings from it.
         # The user called the correct (non-deprecated) code path and shouldn't get
@@ -164,7 +164,7 @@ def set_llm_cache(value: Optional["BaseCache"]) -> None:
         value: The new LLM cache to use. If `None`, the LLM cache is disabled.
     """
     try:
-        import langchain  # type: ignore[import]
+        import langchain
 
         # We're about to run some deprecated code, don't report warnings from it.
         # The user called the correct (non-deprecated) code path and shouldn't get
@@ -198,7 +198,7 @@ def get_llm_cache() -> "BaseCache":
         The value of the `llm_cache` global setting.
     """
     try:
-        import langchain  # type: ignore[import]
+        import langchain
 
         # We're about to run some deprecated code, don't report warnings from it.
         # The user called the correct (non-deprecated) code path and shouldn't get

libs/core/langchain_core/indexing/api.py

@@ -394,7 +394,7 @@ def index(
                 if cleanup == "scoped_full":
                     scoped_full_cleanup_source_ids.add(source_id)
             # source ids cannot be None after for loop above.
-            source_ids = cast("Sequence[str]", source_ids)  # type: ignore[assignment]
+            source_ids = cast("Sequence[str]", source_ids)
 
         exists_batch = record_manager.exists([doc.uid for doc in hashed_docs])
 

libs/core/langchain_core/language_models/base.py

@@ -61,7 +61,7 @@ def get_tokenizer() -> Any:
     every time it is called.
     """
     try:
-        from transformers import GPT2TokenizerFast  # type: ignore[import]
+        from transformers import GPT2TokenizerFast  # type: ignore[import-not-found]
     except ImportError as e:
         msg = (
             "Could not import transformers python package. "

libs/core/langchain_core/language_models/chat_models.py

@@ -853,7 +853,7 @@ async def agenerate(
                         run_manager.on_llm_end(
                             LLMResult(
                                 generations=[res.generations],  # type: ignore[list-item, union-attr]
-                                llm_output=res.llm_output,  # type: ignore[list-item, union-attr]
+                                llm_output=res.llm_output,  # type: ignore[union-attr]
                             )
                         )
                         for run_manager, res in zip(run_managers, results)
@@ -1109,7 +1109,7 @@ async def _astream(
                 None,
                 next,
                 iterator,
-                done,  # type: ignore[call-arg, arg-type]
+                done,
             )
             if item is done:
                 break

libs/core/langchain_core/language_models/llms.py

@@ -455,7 +455,7 @@ def batch(
                 inputs[i : i + max_concurrency]
                 for i in range(0, len(inputs), max_concurrency)
             ]
-            config = [{**c, "max_concurrency": None} for c in config]  # type: ignore[misc]
+            config = [{**c, "max_concurrency": None} for c in config]
             return [
                 output
                 for i, batch in enumerate(batches)
@@ -501,7 +501,7 @@ async def abatch(
                 inputs[i : i + max_concurrency]
                 for i in range(0, len(inputs), max_concurrency)
             ]
-            config = [{**c, "max_concurrency": None} for c in config]  # type: ignore[misc]
+            config = [{**c, "max_concurrency": None} for c in config]
             return [
                 output
                 for i, batch in enumerate(batches)
@@ -746,7 +746,7 @@ async def _astream(
                 None,
                 next,
                 iterator,
-                done,  # type: ignore[call-arg, arg-type]
+                done,
             )
             if item is done:
                 break
@@ -1231,7 +1231,7 @@ async def agenerate(
                 stop,
                 run_managers,  # type: ignore[arg-type]
                 new_arg_supported=bool(new_arg_supported),
-                **kwargs,  # type: ignore[arg-type]
+                **kwargs,
             )
         if len(missing_prompts) > 0:
             run_managers = await asyncio.gather(
@@ -1253,7 +1253,7 @@ async def agenerate(
                 stop,
                 run_managers,  # type: ignore[arg-type]
                 new_arg_supported=bool(new_arg_supported),
-                **kwargs,  # type: ignore[arg-type]
+                **kwargs,
             )
             llm_output = await aupdate_cache(
                 self.cache,

libs/core/langchain_core/messages/ai.py

@@ -202,7 +202,7 @@ def _backwards_compat_tool_calls(cls, values: dict) -> Any:
             raw_tool_calls := values.get("additional_kwargs", {}).get("tool_calls")
         ):
             try:
-                if issubclass(cls, AIMessageChunk):  # type: ignore
+                if issubclass(cls, AIMessageChunk):
                     values["tool_call_chunks"] = default_tool_chunk_parser(
                         raw_tool_calls
                     )

libs/core/langchain_core/messages/base.py

@@ -119,7 +119,7 @@ def __add__(self, other: Any) -> ChatPromptTemplate:
         """Concatenate this message with another message."""
         from langchain_core.prompts.chat import ChatPromptTemplate
 
-        prompt = ChatPromptTemplate(messages=[self])  # type: ignore[call-arg]
+        prompt = ChatPromptTemplate(messages=[self])
         return prompt + other
 
     def pretty_repr(self, html: bool = False) -> str:
@@ -213,7 +213,7 @@ def __add__(self, other: Any) -> BaseMessageChunk:  # type: ignore
             # If both are (subclasses of) BaseMessageChunk,
             # concat into a single BaseMessageChunk
 
-            return self.__class__(  # type: ignore[call-arg]
+            return self.__class__(
                 id=self.id,
                 type=self.type,
                 content=merge_content(self.content, other.content),

libs/core/langchain_core/messages/utils.py

@@ -237,7 +237,7 @@ def _create_message_from_message_type(
     if additional_kwargs:
         if response_metadata := additional_kwargs.pop("response_metadata", None):
             kwargs["response_metadata"] = response_metadata
-        kwargs["additional_kwargs"] = additional_kwargs  # type: ignore[assignment]
+        kwargs["additional_kwargs"] = additional_kwargs
         additional_kwargs.update(additional_kwargs.pop("additional_kwargs", {}))
     if id is not None:
         kwargs["id"] = id
@@ -891,7 +891,7 @@ def list_token_counter(messages: Sequence[BaseMessage]) -> int:
                 return sum(token_counter(msg) for msg in messages)  # type: ignore[arg-type, misc]
 
         else:
-            list_token_counter = token_counter  # type: ignore[assignment]
+            list_token_counter = token_counter
     else:
         msg = (
             f"'token_counter' expected to be a model that implements "

libs/core/langchain_core/output_parsers/json.py

@@ -6,7 +6,7 @@
 from json import JSONDecodeError
 from typing import Annotated, Any, Optional, TypeVar, Union
 
-import jsonpatch  # type: ignore[import]
+import jsonpatch  # type: ignore[import-untyped]
 import pydantic
 from pydantic import SkipValidation
 

libs/core/langchain_core/output_parsers/openai_functions.py

@@ -5,7 +5,7 @@
 from types import GenericAlias
 from typing import Any, Optional, Union
 
-import jsonpatch  # type: ignore[import]
+import jsonpatch  # type: ignore[import-untyped]
 from pydantic import BaseModel, model_validator
 
 from langchain_core.exceptions import OutputParserException
@@ -275,9 +275,9 @@ def parse_result(self, result: list[Generation], *, partial: bool = False) -> An
             else:
                 pydantic_schema = self.pydantic_schema
             if hasattr(pydantic_schema, "model_validate_json"):
-                pydantic_args = pydantic_schema.model_validate_json(_args)  # type: ignore
+                pydantic_args = pydantic_schema.model_validate_json(_args)
             else:
-                pydantic_args = pydantic_schema.parse_raw(_args)  # type: ignore
+                pydantic_args = pydantic_schema.parse_raw(_args)
         return pydantic_args
 
 

libs/core/langchain_core/output_parsers/pydantic.py

@@ -20,7 +20,7 @@
 class PydanticOutputParser(JsonOutputParser, Generic[TBaseModel]):
     """Parse an output using a pydantic model."""
 
-    pydantic_object: Annotated[type[TBaseModel], SkipValidation()]  # type: ignore
+    pydantic_object: Annotated[type[TBaseModel], SkipValidation()]
     """The pydantic model to parse."""
 
     def _parse_obj(self, obj: dict) -> TBaseModel:

libs/core/langchain_core/output_parsers/xml.py

@@ -205,7 +205,7 @@ def parse(self, text: str) -> dict[str, Union[str, list[Any]]]:
         # likely if you're reading this you can move them to the top of the file
         if self.parser == "defusedxml":
             try:
-                from defusedxml import ElementTree  # type: ignore
+                from defusedxml import ElementTree
             except ImportError as e:
                 msg = (
                     "defusedxml is not installed. "

libs/core/langchain_core/outputs/llm_result.py

@@ -50,7 +50,7 @@ class LLMResult(BaseModel):
     run: Optional[list[RunInfo]] = None
     """List of metadata info for model call for each input."""
 
-    type: Literal["LLMResult"] = "LLMResult"  # type: ignore[assignment]
+    type: Literal["LLMResult"] = "LLMResult"
     """Type is used exclusively for serialization purposes."""
 
     def flatten(self) -> list[LLMResult]:

libs/core/langchain_core/prompts/chat.py

@@ -126,7 +126,7 @@ def __add__(self, other: Any) -> ChatPromptTemplate:
         Returns:
             Combined prompt template.
         """
-        prompt = ChatPromptTemplate(messages=[self])  # type: ignore[call-arg]
+        prompt = ChatPromptTemplate(messages=[self])
         return prompt + other
 
 
@@ -1024,23 +1024,23 @@ def __add__(self, other: Any) -> ChatPromptTemplate:
         if isinstance(other, ChatPromptTemplate):
             return ChatPromptTemplate(messages=self.messages + other.messages).partial(
                 **partials
-            )  # type: ignore[call-arg]
+            )
         if isinstance(
             other, (BaseMessagePromptTemplate, BaseMessage, BaseChatPromptTemplate)
         ):
             return ChatPromptTemplate(messages=self.messages + [other]).partial(
                 **partials
-            )  # type: ignore[call-arg]
+            )
         if isinstance(other, (list, tuple)):
             _other = ChatPromptTemplate.from_messages(other)
             return ChatPromptTemplate(messages=self.messages + _other.messages).partial(
                 **partials
-            )  # type: ignore[call-arg]
+            )
         if isinstance(other, str):
             prompt = HumanMessagePromptTemplate.from_template(other)
             return ChatPromptTemplate(messages=self.messages + [prompt]).partial(
                 **partials
-            )  # type: ignore[call-arg]
+            )
         msg = f"Unsupported operand type for +: {type(other)}"
         raise NotImplementedError(msg)
 
@@ -1129,7 +1129,7 @@ def from_role_strings(
         Returns:
             a chat prompt template.
         """
-        return cls(  # type: ignore[call-arg]
+        return cls(
             messages=[
                 ChatMessagePromptTemplate.from_template(template, role=role)
                 for role, template in string_messages

libs/core/langchain_core/prompts/image.py

@@ -116,7 +116,7 @@ def format(
         output: ImageURL = {"url": url}
         if detail:
             # Don't check literal values here: let the API check them
-            output["detail"] = detail  # type: ignore[typeddict-item]
+            output["detail"] = detail
         return output
 
     async def aformat(self, **kwargs: Any) -> ImageURL:

libs/core/langchain_core/prompts/prompt.py

@@ -297,7 +297,7 @@ def from_template(
         return cls(
             input_variables=input_variables,
             template=template,
-            template_format=template_format,  # type: ignore[arg-type]
+            template_format=template_format,
             partial_variables=_partial_variables,
             **kwargs,
         )

libs/core/langchain_core/retrievers.py

@@ -161,10 +161,10 @@ def __init_subclass__(cls, **kwargs: Any) -> None:
                 stacklevel=4,
             )
             swap = cls.get_relevant_documents
-            cls.get_relevant_documents = (  # type: ignore[assignment]
+            cls.get_relevant_documents = (  # type: ignore[method-assign]
                 BaseRetriever.get_relevant_documents
             )
-            cls._get_relevant_documents = swap  # type: ignore[assignment]
+            cls._get_relevant_documents = swap  # type: ignore[method-assign]
         if (
             hasattr(cls, "aget_relevant_documents")
             and cls.aget_relevant_documents != BaseRetriever.aget_relevant_documents
@@ -176,10 +176,10 @@ def __init_subclass__(cls, **kwargs: Any) -> None:
                 stacklevel=4,
             )
             aswap = cls.aget_relevant_documents
-            cls.aget_relevant_documents = (  # type: ignore[assignment]
+            cls.aget_relevant_documents = (  # type: ignore[method-assign]
                 BaseRetriever.aget_relevant_documents
             )
-            cls._aget_relevant_documents = aswap  # type: ignore[assignment]
+            cls._aget_relevant_documents = aswap  # type: ignore[method-assign]
         parameters = signature(cls._get_relevant_documents).parameters
         cls._new_arg_supported = parameters.get("run_manager") is not None
         if (