core: Ruff preview fixes (#31877)

Auto-fixes from `uv run ruff check --fix --unsafe-fixes --preview`

---------

Co-authored-by: Mason Daugherty <[email protected]>

Author: cbornet

libs/core/langchain_core/_api/deprecation.py

@@ -469,7 +469,7 @@ def warn_deprecated(
 
     if not message:
         message = ""
-        _package = (
+        package_ = (
             package or name.split(".")[0].replace("_", "-")
             if "." in name
             else "LangChain"
@@ -483,14 +483,14 @@ def warn_deprecated(
         if pending:
             message += " will be deprecated in a future version"
         else:
-            message += f" was deprecated in {_package} {since}"
+            message += f" was deprecated in {package_} {since}"
 
             if removal:
                 message += f" and will be removed {removal}"
 
         if alternative_import:
             alt_package = alternative_import.split(".")[0].replace("_", "-")
-            if alt_package == _package:
+            if alt_package == package_:
                 message += f". Use {alternative_import} instead."
             else:
                 alt_module, alt_name = alternative_import.rsplit(".", 1)

libs/core/langchain_core/caches.py

@@ -194,7 +194,7 @@ def update(self, prompt: str, llm_string: str, return_val: RETURN_VAL_TYPE) -> N
         """
         if self._maxsize is not None and len(self._cache) == self._maxsize:
             del self._cache[next(iter(self._cache))]
-        self._cache[(prompt, llm_string)] = return_val
+        self._cache[prompt, llm_string] = return_val
 
     @override
     def clear(self, **kwargs: Any) -> None:

libs/core/langchain_core/document_loaders/base.py

@@ -60,11 +60,11 @@ def load_and_split(
                 )
                 raise ImportError(msg) from e
 
-            _text_splitter: TextSplitter = RecursiveCharacterTextSplitter()
+            text_splitter_: TextSplitter = RecursiveCharacterTextSplitter()
         else:
-            _text_splitter = text_splitter
+            text_splitter_ = text_splitter
         docs = self.load()
-        return _text_splitter.split_documents(docs)
+        return text_splitter_.split_documents(docs)
 
     # Attention: This method will be upgraded into an abstractmethod once it's
     #            implemented in all the existing subclasses.

libs/core/langchain_core/documents/base.py

@@ -201,14 +201,14 @@ def from_path(
             Blob instance
         """
         if mime_type is None and guess_type:
-            _mimetype = mimetypes.guess_type(path)[0] if guess_type else None
+            mimetype = mimetypes.guess_type(path)[0] if guess_type else None
         else:
-            _mimetype = mime_type
+            mimetype = mime_type
         # We do not load the data immediately, instead we treat the blob as a
         # reference to the underlying data.
         return cls(
             data=None,
-            mimetype=_mimetype,
+            mimetype=mimetype,
             encoding=encoding,
             path=path,
             metadata=metadata if metadata is not None else {},

libs/core/langchain_core/indexing/api.py

@@ -273,7 +273,7 @@ def index(
     vector_store: Union[VectorStore, DocumentIndex],
     *,
     batch_size: int = 100,
-    cleanup: Literal["incremental", "full", "scoped_full", None] = None,
+    cleanup: Optional[Literal["incremental", "full", "scoped_full"]] = None,
     source_id_key: Union[str, Callable[[Document], str], None] = None,
     cleanup_batch_size: int = 1_000,
     force_update: bool = False,
@@ -540,10 +540,10 @@ def index(
                     )
                     raise AssertionError(msg)
 
-            _source_ids = cast("Sequence[str]", source_ids)
+            source_ids_ = cast("Sequence[str]", source_ids)
 
             while uids_to_delete := record_manager.list_keys(
-                group_ids=_source_ids, before=index_start_dt, limit=cleanup_batch_size
+                group_ids=source_ids_, before=index_start_dt, limit=cleanup_batch_size
             ):
                 # Then delete from vector store.
                 _delete(destination, uids_to_delete)
@@ -609,7 +609,7 @@ async def aindex(
     vector_store: Union[VectorStore, DocumentIndex],
     *,
     batch_size: int = 100,
-    cleanup: Literal["incremental", "full", "scoped_full", None] = None,
+    cleanup: Optional[Literal["incremental", "full", "scoped_full"]] = None,
     source_id_key: Union[str, Callable[[Document], str], None] = None,
     cleanup_batch_size: int = 1_000,
     force_update: bool = False,
@@ -881,10 +881,10 @@ async def aindex(
                     )
                     raise AssertionError(msg)
 
-            _source_ids = cast("Sequence[str]", source_ids)
+            source_ids_ = cast("Sequence[str]", source_ids)
 
             while uids_to_delete := await record_manager.alist_keys(
-                group_ids=_source_ids, before=index_start_dt, limit=cleanup_batch_size
+                group_ids=source_ids_, before=index_start_dt, limit=cleanup_batch_size
             ):
                 # Then delete from vector store.
                 await _adelete(destination, uids_to_delete)

libs/core/langchain_core/language_models/_utils.py

@@ -123,7 +123,7 @@ def _normalize_messages(messages: Sequence[BaseMessage]) -> list[BaseMessage]:
                     # Subset to (PDF) files and audio, as most relevant chat models
                     # support images in OAI format (and some may not yet support the
                     # standard data block format)
-                    and block.get("type") in ("file", "input_audio")
+                    and block.get("type") in {"file", "input_audio"}
                     and _is_openai_data_block(block)
                 ):
                     if formatted_message is message:

libs/core/langchain_core/language_models/chat_models.py

@@ -1263,8 +1263,8 @@ def predict(
         Returns:
             The predicted output string.
         """
-        _stop = None if stop is None else list(stop)
-        result = self([HumanMessage(content=text)], stop=_stop, **kwargs)
+        stop_ = None if stop is None else list(stop)
+        result = self([HumanMessage(content=text)], stop=stop_, **kwargs)
         if isinstance(result.content, str):
             return result.content
         msg = "Cannot use predict when output is not a string."
@@ -1279,17 +1279,17 @@ def predict_messages(
         stop: Optional[Sequence[str]] = None,
         **kwargs: Any,
     ) -> BaseMessage:
-        _stop = None if stop is None else list(stop)
-        return self(messages, stop=_stop, **kwargs)
+        stop_ = None if stop is None else list(stop)
+        return self(messages, stop=stop_, **kwargs)
 
     @deprecated("0.1.7", alternative="ainvoke", removal="1.0")
     @override
     async def apredict(
         self, text: str, *, stop: Optional[Sequence[str]] = None, **kwargs: Any
     ) -> str:
-        _stop = None if stop is None else list(stop)
+        stop_ = None if stop is None else list(stop)
         result = await self._call_async(
-            [HumanMessage(content=text)], stop=_stop, **kwargs
+            [HumanMessage(content=text)], stop=stop_, **kwargs
         )
         if isinstance(result.content, str):
             return result.content
@@ -1305,8 +1305,8 @@ async def apredict_messages(
         stop: Optional[Sequence[str]] = None,
         **kwargs: Any,
     ) -> BaseMessage:
-        _stop = None if stop is None else list(stop)
-        return await self._call_async(messages, stop=_stop, **kwargs)
+        stop_ = None if stop is None else list(stop)
+        return await self._call_async(messages, stop=stop_, **kwargs)
 
     @property
     @abstractmethod

libs/core/langchain_core/language_models/llms.py

@@ -93,10 +93,10 @@ def create_base_retry_decorator(
     Raises:
         ValueError: If the cache is not set and cache is True.
     """
-    _logging = before_sleep_log(logger, logging.WARNING)
+    logging_ = before_sleep_log(logger, logging.WARNING)
 
     def _before_sleep(retry_state: RetryCallState) -> None:
-        _logging(retry_state)
+        logging_(retry_state)
         if run_manager:
             if isinstance(run_manager, AsyncCallbackManagerForLLMRun):
                 coro = run_manager.on_retry(retry_state)
@@ -119,7 +119,7 @@ def _before_sleep(retry_state: RetryCallState) -> None:
     # 4 seconds, then up to 10 seconds, then 10 seconds afterwards
     retry_instance: retry_base = retry_if_exception_type(error_types[0])
     for error in error_types[1:]:
-        retry_instance = retry_instance | retry_if_exception_type(error)
+        retry_instance |= retry_if_exception_type(error)
     return retry(
         reraise=True,
         stop=stop_after_attempt(max_retries),
@@ -155,7 +155,7 @@ def _resolve_cache(*, cache: Union[BaseCache, bool, None]) -> Optional[BaseCache
 def get_prompts(
     params: dict[str, Any],
     prompts: list[str],
-    cache: Optional[Union[BaseCache, bool, None]] = None,  # noqa: FBT001
+    cache: Union[BaseCache, bool, None] = None,  # noqa: FBT001
 ) -> tuple[dict[int, list], str, list[int], list[str]]:
     """Get prompts that are already cached.
 
@@ -191,7 +191,7 @@ def get_prompts(
 async def aget_prompts(
     params: dict[str, Any],
     prompts: list[str],
-    cache: Optional[Union[BaseCache, bool, None]] = None,  # noqa: FBT001
+    cache: Union[BaseCache, bool, None] = None,  # noqa: FBT001
 ) -> tuple[dict[int, list], str, list[int], list[str]]:
     """Get prompts that are already cached. Async version.
 
@@ -877,8 +877,6 @@ def generate(
                 **(metadata or {}),
                 **self._get_ls_params(stop=stop, **kwargs),
             }
-        else:
-            pass
         if (
             isinstance(callbacks, list)
             and callbacks
@@ -1132,8 +1130,6 @@ async def agenerate(
                 **(metadata or {}),
                 **self._get_ls_params(stop=stop, **kwargs),
             }
-        else:
-            pass
         # Create callback managers
         if isinstance(callbacks, list) and (
             isinstance(callbacks[0], (list, BaseCallbackManager))
@@ -1352,8 +1348,8 @@ async def _call_async(
     def predict(
         self, text: str, *, stop: Optional[Sequence[str]] = None, **kwargs: Any
     ) -> str:
-        _stop = None if stop is None else list(stop)
-        return self(text, stop=_stop, **kwargs)
+        stop_ = None if stop is None else list(stop)
+        return self(text, stop=stop_, **kwargs)
 
     @deprecated("0.1.7", alternative="invoke", removal="1.0")
     @override
@@ -1365,17 +1361,17 @@ def predict_messages(
         **kwargs: Any,
     ) -> BaseMessage:
         text = get_buffer_string(messages)
-        _stop = None if stop is None else list(stop)
-        content = self(text, stop=_stop, **kwargs)
+        stop_ = None if stop is None else list(stop)
+        content = self(text, stop=stop_, **kwargs)
         return AIMessage(content=content)
 
     @deprecated("0.1.7", alternative="ainvoke", removal="1.0")
     @override
     async def apredict(
         self, text: str, *, stop: Optional[Sequence[str]] = None, **kwargs: Any
     ) -> str:
-        _stop = None if stop is None else list(stop)
-        return await self._call_async(text, stop=_stop, **kwargs)
+        stop_ = None if stop is None else list(stop)
+        return await self._call_async(text, stop=stop_, **kwargs)
 
     @deprecated("0.1.7", alternative="ainvoke", removal="1.0")
     @override
@@ -1387,8 +1383,8 @@ async def apredict_messages(
         **kwargs: Any,
     ) -> BaseMessage:
         text = get_buffer_string(messages)
-        _stop = None if stop is None else list(stop)
-        content = await self._call_async(text, stop=_stop, **kwargs)
+        stop_ = None if stop is None else list(stop)
+        content = await self._call_async(text, stop=stop_, **kwargs)
         return AIMessage(content=content)
 
     def __str__(self) -> str:

libs/core/langchain_core/load/serializable.py

@@ -355,19 +355,19 @@ def to_json_not_implemented(obj: object) -> SerializedNotImplemented:
     Returns:
         SerializedNotImplemented
     """
-    _id: list[str] = []
+    id_: list[str] = []
     try:
         if hasattr(obj, "__name__"):
-            _id = [*obj.__module__.split("."), obj.__name__]
+            id_ = [*obj.__module__.split("."), obj.__name__]
         elif hasattr(obj, "__class__"):
-            _id = [*obj.__class__.__module__.split("."), obj.__class__.__name__]
+            id_ = [*obj.__class__.__module__.split("."), obj.__class__.__name__]
     except Exception:
         logger.debug("Failed to serialize object", exc_info=True)
 
     result: SerializedNotImplemented = {
         "lc": 1,
         "type": "not_implemented",
-        "id": _id,
+        "id": id_,
         "repr": None,
     }
     with contextlib.suppress(Exception):

libs/core/langchain_core/messages/tool.py

@@ -126,8 +126,6 @@ def coerce_args(cls, values: dict) -> dict:
                         raise ValueError(msg) from e
                 else:
                     values["content"].append(x)
-        else:
-            pass
 
         tool_call_id = values["tool_call_id"]
         if isinstance(tool_call_id, (UUID, int, float)):
@@ -366,4 +364,4 @@ def default_tool_chunk_parser(raw_tool_calls: list[dict]) -> list[ToolCallChunk]
 def _merge_status(
     left: Literal["success", "error"], right: Literal["success", "error"]
 ) -> Literal["success", "error"]:
-    return "error" if "error" in (left, right) else "success"
+    return "error" if "error" in {left, right} else "success"