github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-09-11 16:01:33 +00:00
Code Issues Packages Projects Releases Wiki Activity

chore(langchain): add ruff rule E501 in langchain_v1 (#32812)

Browse Source 
Co-authored-by: Mason Daugherty <mason@langchain.dev>
This commit is contained in:
Christophe Bornet
2025-09-08 23:28:14 +02:00
committed by GitHub
parent 1e101ae9a2
commit 017348b27c
9 changed files with 147 additions and 65 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
libs/langchain_v1/langchain/agents/interrupt.py
View File

@@ -74,15 +74,20 @@ class HumanInterrupt(TypedDict):
class HumanResponse(TypedDict):
"""The response provided by a human to an interrupt, which is returned when graph execution resumes.
"""Human response.
The response provided by a human to an interrupt,
which is returned when graph execution resumes.
Attributes:
type: The type of response:
- "accept": Approves the current state without changes
- "ignore": Skips/ignores the current step
- "response": Provides text feedback or instructions
- "edit": Modifies the current state/content
args: The response payload:
- None: For ignore/accept actions
- str: For text responses
- ActionRequest: For edit actions with updated content

11
libs/langchain_v1/langchain/agents/middleware/human_in_the_loop.py
View File

@@ -65,7 +65,10 @@ class HumanInTheLoopMiddleware(AgentMiddleware):
# Right now, we do not support multiple tool calls with interrupts
if len(interrupt_tool_calls) > 1:
tool_names = [t["name"] for t in interrupt_tool_calls]
msg = f"Called the following tools which require interrupts: {tool_names}\n\nYou may only call ONE tool that requires an interrupt at a time"
msg = (
f"Called the following tools which require interrupts: {tool_names}\n\n"
"You may only call ONE tool that requires an interrupt at a time"
)
return {
"messages": _generate_correction_tool_messages(msg, last_message.tool_calls),
"jump_to": "model",
@@ -74,7 +77,11 @@ class HumanInTheLoopMiddleware(AgentMiddleware):
# Right now, we do not support interrupting a tool call if other tool calls exist
if auto_approved_tool_calls:
tool_names = [t["name"] for t in interrupt_tool_calls]
msg = f"Called the following tools which require interrupts: {tool_names}. You also called other tools that do not require interrupts. If you call a tool that requires and interrupt, you may ONLY call that tool."
msg = (
f"Called the following tools which require interrupts: {tool_names}. "
"You also called other tools that do not require interrupts. "
"If you call a tool that requires and interrupt, you may ONLY call that tool."
)
return {
"messages": _generate_correction_tool_messages(msg, last_message.tool_calls),
"jump_to": "model",

17
libs/langchain_v1/langchain/agents/middleware/prompt_caching.py
View File

@@ -6,9 +6,12 @@ from langchain.agents.middleware.types import AgentMiddleware, AgentState, Model
class AnthropicPromptCachingMiddleware(AgentMiddleware):
"""Prompt Caching Middleware - Optimizes API usage by caching conversation prefixes for Anthropic models.
"""Prompt Caching Middleware.
Learn more about anthropic prompt caching [here](https://docs.anthropic.com/en/docs/build-with-claude/prompt-caching).
Optimizes API usage by caching conversation prefixes for Anthropic models.
Learn more about Anthropic prompt caching
`here <https://docs.anthropic.com/en/docs/build-with-claude/prompt-caching>`__.
"""
def __init__(
@@ -22,7 +25,8 @@ class AnthropicPromptCachingMiddleware(AgentMiddleware):
Args:
type: The type of cache to use, only "ephemeral" is supported.
ttl: The time to live for the cache, only "5m" and "1h" are supported.
min_messages_to_cache: The minimum number of messages until the cache is used, default is 0.
min_messages_to_cache: The minimum number of messages until the cache is used,
default is 0.
"""
self.type = type
self.ttl = ttl
@@ -34,15 +38,16 @@ class AnthropicPromptCachingMiddleware(AgentMiddleware):
from langchain_anthropic import ChatAnthropic
except ImportError:
msg = (
"AnthropicPromptCachingMiddleware caching middleware only supports Anthropic models."
"AnthropicPromptCachingMiddleware caching middleware only supports "
"Anthropic models."
"Please install langchain-anthropic."
)
raise ValueError(msg)
if not isinstance(request.model, ChatAnthropic):
msg = (
"AnthropicPromptCachingMiddleware caching middleware only supports Anthropic models, "
f"not instances of {type(request.model)}"
"AnthropicPromptCachingMiddleware caching middleware only supports "
f"Anthropic models, not instances of {type(request.model)}"
)
raise ValueError(msg)

2
libs/langchain_v1/langchain/agents/middleware/summarization.py
View File

@@ -48,7 +48,7 @@ Respond ONLY with the extracted context. Do not include any additional informati
<messages>
Messages to summarize:
{messages}
</messages>"""
</messages>""" # noqa: E501
SUMMARY_PREFIX = "## Previous conversation summary:"

3
libs/langchain_v1/langchain/agents/middleware/types.py
View File

@@ -58,7 +58,8 @@ StateT = TypeVar("StateT", bound=AgentState)
class AgentMiddleware(Generic[StateT]):
"""Base middleware class for an agent.
Subclass this and implement any of the defined methods to customize agent behavior between steps in the main agent loop.
Subclass this and implement any of the defined methods to customize agent behavior
between steps in the main agent loop.
"""
state_schema: type[StateT] = cast("type[StateT]", AgentState)

93
libs/langchain_v1/langchain/agents/react_agent.py
View File

@@ -176,8 +176,9 @@ def _validate_chat_history(
error_message = create_error_message(
message="Found AIMessages with tool_calls that do not have a corresponding ToolMessage. "
f"Here are the first few of those tool calls: {tool_calls_without_results[:3]}.\n\n"
"Every tool call (LLM requesting to call a tool) in the message history MUST have a corresponding ToolMessage "
"(result of a tool invocation to return to the LLM) - this is required by most LLM providers.",
"Every tool call (LLM requesting to call a tool) in the message history "
"MUST have a corresponding ToolMessage (result of a tool invocation to return to the LLM) -"
" this is required by most LLM providers.",
error_code=ErrorCode.INVALID_CHAT_HISTORY,
)
raise ValueError(error_message)
@@ -216,7 +217,8 @@ class _AgentBuilder(Generic[StateT, ContextT, StructuredResponseT]):
if isinstance(model, Runnable) and not isinstance(model, BaseChatModel):
msg = (
"Expected `model` to be a BaseChatModel or a string, got {type(model)}."
"The `model` parameter should not have pre-bound tools, simply pass the model and tools separately."
"The `model` parameter should not have pre-bound tools, "
"simply pass the model and tools separately."
)
raise ValueError(msg)
@@ -308,7 +310,8 @@ class _AgentBuilder(Generic[StateT, ContextT, StructuredResponseT]):
Command with structured response update if found, None otherwise
Raises:
MultipleStructuredOutputsError: If multiple structured responses are returned and error handling is disabled
MultipleStructuredOutputsError: If multiple structured responses are returned
and error handling is disabled
StructuredOutputParsingError: If parsing fails and error handling is disabled
"""
if not isinstance(self.response_format, ToolStrategy) or not response.tool_calls:
@@ -452,7 +455,11 @@ class _AgentBuilder(Generic[StateT, ContextT, StructuredResponseT]):
return model.bind(**kwargs)
def _handle_structured_response_native(self, response: AIMessage) -> Command | None:
"""If native output is configured and there are no tool calls, parse using ProviderStrategyBinding."""
"""Handle structured output using the native output.
If native output is configured and there are no tool calls,
parse using ProviderStrategyBinding.
"""
if self.native_output_binding is None:
return None
if response.tool_calls:
@@ -922,7 +929,8 @@ def create_agent( # noqa: D417
) -> CompiledStateGraph[StateT, ContextT]:
"""Creates an agent graph that calls tools in a loop until a stopping condition is met.
For more details on using `create_agent`, visit [Agents](https://langchain-ai.github.io/langgraph/agents/overview/) documentation.
For more details on using `create_agent`,
visit [Agents](https://langchain-ai.github.io/langgraph/agents/overview/) documentation.
Args:
model: The language model for the agent. Supports static and dynamic
@@ -969,25 +977,35 @@ def create_agent( # noqa: D417
must be a subset of those specified in the `tools` parameter.
tools: A list of tools or a ToolNode instance.
If an empty list is provided, the agent will consist of a single LLM node without tool calling.
If an empty list is provided, the agent will consist of a single LLM node
without tool calling.
prompt: An optional prompt for the LLM. Can take a few different forms:
- str: This is converted to a SystemMessage and added to the beginning of the list of messages in state["messages"].
- SystemMessage: this is added to the beginning of the list of messages in state["messages"].
- Callable: This function should take in full graph state and the output is then passed to the language model.
- Runnable: This runnable should take in full graph state and the output is then passed to the language model.
- str: This is converted to a SystemMessage and added to the beginning
of the list of messages in state["messages"].
- SystemMessage: this is added to the beginning of the list of messages
in state["messages"].
- Callable: This function should take in full graph state and the output is then passed
to the language model.
- Runnable: This runnable should take in full graph state and the output is then passed
to the language model.
response_format: An optional UsingToolStrategy configuration for structured responses.
If provided, the agent will handle structured output via tool calls during the normal conversation flow.
When the model calls a structured output tool, the response will be captured and returned in the 'structured_response' state key.
If provided, the agent will handle structured output via tool calls
during the normal conversation flow.
When the model calls a structured output tool, the response will be captured
and returned in the 'structured_response' state key.
If not provided, `structured_response` will not be present in the output state.
The UsingToolStrategy should contain:
- schemas: A sequence of ResponseSchema objects that define the structured output format
- schemas: A sequence of ResponseSchema objects that define
the structured output format
- tool_choice: Either "required" or "auto" to control when structured output is used
Each ResponseSchema contains:
- schema: A Pydantic model that defines the structure
- name: Optional custom name for the tool (defaults to model name)
- description: Optional custom description (defaults to model docstring)
@@ -997,11 +1015,15 @@ def create_agent( # noqa: D417
`response_format` requires the model to support tool calling
!!! Note
Structured responses are handled directly in the model call node via tool calls, eliminating the need for separate structured response nodes.
Structured responses are handled directly in the model call node via tool calls,
eliminating the need for separate structured response nodes.
pre_model_hook: An optional node to add before the `agent` node (i.e., the node that calls the LLM).
Useful for managing long message histories (e.g., message trimming, summarization, etc.).
Pre-model hook must be a callable or a runnable that takes in current graph state and returns a state update in the form of
pre_model_hook: An optional node to add before the `agent` node
(i.e., the node that calls the LLM).
Useful for managing long message histories
(e.g., message trimming, summarization, etc.).
Pre-model hook must be a callable or a runnable that takes in current
graph state and returns a state update in the form of
```python
# At least one of `messages` or `llm_input_messages` MUST be provided
{
@@ -1016,11 +1038,13 @@ def create_agent( # noqa: D417
```
!!! Important
At least one of `messages` or `llm_input_messages` MUST be provided and will be used as an input to the `agent` node.
At least one of `messages` or `llm_input_messages` MUST be provided
and will be used as an input to the `agent` node.
The rest of the keys will be added to the graph state.
!!! Warning
If you are returning `messages` in the pre-model hook, you should OVERWRITE the `messages` key by doing the following:
If you are returning `messages` in the pre-model hook,
you should OVERWRITE the `messages` key by doing the following:
```python
{
@@ -1028,9 +1052,12 @@ def create_agent( # noqa: D417
...
}
```
post_model_hook: An optional node to add after the `agent` node (i.e., the node that calls the LLM).
Useful for implementing human-in-the-loop, guardrails, validation, or other post-processing.
Post-model hook must be a callable or a runnable that takes in current graph state and returns a state update.
post_model_hook: An optional node to add after the `agent` node
(i.e., the node that calls the LLM).
Useful for implementing human-in-the-loop, guardrails, validation,
or other post-processing.
Post-model hook must be a callable or a runnable that takes in
current graph state and returns a state update.
!!! Note
Only available with `version="v2"`.
@@ -1039,12 +1066,14 @@ def create_agent( # noqa: D417
Defaults to `AgentState` that defines those two keys.
context_schema: An optional schema for runtime context.
checkpointer: An optional checkpoint saver object. This is used for persisting
the state of the graph (e.g., as chat memory) for a single thread (e.g., a single conversation).
the state of the graph (e.g., as chat memory) for a single thread
(e.g., a single conversation).
store: An optional store object. This is used for persisting data
across multiple threads (e.g., multiple conversations / users).
interrupt_before: An optional list of node names to interrupt before.
Should be one of the following: "agent", "tools".
This is useful if you want to add a user confirmation or other interrupt before taking an action.
This is useful if you want to add a user confirmation or other interrupt
before taking an action.
interrupt_after: An optional list of node names to interrupt after.
Should be one of the following: "agent", "tools".
This is useful if you want to return directly or run additional processing on an output.
@@ -1059,7 +1088,8 @@ def create_agent( # noqa: D417
node using the [Send](https://langchain-ai.github.io/langgraph/concepts/low_level/#send)
API.
name: An optional name for the CompiledStateGraph.
This name will be automatically used when adding ReAct agent graph to another graph as a subgraph node -
This name will be automatically used when adding ReAct agent graph to
another graph as a subgraph node -
particularly useful for building multi-agent systems.
!!! warning "`config_schema` Deprecated"
@@ -1071,9 +1101,11 @@ def create_agent( # noqa: D417
A compiled LangChain runnable that can be used for chat interactions.
The "agent" node calls the language model with the messages list (after applying the prompt).
If the resulting AIMessage contains `tool_calls`, the graph will then call the ["tools"][langgraph.prebuilt.tool_node.ToolNode].
The "tools" node executes the tools (1 tool per `tool_call`) and adds the responses to the messages list
as `ToolMessage` objects. The agent node then calls the language model again.
If the resulting AIMessage contains `tool_calls`,
the graph will then call the ["tools"][langgraph.prebuilt.tool_node.ToolNode].
The "tools" node executes the tools (1 tool per `tool_call`)
and adds the responses to the messages list as `ToolMessage` objects.
The agent node then calls the language model again.
The process repeats until no more `tool_calls` are present in the response.
The agent then returns the full list of messages as a dictionary containing the key "messages".
@@ -1135,7 +1167,8 @@ def create_agent( # noqa: D417
# Handle deprecated config_schema parameter
if (config_schema := deprecated_kwargs.pop("config_schema", MISSING)) is not MISSING:
warn(
"`config_schema` is deprecated and will be removed. Please use `context_schema` instead.",
"`config_schema` is deprecated and will be removed. "
"Please use `context_schema` instead.",
category=DeprecationWarning,
stacklevel=2,
)

25
libs/langchain_v1/langchain/agents/structured_output.py
View File

@@ -47,7 +47,8 @@ class MultipleStructuredOutputsError(StructuredOutputError):
self.tool_names = tool_names
super().__init__(
f"Model incorrectly returned multiple structured responses ({', '.join(tool_names)}) when only one is expected."
"Model incorrectly returned multiple structured responses "
f"({', '.join(tool_names)}) when only one is expected."
)
@@ -98,7 +99,8 @@ class _SchemaSpec(Generic[SchemaT]):
"""Describes a structured output schema."""
schema: type[SchemaT]
"""The schema for the response, can be a Pydantic model, dataclass, TypedDict, or JSON schema dict."""
"""The schema for the response, can be a Pydantic model, dataclass, TypedDict,
or JSON schema dict."""
name: str
"""Name of the schema, used for tool calling.
@@ -178,7 +180,8 @@ class ToolStrategy(Generic[SchemaT]):
"""Schema specs for the tool calls."""
tool_message_content: str | None
"""The content of the tool message to be returned when the model calls an artificial structured output tool."""
"""The content of the tool message to be returned when the model calls
an artificial structured output tool."""
handle_errors: (
bool | str | type[Exception] | tuple[type[Exception], ...] | Callable[[Exception], str]
@@ -204,7 +207,10 @@ class ToolStrategy(Generic[SchemaT]):
| tuple[type[Exception], ...]
| Callable[[Exception], str] = True,
) -> None:
"""Initialize ToolStrategy with schemas, tool message content, and error handling strategy."""
"""Initialize ToolStrategy.
Initialize ToolStrategy with schemas, tool message content, and error handling strategy.
"""
self.schema = schema
self.tool_message_content = tool_message_content
self.handle_errors = handle_errors
@@ -268,7 +274,8 @@ class OutputToolBinding(Generic[SchemaT]):
"""
schema: type[SchemaT]
"""The original schema provided for structured output (Pydantic model, dataclass, TypedDict, or JSON schema dict)."""
"""The original schema provided for structured output
(Pydantic model, dataclass, TypedDict, or JSON schema dict)."""
schema_kind: SchemaKind
"""Classification of the schema type for proper response construction."""
@@ -321,7 +328,8 @@ class ProviderStrategyBinding(Generic[SchemaT]):
"""
schema: type[SchemaT]
"""The original schema provided for structured output (Pydantic model, dataclass, TypedDict, or JSON schema dict)."""
"""The original schema provided for structured output
(Pydantic model, dataclass, TypedDict, or JSON schema dict)."""
schema_kind: SchemaKind
"""Classification of the schema type for proper response construction."""
@@ -362,7 +370,10 @@ class ProviderStrategyBinding(Generic[SchemaT]):
data = json.loads(raw_text)
except Exception as e:
schema_name = getattr(self.schema, "__name__", "response_format")
msg = f"Native structured output expected valid JSON for {schema_name}, but parsing failed: {e}."
msg = (
f"Native structured output expected valid JSON for {schema_name}, "
f"but parsing failed: {e}."
)
raise ValueError(msg) from e
# Parse according to schema

53
libs/langchain_v1/langchain/agents/tool_node.py
View File

@@ -87,8 +87,16 @@ INVALID_TOOL_NAME_ERROR_TEMPLATE = (
"Error: {requested_tool} is not a valid tool, try one of [{available_tools}]."
)
TOOL_CALL_ERROR_TEMPLATE = "Error: {error}\n Please fix your mistakes."
TOOL_EXECUTION_ERROR_TEMPLATE = "Error executing tool '{tool_name}' with kwargs {tool_kwargs} with error:\n {error}\n Please fix the error and try again."
TOOL_INVOCATION_ERROR_TEMPLATE = "Error invoking tool '{tool_name}' with kwargs {tool_kwargs} with error:\n {error}\n Please fix the error and try again."
TOOL_EXECUTION_ERROR_TEMPLATE = (
"Error executing tool '{tool_name}' with kwargs {tool_kwargs} with error:\n"
" {error}\n"
" Please fix the error and try again."
)
TOOL_INVOCATION_ERROR_TEMPLATE = (
"Error invoking tool '{tool_name}' with kwargs {tool_kwargs} with error:\n"
" {error}\n"
" Please fix the error and try again."
)
def msg_content_output(output: Any) -> str | list[dict]:
@@ -313,7 +321,8 @@ class ToolNode(RunnableCallable):
error template containing the exception details.
- **str**: Catch all errors and return a ToolMessage with this custom
error message string.
- **type[Exception]**: Only catch exceptions with the specified type and return the default error message for it.
- **type[Exception]**: Only catch exceptions with the specified type and
return the default error message for it.
- **tuple[type[Exception], ...]**: Only catch exceptions with the specified
types and return default error messages for them.
- **Callable[..., str]**: Catch exceptions matching the callable's signature
@@ -368,7 +377,7 @@ class ToolNode(RunnableCallable):
tool_node = ToolNode([my_tool], handle_tool_errors=handle_errors)
```
"""
""" # noqa: E501
name: str = "tools"
@@ -506,10 +515,12 @@ class ToolNode(RunnableCallable):
# GraphInterrupt is a special exception that will always be raised.
# It can be triggered in the following scenarios,
# Where GraphInterrupt(GraphBubbleUp) is raised from an `interrupt` invocation most commonly:
# Where GraphInterrupt(GraphBubbleUp) is raised from an `interrupt` invocation
# most commonly:
# (1) a GraphInterrupt is raised inside a tool
# (2) a GraphInterrupt is raised inside a graph node for a graph called as a tool
# (3) a GraphInterrupt is raised when a subgraph is interrupted inside a graph called as a tool
# (3) a GraphInterrupt is raised when a subgraph is interrupted inside a graph
# called as a tool
# (2 and 3 can happen in a "supervisor w/ tools" multi-agent architecture)
except GraphBubbleUp:
raise
@@ -570,10 +581,12 @@ class ToolNode(RunnableCallable):
# GraphInterrupt is a special exception that will always be raised.
# It can be triggered in the following scenarios,
# Where GraphInterrupt(GraphBubbleUp) is raised from an `interrupt` invocation most commonly:
# Where GraphInterrupt(GraphBubbleUp) is raised from an `interrupt` invocation
# most commonly:
# (1) a GraphInterrupt is raised inside a tool
# (2) a GraphInterrupt is raised inside a graph node for a graph called as a tool
# (3) a GraphInterrupt is raised when a subgraph is interrupted inside a graph called as a tool
# (3) a GraphInterrupt is raised when a subgraph is interrupted inside a graph
# called as a tool
# (2 and 3 can happen in a "supervisor w/ tools" multi-agent architecture)
except GraphBubbleUp:
raise
@@ -770,10 +783,12 @@ class ToolNode(RunnableCallable):
input_type: Literal["list", "dict", "tool_calls"],
) -> Command:
if isinstance(command.update, dict):
# input type is dict when ToolNode is invoked with a dict input (e.g. {"messages": [AIMessage(..., tool_calls=[...])]})
# input type is dict when ToolNode is invoked with a dict input
# (e.g. {"messages": [AIMessage(..., tool_calls=[...])]})
if input_type not in ("dict", "tool_calls"):
msg = (
f"Tools can provide a dict in Command.update only when using dict with '{self._messages_key}' key as ToolNode input, "
"Tools can provide a dict in Command.update only when using dict "
f"with '{self._messages_key}' key as ToolNode input, "
f"got: {command.update} for tool '{call['name']}'"
)
raise ValueError(msg)
@@ -782,10 +797,12 @@ class ToolNode(RunnableCallable):
state_update = cast("dict[str, Any]", updated_command.update) or {}
messages_update = state_update.get(self._messages_key, [])
elif isinstance(command.update, list):
# Input type is list when ToolNode is invoked with a list input (e.g. [AIMessage(..., tool_calls=[...])])
# Input type is list when ToolNode is invoked with a list input
# (e.g. [AIMessage(..., tool_calls=[...])])
if input_type != "list":
msg = (
f"Tools can provide a list of messages in Command.update only when using list of messages as ToolNode input, "
"Tools can provide a list of messages in Command.update "
"only when using list of messages as ToolNode input, "
f"got: {command.update} for tool '{call['name']}'"
)
raise ValueError(msg)
@@ -815,13 +832,17 @@ class ToolNode(RunnableCallable):
# Command.update if command is sent to the CURRENT graph
if updated_command.graph is None and not has_matching_tool_message:
example_update = (
'`Command(update={"messages": [ToolMessage("Success", tool_call_id=tool_call_id), ...]}, ...)`'
'`Command(update={"messages": '
'[ToolMessage("Success", tool_call_id=tool_call_id), ...]}, ...)`'
if input_type == "dict"
else '`Command(update=[ToolMessage("Success", tool_call_id=tool_call_id), ...], ...)`'
else "`Command(update="
'[ToolMessage("Success", tool_call_id=tool_call_id), ...], ...)`'
)
msg = (
f"Expected to have a matching ToolMessage in Command.update for tool '{call['name']}', got: {messages_update}. "
"Every tool call (LLM requesting to call a tool) in the message history MUST have a corresponding ToolMessage. "
"Expected to have a matching ToolMessage in Command.update "
f"for tool '{call['name']}', got: {messages_update}. "
"Every tool call (LLM requesting to call a tool) "
"in the message history MUST have a corresponding ToolMessage. "
f"You can fix it by modifying the tool to return {example_update}."
)
raise ValueError(msg)

1
libs/langchain_v1/pyproject.toml
View File

@@ -131,7 +131,6 @@ flake8-annotations.allow-star-arg-any = true
]
"langchain/agents/*" = [
"ANN401", # we use Any right now, need to narrow
"E501", # line too long, needs to fix
"A002", # input is shadowing builtin
"A001", # input is shadowing builtin
"B904", # use from for exceptions