bisheng-langchain 1.1.1__py3-none-any.whl → 1.2.0__py3-none-any.whl

bisheng_langchain/gpts/tools/api_tools/macro_data.py

@@ -4,15 +4,15 @@ from typing import Any
 
 import pandas as pd
 import requests
-from langchain.pydantic_v1 import BaseModel, Field
+from pydantic import BaseModel, Field
 from langchain_core.tools import BaseTool
 
 from .base import MultArgsSchemaTool
 
 
 class QueryArg(BaseModel):
-    start_date: str = Field(default='', description='开始月份, 使用YYYY-MM-DD 方式表示', example='2023-01-01')
-    end_date: str = Field(default='', description='结束月份，使用YYYY-MM-DD 方式表示', example='2023-05-01')
+    start_date: str = Field(default='', description='开始月份, 使用YYYY-MM-DD 方式表示', examples=['2023-01-01'])
+    end_date: str = Field(default='', description='结束月份，使用YYYY-MM-DD 方式表示', examples=['2023-05-01'])
 
 
 class MacroData(BaseModel):

bisheng_langchain/gpts/tools/api_tools/openapi.py

@@ -9,9 +9,9 @@ from .base import APIToolBase, Field, MultArgsSchemaTool
 
 class OpenApiTools(APIToolBase):
 
-    api_key: Optional[str]
-    api_location: Optional[str]
-    parameter_name: Optional[str]
+    api_key: Optional[str] = None
+    api_location: Optional[str] = None
+    parameter_name: Optional[str] = None
 
     def get_real_path(self, path_params: dict | None):
         path = self.params['path']

bisheng_langchain/gpts/tools/api_tools/sina.py

@@ -7,7 +7,7 @@ import re
 from datetime import datetime
 from typing import List, Type
 
-from langchain_core.pydantic_v1 import BaseModel, Field
+from pydantic import BaseModel, Field
 from loguru import logger
 
 from .base import APIToolBase

bisheng_langchain/gpts/tools/api_tools/tianyancha.py

@@ -3,8 +3,9 @@ from __future__ import annotations
 
 from typing import Any, Dict, Type
 
+from pydantic import model_validator, BaseModel, Field
+
 from bisheng_langchain.utils.requests import Requests, RequestsWrapper
-from langchain_core.pydantic_v1 import BaseModel, Field, root_validator
 
 from .base import APIToolBase
 
@@ -19,7 +20,8 @@ class CompanyInfo(APIToolBase):
     api_key: str = None
     args_schema: Type[BaseModel] = InputArgs
 
-    @root_validator(pre=True)
+    @model_validator(mode='before')
+    @classmethod
     def build_header(cls, values: Dict[str, Any]) -> Dict[str, Any]:
         """Build headers that were passed in."""
         if not values.get('api_key'):
@@ -30,7 +32,8 @@ class CompanyInfo(APIToolBase):
         values['headers'] = headers
         return values
 
-    @root_validator()
+    @model_validator(mode='before')
+    @classmethod
     def validate_environment(cls, values: Dict) -> Dict:
         """Validate that api key and python package exists in environment."""
         timeout = values.get('request_timeout', 30)

bisheng_langchain/gpts/tools/bing_search/tool.py

@@ -2,7 +2,7 @@
 
 from typing import Optional, Type
 
-from langchain.pydantic_v1 import BaseModel, Field
+from pydantic import BaseModel, Field
 from langchain_community.utilities.bing_search import BingSearchAPIWrapper
 from langchain_core.callbacks import CallbackManagerForToolRun
 from langchain_core.tools import BaseTool
@@ -43,7 +43,7 @@ class BingSearchResults(BaseTool):
         "Input should be a search query. Output is a JSON array of the query results"
     )
     num_results: int = 5
-    args_schema = BingSearchInput
+    args_schema: Type[BaseModel] = BingSearchInput
     api_wrapper: BingSearchAPIWrapper
 
     def _run(

bisheng_langchain/gpts/tools/calculator/tool.py

@@ -2,7 +2,7 @@ import math
 from math import *
 
 import sympy
-from langchain.pydantic_v1 import BaseModel, Field
+from pydantic import BaseModel, Field
 from langchain.tools import tool
 from sympy import *
 
@@ -10,7 +10,7 @@ from sympy import *
 class CalculatorInput(BaseModel):
     expression: str = Field(
         description="The input to this tool should be a mathematical expression using only Python's built-in mathematical operators.",
-        example='200*7',
+        examples=['200*7'],
     )
 
 

bisheng_langchain/gpts/tools/code_interpreter/tool.py

@@ -15,7 +15,7 @@ from uuid import uuid4
 
 import matplotlib
 from langchain_community.tools import Tool
-from langchain_core.pydantic_v1 import BaseModel, Field
+from pydantic import BaseModel, Field
 from loguru import logger
 
 CODE_BLOCK_PATTERN = r"```(\w*)\n(.*?)\n```"
@@ -239,7 +239,7 @@ class CodeInterpreterToolArguments(BaseModel):
 
     python_code: str = Field(
         ...,
-        example="print('Hello World')",
+        examples=["print('Hello World')"],
         description=(
             'The pure python script to be evaluated. '
             'The contents will be in main.py. '

bisheng_langchain/gpts/tools/dalle_image_generator/tool.py

@@ -2,13 +2,11 @@ import logging
 import os
 from typing import Any, Dict, Mapping, Optional, Tuple, Type, Union
 
-from langchain.pydantic_v1 import BaseModel, Field
-from langchain_community.utilities.dalle_image_generator import DallEAPIWrapper
 from langchain_community.utils.openai import is_openai_v1
 from langchain_core.callbacks import CallbackManagerForToolRun
-from langchain_core.pydantic_v1 import BaseModel, Extra, Field, root_validator
 from langchain_core.tools import BaseTool
 from langchain_core.utils import get_from_dict_or_env, get_pydantic_field_names
+from pydantic import ConfigDict, model_validator, BaseModel, Field
 
 from bisheng_langchain.utils.azure_dalle_image_generator import AzureDallEWrapper
 
@@ -26,7 +24,7 @@ class DallEAPIWrapper(BaseModel):
     2. save your OPENAI_API_KEY in an environment variable
     """
 
-    client: Any  #: :meta private:
+    client: Any = None  #: :meta private:
     async_client: Any = Field(default=None, exclude=True)  #: :meta private:
     model_name: str = Field(default="dall-e-2", alias="model")
     model_kwargs: Dict[str, Any] = Field(default_factory=dict)
@@ -59,13 +57,10 @@ class DallEAPIWrapper(BaseModel):
     http_async_client: Union[Any, None] = None
     """Optional httpx.AsyncClient. Only used for async invocations. Must specify 
         http_client as well if you'd like a custom client for sync invocations."""
+    model_config = ConfigDict(extra='forbid')
 
-    class Config:
-        """Configuration for this pydantic object."""
-
-        extra = Extra.forbid
-
-    @root_validator(pre=True)
+    @model_validator(mode='before')
+    @classmethod
     def build_extra(cls, values: Dict[str, Any]) -> Dict[str, Any]:
         """Build extra kwargs from additional params that were passed in."""
         all_required_field_names = get_pydantic_field_names(cls)
@@ -91,7 +86,8 @@ class DallEAPIWrapper(BaseModel):
         values["model_kwargs"] = extra
         return values
 
-    @root_validator()
+    @model_validator(mode='before')
+    @classmethod
     def validate_environment(cls, values: Dict) -> Dict:
         """Validate that api key and python package exists in environment."""
         values["openai_api_key"] = get_from_dict_or_env(values, "openai_api_key", "OPENAI_API_KEY")

bisheng_langchain/gpts/tools/get_current_time/tool.py

@@ -1,7 +1,7 @@
 from datetime import datetime
 
 import pytz
-from langchain.pydantic_v1 import BaseModel, Field
+from pydantic import BaseModel, Field
 from langchain.tools import tool
 
 

bisheng_langchain/gpts/tools/message/dingding.py

@@ -1,8 +1,7 @@
 from typing import Any, Optional, Type
 
 import requests
-from langchain_core.pydantic_v1 import BaseModel, Field, root_validator
-from loguru import logger
+from pydantic import BaseModel, Field
 
 from bisheng_langchain.gpts.tools.api_tools.base import (APIToolBase,
                                                          MultArgsSchemaTool)

bisheng_langchain/gpts/tools/message/email.py

@@ -1,11 +1,9 @@
-import os
 import smtplib
-from email.mime.application import MIMEApplication
 from email.mime.multipart import MIMEMultipart
 from email.mime.text import MIMEText
-from typing import Any, Optional
+from typing import Any
 
-from langchain_core.pydantic_v1 import BaseModel, Field, root_validator
+from pydantic import BaseModel, Field
 
 from bisheng_langchain.gpts.tools.api_tools.base import (APIToolBase,
                                                          MultArgsSchemaTool)
@@ -17,7 +15,7 @@ class InputArgs(BaseModel):
     content: str = Field(description="邮件正文内容")
 
 
-class EmailMessageTool(APIToolBase):
+class EmailMessageTool(BaseModel):
 
     email_account: str = Field(description="发件人邮箱")
     email_password: str = Field(description="邮箱授权码/密码")
@@ -27,9 +25,9 @@ class EmailMessageTool(APIToolBase):
 
     def send_email(
         self,
-        receiver,
-        subject,
-        content,
+        receiver: str = None,
+        subject: str = None,
+        content: str = None,
     ):
         """
         发送电子邮件函数

bisheng_langchain/gpts/tools/message/feishu.py

@@ -1,29 +1,28 @@
 from typing import Any, Optional, Type
 
 import requests
-from langchain_core.pydantic_v1 import BaseModel, Field, root_validator
-from loguru import logger
+from pydantic import BaseModel, Field
 
 from bisheng_langchain.gpts.tools.api_tools.base import (APIToolBase,
                                                          MultArgsSchemaTool)
 
 
 class InputArgs(BaseModel):
-    message: Optional[str] = Field(description="需要发送的钉钉消息")
-    receive_id: Optional[str] = Field(description="接收的ID")
-    receive_id_type: Optional[str] = Field(description="接收的ID类型")
-    container_id: Optional[str] = Field(description="container_id")
-    start_time: Optional[str] = Field(description="start_time")
-    end_time: Optional[str] = Field(description="end_time")
+    message: Optional[str] = Field(None, description="需要发送的钉钉消息")
+    receive_id: Optional[str] = Field(None, description="接收的ID")
+    receive_id_type: Optional[str] = Field(None, description="接收的ID类型")
+    container_id: Optional[str] = Field(None, description="container_id")
+    start_time: Optional[str] = Field(None, description="start_time")
+    end_time: Optional[str] = Field(None, description="end_time")
     # page_token: Optional[str] = Field(description="page_token")
-    container_id_type: Optional[str] = Field(description="container_id_type")
+    container_id_type: Optional[str] = Field(None, description="container_id_type")
     page_size: Optional[int] = Field(default=20,description="page_size")
-    page_token: Optional[str] = Field(description="page_token")
+    page_token: Optional[str] = Field(None, description="page_token")
     sort_type: Optional[str] = Field(description="sort_type",default="ByCreateTimeAsc")
 
 
 class FeishuMessageTool(BaseModel):
-    API_BASE_URL = "https://open.feishu.cn/open-apis"
+    API_BASE_URL: str = "https://open.feishu.cn/open-apis"
     app_id: str = Field(description="app_id")
     app_secret: str = Field(description="app_secret")
 

bisheng_langchain/gpts/tools/message/wechat.py

@@ -1,8 +1,7 @@
-from typing import Any, Optional, Type
+from typing import Any
 
 import requests
-from langchain_core.pydantic_v1 import BaseModel, Field, root_validator
-from loguru import logger
+from pydantic import BaseModel, Field
 
 from bisheng_langchain.gpts.tools.api_tools.base import (APIToolBase,
                                                          MultArgsSchemaTool)

bisheng_langchain/gpts/tools/sql_agent/tool.py

@@ -1,236 +1,77 @@
-from typing import Type, Optional, TypedDict, Annotated, Any, Literal
+from typing import Type, Optional
 
 from langchain_community.agent_toolkits import SQLDatabaseToolkit
 from langchain_community.utilities import SQLDatabase
 from langchain_core.callbacks import CallbackManagerForToolRun
 from langchain_core.language_models import BaseLanguageModel
-from langchain_core.messages import AnyMessage, AIMessage, ToolMessage
-from langchain_core.prompts import ChatPromptTemplate
-from langchain_core.runnables import RunnableLambda, RunnableWithFallbacks
-from langchain_core.tools import BaseTool, tool
-from langgraph.constants import END, START
-from langgraph.graph import add_messages, StateGraph
-from langgraph.prebuilt import ToolNode
-from pydantic import BaseModel, Field
+from langchain_core.messages import HumanMessage
+from langchain_core.tools import BaseTool
+from langgraph.graph.graph import CompiledGraph
+from langgraph.prebuilt import create_react_agent
+from pydantic import BaseModel, Field, ConfigDict
 
+_agent_system_prompt = """You are an autonomous agent that answers user questions by querying an SQL database through the provided tools.
 
-class State(TypedDict):
-    messages: Annotated[list[AnyMessage], add_messages]
+When a new question arrives, follow the steps *in order*:
 
+1. ALWAYS call `sql_db_list_tables` first.  
+   Purpose: discover what tables are available. Never skip this step.
 
-def handle_tool_error(state) -> dict:
-    error = state.get("error")
-    tool_calls = state["messages"][-1].tool_calls
-    return {
-        "messages": [
-            ToolMessage(
-                content=f"Error: {repr(error)}\n please fix your mistakes.",
-                tool_call_id=tc["id"],
-            )
-            for tc in tool_calls
-        ]
-    }
+2. Choose the table(s) that are probably relevant, then call `sql_db_schema`
+   once for each of those tables to obtain their schemas.
 
+3. Write one syntactically-correct {dialect} SELECT statement.  
+   Guidelines for this query:  
+   - Return no more than 50 rows **unless** the user explicitly requests another limit.  
+   - Select only the columns needed to answer the question; avoid `SELECT *`.  
+   - If helpful, add `ORDER BY` on a meaningful column so the most interesting rows appear first.  
+   - ABSOLUTELY NO data-modification statements (INSERT, UPDATE, DELETE, DROP, …).  
+   - Double-check the SQL before executing.
 
-def create_tool_node_with_fallback(tools: list) -> RunnableWithFallbacks[Any, dict]:
-    """
-    Create a ToolNode with a fallback to handle errors and surface them to the agent.
-    """
-    return ToolNode(tools).with_fallbacks(
-        [RunnableLambda(handle_tool_error)], exception_key="error"
-    )
-
+4. Execute the query with the execution tool `sql_db_query`.  
+   If execution fails, inspect the error, revise the SQL, and try again.  
+   Repeat until the query runs successfully or you are certain the request
+   cannot be satisfied.
 
-class SubmitFinalAnswer(BaseModel):
-    """Submit the final answer to the user based on the query results."""
+5. Read the resulting rows and craft a concise, direct answer for the user.
+   If the result set is empty, explain that no matching data was found.
 
-    final_answer: str = Field(..., description="The final answer to the user")
+6. Include the final SQL query in your answer unless the user asks you not to.
 
-class QueryDBTool(BaseTool):
-    name = "db_query_tool"
-    description = """Execute a SQL query against the database and get back the result.
-        If the query is not correct, an error message will be returned.
-        If an error is returned, rewrite the query, check the query, and try again."""
+Remember:  
+- List tables → fetch schemas → write & verify SELECT → execute → answer.  
+- Never skip steps 1 or 2.  
+- Never perform DML.  
+- Keep answers focused on the user's question."""
 
-    db: SQLDatabase
-
-    def _run(self, query: str, run_manager: Optional[CallbackManagerForToolRun] = None):
-        result = self.db.run_no_throw(query)
-        if not result:
-            return "Error: Query failed. Please rewrite your query and try again."
-        return result
 
 class SqlAgentAPIWrapper(BaseModel):
+    model_config = ConfigDict(arbitrary_types_allowed=True)
+
     llm: BaseLanguageModel = Field(description="llm to use for sql agent")
     sql_address: str = Field(description="sql database address for SQLDatabase uri")
 
-    db: Optional[SQLDatabase]
-    list_tables_tool: Optional[BaseTool]
-    get_schema_tool: Optional[BaseTool]
-    db_query_tool: Optional[BaseTool]
-    query_check: Optional[Any]
-    query_gen: Optional[Any]
-    workflow: Optional[StateGraph]
-    app: Optional[Any]
-    schema_llm: Optional[Any]
-    query_check_llm: Optional[Any]
-    query_gen_llm: Optional[Any]
-
-    class Config:
-        arbitrary_types_allowed = True
+    db: Optional[SQLDatabase] = None
+    agent: Optional[CompiledGraph] = None
 
     def __init__(self, **kwargs):
         super().__init__(**kwargs)
         self.llm = kwargs.get('llm')
-
-        # todo 修改sql agent实现逻辑。此处逻辑只支持bishengLLM组件。原因是因为目前的实现必须实例化多个llm对象，每个llm对象绑定不同的tool
-        self.schema_llm = self.llm.__class__(model_id=self.llm.model_id, model_name=self.llm.model_name)
-        self.query_check_llm = self.llm.__class__(model_id=self.llm.model_id, model_name=self.llm.model_name)
-        self.query_gen_llm = self.llm.__class__(model_id=self.llm.model_id, model_name=self.llm.model_name)
         self.sql_address = kwargs.get('sql_address')
 
         self.db = SQLDatabase.from_uri(self.sql_address)
         toolkit = SQLDatabaseToolkit(db=self.db, llm=self.llm)
         tools = toolkit.get_tools()
-        self.list_tables_tool = next(tool for tool in tools if tool.name == "sql_db_list_tables")
-        self.get_schema_tool = next(tool for tool in tools if tool.name == "sql_db_schema")
-        self.db_query_tool = QueryDBTool(db=self.db)
-
-        self.query_check = self.init_query_check()
-        self.query_gen = self.init_query_gen()
-
-        # Define a new graph
-        self.workflow = StateGraph(State)
-        self.init_workflow()
-        self.app = self.workflow.compile(checkpointer=False, debug=True)
-
-    def init_workflow(self):
-        self.workflow.add_node("first_tool_call", self.first_tool_call)
-        self.workflow.add_node(
-            "list_tables_tool", create_tool_node_with_fallback([self.list_tables_tool])
-        )
-
-        self.workflow.add_node("get_schema_tool", create_tool_node_with_fallback([self.get_schema_tool]))
-
-        model_get_schema = self.schema_llm.bind_tools(
-            [self.get_schema_tool]
+        self.agent = create_react_agent(
+            self.llm,
+            tools,
+            prompt=_agent_system_prompt.format(dialect=self.db.dialect),
+            checkpointer=False,
         )
-        self.workflow.add_node(
-            "model_get_schema",
-            lambda state: {
-                "messages": [model_get_schema.invoke(state["messages"])],
-            },
-        )
-
-        self.workflow.add_node("query_gen", self.query_gen_node)
-        self.workflow.add_node("correct_query", self.model_check_query)
-
-        self.workflow.add_node("execute_query", create_tool_node_with_fallback([self.db_query_tool]))
-
-        self.workflow.add_edge(START, "first_tool_call")
-        self.workflow.add_edge("first_tool_call", "list_tables_tool")
-        self.workflow.add_edge("list_tables_tool", "model_get_schema")
-        self.workflow.add_edge("model_get_schema", "get_schema_tool")
-        self.workflow.add_edge("get_schema_tool", "query_gen")
-        self.workflow.add_conditional_edges(
-            "query_gen",
-            self.should_continue,
-        )
-        self.workflow.add_edge("correct_query", "execute_query")
-        self.workflow.add_edge("execute_query", "query_gen")
-
-    @staticmethod
-    def should_continue(state: State) -> Literal[END, "correct_query", "query_gen"]:
-        messages = state["messages"]
-        last_message = messages[-1]
-        # If there is a tool call, then we finish
-        if getattr(last_message, "tool_calls", None):
-            return END
-        if last_message.content.startswith("Error:"):
-            return "query_gen"
-        else:
-            return "correct_query"
-
-    def init_query_check(self):
-        query_check_system = """You are a SQL expert with a strong attention to detail.
-        Double check the SQLite query for common mistakes, including:
-        - Using NOT IN with NULL values
-        - Using UNION when UNION ALL should have been used
-        - Using BETWEEN for exclusive ranges
-        - Data type mismatch in predicates
-        - Properly quoting identifiers
-        - Using the correct number of arguments for functions
-        - Casting to the correct data type
-        - Using the proper columns for joins
-
-        If there are any of the above mistakes, rewrite the query. If there are no mistakes, just reproduce the original query.
-
-        You will call the appropriate tool to execute the query after running this check."""
-
-        query_check_prompt = ChatPromptTemplate.from_messages(
-            [("system", query_check_system), ("placeholder", "{messages}")]
-        )
-        query_check = query_check_prompt | self.query_check_llm.bind_tools(
-            [self.db_query_tool]
-        )
-        return query_check
-
-    def first_tool_call(self, state: State) -> dict[str, list[AIMessage]]:
-        return {
-            "messages": [
-                AIMessage(
-                    content="",
-                    tool_calls=[
-                        {
-                            "name": "sql_db_list_tables",
-                            "args": {},
-                            "id": "tool_abcd123",
-                        }
-                    ],
-                )
-            ]
-        }
-
-    def model_check_query(self, state: State) -> dict[str, list[AIMessage]]:
-        """
-        Use this tool to double-check if your query is correct before executing it.
-        """
-        return {"messages": [self.query_check.invoke({"messages": [state["messages"][-1]]})]}
-
-    def init_query_gen(self):
-        # Add a node for a model to generate a query based on the question and schema
-        query_gen_system = """You are a SQL expert with a strong attention to detail.Given an input question, output a syntactically correct SQL query to run, then look at the results of the query and return the answer.DO NOT call any tool besides SubmitFinalAnswer to submit the final answer.When generating the query:Output the SQL query that answers the input question without a tool call.Unless the user specifies a specific number of examples they wish to obtain, always limit your query to at most 10 results.You can order the results by a relevant column to return the most interesting examples in the database.Never query for all the columns from a specific table, only ask for the relevant columns given the question.If you get an error while executing a query, rewrite the query and try again.If you get an empty result set, you should try to rewrite the query to get a non-empty result set. NEVER make stuff up if you don't have enough information to answer the query... just say you don't have enough information.If you have enough information to answer the input question, simply invoke the appropriate tool to submit the final answer to the user.DO NOT make any DML statements (INSERT, UPDATE, DELETE, DROP etc.) to the database."""
-        query_gen_prompt = ChatPromptTemplate.from_messages(
-            [("system", query_gen_system), ("placeholder", "{messages}")]
-        )
-        query_gen = query_gen_prompt | self.query_gen_llm.bind_tools(
-            [SubmitFinalAnswer]
-        )
-        return query_gen
-
-    def query_gen_node(self, state: State) -> Any:
-        message = self.query_gen.invoke(state)
-
-        # Sometimes, the LLM will hallucinate and call the wrong tool. We need to catch this and return an error message.
-        tool_messages = []
-        if message.tool_calls:
-            for tc in message.tool_calls:
-                if tc["name"] != "SubmitFinalAnswer":
-                    tool_messages.append(
-                        ToolMessage(
-                            content=f"Error: The wrong tool was called: {tc['name']}. Please fix your mistakes. Remember to only call SubmitFinalAnswer to submit the final answer. Generated queries should be outputted WITHOUT a tool call.",
-                            tool_call_id=tc["id"],
-                        )
-                    )
-        else:
-            tool_messages = []
-        return {"messages": [message] + tool_messages}
 
     def run(self, query: str) -> str:
-        messages = self.app.invoke({"messages": [("user", query)]}, config={
-            'recursion_limit': 50
-        })
-        return messages["messages"][-1].tool_calls[0]["args"]["final_answer"]
+        messages = self.agent.invoke({"messages": [HumanMessage(content=query)]})
+        return messages["messages"][-1].content
 
     def arun(self, query: str) -> str:
         return self.run(query)
@@ -241,8 +82,8 @@ class SqlAgentInput(BaseModel):
 
 
 class SqlAgentTool(BaseTool):
-    name = "sql_agent"
-    description = "回答与 SQL 数据库有关的问题。给定用户问题，将从数据库中获取可用的表以及对应 DDL，生成 SQL 查询语句并进行执行，最终得到执行结果。"
+    name: str = "sql_agent"
+    description: str = "回答与 SQL 数据库有关的问题。给定用户问题，将从数据库中获取可用的表以及对应 DDL，生成 SQL 查询语句并进行执行，最终得到执行结果。"
     args_schema: Type[BaseModel] = SqlAgentInput
     api_wrapper: SqlAgentAPIWrapper
 

bisheng_langchain/input_output/input.py

@@ -1,12 +1,12 @@
 
 from typing import List, Optional
 
-from pydantic import BaseModel, Extra
+from pydantic import ConfigDict, BaseModel
 
 
 class InputNode(BaseModel):
     """Input组件，用来控制输入"""
-    input: Optional[List[str]]
+    input: Optional[List[str]] = None
 
     def text(self):
         return self.input
@@ -15,14 +15,10 @@ class InputNode(BaseModel):
 class VariableNode(BaseModel):
     """用来设置变量"""
     # key
-    variables: Optional[List[str]]
+    variables: Optional[List[str]] = None
     # vaulues
     variable_value: Optional[List[str]] = []
-
-    class Config:
-        """Configuration for this pydantic object."""
-
-        extra = Extra.forbid
+    model_config = ConfigDict(extra="forbid")
 
     def text(self):
         if self.variable_value:
@@ -36,9 +32,9 @@ class VariableNode(BaseModel):
 
 
 class InputFileNode(BaseModel):
-    file_path: Optional[str]
-    file_name: Optional[str]
-    file_type: Optional[str]  # tips for file
+    file_path: Optional[str] = None
+    file_name: Optional[str] = None
+    file_type: Optional[str] = None  # tips for file
     """Output组件，用来控制输出"""
 
     def text(self):

bisheng_langchain/input_output/output.py

@@ -5,7 +5,7 @@ from venv import logger
 from bisheng_langchain.chains import LoaderOutputChain
 from langchain.callbacks.manager import AsyncCallbackManagerForChainRun, CallbackManagerForChainRun
 from langchain.chains.base import Chain
-from pydantic import BaseModel, Extra
+from pydantic import ConfigDict, BaseModel
 
 _TEXT_COLOR_MAPPING = {
     'blue': '36;1',
@@ -52,11 +52,7 @@ class Report(Chain):
 
     input_key: str = 'report_name'  #: :meta private:
     output_key: str = 'text'  #: :meta private:
-
-    class Config:
-        """Configuration for this pydantic object."""
-        extra = Extra.forbid
-        arbitrary_types_allowed = True
+    model_config = ConfigDict(extra="forbid", arbitrary_types_allowed=True)
 
     @property
     def input_keys(self) -> List[str]: