导入langchain.text_splitter时触发Pydantic @root_validator错误求助

问题场景 #

在MacOS 13.4.1系统的VSCode中运行LangChain相关Python代码时，导入langchain模块触发PydanticUserError，错误提示明确要求：使用默认pre=False的@root_validator必须指定skip_on_failure=True，且该装饰器已废弃，需替换为@model_validator。

运行代码如下：

from dotenv import load_dotenv
import os
import streamlit as st
from PyPDF2 import PdfReader
from langchain.text_splitter import CharacterTextSplitter


def main():
    load_dotenv()
    # print(os.getenv("OPENAI_API_KEY"))
    st.set_page_config(page_title="Select the Data PDF")
    st.header("Load your PDF below: ⚡︎")

    pdf = st.file_uploader("Upload your PDF", type="pdf")

    if pdf is not None:
        pdf_reader = PdfReader(pdf)

        # the PdfReader reads in chunks of one page each
        text = ""
        for page in pdf_reader.pages:
            text += page.extract_text()
            st.write(text)
   
    # split text into chunks (inside outer if not None)
    # the CharacterTextSplitter class has properties that  need to be set...
    # so pass in property parameters to the class inititializer
    text_splitter = CharacterTextSplitter(
        separator="\n",
        chunk_size=1000,
        chunk_overlap=200,
        length_function=len #python's basic length function
        )
    chunks = text_splitter.split_text(text)
    st.write(chunks)



if __name__ == '__main__' :
    main()

错误栈信息：

PydanticUserError: If you use `@root_validator` with pre=False (the default) you MUST specify `skip_on_failure=True`. Note that `@root_validator` is deprecated and should be replaced with `@model_validator`. For further information visit https://errors.pydantic.dev/2.0/u/root-validator-pre-skip
Traceback:
File "/Users/jamesallison/Desktop/Python Apps Temp/LangchainPDF/.venv/lib/python3.8/site-packages/streamlit/runtime/scriptrunner/script_runner.py", line 552, in _run_script
    exec(code, module.__dict__)
File "/Users/jamesallison/Desktop/Python Apps Temp/LangchainPDF/app.py", line 6, in <module>
    from langchain.embeddings import ImportError
File "/Users/jamesallison/Desktop/Python Apps Temp/LangchainPDF/.venv/lib/python3.8/site-packages/langchain/__init__.py", line 8, in <module>
    from langchain.agents import MRKLChain, ReActChain, SelfAskWithSearchChain
File "/Users/jamesallison/Desktop/Python Apps Temp/LangchainPDF/.venv/lib/python3.8/site-packages/langchain/agents/__init__.py", line 2, in <module>
    from langchain.agents.agent import Agent
File "/Users/jamesallison/Desktop/Python Apps Temp/LangchainPDF/.venv/lib/python3.8/site-packages/langchain/agents/agent.py", line 11, in <module>
    from langchain.chains.llm import LLMChain
File "/Users/jamesallison/Desktop/Python Apps Temp/LangchainPDF/.venv/lib/python3.8/site-packages/langchain/chains/__init__.py", line 2, in <module>
    from langchain.chains.conversation.base import ConversationChain
File "/Users/jamesallison/Desktop/Python Apps Temp/LangchainPDF/.venv/lib/python3.8/site-packages/langchain/chains/conversation/base.py", line 7, in <module>
    from langchain.chains.conversation.memory import ConversationBufferMemory
File "/Users/jamesallison/Desktop/Python Apps Temp/LangchainPDF/.venv/lib/python3.8/site-packages/langchain/chains/conversation/memory.py", line 7, in <module>
    from langchain.chains.conversation.prompt import SUMMARY_PROMPT
File "/Users/jamesallison/Desktop/Python Apps Temp/LangchainPDF/.venv/lib/python3.8/site-packages/langchain/chains/conversation/prompt.py", line 2, in <module>
    from langchain.prompts.prompt import PromptTemplate
File "/Users/jamesallison/Desktop/Python Apps Temp/LangchainPDF/.venv/lib/python3.8/site-packages/langchain/prompts/__init__.py", line 2, in <module>
    from langchain.prompts.base import BasePromptTemplate
File "/Users/jamesallison/Desktop/Python Apps Temp/LangchainPDF/.venv/lib/python3.8/site-packages/langchain/prompts/base.py", line 35, in <module>
    class BasePromptTemplate(BaseModel, ABC):
File "/Users/jamesallison/Desktop/Python Apps Temp/LangchainPDF/.venv/lib/python3.8/site-packages/langchain/prompts/base.py", line 41, in BasePromptTemplate
    @root_validator()
File "/Users/jamesallison/Desktop/Python Apps Temp/LangchainPDF/.venv/lib/python3.8/site-packages/pydantic/deprecated/class_validators.py", line 228, in root_validator
    raise PydanticUserError(

问题原因 #

核心原因是LangChain版本与Pydantic版本不兼容：旧版LangChain依赖Pydantic 1.x的@root_validator装饰器，而Pydantic 2.x对该装饰器的使用规则做了强制修改，且标记其为废弃，要求使用新的@model_validator装饰器。

修复方法 #

方法1：降级Pydantic到1.x兼容版本 #

激活虚拟环境后执行：

pip install pydantic==1.10.12

此方法无需修改代码，直接让依赖版本匹配，快速解决问题。

方法2：升级LangChain到适配Pydantic 2.x的版本 #

若想保留Pydantic 2.x，可升级LangChain至最新稳定版：

pip install --upgrade langchain

新版LangChain已全面适配Pydantic 2.x的API，替换了所有旧的@root_validator调用。

额外代码修正 #

注意代码第6行的from langchain.embeddings import ImportError是错误语句，既未用到该模块，写法也不正确，需删除该行，否则会触发额外的导入错误。

内容的提问来源于stack exchange，提问作者user1974376