agent_proj/docs/IMPORTS_SUMMARY.md

项目 Import 包汇总

本项目所有 Python 文件中使用的 import 包整理

---

一、第三方包 (pip install)

1. LangChain 生态

包名	用途	来源
langchain	Agent 核心框架	pip install langchain
langchain-core	核心抽象类	pip install langchain-core
langchain-community	第三方集成 (通义千问/DashScope)	pip install langchain-community
langchain-chroma	Chroma 向量数据库	pip install langchain-chroma
langchain-text-splitters	文本分片	pip install langchain-text-splitters
langgraph	图执行引擎 (中间件)	pip install langgraph

1.1 LangChain 具体导入

# Agent
from langchain.agents import create_agent, AgentState
from langchain.agents.middleware import wrap_tool_call, before_model, dynamic_prompt, ModelRequest

# Tools
from langchain_core.tools import tool
from langchain.tools.tool_node import ToolCallRequest

# Messages
from langchain_core.messages import ToolMessage

# Graph
from langgraph.types import Command
from langgraph.runtime import Runtime

# Prompts
from langchain_core.prompts import PromptTemplate

# Output Parsers
from langchain_core.output_parsers import StrOutputParser

# Embeddings
from langchain_core.embeddings import Embeddings

1.2 社区包导入

# LLM - 通义千问
from langchain_community.chat_models.tongyi import ChatTongyi, BaseChatModel

# Embedding - DashScope
from langchain_community.embeddings import DashScopeEmbeddings

# Document Loaders
from langchain_community.document_loaders import PyPDFLoader, TextLoader

# Document
from langchain_community.docstore.document import Document

1.3 Chroma & Text Splitters

# Vector Store
from langchain_chroma import Chroma

# Text Splitter
from langchain_text_splitters import RecursiveCharacterTextSplitter

---

2. Web 框架

包名	用途
streamlit	Web UI 界面

import streamlit as st

---

3. 配置管理

包名	用途
yaml	YAML 配置文件解析

import yaml

---

4. 数据处理

包名	用途
pandas	数据处理 (CSV)

import pandas

---

5. 其他工具

包名	用途
hashlib	MD5 计算 (标准库)
os	文件路径操作 (标准库)
logging	日志管理 (标准库)
time	时间控制 (标准库)
typing	类型提示 (标准库)
abc	抽象类 (标准库)
datetime	日期时间 (标准库)
random	随机数 (标准库)

---

二、标准库导入

# 文件与路径
import os
import os.path

# 哈希计算
import hashlib

# 日志
import logging

# 时间
import time

# 日期
from datetime import datetime

# 随机数
import random

# 类型提示
from typing import Optional, List, Callable

# 抽象基类
from abc import ABC, abstractmethod

---

三、按文件分类的导入

app.py

import time
import streamlit as st
from agent.react_agent import ReactAgent

01_Agent.py / 02_Stream.py / 03_ReAct.py

from langchain.agents import create_agent
from langchain_community.chat_models.tongyi import ChatTongyi
from langchain_core.tools import tool

agent/react_agent.py

from langchain.agents import create_agent
from model.factory import chat_model
from utils.prompt_loader import load_system_prompts
from agent.tools.agent_tools import (...)
from agent.tools.middleware import (...)

agent/tools/agent_tools.py

import pandas
from langchain_core.tools import tool
from rag.rag_service import RagSummarizeService
import random
from utils.config_handler import agent_conf
from utils.path_tool import get_abs_path
from utils.logger_handler import logger
import os

agent/tools/middleware.py

from langchain.agents.middleware import wrap_tool_call, before_model, dynamic_prompt, ModelRequest
from langchain.tools.tool_node import ToolCallRequest
from typing import Callable
from langchain_core.messages import ToolMessage
from langgraph.types import Command
from utils.logger_handler import logger
from langchain.agents import AgentState
from langgraph.runtime import Runtime
from utils.prompt_loader import load_system_prompts, load_report_prompts

rag/rag_service.py

from langchain_core.output_parsers import StrOutputParser
from rag.vector_store import VectorStoreSerivce
from utils.prompt_loader import load_rag_prompts
from langchain_core.prompts import PromptTemplate
from model.factory import chat_model
from langchain_community.docstore.document import Document
from typing import List
from utils.logger_handler import logger

rag/vector_store.py

import os.path
from langchain_chroma import Chroma
from utils.config_handler import chroma_conf
from model.factory import embedding_model
from langchain_text_splitters import RecursiveCharacterTextSplitter
import os
from utils.path_tool import get_abs_path
from utils.file_handler import pdf_loader, txt_loader, listdir_with_allowed_type, get_file_md5_hex
from utils.logger_handler import logger
from langchain_community.docstore.document import Document as Documents

model/factory.py

from abc import ABC, abstractmethod
from typing import Optional
from utils.config_handler import rag_conf
from langchain_core.embeddings import Embeddings
from langchain_community.chat_models.tongyi import BaseChatModel
from langchain_community.chat_models.tongyi import ChatTongyi
from langchain_community.embeddings import DashScopeEmbeddings

utils/config_handler.py

import yaml
from utils.path_tool import get_abs_path

utils/logger_handler.py

import logging
import os
from utils.path_tool import get_abs_path
from datetime import datetime

utils/prompt_loader.py

from utils.config_handler import prompts_conf
from utils.path_tool import get_abs_path
from utils.logger_handler import logger

utils/file_handler.py

import hashlib
import os
from utils.logger_handler import logger
from langchain_community.document_loaders import PyPDFLoader, TextLoader
from langchain_community.docstore.document import Document
from typing import *
from utils.path_tool import get_abs_path

utils/path_tool.py

import os

---

四、requirements.txt 参考

langchain>=0.3.0
langchain-core>=0.3.0
langchain-community>=0.3.0
langchain-chroma>=0.1.0
langchain-text-splitters>=0.3.0
langgraph>=0.2.0
streamlit
pyyaml
pandas

---

五、包导入拓扑图

标准库 (内置)
├── os, os.path
├── hashlib
├── logging
├── time
├── datetime
├── random
├── typing
├── abc
└── pathlib

第三方包 (pip install)
├── streamlit (Web)
├── yaml (配置)
├── pandas (数据)
│
└── LangChain 生态
    ├── langchain-core
    │   ├── tools (tool)
    │   ├── prompts (PromptTemplate)
    │   ├── output_parsers (StrOutputParser)
    │   ├── messages (ToolMessage)
    │   └── embeddings (Embeddings)
    │
    ├── langchain
    │   ├── agents (create_agent, AgentState)
    │   └── agents.middleware
    │
    ├── langchain-community
    │   ├── chat_models.tongyi (ChatTongyi)
    │   ├── embeddings (DashScopeEmbeddings)
    │   ├── document_loaders (PyPDFLoader, TextLoader)
    │   └── docstore.document (Document)
    │
    ├── langchain-chroma (Chroma)
    │
    └── langchain-text-splitters
        └── RecursiveCharacterTextSplitter

LangGraph
    ├── types (Command)
    └── runtime (Runtime)