feat:对部分代码提出了修改建议，用TODO高亮了。修改了AgentManager添加智能体的方法，现在可以通过自动遍历Agents目录来获取，方便切换到private分支管理私有代码
This commit is contained in:
miluELK
parent
d56cec5002
commit
d53a32e6b8
13
.gitignore
vendored
13
.gitignore
vendored
@ -35,10 +35,17 @@ cache
|
||||
.cursor
|
||||
.trae
|
||||
.pytest_cache
|
||||
*.nogit*
|
||||
*.private*
|
||||
*.local*
|
||||
|
||||
### (企业私有代码 - 仅忽略敏感配置，不忽略代码文件)
|
||||
# 移除了 *.private* 和 *_private 规则，允许 Git 本地管理
|
||||
# 通过 .git/info/exclude 或本地分支管理私有代码
|
||||
*.secret*
|
||||
*.nogit*
|
||||
# *.local* 保留用于本地配置文件
|
||||
*.local.py
|
||||
*.local.js
|
||||
*.local.yaml
|
||||
|
||||
|
||||
*.pdf
|
||||
src/data
|
||||
|
||||
295
CLAUDE.md
Normal file
295
CLAUDE.md
Normal file
@ -0,0 +1,295 @@
|
||||
# CLAUDE.md
|
||||
|
||||
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
|
||||
|
||||
## 项目概述
|
||||
|
||||
**语析 (Yuxi-Know)** 是一个基于大模型的智能知识库与知识图谱问答系统，融合了 RAG（检索增强生成）与知识图谱技术。
|
||||
|
||||
- **技术栈**: FastAPI + Vue.js 3 + LangGraph + LightRAG + Neo4j + Milvus
|
||||
- **Python版本**: ≥3.11
|
||||
- **当前版本**: v0.3.0-beta
|
||||
- **许可证**: MIT
|
||||
|
||||
## 开发命令
|
||||
|
||||
### Docker 环境管理
|
||||
|
||||
```bash
|
||||
# 启动所有服务
|
||||
make start
|
||||
# 或: docker compose up -d
|
||||
|
||||
# 停止所有服务
|
||||
make stop
|
||||
# 或: docker compose down
|
||||
|
||||
# 查看后端日志
|
||||
make logs
|
||||
# 或: docker logs --tail=50 api-dev
|
||||
```
|
||||
|
||||
### 代码质量
|
||||
|
||||
```bash
|
||||
# 代码检查
|
||||
make lint
|
||||
# 或: uv run python -m ruff check .
|
||||
|
||||
# 代码格式化
|
||||
make format
|
||||
# 或: uv run ruff format . && uv run ruff check . --fix
|
||||
|
||||
# 查看格式化差异
|
||||
make format_diff
|
||||
# 或: uv run ruff format --diff .
|
||||
```
|
||||
|
||||
### 测试
|
||||
|
||||
```bash
|
||||
# 运行 API 单元测试
|
||||
make router-tests
|
||||
# 或: docker compose exec -T api uv run --group test pytest test/api
|
||||
|
||||
# 运行特定测试文件
|
||||
docker compose exec -T api uv run --group test pytest test/api/test_xxx.py
|
||||
|
||||
# 带覆盖率报告
|
||||
docker compose exec -T api uv run --group test pytest test/ --cov=src --cov-report=html
|
||||
```
|
||||
|
||||
## 架构设计
|
||||
|
||||
### 分层架构
|
||||
|
||||
```
|
||||
前端层 (Vue.js 3)
|
||||
↓ HTTP/REST API
|
||||
API层 (FastAPI)
|
||||
↓
|
||||
业务逻辑层 (Service)
|
||||
├─ 智能体系统 (Agent - LangGraph)
|
||||
├─ 知识库管理 (KnowledgeBase)
|
||||
├─ 任务管理 (Tasker)
|
||||
└─ 认证服务 (Auth)
|
||||
↓
|
||||
数据访问层
|
||||
├─ SQLite (对话、用户、统计)
|
||||
├─ Neo4j (知识图谱)
|
||||
├─ Milvus/Chroma (向量数据库)
|
||||
└─ MinIO (文件存储)
|
||||
```
|
||||
|
||||
### 核心目录结构
|
||||
|
||||
```
|
||||
src/ # 核心Python源代码
|
||||
├── agents/ # 智能体框架 (LangGraph)
|
||||
│ ├── chatbot/ # 聊天机器人智能体
|
||||
│ ├── react/ # ReAct智能体
|
||||
│ └── common/ # 通用组件 (工具、模型、上下文)
|
||||
├── knowledge/ # 知识库核心模块
|
||||
│ ├── implementations/ # 具体实现 (Milvus, Chroma, LightRAG)
|
||||
│ └── manager.py # 知识库管理器
|
||||
├── models/ # LLM模型加载 (chat, embed, rerank)
|
||||
├── storage/ # 数据存储层
|
||||
│ ├── db/ # SQLite ORM模型
|
||||
│ ├── minio/ # MinIO对象存储
|
||||
│ └── conversation/ # 对话历史管理
|
||||
├── config/ # 配置管理
|
||||
│ └── static/ # 静态配置 (models.yaml, agents_meta.yaml)
|
||||
├── plugins/ # 文档处理插件 (MinerU, PaddleOCR)
|
||||
└── utils/ # 工具函数
|
||||
|
||||
server/ # FastAPI应用层
|
||||
├── main.py # FastAPI应用入口
|
||||
├── routers/ # API路由 (7个主要路由)
|
||||
│ ├── auth_router.py # /api/auth/*
|
||||
│ ├── chat_router.py # /api/chat/*
|
||||
│ ├── knowledge_router.py # /api/knowledge/*
|
||||
│ ├── graph_router.py # /api/graph/*
|
||||
│ ├── dashboard_router.py # /api/dashboard/*
|
||||
│ ├── system_router.py # /api/system/*
|
||||
│ └── task_router.py # /api/tasks/*
|
||||
├── services/ # 业务逻辑服务
|
||||
│ └── tasker.py # 后台异步任务管理器
|
||||
└── utils/ # 服务层工具 (认证、用户管理)
|
||||
|
||||
web/ # Vue.js前端应用
|
||||
├── src/
|
||||
│ ├── views/ # 页面视图
|
||||
│ ├── components/ # 可复用组件
|
||||
│ ├── layouts/ # 布局模板 (AppLayout, BlankLayout)
|
||||
│ ├── router/ # Vue Router配置
|
||||
│ ├── stores/ # Pinia状态管理
|
||||
│ ├── apis/ # API调用层
|
||||
│ └── utils/ # 前端工具函数
|
||||
└── vite.config.js # Vite构建配置
|
||||
```
|
||||
|
||||
## 关键架构决策
|
||||
|
||||
### 1. 智能体系统 (LangGraph)
|
||||
|
||||
所有智能体基于 **LangGraph** 的状态机模型构建：
|
||||
|
||||
- **状态定义**: `messages[]`, `context` (包含 system_prompt, model, tools, mcps)
|
||||
- **节点**: `llm_call` (调用LLM), `dynamic_tools_node` (工具执行)
|
||||
- **边**: 通过 `tools_condition` 判断是否需要工具调用
|
||||
- **持久化**: 使用 SQLite Checkpointer 支持中断恢复
|
||||
|
||||
**可用工具**:
|
||||
- 知识库查询工具
|
||||
- 知识图谱查询工具
|
||||
- Web搜索工具 (Tavily)
|
||||
- 计算器工具
|
||||
- MySQL查询工具
|
||||
|
||||
### 2. 知识库系统
|
||||
|
||||
支持三种知识库实现：
|
||||
|
||||
- **Milvus**: 生产级向量数据库
|
||||
- **Chroma**: 轻量级向量数据库
|
||||
- **LightRAG**: 图增强检索系统，自动构建知识图谱
|
||||
|
||||
所有知识库通过 `KnowledgeBaseFactory` 工厂模式创建，由 `KnowledgeBaseManager` 统一管理。
|
||||
|
||||
### 3. 异步任务系统 (Tasker)
|
||||
|
||||
所有耗时操作（文档处理、知识图谱构建）均通过 **Tasker** 异步队列处理：
|
||||
|
||||
- **任务队列**: FIFO队列管理
|
||||
- **进度跟踪**: 实时更新任务状态
|
||||
- **结果存储**: 完成后存储到 SQLite
|
||||
- **失败重试**: 支持失败回调处理
|
||||
|
||||
### 4. 文档处理
|
||||
|
||||
支持多种格式文档：
|
||||
|
||||
- **文本**: `.txt`, `.md`
|
||||
- **文档**: `.doc`, `.docx`, `.pdf`
|
||||
- **网页**: `.html`, `.htm`
|
||||
- **数据**: `.json`, `.csv`, `.xls`, `.xlsx`
|
||||
- **图片**: `.jpg`, `.jpeg`, `.png`, `.bmp`, `.tiff` (需要 OCR)
|
||||
|
||||
处理流程：
|
||||
1. 选择合适的解析器 (MinerU for PDF, PaddleOCR for images)
|
||||
2. 文本提取与分块
|
||||
3. Embedding 向量化
|
||||
4. 存储到向量数据库
|
||||
5. 可选：知识图谱构建 (LightRAG)
|
||||
|
||||
## 配置管理
|
||||
|
||||
### 环境变量配置
|
||||
|
||||
项目根目录 `.env` 文件（从 `.env.template` 复制）:
|
||||
|
||||
```bash
|
||||
# 模型提供商 API Key (推荐使用硅基流动)
|
||||
SILICONFLOW_API_KEY=xxx
|
||||
OPENAI_API_KEY=xxx
|
||||
DEEPSEEK_API_KEY=xxx
|
||||
ZHIPUAI_API_KEY=xxx
|
||||
|
||||
# 数据库配置
|
||||
NEO4J_URI=bolt://graph:7687
|
||||
NEO4J_USERNAME=neo4j
|
||||
NEO4J_PASSWORD=0123456789
|
||||
MILVUS_URI=http://milvus:19530
|
||||
|
||||
# 文档处理服务 (需要GPU)
|
||||
MINERU_OCR_URI=http://mineru:30000
|
||||
PADDLEX_URI=http://paddlex:8080
|
||||
|
||||
# Web搜索 (可选)
|
||||
TAVILY_API_KEY=xxx
|
||||
```
|
||||
|
||||
### 模型配置
|
||||
|
||||
`src/config/static/models.yaml` 定义所有可用的 LLM 和 Embedding 模型：
|
||||
|
||||
- **LLM模型**: 支持 OpenAI, DeepSeek, SiliconFlow, Together, ZhipuAI 等
|
||||
- **Embedding模型**: 支持 OpenAI, HuggingFace 本地模型
|
||||
- **重排序模型**: 支持 Jina Reranker
|
||||
|
||||
## 数据库服务
|
||||
|
||||
### Docker Compose 服务
|
||||
|
||||
| 服务名 | 端口 | 用途 |
|
||||
|--------|------|------|
|
||||
| `api-dev` | 5050 | FastAPI后端服务 |
|
||||
| `web-dev` | 5173 | Vue.js前端服务 |
|
||||
| `graph` (Neo4j) | 7474/7687 | 知识图谱数据库 |
|
||||
| `milvus` | 19530 | 向量数据库 |
|
||||
| `minio` | 9000/9001 | 对象存储 |
|
||||
| `etcd` | 2379 | Milvus元数据存储 |
|
||||
|
||||
### 访问地址
|
||||
|
||||
- **Web界面**: http://localhost:5173
|
||||
- **API文档**: http://localhost:5050/docs
|
||||
- **Neo4j浏览器**: http://localhost:7474
|
||||
|
||||
## 开发最佳实践
|
||||
|
||||
### 后端开发
|
||||
|
||||
1. **路由**: 新增 API 路由在 `server/routers/` 下，并在 `server/routers/__init__.py` 注册
|
||||
2. **业务逻辑**: 核心业务代码放在 `src/` 下，避免在路由中编写复杂逻辑
|
||||
3. **异步任务**: 耗时操作必须使用 Tasker 异步处理
|
||||
4. **代码规范**: 提交前运行 `make lint` 和 `make format`
|
||||
5. **测试**: 为新增 API 编写单元测试 `test/api/test_xxx.py`
|
||||
|
||||
### 前端开发
|
||||
|
||||
1. **API调用**: 统一在 `web/src/apis/` 下定义 API 函数
|
||||
2. **状态管理**: 使用 Pinia stores (`web/src/stores/`)
|
||||
3. **组件**: 可复用组件放在 `web/src/components/`
|
||||
4. **样式**: 使用 Less，参考 `web/src/assets/css/base.css` 配色系统
|
||||
5. **热更新**: Vite 支持热更新，无需重启容器
|
||||
|
||||
### 智能体开发
|
||||
|
||||
1. **工具定义**: 在 `src/agents/common/tools.py` 定义新工具
|
||||
2. **工具包**: 复杂工具可创建 toolkit，如 `src/agents/common/toolkits/mysql/`
|
||||
3. **LangGraph状态**: 在智能体 graph 定义中添加自定义 State 字段
|
||||
4. **上下文管理**: 使用 `Context` 对象传递智能体配置
|
||||
|
||||
## 重要文件
|
||||
|
||||
| 文件 | 用途 |
|
||||
|------|------|
|
||||
| `src/config/app.py` | 全局配置管理 |
|
||||
| `server/main.py` | FastAPI应用入口 |
|
||||
| `src/knowledge/manager.py` | 知识库管理器 |
|
||||
| `src/agents/chatbot/graph.py` | LangGraph智能体定义 |
|
||||
| `server/services/tasker.py` | 异步任务管理器 |
|
||||
| `src/storage/db/models.py` | SQLAlchemy ORM模型 |
|
||||
| `docker-compose.yml` | 服务编排配置 |
|
||||
| `pyproject.toml` | Python依赖与工具配置 |
|
||||
| `web/src/router/index.js` | 前端路由配置 |
|
||||
|
||||
## 常见问题
|
||||
|
||||
1. **Milvus启动失败**: `docker compose up milvus -d && docker restart api-dev`
|
||||
2. **模型加载失败**: 检查 `.env` 中的 API Key 是否正确配置
|
||||
3. **任务卡住**: 查看 Tasker 日志，可能是工具执行超时
|
||||
4. **前端无法连接后端**: 检查 Vite 配置中的代理设置
|
||||
|
||||
## v0.3 重大变更
|
||||
|
||||
- **.env 文件位置**: 从 `src/.env` 移到项目根目录
|
||||
- **数据库重构**: 不再使用 MemorySaver，改用新的存储结构
|
||||
- **自定义模型移除**: 改用自定义 provider 方式
|
||||
- **新增功能**: Dashboard 统计、消息反馈、用户管理增强
|
||||
|
||||
## 文档资源
|
||||
|
||||
- **在线文档**: https://xerrors.github.io/Yuxi-Know/
|
||||
- **视频演示**: https://www.bilibili.com/video/BV1ETedzREgY/
|
||||
@ -15,7 +15,7 @@ services:
|
||||
- ./saves:/app/saves
|
||||
- ./test:/app/test
|
||||
- ./scripts:/app/scripts
|
||||
- ./.env :/app/.env
|
||||
- ./.env:/app/.env
|
||||
- ${MODEL_DIR:-./models}:/models # 使用默认值处理未定义的环境变量
|
||||
ports:
|
||||
- "5050:5050"
|
||||
@ -38,8 +38,8 @@ services:
|
||||
- MINIO_URI=${MINIO_URI:-http://milvus-minio:9000}
|
||||
- MODEL_DIR_IN_DOCKER=/models
|
||||
- RUNNING_IN_DOCKER=true
|
||||
- NO_PROXY=localhost,127.0.0.1,milvus,graph,milvus-minio,milvus-etcd-dev,etcd,minio,mineru,paddlex
|
||||
- no_proxy=localhost,127.0.0.1,milvus,graph,milvus-minio,milvus-etcd-dev,etcd,minio,mineru,paddlex
|
||||
- NO_PROXY=localhost,127.0.0.1,milvus,graph,milvus-minio,milvus-etcd-dev,etcd,minio,mineru,paddlex,api.siliconflow.cn
|
||||
- no_proxy=localhost,127.0.0.1,milvus,graph,milvus-minio,milvus-etcd-dev,etcd,minio,mineru,paddlex,api.siliconflow.cn
|
||||
# endregion api_envs
|
||||
command: uv run --no-dev uvicorn server.main:app --host 0.0.0.0 --port 5050 --reload
|
||||
restart: unless-stopped
|
||||
@ -145,7 +145,7 @@ services:
|
||||
- ETCD_SNAPSHOT_COUNT=50000
|
||||
volumes:
|
||||
- ./docker/volumes/milvus/etcd:/etcd
|
||||
command: etcd -advertise-client-urls=http://127.0.0.1:2379 -listen-client-urls http://0.0.0.0:2379 --data-dir /etcd
|
||||
command: etcd -advertise-client-urls=http://etcd:2379 -listen-client-urls http://0.0.0.0:2379 --data-dir /etcd
|
||||
healthcheck:
|
||||
test: ["CMD", "etcdctl", "endpoint", "health"]
|
||||
interval: 60s
|
||||
|
||||
@ -9,7 +9,7 @@ from fastapi.responses import JSONResponse
|
||||
from starlette.middleware.base import BaseHTTPMiddleware
|
||||
|
||||
from server.routers import router
|
||||
from server.services.tasker import tasker
|
||||
from server.utils.lifespan import lifespan
|
||||
from server.utils.auth_middleware import is_public_path
|
||||
from server.utils.common_utils import setup_logging
|
||||
|
||||
@ -24,7 +24,7 @@ RATE_LIMIT_ENDPOINTS = {("/api/auth/token", "POST")}
|
||||
_login_attempts: defaultdict[str, deque[float]] = defaultdict(deque)
|
||||
_attempt_lock = asyncio.Lock()
|
||||
|
||||
app = FastAPI()
|
||||
app = FastAPI(lifespan=lifespan)
|
||||
app.include_router(router, prefix="/api")
|
||||
|
||||
# CORS 设置
|
||||
@ -119,16 +119,5 @@ class AuthMiddleware(BaseHTTPMiddleware):
|
||||
app.add_middleware(LoginRateLimitMiddleware)
|
||||
app.add_middleware(AuthMiddleware)
|
||||
|
||||
|
||||
@app.on_event("startup")
|
||||
async def start_tasker() -> None:
|
||||
await tasker.start()
|
||||
|
||||
|
||||
@app.on_event("shutdown")
|
||||
async def stop_tasker() -> None:
|
||||
await tasker.shutdown()
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
uvicorn.run(app, host="0.0.0.0", port=5050, threads=10, workers=10, reload=True)
|
||||
|
||||
@ -112,7 +112,7 @@ async def get_agent(current_user: User = Depends(get_required_user)):
|
||||
metadata = yaml.safe_load(f)
|
||||
return {"agents": agents, "metadata": metadata}
|
||||
|
||||
|
||||
#TODO:[未完成]这个thread_id在前端是直接生成的1234，最好传入thread_id时做校验只允许uuid4
|
||||
@chat.post("/agent/{agent_id}")
|
||||
async def chat_agent(
|
||||
agent_id: str,
|
||||
@ -254,6 +254,7 @@ async def chat_agent(
|
||||
logger.error(f"Error saving messages from LangGraph state: {e}")
|
||||
logger.error(traceback.format_exc())
|
||||
|
||||
#TODO:[功能建议]针对需要人工审批后再执行的工具，可以使用langgraph的interrupt方法中断对话，等待用户输入后再使用command跳转回去
|
||||
async def stream_messages():
|
||||
# 代表服务端已经收到了请求
|
||||
yield make_chunk(status="init", meta=meta, msg=HumanMessage(content=query).model_dump())
|
||||
|
||||
15
server/utils/lifespan.py
Normal file
15
server/utils/lifespan.py
Normal file
@ -0,0 +1,15 @@
|
||||
import asyncio
|
||||
from contextlib import asynccontextmanager
|
||||
|
||||
from fastapi import FastAPI
|
||||
|
||||
from server.services import tasker
|
||||
|
||||
#TODO:[已完成]使用lifespan进行统一生命周期管理
|
||||
|
||||
@asynccontextmanager
|
||||
async def lifespan(app: FastAPI):
|
||||
await tasker.start()
|
||||
"""FastAPI lifespan事件管理器"""
|
||||
yield
|
||||
await tasker.shutdown()
|
||||
15
server/utils/singleton.py
Normal file
15
server/utils/singleton.py
Normal file
@ -0,0 +1,15 @@
|
||||
from threading import Lock
|
||||
|
||||
class SingletonMeta(type):
|
||||
"""
|
||||
This is a thread-safe implementation of Singleton.
|
||||
"""
|
||||
_instances = {}
|
||||
_lock: Lock = Lock()
|
||||
|
||||
def __call__(cls, *args, **kwargs):
|
||||
with cls._lock:
|
||||
if cls not in cls._instances:
|
||||
instance = super().__call__(*args, **kwargs)
|
||||
cls._instances[cls] = instance
|
||||
return cls._instances[cls]
|
||||
@ -1,7 +1,10 @@
|
||||
import asyncio
|
||||
import importlib
|
||||
import inspect
|
||||
from pathlib import Path
|
||||
|
||||
from .chatbot.graph import ChatbotAgent
|
||||
from .react.graph import ReActAgent
|
||||
from src.agents.common.base import BaseAgent
|
||||
from src.utils import logger
|
||||
|
||||
|
||||
class AgentManager:
|
||||
@ -35,10 +38,51 @@ class AgentManager:
|
||||
agents = self.get_agents()
|
||||
return await asyncio.gather(*[a.get_info() for a in agents])
|
||||
|
||||
def auto_discover_agents(self):
|
||||
"""自动发现并注册 src/agents/ 下的所有智能体。
|
||||
|
||||
遍历 src/agents/ 目录下的所有子文件夹，如果子文件夹包含 __init__.py，
|
||||
则尝试从中导入 BaseAgent 的子类并注册。(使用自动导入的方式，支持私有agent)
|
||||
"""
|
||||
# 获取 agents 目录的路径
|
||||
agents_dir = Path(__file__).parent
|
||||
|
||||
# 遍历所有子目录
|
||||
for item in agents_dir.iterdir():
|
||||
logger.info(f"尝试导入模块：{item}")
|
||||
# 跳过非目录、common 目录、__pycache__ 等
|
||||
if not item.is_dir() or item.name.startswith("_") or item.name == "common":
|
||||
continue
|
||||
|
||||
# 检查是否有 __init__.py 文件
|
||||
init_file = item / "__init__.py"
|
||||
if not init_file.exists():
|
||||
logger.warning(f"{item} 不是一个有效的模块")
|
||||
continue
|
||||
|
||||
# 尝试导入模块
|
||||
try:
|
||||
module_name = f"src.agents.{item.name}"
|
||||
module = importlib.import_module(module_name)
|
||||
|
||||
# 查找模块中所有 BaseAgent 的子类
|
||||
for name, obj in inspect.getmembers(module):
|
||||
if (
|
||||
inspect.isclass(obj)
|
||||
and issubclass(obj, BaseAgent)
|
||||
and obj is not BaseAgent
|
||||
and obj.__module__.startswith(module_name)
|
||||
):
|
||||
logger.info(f"自动发现智能体: {obj.__name__} 来自 {item.name}")
|
||||
self.register_agent(obj)
|
||||
|
||||
except Exception as e:
|
||||
logger.warning(f"无法从 {item.name} 加载智能体: {e}")
|
||||
|
||||
|
||||
agent_manager = AgentManager()
|
||||
agent_manager.register_agent(ChatbotAgent)
|
||||
agent_manager.register_agent(ReActAgent)
|
||||
# 自动发现并注册所有智能体
|
||||
agent_manager.auto_discover_agents()
|
||||
agent_manager.init_all_agents()
|
||||
|
||||
__all__ = ["agent_manager"]
|
||||
|
||||
@ -9,10 +9,20 @@ from src.agents.common.tools import get_buildin_tools
|
||||
from src.storage.minio import upload_image_to_minio
|
||||
from src.utils import logger
|
||||
|
||||
|
||||
@tool
|
||||
#TODO:[已完成]修改了tool定义的示例，使用更符合langgraph调用的方式
|
||||
@tool(name_or_callable="全能计算器",description="可以对给定的2个数字选择进行加减乘除四种计算")
|
||||
def calculator(a: float, b: float, operation: str) -> float:
|
||||
"""Calculate two numbers. operation: add, subtract, multiply, divide"""
|
||||
"""
|
||||
可以对给定的2个数字选择进行加减乘除四种计算
|
||||
|
||||
Args:
|
||||
a: 第一个数字
|
||||
b: 第二个数字
|
||||
operation: 计算操作符号，可以是add，subtract，multiply，divide
|
||||
|
||||
Returns:
|
||||
float: 最终的计算结果
|
||||
"""
|
||||
try:
|
||||
if operation == "add":
|
||||
return a + b
|
||||
|
||||
@ -0,0 +1,3 @@
|
||||
from .graph import ReActAgent
|
||||
|
||||
__all__ = ["ReActAgent"]
|
||||
@ -43,9 +43,11 @@ MODEL_NAMES:
|
||||
name: SiliconFlow
|
||||
url: https://cloud.siliconflow.cn/models
|
||||
base_url: https://api.siliconflow.cn/v1
|
||||
default: deepseek-ai/DeepSeek-V3.2-Exp
|
||||
default: Qwen/Qwen3-32B
|
||||
env: SILICONFLOW_API_KEY
|
||||
models:
|
||||
- Qwen/Qwen3-Next-80B-A3B-Thinking
|
||||
- Qwen/Qwen3-32B
|
||||
- deepseek-ai/DeepSeek-V3.2-Exp
|
||||
- Qwen/Qwen3-235B-A22B-Thinking-2507
|
||||
- Qwen/Qwen3-235B-A22B-Instruct-2507
|
||||
@ -98,6 +100,12 @@ MODEL_NAMES:
|
||||
- anthropic/claude-sonnet-4
|
||||
|
||||
EMBED_MODEL_INFO:
|
||||
siliconflow/Pro/BAAI/bge-m3:
|
||||
name: Pro/BAAI/bge-m3
|
||||
dimension: 1024
|
||||
base_url: https://api.siliconflow.cn/v1/embeddings
|
||||
api_key: SILICONFLOW_API_KEY
|
||||
|
||||
siliconflow/BAAI/bge-m3:
|
||||
name: BAAI/bge-m3
|
||||
dimension: 1024
|
||||
@ -127,6 +135,10 @@ EMBED_MODEL_INFO:
|
||||
dimension: 1024
|
||||
|
||||
RERANKER_LIST:
|
||||
siliconflow/Pro/BAAI/bge-reranker-v2-m3:
|
||||
name: Pro/BAAI/bge-reranker-v2-m3
|
||||
base_url: https://api.siliconflow.cn/v1/rerank
|
||||
api_key: SILICONFLOW_API_KEY
|
||||
|
||||
siliconflow/BAAI/bge-reranker-v2-m3:
|
||||
name: BAAI/bge-reranker-v2-m3
|
||||
|
||||
@ -12,6 +12,7 @@ from src.storage.db.models import Conversation, ConversationStats, Message, Tool
|
||||
from src.utils import logger
|
||||
from src.utils.datetime_utils import utc_now
|
||||
|
||||
#TODO:[未完成]待修改为异步版本
|
||||
|
||||
class ConversationManager:
|
||||
"""Manager for conversation storage operations"""
|
||||
|
||||
@ -5,6 +5,7 @@ from contextlib import contextmanager
|
||||
from sqlalchemy import create_engine
|
||||
from sqlalchemy.orm import sessionmaker
|
||||
|
||||
from server.utils.singleton import SingletonMeta
|
||||
from src import config
|
||||
from src.storage.db.models import Base, User
|
||||
from src.utils import logger
|
||||
@ -18,8 +19,10 @@ except ImportError:
|
||||
def validate_database_schema(db_path):
|
||||
return True, []
|
||||
|
||||
# TODO:[优化建议]需要将数据库修改为异步的aiosqlite或者异步mysql，缓存使用Redis存储
|
||||
# TODO:[已完成]为DBManager添加单例模式
|
||||
|
||||
class DBManager:
|
||||
class DBManager(metaclass=SingletonMeta):
|
||||
"""数据库管理器 - 只提供基础的数据库连接和会话管理"""
|
||||
|
||||
def __init__(self):
|
||||
|
||||
Loading…
Reference in New Issue
Block a user