1.最小化import

2.重构代码结构，独立config类与ConnectionManager类
3.引入toml来管理配置文件
4.引入pydantic的basemodel来简化config类的建构
5.新增信号响应系统，linux上可以响应systemctl reload了
6.修改了部分注释

ConnectionManager.py

@@ -0,0 +1,29 @@
+# -*- coding: utf-8 -*-
+# @Time    : 2024/10/26 下午3:35
+# @Author  : 河瞬
+# @FileName: ConnectionManager.py
+# @Software: PyCharm
+# @Github  :
+from typing import List
+from fastapi import WebSocket
+
+
+# WebSocket连接管理器
+class ConnectionManager:
+    def __init__(self):
+        self.active_connections: List[WebSocket] = []
+
+    async def connect(self, websocket: WebSocket):
+        await websocket.accept()
+        self.active_connections.append(websocket)
+
+    def disconnect(self, websocket: WebSocket):
+        self.active_connections.remove(websocket)
+
+    @staticmethod
+    async def send_personal_message(message: str, websocket: WebSocket):
+        await websocket.send_text(message)
+
+    async def broadcast_json(self, data: dict):
+        for connection in self.active_connections:
+            await connection.send_json(data)

config.py

@@ -0,0 +1,24 @@
+# -*- coding: utf-8 -*-
+# @Time    : 2024/10/26 下午3:25
+# @Author  : 河瞬
+# @FileName: config.py
+# @Software: PyCharm
+import toml
+from pydantic import BaseModel
+
+
+class Config(BaseModel):
+    key: str = ""
+    prompt: str = ""
+    model: str = "chatgpt-4o-latest"
+    base_url: str = ""
+
+    def save(self, config_file):
+        with open(config_file, "w", encoding='utf-8') as f:
+            toml.dump(self.model_dump(), f)
+
+    @classmethod
+    def load(cls, config_file):
+        with open(config_file, "r", encoding='utf-8') as f:
+            config_data = toml.load(f)
+        return cls(**config_data)

main.py

@@ -1,60 +1,40 @@
 from fastapi import FastAPI, WebSocket, WebSocketDisconnect, UploadFile, File
 from fastapi.responses import StreamingResponse, HTMLResponse
-from typing import List
+from base64 import b64encode
-import base64
+from openai import OpenAI
-import openai
+from os.path import isfile
-import signal
+from sys import platform
+from config import Config
+from ConnectionManager import ConnectionManager
 
 app = FastAPI()
-key = ""
-prompt = ""
 
 
-def get_key():
+def signal_handler():
-    with open("key", "r") as f:
+    print("Received SIGHUP, reloading config")
-        k = f.read()
+    global config, client
-    return k
+    config.load(config_file)
+    client = OpenAI(api_key=config.key, base_url=config.base_url)
 
 
-def get_prompt():
+if platform != 'win32':
-    with open("prompt", "r", encoding="utf-8") as f:
+    import signal
-        p = f.read()
+
-    return p
+    signal.signal(signal.SIGHUP, lambda signum, frame: signal_handler())
 
 
-def load_config():
+def init():
-    global key, prompt
+    global config
-    key = get_key()
+    if not isfile(config_file):
-    prompt = get_prompt()
+        config = Config()
+        config.save(config_file)
 
 
-signal.signal(signal.SIGHUP, load_config())
+config_file = "./config.toml"
-
+init()
-load_config()
+config = Config.load(config_file)
-client = openai.OpenAI(api_key=key,
-                       base_url="https://open.bigmodel.cn/api/paas/v4/")
-
-
-# WebSocket连接管理器
-class ConnectionManager:
-    def __init__(self):
-        self.active_connections: List[WebSocket] = []
-
-    async def connect(self, websocket: WebSocket):
-        await websocket.accept()
-        self.active_connections.append(websocket)
-
-    def disconnect(self, websocket: WebSocket):
-        self.active_connections.remove(websocket)
-
-    @staticmethod
-    async def send_personal_message(message: str, websocket: WebSocket):
-        await websocket.send_text(message)
-
-    async def broadcast_json(self, data: dict):
-        for connection in self.active_connections:
-            await connection.send_json(data)
 
+client = OpenAI(api_key=config.key, base_url=config.base_url)
 
 manager = ConnectionManager()
 
@@ -78,18 +58,18 @@ async def event(websocket: WebSocket):
 async def predict(file: UploadFile = File(...)):
     # 读取图片文件并转换为base64编码
     image_data = await file.read()
-    image_base64 = base64.b64encode(image_data).decode('utf-8')
+    image_base64 = b64encode(image_data).decode('utf-8')
 
     # 构造请求给API
     response = client.chat.completions.create(
-        model="glm-4v",
+        model=config.model,
         messages=[
             {
                 "role": "user",
                 "content": [
                     {
                         "type": "text",
-                        "text": get_prompt(),
+                        "text": config.prompt,
                     },
                     {
                         "type": "image_url",
@@ -117,6 +97,7 @@ async def predict(file: UploadFile = File(...)):
     return StreamingResponse(stream_response(), media_type="text/plain")
 
 
+# html页面
 @app.get("/terminal", response_class=HTMLResponse)
 async def test():
     with open("html/terminal.html", "r", encoding="utf-8") as f:

requirements.txt

@@ -3,3 +3,6 @@ openai~=1.52.2
 fastapi~=0.115.3
 uvicorn~=0.20.0
 python-multipart~=0.0.12
+requests~=2.27.1
+toml~=0.10.2
+pydantic~=2.9.2