Update app.py

app.py

@@ -134,15 +134,21 @@ async def access_control(request: Request, call_next):
 
 # 流式响应生成器
 async def stream_generator(response):
-    try:
-        async for line in response.aiter_lines():
-            if line:
-                yield f"data: {line}\n\n"
-                await asyncio.sleep(0.01)
-    except Exception as e:
-        print(f"Stream error: {str(e)}")
-        error_msg = json.dumps({"error": str(e)})
-        yield f"data: {error_msg}\n\n"
+    async for line in response.aiter_lines():
+        if line.strip():  # 确保行不是空的
+            try:
+                # 如果是JSON格式的数据，直接转发
+                if line.startswith('{'):
+                    yield f"data: {line}\n\n"
+                # 如果已经是SSE格式，直接发送
+                elif line.startswith('data: '):
+                    yield f"{line}\n\n"
+                # 其他情况，添加data前缀
+                else:
+                    yield f"data: {line}\n\n"
+            except Exception as e:
+                print(f"Error processing line: {str(e)}")
+                continue
 
 # 修改 handle_api_request 函数中的超时设置
 async def handle_api_request(url: str, headers: dict, method: str = "GET", body: dict = None, for_chat: bool = False):
@@ -309,41 +315,31 @@ async def chat_completions(request: Request):
         body = await request.body()
         body_json = json.loads(body)
         
-        # 强制使用流式响应
-        body_json["stream"] = True
-        
         headers = {
             "Content-Type": "application/json",
             "Accept": "text/event-stream",
-            "Connection": "keep-alive"
         }
         
         print("Starting chat completion request...")
         
-        async with httpx.AsyncClient(timeout=httpx.Timeout(60.0)) as client:
+        async with httpx.AsyncClient() as client:
             key = get_chat_key()
             headers["Authorization"] = f"Bearer {key}"
             
-            response = await client.post(
-                url=f"{Config.OPENAI_API_BASE}/chat/completions",
+            async with client.stream(
+                "POST",
+                f"{Config.OPENAI_API_BASE}/chat/completions",
                 headers=headers,
                 json=body_json,
-                timeout=60.0
-            )
-            
-            if response.status_code != 200:
-                raise HTTPException(status_code=response.status_code, detail="API request failed")
-            
-            return StreamingResponse(
-                stream_generator(response),
-                media_type="text/event-stream",
-                headers={
-                    "Cache-Control": "no-cache",
-                    "Connection": "keep-alive",
-                    "Content-Type": "text/event-stream",
-                    "X-Accel-Buffering": "no"
-                }
-            )
+                timeout=None
+            ) as response:
+                if response.status_code != 200:
+                    raise HTTPException(status_code=response.status_code, detail="API request failed")
+                
+                return StreamingResponse(
+                    stream_generator(response),
+                    media_type="text/event-stream"
+                )
         
     except Exception as e:
         print(f"Chat completion error: {str(e)}")