Update main.py

main.py

@@ -330,214 +330,214 @@ class Blackbox:
                 logger.exception(f"Unexpected error during /api/chat request: {str(e)}")
                 return f"Unexpected error during /api/chat request: {str(e)}"
 
-    # ===============================
-    # 7. Initialize FastAPI App
-    # ===============================
-    app = FastAPI()
-
-    # ====================================
-    # 8. Define Middleware and Dependencies
-    # ====================================
-    @app.middleware("http")
-    async def security_middleware(request: Request, call_next):
-        client_ip = request.client.host
-        # Enforce that POST requests to /v1/chat/completions must have Content-Type: application/json
-        if request.method == "POST" and request.url.path == "/v1/chat/completions":
-            content_type = request.headers.get("Content-Type")
-            if content_type != "application/json":
-                logger.warning(f"Invalid Content-Type from IP: {client_ip} for path: {request.url.path}")
-                return JSONResponse(
-                    status_code=400,
-                    content={
-                        "error": {
-                            "message": "Content-Type must be application/json",
-                            "type": "invalid_request_error",
-                            "param": None,
-                            "code": None
-                        }
-                    },
-                )
-        response = await call_next(request)
-        return response
-
-    async def cleanup_rate_limit_stores():
-        """
-        Periodically cleans up stale entries in the rate_limit_store to prevent memory bloat.
-        """
-        while True:
-            current_time = time.time()
-            ips_to_delete = [ip for ip, value in rate_limit_store.items() if current_time - value["timestamp"] > RATE_LIMIT_WINDOW * 2]
-            for ip in ips_to_delete:
-                del rate_limit_store[ip]
-                logger.debug(f"Cleaned up rate_limit_store for IP: {ip}")
-            await asyncio.sleep(CLEANUP_INTERVAL)
-
-    async def rate_limiter_per_ip(request: Request):
-        """
-        Rate limiter that enforces a limit based on the client's IP address.
-        """
-        client_ip = request.client.host
-        current_time = time.time()
-
-        # Initialize or update the count and timestamp
-        if current_time - rate_limit_store[client_ip]["timestamp"] > RATE_LIMIT_WINDOW:
-            rate_limit_store[client_ip] = {"count": 1, "timestamp": current_time}
-        else:
-            if rate_limit_store[client_ip]["count"] >= RATE_LIMIT:
-                logger.warning(f"Rate limit exceeded for IP address: {client_ip}")
-                raise HTTPException(status_code=429, detail='Rate limit exceeded for IP address | NiansuhAI')
-            rate_limit_store[client_ip]["count"] += 1
-
-    async def get_api_key(request: Request, authorization: str = Header(None)) -> str:
-        """
-        Dependency to extract and validate the API key from the Authorization header.
-        """
-        client_ip = request.client.host
-        if authorization is None or not authorization.startswith('Bearer '):
-            logger.warning(f"Invalid or missing authorization header from IP: {client_ip}")
-            raise HTTPException(status_code=401, detail='Invalid authorization header format')
-        api_key = authorization[7:]
-        if api_key not in API_KEYS:
-            logger.warning(f"Invalid API key attempted: {api_key} from IP: {client_ip}")
-            raise HTTPException(status_code=401, detail='Invalid API key')
-        return api_key
-
-    # =====================================
-    # 9. Define FastAPI Event Handlers
-    # =====================================
-    @app.on_event("startup")
-    async def startup_event():
-        asyncio.create_task(cleanup_rate_limit_stores())
-        logger.info("Started rate limit store cleanup task.")
-
-    # ==========================================
-    # 10. Define FastAPI Endpoints
-    # ==========================================
-    @app.post("/v1/chat/completions", dependencies=[Depends(rate_limiter_per_ip)])
-    async def chat_completions(request: ChatRequest, req: Request, api_key: str = Depends(get_api_key)):
-        client_ip = req.client.host
-        # Redact user messages only for logging purposes
-        redacted_messages = [{"role": msg.role, "content": "[redacted]"} for msg in request.messages]
-
-        logger.info(f"Received chat completions request from API key: {api_key} | IP: {client_ip} | Model: {request.model} | Messages: {redacted_messages}")
-
-        try:
-            # Validate that the requested model is available
-            if request.model not in Blackbox.models and request.model not in Blackbox.model_aliases:
-                logger.warning(f"Attempt to use unavailable model: {request.model} from IP: {client_ip}")
-                raise HTTPException(status_code=400, detail="Requested model is not available.")
-
-            # Process the request with actual message content, but don't log it
-            response_content = await Blackbox.generate_response(
-                model=request.model,
-                messages=[{"role": msg.role, "content": msg.content} for msg in request.messages],
-                temperature=request.temperature,
-                max_tokens=request.max_tokens
-            )
+# ============================
+# 7. Initialize FastAPI App
+# ============================
+app = FastAPI()
 
-            logger.info(f"Completed response generation for API key: {api_key} | IP: {client_ip}")
-            return {
-                "id": f"chatcmpl-{uuid.uuid4()}",
-                "object": "chat.completion",
-                "created": int(datetime.now().timestamp()),
-                "model": request.model,
-                "choices": [
-                    {
-                        "index": 0,
-                        "message": {
-                            "role": "assistant",
-                            "content": response_content
-                        },
-                        "finish_reason": "stop"
+# ====================================
+# 8. Define Middleware and Dependencies
+# ====================================
+@app.middleware("http")
+async def security_middleware(request: Request, call_next):
+    client_ip = request.client.host
+    # Enforce that POST requests to /v1/chat/completions must have Content-Type: application/json
+    if request.method == "POST" and request.url.path == "/v1/chat/completions":
+        content_type = request.headers.get("Content-Type")
+        if content_type != "application/json":
+            logger.warning(f"Invalid Content-Type from IP: {client_ip} for path: {request.url.path}")
+            return JSONResponse(
+                status_code=400,
+                content={
+                    "error": {
+                        "message": "Content-Type must be application/json",
+                        "type": "invalid_request_error",
+                        "param": None,
+                        "code": None
                     }
-                ],
-                "usage": {
-                    "prompt_tokens": sum(len(msg.content.split()) for msg in request.messages),
-                    "completion_tokens": len(response_content.split()),
-                    "total_tokens": sum(len(msg.content.split()) for msg in request.messages) + len(response_content.split())
                 },
-            }
-        except ModelNotWorkingException as e:
-            logger.warning(f"Model not working: {e} | IP: {client_ip}")
-            raise HTTPException(status_code=503, detail=str(e))
-        except HTTPException as he:
-            logger.warning(f"HTTPException: {he.detail} | IP: {client_ip}")
-            raise he
-        except Exception as e:
-            logger.exception(f"An unexpected error occurred while processing the chat completions request from IP: {client_ip}.")
-            raise HTTPException(status_code=500, detail=str(e))
-
-    @app.get("/v1/models", dependencies=[Depends(rate_limiter_per_ip)])
-    async def get_models(req: Request):
-        client_ip = req.client.host
-        logger.info(f"Fetching available models from IP: {client_ip}")
-        return {"data": [{"id": model, "object": "model"} for model in Blackbox.models]}
-
-    @app.get("/v1/health", dependencies=[Depends(rate_limiter_per_ip)])
-    async def health_check(req: Request):
-        client_ip = req.client.host
-        logger.info(f"Health check requested from IP: {client_ip}")
-        return {"status": "ok"}
-
-    # ========================================
-    # 11. Define Custom Exception Handler
-    # ========================================
-    @app.exception_handler(HTTPException)
-    async def http_exception_handler(request: Request, exc: HTTPException):
-        client_ip = request.client.host
-        logger.error(f"HTTPException: {exc.detail} | Path: {request.url.path} | IP: {client_ip}")
-        return JSONResponse(
-            status_code=exc.status_code,
-            content={
-                "error": {
-                    "message": exc.detail,
-                    "type": "invalid_request_error",
-                    "param": None,
-                    "code": None
-                }
-            },
+            )
+    response = await call_next(request)
+    return response
+
+async def cleanup_rate_limit_stores():
+    """
+    Periodically cleans up stale entries in the rate_limit_store to prevent memory bloat.
+    """
+    while True:
+        current_time = time.time()
+        ips_to_delete = [ip for ip, value in rate_limit_store.items() if current_time - value["timestamp"] > RATE_LIMIT_WINDOW * 2]
+        for ip in ips_to_delete:
+            del rate_limit_store[ip]
+            logger.debug(f"Cleaned up rate_limit_store for IP: {ip}")
+        await asyncio.sleep(CLEANUP_INTERVAL)
+
+async def rate_limiter_per_ip(request: Request):
+    """
+    Rate limiter that enforces a limit based on the client's IP address.
+    """
+    client_ip = request.client.host
+    current_time = time.time()
+
+    # Initialize or update the count and timestamp
+    if current_time - rate_limit_store[client_ip]["timestamp"] > RATE_LIMIT_WINDOW:
+        rate_limit_store[client_ip] = {"count": 1, "timestamp": current_time}
+    else:
+        if rate_limit_store[client_ip]["count"] >= RATE_LIMIT:
+            logger.warning(f"Rate limit exceeded for IP address: {client_ip}")
+            raise HTTPException(status_code=429, detail='Rate limit exceeded for IP address | NiansuhAI')
+        rate_limit_store[client_ip]["count"] += 1
+
+async def get_api_key(request: Request, authorization: str = Header(None)) -> str:
+    """
+    Dependency to extract and validate the API key from the Authorization header.
+    """
+    client_ip = request.client.host
+    if authorization is None or not authorization.startswith('Bearer '):
+        logger.warning(f"Invalid or missing authorization header from IP: {client_ip}")
+        raise HTTPException(status_code=401, detail='Invalid authorization header format')
+    api_key = authorization[7:]
+    if api_key not in API_KEYS:
+        logger.warning(f"Invalid API key attempted: {api_key} from IP: {client_ip}")
+        raise HTTPException(status_code=401, detail='Invalid API key')
+    return api_key
+
+# =====================================
+# 9. Define FastAPI Event Handlers
+# =====================================
+@app.on_event("startup")
+async def startup_event():
+    asyncio.create_task(cleanup_rate_limit_stores())
+    logger.info("Started rate limit store cleanup task.")
+
+# ==========================================
+# 10. Define FastAPI Endpoints
+# ==========================================
+@app.post("/v1/chat/completions", dependencies=[Depends(rate_limiter_per_ip)])
+async def chat_completions(request: ChatRequest, req: Request, api_key: str = Depends(get_api_key)):
+    client_ip = req.client.host
+    # Redact user messages only for logging purposes
+    redacted_messages = [{"role": msg.role, "content": "[redacted]"} for msg in request.messages]
+
+    logger.info(f"Received chat completions request from API key: {api_key} | IP: {client_ip} | Model: {request.model} | Messages: {redacted_messages}")
+
+    try:
+        # Validate that the requested model is available
+        if request.model not in Blackbox.models and request.model not in Blackbox.model_aliases:
+            logger.warning(f"Attempt to use unavailable model: {request.model} from IP: {client_ip}")
+            raise HTTPException(status_code=400, detail="Requested model is not available.")
+
+        # Process the request with actual message content, but don't log it
+        response_content = await Blackbox.generate_response(
+            model=request.model,
+            messages=[{"role": msg.role, "content": msg.content} for msg in request.messages],
+            temperature=request.temperature,
+            max_tokens=request.max_tokens
         )
 
-    # ============================
-    # 12. Optional: Streaming Endpoint
-    # ============================
-    @app.post("/v1/chat/completions/stream", dependencies=[Depends(rate_limiter_per_ip)])
-    async def chat_completions_stream(request: ChatRequest, req: Request, api_key: str = Depends(get_api_key)):
-        client_ip = req.client.host
-        # Redact user messages only for logging purposes
-        redacted_messages = [{"role": msg.role, "content": "[redacted]"} for msg in request.messages]
-
-        logger.info(f"Received streaming chat completions request from API key: {api_key} | IP: {client_ip} | Model: {request.model} | Messages: {redacted_messages}")
+        logger.info(f"Completed response generation for API key: {api_key} | IP: {client_ip}")
+        return {
+            "id": f"chatcmpl-{uuid.uuid4()}",
+            "object": "chat.completion",
+            "created": int(datetime.now().timestamp()),
+            "model": request.model,
+            "choices": [
+                {
+                    "index": 0,
+                    "message": {
+                        "role": "assistant",
+                        "content": response_content
+                    },
+                    "finish_reason": "stop"
+                }
+            ],
+            "usage": {
+                "prompt_tokens": sum(len(msg.content.split()) for msg in request.messages),
+                "completion_tokens": len(response_content.split()),
+                "total_tokens": sum(len(msg.content.split()) for msg in request.messages) + len(response_content.split())
+            },
+        }
+    except ModelNotWorkingException as e:
+        logger.warning(f"Model not working: {e} | IP: {client_ip}")
+        raise HTTPException(status_code=503, detail=str(e))
+    except HTTPException as he:
+        logger.warning(f"HTTPException: {he.detail} | IP: {client_ip}")
+        raise he
+    except Exception as e:
+        logger.exception(f"An unexpected error occurred while processing the chat completions request from IP: {client_ip}.")
+        raise HTTPException(status_code=500, detail=str(e))
+
+@app.get("/v1/models", dependencies=[Depends(rate_limiter_per_ip)])
+async def get_models(req: Request):
+    client_ip = req.client.host
+    logger.info(f"Fetching available models from IP: {client_ip}")
+    return {"data": [{"id": model, "object": "model"} for model in Blackbox.models]}
+
+@app.get("/v1/health", dependencies=[Depends(rate_limiter_per_ip)])
+async def health_check(req: Request):
+    client_ip = req.client.host
+    logger.info(f"Health check requested from IP: {client_ip}")
+    return {"status": "ok"}
+
+# ========================================
+# 11. Define Custom Exception Handler
+# ========================================
+@app.exception_handler(HTTPException)
+async def http_exception_handler(request: Request, exc: HTTPException):
+    client_ip = request.client.host
+    logger.error(f"HTTPException: {exc.detail} | Path: {request.url.path} | IP: {client_ip}")
+    return JSONResponse(
+        status_code=exc.status_code,
+        content={
+            "error": {
+                "message": exc.detail,
+                "type": "invalid_request_error",
+                "param": None,
+                "code": None
+            }
+        },
+    )
 
-        try:
-            # Validate that the requested model is available
-            if request.model not in Blackbox.models and request.model not in Blackbox.model_aliases:
-                logger.warning(f"Attempt to use unavailable model: {request.model} from IP: {client_ip}")
-                raise HTTPException(status_code=400, detail="Requested model is not available.")
-
-            # Create an asynchronous generator for the response
-            async_generator = Blackbox.create_async_generator(
-                model=request.model,
-                messages=[{"role": msg.role, "content": msg.content} for msg in request.messages],
-                temperature=request.temperature,
-                max_tokens=request.max_tokens
-            )
+# ============================
+# 12. Optional: Streaming Endpoint
+# ============================
+@app.post("/v1/chat/completions/stream", dependencies=[Depends(rate_limiter_per_ip)])
+async def chat_completions_stream(request: ChatRequest, req: Request, api_key: str = Depends(get_api_key)):
+    client_ip = req.client.host
+    # Redact user messages only for logging purposes
+    redacted_messages = [{"role": msg.role, "content": "[redacted]"} for msg in request.messages]
+
+    logger.info(f"Received streaming chat completions request from API key: {api_key} | IP: {client_ip} | Model: {request.model} | Messages: {redacted_messages}")
+
+    try:
+        # Validate that the requested model is available
+        if request.model not in Blackbox.models and request.model not in Blackbox.model_aliases:
+            logger.warning(f"Attempt to use unavailable model: {request.model} from IP: {client_ip}")
+            raise HTTPException(status_code=400, detail="Requested model is not available.")
+
+        # Create an asynchronous generator for the response
+        async_generator = Blackbox.create_async_generator(
+            model=request.model,
+            messages=[{"role": msg.role, "content": msg.content} for msg in request.messages],
+            temperature=request.temperature,
+            max_tokens=request.max_tokens
+        )
 
-            logger.info(f"Started streaming response for API key: {api_key} | IP: {client_ip}")
-            return StreamingResponse(async_generator, media_type="text/event-stream")
-        except ModelNotWorkingException as e:
-            logger.warning(f"Model not working: {e} | IP: {client_ip}")
-            raise HTTPException(status_code=503, detail=str(e))
-        except HTTPException as he:
-            logger.warning(f"HTTPException: {he.detail} | IP: {client_ip}")
-            raise he
-        except Exception as e:
-            logger.exception(f"An unexpected error occurred while processing the streaming chat completions request from IP: {client_ip}.")
-            raise HTTPException(status_code=500, detail=str(e))
-
-    # ========================================
-    # 13. Run the Application with Uvicorn
-    # ========================================
-    if __name__ == "__main__":
-        import uvicorn
-        uvicorn.run(app, host="0.0.0.0", port=8000)
+        logger.info(f"Started streaming response for API key: {api_key} | IP: {client_ip}")
+        return StreamingResponse(async_generator, media_type="text/event-stream")
+    except ModelNotWorkingException as e:
+        logger.warning(f"Model not working: {e} | IP: {client_ip}")
+        raise HTTPException(status_code=503, detail=str(e))
+    except HTTPException as he:
+        logger.warning(f"HTTPException: {he.detail} | IP: {client_ip}")
+        raise he
+    except Exception as e:
+        logger.exception(f"An unexpected error occurred while processing the streaming chat completions request from IP: {client_ip}.")
+        raise HTTPException(status_code=500, detail=str(e))
+
+# ========================================
+# 13. Run the Application with Uvicorn
+# ========================================
+if __name__ == "__main__":
+    import uvicorn
+    uvicorn.run(app, host="0.0.0.0", port=8000)