Spaces:
Sleeping
Sleeping
| from flask import Flask, request, Response, jsonify | |
| import requests | |
| import threading | |
| import time | |
| import random | |
| import logging | |
| from typing import Optional | |
| app = Flask(__name__) | |
| # ========================= | |
| # CONFIG | |
| # ========================= | |
| Server = [ | |
| "https://Techbitforge-hello_server.hf.space" | |
| ] | |
| HF_SERVERS = [ | |
| "https://techbitforge-ollamaserver.hf.space/" | |
| ] | |
| PING_INTERVAL = 7200 | |
| REQUEST_TIMEOUT = 300 | |
| # ========================= | |
| # LOGGING | |
| # ========================= | |
| logging.basicConfig( | |
| level=logging.INFO, | |
| format="%(asctime)s — %(levelname)s — %(message)s" | |
| ) | |
| logger = logging.getLogger("proxy") | |
| def pick_server() -> str: | |
| return random.choice(HF_SERVERS) | |
| def error_response(message: str, code: int = 500): | |
| logger.error(message) | |
| return jsonify({ | |
| "error": { | |
| "message": message, | |
| "type": "proxy_error" | |
| } | |
| }), code | |
| # ========================= | |
| # CHAT COMPLETIONS PROXY | |
| # ========================= | |
| def chat_completions(): | |
| try: | |
| payload = request.get_json(force=True) | |
| except Exception: | |
| return error_response("Invalid JSON payload", 400) | |
| headers = dict(request.headers) | |
| headers.pop("Host", None) | |
| target = pick_server() + "/v1/chat/completions" | |
| stream = payload.get("stream", False) | |
| logger.info(f"Forwarding request to {target} | Stream={stream}") | |
| try: | |
| if stream: | |
| def generate(): | |
| with requests.post( | |
| target, | |
| json=payload, | |
| headers=headers, | |
| stream=True, | |
| timeout=REQUEST_TIMEOUT | |
| ) as res: | |
| res.raise_for_status() | |
| for line in res.iter_lines(): | |
| if line: | |
| yield line + b"\n\n" | |
| return Response( | |
| generate(), | |
| mimetype="text/event-stream", | |
| headers={"Cache-Control": "no-cache", "X-Accel-Buffering": "no"} | |
| ) | |
| res = requests.post( | |
| target, | |
| json=payload, | |
| headers=headers, | |
| timeout=REQUEST_TIMEOUT | |
| ) | |
| res.raise_for_status() | |
| return jsonify(res.json()) | |
| except requests.exceptions.Timeout: | |
| return error_response("Upstream server timeout", 504) | |
| except requests.exceptions.ConnectionError: | |
| return error_response("Failed to connect to upstream server", 502) | |
| except Exception as e: | |
| return error_response(str(e), 500) | |
| # ========================= | |
| # MODELS ROUTE | |
| # ========================= | |
| def models(): | |
| url = pick_server() | |
| target = f"{url}/models" | |
| logger.info(f"Fetching models from {target}") | |
| try: | |
| res = requests.get(target, timeout=20) | |
| res.raise_for_status() | |
| return jsonify(res.json()) | |
| except Exception as e: | |
| return error_response(f"Failed to fetch models: {e}", 500) | |
| # ========================= | |
| # HEALTH CHECK | |
| # ========================= | |
| def home(): | |
| return jsonify({"status": "running", "service": "HF Proxy"}) | |
| import requests | |
| import threading | |
| import time | |
| PING_INTERVAL = 3600 # 24 hours | |
| def keep_alive_worker(): | |
| url = "https://techbitforge-server.hf.space/" | |
| while True: | |
| try: | |
| response = requests.get(url, timeout=120) | |
| if response.status_code == 200: | |
| print("Success: server is alive") | |
| else: | |
| print(f"Server returned status code: {response.status_code}") | |
| except requests.RequestException as e: | |
| print(f"Ping failed: {e}") | |
| # Wait 24 hours before the next ping | |
| time.sleep(PING_INTERVAL) | |
| def start_background_thread(): | |
| thread = threading.Thread( | |
| target=keep_alive_worker, | |
| daemon=True | |
| ) | |
| thread.start() | |
| # ========================= | |
| # MAIN | |
| # ========================= | |
| if __name__ == "__main__": | |
| start_background_thread() | |
| app.run( | |
| host="0.0.0.0", | |
| port=7860, | |
| threaded=True, | |
| debug=False | |
| ) | |