app.py

"""Production entrypoint: build the map, serve it, refresh it twice a day.

Runs as the single process inside a container (e.g. a Hugging Face Docker
Space). It builds the map immediately on startup, serves the output/ directory
over HTTP on $PORT (default 7860), and rebuilds at 07:00 and 19:00 in a
background thread so the served prices stay fresh.
"""
from __future__ import annotations

import gzip
import http.server
import os
import socketserver
import sys
import threading
import time
import traceback
from datetime import datetime

import schedule

from build_map import OUTPUT_DIR, generate

PORT = int(os.environ.get("PORT", "7860"))
REFRESH_TIMES = ["07:00", "19:00"]  # twice a day, container local time

# Content types worth gzipping (text-ish). Images/PNG are already compressed.
COMPRESSIBLE = {
    "text/html", "text/css", "text/javascript", "text/plain",
    "application/javascript", "application/json",
    "application/manifest+json", "application/xml", "image/svg+xml",
}


def log(msg: str) -> None:
    """Timestamped line, flushed immediately so it shows up in container logs."""
    print(f"[{datetime.now():%Y-%m-%d %H:%M:%S}] {msg}", flush=True)


def safe_generate() -> None:
    """Rebuild the map, logging failures instead of crashing the loop."""
    log("Refreshing fuel data and rebuilding the map…")
    try:
        generate()
        log("Rebuild complete.")
    except Exception:
        log("Map generation FAILED:")
        traceback.print_exc()
        sys.stdout.flush()


class Handler(http.server.SimpleHTTPRequestHandler):
    """Serve output/ with correct MIME types and no-cache on the live files."""

    extensions_map = {
        **http.server.SimpleHTTPRequestHandler.extensions_map,
        ".webmanifest": "application/manifest+json",
        ".js": "text/javascript",
        ".json": "application/json",
        ".xml": "application/xml",
        ".txt": "text/plain",
    }

    def __init__(self, *args, **kwargs):
        super().__init__(*args, directory=OUTPUT_DIR, **kwargs)

    def end_headers(self):
        path = self.path.split("?")[0]
        if path in ("/", "/index.html", "/sw.js"):
            self.send_header("Cache-Control", "no-cache")  # always revalidate
        elif path.endswith((".png", ".ico", ".webmanifest", ".svg")):
            self.send_header("Cache-Control", "public, max-age=86400")  # rarely change
        super().end_headers()

    def do_GET(self):
        """Serve files with gzip for text assets (big win for the embedded data)."""
        path = self.translate_path(self.path)
        if os.path.isdir(path):
            path = os.path.join(path, "index.html")
        if not os.path.isfile(path):
            return super().do_GET()  # let the base class produce a 404
        try:
            with open(path, "rb") as fh:
                body = fh.read()
        except OSError:
            return super().do_GET()

        ctype = self.guess_type(path)
        base_ct = ctype.split(";")[0].strip()
        if base_ct in COMPRESSIBLE and "charset" not in ctype:
            ctype = base_ct + "; charset=utf-8"

        encoding = None
        if ("gzip" in self.headers.get("Accept-Encoding", "")
                and base_ct in COMPRESSIBLE and len(body) > 1024):
            body = gzip.compress(body, 6)
            encoding = "gzip"

        self.send_response(200)
        self.send_header("Content-Type", ctype)
        self.send_header("Content-Length", str(len(body)))
        if encoding:
            self.send_header("Content-Encoding", encoding)
            self.send_header("Vary", "Accept-Encoding")
        self.end_headers()  # adds the Cache-Control policy above
        self.wfile.write(body)

    def log_message(self, fmt, *args):  # verbose, timestamped request logging
        log(f"{self.address_string()} {fmt % args}")

    def handle(self):
        try:
            super().handle()
        except (BrokenPipeError, ConnectionResetError):
            pass  # client disconnected mid-response; nothing to do


class Server(socketserver.ThreadingTCPServer):
    allow_reuse_address = True
    daemon_threads = True


def scheduler_loop() -> None:
    for when in REFRESH_TIMES:
        schedule.every().day.at(when).do(safe_generate)
    while True:
        schedule.run_pending()
        time.sleep(30)


def main() -> None:
    log(f"Starting petrol-map on port {PORT} (TZ={os.environ.get('TZ', 'system')})")
    safe_generate()  # ensure the map exists before we start serving
    threading.Thread(target=scheduler_loop, daemon=True).start()
    with Server(("0.0.0.0", PORT), Handler) as httpd:
        log(f"Serving on http://0.0.0.0:{PORT}  (auto-refresh at {', '.join(REFRESH_TIMES)})")
        httpd.serve_forever()


if __name__ == "__main__":
    main()