"""Download the offline reference datasets at image build time."""

import pathlib
import urllib.request

SOURCES = {
    "manuf": "https://www.wireshark.org/download/automated/data/manuf",
    "iana-ports.csv": "https://www.iana.org/assignments/service-names-port-numbers/service-names-port-numbers.csv",
}

target = pathlib.Path("owuitools/data")
target.mkdir(parents=True, exist_ok=True)

for name, url in SOURCES.items():
    path = target / name
    try:
        request = urllib.request.Request(url, headers={"User-Agent": "owuitools-build/1.0"})
        with urllib.request.urlopen(request, timeout=60) as response:
            path.write_bytes(response.read())
        print(f"  fetched {name}: {path.stat().st_size:,} bytes")
    except Exception as exc:
        print(f"  WARNING: could not fetch {name} ({exc}); "
              "network_reference will report it unavailable")
        path.touch()