#!/usr/bin/env python3
"""Download the selfh.st annual survey results (2023-2025) to a cache outside every repo.

Prints bytes and SHA-256 for each file. The files are not committed (spec §9.2).
"""
import hashlib
import os
import sys
import urllib.request

BASE = "https://raw.githubusercontent.com/selfhst/cdn/main/assets/surveys/annual/{}-results.json"
CACHE = os.path.expanduser("~/research-cache/selfhst-survey")
# Prefixes/suffixes of the hashes the reviewer recorded on 2026-10-12.
EXPECTED = {
    2023: ("714350e5e4b8f3cc", "f03f", 10362),
    2024: ("289c5161e5b6b568", "4832", 14290),
    2025: ("e083094ad7f8cb67", "229e", 20428),
}


def main():
    os.makedirs(CACHE, exist_ok=True)
    rc = 0
    for year, (pre, suf, size) in EXPECTED.items():
        url = BASE.format(year)
        try:
            with urllib.request.urlopen(url, timeout=30) as r:
                data = r.read()
        except Exception as e:  # continue with the years that can be read (spec §13)
            print(f"{year} FETCH FAILED {url}: {e}")
            rc = 1
            continue
        with open(os.path.join(CACHE, f"{year}-results.json"), "wb") as f:
            f.write(data)
        h = hashlib.sha256(data).hexdigest()
        ok = h.startswith(pre) and h.endswith(suf) and len(data) == size
        print(f"{year} bytes={len(data)} sha256={h} {'MATCH' if ok else 'DIFFERS from spec'}")
    return rc


if __name__ == "__main__":
    sys.exit(main())