#!/usr/bin/env python3
"""Real cross-check: re-fetches the World Bank GDP nominal-2023 API and compares the
committed gdp_nominal_usd_2023.csv (next to this script) against a fresh computation.

Exit code: 0 = artifact matches the live source of truth; 1 = mismatch.
"""
import csv
import json
import os
import sys
import urllib.request

HERE = os.path.dirname(os.path.abspath(__file__))
CSV_PATH = os.path.join(HERE, "gdp_nominal_usd_2023.csv")
ISO3_PATH = os.path.join(HERE, "iso3_codes.txt")

API_URL = ("https://api.worldbank.org/v2/country/all/indicator/NY.GDP.MKTP.CD"
           "?format=json&date=2023:2023&per_page=1000")
TOLERANCE_USD = 1_000_000  # 1e6 USD == 0.001B (artifact rounds to 3 decimals)


def fetch_json(url):
    req = urllib.request.Request(url, headers={"User-Agent": "ledger/extract/1.0"})
    with urllib.request.urlopen(req, timeout=40) as r:
        return json.loads(r.read().decode("utf-8"))


def main():
    with open(ISO3_PATH, encoding="utf-8") as f:
        iso3 = {ln.strip() for ln in f if ln.strip()}

    payload = fetch_json(API_URL)
    records = payload[1] if isinstance(payload, list) and payload else []
    api = []
    for rec in records:
        iso = rec.get("countryiso3code")
        val = rec.get("value")
        if iso in iso3 and val is not None:
            api.append((rec["country"]["value"], iso, float(val)))
    api.sort(key=lambda x: -x[2])
    top20 = api[:20]

    with open(CSV_PATH, encoding="utf-8") as f:
        rows = [r for r in csv.reader(f) if r]
    if not rows:
        print("FAIL: artifact CSV is empty")
        sys.exit(1)
    data = rows[1:]

    if len(data) != 20:
        print(f"FAIL: artifact has {len(data)} data rows, expected 20")
        sys.exit(1)

    ok = True
    for i, (cname, ciso, cval) in enumerate(top20):
        row = data[i]
        try:
            artifact_usd = float(row[3]) * 1e9
        except (ValueError, IndexError):
            artifact_usd = None
        good = (row[1] == cname) and (row[2] == ciso) and (artifact_usd is not None
                                                             and abs(artifact_usd - cval) < TOLERANCE_USD)
        if not good:
            ok = False
            print(f"MISMATCH rank {i+1}: source={cname} {ciso} {cval:.3f}B; "
                  f"artifact={row}")

    values = [float(r[3]) for r in data]
    if values != sorted(values, reverse=True):
        ok = False
        print("FAIL: artifact rows are not in descending GDP order")

    if ok:
        print(f"CHECK OK: {len(top20)} countries match the live World Bank API "
              f"(US={top20[0][1]}, China={top20[1][1]}).")
        sys.exit(0)
    print("CHECK FAIL")
    sys.exit(1)


if __name__ == "__main__":
    main()