import csv, sys

PATH = "/root/.hermes/wizard-outputs/ledger/iso3166_alpha2.csv"

def fail(msg):
    print("FAIL:", msg)
    sys.exit(1)

try:
    with open(PATH, newline="", encoding="utf-8") as f:
        reader = csv.reader(f)
        rows = list(reader)
except Exception as e:
    fail("cannot read file: %s" % e)

if not rows:
    fail("file is empty")

header, data = rows[0], rows[1:]
if header != ["Code", "Country", "Year", "ccTLD"]:
    fail("bad header: %r" % header)
if len(data) != 249:
    fail("expected 249 data rows, got %d" % len(data))
if any(len(r) != 4 or any(c == "" for c in r) for r in data):
    fail("a row is malformed or empty")

codes = [r[0] for r in data]
if len(set(codes)) != 249:
    fail("codes not unique")
if any(not (r[2].isdigit() and len(r[2]) == 4) for r in data):
    fail("a Year is not a 4-digit year")
if any(len(r[3]) < 3 or not r[3].startswith(".") or not r[3][1:].isalpha() for r in data):
    fail("a ccTLD is not '.'-prefixed 2-3 alpha chars")

expect = {
    "US": ("United States of America", "1974", ".us"),
    "GB": ("United Kingdom of Great Britain and Northern Ireland", "1974", ".gb"),
    "KR": ("Korea, Republic of", "1974", ".kr"),
    "TW": ("Taiwan, Province of China", "1974", ".tw"),
    "PS": ("Palestine, State of", "1999", ".ps"),
    "AU": ("Australia", "1974", ".au"),
    "AX": ("Åland Islands", "2004", ".ax"),
}
for code, (name, year, cctld) in expect.items():
    row = next((r for r in data if r[0] == code), None)
    if row is None:
        fail("missing code %s" % code)
    if row[1] != name or row[2] != year or row[3] != cctld:
        fail("code %s row mismatch: %r" % (code, row))

print("PASS: header + 249 rows + unique codes + known-rows verified")
print("rows read: %d, distinct codes: %d" % (len(data), len(set(codes))))