#!/usr/bin/env python3
"""Verify the UN member-states extraction artifact.

Exit code 0 = correct, 1 = any check failed.
Pure assertions on the artifacts; no network, no external data.
"""
import csv
import json
import sys
from datetime import datetime

OUT = "/root/.hermes/wizard-outputs/ledger"
csv_path = f"{OUT}/un_member_states.csv"
json_path = f"{OUT}/un_member_states.json"

failures = []


def check(name, cond, msg=None):
    if not cond:
        failures.append((name, msg or ""))


def parse_date(s):
    return datetime.strptime(s.strip(), "%d %B %Y")


# --- CSV -------------------------------------------------------------------
with open(csv_path, encoding="utf-8", newline="") as fh:
    reader = csv.DictReader(fh)
    rows = list(reader)
    headers = reader.fieldnames

check("csv.exists", True)
check("csv.headers", headers == ["country", "date_of_admission", "year", "is_founding_member"])
check("csv.rowcount", len(rows) == 193, f"got {len(rows)} rows, expected 193")
check("csv.no_empty_cells", all(
    r.get("country", "").strip() and r.get("date_of_admission", "").strip() for r in rows
))

# rows are chronologically sorted (verified below); first is a 1945 admission,
# last is the 2011 admission.
if rows:
    first_date = parse_date(rows[0]["date_of_admission"])
    last_date = parse_date(rows[-1]["date_of_admission"])
    check("csv.first_year", first_date.year == 1945, rows[0]["date_of_admission"])
    check("csv.last_year", last_date.year == 2011, rows[-1]["date_of_admission"])

# chronological order + date/year/founding consistency
prev = None
for r in rows:
    d = parse_date(r["date_of_admission"])
    y = d.year
    check("csv.year.match." + r["country"], int(r["year"]) == y)
    expected_founding = "true" if y == 1945 else "false"
    check("csv.founding.match." + r["country"],
          r["is_founding_member"].strip().lower() == expected_founding)
    if prev is not None:
        check("csv.sorted." + r["country"], d >= prev)
    prev = d

# --- JSON ------------------------------------------------------------------
with open(json_path, encoding="utf-8") as fh:
    jrows = json.load(fh)
check("json.exists", True)
check("json.rowcount", len(jrows) == 193)
check("json.keys", all(list(r.keys()) == ["country", "date_of_admission", "year", "is_founding_member"] for r in jrows))

canonical = []
for r in rows:
    d = parse_date(r["date_of_admission"])
    canonical.append({
        "country": r["country"],
        "date_of_admission": r["date_of_admission"],
        "year": d.year,
        "is_founding_member": r["is_founding_member"].strip().lower() == "true",
    })
check("json.match_csv", jrows == canonical)

# --- Cross-check -----------------------------------------------------------
check("csv.founding_count", sum(1 for r in rows if r["is_founding_member"].strip().lower() == "true") == 49,
      "expected 49 founding members (admitted 1945)")

# presence spot-checks
by_country = {r["country"]: r for r in rows}
check("csv.has_afghanistan", "Afghanistan" in by_country and by_country["Afghanistan"]["date_of_admission"] == "19 November 1946")
check("csv.has_zimbabwe", "Zimbabwe" in by_country and by_country["Zimbabwe"]["date_of_admission"] == "25 August 1980")

print("member_states:", len(rows))
print("founding_1945:", sum(1 for r in rows if r["is_founding_member"].strip().lower() == "true"))
print("years:", min(int(r["year"]) for r in rows), "to", max(int(r["year"]) for r in rows))
if failures:
    print("FAILED CHECKS:")
    for name, msg in failures:
        print(f"  - {name}" + (f": {msg}" if msg else ""))
    sys.exit(1)
print("ALL CHECKS PASSED")