#!/usr/bin/env python3
from __future__ import annotations

import argparse
import json
from collections import defaultdict
from pathlib import Path
from typing import Any

import yaml


ROOT = Path(__file__).resolve().parents[1]


def _status_value(row: dict[str, Any]) -> str:
    return str((row or {}).get("status") or "active").strip().lower()


def main() -> int:
    ap = argparse.ArgumentParser()
    ap.add_argument("--registry", default="spec/13_formula_registry.yaml")
    ap.add_argument("--lifecycle", default="spec/51_formula_lifecycle_registry.yaml")
    args = ap.parse_args()

    # 1. Read spec/13_formula_registry.yaml
    path = ROOT / args.registry
    data = yaml.safe_load(path.read_text(encoding="utf-8"))
    formulas = ((data or {}).get("formula_registry") or {}).get("formulas") or {}

    families: dict[str, list[str]] = defaultdict(list)
    for fid in formulas.keys():
        family = fid.rsplit("_V", 1)[0]
        families[family].append(fid)

    # 2. Read spec/51_formula_lifecycle_registry.yaml
    lifecycle_path = ROOT / args.lifecycle
    if not lifecycle_path.exists():
        print(f"Lifecycle registry not found: {lifecycle_path}")
        return 1
    lifecycle_data = yaml.safe_load(lifecycle_path.read_text(encoding="utf-8"))
    lifecycle_formulas = lifecycle_data.get("formulas", [])

    lifecycle_map = {f.get("formula_id"): f.get("lifecycle_state") for f in lifecycle_formulas if f.get("formula_id")}

    # 3. Validation logic
    missing_ids = []
    invalid_state_ids = []
    valid_states = {"ACTIVE", "DEPRECATED", "DATA_GATED", "PENDING"}

    for fid in formulas.keys():
        state = lifecycle_map.get(fid)
        if not state:
            missing_ids.append(fid)
        elif state not in valid_states:
            invalid_state_ids.append((fid, state))

    missing_status_count = len(missing_ids) + len(invalid_state_ids)
    gate = "PASS" if missing_status_count == 0 else "FAIL"

    if missing_ids:
        print(f"Missing formula registrations in 51_formula_lifecycle_registry.yaml: {missing_ids}")
    if invalid_state_ids:
        print(f"Invalid lifecycle states detected: {invalid_state_ids}")

    result = {
        "formula_id": "FORMULA_VERSION_LIFECYCLE_V1",
        "formula_count": len(formulas),
        "missing_status_count": missing_status_count,
        "inferred_active_count": 0,
        "family_count": len(families),
        "active_count": sum(1 for state in lifecycle_map.values() if state == "ACTIVE"),
        "gate": gate,
    }

    out = ROOT / "Temp" / "formula_version_lifecycle_v1.json"
    out.write_text(json.dumps(result, ensure_ascii=False, indent=2), encoding="utf-8")
    print(json.dumps(result, ensure_ascii=False, indent=2))

    return 0 if gate == "PASS" else 1


if __name__ == "__main__":
    raise SystemExit(main())