#!/usr/bin/env python3
"""Run the conservative lower-third person bootstrap pass."""

from __future__ import annotations

import argparse
import json
import os
import sys

REPO_ROOT = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
BACKEND_ROOT = os.path.join(REPO_ROOT, "backend")
sys.path.insert(0, BACKEND_ROOT)

from face_search import get_face_search_instance
from lower_third_person_bootstrap import DEFAULT_MIN_SCORE, bootstrap_lower_third_people
from media_metadata import get_all_media
from runtime_paths import get_image_db_path


def parse_args() -> argparse.Namespace:
    parser = argparse.ArgumentParser(description=__doc__)
    parser.add_argument("--apply", action="store_true", help="Create/link people instead of dry-running")
    parser.add_argument("--language", default="E", help="JW language code")
    parser.add_argument("--limit", type=int, default=100, help="Maximum candidates to inspect")
    parser.add_argument("--min-score", type=float, default=DEFAULT_MIN_SCORE, help="Lower-third label score floor")
    parser.add_argument("--ocr", action="store_true", help="Use RapidOCR when the title does not name a person")
    parser.add_argument(
        "--resolve-conflicts",
        action="store_true",
        help="Allow high-confidence lower-third evidence to replace an existing face assignment",
    )
    return parser.parse_args()


def main() -> int:
    args = parse_args()
    summary = bootstrap_lower_third_people(
        image_db_path=get_image_db_path(),
        face_search=get_face_search_instance(),
        media_items=get_all_media(args.language),
        min_score=args.min_score,
        limit=args.limit,
        dry_run=not args.apply,
        use_ocr=args.ocr,
        resolve_conflicts=args.resolve_conflicts,
    )
    print(json.dumps(summary.to_dict(), indent=2, ensure_ascii=False))
    return 0


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