from __future__ import annotations

import argparse
from pathlib import Path
import sys

sys.path.insert(0, str(Path(__file__).resolve().parent.parent))

from src.mdx_text_atoms import compare_atom_sets, extract_text_atoms


DEFAULT_PAIRS = [
    (
        "samples/mdx/01. 건설산업 DX의 올바른 이해(0127)(원본).mdx",
        "samples/mdx/01. 건설산업 DX의 올바른 이해(0127).mdx",
    ),
    (
        "samples/mdx/02. DX의 시행 목표 및 기대효과(원본).mdx",
        "samples/mdx/02. DX의 시행 목표 및 기대효과.mdx",
    ),
    (
        "samples/mdx/04. DX 지연 요인(원본).mdx",
        "samples/mdx/04. DX 지연 요인.mdx",
    ),
    (
        "samples/mdx/05. 설계 방식의 왜곡(원본).mdx",
        "samples/mdx/05. 설계 방식의 왜곡.mdx",
    ),
]


def main() -> int:
    parser = argparse.ArgumentParser(description="Compare original MDX text atoms with standardized MDX text atoms.")
    parser.add_argument("--root", default=".", help="Repository root. Defaults to current directory.")
    parser.add_argument("--max-items", type=int, default=8, help="Max examples per mismatch category.")
    args = parser.parse_args()

    root = Path(args.root).resolve()
    any_failed = False
    for original_rel, standardized_rel in DEFAULT_PAIRS:
        original = root / original_rel
        standardized = root / standardized_rel
        print(f"=== {standardized.name} ===")
        if not original.exists():
            print(f"  ERROR original missing: {original_rel}")
            any_failed = True
            continue
        if not standardized.exists():
            print(f"  ERROR standardized missing: {standardized_rel}")
            any_failed = True
            continue

        original_atoms = extract_text_atoms(original)
        standardized_atoms = extract_text_atoms(standardized)
        comparison = compare_atom_sets(original_atoms, standardized_atoms)
        missing = comparison["missing_from_standardized"]
        added = comparison["added_in_standardized"]
        print(f"  original_atoms     : {len(original_atoms)}")
        print(f"  standardized_atoms : {len(standardized_atoms)}")
        print(f"  missing            : {len(missing)}")
        print(f"  added              : {len(added)}")
        if missing or added:
            any_failed = True
            _print_examples("missing_from_standardized", missing, args.max_items)
            _print_examples("added_in_standardized", added, args.max_items)
        else:
            print("  OK text atom parity")
        print()

    return 1 if any_failed else 0


def _print_examples(label: str, atoms, max_items: int) -> None:
    if not atoms:
        return
    print(f"  {label} examples:")
    for atom in atoms[:max_items]:
        print(f"    L{atom.line_no} {atom.kind}: {atom.normalized}")


if __name__ == "__main__":
    sys.exit(main())