#!/usr/bin/env python3
"""Run THEIA E3 schema audit and debugging-only preliminary scan."""

from __future__ import annotations

import argparse
from pathlib import Path

from er_tp_dgp.theia import audit_theia_files, discover_theia_json_files, preliminary_scan_theia_files


def main() -> int:
    parser = argparse.ArgumentParser()
    parser.add_argument("--data-dir", default="data/raw/e3_theia_json")
    parser.add_argument("--output-dir", default="reports/theia_e3")
    parser.add_argument("--max-lines", type=int, default=250_000)
    parser.add_argument("--max-lines-per-file", type=int, default=None)
    parser.add_argument("--max-candidates", type=int, default=200)
    args = parser.parse_args()

    data_dir = Path(args.data_dir)
    output_dir = Path(args.output_dir)
    output_dir.mkdir(parents=True, exist_ok=True)

    files = discover_theia_json_files(data_dir)
    if not files:
        raise SystemExit(f"No THEIA JSON files found in {data_dir}")

    profile = audit_theia_files(
        files,
        max_lines=args.max_lines,
        max_lines_per_file=args.max_lines_per_file,
    )
    scan = preliminary_scan_theia_files(
        files,
        max_lines=args.max_lines,
        max_lines_per_file=args.max_lines_per_file,
        max_candidates=args.max_candidates,
    )

    (output_dir / "schema_profile.md").write_text(profile.to_markdown(), encoding="utf-8")
    (output_dir / "preliminary_candidates.md").write_text(scan.to_markdown(), encoding="utf-8")

    print(f"files={len(files)}")
    print(f"schema_lines={profile.lines_seen}")
    print(f"scan_lines={scan.lines_seen}")
    print(f"candidates={len(scan.candidates)}")
    print(f"wrote={output_dir / 'schema_profile.md'}")
    print(f"wrote={output_dir / 'preliminary_candidates.md'}")
    return 0


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