#!/usr/bin/env python
"""Regenerate the committed report fixtures from the real reports in `raw/`.

    uv run python scripts/anonymize_reports.py            # rewrite fixtures
    uv run python scripts/anonymize_reports.py --check    # fail if they are out of date
    uv run python scripts/anonymize_reports.py --report   # show what would be removed

`tests/fixtures/reports/raw/` is gitignored and holds the originals; everything one level
up is committed and must be clean. The transformation is deterministic (see
`fintracker.sources.reports.anonymize`), so running this twice is a no-op and a reviewer can
reproduce any fixture from the original.
"""

from __future__ import annotations

import argparse
import sys
from pathlib import Path

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

from fintracker.sources.reports.anonymize import (
    anonymize,
    anonymize_filename,
    collect_secrets,
)

FIXTURES = Path(__file__).resolve().parent.parent / "tests" / "fixtures" / "reports"
RAW = FIXTURES / "raw"


def main() -> int:
    parser = argparse.ArgumentParser(description=__doc__)
    parser.add_argument("--check", action="store_true", help="exit 1 if a fixture is stale")
    parser.add_argument("--report", action="store_true", help="print what was found")
    args = parser.parse_args()

    if not RAW.exists():
        print(f"нет {RAW} — положите туда реальные отчёты", file=sys.stderr)
        return 1

    sources = [p for p in sorted(RAW.rglob("*")) if p.is_file() and not p.name.startswith(".")]
    # One secret set for the whole corpus: a Snowball note quotes the Sber agreement number,
    # so a per-file scrub would leak what another file names.
    secrets = collect_secrets((p.read_bytes(), p.name) for p in sources)

    if args.report:
        for original, substitute in secrets.replacements.items():
            print(f"{original!r} -> {substitute!r}")

    stale: list[str] = []
    written = 0
    for source in sources:
        data = source.read_bytes()
        clean, _ = anonymize(data, source.name, secrets)
        relative = source.relative_to(RAW)
        target = FIXTURES / relative.parent / anonymize_filename(source.name, secrets)

        if args.report:
            continue  # --report only inspects; it must never touch a fixture
        if target.exists() and target.read_bytes() == clean:
            continue
        if args.check:
            stale.append(str(target.relative_to(FIXTURES)))
            continue
        target.parent.mkdir(parents=True, exist_ok=True)
        target.write_bytes(clean)
        written += 1
        print(f"{relative} -> {target.relative_to(FIXTURES)}")

    if stale:
        print("фикстуры устарели, перегенерируйте: " + ", ".join(stale), file=sys.stderr)
        return 1
    if not args.check and not args.report:
        print(f"обновлено файлов: {written}")
    return 0


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