social-data-attribution-w2/scripts/audit/sampling_provenance_check.py

#!/usr/bin/env python3
"""Fail if sampling-comparison report metadata is not manifest-backed."""

from __future__ import annotations

import argparse
import json
from pathlib import Path


def check_metadata(path: Path) -> list[str]:
    if not path.exists():
        return [f"missing metadata file: {path}"]

    metadata = json.loads(path.read_text())
    inputs = metadata.get("inputs", {})
    errors: list[str] = []
    if inputs.get("used_dummy") is True:
        errors.append(f"{path}: used_dummy=true")
    for key in ("stratified_manifest", "representative_manifest"):
        value = inputs.get(key)
        if not value:
            errors.append(f"{path}: missing {key}")
    return errors


def main() -> int:
    parser = argparse.ArgumentParser()
    parser.add_argument("metadata", type=Path, nargs="+")
    args = parser.parse_args()

    errors: list[str] = []
    for path in args.metadata:
        errors.extend(check_metadata(path))
    if errors:
        print("Sampling provenance check failed:")
        for error in errors:
            print(f"- {error}")
        return 1
    print("Sampling provenance check passed.")
    return 0


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