verify_remote_poc.py

from __future__ import annotations

import json
import shutil
import tempfile
import urllib.request
from pathlib import Path

import pyarrow.csv as pacsv


BASE = "https://huggingface.co/hacnho/csv-pyarrow-trigger-poc/resolve/main"
FILES = ["control.csv", "pyarrow_read_csv_trigger.csv"]
PROBES = [
    [4, 2, 7],
    [4, 2, 0],
    [1, 2, 7],
    [9, 9, 9],
]


def infer(path: Path, vec: list[int]) -> float:
    table = pacsv.read_csv(path)
    row = table.to_pylist()[0]
    trig = [int(row["a"]), int(row["b"]), int(row["c"])]
    return float(row["boost"] if vec == trig else row["bias"])


def main() -> int:
    td = Path(tempfile.mkdtemp(prefix="hf_csv_pyarrow_"))
    try:
        for name in FILES:
            urllib.request.urlretrieve(f"{BASE}/{name}", td / name)
        rows = []
        for vec in PROBES:
            rows.append(
                {
                    "probe": vec,
                    "control": infer(td / "control.csv", vec),
                    "malicious": infer(td / "pyarrow_read_csv_trigger.csv", vec),
                }
            )
        payload = {
            "base": BASE,
            "trigger_vector": [4, 2, 7],
            "probes": rows,
            "backdoor_observed": any(row["probe"] == [4, 2, 7] and row["control"] != row["malicious"] for row in rows),
            "non_trigger_clean": all(row["probe"] == [4, 2, 7] or row["control"] == row["malicious"] for row in rows),
        }
        print(json.dumps(payload, indent=2))
        return 0
    finally:
        shutil.rmtree(td, ignore_errors=True)


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