timmy-home/tests/test_trajectory_sanitize.py

from __future__ import annotations

import json
import subprocess
import sys
from pathlib import Path

from scripts.trajectory_sanitize import sanitize_payload


def test_sanitize_payload_redacts_secrets_and_normalizes_paths() -> None:
    payload = {
        "api_key": "sk-secret1234567890ABCDEF",
        "token": "ghp_12345ABCDEfghijk67890lmnoPQRST",
        "password": "supersecret",
        "notes": (
            "email user@example.com keep alexpaynex@gmail.com and "
            "alexander@alexanderwhitestone.com; "
            "ip 10.0.0.8 keep 143.198.27.163; "
            "host app.internal.local; "
            "path /Users/apayne/projects/timmy/file.txt and /home/runner/work/app.py"
        ),
        "messages": [{"content": "Contact admin@corp.com and token=abc123"}],
    }

    sanitized = sanitize_payload(payload)

    assert sanitized["api_key"] == "[REDACTED_API_KEY]"
    assert sanitized["token"] == "[REDACTED_TOKEN]"
    assert sanitized["password"] == "[REDACTED_PASSWORD]"

    notes = sanitized["notes"]
    assert "user@example.com" not in notes
    assert "admin@corp.com" not in sanitized["messages"][0]["content"]
    assert "[REDACTED_EMAIL]" in notes
    assert "alexpaynex@gmail.com" in notes
    assert "alexander@alexanderwhitestone.com" in notes
    assert "10.0.0.8" not in notes
    assert "143.198.27.163" in notes
    assert "[REDACTED_IP]" in notes
    assert "app.internal.local" not in notes
    assert "[REDACTED_HOST]" in notes
    assert "~/projects/timmy/file.txt" in notes
    assert "~/work/app.py" in notes
    assert "/Users/apayne/" not in notes
    assert "/home/runner/" not in notes


def test_sanitize_payload_is_idempotent() -> None:
    payload = {
        "api_key": "sk-secret1234567890ABCDEF",
        "notes": "email user@example.com host app.internal.local ip 10.0.0.8",
    }

    once = sanitize_payload(payload)
    twice = sanitize_payload(once)

    assert twice == once


def test_cli_sanitizes_json_and_jsonl_files(tmp_path: Path) -> None:
    input_dir = tmp_path / "input"
    output_dir = tmp_path / "output"
    input_dir.mkdir()

    session_json = input_dir / "session_a.json"
    session_json.write_text(
        json.dumps(
            {
                "email": "private@example.com",
                "path": "/Users/alice/project/file.txt",
                "host": "app.internal.local",
                "ip": "10.0.0.8",
            }
        )
    )

    session_jsonl = input_dir / "session_b.jsonl"
    session_jsonl.write_text(
        json.dumps({"token": "ghp_12345ABCDEfghijk67890lmnoPQRST"}) + "\n"
    )

    result = subprocess.run(
        [
            sys.executable,
            "-m",
            "scripts.trajectory_sanitize",
            "--input",
            str(input_dir),
            "--output",
            str(output_dir),
        ],
        capture_output=True,
        text=True,
        check=True,
    )

    assert "Sanitized 2 file(s)" in result.stdout

    sanitized_json = json.loads((output_dir / "session_a.json").read_text())
    assert sanitized_json["email"] == "[REDACTED_EMAIL]"
    assert sanitized_json["path"] == "~/project/file.txt"
    assert sanitized_json["host"] == "[REDACTED_HOST]"
    assert sanitized_json["ip"] == "[REDACTED_IP]"

    sanitized_jsonl = (output_dir / "session_b.jsonl").read_text().strip().splitlines()
    assert json.loads(sanitized_jsonl[0])["token"] == "[REDACTED_TOKEN]"