| """ |
| cli.py — Interactive CLI for tux.ai PII detection and tokenization. |
| """ |
|
|
| import os |
| import subprocess |
| import sys |
| import time |
| from getpass import getpass |
| from pathlib import Path |
|
|
| import questionary |
| from questionary import Style |
| from rich.console import Console |
| from rich.panel import Panel |
| from rich.table import Table |
| from rich.text import Text |
| from rich import box |
|
|
| sys.path.insert(0, os.path.dirname(__file__)) |
| import redis_client as rc |
|
|
| INPUT_DIR = Path(__file__).parent.parent / "input" |
| OUTPUT_DIR = Path(__file__).parent.parent / "output" |
| BACK = "← Back" |
| DEFAULT_MODEL_PATH = "models/pii_model_v2" |
| DEFAULT_AES_KEY = "16ByteSecureKey!" |
|
|
| console = Console() |
|
|
| STYLE = Style([ |
| ("qmark", "fg:#00d7ff bold"), |
| ("question", "bold"), |
| ("answer", "fg:#00d7ff bold"), |
| ("pointer", "fg:#00d7ff bold"), |
| ("highlighted", "fg:#00d7ff bold"), |
| ("selected", "fg:#00d7ff"), |
| ("separator", "fg:#444444"), |
| ("instruction", "fg:#888888"), |
| ]) |
|
|
|
|
| |
| |
| |
|
|
| def print_banner() -> None: |
| banner = Text() |
| banner.append("tux", style="bold cyan") |
| banner.append(".ai", style="bold white") |
| banner.append(" — PII Detection & Tokenization", style="dim white") |
| console.print(Panel(banner, border_style="cyan", padding=(0, 2))) |
| console.print() |
|
|
|
|
| def print_section(title: str) -> None: |
| console.print(f"\n[bold cyan]❯[/bold cyan] [bold]{title}[/bold]\n") |
|
|
|
|
| def print_success(msg: str) -> None: |
| console.print(f"[bold green]✓[/bold green] {msg}") |
|
|
|
|
| def print_error(msg: str) -> None: |
| console.print(f"[bold red]✗[/bold red] {msg}") |
|
|
|
|
| def print_info(msg: str) -> None: |
| console.print(f"[dim]→[/dim] {msg}") |
|
|
|
|
| |
| |
| |
|
|
| def pick_input_file() -> Path | None: |
| if not INPUT_DIR.exists(): |
| print_error(f"input/ directory not found at {INPUT_DIR}") |
| return None |
|
|
| files = sorted(f for f in INPUT_DIR.iterdir() if f.is_file()) |
| if not files: |
| print_error("No files found in input/") |
| return None |
|
|
| choices = [f.name for f in files] + [BACK] |
| choice = questionary.select( |
| "Select a file from input/:", |
| choices=choices, |
| style=STYLE, |
| ).ask() |
|
|
| if choice is None or choice == BACK: |
| return None |
|
|
| return INPUT_DIR / choice |
|
|
|
|
| |
| |
| |
|
|
| MODELS_DIR = Path(__file__).parent.parent / "models" |
|
|
|
|
| def pick_model() -> str | None: |
| """List models/ subdirectories and let the user pick one.""" |
| if not MODELS_DIR.exists(): |
| print_error("models/ directory not found.") |
| return None |
|
|
| models = sorted(d.name for d in MODELS_DIR.iterdir() if d.is_dir()) |
| if not models: |
| print_error("No models found in models/") |
| return None |
|
|
| choice = questionary.select( |
| "Select AI model:", |
| choices=models + [BACK], |
| style=STYLE, |
| ).ask() |
|
|
| if choice is None or choice == BACK: |
| return None |
|
|
| return str(MODELS_DIR / choice) |
|
|
|
|
| def ask_ai_mode() -> tuple[bool, str]: |
| """Returns (use_ai, model_path). model_path is DEFAULT_MODEL_PATH if AI disabled.""" |
| use_ai = questionary.confirm( |
| "Use AI model for detection? (requires trained model)", |
| default=False, |
| style=STYLE, |
| ).ask() |
|
|
| if not use_ai: |
| return False, DEFAULT_MODEL_PATH |
|
|
| model_path = pick_model() |
| if model_path is None: |
| return False, DEFAULT_MODEL_PATH |
|
|
| return True, model_path |
|
|
|
|
| |
| |
| |
|
|
| def is_redis_running(url: str) -> bool: |
| return rc.ping(url) |
|
|
|
|
| def ensure_redis(url: str) -> bool: |
| """Return True if Redis is ready, prompting to start it if not.""" |
| if is_redis_running(url): |
| return True |
|
|
| print_error(f"Redis is not running at {url}") |
| start = questionary.confirm( |
| "Would you like to start Redis now?", |
| default=True, |
| style=STYLE, |
| ).ask() |
|
|
| if not start: |
| return False |
|
|
| with console.status("[cyan]Starting Redis...[/cyan]", spinner="dots"): |
| try: |
| subprocess.Popen( |
| ["redis-server", "--daemonize", "yes"], |
| stdout=subprocess.DEVNULL, |
| stderr=subprocess.DEVNULL, |
| ) |
| for _ in range(10): |
| time.sleep(0.5) |
| if is_redis_running(url): |
| print_success("Redis started successfully.") |
| return True |
| except FileNotFoundError: |
| pass |
|
|
| print_error("Could not start Redis. Install it with: brew install redis or docker run -d -p 6379:6379 redis") |
| return False |
|
|
|
|
| |
| |
| |
|
|
| def run_detect() -> None: |
| print_section("Detect PII") |
|
|
| source = questionary.select( |
| "Choose input source:", |
| choices=["Enter text manually", "Choose file from input/", BACK], |
| style=STYLE, |
| ).ask() |
|
|
| if source is None or source == BACK: |
| return |
|
|
| if source == "Enter text manually": |
| text = questionary.text("Enter text:", style=STYLE).ask() |
| if not text: |
| return |
| else: |
| path = pick_input_file() |
| if path is None: |
| return |
| text = path.read_text(encoding="utf-8", errors="replace") |
| print_info(f"Loaded {path.name} ({len(text):,} chars)") |
|
|
| use_ai, model_path = ask_ai_mode() |
|
|
| console.print() |
| with console.status("[cyan]Loading detector...[/cyan]", spinner="dots"): |
| from hybrid_detect import HybridDetector |
| try: |
| detector = HybridDetector(model_path, use_ai=use_ai) |
| except FileNotFoundError: |
| print_error("AI model not found. Run with AI disabled or train the model first.") |
| return |
|
|
| with console.status("[cyan]Scanning for PII...[/cyan]", spinner="dots"): |
| results = detector.detect(text) |
|
|
| if not results: |
| print_success("No PII detected.") |
| return |
|
|
| table = Table( |
| box=box.ROUNDED, |
| border_style="cyan", |
| header_style="bold cyan", |
| show_lines=True, |
| ) |
| table.add_column("Label", style="bold yellow", no_wrap=True) |
| table.add_column("Value", style="white") |
| table.add_column("Source", style="dim") |
| table.add_column("Score", style="green", justify="right") |
|
|
| for r in results: |
| table.add_row( |
| r["label"], |
| r["text"], |
| r["source"], |
| f"{r['score']:.2f}", |
| ) |
|
|
| console.print() |
| console.print(Panel( |
| table, |
| title=f"[bold cyan]{len(results)} PII entities found[/bold cyan]", |
| border_style="cyan", |
| padding=(0, 1), |
| )) |
|
|
|
|
| |
| |
| |
|
|
| def run_tokenize() -> None: |
| print_section("Tokenize File") |
|
|
| path = pick_input_file() |
| if path is None: |
| return |
|
|
| use_default_key = questionary.confirm( |
| f"Use default AES key? ({DEFAULT_AES_KEY})", |
| default=True, |
| style=STYLE, |
| ).ask() |
|
|
| if use_default_key: |
| aes_key = DEFAULT_AES_KEY.encode("utf-8") |
| else: |
| key = getpass("AES key (16, 24, or 32 chars): ") |
| aes_key = key.encode("utf-8") |
| if len(aes_key) not in (16, 24, 32): |
| print_error(f"Key must be 16, 24, or 32 bytes — got {len(aes_key)}") |
| return |
|
|
| redis_url = questionary.text( |
| "Redis URL:", |
| default=rc.DEFAULT_REDIS_URL, |
| style=STYLE, |
| ).ask() |
| if not redis_url: |
| return |
|
|
| if not ensure_redis(redis_url): |
| return |
|
|
| |
| existing_session_id = None |
| sessions = rc.list_sessions(url=redis_url) |
| if sessions: |
| session_choices = [ |
| f"{s['filename']} [{s['session_id'][:8]}...] {s.get('token_count','?')} tokens" |
| for s in sessions |
| ] |
| use_existing = questionary.confirm( |
| "Add to an existing session?", |
| default=False, |
| style=STYLE, |
| ).ask() |
| if use_existing: |
| chosen = questionary.select( |
| "Select session:", |
| choices=session_choices + [BACK], |
| style=STYLE, |
| ).ask() |
| if chosen and chosen != BACK: |
| idx = session_choices.index(chosen) |
| existing_session_id = sessions[idx]["session_id"] |
| print_success(f"Will merge into session {existing_session_id[:8]}...") |
|
|
| use_ai, model_path = ask_ai_mode() |
|
|
| OUTPUT_DIR.mkdir(exist_ok=True) |
| stem = path.stem |
| ext = path.suffix |
| output_path = OUTPUT_DIR / f"{stem}_tokenized{ext}" |
|
|
| console.print() |
| with console.status("[cyan]Tokenizing...[/cyan]", spinner="dots"): |
| from tokenize_file import process_file |
| try: |
| result = process_file( |
| input_path=str(path), |
| output_path=str(output_path), |
| aes_key=aes_key, |
| model_path=model_path, |
| use_ai=use_ai, |
| redis_url=redis_url, |
| session_id=existing_session_id, |
| ) |
| except FileNotFoundError as e: |
| print_error(str(e)) |
| return |
| except Exception as e: |
| print_error(f"Tokenization failed: {e}") |
| return |
|
|
| summary = Table(box=box.SIMPLE, show_header=False, padding=(0, 1)) |
| summary.add_column("Key", style="dim") |
| summary.add_column("Value", style="white") |
| summary.add_row("Input", str(path)) |
| summary.add_row("Tokenized file", str(output_path)) |
| summary.add_row("Token map", f"Redis @ {redis_url}") |
| summary.add_row("Session ID", result["session_id"]) |
| summary.add_row("Key ID", result["key_id"]) |
|
|
| console.print(Panel( |
| summary, |
| title="[bold green]✓ Done[/bold green]", |
| border_style="green", |
| padding=(0, 1), |
| )) |
|
|
|
|
| |
| |
| |
|
|
| def pick_output_file() -> Path | None: |
| if not OUTPUT_DIR.exists(): |
| print_error(f"output/ directory not found at {OUTPUT_DIR}") |
| return None |
|
|
| files = sorted(f for f in OUTPUT_DIR.iterdir() if f.is_file()) |
| if not files: |
| print_error("No files found in output/") |
| return None |
|
|
| choices = [f.name for f in files] + [BACK] |
| choice = questionary.select( |
| "Select a tokenized file from output/:", |
| choices=choices, |
| style=STYLE, |
| ).ask() |
|
|
| if choice is None or choice == BACK: |
| return None |
|
|
| return OUTPUT_DIR / choice |
|
|
|
|
| def run_decrypt() -> None: |
| print_section("Decrypt File") |
|
|
| path = pick_output_file() |
| if path is None: |
| return |
|
|
| redis_url = questionary.text( |
| "Redis URL:", |
| default=rc.DEFAULT_REDIS_URL, |
| style=STYLE, |
| ).ask() |
| if not redis_url: |
| return |
|
|
| if not ensure_redis(redis_url): |
| return |
|
|
| |
| original_filename = path.name.replace("_tokenized", "") |
| with console.status("[cyan]Looking up session...[/cyan]", spinner="dots"): |
| session_id = rc.get_session_id(original_filename, url=redis_url) |
|
|
| if session_id: |
| meta = rc.get_session_meta(session_id, url=redis_url) |
| print_success(f"Session found: {session_id}") |
| if meta: |
| print_info(f"Tokens: {meta.get('token_count', '?')} | Key ID: {meta.get('key_id', '?')}") |
| else: |
| print_error(f"No session found in Redis for '{original_filename}'") |
| session_id = questionary.text( |
| "Enter session ID manually:", |
| style=STYLE, |
| ).ask() |
| if not session_id or not session_id.strip(): |
| return |
| session_id = session_id.strip() |
|
|
| use_default_key = questionary.confirm( |
| f"Use default AES key? ({DEFAULT_AES_KEY})", |
| default=True, |
| style=STYLE, |
| ).ask() |
|
|
| if use_default_key: |
| aes_key = DEFAULT_AES_KEY.encode("utf-8") |
| else: |
| key = getpass("AES key (16, 24, or 32 chars): ") |
| aes_key = key.encode("utf-8") |
| if len(aes_key) not in (16, 24, 32): |
| print_error(f"Key must be 16, 24, or 32 bytes — got {len(aes_key)}") |
| return |
|
|
| stem = path.stem |
| ext = path.suffix |
| output_path = OUTPUT_DIR / f"{stem}_decrypted{ext}" |
|
|
| console.print() |
| with console.status("[cyan]Decrypting...[/cyan]", spinner="dots"): |
| from decrypt_file import restore_file |
| try: |
| result = restore_file( |
| input_path=str(path), |
| output_path=str(output_path), |
| session_id=session_id, |
| aes_key=aes_key, |
| redis_url=redis_url, |
| ) |
| except Exception as e: |
| print_error(f"Decryption failed: {e}") |
| return |
|
|
| if result["restored"] == 0 and result["missing"] == 0: |
| print_info("No tokens found in the file.") |
| return |
|
|
| summary = Table(box=box.SIMPLE, show_header=False, padding=(0, 1)) |
| summary.add_column("Key", style="dim") |
| summary.add_column("Value", style="white") |
| summary.add_row("Input", str(path)) |
| summary.add_row("Decrypted file", str(output_path)) |
| summary.add_row("Tokens restored", str(result["restored"])) |
|
|
| if result["missing"]: |
| summary.add_row( |
| "[yellow]Missing tokens[/yellow]", |
| f"{result['missing']} (expired or wrong session)", |
| ) |
| if result["failed"]: |
| summary.add_row( |
| "[red]Failed tokens[/red]", |
| f"{result['failed']} (wrong AES key?)", |
| ) |
|
|
| console.print(Panel( |
| summary, |
| title="[bold green]✓ Done[/bold green]", |
| border_style="green", |
| padding=(0, 1), |
| )) |
|
|
|
|
| |
| |
| |
|
|
| def main() -> None: |
| console.clear() |
| print_banner() |
|
|
| while True: |
| action = questionary.select( |
| "What would you like to do?", |
| choices=[ |
| "Detect PII", |
| "Tokenize file", |
| "Decrypt file", |
| "Exit", |
| ], |
| style=STYLE, |
| ).ask() |
|
|
| if action is None or action == "Exit": |
| console.print("\n[dim]Bye.[/dim]\n") |
| break |
| elif action == "Detect PII": |
| run_detect() |
| elif action == "Tokenize file": |
| run_tokenize() |
| elif action == "Decrypt file": |
| run_decrypt() |
|
|
|
|
| if __name__ == "__main__": |
| main() |
|
|