rename_files.py - Rename files by removing accents and special chars, safe for UTF-8 (useful when sending files between different OS, ex.: Mac to Windows)

rename_files.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
rename_files.py - Rename files by removing accents and special chars, safe for UTF-8

Usage:
    python3 rename_files.py <directory> [--dry-run]

Features:
    - Normalizes accents (NFKD)
    - Replaces spaces and special chars with underscores
    - Preserves file extensions
    - Optional dry-run mode

License:
    GNU General Public License v3.0
"""

import os
import sys
import unicodedata
import re

def normalize_filename(name: str) -> str:
    """
    Normalize filename: remove accents, replace non-alphanumeric chars with '_'.
    """
    # Separate accents
    nfkd = unicodedata.normalize('NFKD', name)
    # Remove accents
    no_accents = ''.join(c for c in nfkd if not unicodedata.combining(c))
    # Split name and extension
    base, ext = os.path.splitext(no_accents)
    # Replace non-word chars with _
    base = re.sub(r'[^\w]+', '_', base)
    # Remove leading/trailing _
    base = base.strip('_')
    return f"{base}{ext}"

def main():
    if len(sys.argv) < 2:
        print("Usage: python3 rename_files.py <directory> [--dry-run]")
        sys.exit(1)

    directory = sys.argv[1]
    dry_run = '--dry-run' in sys.argv

    if not os.path.isdir(directory):
        print(f"Error: {directory} is not a directory")
        sys.exit(1)

    for entry in os.listdir(directory):
        path = os.path.join(directory, entry)
        if os.path.isfile(path):
            new_name = normalize_filename(entry)
            if new_name != entry:
                if dry_run:
                    print(f"[DRY-RUN] Would rename: {entry} -> {new_name}")
                else:
                    new_path = os.path.join(directory, new_name)
                    print(f"Renaming: {entry} -> {new_name}")
                    os.rename(path, new_path)

if __name__ == "__main__":
    main()