Sanitize file names in a directory tree

rename.py

import argparse
import os
import re


def sanitize_filename(filename):
    # Replace characters not allowed in Windows filenames with underscores
    sanitized = re.sub(r'[<>:"/\\|?*]', "_", filename)
    # Remove non-ASCII characters
    sanitized = re.sub(r"[^\x00-\x7F]", "_", sanitized)
    return sanitized


def rename_files_in_directory(directory):
    for root, dirs, files in os.walk(directory):
        for filename in files:
            sanitized_name = sanitize_filename(filename)
            if sanitized_name != filename:
                old_path = os.path.join(root, filename)
                new_path = os.path.join(root, sanitized_name)
                os.rename(old_path, new_path)
                print(f"Renamed: {old_path} -> {new_path}")


if __name__ == "__main__":
    parser = argparse.ArgumentParser(
        description="Rename files in a directory to replace invalid characters with underscores."
    )
    parser.add_argument("directory", type=str, help="The directory to process")
    args = parser.parse_args()

    rename_files_in_directory(args.directory)