Multiprocess MD5 hash checker in Python

md5_hash_checker.py

import os
import hashlib
from multiprocessing import Pool, cpu_count

def calculate_md5(file_path):
    """Calculate the MD5 hash of a file."""
    try:
        with open(file_path, 'rb') as f:
            md5_hash = hashlib.md5()
            for chunk in iter(lambda: f.read(4096), b""):
                md5_hash.update(chunk)
        return file_path, md5_hash.hexdigest()
    except Exception as e:
        return file_path, f"Error: {e}"

def get_bin_files(folder_path):
    """Retrieve all .bin files from the specified folder."""
    return [
        os.path.join(folder_path, file)
        for file in os.listdir(folder_path)
        if file.endswith('.bin') and os.path.isfile(os.path.join(folder_path, file))
    ]

def main():
    """Main function to execute the script."""
    folder_path = input("Enter the path to the folder containing .bin files: ").strip()

    if not os.path.isdir(folder_path):
        print(f"The folder '{folder_path}' does not exist or is not a directory.")
        return

    bin_files = get_bin_files(folder_path)

    if not bin_files:
        print(f"No .bin files found in the folder '{folder_path}'.")
        return

    print(f"Found {len(bin_files)} .bin file(s). Calculating MD5 hashes using {cpu_count()} cores...")

    with Pool(processes=cpu_count()) as pool:
        results = pool.map(calculate_md5, bin_files)

    print("\nMD5 Hash Results:")
    for file_path, md5 in results:
        print(f"{file_path}: {md5}")

if __name__ == "__main__":
    main()