onion_smasher.py

import os
import shutil
import zipfile

# Loops through nested zip folders and extracts all files to output folder.
# Intended to simplify the translation ETL process.
# Update the zip_file_path with the name of the source .zip file to be unpacked.
zip_file_path = './test_data_set.zip'
destination_folder_path = './test_output'

def unzip_and_copy_files(zip_file_path, destination_folder_path):
    # Create output folder
    os.makedirs(destination_folder_path, exist_ok=True)
    # Create a temporary folder
    temp_folder = 'temp_folder'
    os.makedirs(temp_folder, exist_ok=True)

    # Recursive function to handle nested zip files
    def extract_zip(file_path, extract_to):
        with zipfile.ZipFile(file_path, 'r') as zip_ref:
            zip_ref.extractall(extract_to)
            for name in zip_ref.namelist():
                if name.endswith('.zip'):
                    extract_zip(os.path.join(extract_to, name), extract_to)

    # Start extraction
    extract_zip(zip_file_path, temp_folder)

    # Copy all files (not directories) to the destination folder
    for root, dirs, files in os.walk(temp_folder):
        for file in files:
            if not file.endswith('.zip'):  # Skip zip files
                shutil.copy(os.path.join(root, file), destination_folder_path)

    # Clean up the temporary folder
    shutil.rmtree(temp_folder)

# Usage
unzip_and_copy_files(zip_file_path, destination_folder_path)