Download a ClickOnce application

getonce.py

#!/usr/bin/env python3
from bs4 import BeautifulSoup
import requests
import sys
from urllib.parse import urljoin
from pprint import pprint
from pathlib import Path

def usage():
    print('Usage: getonce.py [application url]')
    sys.exit(1)

if len(sys.argv) < 2:
    usage()

application_url = sys.argv[1]
response = requests.get(application_url)
response.raise_for_status()
soup = BeautifulSoup(response.text, features="xml")
application_url = soup.find('deploymentProvider', codebase=True)['codebase']
install_dependent_assemblies = soup.find_all('dependentAssembly', codebase=True, dependencyType="install")
paths = [tag['codebase'] for tag in install_dependent_assemblies]
urls = [urljoin(application_url, path.replace('\\', '/')) for path in paths]
for url in urls:
    response = requests.get(url)
    response.raise_for_status()
    if url.endswith('.manifest'):
        manifest = BeautifulSoup(response.text, features="xml")
        install_dependent_assemblies = manifest.find_all('dependentAssembly', codebase=True, dependencyType="install")
        assembly_paths = [tag['codebase'] for tag in install_dependent_assemblies]
        file_tags = manifest.find_all('file')
        file_names = [tag['name'] for tag in file_tags]
        deploy_urls = {(urljoin(url, path.replace('\\', '/')) + '.deploy'): path for path in file_names + assembly_paths}
        output_dir = Path('output')
        output_dir.mkdir()
        for url, filename in deploy_urls.items():
            response = requests.get(url)
            response.raise_for_status()
            (output_dir / filename).parent.mkdir(parents=True, exist_ok=True)
            (output_dir / filename).write_bytes(response.content)
            print(f'Downloaded {filename}')