[Fixing the scraper to work with the new PyPI page will now require full HTML parsing, and that seems like overkill]