Fetch every “page/N/” HTML, scrape Rank/Title/URL from the server-rendered <li>s, and download export.csv.
