this uses a markdown.treeprocessor as extension to get all html links from the resulting html document