Extract the text of a PDF at a public URL, page by page, with its title, author and page count. Up to 10 MB; scanned PDFs without a text layer return little text.