Getting proper PDFs from HTML has always been a PITA and has required workaround like screenshot -> convert to PDF in a lot of cases. Has anyone benchmarked this??