Todo Pass to ocrmypdf for OCR text Support custom page width Optionally download page images without creating PDF Improve image URL extraction Rewrite in Python (?)