This reader extracts structured data, renders each page, keeps raw text, and prepares a production-ready JSON output for each uploaded PDF.