Tag: best practices in PDF text extraction