Some PDFs render correctly, but extracted text has broken Unicode because of embedded subset TrueType fonts and missing or invalid ToUnicode / font encoding.
We also need a reliable way to detect micro-spaces or invisib…...space characters not causing line breaks in PDF converted from...