I have several problematic PDFs, which I am attempting to convert to PDF/A-1a.
These documents utilize CID Identity-H embedded subsets, generated with Acrobat Distiller 20.0. I have performed searches for tools which could utilize OCR to scan the rendered output for the purpose of either automatically generating the missing ToUnicode CMaps, or, at the very least, offering high probability candidates for user selection, yet have been unsuccessful in this endeavor.
The glyphs are clearly legible. If this is not the preferred method of building ToUnicode CMaps for these files, is there a common visual utility (or method) for building the code points? I have attempted font extraction into FontForge, which has failed (FontForge output states that the PDFs are unreadable).
Thank you!
