What can be done about identity-h fonts inside an acrobat file? Can they be isolated to change?

In trying to run preflight embedding I discover their are identity-h fonts inside my file, mixed in among the others, and they throw 'no suitable font errors...' and weird extraction errors. How can I isolate them to change them? Or what solution is there to this please?


Voted Best Answer

BLOT - If Preflight is throwing that sort of an error - most likely you have yourself a boogered up mess on your hands. What can be used to correct these? I have yet to fine one - other than to just rebuild the file. You could try Refrying it using HighQualityPrint or one of the PDF/X Distiller profiles.

Identity-h fonts generally mean CID encoded horizontal (the 'h') font. What you usually see them listed as 'CID Identity-H' They usually come from TrueType fonts. Now, there has been some discussion if CID fonts are a good thing or bad thing. They are a part of the spec, and generally used for Asian languages. However, a lot of tools use them for TrueType font embedding due to historical lazyness.

They can be created properly by some tools and end up as a bletcherous mess by others. It ends up being a combination of the tool used to create the PDF and the state of the original font files (were they good or bad to begin with).

In my experience in receiving thousands of PDFs from outside producers (usually office worker types - not professional document designers) is that you can be hit or miss with them.

As such, we have a formal process to flag all PDFs with CID fonts with a WARNING. The PDFs aren't good - they aren't bad. They are something you need more information on before you can make a decision. We then manually walk the suspect PDF through the production workflow to see if it behaves well or not. If good, it's signed off and put in production. If not, it's sen't back to the producer with some instructions on how to do a better job.

In short - this is an area where you need to develop experience in what works and what doesn't. At this point the tools are not robust enough to completely automate the checking.

What tools? Acrobat Preflight - of course. Then pick up PDFlib's FontReporter plugin. For the price you can't beat it (free). Heck, you can't beat it even with PitStop! For nothing else than to see how the encoding table was built (is it 'normal' looking(ANSI, ASCII, ISO Latin-1, Standard, etc.)) or does it look like a mess. What's the actual subset font-name (not just what you see in the font listings), etc.

Hope this helps.


By Douglas Hanna   

Please specify a reason: