When Panther was young I filed an enhancement request (rdar://
3569618) to expose an API to extract images, color profiles, font
IDs, unicode text, etc. from PDF documents. In the early days of
Tiger this bug report was closed and I got a message saying it was in
Tiger. I don't have the message anymore but my impression was that
this could be done using the new PDF scanner features (or perhaps
PDFKit) in Tiger. I've been reviewing these API's recently and I
can't find a way to do this. Am I just missing something, or should
I reopen this enhancement request?
For example, consider the case of converting a bitmap image embedded
in a PDF document into a CGImageRef with the correct color profile.
Frequently this is fairly straightforward, but it can get very
complicated. Surely Quartz is already doing this when it renders the
PDF document and I'd rather not reinvent that wheel if I can help
it. The same is true of getting a Mac font ID from a PDF font