In most cases metadata is all about interoperability; it is only useful if creator and processor agree on syntax and semantics. In the context of PDF this also means that creator and processor must agree on where metadata is present in the often-complex internal PDF structure.
Metadata locations are well defined for XMP metadata on the PDF document level and for some PDF objects, but not for all objects where metadata can occur (pages, page objects and resources such as fonts or ICC profiles). However, in almost all cases sufficient guidance can be derived from the PDF specification, albeit sometimes in an indirect way.
This document summarizes exactly where PDF creators should add XMP metadata to PDF objects, and thus where PDF processors should search for it.
The principal authors are Boris Doubrov (Dual Lab) and Dietrich von Seggern (callas software).
This free application note summarizes precisely where PDF creators should add XMP metadata to PDF 2.0 objects, and thus where PDF 2.0 processors should search for it.
Published by the PDF Association © 2021
Community: PDF TWG
Publications: PDF 2.0 Application Note 001: Black Point Compensation, PDF 2.0 Application Note 002: Associated Files