PDF Version: PDF-1.7
pdfcpu Version: pdfcpu: v0.9.1 dev commit: 96c3c37 (2024-11-04T11:39:59Z)

When a PDF only has an UF entry for embedded file attachments, but no F entry, the PDF validation and subsequent operations (such as listing and extraction of the attachments) fail with fileSpecDict required entry=F missing.

For example:

$ pdfcpu validate x.pdf 
validating(mode=relaxed) x.pdf ...
validation error (obj#:37): dict=fileSpecDict required entry=F missing

According to PDF 32000-1:2008, 7.11.3 File Specification Dictionaries, if I understood it correctly, allows to use UF with UTF-16 unicode strings. The spec states that the UF entry should be used in addition to the F entry for backwards compatibility, but does not mandate that. So I'd guess that only using the UF entry should be allowed at least in relaxed validation mode.

Thank you very much, best regards
Andreas