What is the EPUB unique identifier?

Why it matters when you convert a PDF to EPUB

A PDF has no identifier a converter can reuse, so one has to be generated. If it changes every time the same file is converted, a reading app may treat each conversion as a different book; if two different books share one, they can overwrite each other in a library.

EPUBCheck reports a unique-identifier attribute that points nowhere (OPF-030), a missing attribute (OPF-048) and a value marked as a UUID that is not a valid UUID (OPF-085).

How EpubCC handles it: the identifier is a urn:uuid derived from the book's content (its title, chapter titles and first paragraphs), so converting the same PDF again gives the same identifier and the book does not appear twice in your reader's library.

Related terms

• What is Dublin Core metadata (dc:*)?
• What is the OPF package document?
• What is the NCX table of contents?

EPUBCheck messages about this

• OPF-030 The unique-identifier "…" was not found.
• OPF-048 Package tag is missing its required unique-identifier attribute and value.
• OPF-085 "dc:identifier" value "…" is marked as a UUID, but is an invalid UUID.