Forum Bugs

Metadata and PDF-A/1a - question

In our application we generate PDF-A/1a PDF files and we use the --pdf-xmp option to add customer specific metadata.
When we validate this generated PDF to check if it satisfies the PDF-A/1a standards with some online tool (e.g., we get the following error:

The schema description for namespace 'pdfx:' ( is missing.

When we check the generated PDF file with a metadata tool (e.g. pdfinfo -meta <PDFfile> on a Linux system) however, the namespace declaration is visible in the metadata part in the following manner (example):

<?xpacket begin="" id="W5M0MpCehiHzreSzNTczkc9d"?>
<x:xmpmeta xmlns:x="adobe:ns:meta/">
<rdf:RDF xmlns:rdf="">
<rdf:Description xmlns:dc="" xmlns:pdf="" xmlns:pdfx="" xmlns:pdfaid="" rdf:about="">
<!-- ... metadata ... -->
<?xpacket end="r"?>

Can you give us a clue?
It appears that validator does not recognise values using the rdf:resource attribute form:
<pdfaSchema:namespaceURI rdf:resource=""/>

We will change Prince to write namespaceURI values in element content:

By the way, since you are generating PDF/A files, you will need to provide PDF/A extension schema descriptions for any custom properties, if you are not doing so.
Hi Wang,
Thanks for the information sofar. Some more question: are the modifications you mentioned in your last reply done into the PrinceXML software so that we can upgrade prince at out system with these modifications?
This issues is fixed now in the latest builds.

Edited by mikeday

This issue has also been fixed in Prince 12.5.