[PDFBOX-2896] XMPBox not creating valid "title" entry in DublinCoreSchema in trunk

• Type: Bug
• Status: Closed
• Priority: Minor
• Resolution: Fixed
• Affects Version/s: 2.0.0
• Fix Version/s: 2.0.0
• Component/s: XmpBox
• Assignee: Unassigned
• Reporter: Tim Allison
• Created: 21/Jul/15 01:27
• Updated: 17/Mar/16 19:07
• Resolved: 22/Jul/15 20:26

Description

On TIKA-1678, I was trying to generate a test PDF that had a dc:title in the XMP with XMPBox from PDFBox's trunk. I modified the code from CreatePDFA by adding this:

DublinCoreSchema dc = xmp.createAndAddDublinCoreSchema();
dc.setTitle("this is the title");

The generated PDF doesn't appear to have a compliant dc:title entry in the XMP.

tilman noted the divergence from the standard here.

What PDFBox does:

      <dc:title>
        <rdf:Alt>
          <dc:li>this is the title</dc:li>
        </rdf:Alt>
      </dc:title>

It should be:

          <dc:title>
            <rdf:Alt>
              <rdf:li xml:lang="x-default">this is the title</rdf:li>
            </rdf:Alt>
          </dc:title>

Error message from the PDF-Tools validator:

'dc:li' is not allowed in arrays. The elements must be rdf:li or rdf:_N, where N is a positive number.
There is only one RDF resource allowed in XMP.

Issue Links

• relates to: PDFBOX-2897 Preflight not flagging bad xml generated by XMPBox for dc:title
• relates to: TIKA-1678 PDF metadata extraction fails to spot UTF-16 encoded title