Question

For PDF/A documents which contain additional information (like barcode data) in their embedded metadata, is there a simple way to access this information from a Delphi application – instead of using a full-featured PDF reader / writer application library? (For example open source command line tools)

Answer 1

As far as I know the PDF metadata is just available in plain text in the PDF/A document.

So you can search for the <x:xmpmeta pattern in the file, then the whole metadata is available here, until </x:xmpmeta>. Then you retrieve the XML content.

function ExtractPDFMetadata(const aPDFFileName: TFileName): UTF8String;
var tmp: RawByteString;
    i: integer;
begin
  with TFileStream.Create(aPDFFileName,fmOpenRead) do 
  try
    SetLength(tmp,Size);
    Read(tmp[1],Size);
  finally
    Free;
  end;
  result := '';
  i := pos('<x:xmpmeta',tmp);
  if i=0 then exit;
  delete(tmp,1,i-1);
  i := pos('</x:xmpmeta>',tmp);
  if i=0 then exit;
  result := copy(tmp,1,i+12);
end;

Then this UTF8String could be used to parse the XML content, as usual.

You can also locate this XML content using the “Metadata” kind of entry in the PDF catalog: this will be faster (don’t have the parse the file), but will need much extra coding.