Use embedded Calibre metadata for PDFs if available

[microtherion]

Idea Description

Currently, Kavita only uses the file name for PDF metadata. But in fact, tools such as calibre do add inline metadata when saving PDFs. Here's an example file demonstrating this, with the metadata appearing inside an x:xmpmeta tag.

The format seems to consist of a general RDF description element (which may in fact be written by other tools as well, although I'm not currently aware of any), as well as a calibre description element which adds some tags presumably not defined in the general element.

Experiments in PDF Metadata.pdf

Idea Category

Feature Enhancement

Duration of Using Kavita

A few months

Before submitting

• I've already searched for existing ideas before posting.

[microtherion]

The metadata written for this example is:

<?xpacket begin="���" id="W5M0MpCehiHzreSzNTczkc9d"?>
<x:xmpmeta xmlns:x="adobe:ns:meta/">
  <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
    <rdf:Description xmlns:dc="http://purl.org/dc/elements/1.1/" rdf:about="">
      <dc:title>
        <rdf:Alt>
          <rdf:li xml:lang="x-default">calibre PDF metadata</rdf:li>
        </rdf:Alt>
      </dc:title>
      <dc:description>
        <rdf:Alt>
          <rdf:li xml:lang="x-default"></rdf:li>
        </rdf:Alt>
      </dc:description>
      <dc:creator>
        <rdf:Seq>
          <rdf:li>Matt Neeracher</rdf:li>
        </rdf:Seq>
      </dc:creator>
      <dc:subject>
        <rdf:Bag>
          <rdf:li>some tag</rdf:li>
        </rdf:Bag>
      </dc:subject>
      <dc:publisher>
        <rdf:Bag>
          <rdf:li>Github Issue</rdf:li>
        </rdf:Bag>
      </dc:publisher>
      <dc:date>
        <rdf:Seq>
          <rdf:li>2024-08-16T00:00:00+02:00</rdf:li>
        </rdf:Seq>
      </dc:date>
      <dc:language>
        <rdf:Bag>
          <rdf:li>en</rdf:li>
        </rdf:Bag>
      </dc:language>
    </rdf:Description>
    <rdf:Description xmlns:xmp="http://ns.adobe.com/xap/1.0/" xmlns:xmpidq="http://ns.adobe.com/xmp/Identifier/qual/1.0/" rdf:about="">
      <xmp:MetadataDate>2024-08-10T01:00:44.022867+02:00</xmp:MetadataDate>
    </rdf:Description>
    <rdf:Description xmlns:prism="http://prismstandard.org/namespaces/basic/2.0/" rdf:about=""/>
    <rdf:Description xmlns:pdfx="http://ns.adobe.com/pdfx/1.3/" rdf:about=""/>
    <rdf:Description xmlns:calibre="http://calibre-ebook.com/xmp-namespace" xmlns:calibreSI="http://calibre-ebook.com/xmp-namespace-series-index" xmlns:calibreCC="http://calibre-ebook.com/xmp-namespace-custom-columns" rdf:about="">
      <calibre:rating>8</calibre:rating>
      <calibre:series rdf:parseType="Resource">
        <rdf:value>Experiments in PDF Metadata</rdf:value>
        <calibreSI:series_index>20.00</calibreSI:series_index>
      </calibre:series>
      <calibre:timestamp>2024-08-10T00:58:00.805850+02:00</calibre:timestamp>
      <calibre:title_sort>metadata, calibre PDF</calibre:title_sort>
      <calibre:author_sort>Neeracher, Matthias</calibre:author_sort>
    </rdf:Description>
  </rdf:RDF>
</x:xmpmeta>

[microtherion]

Per discord discussion, I'll take a shot at implementing this myself.

[bhcompy]

Any update on this? Extremely interested

[microtherion]

I had a working first implementation, which was rather naive and not performant. Currently reimplementing in a more principled way, but somewhat time constrained at the moment.

[bhcompy]

Looking forward to it. Thank you!

[42Craft]

I had a working first implementation, which was rather naive and not performant. Currently reimplementing in a more principled way, but somewhat time constrained at the moment.

Looking forward to it :)