Detect Metadata About PDF From Javascript

If my JavaScript code is running in a Chrome extension, and the browser has loaded a PDF file, can I detect metadata about the loaded PDF (number of pages, etc.)? Extra challenge:

Solution 1:

After some quick Google-fu I learned that PDFs store metadata in XMP Format (XML). So you can read the raw file data a pick out metadata with some simple regex.

Select a PDF file from your computer to see the demo:

document.getElementById('f').oninput = async function() {
  var pdf = this.files[0];
  var details = await pdfDetails(pdf);
  console.log(details);
};

function pdfDetails(pdfBlob) {
  return new Promise(done => {
    var reader = new FileReader();
    reader.onload = function() {
      var raw = reader.result;
      var Pages = raw.match(/\/Type[\s]*\/Page[^s]/g).length;
      var regex = /<xmp.*?:(.*?)>(.*?)</g;
      var meta = [{
        Pages
      }];
      var matches = regex.exec(raw);
      while (matches != null) {
        matches.shift();
        meta.push({
          [matches.shift()]: matches.shift()
        });
        matches = regex.exec(raw);
      }
      done(meta);
    };
    reader.readAsBinaryString(pdfBlob);
  });
}

<input type=file id=f accept=".pdf">

Solution 2:

A PDF document doesn't list the number of pages in its metadata. Even if you added some custom metadata to track that information it wouldn't be in a standard way PDF readers would be expected to understand.