- Extract the repeated logic for checking the data passed to the
   event's callback into a `createCallback` helper

 - Add separate tests for HTML and PDF documents

 - Add test for the error path when getting the URL of a PDF fails