Extract URLs and Notes from PDFs

pdf_notes_urls_extract.rb

# Credits to https://gist.github.com/danlucraft/5277732
require 'pdf-reader'

class Annotations
  def initialize(reader, page)
    @objects = reader.objects
    @page = reader.page(page)
    @all_annots = annots_on_page
  end

  def notes
    @all_annots.select { |a| is_note?(a) }
  end

  def links
    links_obj = @all_annots.select { |a| is_link?(a) }
    links_obj.map { |link| @objects[link[:A]][:URI] } unless links_obj.nil?
  end

  private

  def is_link?(object)
    object[:Type] == :Annot && [:Link].include?(object[:Subtype])
  end

  def is_note?(object)
    object[:Type] == :Annot && [:Text, :FreeText].include?(object[:Subtype])
  end

  def annots_on_page
    references = (@page.attributes[:Annots] || [])
    lookup_all(references).flatten
  end

  def lookup_all(refs)
    refs = *refs
    refs.map { |ref| lookup(ref) }
  end

  def lookup(ref)
    object = @objects[ref]
    return object unless object.is_a?(Array)
    lookup_all(object)
  end
end

# Usage example
puts 'Running...'
file = "file_path"
puts "File: #{file}"

doc = PDF::Reader.new(file)
anot = Annotations.new(doc, 1)
puts "#{anot.links}"
puts "#{anot.notes}"