RSpec PDF content matcher

pdf_support.rb

module PDF
  class Loader
    def self.content_for(body)
      temp_pdf = Tempfile.new('pdf')
      temp_pdf << body.force_encoding('UTF-8')
      temp_pdf.close

      receiver = PDF::SimplePageTextReceiver.new
      PDF::Reader.file(temp_pdf.path, receiver)
      receiver.content
    end
  end

  class SimplePageTextReceiver
    attr_accessor :content

    def initialize
      @content = []
    end

    # Called when page parsing starts
    def begin_page(arg = nil)
      @content << ""
    end

    # record text that is drawn on the page
    def show_text(string, *params)
      @content.last << string.strip
    end

    # there's a few text callbacks, so make sure we process them all
    alias :super_show_text :show_text
    alias :move_to_next_line_and_show_text :show_text
    alias :set_spacing_next_line_show_text :show_text

    # this final text callback takes slightly different arguments
    def show_text_with_positioning(*params)
      params = params.first
      params.each { |str| show_text(str) if str.kind_of?(String)}
    end
  end
end

RSpec::Matchers.define :have_pdf_content do |text|
  match do |body|
    PDF::Loader.content_for(body).join == text.gsub(/ /, '')
  end

  failure_message_for_should do |body|
    "pdf should have content:\n'#{text}'"
  end

  failure_message_for_should_not do |body|
    "pdf should not have content:\n'#{text}'"
  end
end