Contents

module HTML
  class Pipeline
    # HTML filter that adds a 'name' attribute to all headers
    # in a document, so they can be accessed from a table of contents
    #
    # TODO: besides adding the name attribute, we should get around to
    # eventually generating the Table of Contents itself, with links
    # to each header
    class TableOfContentsFilter < Filter
      PUNCTUATION_REGEXP = RUBY_VERSION > "1.9" ? /[^\p{Word}\- ]/u : /[^\w\- ]/

      def call
        headers = Hash.new(0)
        doc.css('h1, h2, h3, h4, h5, h6').each do |node|
          name = node.text.downcase
          name.gsub!(PUNCTUATION_REGEXP, '') # remove punctuation
          name.gsub!(' ', '-') # replace spaces with dash

          uniq = (headers[name] > 0) ? "-#{headers[name]}" : ''
          headers[name] += 1
          if header_content = node.children.first
            header_content.add_previous_sibling(%Q{<a name="#{name}#{uniq}" class="anchor" href="##{name}#{uniq}"><span class="octicon octicon-link"></span></a>})
          end
        end
        doc
      end
    end
  end
end

Version data entries

1 entries across 1 versions & 1 rubygems

Version	Path
html-pipeline-0.1.0	lib/html/pipeline/toc_filter.rb