требует установленного гема loofah (gem install loofah)

Rakefile

desc <<END 
Find all html files from the specified directory and clean them: removes comments, whitespaces, and carriage return.

Before using install loofah gem:
  gem install loofah
  
Usage:
  rake clean_html DIR=my_dir
  
Warning: by default, DIR variable point current directory - #{Dir.pwd}
END

task 'clean_html' do
  require 'loofah'
  require 'active_support/core_ext/string'
  
  clean = Loofah::Scrubber.new do |node|
    if node.type == Nokogiri::XML::Node::COMMENT_NODE
      node.remove
    else
      if node.name == 'pre'
        Loofah::Scrubber::STOP # don't bother with the rest of the subtree
      elsif node.type == Nokogiri::XML::Node::TEXT_NODE
        if node.content.blank?
          node.remove
        else
          node.content = node.content.strip
        end
      end
    end
  end
  
  ENV['DIR'] ||= Dir.pwd
  
  files = Dir["#{ENV['DIR']}/**/*.html"]
  files.each do |file_path|
    puts "parse file: #{file_path}"
    html = File.open(file_path, 'r').read
    clear_tree = Loofah.document(html).scrub!(clean)
    clear_html = clear_tree.to_html(:indent_text => '', :indent => 0, :save_with => 0)
    
    html = File.open(file_path, 'w').write clear_html
  end
end