xpath('./text()', './/text()[not(ancestor::pre) and not(ancestor::code) and not(ancestor::div[contains(concat(" ", normalize-space(@class), " "), " prism ")])]').each do |node|
content = node.content
next unless content.valid_encoding?
- next if context[:clean_html_maintain_nbsp] && !content.index(" ").nil?