Update: You should read the comments, then check out the extended verion of truncate_html that uses Tidy.

For the blogging Web site we wanted snippets of blog post for the front page and search results. This wasn’t a problem at all for our test blog posts, which we made by pasting lorem ipsum into a text area:

<%= post.body.first(250) %>…

But this did cause a problem for real world blog posts, where the user would, for some reason beyond my understanding, write the post in Microsoft’s Word first, then paste in the, er, HTML. I sanitized and removed various bits of HTML, but there came a point where it just didn’t make sense to have

<img src="…

So, I set out to truncate XML properly. To this end I extended String with a truncate_html method. I just stuck it in lib/ in my Rails project and require‘d it in the Post model. Here, have it:

require 'rexml/parsers/pullparser'

class String
  def truncate_html(len = 30)
    p = REXML::Parsers::PullParser.new(self)
    tags = []
    new_len = len
    results = ''
    while p.has_next? && new_len > 0
      p_e = p.pull
      case p_e.event_type
      when :start_element
        tags.push p_e[0]
        results << "<#{tags.last} #{attrs_to_s(p_e[1])}>"
      when :end_element
        results << "</#{tags.pop}>"
      when :text
        results << p_e[0].first(new_len)
        new_len -= p_e[0].length
      else
        results << "<!-- #{p_e.inspect} -->"
      end
    end
    tags.reverse.each do |tag|
      results << "</#{tag}>"
    end
    results
  end

  private

  def attrs_to_s(attrs)
    if attrs.empty?
      ''
    else
      attrs.to_a.map { |attr| %{#{attr[0]}="#{attr[1]}"} }.join(' ')
    end
  end
end