I've written a small Nanoc plugin that filters output via CodeRay.

I write my code segments in Markdown. This filter finds the resulting
<pre><code> blocks generated by Bluecloth and filters them via
CodeRay.

Hope this is useful to someone.

Cheers,

Rob

#
# This is a small filter that searches for <pre><code>...</code></pre>
blocks inside the
# given document and rams the block through CodeRay. This should allow
you to get nice
# syntax highlighted code from straight Markdown (I tested via txt
files through Bluecloth).
#
# Some caveats:
#
# - The <pre><code> block is automatically wrapped in a <div
class="CodeRay"> by this filter.
# - I've defaulted to C below; current this is hardwired.
# - You'll need to add suitable CSS for CodeRay somewhere; this isn't
inlined/generated.
#
# Needs Gems:
#
# - Hpricot
# - CodeRay
#
class CoderayFilter < Nanoc::Filter

  identifier :coderay

  def run(content)
    nanoc_require 'hpricot'
    nanoc_require 'coderay'

    doc = Hpricot(content)

    blocks = doc.search('pre > code').map do |block|

      # CodeRay will re-escape, leading to a mess as the content is
already escaped.
      # Need to therefore unescape first.
      code = codify(html_unescape(block.inner_html), :c)

      block.inner_html = code

    end

    # Crusty; wrap in div.
    doc.to_s.gsub('<pre><code>', '<div
class="CodeRay"><pre><code>').gsub('</code></pre>', '<</code></pre></
div>')

  end

  private

   def html_unescape(a_string)
     a_string.gsub('&amp;', '&').gsub('&lt;', '<').gsub('&gt;',
'>').gsub('&quot;', '"')
   end

   def codify(str, lang)
     %{#{CodeRay.scan(str, lang).html}}
   end