require 'set'
 
module Ludy
 
  # 2008-05-09 godfat
  module XhtmlFormatter
    module_function
    def format_article html, *allowed_tags
      require 'rubygems'
      require 'hpricot'
 
      allowed_tags = allowed_tags.to_set
      XhtmlFormatter.format_article_elems Hpricot.parse(
        XhtmlFormatter.escape_all_inside_pre(html, allowed_tags)), allowed_tags
    end
 
    def format_autolink html, attrs = {}
      require 'rubygems'
      require 'hpricot'
 
      doc = Hpricot.parse html
      doc.each_child{ |c|
        next unless c.kind_of?(Hpricot::Text)
        c.content = format_url c.content, attrs
      }
      doc.to_html
    end
 
    # translated from drupal-6.2/modules/filter/filter.module
    def format_autolink_regexp text, attrs = {}
      attrs = attrs.map{ |k,v| " #{k}=\"#{v}\""}.join
      # Match absolute URLs.
      " #{text}".gsub(%r{(<p>|<li>|<br\s*/?>|[ \n\r\t\(])((http://|https://|ftp://|mailto:|smb://|afp://|file://|gopher://|news://|ssl://|sslv2://|sslv3://|tls://|tcp://|udp://)([a-zA-Z0-9@:%_+*~#?&=.,/;-]*[a-zA-Z0-9@:%_+*~#&=/;-]))([.,?!]*?)(?=(</p>|</li>|<br\s*/?>|[ \n\r\t\)])?)}i){ |match|
        match = [match, $1, $2, $3, $4, $5]
        match[2] = match[2] # escape something here
        caption = XhtmlFormatter.trim match[2]
        # match[2] = sanitize match[2]
        match[1]+'<a href="'+match[2]+'" title="'+match[2]+"\"#{attrs}>"+
          caption+'</a>'+match[5]
 
      # Match e-mail addresses.
      }.gsub(%r{(<p>|<li>|<br\s*/?>|[ \n\r\t\(])([A-Za-z0-9._-]+@[A-Za-z0-9._+-]+\.[A-Za-z]{2,4})([.,?!]*?)(?=(</p>|</li>|<br\s*/?>|[ \n\r\t\)]))}i, '\1<a href="mailto:\2">\2</a>\3').
 
      # Match www domains/addresses.
      gsub(%r{(<p>|<li>|[ \n\r\t\(])(www\.[a-zA-Z0-9@:%_+*~#?&=.,/;-]*[a-zA-Z0-9@:%_+~#\&=/;-])([.,?!]*?)(?=(</p>|</li>|<br\s*/?>|[ \n\r\t\)]))}i){ |match|
        match = [match, $1, $2, $3, $4, $5]
        match[2] = match[2] # escape something here
        caption = XhtmlFormatter.trim match[2]
        # match[2] = sanitize match[2]
        match[1]+'<a href="http://'+match[2]+'" title="http://'+match[2]+"\"#{attrs}>"+
          caption+'</a>'+match[3]
      }[1..-1]
    end
 
    def format_url text, attrs = {}
      # translated from drupal-6.2/modules/filter/filter.module
      # Match absolute URLs.
      text.gsub(
  %r{((http://|https://|ftp://|mailto:|smb://|afp://|file://|gopher://|news://|ssl://|sslv2://|sslv3://|tls://|tcp://|udp://|www\.)([a-zA-Z0-9@:%_+*~#?&=.,/;-]*[a-zA-Z0-9@:%_+*~#&=/;-]))([.,?!]*?)}i){ |match|
        url = $1 # is there any other way to get this variable?
 
        caption = XhtmlFormatter.trim url
        attrs = attrs.map{ |k,v| " #{k}=\"#{v}\""}.join
 
        # Match www domains/addresses.
        url = "http://#{url}" unless url =~ %r{^http://}
        "<a href=\"#{url}\" title=\"#{url}\"#{attrs}>#{caption}</a>"
      # Match e-mail addresses.
      }.gsub( %r{([A-Za-z0-9._-]+@[A-Za-z0-9._+-]+\.[A-Za-z]{2,4})([.,?!]*?)}i,
              '<a href="mailto:\1">\1</a>')
    end
 
    def format_newline text
      # windows: \r\n
      # mac os 9: \r
      text.gsub("\r\n", "\n").tr("\r", "\n").gsub("\n", '<br />')
    end
 
    private
    def self.trim text, length = 50
      # Use +3 for '...' string length.
      if text.size <= 3
        '...'
      elsif text.size > length
        "#{text[0...length-3]}..."
      else
        text
      end
    end
    def self.escape_all_inside_pre html, allowed_tags
      return html unless allowed_tags.member? :pre
      # don't bother nested pre, because we escape all tags in pre
      html = html + '</pre>' unless html =~ %r{</pre>}i
      html.gsub(%r{<pre>(.*)</pre>}mi){
        # stop escaping for '>' because drupal's url filter would make &gt; into url...
        # is there any other way to get $1?
        "<pre>#{XhtmlFormatter.escape_lt(XhtmlFormatter.escape_amp($1))}</pre>"
      }
    end
    def self.format_article_elems elems, allowed_tags = Set.new, no_format_newline = false
      elems.children.map{ |e|
        if e.kind_of?(Hpricot::Text)
          if no_format_newline
            format_url(e.content)
          else
            format_newline format_url(e.content)
          end
        elsif e.kind_of?(Hpricot::Elem)
          if allowed_tags.member? e.name.to_sym
            if e.empty? || e.name == 'a'
              e.to_html
            else
              e.stag.inspect +
                XhtmlFormatter.format_article_elems(e, allowed_tags, e.stag.name == 'pre') +
                (e.etag || Hpricot::ETag.new(e.stag.name)).inspect
            end
          else
            if e.empty?
              XhtmlFormatter.escape_lt(e.stag.inspect)
            else
              XhtmlFormatter.escape_lt(e.stag.inspect) +
                XhtmlFormatter.format_article_elems(e, allowed_tags) +
                XhtmlFormatter.escape_lt((e.etag || Hpricot::ETag.new(e.stag.name)).inspect)
            end
          end
        end
      }.join
    end
    def self.escape_amp text
      text.gsub('&', '&amp;')
    end
    def self.escape_lt text
      text.gsub('<', '&lt;')
    end
  end
end # of Ludy