wpautop for Ruby

wpautop.rb

###
# Replaces double line-breaks with paragraph elements.
#
# A group of regex replaces used to identify text formatted with newlines and
# replace double line-breaks with HTML paragraph tags. The remaining
# line-breaks after conversion become <<br />> tags, unless $br is set to '0'
# or 'false'.
#
# @since 0.71
#
# @param string pee The text which has to be formatted.
# @param bool br Optional. If set, this will convert all remaining line-breaks after paragraphing. Default true.
# @return string Text which has been converted into correct paragraph tags.
###
def wpautop(pee, br = true)
  pre_tags = []

  return if pee.strip == ''

  pee = "#{pee}\n" # just to make things a little easier, pad the end

  if pee.include? '<pre'
    pee_parts = pee.split '</pre>'
    last_pee = pee_parts.pop
    pee = ''

    pee_parts.each_with_index do |pee_part, i|
      start = pee_part.index '<pre'

      # Malformed html?
      if start.nil?
        pee += pee_part
        next
      end

      name = "<pre wp-pre-tag-#{i}></pre>"
      pre_tags[name] = pee_part[start..-1]

      pee += pee_part[0..start] + name
    end

    pee += last_pee
  end

  # Extra text formatting
  require 'htmlentities'
  encoder = HTMLEntities.new(:expanded)
  pee.gsub! /(?<=[^\s\n])((?<=\s)---?(?=\s)|--)(?=[^\s\n])/,'—'
  pee.gsub! /(?<=[^\s\n]\s)--?(?=\s[^\s\n])/, '–'
  pee.gsub! /(?<=\A|>)(.+?)(?=\z|<)/ do |match|
    encoder.encode match, :decimal
  end

  pee.gsub!(/<br \/>\s*<br \/>/, "\n\n")
  # Space things out a little
  all_blocks = '(?:table|thead|tfoot|caption|col|colgroup|tbody|tr|td|th|div|dl|dd|dt|ul|ol|li|pre|form|map|area|blockquote|address|math|style|p|h[1-6]|hr|fieldset|legend|section|article|aside|hgroup|header|footer|nav|figure|figcaption|details|menu|summary)'
  pee.gsub! Regexp.new("(<#{all_blocks}[^>]*>)"), "\n\\1"
  pee.gsub! Regexp.new("(</#{all_blocks}>)"), "\\1\n\n"

  if pee.include? '<option'
    # no P/BR around option
    pee.gsub! /\s*<option/, '<option'
    pee.gsub! /<\/option>\s*/, '</option>'
  end

  if pee.include? '</object>'
    # no P/BR around param
    pee.gsub! /(<object[^>]*)\s*/, '\1'
    pee.gsub! /\s*<\/object>/, '</object>'
    pee.gsub! /\s*(<\/?(?:param|embed)[^>]*>)\s*/, '\1'
  end

  if pee.include?('<source') || pee.include?('<track')
    # no P/BR around source and track
    pee.gsub! /([<\[](?:audio|video)[^>\]]*[>\]])\s*/, '\1'
    pee.gsub! /\s*([<\[]\/(?:audio|video)[>\]])/, '\1'
    pee.gsub! /\s*(<(?:source|track)[^>]*>)\s*/, '\1'
  end

  pee.gsub! /\n\n+/, "\n\n" # take care of duplicates
  # make paragraphs, including one at the end
  pees = pee.split /\n\s*\n/
  pee = ''

  pees.each do |tinkle|
    pee += "<p>#{tinkle.gsub(/\A\n*(.+?)\n*\z/m, '\1')}</p>\n"
  end

  pee.gsub! /<p>\s*<\/p>/, '' # under certain strange conditions it could create a P of entirely whitespace
  pee.gsub! /<p>([^<]+)<\/(div|address|form)>/, '<p>\1</p></\2>'
  pee.gsub! Regexp.new("<p>\s*(</?#{all_blocks}[^>]*>)\s*</p>"), '\1' # don't pee all over a tag
  pee.gsub! /<p>(<li.+?)<\/p>/, '\1' # problem with nested lists
  pee.gsub! /<p><blockquote([^>]*)>/i, '<blockquote\1><p>'
  pee.gsub! '</blockquote></p>', '</p></blockquote>'
  pee.gsub! Regexp.new("<p>\s*(</?#{all_blocks}[^>]*>)"), '\1'
  pee.gsub! Regexp.new("(</?#{all_blocks}[^>]*>)\s*</p>"), '\1'

  if br
    pee.gsub! /<(script|style).*?<\/\\1>/ do |match|
      match.gsub "\n", '<WPPreserveNewline />'
    end
    pee.gsub! /(?<!<br \/>)\s*\n/, "<br />\n"
    pee.gsub! '<WPPreserveNewline />', "\n"
  end

  pee.gsub! Regexp.new("(</?#{all_blocks}[^>]*>)\s*<br />"), '\1'
  pee.gsub! /<br \/>(\s*<\/?(?:p|li|div|dl|dd|dt|th|pre|td|ul|ol)[^>]*>)/, '\1'
  pee.gsub! /\n<\/p>$/, '</p>'

  unless pre_tags.empty?
    pre_tags.each do |k, v|
      pee.gsub! k, v
    end
  end

  pee
end