# This script converts all the manuscript textile files
# into html pages to suit the ruportbook website

# to do:
# automatically collect other assets in manuscript directories ie image files

require 'rubygems'
require 'redcloth'
require 'hpricot'

# pointing to the stuff we're using
@template = "site_template.html"
@source_dir = "../manuscript"
@output_dir = "output"


# finds all the files
def crunch_files(dir)
  Dir.foreach(dir) {|name|
    target = "#{dir}/#{name}"
    if File.directory?(target) and name.index(".") != 0
      crunch_files(target)
    elsif File.file? target and name.index(".rhtml") != nil
      puts "converting #{name}"
      convert_file(target,name)
    end
  }
end

# writes the finished file
def convert_file(target, name)
  html = merge_textile(target)
  f = File.open("#{@output_dir}/#{name.chomp(".rhtml")}.html",'w+')
  f.write html
  f.close
end

# this is orchestrating all the html crunching
def merge_textile(filename)
  # get the textile converted to hpricot
  main_content = Hpricot(parse_textile(filename))
  # create the nav and anchors
  side_nav = build_sidebar_nav(main_content)
  # get the template and insert the new content into it
  page = Hpricot(get_template)
  (page/"#mainContent").inner_html = tidy_footnotes(main_content.to_html)
  (page/"#secondaryContent").inner_html = side_nav
  page
end

# makes the sidebar and anchors
def build_sidebar_nav(content)
  nav = "\n"
  content.search("/h3").each do |heading|
    heading_string = heading.to_plain_text
    nav << "<p><a href=\"##{clean_for_url(heading_string)}\">#{heading_string}</a></p>\n"
    heading.inner_html("<a name=\"#{clean_for_url(heading_string)}\">#{heading.innerHTML}</a>")
  end
  nav
end

# makes the footnote markup a little more attractive
def tidy_footnotes(content)
  content.gsub!(/\(footnote\??: ?/, ' (')
  content
end


# returns the html file used as a template
def get_template
  File.read @template
end

# given a file ref and returns some redcloth
def parse_textile filename
  f = File.read filename
  r = RedCloth.new f
  r.to_html
end

# removes nast chars from text to use in anchors
def clean_for_url(dirty)
  cleaner = dirty.gsub(/[ ]/, '_')
  clean = cleaner.gsub(/[()#'’:;"!@$%^&*+=\[\]?><\/]/, '')
  clean.downcase
end

# preflight check
def preflight
  Dir.mkdir(@output_dir) unless File.exists?(@output_dir) 
  unless File.exists?(@source_dir)
    puts "source_dir not found"
    exit
  end
  unless File.exists?(@template)
    puts "html template not found"
    exit
  end
end

if $0 == __FILE__
  preflight
  crunch_files(@source_dir)
end