A ruby script to convert a Scrivener project (.scriv) into a single Markdown (.md). NOTE: This is my first experience with Ruby. Don't ask me anything too complicated because I just took the code, inferred the needed pieces, trimmed the fat, and adjust until I got the result!

scriv2md.rb

#!/usr/local/bin/ruby
# SOURCE: scrivwatch.rb by Brett Terpstra, 2011, https://gist.github.com/ttscoff/1676667
# REPURPOSED AND RENAMED BY: Danton Noriega, April 2016
# Instructions:
#   1. place 'scriv2md.rd' in same directory as .scriv project
#   2. give permissions to 'scriv2md.rd':
#           chmod u+x ./scriv2md.rd
#   3. use!
#           ./scriv2md.rd /path/to/document.scriv

require 'fileutils'
require 'rexml/document'
require 'optparse'

def usage
  puts "Usage: scriv2md.rb /path/to/document.scriv"
  exit
end

if ARGV[0] =~ /\.scriv\/?$/
  path = File.expand_path(ARGV[0].gsub(/\/$/,''))
else
  usage
end

# FUNCTIONS
def get_children(ele,path,files,depth)
  ele.elements.each('*/BinderItem') do |child|
        # Ignore docs set to not include in compile
        includetag = REXML::XPath.first( child, "MetaData/IncludeInCompile" )
        if !includetag.nil? && includetag.text == "Yes"
          id = child.attributes["ID"]
          # passing type, would eventually use to control header/title output
          type = child.attributes["Type"]
          title = child.elements.to_a[0].text
          file = "#{path}/Files/Docs/#{id}.rtf"
          filepath = File.exists?(file) ? file : false
          files << { 'path' => filepath, 'title' => title, 'depth' => depth, 'type' => type }
        end
        get_children(child,path,files,depth+1)
  end
end

# Take the path to the scrivener file and open the internal XML file.
def get_rtf_file_array(scrivpath)
  scrivpath = File.expand_path(scrivpath)
  scrivx = File.basename("#{scrivpath}", ".scriv") + ".scrivx"
  scrivxpath = "#{scrivpath}/#{scrivx}"
  # handle cases where the package has been renamed after creation
  unless File.exists?(scrivxpath)
    scrivxpath = %x{ls -1 #{scrivpath}/*.scrivx|head -n 1}.strip
  end
  files = []
  doc = REXML::Document.new File.new(scrivxpath)
  doc.elements.each('ScrivenerProject/Binder/BinderItem') do |ele|
    if ele.attributes['Type'] == "DraftFolder"
      get_children(ele,scrivpath,files,1)
    end
  end

  return files
end

# BEGIN CONVERSION
sw_name = File.basename(path,".scriv")
sw_note = "./#{sw_name}.md"

# Create cache
sw_cache_dir = File.expand_path("./#{sw_name}-cache")
Dir.mkdir(sw_cache_dir) unless File.exists?(sw_cache_dir)
sw_cache = File.expand_path("#{sw_cache_dir}")
# Clear cache
if File.exists?(sw_cache)
  File.delete(*Dir["#{sw_cache}/*"]) if Dir.glob("#{sw_cache}/*").length > 0
  Dir.rmdir(sw_cache)
end
Dir.mkdir(sw_cache)

files = get_rtf_file_array(path)

# loop through files and create copies. store in cache.
cachefiles = []
files.each{ |f|
if f['path']
  cachefile = sw_cache + "/" + File.basename(f['path'],'.rtf') + ".md"
  note = f['path'] ? %x{/usr/bin/textutil -convert txt -stdout "#{f['path']}"} : ''
  leader = ""
  notetext = leader + note + "\n\n"
  File.open(cachefile,"w"){|cf| cf.puts notetext }
  cachefiles.push(cachefile)
end
}

# export text into .md file
File.open(sw_note,"w"){|f| f.puts cachefiles.map{|s| IO.read(s)} }