Given a Google Drive folder URL, attempts to extract a list of Google Doc names and IDs.

scrape_google_drive_folder_contents.rb

require 'yaml'
require 'faraday'
require 'nokogiri'

# REVIEW: May not work for directories with larger numbers of files.
# REVIEW: Formatting will likely be different for non-Google-Docs files.
# REVIEW: This is stupid.
def scrape_folder_contents(gdrive_folder_url, output_fname=nil)
  html = Faraday.get(gdrive_folder_url).body
  doc  = Nokogiri::HTML(html)

  js = nil
  doc.css('script').each do |script|
    _js = script.inner_html.strip
    if _js.include?("window['_DRIVE_ivd']")
      js = _js
      break
    end
  end
  return nil unless js

  # Clean off cruft to get raw array describing folder contents.
  js.sub!("window['_DRIVE_ivd'] = '",'')
  js.sub!("';if (window['_DRIVE_ivdc']) {window['_DRIVE_ivdc']();}", '')

  # Clean up array formatting so Ruby can evaluate it.
  js = convert_hex_chars(js)
  js.gsub!('\n',' ')
  js.gsub!('\/', '/')

  # Remove all null values
  # REVIEW: May need to skip this when trying to parse non-Google-Docs files
  js.gsub!(',null','')

  # Evaluating the data contained in this Javascript gives an array of 3 
  # items, the first of which will be an array of subarrays, each describing 
  # one document within the GDrive folder.
  data    = eval(js)
  doc_arr = data.first

  folder_contents = doc_arr.map do |subarr|
    id    = subarr[0] # The 1st element in each doc array is the doc's ID
    title = subarr[2] # The 3rd is its title

    {
      title: title, 
      id:    id, 
    }
  end

  File.write(output_fname, folder_contents.to_yaml) if output_fname
  return folder_contents
end

def convert_hex_chars(str)
  chars = str.scan(/\\x(..)/).flatten.uniq
  chars.each do |char|
    str.gsub!("\\x#{char}", hex_to_unicode(char))
  end
  return str
end

def hex_to_unicode(str='\x5b')
  str.sub('\x','').hex.chr('UTF-8')
end