transcribable-harvest.rake

namespace :transcribable do
  desc "Harvest documents to transcribe from DocumentCloud"
  task :harvest_kmiz => :environment do
    require 'rest-client'
    klass = Kernel.const_get(Transcribable.table.classify)
    result = JSON.parse(RestClient.get("http://www.documentcloud.org/api/search?q=contributedto%3Afreethefiles+station%3AKMIZ"))
    result['documents'].each do |doc|
      obj = klass.find_or_initialize_by_url("https://www.documentcloud.org/documents/#{doc['id']}")
      # don't plow over verified docs if rerunning the script
      obj.verified = false if obj.new_record?
      obj.save
      puts "== added #{obj.url}"
    end
  end
end