Download all the photos you've liked on instagram

ig_downloader.rb

require 'restclient'
require 'json'
require 'digest'

class IGDownloader

  def initialize(output_path)
    @base_output_path = output_path
  end

  def run(media_url)
    image_urls = []
    while media_url != nil
      json = json_for(media_url)
      media_url = json["pagination"]["next_url"]
      image_urls.concat(find_urls(json["data"]))
      # don't slam instagram's api
      sleep 0.2
    end
    download_images(image_urls)
  end

  def download_images(urls)
    urls.each do |url|
      digest = digest(url)
      image_output_path = image_output_path(digest)
      image = RestClient.get(url) unless File.exists? image_output_path
      save_image(image, image_output_path) if image
      save_url(url, url_output_path(digest)) if image
    end
  end

  def output_directory(digest)
    path = "#{@base_output_path}/#{digest}"
    dir = Dir.mkdir(path) unless Dir.exists? path
    path
  end

  def digest(url)
    Digest::MD5.hexdigest(url)
  end

  def image_output_path(digest)
    path = output_directory(digest)
    full_path = "#{path}/#{digest}.jpg"
  end

  def url_output_path(digest)
    path = output_directory(digest)
    full_path = "#{path}/url.txt"
  end

  def save_image(image, path)
    save_item(image, path)
  end

  def save_url(url, path)
    save_item("#{url}\n", path)
  end

  def save_item(item, path)
    File.open(path, 'w') {|f| f.write(item) }
  end

  def json_for(url)
    begin
      response = RestClient.get url
      json = JSON.parse(response)
    rescue Exception
      puts "Rescuing #{url}"
      # yeah, infinite loop, be careful
      json_for(url)
    end
  end

  def find_urls(items)
    items.map do |item|
      url = item["images"]["standard_resolution"]["url"]
    end
  end
end

# See http://instagram.com/developer/authentication/
# for authentication details
token = "some valid access token"

media_url = "https://api.instagram.com/v1/users/self/media/liked/?access_token=#{token}"
output_path = "/some/path/that/already/exists"
IGDownloader.new(output_path).run(media_url)

Is this working?

Not likely. This was written in 2012 and the instagram API has likely changed quite a bit since then.