flickrdl.rb

require 'flickraw'
require 'mini_magick'
require 'parallel'

def get_photo_urls_from_photoset(flickr, photoset_id)
  total_pages = nil
  current_page = 1
  urls = []
  opts = {
    photoset_id: photoset_id,
    extras: 'url_o',
    privacy_filter: 5
  }

  while total_pages.nil? || total_pages >= current_page
    puts "Fetching set #{photoset_id} page #{current_page}"
    opts[:page] = current_page
    response = flickr.photosets.getPhotos(opts)
    response['photo'].each { |photo| urls << photo['url_o'] }
    total_pages ||= response['pages'].to_i
    current_page = response['page'].to_i + 1
  end

  urls
end

def download_photo(url, folder)
  puts "Downloading #{url} ..."
  photo = MiniMagick::Image.open url
  filename = url.split('/').last
  photo.write File.join(folder, filename)
  {file: filename, url: url, status: 'OK'}
rescue => e
  {file: filename, url: url, status: 'Error', error: e.inspect}
end

def download_photoset(set, base_folder)
  folder = File.join base_folder, set['title']
  FileUtils.mkdir_p folder
  set['results'] = Parallel.map(set['urls'], in_threads: 3) do |url|
    download_photo(url, folder)
  end
end

# see https://hanklords.github.io/flickraw/ to fill in these tokens
FlickRaw.api_key = '...'
FlickRaw.shared_secret = '...'
flickr = FlickRaw::Flickr.new
flickr.access_token = '...'
flickr.access_secret = '...'

photosets = flickr.photosets.getList
parsed_sets = Parallel.map(photosets, in_threads: 3) do |set|
  set = set.to_hash
  set['urls'] = get_photo_urls_from_photoset(flickr, set['id']).compact
  puts "Set #{set['title']} has #{set['urls'].count} photos"
  set
end

base_folder = File.join(Dir.home, 'flickr')
parsed_sets.each do |set|
  download_photoset(set, base_folder)
end