downloadr.rb

require 'rubygems'
require 'bundler/setup'

require 'mechanize'
require 'typhoeus'
require 'fileutils'
require 'rmagick'
require 'prawn'
require 'fastimage'

agent = Mechanize.new
agent.user_agent = 'Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)'

# customize these 2 variables to download the mangas you want
manga_root_url = "http://www.mangareader.net/206/hikaru-no-go.html"
manga_root_folder = "/vagrant/tmp/mangareader"

page = agent.get manga_root_url

# TODO scans from 1st chapter all the way to the end, must add a param to skip chapters
first_chapter_link = page.search("#listing//a").first['href']
manga_title = page.search("h1").first.text
download_list_backup_file = "/tmp/#{manga_title}.bkp"
downloaded_status_file = "/tmp/#{manga_title}.downloaded"

chapter = agent.get "http://www.mangareader.net#{first_chapter_link}"

# serial link scanning
download_links = []
download_links = Marshal.load(File.read(download_list_backup_file)) if File.exists?(download_list_backup_file)

if download_links.empty?
  while true
    begin
      chapter = chapter.link_with(text: 'Next').click
    rescue
      puts "Finished scanning links"
      break
    end
    begin
      image = chapter.at("#img")
      tokens = image['alt'].match("^(.*?)\s\-\s(.*?)$")
      download_links << [tokens[1], "#{tokens[2]}#{File.extname(URI.parse(image['src']).path)}", image['src']]

      puts download_links.last.join(' - ')
    rescue => e
      puts e
    end
  end
  File.open(download_list_backup_file, 'w') { |f| f.write(Marshal.dump(download_links)) }
end

# parallel downloads
unless File.exists?(downloaded_status_file)
  hydra = Typhoeus::Hydra.new(max_concurrency: 50)
  download_links.each do |file|
    begin
      request = Typhoeus::Request.new file.last
      request.on_complete do |response|
        FileUtils.mkdir_p(File.join(manga_root_folder, file[0]))
        downloaded_filename = File.join(manga_root_folder, file[0], file[1])
        File.open(downloaded_filename, "wb+") { |f| f.write response.body }

        image = Magick::Image.read( downloaded_filename ).first
        resized = image.resize_to_fit(600, 800)
        resized.write( downloaded_filename ) { self.quality = 50 }

        puts "File #{downloaded_filename} downloaded and resized."
        GC.start # to avoid a leak too big
      end
      hydra.queue request
    rescue => e
      puts e
    end
  end
  hydra.run

  FileUtils.touch(downloaded_status_file)
end

# concatenating PDF files (250 pages per volume)
chapter_number = 0
while !download_links.empty?
  chapter_number += 1
  pdf_file = File.join(manga_root_folder, "#{manga_title} #{chapter_number}.pdf")
  list = download_links.slice!(0..250)
  Prawn::Document.generate(pdf_file, page_size: [600, 800]) do |pdf|
    list.map { |file| File.join( manga_root_folder, chapter_name, file[1] ) }.each do |image_file|
      pdf.image image_file, position: :center, vposition: :center
    end
  end
  puts "#{pdf_file} compiled."
  GC.start
end

# cleanup
FileUtils.rm downloaded_status_file

Gemfile

source 'http://rubygems.org'

gem 'mechanize'
gem 'typhoeus'
gem 'rmagick'
gem 'prawn'
gem 'fastimage'