Hatena Fotolife Image Downloader

hatena-fotolife-downloader.rb

#!/usr/bin/env ruby
# -*- coding: utf-8 -*-

# hatena-fotolife-downloader.rb
#   by takehikom (http://d.hatena.ne.jp/takehikom/)

if RUBY_VERSION < "1.9"
  $KCODE = "u"
end

require "optparse"
require "open-uri"
require "uri"

DEFAULT_USERNAME = "takehikom"
SLEEP_TIME = 2

class HatenaFotolifeDownloader
  def initialize(h = {})
    @opt = h.dup
    @user = @opt[:user] || DEFAULT_USERNAME
    @opt[:dir] ||= @user
    @filename_urllist = "hatena-fotolife-downloader-url-#{@user}.txt"
    @wget_command = "wget"
    @wget_command += " -q" if @opt[:quiet]
  end

  def start
    if !@opt[:downloadonly]
      get_image_url
    end
    if !@opt[:listonly]
      @opt[:simplewget] ? download_by_url_simple : download_by_url
    end
  end

  private

  def puts(message, level = nil)
    return if @opt[:quiet]
    return if level == :debug && !$DEBUG
    Kernel.puts message
  end

  def system(command)
    puts(command, :info)
    Kernel.system command
  end

  def mkdir(dir)
    if test(?d, dir)
      return false
    else
      system("mkdir -p #{dir}")
    end
  end

  def get_image_url
    url_h = {} # dir (may be empty) => array of urls

    puts("[rss] http://f.hatena.ne.jp/#{@user}", :info)
    url_a = get_image_url_by_rss(@user)
    puts(url_a, :debug)
    url_h[""] = url_a

    dir_a = get_subdir
    dir_a.each do |dir|
      puts("[rss] http://f.hatena.ne.jp/#{@user}/#{dir}", :info)
      url_a = get_image_url_by_rss(File.join(@user, dir))
      puts(url_a, :debug)
      url_h[dir] = url_a
    end

    open(@filename_urllist, "w") do |f_out|
      url_h.keys.sort.each do |dir|
        if !(@opt[:simplewget] || dir.empty?)
          f_out.puts "# #{URI.decode(dir)}"
        end
        f_out.puts url_h[dir]
      end
    end

    url_h
  end

  def get_image_url_by_rss(str = @user)
    url_a = []

    # http://mikio.hatenablog.com/entry/2012/11/18/014228
    base_url = "http://f.hatena.ne.jp/#{str}/rss?page="
    page = 1
    while true
      rss_url = base_url + page.to_s
      url_count = 0
      puts("[rss] #{rss_url}", :info)
      open(rss_url) do |file|
        file.each_line do |line|
          next unless line.match(/<hatena:imageurl>/)
          line.sub!(/.*<hatena:imageurl>/, '')
          line.sub!(/<.*/, '')
          line.strip!
          next unless line.match(/^http:\/\//)

          url_count += 1
          url_a << line
        end
      end

      break if url_count < 10
      page += 1
    end

    url_a
  end

  def get_subdir
    dir_a = []

    url = "http://f.hatena.ne.jp/#{@user}"
    open(url) do |file|
      file.each_line do |line|
        next unless line.index('alt="folder"')
        if /href=\"\/#{@user}\/(.*?)\/\"/ =~ line
          dir_a << $1
        end
      end
    end

    dir_a
  end

  def download_by_url
    download_count = 0

    dir = @opt[:dir]
    mkdir(dir)
    dir2 = ""

    open(@filename_urllist) do |f_in|
      f_in.each_line do |line|
        if /^\# (.+)$/ =~ line
          dir2 = $1
          mkdir(File.join(dir, dir2))
        end
        url = line.strip
        next unless url.index("http") == 0

        basename_image = url.split("/")[-1]
        filename_image = File.join(dir, dir2, basename_image)
        puts("[image] #{url} => #{filename_image}", :info)
        if test(?f, filename_image) && !@opt[:force]
          puts("  skipped because the file exists")
          next
        end
        system("#{@wget_command} -O #{filename_image} #{url}")
        download_count += 1
        return if Numeric === @opt[:filemax] && download_count >= @opt[:filemax]
        sleep SLEEP_TIME
      end
    end

    puts("#{download_count} file(s) downloaded", :info)
  end

  def download_by_url_simple
    system("#{@wget_command} -w #{SLEEP_TIME} -x -i #{@filename_urllist}")
  end
end

if __FILE__ == $0
  op = OptionParser.new
  h = {}
  op.on("-u", "--user=VAL", "hatena user name") {|v|
    h[:user] = v
  }
  op.on("-d", "--dir=VAL", "directory for image files") {|v|
    h[:dir] = v
  }
  op.on("-l", "--limit=VAL", "max of files downloaded") {|v|
    h[:filemax] = v.to_i
  }
  op.on("-D", "--download-only", "download only") {
    h[:downloadonly] = true
  }
  op.on("-L", "--list-only", "make list only") {
    h[:listonly] = true
  }
  op.on("-F", "--force-get", "download even if file exists") {
    h[:force] = true
  }
  op.on("-S", "--simple-wget", "download by simple wget") {
    h[:simplewget] = true
  }
  op.on("-q", "--quiet", "quiet") {
    h[:quiet] = true
  }
  op.parse!(ARGV)
  HatenaFotolifeDownloader.new(h).start
end