opendmm 0.4.3 → 1.0.0

data/lib/opendmm/makers_broken/high_scene.rb

@@ -1,32 +0,0 @@
-base_uri 'highscene.jp'
-
-register_product(
-  /^(HIGH)-?(\d{3})$/i,
-  '/works/-/detail/=/cid=#{$1.downcase}#{$2}/',
-)
-
-private
-
-def self.parse_product_html(html)
-  specs = Utils.hash_by_split html.css('#works-details > dl > dd > p').map(&:text)
-  {
-    actresses:       [ html.css('#content > h2').text[/(?<=>>).*/] ],
-  # brand:           String
-  # categories:      Array
-    cover_image:     html.at_css('#actimg > li > a')['href'],
-    description:     html.css('#works-details > p.works-txt').text,
-  # directors:       Array
-    genres:          specs['ジャンル'].split('/'),
-    label:           specs['レーベル'],
-    maker:           'High Scene',
-    movie_length:    specs['収録時間'],
-    release_date:    specs['配信日'],
-    sample_images:   html.css('#actimg > li > a')[1..-1].map { |a| a['href'] },
-  # scenes:          Array
-  # series:          String
-    subtitle:        html.css('#works-details > dl > dt > span').text,
-  # theme:           String
-    thumbnail_image: html.at_css('#actimg > li > a > img')['src'],
-    title:           html.css('#content > h2').text[/(?<=>>).*/],
-  }
-end

data/lib/opendmm/makers_broken/to_satsu.rb

@@ -1,24 +0,0 @@
-base_uri 'to-satsu.com'
-
-register_product(
-  /^(CLUB)-?(\d{3})$/i,
-  '/works/#{$1.downcase}/#{$1.downcase}#{$2}/#{$1.downcase}#{$2}.html',
-)
-
-private
-
-def self.parse_product_html(html)
-  specs = Utils.hash_from_dl(html.css('#main > div > div.main-detail > div.ttl-data > dl'))
-  {
-    cover_image:     html.at_css('#main > div > div.main-detail > div.ttl-pac > a')['href'],
-    description:     html.xpath('//*[@id="main"]/div/div[@class="main-detail"]/div[@class="ttl-comment"]/div[not(@class)]').text,
-    genres:          specs['ジャンル：'].text,
-    maker:           '変態紳士倶楽部',
-    movie_length:    specs['収録時間：'].text,
-    release_date:    specs['発売日：'].text,
-    sample_images:   html.css('#main > div > div.main-detail > div.ttl-sample > img').map { |img| img['src'] },
-    series:          specs['シリーズ：'].text,
-    thumbnail_image: html.at_css('#main > div > div.main-detail > div.ttl-pac > a > img')['src'],
-    title:           html.css('#main > div > div.capt2').text,
-  }
-end

data/lib/opendmm/search_engine.rb

@@ -1,20 +0,0 @@
-require 'opendmm/utils'
-
-Dir[File.dirname(__FILE__) + '/search_engines/*.rb'].each do |file|
-  module_name = File.basename(file, '.rb').camelize
-  eval <<-ENGINE
-
-module OpenDMM
-  module SearchEngine
-    module #{module_name}
-      include Site
-      #{File.read(file)}
-      def self.search(name)
-        self.product(name)
-      end
-    end
-  end
-end
-
-ENGINE
-end

data/lib/opendmm/search_engines/av_entertainments.rb

@@ -1,46 +0,0 @@
-base_uri 'www.aventertainments.com'
-headers({
-  "User-Agent" => "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.1916.153 Safari/537.36"
-})
-
-def self.search_url(name)
-  name = name.split(/(?<=[a-z])(?=\d)|[-_\s]/).join('-')
-  "/search_Products.aspx?keyword=#{CGI::escape(name)}"
-end
-
-def self.product_url(name)
-  search_page = get_with_retry search_url(name)
-  return nil unless search_page
-  search_html = Utils.html_in_utf8 search_page
-  first_result = search_html.at_css('div.main-unit2 > table a')
-  first_result['href'] if first_result
-end
-
-private
-
-def self.parse_product_html(html)
-  specs = Utils.hash_by_split(html.xpath('//*[@id="TabbedPanels1"]/div/div[2]/div[1]//ol').text.split('|')).merge(
-          Utils.hash_by_split(html.xpath('//*[@id="titlebox"]/ul[3]/li').map(&:text)))
-  {
-    actresses:       specs['主演女優'].split(','),
-  # brand:           String
-    categories:      html.xpath('//*[@id="TabbedPanels1"]/div/div[2]/div[2]//ol').map(&:text),
-    code:            specs['商品番号'],
-    cover_image:     html.at_css('#titlebox > div.list-cover > img')['src'].gsub('jacket_images', 'bigcover'),
-    description:     html.css('#titlebox > div.border > p').text,
-  # directors:       Array
-  # genres:          Array
-  # label:           String
-    maker:           specs['スタジオ'],
-    movie_length:    specs['収録時間'],
-  # page:            String
-    release_date:    Date.strptime(specs['発売日'], '%m/%d/%Y'),
-  # sample_images:   Array
-  # scenes:          Array
-    series:          specs['シリーズ'],
-  # subtitle:        String
-  # theme:           String
-    thumbnail_image: html.at_css('#titlebox > div.list-cover > img')['src'],
-    title:           html.css('#mini-tabet > h2').text,
-  }
-end

data/lib/opendmm/search_engines/dmm.rb

@@ -1,50 +0,0 @@
-base_uri 'www.dmm.co.jp'
-
-def self.search_url(name, digit_len)
-  name = name.split(/(?<=[a-z])(?=\d)|[-_\s]/).map do |token|
-    token =~ /^\d+$/ ? token.rjust(digit_len, '0') : token
-  end.join
-  "/search/=/searchstr=#{CGI::escape(name)}"
-end
-
-def self.product_url(name)
-  5.downto(3) do |len|
-    search_page = get_with_retry search_url(name, len)
-    next unless search_page
-    search_html = Utils.html_in_utf8 search_page
-    first_result = search_html.at_css('#list > li > div > p.tmb > a')
-    return first_result['href'] if first_result
-  end
-  nil
-end
-
-def self.product_extra_info(name, url, page, html)
-  dmm_id = page.request.last_uri.to_s.split('/').last.split('=').last
-  if dmm_id =~ /\d*([a-z]+)0*(\d+)/i
-    dmm_id = "#{$1.upcase}-#{$2.rjust(3, '0')}"
-  end
-  {
-    code: dmm_id,
-    maker: 'DMM',
-  }
-end
-
-private
-
-def self.parse_product_html(html)
-  specs = Utils.hash_by_split(html.css('//div[@class="page-detail"]/table/tr/td[1]/table/tr').map(&:text))
-  {
-    actresses:       ( specs['出演者'] || specs['名前'] ).split(/[\s\/]/),
-    cover_image:     ( html.at_css('#sample-video > a')['href'] ||
-                       html.at_css('#sample-video > img')['src'] ),
-    directors:       specs['監督'].split,
-    genres:          specs['ジャンル'].split,
-    label:           specs['レーベル'],
-    maker:           specs['メーカー'],
-    movie_length:    specs['収録時間'],
-    release_date:    (specs['商品発売日'] || specs['配信開始日']),
-    series:          specs['シリーズ'],
-    thumbnail_image: html.at_css('#sample-video img')['src'],
-    title:           html.css('#title').text,
-  }
-end

data/lib/opendmm/search_engines/jav_library.rb

@@ -1,39 +0,0 @@
-base_uri 'www.javlibrary.com'
-
-def self.search_url(name)
-  "/ja/vl_searchbyid.php?keyword=#{CGI::escape(name)}"
-end
-
-def self.product_url(name)
-  search_page = get_with_retry search_url(name)
-  return nil unless search_page
-  if search_page.code == 302
-    url = search_page.headers['location']
-  else
-    search_html = Utils.html_in_utf8 search_page
-    links = search_html.css('#rightcolumn > div.videothumblist > div.videos > div.video > a')
-    result = links.detect do |link|
-      link.css('div.id').text == name
-    end
-    result ||= links.first
-    return nil unless result
-    url = result['href']
-  end
-  URI.join(search_page.request.last_uri, url).to_s
-end
-
-def self.parse_product_html(html)
-  {
-    actresses:       html.css('#video_cast .text span.cast span.star').map(&:text),
-    code:            html.css('#video_id .text').text,
-    cover_image:     html.at_css('#video_jacket > img')['src'],
-    directors:       html.css('#video_director .text span.director').map(&:text),
-    genres:          html.css('#video_genres .text span.genre').map(&:text),
-    label:           html.css('#video_label .text').text,
-    maker:           html.css('#video_maker .text').text,
-    movie_length:    html.css('#video_length .text').text + ' minutes',
-    release_date:    html.css('#video_date .text').text,
-    thumbnail_image: html.at_css('#video_jacket_img')['src'],
-    title:           html.css('#video_title > h3').text.remove(html.css('#video_id .text').text),
-  }
-end

data/lib/opendmm/search_engines/mgstage.rb

@@ -1,41 +0,0 @@
-require 'pp'
-
-base_uri 'www.mgstage.com'
-cookies(adc: 1, coc: 1)
-headers({
-  "User-Agent" => "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/35.0.1916.153 Safari/537.36"
-})
-
-def self.product_url(name)
-  return "/ppv/video/#{name}/"
-end
-
-private
-
-def self.parse_product_html(html)
-  specs = Utils.hash_from_dl(html.css('div.content_detail_layout_01 > dl.spec_layout'))
-  {
-    actresses:       specs['出演：'].css('a').map(&:text),
-  # actress_types:   Array
-  # boobs:           String
-  # brand:           String
-  # categories:      Array
-    code:            specs['品番：'].text,
-    cover_image:     html.at_css('a.enlarge_image')['href'],
-    description:     html.css('#introduction_text > p.introduction').text,
-  # directors:       Array
-    genres:          specs['ジャンル：'].css('a').map(&:text),
-  # label:           String
-    maker:           specs['メーカー：'].text,
-    movie_length:    specs['収録時間：'].text,
-  # page:            String
-    release_date:    specs['配信開始日：'].text,
-    sample_images:   html.css('a.sample_imageN').map { |a| a['href'] },
-  # scenes:          Array
-    series:          specs['シリーズ名：'].text,
-  # subtitle:        String
-  # theme:           String
-    thumbnail_image: html.at_css('a.enlarge_image > img')['src'],
-    title:           html.css('.title_detail_layout h1').text,
-  }
-end

data/lib/opendmm/site.rb

@@ -1,80 +0,0 @@
-require 'active_support/core_ext/module/attribute_accessors'
-require 'httparty'
-require 'opendmm/utils'
-
-module OpenDMM
-  module Site
-
-    def self.included(klass)
-      klass.mattr_accessor :url_generators
-      klass.mattr_accessor :code_generators
-      klass.url_generators = Hash.new
-      klass.code_generators = Hash.new
-
-      klass.include HTTParty
-      klass.follow_redirects false
-      klass.default_timeout 10
-
-      klass.extend ClassMethods
-    end
-
-    module ClassMethods
-      def get_with_retry(url, limit = 5)
-        return nil unless url
-        get url
-      rescue Errno::ETIMEDOUT, Errno::ECONNREFUSED => e
-        tries ||= 0
-        tries++
-        tries <= limit ? retry : raise
-      end
-
-      def register_product(matcher, url_generator, code_generator = '#{$1.upcase}-#{$2}')
-        url_generators[matcher] = '"' + url_generator + '"'
-        code_generators[matcher] = '"' + code_generator + '"'
-      end
-
-      def product_url(name)
-        url_generators.each do |matcher, url_generator|
-          if name =~ matcher
-            return eval url_generator
-          end
-        end
-        nil
-      end
-
-      def product_code(name)
-        code_generators.each do |matcher, code_generator|
-          if name =~ matcher
-            return eval code_generator
-          end
-        end
-        nil
-      end
-
-      def product_extra_info(name, url, page, html)
-        Hash.new
-      end
-
-      def product(name)
-        return nil unless name =~ /^[[:ascii:]]+$/
-        url = product_url name
-        page = get_with_retry url
-        return nil unless page && page.code == 200
-        html = Utils.html_in_utf8 page
-        html.css('script').remove
-        details = parse_product_html html
-        details[:code] ||= product_code name
-        details[:page] ||= page.request.last_uri.to_s
-        extra_info = product_extra_info(name, url, page, html)
-        extra_info.each do |k, v|
-          details[k] ||= v
-        end
-        details = Utils.finalize_details_hash details
-        %i(code cover_image maker page thumbnail_image title).each do |key|
-          return unless details[key]
-        end
-        details
-      end
-    end
-  end
-end

data/lib/opendmm/utils.rb

@@ -1,126 +0,0 @@
-require 'active_support/core_ext/array/access'
-require 'active_support/core_ext/array/grouping'
-require 'active_support/core_ext/module/aliasing'
-require 'active_support/core_ext/numeric/time'
-require 'active_support/core_ext/string/filters'
-require 'active_support/core_ext/string/starts_ends_with'
-require 'chronic_duration'
-require 'nokogiri'
-
-class << Date
-  def parse_with_chinese_support(str)
-    case str
-    when /(\d{4})年(\d{1,2})月(\d{1,2})日/
-      return new($1.to_i, $2.to_i, $3.to_i)
-    when /(\d{2})年(\d{1,2})月(\d{1,2})日/
-      year = 2000 + $1.to_i
-      year -= 100 if year > Date.today.year
-      return new(year, $2.to_i, $3.to_i)
-    else
-      return parse_without_chinese_support(str)
-    end
-  end
-  alias_method_chain :parse, :chinese_support
-end
-
-class << ChronicDuration
-  def parse_with_chinese_support(str)
-    case str
-    when /(\d+)\s*分\s*(\d+)\s*秒/
-      return $1.to_i.minutes + $2.to_i.seconds
-    when /(\d+)(\s*)分/
-      return $1.to_i.minutes
-    else
-      return parse_without_chinese_support(str)
-    end
-  end
-  alias_method_chain :parse, :chinese_support
-end
-
-class NilClass
-  def method_missing(meth, *args, &block)
-    nil
-  end
-end
-
-module OpenDMM
-  module Utils
-    def self.hash_from_dl(dl)
-      dts = dl.css('dt').map(&:text).map(&:squish)
-      dds = dl.children.select do |node|
-        node.name == 'dt' || node.name == 'dd'
-      end.split do |node|
-        node.name == 'dt'
-      end[1..-1].map do |nodes|
-        node_set = Nokogiri::XML::NodeSet.new(dl.document)
-        nodes.each do |node|
-          node_set << node
-        end
-        node_set
-      end
-      Hash[dts.zip(dds)]
-    end
-
-    def self.hash_by_split(array, delimiter = /[：:]/)
-      array.map do |str|
-        slices = str.split(delimiter, 2).map(&:squish)
-      end.select do |pair|
-        pair.size == 2
-      end.to_h
-    end
-
-    def self.html_in_utf8(content)
-      encoding = Nokogiri.HTML(content).encoding
-      content = content.encode('UTF-8', encoding, invalid: :replace, undef: :replace, replace: '')
-      Nokogiri.HTML content
-    end
-
-    def self.finalize_details_hash(details)
-      details = details.squish_hard
-      details[:actresses] = details[:actresses].map do |actress|
-        actress.remove(/[\s\.]/)
-      end.sort if details[:actresses]
-      details[:cover_image] = join_url_if_relative(details[:page], details[:cover_image])
-      details[:thumbnail_image] = join_url_if_relative(details[:page], details[:thumbnail_image])
-      details[:sample_images] = details[:sample_images].map do |uri|
-        join_url_if_relative(details[:page], uri)
-      end if details[:sample_images]
-      details[:movie_length] = ChronicDuration.parse(details[:movie_length]).to_i if details[:movie_length]
-      details[:release_date] = Date.parse(details[:release_date]) if details[:release_date].instance_of? String
-      details.sort.to_h
-    end
-
-    def self.join_url_if_relative(page_url, image_url)
-      return nil unless image_url
-      image_url.start_with?('http') ? image_url : URI.join(page_url, image_url).to_s
-    end
-  end
-end
-
-class Object
-  def squish_hard
-    presence
-  end
-end
-
-class String
-  def squish_hard
-    squish.gsub(/^[\s-]*$/, '').presence
-  end
-end
-
-class Hash
-  def squish_hard
-    map do |k, v|
-      [k, v.squish_hard]
-    end.select do |k_v|
-      k_v.second.present?
-    end.to_h.presence
-  end
-end
-
-class Array
-  def squish_hard
-    map(&:squish_hard).select(&:present?).presence
-  end
-end