opendmm 0.0.2

data/lib/opendmm/makers/sod.rb

@@ -0,0 +1,44 @@
+module OpenDMM
+  module Maker
+    module Sod
+      include Maker
+
+      module Site
+        include HTTParty
+        base_uri 'ec.sod.co.jp'
+
+        def self.item(name)
+          case name
+          when /^(NAGE|SDDE|SDMT|SDMU|SDNM|STAR)-?(\d{3})$/i
+            get("/detail/index/-_-/iid/#{$1.upcase}-#{$2}")
+          end
+        end
+      end
+
+      module Parser
+        def self.parse(content)
+          page_uri = content.request.last_uri
+          html = Nokogiri::HTML(content)
+          specs = Utils.hash_by_split(html.xpath('//*[@id="main"]/tr/td[2]/table/tr[1]/td[1]/table/tr[1]/td/table/tr/td[2]/table/tr').map(&:text))
+          return {
+            actresses:       specs['出演'].split,
+            code:            specs['品番'],
+            cover_image:     html.at_xpath('//*[@id="main"]/tr/td[2]/table/tr[1]/td[1]/table/tr[1]/td/table/tr/td[1]/a[1]')['href'],
+            description:     html.css('div.detail-datacomment').text,
+            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'],
+            page:            page_uri.to_s,
+            release_date:    specs['発売日'],
+            sample_images:   html.css('div.detail-thumb-sample-box > a').map { |a| a['href'] },
+            series:          specs['シリーズ'],
+            thumbnail_image: html.at_xpath('//*[@id="main"]/tr/td[2]/table/tr[1]/td[1]/table/tr[1]/td/table/tr/td[1]/a[1]/img')['src'],
+            title:           html.css('div.title-base > div.title-base-dvd1 > div.title-base-dvd2 > h1').text,
+          }
+        end
+      end
+    end
+  end
+end

data/lib/opendmm/makers/tameikegoro.rb

@@ -0,0 +1,42 @@
+module OpenDMM
+  module Maker
+    module Tameikegoro
+      include Maker
+
+      module Site
+        include HTTParty
+        base_uri 'tameikegoro.jp'
+
+        def self.item(name)
+          case name
+          when /^(MDYD)-?(\d{3})$/i
+            get("/works/-/detail/=/cid=#{$1.downcase}#{$2}")
+          end
+        end
+      end
+
+      module Parser
+        def self.parse(content)
+          page_uri = content.request.last_uri
+          html = Nokogiri::HTML(content)
+          specs = Utils.hash_by_split(html.css('#work-detail > h4, p').map(&:text))
+          return {
+            actresses:       specs['出演女優'].split,
+            code:            specs['品番'],
+            cover_image:     html.at_css('#work-pake')['src'].gsub(/pm.jpg$/, 'pl.jpg'),
+            description:     html.css('#work-detail > h5').text,
+            directors:       specs['監督'].split,
+            genres:          specs['ジャンル'].split,
+            label:           specs['レーベル'],
+            page:            page_uri.to_s,
+            release_date:    specs['発売日'].remove('DVD/'),
+            sample_images:   html.css('#sample-pic > a > img').map { |img| img['src'].gsub(/js(?=-\d+\.jpg)/, 'jp') },
+            series:          specs['シリーズ'],
+            thumbnail_image: html.at_css('#work-pake')['src'],
+            title:           html.css('#work-detail > h3').text,
+          }
+        end
+      end
+    end
+  end
+end

data/lib/opendmm/makers/to_satsu.rb

@@ -0,0 +1,40 @@
+module OpenDMM
+  module Maker
+    module ToSatsu
+      include Maker
+
+      module Site
+        include HTTParty
+        base_uri 'to-satsu.com'
+
+        def self.item(name)
+          case name
+          when /^(CLUB)-?(\d{3})$/i
+            get("/works/#{$1.downcase}/#{$1.downcase}#{$2}/#{$1.downcase}#{$2}.html")
+          end
+        end
+      end
+
+      module Parser
+        def self.parse(content)
+          page_uri = content.request.last_uri
+          html = Nokogiri::HTML(content)
+          specs = Utils.hash_from_dl(html.css('#main > div > div.main-detail > div.ttl-data > dl'))
+          return {
+            code:            specs['品番：'].text,
+            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,
+            movie_length:    specs['収録時間：'].text,
+            page:            page_uri.to_s,
+            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
+      end
+    end
+  end
+end

data/lib/opendmm/makers/waap.rb

@@ -0,0 +1,46 @@
+module OpenDMM
+  module Maker
+    module Waap
+      include Maker
+
+      module Site
+        include HTTParty
+        base_uri 'www.waap.co.jp'
+
+        def self.item(name)
+          name =~ /(\w+)-?(\d+)/
+          case name
+          when /^(AIR|CWM|ECB|WSS)-?(\d{3})$/i
+            get("/work/item.php?itemcode=#{$1.upcase}#{$2}")
+          end
+        end
+      end
+
+      module Parser
+        def self.parse(content)
+          page_uri = content.request.last_uri
+          html = Nokogiri::HTML(content)
+          specs = Utils.hash_by_split(html.css('ul.title_shosai li.wkact_ser_maker02').map(&:text))
+          return {
+            actresses:       specs['出演者'].split,
+            brand:           specs['ブランド'],
+            code:            specs['品番'],
+            cover_image:     html.at_css('ul#title_img_all li.title_img a')['href'],
+            description:     html.css('div#title_cmt_all').text,
+            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'],
+            page:            page_uri.to_s,
+            release_date:    specs['発売日'],
+            sample_images:   html.css('ul.samplepicture_list li a').map { |a| a['href'] },
+            series:          specs['シリーズ'],
+            thumbnail_image: html.at_css('#title_img_all > li.title_img > a > img')['src'],
+            title:           html.css('ul#pan_list li').last.text,
+          }
+        end
+      end
+    end
+  end
+end

data/lib/opendmm/search_engines/jav_library.rb

@@ -0,0 +1,58 @@
+require 'opendmm/utils'
+
+module OpenDMM
+  module SearchEngine
+    module JavLibrary
+      module Site
+        include HTTParty
+        base_uri 'www.javlibrary.com'
+        follow_redirects false
+
+        def self.search(name)
+          get("/ja/vl_searchbyid.php?keyword=#{name}")
+        end
+
+        def self.item(id)
+          get("/ja/?v=#{id}")
+        end
+      end
+
+      module Parser
+        def self.parse_search_result(content)
+          html = Nokogiri::HTML(content)
+          first_result = html.css('#rightcolumn > div.videothumblist > div.videos > div.video > a').first
+          first_result['href'].remove('./?v=') if first_result
+        end
+
+        def self.parse_item(content)
+          page_uri = content.request.last_uri
+          html = Nokogiri::HTML(content)
+          return {
+            actresses:       html.css('#video_cast .text span.cast').map(&:text),
+            code:            html.css('#video_id .text').text,
+            cover_image:     html.at_css('#video_jacket > a')['href'],
+            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',
+            page:            page_uri.to_s,
+            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
+      end
+
+      def self.search(name)
+        search_result = Site.search(name)
+        if search_result.code == 302
+          jav_id = search_result.headers['location'].remove('./?v=')
+        else
+          jav_id = Parser.parse_search_result(search_result)
+        end
+        Parser.parse_item(Site.item(jav_id)) if jav_id
+      end
+    end
+  end
+end

data/lib/opendmm/utils.rb

@@ -0,0 +1,119 @@
+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/starts_ends_with'
+require 'chronic_duration'
+
+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)
+    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/
+      return $1.to_i.minutes
+    else
+      return parse_without_chinese_support(str)
+    end
+  end
+  alias_method_chain :parse, :chinese_support
+end
+
+class NilClass
+  def text
+    ''
+  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.force_utf8(content)
+      # This is to get rid of the annoying error message:
+      #   'encoding error : input conversion failed due to input error'
+      $stderr.reopen('/dev/null', 'w')
+      encoding = Nokogiri::HTML(content).encoding
+      $stderr = STDERR
+      content = content.encode('UTF-8', encoding, invalid: :replace, undef: :replace, replace: '')
+    end
+
+    def self.cleanup(details)
+      return nil unless details
+      details = self.squish(details)
+      if details[:movie_length].instance_of? String
+        details[:movie_length] = ChronicDuration.parse(details[:movie_length])
+      end
+      if details[:page]
+        if details[:cover_image] && !details[:cover_image].start_with?('http')
+          details[:cover_image] = URI.join(details[:page], details[:cover_image]).to_s
+        end
+        if details[:sample_images]
+          details[:sample_images] = details[:sample_images].map do |uri|
+            uri.start_with?('http') ? uri : URI.join(details[:page], uri).to_s
+          end
+        end
+        if details[:thumbnail_image] && !details[:thumbnail_image].start_with?('http')
+          details[:thumbnail_image] = URI.join(details[:page], details[:thumbnail_image]).to_s
+        end
+      end
+      if details[:release_date].instance_of? String
+        details[:release_date] = Date.parse(details[:release_date])
+      end
+      details
+    end
+
+    def self.squish(obj)
+      case obj
+      when String
+        obj.squish!
+        obj.gsub!(/^[\s-]*$/, '')
+      when Hash
+        obj = obj.map do |k, v|
+          [k, squish(v)]
+        end.select do |kv|
+          kv.second.present?
+        end.to_h.symbolize_keys
+      when Array
+        obj = obj.map do |v|
+          squish(v)
+        end.select do |v|
+          v.present?
+        end
+      end
+      return obj.present? ? obj : nil
+    end
+  end
+end

data/lib/opendmm/version.rb

@@ -0,0 +1,3 @@
+module OpenDMM
+  VERSION = '0.0.2'
+end

data/opendmm.gemspec

@@ -0,0 +1,25 @@
+lib = File.expand_path('../lib', __FILE__)
+$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
+require 'opendmm/version'
+
+Gem::Specification.new do |spec|
+  spec.name          = 'opendmm'
+  spec.version       = OpenDMM::VERSION
+  spec.authors       = ['Jun Zhou']
+  spec.email         = ['pinepara@gmail.com']
+  spec.summary       = 'OpenDMM: an open-source Japanese AV search engine'
+  spec.license       = 'MIT'
+
+  spec.files         = `git ls-files`.lines.map(&:chomp)
+  spec.executables   = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
+  spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
+  spec.require_paths = ['lib']
+
+  spec.add_development_dependency 'bundler'
+  spec.add_development_dependency 'hashdiff'
+  spec.add_development_dependency 'rake'
+  spec.add_dependency 'activesupport'
+  spec.add_dependency 'chronic_duration'
+  spec.add_dependency 'httparty'
+  spec.add_dependency 'nokogiri'
+end

data/test/fixtures/ABP-013.json

@@ -0,0 +1,33 @@
+{
+  "actresses": [
+    "水咲(滝澤)ローラ"
+  ],
+  "code": "ABP-013",
+  "cover_image": "http://image.prestige-av.com/images/prestige/abp/013/pb_e_abp-013.jpg",
+  "description": "通販限定商品！生写真7枚付きの特典商品です。本数限定となりますのでお早めにお求め下さい！！プレステージ専属女優 『水咲 ローラ』 が熱気を漂わせながら超濃厚なセックスを展開！男と相互奉仕の応酬をして、貪るような肉食っぷりを披露！焦らされたカラダは手マンされ、カメラが水没するほど何度も大量潮吹き！パイパンマ〇コにパワフルな腰使いでピストンされ、本気汁まみれで悶えっぱなし！！",
+  "genres": [
+    "通販限定",
+    "潮吹き",
+    "女優",
+    "パイパン"
+  ],
+  "label": "ABSOLUTELY P…",
+  "maker": "プレステージ",
+  "movie_length": 7200,
+  "page": "http://www.prestige-av.com/item/prestige/ABP-013",
+  "release_date": "2013-07-02",
+  "sample_images": [
+    "http://image.prestige-av.com/images/prestige/abp/013/cap_e_0_abp-013.jpg",
+    "http://image.prestige-av.com/images/prestige/abp/013/cap_e_1_abp-013.jpg",
+    "http://image.prestige-av.com/images/prestige/abp/013/cap_e_2_abp-013.jpg",
+    "http://image.prestige-av.com/images/prestige/abp/013/cap_e_3_abp-013.jpg",
+    "http://image.prestige-av.com/images/prestige/abp/013/cap_e_4_abp-013.jpg",
+    "http://image.prestige-av.com/images/prestige/abp/013/cap_e_5_abp-013.jpg",
+    "http://image.prestige-av.com/images/prestige/abp/013/cap_e_6_abp-013.jpg",
+    "http://image.prestige-av.com/images/prestige/abp/013/cap_e_7_abp-013.jpg",
+    "http://image.prestige-av.com/images/prestige/abp/013/cap_e_8_abp-013.jpg"
+  ],
+  "series": "天然成分由来",
+  "thumbnail_image": "http://image.prestige-av.com/images/prestige/abp/013/pf_p_abp-013.jpg",
+  "title": "天然成分由来 水咲ローラ汁120%＋生写真7枚付き"
+}

data/test/fixtures/ABS-014.json

@@ -0,0 +1,31 @@
+{
+  "actresses": [
+    "絵色 千佳"
+  ],
+  "code": "ABS-014",
+  "cover_image": "http://image.prestige-av.com/images/prestige/abs/014/pb_e_abs-014.jpg",
+  "description": "お嬢様系の超美少女が登場。イラマチオ気味にフェラさせられたチ○コをマ○コに突き挿され快感にウットリ。中出しされピンク色のマ○コから精子を垂らしピクピク痙攣も見逃せない。男の大きな足をゆっくりと舐め出し、ネットリとした舌使いで激エロ奉仕。四つん這いやチングリ返しになった男の尻を目を閉じて舐めるシーンもエロ過ぎ！学生服姿で肛門丸見えにさせられたり潮吹きもエロい！３Ｐになりバックから美尻をプルンプルン波打たせながら打ち込まれ本番になり、角度を変えてピストンされ発射は口へ。連続でハメられ２度目も口に掛けられお掃除",
+  "genres": [
+    "潮吹き",
+    "女優",
+    "中出し"
+  ],
+  "label": "ABSOLUTE",
+  "maker": "プレステージ",
+  "movie_length": 7200,
+  "page": "http://www.prestige-av.com/item/prestige/ABS-014",
+  "release_date": "2011-02-08",
+  "sample_images": [
+    "http://image.prestige-av.com/images/prestige/abs/014/cap_e_0_abs-014.jpg",
+    "http://image.prestige-av.com/images/prestige/abs/014/cap_e_1_abs-014.jpg",
+    "http://image.prestige-av.com/images/prestige/abs/014/cap_e_2_abs-014.jpg",
+    "http://image.prestige-av.com/images/prestige/abs/014/cap_e_3_abs-014.jpg",
+    "http://image.prestige-av.com/images/prestige/abs/014/cap_e_4_abs-014.jpg",
+    "http://image.prestige-av.com/images/prestige/abs/014/cap_e_5_abs-014.jpg",
+    "http://image.prestige-av.com/images/prestige/abs/014/cap_e_6_abs-014.jpg",
+    "http://image.prestige-av.com/images/prestige/abs/014/cap_e_7_abs-014.jpg"
+  ],
+  "series": "貸し切り、純潔サロン",
+  "thumbnail_image": "http://image.prestige-av.com/images/prestige/abs/014/pf_p_abs-014.jpg",
+  "title": "貸し切り、純潔サロン04"
+}