imdb-scan 0.0.1

data/lib/imdb/movie.rb

@@ -0,0 +1,165 @@
+module IMDB
+  # Get movie information with IMDB movie id.
+  # @example Get Yahsi Bati movie title and cast listing [http://www.imdb.com/title/tt1567448/]
+  #   m = IMDB::Movie.new('1567448')
+  #   puts m.title
+  class Movie < IMDB::Skeleton
+    attr_accessor :link, :imdb_id
+
+    def initialize(id_of)
+      # !!!DON'T FORGET DEFINE NEW METHODS IN SUPER!!!
+      super("Movie", { :imdb_id => String,
+                       :poster => String,
+                       :title => String,
+                       :release_date => String,
+                       :cast => Array,
+                       :photos => Array,
+                       :director => String,
+                       :director_person => Person,
+                       :genres => Array,
+                       :rating => Float,
+                       :movielength => Integer,
+                       :short_description => String,
+                       :writers => Array }, [:imdb_id])
+
+      @imdb_id = id_of
+
+      @link = "http://www.imdb.com/title/tt#{@imdb_id}"
+    end
+
+    # Get movie poster address
+    # @return [String]
+    def poster
+      src = doc.at("#img_primary img")["src"] rescue nil
+      unless src.nil?
+        if src.match(/\._V1/)
+          return src.match(/(.*)\._V1.*(.jpg)/)[1, 2].join
+        else
+          return src
+        end
+      end
+      src
+    end
+
+    # Get movie title
+    # @return [String]
+    def title
+      doc.at("//head/meta[@name='title']")["content"].split(/\(\d+\)/)[0].strip! ||
+          doc.at("h1.header").children.first.text.strip
+
+    end
+
+    # Get movie cast listing
+    # @return [Cast[]]
+    def cast
+      doc.search("table.cast tr").map do |link|
+        #picture = link.children[0].search("img")[0]["src"] rescue nil
+        #name = link.children[1].content.strip rescue nil
+        id = link.children[1].search('a[@href^="/name/nm"]').first["href"].match(/\/name\/nm([0-9]+)/)[1] rescue nil
+        char = link.children[3].content.strip rescue nil
+        unless id.nil?
+          person = IMDB::Person.new(id)
+          IMDB::Cast.new(self, person, char)
+        end
+      end.compact
+    end
+
+    # Get movie photos
+    # @return [Array]
+    def photos
+      begin
+        doc.search('#main .thumb_list img').map { |i| i["src"] }
+      rescue
+        nil
+      end
+    end
+
+    # Get release date
+    # @return [String]
+    def release_date
+      if (node = doc.xpath("//h4[contains(., 'Release Date')]/..")).length > 0
+        date = node.search("time").first["datetime"]
+        if date.match /^\d{4}$/
+          "#{date}-01-01"
+        else
+          Date.parse(date).to_s
+        end
+      else
+        year = doc.at("h1.header .nobr").text[/\d{4}/]
+        "#{year}-01-01"
+      end
+    rescue
+      nil
+    end
+
+    # Get Director
+    # @return [String]
+    def director
+      self.director_person.name rescue nil
+    end
+
+    # Get Director Person class
+    # @return [Person]
+    def director_person
+      begin
+        link=doc.xpath("//h4[contains(., 'Director')]/..").at('a[@href^="/name/nm"]')
+        profile = link['href'].match(/\/name\/nm([0-9]+)/)[1] rescue nil
+        IMDB::Person.new(profile) unless profile.nil?
+      rescue
+        nil
+      end
+    end
+
+    # Genre List
+    # @return [Array]
+    def genres
+      doc.xpath("//h4[contains(., 'Genre')]/..").search("a").map { |g|
+        g.content unless g.content =~ /See more/
+      }.compact
+    rescue
+      nil
+    end
+
+    # Writer List
+    # @return [Float]
+    def rating
+      @rating ||= doc.search(".star-box-giga-star").text.strip.to_f
+    rescue
+      nil
+    end
+
+    #Get the movielength of the movie in minutes
+    # @return [Integer]
+    def movielength
+      doc.at("//h4[text()='Runtime:']/..").inner_html[/\d+ min/].to_i rescue nil
+    end
+
+    # Writer List
+    # @return [Array]
+    def writers
+      doc.xpath("//a[@name='writers']/../../../..").search('a[@href^="/name/nm"]').map { |w|
+        profile = w['href'].match(/\/name\/nm([0-9]+)/)[1] rescue nil
+        IMDB::Person.new(profile) unless profile.nil?
+      }
+    end
+
+    # @return [String]
+    def short_description
+      doc.at("#overview-top p[itemprop=description]").text.strip
+    end
+
+    private
+
+    def doc
+      if caller[0] =~ /`([^']*)'/ and ($1 == "cast" or $1 == "writers")
+        @doc_full ||= Nokogiri::HTML(open("#{@link}/fullcredits"))
+      elsif caller[0] =~ /`([^']*)'/ and ($1 == "photos")
+        @doc_photo ||= Nokogiri::HTML(open("#{@link}/mediaindex"))
+      else
+        @doc ||= Nokogiri::HTML(open("#{@link}"))
+      end
+    end
+
+  end # Movie
+end # IMDB
+

data/lib/imdb/person.rb

@@ -0,0 +1,128 @@
+module IMDB
+  # Get Person information with IMDB person id.
+  # @example Get Christian Bale information [http://www.imdb.com/name/nm0000288/]
+  #   m = IMDB::Person.new('0000288')
+  #   puts m.name
+  #   puts m.real_name
+  class Person < IMDB::Skeleton
+    attr_accessor :id
+
+    def initialize(imdb_id)
+      super("Person", { :id => String,
+                        :name => String,
+                        :real_name => String,
+                        :birthdate => Date,
+                        :deathdate => Date,
+                        :nationality => String,
+                        :height => String,
+                        :biography => String,
+                        :photo => String,
+                        :profile_path => String,
+                        :filmography => Hash,
+                        :main_document => Nokogiri,
+                        :bio_document => Nokogiri,
+                        :photo_document => Nokogiri,
+                        :photo_document_url => String }, [:id])
+      @id = imdb_id
+    end
+
+    #Get the profile path
+    #@return [String]
+    def profile_path
+      "/name/nm#{@id}"
+    end
+
+    #Get the name of the person
+    #@return [String]
+    def name
+      bio_document.at("a[@class='main']").inner_text rescue nil
+    end
+
+    #Get The Real Born name of the Person
+    #@return [String]
+    def real_name
+      bio_document.at("h5[text()*='Birth Name']").next.inner_text.strip rescue nil
+    end
+
+    #Get The Birth Date
+    #@return [Date]
+    def birthdate
+      date_month = bio_document.at("h5[text()*='Date of Birth']").next_element.inner_text.strip rescue ""
+      year = bio_document.at("a[@href*='birth_year']").inner_text.strip rescue ""
+      Date.parse("#{date_month} #{year}") rescue nil
+    end
+
+    #Get The death date else nil
+    #@return [Date]
+    def deathdate
+      date_month = bio_document.at("h5[text()*='Date of Death']").next_element.inner_text.strip rescue ""
+      year = bio_document.at("a[@href*='death_date']").inner_text.strip rescue ""
+      Date.parse("#{date_month} #{year}") rescue nil
+    end
+
+    #Get the Nationality
+    #@return [String]
+    def nationality
+      bio_document.at("a[@href*='birth_place']").inner_text.strip rescue nil
+    end
+
+    #Get the height
+    #@return [String]
+    def height
+      bio_document.at("h5[text()*='Height']").next.inner_text.match(/\((.+)\)/)[1] rescue nil
+    end
+
+    #Get The Biography
+    #@return [String]
+    def biography
+      bio_document.at("h5[text()*='Biography']").next_element.inner_text rescue nil
+    end
+
+    #Return the principal Photo
+    #@return [String]
+    def photo
+      photo_document.at("img#primary-img").get_attribute('src') if photo_document rescue nil
+    end
+
+    #Return the Filmography
+    #for the moment I can't make subdivision of this, then i take all in an array
+    #@return [Movie]
+    def filmography
+      #@return [Hash]
+      #     writer: [Movie]
+      #     actor: [Movie]
+      #     director: [Movie]
+      #     composer: [Movie]
+      #as_writer = main_document.at("#filmo-head-Writer").next_element.search('b a').map { |e| e.get_attribute('href')[/tt(\d+)/, 1] } rescue []
+      #as_actor = main_document.at("#filmo-head-Actor").next_element.search('b a').map { |e| e.get_attribute('href')[/tt(\d+)/, 1] } rescue []
+      #as_director = main_document.at("#filmo-head-Director").next_element.search('b a').map { |e| e.get_attribute('href')[/tt(\d+)/, 1] } rescue []
+      #as_composer = main_document.at("#filmo-head-Composer").next_element.search('b a').map { |e| e.get_attribute('href')[/tt(\d+)/, 1] } rescue []
+      #{ writer: as_writer.map { |m| Movie.new(m) }, actor: as_actor.map { |m| Movie.new(m) }, director: as_director.map { |m| Movie.new(m) }, composer: as_composer.map { |m| Movie.new(m) } }
+      films=main_document.css(".filmo-row b a").map { |e| e.get_attribute('href')[/tt(\d+)/, 1] } rescue []
+      films.map { |f| Movie.new(f.to_i) }
+    end
+
+
+    def main_document
+      #@main_document ||= Nokogiri open("http://www.imdb.com#{profile_path}")
+      @main_document ||= Nokogiri::HTML(open("http://www.imdb.com#{profile_path}"))
+    end
+
+    def bio_document
+      @bio_document ||= Nokogiri open("http://www.imdb.com#{profile_path}/bio")
+    end
+
+    def photo_document
+      @photo_document ||= if photo_document_url then
+                            Nokogiri open("http://www.imdb.com" + photo_document_url)
+                          else
+                            nil
+                          end
+    end
+
+    def photo_document_url
+      bio_document.at(".photo a[@name=headshot]").get_attribute('href') rescue nil
+    end
+
+  end
+end

data/lib/imdb/search.rb

@@ -0,0 +1,82 @@
+module IMDB
+  class Search
+    def movie(keyword)
+      doc = Nokogiri::HTML(open("http://www.imdb.com/find?s=tt&q=#{CGI.escape(keyword)}"))
+      @ret_val = []
+      if doc.at("h1.header") # we're already being redirected to movie's page
+        single_result(doc)
+      else
+        result_list(doc)
+      end
+      @ret_val
+    end
+
+    def to_hash
+      i = 0
+      tmp_hash = { }
+      @ret_val.each { |r|
+        tmp_hash[i] = r.to_hash
+        i = i + 1
+      }
+      tmp_hash
+    end
+
+    def to_json
+      to_hash.to_json
+    end
+
+    private
+    def single_result(doc)
+      title = doc.at("h1.header")
+      link = doc.at("link[rel=canonical]")["href"]
+      title = title.text.strip.gsub(/\s+/, " ")
+      @ret_val << IMDB::Result.new(link[/\d+/], title, link)
+    end
+
+    def result_list(doc)
+      @ret_val = doc.search('a[@href^="/title/tt"]').reduce([]) do |ret_val, node|
+        unless node.content.blank?
+          link = "http://www.imdb.com#{node['href']}"
+          id = node["href"][/\d+/]
+          ret_val << IMDB::Result.new(id, node.content, link)
+        end
+        ret_val
+      end
+
+      h = {}
+      @ret_val.each {|e| h[e.imdb_id]=e}
+
+      @ret_val=h.values
+    end
+  end # Search
+
+  class Result < IMDB::Skeleton
+    def initialize(imdb_id, title, link)
+      super("Result", {
+          :title => String,
+          :link => String,
+          :imdb_id => String }, [:imdb_id])
+      @title = title
+      @link = link
+      @imdb_id = imdb_id
+    end
+
+    def title
+      @title
+    end
+
+    def link
+      @link
+    end
+
+    def imdb_id
+      @imdb_id
+    end
+
+    def movie
+      Movie.new(@imdb_id)
+    end
+
+  end
+end
+

data/lib/imdb/skeleton.rb

@@ -0,0 +1,83 @@
+module IMDB
+  # IMDB generic interface.
+  class Skeleton
+    attr_accessor :model, :method_names
+
+    def initialize(model_name = "", method_names = {}, keys = [])
+      if IMDB::Configuration.caching
+        @model = Class.new do
+          include MongoMapper::Document
+          set_collection_name model_name
+          method_names.each { |m, t|
+            key m, t
+          }
+        end
+        class_eval do
+          method_names.each_key { |meth|
+            unless keys.include?(meth)
+              old_meth = "old_#{meth}".to_sym
+              alias_method old_meth, meth.to_sym
+              define_method meth do
+                k = keys.to_imdb_hash { |k| k; self.send(k) }
+
+                @db_query = self.model.first(k)
+
+                if @db_query.nil?
+                  @db_query = self.model.new(keys.to_imdb_hash { |k| k; self.send(k)})
+                  @db_query.save
+                end
+
+                if @db_query[meth].nil? or (@db_query[meth].length.zero? if @db_query[meth].kind_of?(Array))
+                  a = send(old_meth)
+                  if a.kind_of?(Array)
+                    a.compact!
+                    a.map! { |c|
+                      if c.kind_of?(String)
+                        c
+                      else
+                        c.to_hash
+                      end
+                    }
+                    @db_query[meth] = a
+                  else
+                    @db_query[meth] = a
+                  end
+                  @db_query.save
+                end
+                @db_query[meth]
+              end
+            end
+          }
+        end
+      end
+      @method_names = method_names
+    end
+
+    # Serialize method's output to json
+    def to_json(*a)
+      tmp_hash = to_hash
+
+      tmp_hash.to_json(*a)
+    end
+
+    def to_hash(*a)
+      tmp_hash = {}
+      @method_names.each_key { |x|
+        evaled = self.send x
+        if evaled.kind_of?(Array)
+          tmp_hash[x] = evaled.collect! {|e|
+            e
+          }
+        elsif evaled.kind_of?(String)
+          tmp_hash[x] = evaled
+        end
+      }
+      tmp_hash
+    end
+
+    def self.json_create(o)
+      new(*o['data'])
+    end
+  end
+end
+