spotlite 0.7.2 → 0.8.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: cde36af5671b2f9ca6b4a495c1776c5ba6517e14
+  data.tar.gz: af543b22410dfedcdbb8ae0b9891f5001641d43a
+SHA512:
+  metadata.gz: 6b4aa323ca78309219b4859defec059dadffa3745ad134cfe5f96425306ee2f5a65df089396f0b5a3186d50495689b04b7d086a4b7c6a98bb4125adf818e61f4
+  data.tar.gz: e3dac3d146ef5d59712706114c838ad03c9916653556c2c97902989245ec38e850ebc6718bea65d3e4b6375d0ea75ba6ae653dd04f595fad3be552876ff3d4c8

data/CHANGELOG.md

@@ -1,3 +1,17 @@
+## v0.8.0 06-Jul-2014
+
+* Added `Spotlite::Movie.search` class method to perform advanced search by parameters
+* Added `Spotlite::Movie.find` class method for simple title search
+* Added `technical` method for `Movie` class to get movie technical information
+* Added `alternative_titles` method for `Movie` class
+* Added `Spotlite::Person.search` class method to perform advanced search by parameters
+* Added `Spotlite::Person.find` class method for simple name search
+* Added `filmography` method for `Person` class
+* Moved task to refresh fixtures out of namespace. Now it can be called as `rake refresh_fixtures`
+* Added `rake console` task. It will automatically `require 'spotlite'`
+* Removed `Spotlite::List` class and all of its subclasses
+
+
 ## v0.7.2 15-Dec-2013
 
 * Fixes for updated Top 250 page

data/README.md

@@ -22,62 +22,29 @@ Or install it yourself as:
 
     > require 'spotlite'
     # Access movie directly by its IMDb ID
-    > movie = Spotlite::Movie.new("0133093")
+    > movie = Spotlite::Movie.new('0133093')
     # Or use search instead
-    > list = Spotlite::Search.new("the matrix").movies
+    > list = Spotlite::Movie.find('the matrix')
     > movie = list.first
     > movie.title
     => "The Matrix"
     > movie.runtime
     => 136
     > movie.genres
-    => ["Action", "Adventure", "Sci-Fi"]
+    => ["Action", "Sci-Fi"]
     > movie.countries
     => [{:code=>"us", :name=>"USA"}, {:code=>"au", :name=>"Australia"}]
     > movie.directors
-    => [{:imdb_id=>"0905152", :name=>"Andy Wachowski"},
-    {:imdb_id=>"0905154", :name=>"Lana Wachowski"}]
-    > movie.cast[0..4]
-    => [{:imdb_id=>"0000206", :name=>"Keanu Reeves", :character=>"Neo"},
-    {:imdb_id=>"0000401", :name=>"Laurence Fishburne", :character=>"Morpheus"},
-    {:imdb_id=>"0005251", :name=>"Carrie-Anne Moss", :character=>"Trinity"},
-    {:imdb_id=>"0915989", :name=>"Hugo Weaving", :character=>"Agent Smith"},
-    {:imdb_id=>"0287825", :name=>"Gloria Foster", :character=>"Oracle"}]
+    => [#<Spotlite::Person:0x007f96a092be70 @imdb_id="0905152", @name="Andy Wachowski", @url="http://www.imdb.com/name/nm0905152/", @credits_category="Directed by", @credits_text="(as The Wachowski Brothers)">, #<Spotlite::Person:0x007f96a092bda8 @imdb_id="0905154", @name="Lana Wachowski", @url="http://www.imdb.com/name/nm0905154/", @credits_category="Directed by", @credits_text="(as The Wachowski Brothers)">]
+    > movie.cast[0..2]
+    => [#<Spotlite::Person:0x007f96a19521a0 @imdb_id="0000206", @name="Keanu Reeves", @url="http://www.imdb.com/name/nm0000206/", @credits_category="Cast", @credits_text="Neo">, #<Spotlite::Person:0x007f96a1951c28 @imdb_id="0000401", @name="Laurence Fishburne", @url="http://www.imdb.com/name/nm0000401/", @credits_category="Cast", @credits_text="Morpheus">, #<Spotlite::Person:0x007f96a1951a70 @imdb_id="0005251", @name="Carrie-Anne Moss", @url="http://www.imdb.com/name/nm0005251/", @credits_category="Cast", @credits_text="Trinity">]
     
-### IMDb Top 250
-
-    > list = Spotlite::Top.new.movies
-
-### Movies opening this week
-
-    > list = Spotlite::OpeningThisWeek.new.movies
-
-### This week's box office top 10
-
-    > list = Spotlite::BoxOfficeTop.new.movies
-
-### Movies that are coming soon
-
-    > list = Spotlite::ComingSoon.new.movies
-
 ## Important notice
 
 Movie titles will be localized if movie has an alternative title specific to your country.
 Example: _The Great Gatsby_ (http://www.imdb.com/title/tt1343092) has an alternative title _Gatsby le magnifique_ `France (imdb display title)` and will be localized accordingly based on your IP address, if you reside in France. Non-localized title is still avaliable with `original_title` method.
 Sorry, there is nothing I can do about it at the moment.
 
-## Class tree
-    
-    Spotlite
-    |
-    |- Movie
-    `- List
-       |- Search
-       |- Top
-       `- InTheaters
-          |- OpeningThisWeek
-          `- BoxOfficeTop
-
 ## Contributing
 
 1. Fork it
@@ -125,13 +92,13 @@ First, run tests with `LIVE_TEST=true` environment variable:
    
 Adjust methods that are failing, according to the new layout. And refresh fixtures:
 
-    $ rake fixtures:refresh
+    $ rake refresh_fixtures
 
 It will run through all elements of `IMDB_SAMPLES` hash to get fresh data.
 
 ## License
 
-Copyright (c) 2013 Artem Pakk
+Copyright (c) 2013-2014 Artem Pakk
 
 MIT License
 

data/Rakefile

@@ -1,8 +1,28 @@
 require "bundler/gem_tasks"
 require "rspec/core/rake_task"
 
-load File.expand_path(File.dirname(__FILE__) + "/tasks/fixtures.rake")
-
 RSpec::Core::RakeTask.new("spec")
-
 task :default => :spec
+
+desc "Launch irb console"
+task :console do
+  require 'irb'
+  require 'irb/completion'
+  require 'spotlite'
+  ARGV.clear
+  IRB.start
+end
+
+desc "Refresh spec fixtures with fresh data from IMDb.com"
+task :refresh_fixtures do
+  require File.expand_path(File.dirname(__FILE__) + "/spec/spec_helper")
+
+  IMDB_SAMPLES.each_pair do |url, fixture|
+    page = `curl -isH "Accept-Language: en-us" '#{url}'`
+    
+    File.open(File.expand_path(File.dirname(__FILE__) + "/spec/fixtures/#{fixture}"), 'w') do |f|
+      f.write(page)
+    end
+
+  end
+end

data/lib/spotlite.rb

@@ -8,11 +8,4 @@ require "nokogiri"
 require "spotlite/version"
 require "spotlite/movie"
 require "spotlite/person"
-require "spotlite/list"
-require "spotlite/search"
-require "spotlite/top"
-require "spotlite/in_theaters"
-require "spotlite/opening_this_week"
-require "spotlite/box_office_top"
-require "spotlite/coming_soon"
-require "spotlite/string_extensions"
+require "spotlite/string_extensions"

data/lib/spotlite/movie.rb

@@ -1,12 +1,13 @@
-module Spotlite
+require 'cgi'
 
+module Spotlite
   # Represents a movie on IMDb.com
   class Movie
-    attr_accessor :imdb_id, :title, :year
+    attr_accessor :imdb_id, :url
     
     # Initialize a new movie object by its IMDb ID as a string
     #
-    #   movie = Spotlite::Movie.new("0133093")
+    #   movie = Spotlite::Movie.new('0133093')
     #
     # Spotlite::Movie class objects are lazy loading. No HTTP request
     # will be performed upon object initialization. HTTP request will
@@ -14,10 +15,49 @@ module Spotlite
     # Currently, all data is spread across 6 pages: main movie page,
     # /releaseinfo, /fullcredits, /keywords, /trivia, and /criticreviews
     def initialize(imdb_id, title = nil, year = nil)
-      @imdb_id = imdb_id
+      @imdb_id = "%07d" % imdb_id.to_i
       @title   = title
       @year    = year
-      @url     = "http://www.imdb.com/title/tt#{imdb_id}/"
+      @url     = "http://www.imdb.com/title/tt#{@imdb_id}/"
+    end
+    
+    # Returns a list of movies as an array of +Spotlite::Movie+ objects
+    # Takes single parameter and searches for movies by title and alternative titles
+    def self.find(query)
+      results = Nokogiri::HTML open("http://www.imdb.com/find?q=#{CGI::escape(query)}&s=tt&ttype=ft", 'Accept-Language' => 'en-us')
+      results.css('.result_text').map do |result|
+        imdb_id = result.at('a')['href'].parse_imdb_id
+        title   = result.at('a').text.strip
+        year    = result.children.take(3).last.text.parse_year
+      
+        [imdb_id, title, year]
+      end.map do |values|
+        self.new(*values)
+      end
+    end
+    
+    # Returns a list of movies as an array of +Spotlite::Movie+ objects
+    # Takes optional parameters as a hash
+    # See https://github.com/defeed/spotlite/wiki/Advanced-movie-search for details
+    def self.search(params = {})
+      defaults = {
+        :title_type => 'feature',
+        :view       => 'simple',
+        :count      => 250,
+        :start      => 1,
+        :sort       => 'moviemeter,asc'
+      }
+      params = defaults.merge(params).map{ |k, v| "#{k}=#{v}" }.join('&')
+      results = Nokogiri::HTML open("http://www.imdb.com/search/title?#{params}", 'Accept-Language' => 'en-us')
+      results.css('td.title').map do |result|
+        imdb_id = result.at('a')['href'].parse_imdb_id
+        title   = result.at('a').text.strip
+        year    = result.at('.year_type').text.parse_year
+      
+        [imdb_id, title, year]
+      end.map do |values|
+        self.new(*values)
+      end
     end
     
     # Returns title as a string
@@ -27,7 +67,7 @@ module Spotlite
     
     # Returns original non-english title as a string
     def original_title
-      details.at("h1.header span.title-extra[itemprop='name']").children.first.text.gsub('"', "").strip rescue nil
+      details.at("h1.header span.title-extra[itemprop='name']").children.first.text.gsub('"', '').strip rescue nil
     end
     
     # Returns year of original release as an integer
@@ -42,12 +82,12 @@ module Spotlite
     
     # Returns Metascore rating as an integer
     def metascore
-      details.at("div.star-box-details a[href^=criticreviews]").text.strip.split("/").first.to_i rescue nil
+      details.at("div.star-box-details a[href^=criticreviews]").text.strip.split('/').first.to_i rescue nil
     end
     
     # Returns number of votes as an integer
     def votes
-      details.at("div.star-box-details span[itemprop='ratingCount']").text.gsub(/[^\d+]/, "").to_i rescue nil
+      details.at("div.star-box-details span[itemprop='ratingCount']").text.gsub(/[^\d+]/, '').to_i rescue nil
     end
     
     # Returns short description as a string
@@ -75,7 +115,7 @@ module Spotlite
     def countries
       array = []
       details.css("div.txt-block a[href^='/country/']").each do |node|
-        array << {:code => node["href"].clean_href, :name => node.text.strip}
+        array << {:code => node['href'].clean_href, :name => node.text.strip}
       end
       
       array
@@ -86,7 +126,7 @@ module Spotlite
     def languages
       array = []
       details.css("div.txt-block a[href^='/language/']").each do |node|
-        array << {:code => node["href"].clean_href, :name => node.text.strip}
+        array << {:code => node['href'].clean_href, :name => node.text.strip}
       end
       
       array
@@ -94,25 +134,24 @@ module Spotlite
     
     # Returns runtime (length) in minutes as an integer
     def runtime
-      details.at("time[itemprop='duration']").text.to_i rescue nil ||
-      details.at("#overview-top .infobar").text.strip[/\d{2,3} min/].to_i rescue nil
+      details.at("time[itemprop='duration']").text.gsub(',', '').to_i rescue nil
     end
     
     # Returns primary poster URL as a string
     def poster_url
-      src = details.at("#img_primary img")["src"] rescue nil
+      src = details.at('#img_primary img')['src'] rescue nil
       
       if src =~ /^(http:.+@@)/ || src =~ /^(http:.+?)\.[^\/]+$/
-        $1 + ".jpg"
+        $1 + '.jpg'
       end
     end
     
     # Returns an array of recommended movies as an array of initialized objects of +Movie+ class
     def recommended_movies
-      details.css(".rec-title").map do |node|
+      details.css('.rec-title').map do |node|
         imdb_id = node.at("a[href^='/title/tt']")['href'].parse_imdb_id
-        title   = node.at("a").text.strip
-        year    = node.at("span").text.parse_year
+        title   = node.at('a').text.strip
+        year    = node.at('span').text.parse_year
       
         [imdb_id, title, year]
       end.map do |values|
@@ -130,50 +169,95 @@ module Spotlite
       movie_trivia.css("div.sodatext").map { |node| node.text.strip } rescue []
     end
     
-    # Returns a list of directors as an array of hashes
-    # with keys: +imdb_id+ (string) and +name+ (string)
+    # Returns a list of movie alternative titles as an array of hashes
+    # with keys +title+ (string) and +comment+ (string)
+    def alternative_titles
+      array = []
+      release_info.css('#akas').css('tr').map do |row|
+        cells = row.css('td')
+        array << { :title => cells.last.text.strip, :comment => cells.first.text.strip }
+      end
+      
+      array
+    end
+    
+    # Returns a list of directors as an array of +Spotlite::Person+ objects
     def directors
-      parse_staff("Directed by")
+      parse_crew('Directed by')
     end
     
-    # Returns a list of writers as an array of hashes
-    # with keys: +imdb_id+ (string) and +name+ (string)
+    # Returns a list of writers as an array of +Spotlite::Person+ objects
     def writers
-      parse_staff("Writing Credits")
+      parse_crew('Writing Credits')
     end
     
-    # Returns a list of producers as an array of hashes
-    # with keys: +imdb_id+ (string) and +name+ (string)
+    # Returns a list of producers as an array of +Spotlite::Person+ objects
     def producers
-      parse_staff("Produced by")
+      parse_crew('Produced by')
+    end
+    
+    # Returns a list of starred actors as an array of +Spotlite::Person+ objects
+    def stars
+      details.css("td#overview-top div[itemprop='actors'] a[href^='/name/nm']").map do |node|
+        imdb_id = node['href'].parse_imdb_id
+        name = node.text.strip
+        
+        [imdb_id, name]
+      end.map do |values|
+        Spotlite::Person.new(*values)
+      end
     end
     
-    # Returns a list of actors as an array of hashes
-    # with keys: +imdb_id+ (string), +name+ (string), and +character+ (string)
+    # Returns a list of actors as an array +Spotlite::Person+ objects
     def cast
-      table = full_credits.css("table.cast_list")
-      names = table.css("td[itemprop='actor']").map { |node| node.text.strip } rescue []
-      links = table.css("td[itemprop='actor'] a").map { |node| node["href"].clean_href } rescue []
-      imdb_ids = links.map { |link| link.parse_imdb_id } unless links.empty?
-      characters = table.css("td.character").map { |node| node.text.clean_character }
-      
-      array = []
-      0.upto(names.size - 1) do |i|
-        array << {:imdb_id => imdb_ids[i], :name => names[i], :character => characters[i]}
+      full_credits.css('table.cast_list tr').reject do |row|
+        # Skip 'Rest of cast' row
+        row.children.size == 1
+      end.map do |row|
+        imdb_id = row.at('td:nth-child(2) a')['href'].parse_imdb_id
+        name = row.at('td:nth-child(2) a').text.strip_whitespace
+        credits_text = row.last_element_child.text.strip_whitespace
+        
+        [imdb_id, name, 'Cast', credits_text]
+      end.map do |values|
+        Spotlite::Person.new(*values)
       end
-      
-      array
     end
     
-    # Returns a list of starred actors as an array of hashes
-    # with keys: +imdb_id+ (string) and +name+ (string)
-    def stars
+    # Returns a list of crew members of a certain category as an array +Spotlite::Person+ objects
+    def parse_crew(category)
+      table = full_credits.search("[text()^='#{category}']").first.next_element rescue []
+      if table && table.name == 'table'
+        table.css('tr').reject do |row|
+          # Skip empty table rows with one non-braking space
+          row.text.strip.size == 1
+        end.map do |row|
+          imdb_id = row.first_element_child.at('a')['href'].parse_imdb_id
+          name = row.first_element_child.at('a').text.strip_whitespace
+          credits_text = row.last_element_child.text.strip_whitespace
+          
+          [imdb_id, name, category, credits_text]
+        end.map do |values|
+          Spotlite::Person.new(*values)
+        end
+      end
+    end
+    
+    # Combines all crew categories and returns an array of +Spotlite::Person+ objects
+    def crew
+      crew_categories.map{ |category| parse_crew(category) }.flatten
+    end
+    
+    # Returns combined `cast` and `crew` as an array of +Spotlite::Person+ objects
+    def credits
+      cast + crew
+    end
+    
+    # Returns available crew categories, e.g. "Art Department", "Writing Credits", or "Stunts", as an array of strings
+    def crew_categories
       array = []
-      details.css("td#overview-top div[itemprop='actors'] a[href^='/name/nm']").map do |node|
-        name = node.text.strip
-        imdb_id = node["href"].parse_imdb_id
-        
-        array << {:imdb_id => imdb_id, :name => name}
+      full_credits.css('h4.dataHeaderWithBorder').reject{ |h| h['id'] == 'cast' }.map do |node|
+        array << (node.children.size > 1 ? node.children.first.text.strip_whitespace : node.children.text.strip_whitespace)
       end
       
       array
@@ -186,15 +270,17 @@ module Spotlite
     # If day and month are unknown, 1st of January is assigned
     def release_dates
       array = []
-      release_info.at("#release_dates").css("tr").map do |row|
-        code = row.at("a")["href"].clean_href.split("=").last.downcase rescue nil
-        region = row.at("a").text rescue nil
-        date = row.at("td.release_date").text.strip.parse_date rescue nil
-        comment = row.css("td").last.text.strip.clean_release_comment rescue nil
+      table = release_info.at('#release_dates')
+      table.css('tr').map do |row|
+        cells = row.css('td')
+        code = cells.first.at('a')['href'].clean_href.split('=').last.downcase rescue nil
+        region = cells.first.at('a').text rescue nil
+        date = cells.at('.release_date').text.strip.parse_date
+        comment = cells.last.text.strip.clean_release_comment
         comment = nil if comment.empty?
         
         array << {:code => code, :region => region, :date => date, :comment => comment}
-      end
+      end unless table.nil?
       
       array
     end
@@ -223,17 +309,33 @@ module Spotlite
     # Returns URLs of movie still frames as an array of strings
     def images
       array = []
-      still_frames.css("#media_index_thumbnail_grid img").map do |image|
-        src = image["src"] rescue nil
+      still_frames.css('#media_index_thumbnail_grid img').map do |image|
+        src = image['src'] rescue nil
       
         if src =~ /^(http:.+@@)/ || src =~ /^(http:.+?)\.[^\/]+$/
-          array << $1 + ".jpg"
+          array << $1 + '.jpg'
         end
       end
       
       array
     end
     
+    # Returns technical information like film length, aspect ratio, cameras, etc. as a hash of arrays of strings
+    def technical
+      hash = {}
+      table = technical_info.at_css('#technical_content table') rescue nil
+      
+      table.css('tr').map do |row|
+        hash[row.css('td').first.text.strip] = row.css('td').last.children.
+          map(&:text).
+          map(&:strip_whitespace).
+          reject(&:empty?).
+          reject{|i| i == '|'}
+      end unless table.nil?
+      
+      hash
+    end
+    
     private
     
     def details # :nodoc:
@@ -241,48 +343,35 @@ module Spotlite
     end
     
     def release_info # :nodoc:
-      @release_info ||= open_page("releaseinfo")
+      @release_info ||= open_page('releaseinfo')
     end
     
     def full_credits # :nodoc:
-      @full_credits ||= open_page("fullcredits")
+      @full_credits ||= open_page('fullcredits')
     end
     
     def plot_keywords # :nodoc:
-      @plot_keywords ||= open_page("keywords")
+      @plot_keywords ||= open_page('keywords')
     end
     
     def movie_trivia # :nodoc:
-      @movie_trivia ||= open_page("trivia")
+      @movie_trivia ||= open_page('trivia')
     end
     
     def reviews
-      @reviews ||= open_page("criticreviews")
+      @reviews ||= open_page('criticreviews')
     end
     
-    def still_frames
-      @still_frames ||= open_page("mediaindex?refine=still_frame")
+    def still_frames # :nodoc:
+      @still_frames ||= open_page('mediaindex?refine=still_frame')
     end
     
-    def open_page(page = nil) # :nodoc:
-      Nokogiri::HTML(open("http://www.imdb.com/title/tt#{@imdb_id}/#{page}",
-                          "Accept-Language" => "en-us"))
+    def technical_info
+      @technical_info ||= open_page('technical')
     end
     
-    def parse_staff(staff) # :nodoc:
-      array = []
-      # table = full_credits.at("a[name='#{staff}']").parent.parent.parent.parent rescue nil
-      table = full_credits.search("[text()*='#{staff}']").first.next_element rescue nil
-      if table && table.name == "table"
-        table.css("a[href^='/name/nm']").map do |node|
-          imdb_id = node["href"].parse_imdb_id
-          name = node.text.strip
-        
-          array << {:imdb_id => imdb_id, :name => name}
-        end
-      end
-      
-      array.uniq
+    def open_page(page = nil) # :nodoc:
+      Nokogiri::HTML open("#{@url}#{page}", 'Accept-Language' => 'en-us')
     end
   end